List of Contributors

A. Arenas, Department of Rehabilitation Medicine, University of Miami School of Medicine, P.O. Box
016960 (D-461), Miami, FL 33101, USA
K. Anderson, Reeve-Irvine Research Center, 1216 GNRF, University of California, Irvine, Irvine, CA
92697-4292, USA
H.W.G. Baker, University of Melbourne Department of Obstetrics and Gynaecology Melbourne IVF
Reproductive Services, Royal Women’s Hospital, Carlton, Vic. 3058, Australia
M.S. Beattie, Department of Neuroscience, laboratory of CNS Repair and Spinal Trauma and Repair
Laboratories, The Ohio State University College of Medicine and Pulic Health, Columbus, OH, USA
L.A. Birder, Departments of Medicine and Pharmacology, University of Pittsburgh School of Medicine,
Pittsburgh, PA 15261, USA
A.F. Brading, Oxford Continence Group, University Department of Pharmacology, Mansfield Road,
Oxford OX1 3QT, UK
J.C. Bresnahan, Department of Neuroscience, Laboratory of CNS Repair and Spinal Trauma and Repair
Laboratories, The Ohio State University College of Medicine and Public Health, Columbus, OH, USA
J.A. Brock, Spinal Injuries Research Centre, Prince of Wales Medical Research Institute, Gate 1, Barker
Street, Randwick, NSW 2031, Australia
A. Brown, Biotherapeutics Research Group, The Spinal Cord Injury Team, Robarts Research Institute and
The Graduate Program in Neuroscience, The University of Western Ontario, P.O. Box 5015, 100 Perth
Drive, London, ON N6A 5K8, Canada
D.J. Brown, Victorian Spinal Cord Service, Austin Health, Heidelberg, Vic., Australia
E.A.L. Chung, St Mark’s Hospital, Northwick Park, Watford Road, Harrow, Middlesex, HA1 3UJ, UK
V.E. Claydon, International Collaboration on Repair Discoveries (ICORD), University of British Co-
lumbia, Vancouver, BC V6 T 1Z4, Canada
H.L. Collins, Department of Physiology, Wayne State University School of Medicine, Detroit, MI 48201,
USA
M.D. Craggs, Centre for Spinal Research, Functional Assessment and Restoration, London Spinal Cord
Injuries Unit, Royal National Orthopaedic Hospital NHS Trust, Brockley Hill, Stanmore, Middlesex
HA7 4LP, UK
W.C. de Groat, Departments of Pharmacology and Urology, University of Pittsburgh School of Medicine,
Pittsburgh, PA, USA
G.A. Dekaban, Spinal Cord Injury Team, BioTherapeutics Research Group, Robarts Research Institute,
100 Perth Drive, P.O. Box 5015, London, ON N6A 5K8, Canada
S.E. DiCarlo, Department of Physiology, Wayne State University School of Medicine, Detroit, MI 48201,
USA
J.W. Downie, Department of Urology and Pharmacology, Faculty of Medicine, Dalhousie University,
5850 College St., Halifax, NS B3 H 1X5, Canada
S.L. Elliott, Departments of Psychiatry and Urology, University of British Columbia, British Columbia
Center for Sexual Medicine, Echelon-5, 855 West 12th Avenue, Vancouver, Vancouver Sperm Retrieval
Clinic, Vancouver Hospital and G.F. Strong Rehabilitation Centre, Vancouver, BC, Canada

v
vi

G.S. Emch, Department of Neuroscience, Georgetown University Medical Center, TRB EP04, Washing-
ton, DC 20057, USA
A.V. Emmanuel, St Mark’s Hospital, Northwick Park, Watford Road, Harrow, Middlesex, HA1 3UJ, UK
P. Enck, Department of Psychosomatic Medicine, University Hospitals Tuingen, Schaffhausenstr 113,
72072 Tubingen, Germany
F.A. Frizelle, Colorectal Unit, Department of Surgery, Christchurch Hospital and Burwood Spinal Unit,
Christchurch, New Zealand
R.A. Gaunt, Department of Biomedical Engineering and Center for Neuroscience, University of Alberta,
507 HMRC, Edmonton, AB T6G 2S2, Canada
I. Greving, Department of Internal Medicine, Elisabeth Hospital, Gelsenkirchen, Germany
D. Gris, Spinal Cord Injury Team, BioTherapeutics Research Group, Robarts Research Institute, 100
Perth Drive, P.O. Box 5015, London, ON N6A 5K8, Canada
L.A. Havton, Department of Neurology, David Geffen School of Medicine at University of California Los
Angeles, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA
S.T. Hill, Victorian Spinal Cord Service, Austin Health, Heidelberg and Melbourne IVF Reproductive
Services, Royal Women’s Hospital, Melbourne, Vic., Australia
T.X. Hoang, Department of Neurology, David Geffen School of Medicine at University of California Los
Angeles, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA
C.H. Hubscher, Department of Anatomical Sciences and Neurobiology, University of Louisville School of
Medicine, Louisville, KY 40292, USA
J.E. Jacob, Biotherapeutics Research Group, The Spinal Cord Injury Team, Robarts Research Institute
and The Graduate Program in Neuroscience, The University of Western Ontario, P.O. Box 5015, 100
Perth Drive, London, ON N6A 5K8, Canada
R.D. Johnson, Department of Physiological Sciences, College of Veterinary Medicine and the McKnight
Brain Institute, University of Florida, Gainesville, FL 32610-0144, USA
A.K. Karlsson, Spinal Injuries Unit, Sahlgrenska University Hospital, Institute of Clinical Neuroscience,
Sahlgrenska Academy, S 413 45 Goteborg, Sweden
J.R. Keast, Pain Management Research Institute, University of Sydney at Royal North Shore Hospital, St
Leonards, NSW, Australia
S. Klosterhalfen, Institute of Medical Psychology, University Hospitals Dusseldorf, Germany
A. Krassioukov, International Collaboration on Repair Discoveries (ICORD), Division of Physical Med-
icine, School of Rehabilitation and Department of Medicine, University of British Columbia, Vancou-
ver, BC V6 T 1Z4, Canada and Department of Physical Medicine and Rehabilitation, University of
Western Ontario, London, ON, Canada
G.M. Leedy, Division of Social Work, University of Wyoming, Laramie, WY, USA
I.J. Llewellyn-Smith, Cardiovascular Medicine and Centre for Neuroscience, Flinders University, Bedford
Park, SA 5042, Australia
A.C. Lynch, Colorectal Unit, Department of Surgery, Christchurch Hospital and Burwood Spinal Unit,
Christchurch, New Zealand
D.R. Marsh, Department of Anatomy and Cell Biology, Dalhousie University, Halifax, NS, Canada
C.J. Mathias, Neurovascular Medicine Unit, Faculty of Medicine, Imperial College London at St. Mary’s
Hospital, London W2 1NY, UK, Autonomic Unit, National Hospital for Neurology and Neurosurgery,
Queen Square, and Institute of Neurology, University College London, London, UK
E.M. McLachlan, Spinal Injuries Research Centre, Prince of Wales Medical Research Institute, Gate 1,
Barker Street, Randwick, NSW 2031, Australia
Y.S. Nout, Department of Neuroscience, Laboratory of CNS Repair and Spinal Trauma and Repair
Laboratories, The Ohio State University College of Medicine and Public Health, Columbus, OH, USA
 vii

P.J. Potter, Regional Spinal Cord Injury Rehabilitation Program, and Physical Medicine and Rehabil-
itation, St. Joseph’s Health Center, The University of Western Ontario, London, ON, Canada
A. Prochazka, Department of Biomedical Engineering and Center for Neuroscience, University of Alberta,
507 HMRC, Edmonton, AB T6G 2S2, Canada
A.G. Rabchevsky, University of Kentucky, Spinal Cord & Brain Injury Research Center and Department
of Physiology, 741 South Limestone Street, B 371 BBSRB, Lexington, KY 40536-0509, USA
T. Ramalingam, Specialist Registrar in Colorectal Surgery, Headquarters Army Medical Directorate,
Former Army Staff College, Camberley, Surrey GU16 4NP, UK
D.W. Rodenbaugh, Department of Molecular and Integrative Physiology, University of Michigan, Ann
Arbor, MI 48109, USA
L.P. Schramm, Departments of Biomedical Engineering and Neuroscience, The Johns Hopkins University
School of Medicine, 606 Traylor Building, 720 Rutland Avenue, Baltimore, MD 21205, USA
S.J. Shefchyk, Department of Physiology, Faculty of Medicine, University of Manitoba, Winnipeg, MB
R3E 3J7, Canada
M.L. Sipski, Veterans Administration Rehabilitation Research and Development, Center of Excellence in
Functional Recovery and Spinal Cord Injury, Miami, FL 33101, USA
M.A. Vizzard, Departments of Neurology and Anatomy and Neurobiology, University of Vermont College
of Medicine, Burlington, VT 05405, USA
L.C. Weaver, Spinal Cord Injury Laboratory, BioTherapeutics Research Group, Robarts Research In-
stitute, 100 Perth Drive, P.O. Box 5015, London, ON N6A 5K8, Canada
B. Wietek, Department of Radiology, University Hospitals Tubingen, Germany
J.R. Wrathall, Department of Neuroscience, Georgetown University Medical Center, TRB EP04, Wash-
ington, DC 20057, USA
N. Yoshimura, Departments of Pharmacology and Urology, University of Pittsburgh School of Medicine,
Pittsburgh, PA, USA
N.D.T. Zinck, Department of Pharmacology, Faculty of Medicine, Dalhousie University, 5850 College St.,
Halifax, NS B3 H 1X5, Canada
 Dedication

‘‘Spinal cord injury is a ferocious assault on the body that leaves havoc in its wake. Paralysis is
certainly part of its legacy, but there are other equally devastating consequences including
autonomic dysfunction: compromised cardiovascular, bowel, bladder, and sexual function.
Treatments and cures for these losses would greatly improve the quality of life for all of us living
with spinal cord injury. I am hopeful that the multi-faceted and collaborative approach to spinal
cord repair evidenced by this book and its contributors means that there will be effective
therapies for autonomic dysfunction in the not too distant future.’’
Christopher Reeve, September 30, 2004

Christopher Reeve sent this endorsement to us only 10 days before his death. His passionate advocacy for
research that would better the lives of all who have suffered spinal cord injury has affected all of us who
work in this field. We dedicate this book to Christopher Reeve and to all, who like him, strive to overcome
the tragedy of spinal cord injury. As Canadian editors, we particularly would like to thank and
acknowledge the efforts of our advocates, Mr. Rick Hansen and Ms. Barbara Turnbull for their tireless
efforts to bring awareness, expertise and funding to the field of spinal cord injury research, in all of its
dimensions. Finally, we dedicate this book to everyone who has sustained a spinal cord injury and lives
courageously, hoping that the efforts of science will bring timely rewards.

Lynne Weaver and Canio Polosa

on behalf of the contributors,
March 1, 2005

ix
 Foreword

Autonomic dysfunction after spinal cord injury: the perspective of a person with a spinal cord injury.

‘‘Something’s wrong. I’m hot. No, wait, now I’m cold. But why am I sweating? I never sweat. My
legs won’t stop jumping. Now my hands are in tight fists and my torso is tight as well. Why does
my scalp itch? Oh no, the headache is starting. I’m getting dysreflexic. What is causing it? My
catheter seems okay, I don’t feel any kinks. But why isn’t there any urine in the leg bag? Oh no, is
the catheter clogged up? The headache is getting worse. It feels like a nail is being hammered into
my head. What am I going to do? It’s going to get worse. I know it, I know what’s going to
happen. But I’m all alone. I can’t reach anyone to ask for help. How am I going to fix this? I’m
going to have to drive myself to the doctor. I’m trying to adjust everything, but it’s not working.
My head, it hurts so bad. I can’t think straight. Everything looks a little bit blurry. It’s hard to
breathe now, the spasms are so severe. I feel like I’m going to throw up. Why is this happening to
me? Why? I’m crying now, I’m so afraid. I know what can happen if I don’t stop this. My heart is
beating wildly, my head is hurting more and more. It feels like it’s going to explode. I have to
hurry to the doctor before it’s too late. I could die from this y’’

This is what happened to me a few years ago during one of my worst attacks of autonomic dysreflexia.
Fortunately, I was able to make it to the doctor’s office and convince him of the urgency of the situation. He
changed the catheter, which was indeed clogged with sediment, and within a matter of seconds after the
volume of my bladder was reduced, the grossly obvious symptoms vanished. There are after effects from
such an episode, however. The biggest being extreme exhaustion, which is not trivial when you’re paralyzed.
For a little background on my spinal cord injury, I’m classified as a C5 ASIA grade B and was injured in
a motor vehicle accident in December of 1988. The most prominent source of my autonomic dysreflexia
stems from my bladder. While in the hospital, I tried various methods of managing my bladder and finally
settled on using a suprapubic catheter. This provides me with the greatest level of independence, comfort,
and reliability. I do use an anti-cholinergic medication to reduce spasticity in my bladder. This is a very
important point to stress. If I miss my medication or am delayed in taking it, I experience a continuous state
of mild dysreflexia. That consists of increased spasticity in my body, hot/cold flashes, and an itchy scalp. I
do not develop the severe headache, however, but I also do not know what fluctuations are occurring with
my blood pressure when in that state. These symptoms disappear shortly after I resume the medication.
Just as every spinal cord injury is different, the primary stimulus and pattern of symptoms of autonomic
dysreflexia experienced by each individual are different. The two most common stimuli appear to be
bladder or bowel distension. One anecdotal observation is that the onset and intensity of the symptoms
seem to occur faster and more severely with increasing time post-injury. I have experienced this first-hand
and have been told the same thing by many other people with spinal cord injury. Is this truly common
among the majority of people who develop autonomic dysreflexia? If so, what is the underlying biological
mechanism? And, what long-term damage develops in people with chronic spinal cord injury who repeat-
edly experience episodes of dysreflexia? These are but a few of the problems that need to be addressed in the
scientific and clinical settings.

xi
xii

Aside from acute autonomic dysreflexia episodes, there are many autonomic dysfunctions that present
difficulties that people with any level of spinal cord injury have to deal with on a daily basis. Some of the
most prominent problems include impairments in bladder and bowel control, sexual function, body tem-
perature regulation, cardiovascular control, and metabolism. Any of these dysfunctions can significantly
reduce a person’s quality of life.
Now, in addition to being a quadriplegic, I am also a scientist and, when I first entered the field of spinal
cord injury science in 2000, I observed that research regarding autonomic dysfunctions resulting from
spinal cord injury was not very prevalent. Yet these are problems that everybody with cord injury ex-
periences to some degree. This is a perplexing paradox. In an effort to address this issue, I conducted a
study to determine what areas of functional recovery were most important to people living with spinal cord
injury. Regaining bladder/bowel function and eliminating autonomic dysreflexia was the first or second
highest priority for approximately 40% of quadriplegics and paraplegics and, similarly, regaining sexual
function was the first or second highest priority to 28.3% of quadriplegics and 45.5% of paraplegics
(Anderson, 2004). These results demonstrate that research regarding autonomic dysfunctions is extremely
important. To that end, this book has been written about what is already known and to serve as a platform
for fueling future research.
After all, it is all of these autonomic functions that we take for granted when we have them and that
dominate our lives when we lose them.

References

Anderson, K.D. (2004) Targeting recovery: priorities of the spinal cord injured population. J Neurotrauma, 21: 1371–1383.

Kim Anderson
 Contents

List of Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Dedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Foreword by Kim Anderson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii

Overview: Autonomic dysfunction in spinal cord injury: clinical presentation of symptoms and
signs
A-.K. Karlsson (Goteborg, Sweden) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Section I. Anatomical Changes Mediating Autonomic Dysfunction After Cord Injury

1. Effects of spinal cord injury on synaptic inputs to sympathetic preganglionic neurons

I.J. Llewellyn-Smith, L.C. Weaver and J.R. Keast (Bedford Park, SA, London, ON,
Canada and St. Leonards, NSW, Australia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2. Spinal sympathetic interneurons: Their identification and roles after spinal cord injury
L.P. Schramm (Baltimore, MD, USA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3. Which pathways must be spared in the injured human spinal cord to retain cardiovascular
control?
A. Krassioukov (Vancouver, BC, Canada) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Section II. Urinary Bladder Dysfunction

4. Disordered control of the urinary bladder after human spinal cord injury: what are the
problems?
P.J. Potter (London, ON, Canada) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

5. Mechanisms underlying the recovery of lower urinary tract function following spinal cord
injury
W.C. de Groat and N.Yoshimura (Pittsburgh, PA, USA). . . . . . . . . . . . . . . . . . . . . 59

6. Spinal mechanisms contributing to urethral striated sphincter control during continence and
micturition: ‘‘How good things might go bad’’
S.J. Shefchyk (Winnipeg, MB, Canada) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

xiii
xiv

7. Neurochemical plasticity and the role of neurotrophic factors in bladder reflex pathways
after spinal cord injury
M.A. Vizzard (Burlington, VT, USA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

8. Effect of injury severity on lower urinary tract function after experimental spinal
cord injury
J.R. Wrathall and G.S. Emch (Washington, DC, USA) . . . . . . . . . . . . . . . . . . . . . . 117

9. Role of the urothelium in urinary bladder dysfunction following spinal cord injury
L.A. Birder (Pittsburgh, PA, USA). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

10. Plasticity in the injured spinal cord: can we use it to advantage to reestablish effective
bladder voiding and continence?
N.D.T. Zinck and J.W. Downie (Halifax, NS, Canada) . . . . . . . . . . . . . . . . . . . . . . 147

11. Control of urinary bladder function with devices: successes and failures
R.A. Gaunt and A. Prochazka (Edmonton, AB, Canada). . . . . . . . . . . . . . . . . . . . . 163

12. Novel repair strategies to restore bladder function following cauda equina/ conus medullaris
injuries
T.X. Hoang and L.A. Havton (Los Angeles, CA, USA) . . . . . . . . . . . . . . . . . . . . . . 195

13. Pelvic somato-visceral reflexes after spinal cord injury: measures of functional loss and
partial preservation
M.D. Craggs (Stanmore, UK) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

Section III. Cardiovascular Dysfunction

14. The clinical problems in cardiovascular control following spinal cord injury: an overview
A. Krassioukov and V.E. Claydon (Vancouver, BC, Canada) . . . . . . . . . . . . . . . . . . 223

15. Orthostatic hypotension and paroxysmal hypertension in humans with high spinal cord
injury
C.J. Mathias (London, UK). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

16. Autonomic dysreflexia after spinal cord injury: central mechanisms and strategies for
prevention
L.C. Weaver, D.R. Marsh, D. Gris, A. Brown and G.A. Dekaban (London, ON,
Canada and Halifax, NS, Canada) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

17. Segmental organization of spinal reflexes mediating autonomic dysreflexia after spinal cord
injury
A.G. Rabchevsky (Lexington, KY, USA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

18. Spinal cord injury alters cardiac electrophysiology and increases the susceptibility to
ventricular arrhythmias
H.L. Collins, D.W. Rodenbaugh and S.E. DiCarlo (Detroit and Ann Arbor,
MI, USA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
 xv

19. Adaptations of peripheral vasoconstrictor pathways after spinal cord injury

E.M. McLachlan and J.A. Brock (Randwick, NSW, Australia). . . . . . . . . . . . . . . . . 289

20. Genetic approaches to autonomic dysreflexia

A. Brown and J.E. Jacob (London, ON, Canada) . . . . . . . . . . . . . . . . . . . . . . . . . . 299

Section IV. Bowel Dysfunction

21. Gastrointestinal symptoms related to autonomic dysfunction following spinal cord injury
E.A.L. Chung and A.V. Emmanuel (Harrow, UK) . . . . . . . . . . . . . . . . . . . . . . . . . 317

22. Colorectal motility and defecation after spinal cord injury in humans
A.C. Lynch and F.A. Frizelle (Christchurch, New Zealand) . . . . . . . . . . . . . . . . . . . 335

23. Mechanisms controlling normal defecation and the potential effects of spinal cord injury
A.F. Brading and T. Ramalingam (Oxford and Camberley, UK) . . . . . . . . . . . . . . . 345

24. Alterations in eliminative and sexual reflexes after spinal cord injury: defecatory function
and development of spasticity in pelvic floor musculature
Y.S. Nout, G.M. Leedy, M.S. Beattie and J.C. Bresnahan (Columbus, OH and
Laramie, WY, USA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

25. Upper and lower gastrointestinal motor and sensory dysfunction after human spinal cord
injury
P. Enck, I. Greving, S. Klosterhalfen and B. Wietek (Tubingen, Gelsenkirchen and
Dusseldorf, Germany) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373

Section V. Sexual Dysfunction

26. Problems of sexual function after spinal cord injury

S.L. Elliott (Vancouver, BC, Canada) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387

27. Ascending spinal pathways from sexual organs: effects of chronic spinal lesions
C.H. Hubscher (Louisville, KY, USA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

28. Descending pathways modulating the spinal circuitry for ejaculation: effects of chronic
spinal cord injury
R.D. Johnson (Gainesville, FL, USA). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415

29. Male fertility and sexual function after spinal cord injury
D.J. Brown, S.T. Hill and H.W.G. Baker (Heidelberg, Melbourne and Carlton,
Australia) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427

30. Female sexual function after spinal cord injury

M.L. Sipski and A. Arenas (Miami, FL, USA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449

 Introduction

This book was inspired by a gathering of basic and clinical scientists, and healthcare providers, for a
workshop on autonomic dysfunction after spinal cord injury held in Banff, Alberta, Canada in July, 2003.
The discussions in this meeting highlighted the contrast between the high priority assigned by people with
spinal cord injury to finding a cure for their autonomic dysfunctions, and the limited awareness of these
issues or attention given to them by the scientific and medical community, other than care providers who
interact with cord-injured people regularly. Research is needed to gain greater understanding of the
mechanisms of these problems and to develop treatments and prevention strategies for them. To provide a
foundation for such endeavours, this book contains a compilation of what is known about bladder,
cardiovascular, bowel and sexual dysfunction after spinal cord injury, as it relates to the changes within the
autonomic nervous system control of these functions.
The book is organized into sections that focus on each of the affected visceral functions: urinary bladder,
cardiovascular, gastrointestinal and sexual. The book begins with a description of the time course of
autonomic dysfunctions and their ramifications from the first hours after a spinal cord injury to the more
stable chronic states. The next section contains three chapters that address anatomical findings that may
provide some of the foundation for autonomic dysfunctions in many of the systems. The system-specific
chapters then follow in four sections. Each section begins with a chapter or two defining the clinical
problems experienced by people with cord injury. The following chapters present research, basic and
clinical, that address the autonomic dysfunctions.
We have noted themes that transcend the different sections and can pertain to bladder, bowel, cardio-
vascular and sexual functions. For example, sprouting of axons, including the central processes of sensory
neurons, within the injured spinal cord can be advantageous or detrimental, depending on the amount,
location and potential for new contacts of this sprouting. This may also pertain to changes in the
autonomic ganglia outside the central nervous system. Another theme is loss of coordination and balance
of parasympathetic, sympathetic and somatic systems in the absence of modulatory influences from
supraspinal neuronal systems. Bladder dyssynergia and autonomic dysreflexia, with its episodic hyperten-
sion, have much in common. Lack of coordinated control of pelvic neurons leads to failure of defaecation
and ejaculation. Spinal cord injury affects more than spinal neurons; it impacts on peripheral ganglia and
target tissues such as blood vessels and the wall of the urinary bladder. Growth factors that one would
think should be advantageous to repair of the injured spinal cord, may actually promote development of
circuits that impair rather than support recovered function. Many other themes also thread through this
book.
Finally, we would like to acknowledge several people who have helped us during the preparation of this
book. Ms. Bibi Pettypiece organized the meeting in Banff that started this project and was in commu-
nication with all of the contributing authors to coordinate the details that can so easily become a burden.
Her assistance was invaluable. Ms. Eilis Hamilton applied her considerable skill with graphic presentation
to many of the illustrations in this book, adding to their clarity. The authors who have benefited from her
assistance are very appreciative. Mr. Tom Merriweather and Ms. Maureen Twaig from Progress in Brain
Research, Elsevier, have been encouraging and helpful throughout our effort. Lastly, we realize that we
have not included all of the work that has been done on the subject of autonomic dysfunction after spinal

xvii
xviii

cord injury. Excellent work is being done in addition to that described in these chapters, and we apologize
to those who did not have the opportunity to contribute to this book.

Lynne Weaver
Canio Polosa
Editors
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

Overview

Autonomic dysfunction in spinal cord injury: clinical

presentation of symptoms and signs

Ann-Katrin Karlsson

Spinal Injuries Unit, Sahlgrenska University Hospital, Institute of Clinical Neuroscience, Sahlgrenska Academy, S 413 45
Goteborg, Sweden

Abstract: Spinal cord injury and especially cervical spinal cord injury implies serious disturbances in
autonomic nervous system function. The clinical effects of these disturbances are striking. In the acute
phase, the autonomic imbalance and its effect on cardiovascular, respiratory system and temperature
regulation may be life threatening. Serious complications such as over-hydration with the risk of pulmo-
nary edema or hyponatremia are seen. The cord-injured person suffers from autonomic nervous system
dysfunction also affecting bladder and bowel control, renal and sexual function. Paralytic ileus may cause
vomiting and aspiration, which in turn interferes with respiratory function in those with cervical spinal cord
injury. The cord-injured person is at risk to develop pressure sores from the moment of the accident. Two to
three months post-injury the cord-injured person with a lesion level above the fifth thoracic segment may
develop autonomic dysreflexia, characterised by sympathetically mediated vasoconstriction in muscular,
skin, renal and presumably gastrointestinal vascular beds induced by an afferent peripheral stimulation
below lesion level. The reaction might cause cerebrovascular complications and has effects on metabolism.
Some of the autonomic disturbances are transient and a new balance is reached months post-injury, while
others persist for life.

Spinal cord injury, in a moment, dramatically
changes the life of the affected person. The loss of
control of skeletal muscle, as well as of sensations
from below the injury, together with the impair-
ment of thermoregulation, urinary bladder and
bowel function produce a profound deterioration
in the quality of life for people after spinal cord
injury. Tetraplegic subjects rank improvement in
hand function as the most important factor to en-
hance quality of life (Anderson, 2004). However,
paraplegic subjects give normal sexual function the
highest priority and, when the first and second
choice was combined, recovery of normal bladder
and bowel function were given the highest priority
in both groups of subjects. This shows that people
Corresponding author. Tel.:+46 31 3421000;
Fax: +46 31 415835; E-mail: ann-katrin.karlsson@neuro.gu.se
with spinal cord injury consider the disturbances in
autonomic nervous system function even more
devastating than the loss of motor and sensory
function.
The moment of the accident
A cervical spinal cord injury may be life threaten-
ing. When the level of lesion is above the third
cervical segment (C3), the injured person needs
immediate assistance of respiration due to the loss
of the supraspinal excitatory drive of the phrenic
motor neurons located at C3–C4. Even when the
level of the lesion is below C3, the cord-injured
person may suffer from life-threatening conditions
due to autonomic nervous system dysfunction.
Cardiac arrest may be one of the causes of death in

DOI: 10.1016/S0079-6123(05)52034-X 1
2
the first few minutes following a cervical spinal
cord injury due to the disruption of central sym-
pathetic control and the concomitant unopposed
vagal outflow. The incidence of this complication
is difficult to estimate, although some data indicate
a decreasing incidence during the last decades. The
number of patients who reach hospital alive has
increased more than twofold from the 1940s to the
late 1970s. This probably reflects improved skill in
treatment of a potential spinal cord injury by the
first aid team, since unskilled handling of an un-
stable neck might result in an ascending neurolog-
ical level of the lesion, making it life threatening.
In the emergency room
When arriving at the hospital the person with a
cervical spinal cord injury presents with the fol-
lowing symptoms: flaccid paresis, exclusively dia-
phragmatic respiration, low blood pressure and
mostly bradycardia. It is well recognized that spi-
nal cord injury implies inability to empty the blad-
der voluntary and, accordingly, an indwelling
catheter is always placed in the urinary bladder
in the emergency room when a spinal cord injury is
suspected. At this stage, new risks appear. Because
the spinal cord injury is often part of a multitrau-
ma the low blood pressure after cervical spinal
cord injury may be misinterpreted as consequent
to extensive blood loss. A treatment with rapid
infusion of intravenous fluids might lead to pul-
monary edema. On the other hand, the cord-in-
jured person may be bleeding in the abdomen and
this may be difficult to diagnose because of the
pre-existing low blood pressure, absence of tonic
contraction of the abdominal muscles (guarding
reaction) and absence of pain. For this and other
reasons, the cord-injured person needs to be care-
fully investigated by computerized tomography
scanning and magnetic resonance imaging. An un-
stable fracture in the spinal column with the risk of
deterioration of the neurological outcome some-
times results in placing the patient on a hard table,
a so-called ‘‘spine board.’’ Then the patient can be
moved from the emergency room to X-ray or to
the intensive care unit without having to be moved
from bed to examination tables and back. However,
if the injured person lies more than 2 h in the same
position on a hard spine board, he/she is at risk of
developing pressure sores. This risk is attributed to
the loss of sensory inputs to the brain from below
the level of the lesion, but this may not be the
whole explanation since unconscious patients
without spinal cord injury do not seem to be at
as high risk of developing pressure sore as are the
cord injury patients, conscious or unconscious.
Regulation of blood flow in the skin exposed to
pressure seems to be deranged in the decentralized
areas of the body, even though subcutaneous
adipose tissue blood flow during resting conditions
shows no difference when compared to that in
able-bodied people (Karlsson et al., 1997a).
Blood pressure control and the syndrome of
inappropriate anti-diuretic hormone secretion
After the extent of the injury is visualized radio-
graphically, the injured patient is transported ei-
ther to an operating theatre, an intensive care unit
or to a spinal cord injury unit. Careful monitoring
of blood pressure, heart rate, respiratory rate and
body temperature begins. A mean arterial blood
pressure above 80 mmHg is recommended (Hadley,
2002), and this level is sometimes maintained
by intravenous fluid supply and/or by the use of
pressor agents. Urine output in the cord-injured
patient usually is low during the first days post-
injury, probably due to an inappropriate secretion
of anti-diuretic hormone. Three to six days post-
injury urine output reaches 5–6 l/day; the accu-
mulated water is excreted. If urine output is strictly
monitored daily, this polyuria may be misinter-
preted as a sign of inability to concentrate the
urine, and if the loss of water is fully substituted by
intravenous fluids, or vasopressin is given, hypo-
natremia may develop. Some years ago a 28-year-
old man sustained a C7 spinal cord injury resulting
in tetraplegia and developed a serious hypo-
natremia with concomitant loss of vision (Karlsson
and Krassioukov, 2004). His cervical fracture was
stabilized surgically and during the day of surgery
he received 8.6 l of fluid intravenously. When his
urine output some days later increased to 6 l, it was
misinterpreted as inability to concentrate the urine

and vasopressin was given. Serum sodium concen-
tration decreased to 121 mmol/l (at the lowest lev-
el), and he was treated by restriction of fluid intake
and by mineral corticoids. His sodium level nor-
malized slowly. As soon as his water intake ex-
ceeded 0.7 l/day his serum level of sodium
decreased. As a result of these problems, he had
a partial loss of vision that was permanent. Urine
output must be calculated for a longer period than
the previous day to avoid this risk of overcom-
pensation.
Bradycardia
Another effect of the loss of supraspinal control of
the sympathetic nervous system is the bradycardia
that is seen sometimes during the first 2–3 weeks
post-injury. An example of this is a man, who at
the age of 39 years sustained a spinal cord injury at
the C5 level, an ASIA C type injury. He was
treated in the intensive care unit for 2 days and his
condition was uncomplicated except for brad-
ycardia. At the spinal cord injury unit, he showed
signs of a decreased arterial oxygen tension a tra-
cheal suction induced a cardiac arrest. He was
transferred to the intensive care unit, where he
stayed for 12 days. He had a prolonged period of
bradycardia with a mean heart rate of 48 bpm that
lasted for 2–3 weeks. The cardiologist was con-
sulted and he prescribed a 24 h electrocardiogram
recording. The recording showed sinus brad-
ycardia with a mean rate of 48 bpm and the in-
stallation of a pacemaker was suggested. However,
a week later the bradycardia resolved spontane-
ously and during the next 2 weeks the heart rate
increased to a mean of 57 bpm. This imbalance in
the acute phase seems to be replaced by a new
balance later since, when measuring heart rate
variability in the chronic phase, no difference is
found when comparing cord-injured subjects to
able-bodied subjects (Gao et al., 2002).
Respiratory system
In the first weeks post-injury there is a risk of res-
piratory failure in the cervical spinal cord injury
patient. This is sometimes due to obvious reasons
3
such as simultaneous injuries in the chest. How-
ever, the 40% loss of vital capacity, due to the
paralysis of the inter-costal muscles, is also of im-
portance. The loss of sympathetically mediated
bronchial dilatation may add further to the risk of
developing respiratory failure. Inhalation of
bronchodilators is usually used in the acute phase
following spinal cord injury. In spite of this treat-
ment, excessive mucus production and stagnation
of secretion is seen. The autonomic nervous system
imbalance might be life threatening during this
condition, since a person with cervical injury who
has a tendency to hypoxia might sustain a severe
bradycardia or heart arrest during tracheal suc-
tioning. Irritation of the trachea is a heavy stim-
ulus of vagal outflow even in able-bodied people
and the reaction in cord-injured patients might be
an exaggeration of this reaction due to the loss of
supraspinal control of the sympathetic nervous
system. Pretreatment with anti-cholinergic drugs is
sometimes needed before tracheal suctioning.
Temperature regulation
The respiratory problems might lead to pneumo-
nia with high fever. Then the cord-injured patient
is faced with another effect of autonomic distur-
bances: that is the inability to lose excess heat by
sweating. This inability might be life threatening
during high fever or in an extremely warm climate.
Several years ago a young man was treated at Sa-
hlgrenska University Hospital. He had sustained a
C4 spinal cord injury and had an aspiration of
fluid into his lungs during the transport to hospi-
tal. He developed pneumonia and his temperature
increased from 41.0 to 42.41 C and, at this high
temperature, he died of a cardiac arrest.
Even though central temperature control is un-
affected by cervical spinal cord injury, we some-
times see a prolonged period of increased
temperature in newly injured patient. Careful ex-
amination reveals no signs of infection or inflam-
mation and 3–6 weeks post-injury, the temperature
normalizes. Some people with cervical spinal cord
injury complain of feeling very cold. This is very
marked after a shower, when the patient some-
times needs a heating quilt or a heater to feel
4
comfortable. The ability to increase temperature
by shivering is lost below the lesion level and may
explain some of the sensation. However, some
patients suffer from coldness all the time.
Blood pressure and mobilization
When the fracture is stabilized and the neurological
level of lesion is stable, the person with spinal cord
injury needs to be mobilized. The low blood pres-
sure and the inability to increase blood pressure by
vasoconstriction below lesion level make mobiliza-
tion of the person with cervical injury difficult.
It has to be done gradually by tilting the patient
101/day while blood pressure and neurological sta-
tus are continuously monitored. When the patient
tolerates a 401 tilt, they are usually able to sit in a
wheelchair. Age at injury seems to affect the ability
to mobilize the patients, since the elderly usually
need more time to become mobilized. Whether this
is due to a lower tolerance to low blood pressure or
to greater decreases in blood pressure during mo-
bilization is not known. The renin–angiotensin sys-
tem plays a role in blood pressure control in
cervical spinal cord injury (Johnson et al., 1971;
Sutters et al., 1992). This was clearly demonstrated
when we treated a man in his 40 s with a cervical
injury who had suffered from renal failure prior to
his injury. His mobilization was prolonged and he
suffered from symptoms of low blood pressure for
several months after his injury.
Skin and sensation
Very few patients are able to lie or sit more than 4 h
in the same position without getting redness in the
skin of areas used for body support; this is the first
sign of pressure sores. The patient needs to be
turned every fourth hour even during a skull trac-
tion period. Four to five caregivers are needed to do
a safe log-roll of the patient. When the patient is
mobilized to a wheel chair sitting position, selection
of cushions is of great importance in order to avoid
pressure sores. The risk of developing pressure sores
persists in the cord-injured person and is increased
during severe infections with increased body tem-
perature. Under these circumstances the patient
needs to be turned every second hour in order to
avoid skin problems. What makes the skin more
vulnerable to pressure during fever is not known.
Another problem with the skin, seen during the
first months post-injury, is acne vulgaris that
sometimes flares up in the person with cervical in-
jury. This condition is not life threatening and oc-
curs above as well as below lesion level. Whether
this is due solely to hormonal disturbances elicited
by a stress reaction or to a combination with au-
tonomic dysfunction is not known.
Urinary system, bladder control
The autonomic nervous system dysfunction in-
volves the urinary system during the initial post-
injury stage of ‘‘spinal shock’’ and for the lifetime
of the person. The dysfunction entails loss of con-
trol of the urinary outlet and, during spinal shock,
loss of sensation from the bladder wall making the
patient at risk of over-distension of the bladder.
During spinal shock the bladder is atonic irrespec-
tive of level of lesion. When the stage of spinal
shock is past, which may take 3–4 months, people
with cervical or thoracic lesions develop a spinal
reflex bladder that expels urine under high intra-
vesical pressure at a certain amount of bladder
filling, a condition categorized as upper motor
neuron lesion. People with lower lumbar and/or
sacral levels of lesion retain an atonic bladder, a
lower motor neuron lesion. Regulation of bladder
emptying appears rather robust and might be nor-
malized even if the person suffers from some de-
gree of paresis and loss of sensation. In a
retrospective chart review of 249 patients with up-
per motor neuron lesions who had been treated at
the Spinal Cord Injury Unit in Göteborg,
we found that almost 30% of the individuals
recovered normal micturition and most of them
had injuries classified as ASIA C and D (Karin
Pettersson, personal communication).
Urinary system, renal function
Renal dysfunction has previously been the major
cause of death following spinal cord injury. The
mortality rate due to renal failure has decreased

from about 40% in the late 1940s (Whiteneck
et al., 1992) to 3–5% during the last decade (Webb
et al., 1984). This dramatic change is probably due
to improvements in bladder emptying regime as
well as to the introduction of antibiotics to treat
urinary tract infections. Even better, it seems that
renal function has a capacity to improve during
the first years post-injury. In a retrospective chart
review, we found that glomerular filtration rate
was low in those with cervical spinal cord injury in
the first months after injury. However, at a follow-
up 2–3 years later we found that the glomerular
filtration rate had increased, at least in the group
who emptied the bladder by clean intermittent
catheterization (Karin Pettersson, personal com-
munication).
Gastrointestinal system
The gastrointestinal system is also affected by the
spinal cord injury. Newly injured patients are at
increased risk of developing stress related gastric
ulcers and are regularly offered anti-acid treat-
ment. It might be that the unopposed vagal out-
flow plays a role and increases the risk of ulcer
formation. Other problems from the intestinal sys-
tem are obvious to the patient soon after injury.
The bowel is silent and the voluntary control of
bowel emptying is lost. This paralytic ileus ceases
within 1–2 weeks, but if liquids or solid food is
given prior to this there is a clear risk of a pro-
longed period of paralytic ileus, with the concom-
itant risk of nausea and vomiting. A patient who is
placed in skull traction is hard to manage properly
during vomiting and there is a great risk of aspi-
ration. Furthermore, a paralytic ileus could give
rise to a meteoristic abdomen, which might inter-
fere with respiration by interference with the
breathing movement of the diaphragm.
Programs for bowel emptying must be intro-
duced and we choose a rather conservative way of
treatment initially: no ingestion of food or drink
until the bowel has been emptied. In people with
lesions at the cervical and thoracic level this takes
about 3–6 days, but in the low lumbar lesion level
the emptying might be further delayed by several
days. By this regime, the risk of vomiting and
5
aspiration and the influence on respiration are
controlled. The pathogenesis of this temporary
paralytic ileus is unclear and the time frame for
return of bowel activity does not correspond to
return of reflex activity in the bladder or return of
tonus and reflexes in the skeleto-muscular system.
Even though a program for bowel emptying is in-
troduced, the evacuation of stool may be protract-
ed. This might be due to the new balance of
parasympathetic control, with an intact innerva-
tion of the ascending colon via the vagus nerve and
the loss of supraspinal control of the sacral para-
sympathetic supply to the colon.
When the bowel program has started the cord-
injured person may face new problems. There is a
risk of developing anal incontinence, and the risk
is highest with lesions in the lower lumbar level
due to a low tonus in the external anal sphincter.
This is perhaps one of the most devastating effects
of the injury to the spinal cord and may be one
important reason why cord-injured people do not
return to work after their injury.
The higher tonus and the uninhibited activity in
the anal sphincter after cervical and thoracic le-
sions might give rise to severe pain that seems to
originate from the anal sphincter. The pain is
made worse by anal fissures and hemorrhoids, and
this pain, as well as anal incontinence and consti-
pation, might later lead to colostomy. However,
the pain problem is not always resolved even
though the rectum and anal region are bypassed.
Sexual function
Some cord-injured men already in the intensive
care unit ask about their ability to have an active
sex life and become fathers. The erectile dysfunc-
tion in men after spinal cord injury has different
characteristics depending on level of lesion, and
mainly follows the pattern of bladder dysfunction.
The person with an upper motor neuron lesion
usually has the capacity for reflex erection by tac-
tile stimulation. The person with a lower motor
neuron lesion has loss of all erectile function. The
capacity of psychogenic erection is lost in all cord-
injured men with a complete lesion. Retrograde
ejaculation is the rule when there is an ejaculation
6
at all. Today, we can offer drugs and different
stimulations in order to improve erectile function
and to produce an anterograde ejaculation. Vibra-
tion and electro-stimulation need careful monitor-
ing of blood pressure since these methods readily
evoke autonomic dysreflexia. By the means of vi-
bration and electro-stimulation in combination
with insemination or in vitro fertilization, men af-
ter spinal cord injury can become fathers. At our
Spinal Cord Injury Unit, with 35–50 newly injured
patients admitted every year, we have about 75
children who have cord-injured fathers (Agneta
Siösteen, personal communication).
Sexual function is impaired also in women. The
female analogue to erectile dysfunction, that is,
loss of lubrication, needs treatment. Fertility is
unaffected, but the autonomic disturbances make
the woman with spinal cord injury at increased
risk of urinary incontinence, urinary tract infection
and pressure sores during child bearing. They are
also at risk of developing severe autonomic
dysreflexia during labor.
Autonomic dysreflexia
Two to four months post-injury the person with
cervical spinal cord injury may suddenly experi-
ence a flushing in the face and complain of severe
headache. The blood pressure increases from 100/60
to 240/120 mmHg. When looking for triggering
factors, the staff might find an obstruction of the
urinary outlet. When urine passes again the blood
pressure returns to normal. This reaction, the so-
called autonomic dysreflexia reaction, is seen in
cord-injured people with a lesion level above T6.
The clinical reaction is not an all-or-none reaction
but graded; in mild cases the person just feels a
small chill. Investigations have shown that the re-
action is caused by a severe vasoconstriction below
lesion level in skin, muscular (Karlsson et al.,
1998) and renal vascular beds (Gao et al., 2002).
Presumably also the splanchnic/gastrointestinal
vascular bed is involved. The reaction is mediat-
ed by the sympathetic nervous system as shown by
a profound increase in noradrenalin spillover be-
low lesion level (Karlsson et al., 1998; Gao et al.,
2002). The peripheral afferent stimulation of the
sympathetic neurons may be asymptomatic and
much more frequent than previously known. Con-
tinuous measurement of noradrenalin in blood
samples collected every 30 min in 24 h revealed
several peaks even in subjects who were asympto-
matic (Karlsson et al., 1997b).
The reaction is triggered by distension of hollow
organs below lesion level such as the urinary blad-
der, the gall bladder, the renal pelvis and ureters or
the gastrointestinal system. A bone fracture below
lesion level may also give rise to the reaction.
When the triggering factor is withdrawn, the blood
pressure returns to normal. However, if the cord-
injured person or the treating staff is unfamiliar
with the reaction, there is a serious risk of com-
plications as intracranial hemorrhage or cerebral
infarction. In our clinical practice there have been
two cases in the last 2 years. The first was a woman
in her 50s who sustained a C4–C5 level spinal cord
injury and tetraplegia a few years ago. She was
discharged to home and some months later be-
came severely constipated and developed an auto-
nomic dysreflexia reaction. She was treated at a
local hospital where there was inadequate knowl-
edge of autonomic dysreflexia. The blood pressure
stayed very high and the patient developed a cer-
ebral infarction. Another case was seen recently. A
person with cervical spinal cord injury who had
been injured for more than 30 years was treated
for a serious infection. He developed an abscess in
the abdomen and suffered from a prolonged ep-
isode of dysreflexia. During this period we meas-
ured his cerebral blood flow and found signs of
decreased perfusion in parts of the brain, indicat-
ing vasoconstriction. We speculated that this vaso-
constriction might have been an autoregulatory
response of the cerebral vessels in the presence of
greatly increased systemic arterial pressure.
The autonomic dysreflexia reaction has other
side effects and can influence metabolism. Region-
al investigation of adipose tissue metabolism
above and below lesion level showed an increase
in glycerol release — activation of lipolysis —
during induced dysreflexia, that is, during sympa-
thetic activation below lesion level (Karlsson et al.,
1998). Whether this is of importance for the insulin
resistance sometimes seen after spinal cord injury
(Karlsson, 1999) is not known.

Another associated risk with the autonomic
dysreflexia reaction is the vasoconstriction in the
renal vascular bed (Gao et al., 2002). This sympa-
thetic activation below lesion level might contrib-
ute to the development of renal failure. Previously,
many tetraplegic patients used bladder tapping to
condom drainage as a method of bladder empty-
ing. Bladder tapping is known to induce auto-
nomic dysreflexia every time it is performed. The
introduction of intermittent catheterization may
thus have the advantage of maintaining renal
function both by the intermittent total emptying of
the bladder, with markedly lower risk of urinary
tract infections, and by the absence of triggering
factors for renal vasoconstriction.
Visceral sensation
The loss of sensation includes not only skin and
joints but also loss of visceral sensation. This is of
importance during pathological processes in the
internal organs, and also seems to include loss of
normal sensation of hunger and satiety, a condi-
tion that might contribute to disturbances in body
weight sometimes seen in tetraplegic people. A
weight gain implies the risk of diabetes, impaired
glucose control and increased levels of serum lip-
ids. A higher risk in cord-injured than in able-
bodied subjects has been reported (Duckworth
et al., 1983; Bauman and Spungen, 2001), although
a study that controlled for inheritance by compar-
ing cord-injured subjects to their siblings showed
no difference in glucose tolerance evaluated by the
hyper-insulinemic, normo-glycemic clamp method
(Karlsson, 1999).
Thrombo-emboli
During the first months post-injury the cord-in-
jured person is at increased risk of developing deep
vein thrombosis. Treatment with anti-coagulants
is given for 3–6 months. The role of the autonomic
nervous system in this increased risk is unclear.
Later in life, cord-injured people appear not to
have increased risk of deep vein thrombosis.
7
Long-term effects
Gradually after spinal cord injury some risks seem
to diminish, others persist and new ones develop.
By 15–20 years post-injury new complications
from the circulatory system are sometimes seen.
Following large meals the cord-injured person with
a cervical lesion is sometimes unable to sit upright
due to rapid fall in blood pressure. Whether this is
merely an effect of duration of injury or a marker
of concomitant disease is not known. The muscle
and skin vascular beds are unable to vasoconstrict
during the post-prandial increase in blood flow in
the intestinal vascular bed, but this limitation is
present since the onset of injury and therefore,
does not solely explain the new complication.
People with cervical spinal cord injury sometimes
experience low urine output during the daytime
when they are in sitting position. During the night
there is a huge urine flow, making the patient at risk
of incontinence or over-distension of the bladder.
Time course of autonomic nervous system changes
— transitional stage?
Bradycardia lasts for few weeks. Paralytic ileus
lasts for 1–2 weeks. Signs of the inappropriate an-
ti-diuretic hormone secretion syndrome appear
during the first week. The risk of respiratory fail-
ure is great during the first weeks after injury. The
bladder is flaccid for 3–4 months. Body tempera-
ture is increased for 3–4 weeks. The risk of deep
vein thrombosis is increased for weeks or months.
There are many signs of a transitional stage in
autonomic nervous system dysfunction following
spinal cord injury. However, as described above,
the time frame is highly divergent. Some of the
changes are attributed to the spinal shock, or more
precisely to the recovery from spinal shock. How-
ever, our knowledge of the pathology behind spi-
nal shock is sparse, and it seems that the return of
reflex activity in the spinal cord follows a different
time course. Muscular tone and tendon reflexes
appear within 6–8 weeks, whereas reflexes for
bladder emptying may return much later, up to
3–4 months post-injury. The imbalance in heart
rate regulation has a time course of its own. We
8
have some knowledge from animal and human
research regarding the plasticity of spinal neural
circuits after injury (Krassioukov and Weaver,
1995; Krassioukov et al., 1999) but the clinical
implications of these findings are unclear.
After spinal cord injury, new balances are cre-
ated in the autonomic nervous system and we need
to create more knowledge about how this is done.
We also have to develop more understanding of
the overall disturbances in the autonomic nervous
system of cord-injured people since these distur-
bances have a profound impact on their life.
References
Anderson, K.D. (2004) Targeting recovery: Priorities of the spi-
nal cord-injured population. J. Neurotrauma, 21: 1371–1383.
Bauman, W.A. and Spungen, A.M. (2001) Carbohydrate and
lipid metabolism in chronic spinal cord injury. J. Spinal Cord
Med., 24: 266–277.
Duckworth, W.C., Jallepalli, P. and Solomon, S.S. (1983) Glu-
cose intolerance in spinal cord injury. Arch. Phys. Med. Re-
habil., 64: 107–110.
Gao, S.A., Ambring, A., Lambert, G. and Karlsson, A.K.
(2002) Autonomic control of the heart and renal vascular bed
during autonomic dysreflexia in high spinal cord injury. Clin.
Auton. Res., 12: 457–464.
Hadley, M.N., Walters, B.C., Grabb, P.A., Oyesiku, N.M.,
Przbylski, G.J., Resnick, D.K. and Ryken, T.C. (2002) Man-
agement of acute spinal cord injuries in an intensive care unit
or other monitored settings. Neurosurgery, 50: S51–S57.
Johnson, R.H., Park, D.M. and Frankel, H.L. (1971) Ortho-
static hypotension and the renin-angiotensin system in par-
aplegia. Paraplegia, 9: 146–152.
Karlsson, A.K. (1999) Insulin resistance and sympathetic func-
tion in high spinal cord injury. Spinal Cord, 37: 494–500.
Krassioukov, A.V., Bunge, R.P., Pucket, W.R. and Bygrave,
M.A. (1999) The changes in human spinal sympathetic pre-
ganglionic neurons after spinal cord injury. Spinal Cord, 37:
6–13.
Karlsson, A.K., Elam, M., Friberg, P., Biering-Sorensen, F.,
Sullivan, L. and Lonnroth, P. (1997a) Regulation of lipolysis
by the sympathetic nervous system: a microdialysis study in
normal and spinal cord-injured subjects. Metabolism, 46:
388–394.
Karlsson, A.K., Elam, M., Friberg, P., Sullivan, L., Attvall, S.
and Lonnroth, P. (1997b) Peripheral afferent stimulation of
decentralized sympathetic neurons activates lipolysis in spinal
cord-injured subjects. Metabolism, 46: 1465–1469.
Karlsson, A.K., Friberg, P., Lonnroth, P., Sullivan, L. and
Elam, M. (1998) Regional sympathetic function in high spi-
nal cord injury during mental stress and autonomic dysre-
flexia. Brain, 121: 1711–1719.
Karlsson, A.K. and Krassioukov, A.V. (2004) Hyponatremia-
induced transient visual disturbances in acute spinal cord in-
jury. Spinal Cord, 42(3): 204–207.
Krassioukov, A.V. and Weaver, L.C. (1995) Reflex and mor-
phological changes in spinal preganglionic neurons after cord
injury in rats. Clin. Exp. Hypertens., 17: 361–373.
Sutters, M., Wakefield, C., O’Neil, K., Appleyard, M., Frankel,
H., Mathias, C.J. and Peart, W.S. (1992) The cardiovascular,
endocrine and renal response of tetraplegic and paraplegic
subjects to dietary sodium restriction. J. Physiol., 457:
515–523.
Webb, D.R., Fitzpatrick, J.M. and O’Flynn, J.D. (1984) A 15-
year follow-up of 406 consecutive spinal cord injuries. Br. J.
Urol., 56: 614–617.
Whiteneck, G.G., Charlifue, S.W., Frankel, H.L., Fraser, M.H.,
Gardner, B.P., Gerhart, K.A., Krishnan, K.R., Menter, R.R.,
Nuseibeh, I. and Short, D.J. (1992) Mortality, morbidity, and
psychosocial outcomes of persons spinal cord injured more
than 20 years ago. Paraplegia, 30: 617–630.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 1

Effects of spinal cord injury on synaptic inputs to

sympathetic preganglionic neurons

Ida J. Llewellyn-Smith1,, Lynne C. Weaver2 and Janet R. Keast3

1
Cardiovascular Medicine and Centre for Neuroscience, Flinders University, Bedford Park, SA 5042, Australia
2
Spinal Cord Injury Laboratory, BioTherapeutics Research Group, Robarts Research Institute, London, ON, Canada
3
Pain Management Research Institute, University of Sydney at Royal North Shore Hospital, St Leonards,
NSW, Australia

Abstract: Spinal cord injuries often lead to disorders in the control of autonomic function, including
problems with blood pressure regulation, voiding, defecation and reproduction. The root cause of all these
problems is the destruction of brain pathways that control spinal autonomic neurons lying caudal to the
lesion. Changes induced by spinal cord injuries have been most extensively studied in sympathetic pre-
ganglionic neurons, cholinergic autonomic neurons with cell bodies in the lateral horn of thoracic and
upper lumbar spinal cord that are the sources of sympathetic outflow. After an injury, sympathetic pre-
ganglionic neurons in mid-thoracic cord show plastic changes in their morphology. There is also extensive
loss of synaptic input from the brain, leaving these neurons profoundly denervated in the acute phase of
injury. Our recent studies on sympathetic preganglionic neurons in lower thoracic and upper lumbar cord
that regulate the pelvic viscera suggest that these neurons are not so severely affected by spinal cord injury.
Spinal interneurons appear to contribute most of the synaptic input to these neurons so that injury does not
result in extensive denervation. Since intraspinal circuitry remains intact after injury, drug treatments
targeting these neurons should help to normalize sympathetically mediated pelvic visceral reflexes. Fur-
thermore, sympathetic pelvic visceral control may be more easily restored after an injury because it is less
dependent on the re-establishment of direct synaptic input from regrowing brain axons.

Every year, tens of thousands of people worldwide
suffer a spinal cord injury, often with devastating
consequences. Injured people can lose mobility
because they become partially or totally paralyzed
and experience persistent pain and spasticity. As
this volume documents, their quality of life can
also be significantly damaged by injury-induced
disorders in the control of autonomic function,
including problems with blood pressure regulation,
voiding, defecation and reproduction. The presence
Corresponding author. Tel.: +61-8-8204-4456;
Fax: +61-8-8204-5268;
E-mail: Ida.Llewellyn-Smith@flinders.edu.au
and severity of these problems is dependent
on the level and completeness of the injury. Two
cardiovascular consequences of spinal cord injury
are resting or postural hypotension and autonomic
dysreflexia, a condition in which strokes or death
can occur when noxious or innocuous sensory
stimuli entering the cord below the injury reflexly
induce episodes of hypertension. Spinal cord
injury can also produce a variety of difficulties
that impair the voiding of urine, including
detrusor hyperreflexia, detrusor sphincter dyssy-
nergia and detrusor areflexia. Fecal incontinence
and constipation are also outcomes of spinal cord
injury; and inability to achieve psychogenic and

DOI: 10.1016/S0079-6123(05)52001-6 11
12
reflexogenic erections and ejaculatory dysfunction
in men and failure of vaginal lubrication in women
have a significant impact on sexual life after injury.
The root cause of all these autonomic sequelae
of spinal cord injury is the disruption of the brain
pathways that control spinal autonomic neurons.
Subsequently, when axons conveying supraspinal
drive are dying or dead, neuronal circuits caudal to
the injury are able to reorganize. Primary afferent
neurons sprout (Christensen and Hulsebosch,
1997; Krenz and Weaver, 1998b; Krenz et al.,
1999; Wong et al., 2000; Weaver et al., 2001; On-
darza et al., 2003) and spinal autonomic interneu-
rons can become key regulators of some of the
information that is conveyed from the spinal cord
to the periphery (see Schramm, this volume).
Changes also occur in the synaptic circuitry con-
trolling sympathetic preganglionic neurons that
have axons providing central drive to all sympa-
thetic postganglionic neuron in paravertebral and
prevertebral ganglia and to chromaffin cells in the
adrenal medulla.
Location and morphology of sympathetic
preganglionic neurons
Sympathetic preganglionic neurons are small- to
medium-sized cholinergic neurons and their cell
bodies are located in the thoracic and upper lumbar
spinal cord in four distinct subnuclei within the
lateral horn (Cabot, 1990). The majority of sym-
pathetic preganglionic somata occur in the inter-
mediolateral cell column, which lies at the border
between the grey and white matter of the spinal
cord. In the intermediolateral cell column, the cell
bodies of sympathetic preganglionic neurons are
spindle-shaped or fusiform and occur in groups or
‘‘nests’’ that are spaced at short intervals along the
grey–white boundary. Most of the dendrites of the
neurons in the intermediolateral cell column run
rostrally or caudally and can be hundreds of
micrometers long. In addition, some sympathetic
preganglionic neurons in the intermediolateral
cell column have dendrites that are oriented med-
iolaterally, traveling either into the dorsolateral
funiculus or toward the central canal. Other sym-
pathetic preganglionic somata are situated in the
dorsolateral funiculus in the white matter; sympa-
thetic preganglionic neurons in the dorsolateral
funiculus are more common in the rostral than the
caudal thoracic cord (Strack et al., 1988). A small
proportion of sympathetic preganglionic neurons
are associated with the bundles of dendrites that
course mediolaterally between the intermediolateral
cell column and central canal. These sympathetic
preganglionic neurons have spindle-shaped cell
bodies and comprise the intercalated nucleus. The
somata and dendrites of the sympathetic pregangl-
ionic neurons in the intercalated nucleus are ori-
ented parallel to the dendritic bundles with which
they are associated. A final concentration of sym-
pathetic preganglionic cell bodies, which are usually
fusiform in shape, occurs in the central autonomic
area dorsal to the central canal. Somata occupying
this position are most frequently encountered at the
caudal end of the sympathetic preganglionic neuron
distribution, i.e., in the lowest thoracic and upper
lumbar segments.
Morphological changes after spinal cord injury
Spinal cord injury can evoke significant changes in
sympathetic preganglionic neurons, as well as de-
priving them of input from the brain. While neu-
rons at the site of a cord injury can be killed,
sympathetic preganglionic neurons that lie distant
from the lesion site may also be affected, at least in
the acute phase of injury. Within 3 days of a com-
plete spinal cord transection at thoracic segment 4/5,
the dendrites of sympathoadrenal preganglionic
neurons in the intermediolateral cell column of
mid-thoracic cord have retracted to about one-
third of their original length and the diameter of
their cell bodies has decreased to about 60% of
that in intact cord (Llewellyn-Smith and Weaver,
2001). This reduction in soma size and dendrite
length is less pronounced 7 days after transection;
and, by 14 days post-operatively, the sympathetic
preganglionic neurons and their dendrites are not
significantly different in size from those in intact
cord (Krassioukov and Weaver, 1996; Krenz and
Weaver, 1998a). The shrinkage and regrowth of
sympathetic preganglionic somata and dendrites
correlate with the degeneration and clearance from

the intermediolateral cell column of axons de-
tached from their cell bodies by the transection.
Many degenerating profiles of severed axons can
be seen ultrastructurally at 3 days after injury, but
virtually none are present at 14 days (Weaver et al.,
1997). Although new synapses form on sympatho-
adrenal preganglionic neurons after a spinal cord
injury (Weaver et al., 1997), reinnervation by ax-
ons of spinal neurons is unlikely to be the spur for
regrowth of dendrites. Even at 14 days after a
complete transection, sympathetic preganglionic
neurons in the mid-thoracic cord are profoundly
denervated. The density of synapses on their cell
bodies has been cut to half of that in intact cord
and their axodendritic input is reduced by 70%
(Llewellyn-Smith and Weaver, 2001). Whether or
not sympathetic preganglionic neurons continue to
be reinnervated after 2 weeks of injury has not
been studied ultrastructurally. However, anastomo-
sing networks of axons immunoreactive for growth-
associated protein 43, a marker for developing and
regenerating axons, are present at least as long as 6
weeks after spinal cord injury (Cassam et al., 1999).
Continuing reorganization of synaptic circuitry
controlling the activity of sympathetic preganglion-
ic neurons may underlie the increasing severity of
attacks of autonomic dysreflexia (Maiorov et al.,
1997a, b; Marsh and Weaver, 2004).
Innervation of sympathetic preganglionic neurons in
intact and injured cord
In intact cord, sympathetic preganglionic neurons
are innervated by both supraspinal and intraspinal
neurons. Virus tracing studies have been particu-
larly useful for revealing the locations of presym-
pathetic neurons, i.e., those that are directly
antecedent to sympathetic preganglionic neurons
and are likely to be involved in regulating their
activity. Supraspinal inputs to sympathetic pre-
ganglionic neurons come from five main brain re-
gions, including the rostral ventrolateral medulla,
the rostral ventromedial medulla, the caudal raphe
nuclei, the A5 region and the paraventricular nu-
cleus of the hypothalamus (Strack et al., 1989;
Sved et al., 2001). Spinal neurons that project to
sympathetic preganglionic neurons occur within
13
the intermediolateral cell column, around it in
spinal cord laminae V and VII and in lamina X
dorsal to the central canal (Cabot et al., 1994; Joshi
et al., 1995; Clarke et al., 1998; Cano et al., 2001;
Deuchars et al., 2001; Tang et al., 2004). The axons
of presympathetic supraspinal and intraspinal neu-
rons contain a diverse array of neurotransmitters,
many of which have been shown to have direct
effects on sympathetic preganglionic neurons.
Neurons in the brain are probably the exclusive
source of monoamine innervation, whereas both
supraspinal and intraspinal neurons contribute ax-
ons containing amino acids and neuropeptides.
Amino acids
Glutamate, g-aminobutyric acid (GABA) and
glycine, all produce fast synaptic responses in sym-
pathetic preganglionic neurons (e.g., Mo and Dun,
1987a, b; Inokuchi et al., 1992a, b; Krupp and Feltz,
1995; Krupp et al., 1997) and are considered the
main fast-acting transmitters regulating their activ-
ity (Dampney, 1994). Axons immunoreactive for
these amino acids synapse on sympathetic pregangl-
ionic neurons (Bacon and Smith, 1988; Bogan et al.,
1989; Cabot et al., 1992) and quantitative ultra-
structural studies have demonstrated that synaptic
vesicles containing at least one type of amino acid
are present in virtually all of the axons that provide
input to these neurons (Llewellyn-Smith et al., 1992,
1995b, 1998). Brainstem neurons innervate sympa-
thetic preganglionic neurons monosynaptically
(Zagon and Smith, 1993; Deuchars et al., 1995,
1997); and these spinally projecting neurons contain
markers for glutamate and GABA axons, including
phosphate activated glutaminase, vesicular gluta-
mate transporter 2 and glutamic acid decarboxylase
(Minson et al., 1991; Stornetta et al., 2002, 2004).
Furthermore, immunoreactivity for glutamate or
GABA occurs in boutons in the intermediolateral
cell column that have been anterogradely labeled
from the medulla (Llewellyn-Smith et al., 1995b).
Although neurons in the brain provide many of the
glutamate- and GABA-immunoreactive axons in
the intermediolateral cell column, spinal cord injury
does not deprive sympathetic preganglionic neurons
of amino acid-containing inputs. Ultrastructural
14
studies at times when severed supraspinal axons are
degenerating or have just been removed from below
a lesion show that glutamate- and GABA-immuno-
reactive synaptic contacts persist on sympathetic
preganglionic neurons caudal to either 3- or 7-day
complete spinal cord transections (Llewellyn-Smith
et al., 1997; Llewellyn-Smith and Weaver, 2001).
Hence, intraspinal neurons as well as supraspinal
neurons provide amino acid-containing inputs to
sympathetic preganglionic neurons.
Monoamines
Adrenaline and noradrenaline evoke both
excitatory and inhibitory responses in sympathet-
ic preganglionic neurons (Coote et al., 1981;
Kadzielawa, 1983; Ma and Dun, 1985a; Miyazaki
et al., 1989; Lewis and Coote, 1990a). At the light
microscope level, sympathetic preganglionic neu-
rons at all levels of intact cord are surrounded by
networks of nerve fibers immunoreactive for en-
zymes of catecholamine synthesis, such as tyrosine
hydroxylase (Fig. 1A) and phenylethanolamine
N-methyltransferase (Fig. 1B). In intact cord, phe-
nylethanolamine N-methyltransferase-immunore-
active axons have been confirmed to synapse on
sympathetic preganglionic neurons at the electron
microscope level (Milner et al., 1988; Bernstein-
Goral and Bohn, 1989) and we have found
that sympathetic preganglionic neurons receive
synapses from axons with immunoreactivity for
tyrosine hydroxylase (Fig. 2A). All of this cat-
echolamine input to sympathetic preganglionic
neurons probably originates in the brain (see be-
low). Major sources are the C1 adrenergic neurons
of the rostral ventrolateral medulla and the nor-
adrenergic neurons of the A5 group (Jansen et al.,
1995). The supraspinal origin of catecholamine in-
put is supported by the disappearance of immuno-
reactivity for catecholamine-synthesizing enzymes
caudal to a complete spinal cord transection.
Staining for tyrosine hydroxylase and phenyl-
ethanolamine N-methyltransferase is substantially
reduced at 2 weeks post-operatively (Figs. 1C and
D) and is absent by 11 weeks (Figs. 1E and F).
Interestingly, after spinal cord injury, some neurons
in the intermediate grey of the spinal cord become
immunoreactive for dopamine b-hydroxylase
(Cassam et al., 1997), the enzyme that produces
noradrenaline from dopamine; and we have de-
scribed axons caudal to a 2-week transection that
contain tyrosine hydroxylase and form synapses
in the intermediolateral cell column (Fig. 2B;
Llewellyn-Smith et al., 1995a). Hence, the cat-
echolamine enzyme-immunoreactive fibers present
at 2 weeks may arise from these neurons. Some
immunoreactivity may also be present in the non-
terminal portions of severed axons since, at 2
weeks, degenerating terminals cannot be found in
the intermediolateral cell column at the ultrastruc-
tural level (Weaver et al., 1997). Determining
whether the spinal neurons that express catechola-
mine enzymes at 2 weeks after transection synthe-
size dopamine, adrenaline or noradrenaline will
require different experimental strategies, such as
fluorescence histochemistry, or multiple-label
immunofluorescence for investigating coexistence
of relevant synthetic enzymes or amine transport-
ers. A more detailed anatomical analysis of
11-week transected cord will also be needed to as-
certain whether there are any enzyme-immunore-
active fibers present at the chronic stage of injury.
Pharmacological and physiological studies indi-
cate that, in general, serotonin (5-hydroxytrypta-
mine (5-HT)) has a sympathoexcitatory action on
sympathetic nerve activity and directly on sympa-
thetic preganglionic neurons (e.g., Ma and Dun,
1986; Yusof and Coote, 1988; Lewis and Coote,
1990b; Pickering et al., 1994). Serotonergic axons,
marked by either immunoreactivity for 5-HT or
the serotonin transporter, also form a dense plexus
of axons around sympathetic preganglionic neu-
rons (Figs. 3A and C) and synapses by 5-HT-
immunoreactive axons have been demonstrated on
sympathetic preganglionic neurons that project to
the superior cervical ganglion and adrenal medulla
(Bacon and Smith, 1988; Vera et al., 1990; Jensen
et al., 1995). Retrograde and viral tracing studies
indicate that the serotonergic axons surrounding
sympathetic preganglionic neurons arise from rap-
he neurons, mainly those in the medullary nuclei
(Loewy and McKellar, 1981; Bowker et al., 1982;
Jansen et al., 1995). Very rare 5-HT-immunoreac-
tive neurons have been detected in the spinal cord
(Newton et al., 1986). However, these are unlikely
 15

Fig. 1. Axons containing immunoreactivity for the catecholamine synthesizing enzymes, tyrosine hydroxylase (TH) and phenyl-
ethanolamine N-methyltransferase (PNMT), disappear from the intermediolateral cell column caudal to a complete spinal cord
transection (TX). (A, B) Intact cord. (C, D) 2-week transected cord. (E, F) 11-week transected cord. (A, C, E) Stained for tyrosine
hydroxylase and (B, D, F) stained for phenylethanolamine N-methyltransferase (PNMT). Transections were located in caudal thoracic
segment 4/rostral thoracic segment 5. All micrographs show the intermediolateral cell column in thoracic segment 6. Bars, 100 mm.

to nnervate sympathetic preganglionic neurons be- Neuropeptides

cause 5-HT-immunoreactive and serotonin trans-
porter-immunoreactive axons are absent from
autonomic areas caudal to a 2-week transection
(Figs. 3B and D). Hence, severed serotonergic ax-
ons disappear from the cord before axons that are
immunoreactive for catecholamine enzymes.
Axons containing a substantial array of neuro-
peptides have been demonstrated by light micros-
copy in regions of the cord where the cell bodies
and dendrites of sympathetic preganglionic neu-
rons are located (Table 1) and a number of these
16

Fig. 2. Axons immunoreactive for tyrosine hydroxylase form synapses in intact and transected cord. (A) In the intermediolateral cell
column of intact cord, a tyrosine hydroxylase-immunoreactive varicosity (TH) forms a synapse (arrowhead) on a dendrite that
contains cholera toxin B subunit (CTB) retrogradely transported from the adrenal medulla. An adjacent non-immunoreactive var-
icosity (asterisk) synapses (arrowheads) on the same dendrite. (B) In the intermediolateral cell column of thoracic segment 8, a tyrosine
hydroxylase-immunoreactive varicosity (TH) forms a synapse (arrowheads) on a dendrite 2 weeks after a complete spinal cord
transection (TX) in caudal thoracic segment 4/rostral thoracic segment 5. An adjacent non-immunoreactive varicosity (asterisk)
synapses (arrowhead) on the same dendrite. Bars, 500 nm.

Fig. 3. Axons containing the serotonergic markers, 5-HT or the serotonin transporter, disappear from the intermediolateral cell
column caudal to a complete spinal cord transection (TX). (A, C) Intact cord. (B, D) 2-week transected cord. (A, B) Stained for 5-HT
and (C, D) stained for the serotonin transporter (SERT). Transections were located in caudal thoracic segment 4/rostral thoracic
segment 5. All micrographs show the intermediolateral cell column in thoracic segment 6. Bar in (D) applies to (A—D), 100 mm.

Table 1. Neuropeptide immunoreactivity identified in axons in autonomic regions of the thoracic and upper lumbar spinal cord
Angiotensin II
Avian pancreatic polypeptide (APP)
Calcitonin gene-related peptide (CGRP)
Cholecystokinin (CCK)
Cocaine and amphetamine regulated transcript (CART)a
Corticotropin releasing factor (CRF)
Enkephalina
Galanin
Neuropeptide Y (NPY)a
Neurophysin
a
Axons containing these neuropeptides have been shown to form synapses either in the intermediolateral cell column or on identified
sympathetic preganglionic neurons.
have been shown to synapse on sympathetic pre-
ganglionic neurons (Bacon and Smith, 1988; Vera
et al., 1990; Llewellyn-Smith et al., 1991; Pilowsky
et al., 1992). When applied to sympathetic pre-
ganglionic neurons, many of these neuropeptides
evoke synaptic responses (e.g., Ma and Dun,
1985b; Dun and Mo, 1988; Kolaj et al., 1997;
Lai et al., 1997; Antunes et al., 2001; van den Top
et al., 2003). Some of the neuropeptide-immuno-
reactive axons supplying sympathetic preganglion-
ic neurons arise exclusively from neurons in the
brain, including those containing oxytocin, vaso-
pressin and orexin. However, other neuropeptides,
such as substance P, enkephalin and neuropeptide
Y, occur in autonomic areas of intact and
transected cord (Figs. 4–6), suggesting that both
supraspinal and intraspinal neurons supply the
lateral horn. Substance P is co-localized with se-
rotonin in brainstem neurons that innervate the
spinal cord (Sasek et al., 1990). A large subset
of spinally projecting cardiovascular neurons in
the medulla contain preproenkephalin mRNA
(Stornetta et al., 1999) and immunoreactivity for
neuropeptide Y and mRNA for preproneuropep-
tide Y have been identified in medullospinal neu-
rons (Minson et al., 1994; Stornetta et al., 1999).
Nevertheless, complete spinal cord transection
does not destroy all varicose axons in the inter-
mediolateral cell column immunoreactive for sub-
stance P, enkephalin or neuropeptide Y (Davis
and Cabot, 1984; Romagnano et al., 1987; Cassam
et al., 1997) and we have shown that, caudal to a 7-
day complete transection, axons containing each
of these neuropeptides form synapses on choline
17
Neurotensin
Nociceptin
Orexina
Oxytocin
Pituitary adenylate cyclase activating polypeptide (PACAP)a
Somatostatin
Substance Pa
Thyrotropin releasing hormone
Vasoactive intestinal peptide (VIP)
Vasopressin
acetyltransferase-immunoreactive (i.e., choliner-
gic) neurons in the intermediolateral cell column
(Llewellyn-Smith and Weaver, 2004). Hence, some
of the axons containing these neuropeptides come
from spinal interneurons to innervate the choli-
nergic sympathetic preganglionic neurons. The
source of the neuropeptide Y-immunoreactive ax-
ons that persist in the lateral horn may be the
neurons in laminae V and VII in intact cord that
express neuropeptide Y mRNA (Minson et al.,
2001). The intraspinal enkephalin innervation of
autonomic regions may arise from small enkepha-
lin-immunoreactive neurons in lamina X that
we have detected in sections from intact cord
fixed with high concentrations of glutaraldehyde
(Llewellyn-Smith and Keast, unpublished obser-
vations), as neurons in this location are known to
communicate with sympathetic preganglionic neu-
rons (Cano et al., 2001; Tang et al., 2004). The cell
bodies of origin of the intraspinal substance P in-
put have yet to be defined.
Rostrocaudal differences in sympathetic
preganglionic neurons and their innervation
The sympathetic nervous system was original-
ly thought to act in an undifferentiated way to
allow an animal to respond appropriately to life-
threatening situations. However, over the past two
decades, it has become increasingly clear that cen-
tral control of sympathetic outflow is differential,
permitting specific functional groups of sympathetic
preganglionic neurons to respond in different ways
18

Fig. 4. Caudal to a complete spinal cord transection (TX), axons containing the neuropeptides, substance P, enkephalin and ne-
uropeptide Y, are still present in the intermediolateral cell column of mid-thoracic spinal cord segments. (A, C, E) Intact cord. (B, D,
F) 2-week transected cord. (A, B) Stained for substance P (SP), (C, D) stained for enkephalin (ENK) and (E, F) stained for
neuropeptide Y (NPY). Transections were located in caudal thoracic segment 4/rostral thoracic segment 5. Micrographs show the
intermediolateral cell column from thoracic segments 7, 8 or 9. Bars, 100 mm.

to the same homeostatic challenge (reviewed by
Morrison, 2001). Differences in the spatial ar-
rangement of sympathetic preganglionic neurons
and in their innervation are likely to be the ana-
tomical basis for these differentiated physiological
responses. The differences in the innervation of
sympathetic preganglionic neurons suggest that
the outcomes of spinal cord injury will vary de-
pending on the functional group of sympathetic
preganglionic neurons that are deprived of their
supraspinal input.
Sympathetic preganglionic neurons are topo-
graphically organized along the rostrocaudal axis
of the spinal cord (Strack et al., 1988). Preganglionic
 19

Fig. 5. Caudal to a complete spinal cord transection (TX), axons containing the neuropeptides, substance P, enkephalin and ne-
uropeptide Y, are still present in the intermediolateral cell column of lower thoracic and upper lumbar segments. (A, C, E) Intact cord.
(B, D, F) 2-week transected cord. (A, B) Stained for substance P (SP), (C, D) stained for enkephalin (ENK) and (E, F) stained for
neuropeptide Y (NPY). Transections were located in caudal thoracic segment 4/rostral thoracic segment 5. Micrographs show the
intermediolateral cell column from thoracic segment 13 or lumbar segment 1. Bars, 100 mm.

neurons in the rostral thoracic cord supply rostral
sympathetic ganglia (e.g., the superior cervical
ganglion and the stellate ganglion) and participate
in the regulation of targets in the upper body, like
pupils, salivary glands and the heart. The mid-
thoracic cord contains sympathetic preganglionic
neurons that project to the celiac ganglion as part
of the circuitry controlling mesenteric vasculature,
gut motility and gut secretion as well as sympa-
thetic preganglionic neurons projecting to the ad-
renal medulla to regulate release of noradrenaline
and adrenaline from chromaffin cells. The caudal
end of the range includes sympathetic pregangl-
ionic neurons involved in regulating the activity of
organs in the lower body, like the urinary bladder,
lower bowel and reproductive organs. Despite this
general topographical organization, the distribu-
tions of sympathetic preganglionic neurons pro-
jecting to different target ganglia or adrenal
chromaffin cells overlap so that sympathetic pre-
ganglionic neurons of different functions are in-
termixed within each spinal segment. For example,
20

Fig. 6. Caudal to a complete spinal cord transection (TX), axons containing the neuropeptides, substance P, enkephalin and ne-
uropeptide Y, are still present in the central autonomic area of upper lumbar segments. (A, C, E) Intact cord. (B, D, F) 2-week
transected cord. (A, B) Stained for substance P (SP), (C, D) stained for enkephalin (ENK) and (F) stained for neuropeptide Y (NPY).
Transections were located in caudal thoracic segment 4/rostral thoracic segment 5. All micrographs show the central autonomic area in
lumbar segment 2. cc, central canal. Bars, 100 mm.

in the rat, thoracic segment 6 contains a mixture of
sympathetic preganglionic neurons that send ax-
ons to the superior cervical ganglion, stellate gan-
glion, celiac ganglion or adrenal medulla.
Although not very well studied, rostrocaudal
differences in the innervation of sympathetic pre-
ganglionic neurons parallel the target-based
rostrocaudal arrangement of their cell bodies. In
autonomic areas of cat thoracolumbar cord, nerve
fibers immunoreactive for 5-HT, substance P,
somatostatin, oxytocin, neurotensin or neurophy-
sin show a non-uniform rostrocaudal distribution
(Krukoff et al., 1985) as do 5-HT-immunoreactive
axons in rabbit intermediolateral cell column

(Jensen et al., 1995). In rats, oxytocin-immunore-
active axons closely appose sympathetic pregangl-
ionic neurons retrogradely labeled from the
cervical sympathetic trunk; but sympathoadrenal
preganglionic neurons are not innervated by ox-
ytocin fibers (Holets and Elde, 1982; Appel and
Elde, 1988). Inputs to choline acetyltransferase-
immunoreactive sympathetic preganglionic neu-
rons that express Fos in response to hypotension
also show a heterogenous pattern of innervation
by some types of axons (Minson et al., 2002).
Although apposing almost all Fos-positive sym-
pathetic preganglionic neurons in upper and
middle thoracic cord, neuropeptide Y or phenyl-
ethanolamine N-methyltransferase-immunoreac-
tive axons avoid significant proportions of these
neurons in lower thoracic segments. More than
half of the hypotension-sensitive sympathetic
preganglionic neurons in the middle and lower
thoracic cord lacked appositions from galanin-
immunoreactive axons, whereas some choline ace-
tyltransferase-positive, Fos-negative neurons in
the lumbar cord lay in dense baskets of galanin-
positive fibers.
Recently, we have examined enkephalin-
immunoreactive inputs to sympathetic pregangl-
ionic neurons retrogradely labeled with cholera
toxin B subunit from the major pelvic ganglion
(Llewellyn-Smith et al., 2005), which contains
sympathetic and parasympathetic preganglionic
neurons innervating the urinary bladder, lower
bowel and reproductive organs. This work has re-
vealed a striking difference between the reaction to
spinal cord injury of these neurons, which have
somata in thoracic segment 12 to lumbar segment
2, and more rostral sympathetic preganglionic
neurons. In contrast to choline acetyltransferase-
immunoreactive sympathetic preganglionic neu-
rons in thoracic segment 8 (Llewellyn-Smith and
Weaver, 2001), sympathetic preganglionic neurons
that are in circuits controlling pelvic viscera appear
to retain most of their innervation after a complete
spinal cord transection. In intact cord, we found
that sympathetic preganglionic neurons projecting
to the major pelvic ganglion from the intermedio-
lateral cell column, the intercalated nucleus and
central autonomic area were surrounded by very
dense baskets of enkephalin-immunoreactive
21
axons at the light microscopic level (Figs. 5C,
and 6C). Similarly, dense baskets were present
around retrogradely labeled neurons in cords at
2 and 11 weeks after transection (e.g., Figs. 7 and 8).
These observations imply that most of the
enkephalin input to pelvic visceral sympathetic
preganglionic neurons comes from spinal neurons
below the transection. Furthermore, the density of
the enkephalin innervation of neurons projecting
to the major pelvic ganglion suggests that inter-
neurons are likely to provide the predominant in-
put to sympathetic preganglionic neurons that
control the pelvic viscera.
At the electron microscope level, we observed
that the density of synapses on sympathetic pre-
ganglionic neurons projecting to the major pelvic
ganglion did not appear to differ in intact and
transected cord, although this observation was not
quantified. This conclusion was supported by
quantification of enkephalin-immunoreactive in-
put to these neurons. In intact cord, sympathetic
preganglionic neurons that projected to the major
pelvic ganglion received many synapses from
enkephalin-immunoreactive axon terminals. In
the intermediolateral cell column, 52% of the
synaptic input to retrogradely labeled cell bodies
was enkephalin-immunoreactive. Furthermore,
this enkephalin innervation was targeted to cell
bodies in preference to dendrites since only 29% of
the input to retrogradely labeled dendrites was
enkephalin positive. In the 2-week transected cord,
enkephalin occurred in 65% of the varicosities that
synapsed on sympathetic preganglionic somata
that projected to the major pelvic ganglion from
the intermediolateral cell column. The proportion-
al change in input between cell bodies in intact and
transected cord was not statistically significant.
However, the increase in enkephalin input from
52% to 65% suggests a small loss of synapses due
to transection. This loss might have been revealed
if data had been collected from a larger number of
rats. These data indicate that the enkephalin input
to pelvic visceral sympathetic preganglionic neu-
rons is not significantly affected by transection,
due to the fact that it is predominantly intraspinal.
Since pelvic visceral sympathetic preganglionic
neurons are not substantially denervated after spi-
nal cord injury, their somata and dendrites may
22

Fig. 7. In transected cord, enkephalin-immunoreactive axons closely appose sympathetic preganglionic neurons projecting to the
major pelvic ganglion. Transections (TX) were located in caudal thoracic segment 4/rostral thoracic segment 5. (A) A sympathetic
preganglionic neuron (asterisk) that has retrogradely transported cholera toxin B subunit (CTB) from the major pelvic ganglion
(MPG) lies at the lateral edge of the intermediolateral cell column (IML) in 11-week transected cord. A host of enkephalin (ENK)-
immunoreactive varicosities form close appositions on the sympathetic preganglionic neuron. A retrogradely labeled dendrite in the
white matter (WM) also receives many close appositions from enkephalin-containing terminals. Some appositions are indicated by
arrowheads. Bar, 20 mm. (B) Retrogradely labeled sympathetic preganglionic neurons (asterisks) in the central autonomic area lie
within a very dense network of enkephalin (ENK)-immunoreactive axons. Bar, 50 mm.

not undergo the shrinkage and regrowth that we
have previously documented in more rostral sym-
pathetic preganglionic neurons immediately after
injury. However, further studies are needed to ex-
plore this possibility.
The dominance of intraspinal pathways in the
control of sympathetic preganglionic neurons sup-
plying the major pelvic ganglion has important im-
plications for the restoration of pelvic visceral
function after spinal cord injury. Since intraspinal
circuits controlling pelvic visceral sympathetic pre-
ganglionic neurons are relatively unaffected by spi-
nal cord injury, drug treatments that target this
persistent circuitry should help to normalize sym-
pathetically mediated pelvic visceral reflexes. Fur-
thermore, since sympathetic components are less
affected, they should be more easily restored after
an injury because there will be less dependence on
the re-establishment of direct synaptic input from
regrowing supraspinal axons. It will be interesting to
see whether spinal interneurons are equally impor-
tant in the regulation of parasympathetic pregangl-
ionic neurons, which project to the major pelvic
ganglion from the lower lumbar and upper sacral
cord and are also critical for pelvic visceral function.
Acknowledgments
Project Grants (#229907 to ILS and #000044 to
JRK) and Research Fellowships (#229921 to ILS
and #358709 to JRK) from the National Health
and Medical Research Council of Australia, grants

Fig. 8. In transected cord, enkephalin-immunoreactive axons synapse on sympathetic preganglionic neurons projecting to the major
pelvic ganglion. In the central autonomic area of 2-week transected cord, an enkephalin-immunoreactive varicosity (ENK) forms a
synapse (arrowheads) on a dendrite that contains a crystal due to retrograde transport of cholera toxin B subunit (CTB) from the
major pelvic ganglion (MPG). The transection was located in caudal thoracic segment 4/rostral thoracic segment 5. Bar, 500 nm.
from the National Heart Foundation of Australia
(#G98A0097 and #G00A0512 to ILS), a Visiting
Scientist Award from the Heart and Stroke Foun-
dation of Canada (ILS), Ontario Heart and Stroke
Foundation (LCW) and the Canadian Institutes of
Health Research (LCW) supported this work.
Carolyn Martin, Natalie Fenwick and Lee Travis
provided expert technical assistance.
References
Antunes, V.R., Brailoiu, G.C., Kwok, E.H., Scruggs, P. and
Dun, N.J. (2001) Orexins/hypocretins excite rat sympathetic
preganglionic neurons in vivo and in vitro. Am. J. Physiol.
Regul. Integr. Comp. Physiol., 281: R1801–R1807.
Appel, N.M. and Elde, R.P. (1988) The intermediolateral cell
column of the thoracic spinal cord is comprised of target-
specific subnuclei: evidence from retrograde transport studies
and immunohistochemistry. J. Neurosci., 8: 1767–1775.
Bacon, S.J. and Smith, A.D. (1988) Preganglionic sympathetic
neurones innervating the rat adrenal medulla: immunocyto-
chemical evidence of synaptic input from nerve terminals
containing substance P, GABA or 5-hydroxytryptamine.
J. Auton. Nerv. Syst., 24: 97–122.
Bernstein-Goral, H. and Bohn, M.C. (1989) Phenylethanola-
mine N-methyltransferase-immunoreactive terminals synapse
on adrenal preganglionic neurons in the rat spinal cord. Ne-
uroscience, 32: 521–537.
Bogan, N., Mennone, A. and Cabot, J.B. (1989) Light micro-
scopic and ultrastructural localization of GABA-like
immunoreactive input to retrogradely labeled sympathetic
preganglionic neurons. Brain Res., 505: 257–270.
23
Bowker, R.M., Westlund, K.N., Sullivan, M.C. and Coulter,
J.D. (1982) Organization of descending serotonergic projec-
tions to the spinal cord. Prog. Brain Res., 57: 239–265.
Cabot, J.B. (1990) Sympathetic preganglionic neurons: Cyto-
architecture, ultrastructure and biophysical properties. In:
Loewy A.D. and Spyer K.M. (Eds.), Central Regulation of
Autonomic Functions. Oxford University Press, pp. 44–67.
Cabot, J.B., Alessi, V. and Bushnell, A. (1992) Glycine-like
immunoreactive input to sympathetic preganglionic neurons.
Brain Res., 571: 1–18.
Cabot, J.B., Alessi, V., Carroll, J. and Ligorio, M. (1994) Spinal
cord lamina V and lamina VII interneuronal projections to
sympathetic preganglionic neurons. J. Comp. Neurol., 347:
515–530.
Cano, G., Sved, A.F., Rinaman, L., Rabin, B.S. and Card, J.P.
(2001) Characterization of the central nervous system inner-
vation of the rat spleen using viral transneuronal tracing.
J. Comp. Neurol., 439: 1–18.
Cassam, A.K., Llewellyn-Smith, I.J. and Weaver, L.C. (1997)
Catecholamine enzymes and neuropeptides are expressed in
fibres and somata in the intermediate gray matter in chronic
spinal rats. Neuroscience, 78: 829–841.
Cassam, A.K., Rogers, K.A. and Weaver, L.C. (1999) Co-lo-
calization of substance P and dopamine beta-hydroxylase
with growth-associated protein-43 is lost caudal to a spinal
cord transection. Neuroscience, 88: 1275–1288.
Christensen, M.D. and Hulsebosch, C.E. (1997) Spinal cord
injury and anti-NGF treatment results in changes in CGRP
density and distribution in the dorsal horn in the rat. Exp.
Neurol., 147: 463–475.
Clarke, H.A., Dekaban, G.A. and Weaver, L.C. (1998) Iden-
tification of lamina V and VII interneurons presynaptic to
adrenal sympathetic preganglionic neurons in rats using a
recombinant herpes simplex virus type 1. Neuroscience, 85:
863–872.
24
Coote, J.H., Macleod, V.H., Fleetwood-Walker, S.M. and
Gilbey, M.P. (1981) The response of individual sympathetic
preganglionic neurons to microiontophoretically applied en-
dogenous monoamines. Brain Res., 215: 135–145.
Dampney, R.A.L. (1994) Functional organization of central
pathways regulating the cardiovascular system. Physiol.
Rev., 74: 323–364.
Davis, B.M. and Cabot, J.B. (1984) Substance P-containing
pathways to avian sympathetic preganglionic neurons: evi-
dence for major spinalspinal circuitry. J. Neurosci., 4:
2145–2159.
Deuchars, S.A., Brooke, R.E., Frater, B. and Deuchars, J.
(2001) Properties of interneurones in the intermediolateral
cell column of the rat spinal cord: role of the potassium
channel subunit Kv3.1. Neuroscience, 106: 433–446.
Deuchars, S.A., Morrison, S.F. and Gilbey, M.P. (1995)
Medullary-evoked EPSPs in neonatal rat sympathetic
preganglionic neurones in vitro. J. Physiol., 487(Part 2):
453–463.
Deuchars, S.A., Spyer, K.M. and Gilbey, M.P. (1997) Stimu-
lation within the rostral ventrolateral medulla can evoke
monosynaptic GABAergic IPSPs in sympathetic pregangl-
ionic neurons in vitro. J. Neurophysiol., 77: 229–235.
Dun, N.J. and Mo, N. (1988) In vitro effects of substance P on
neonatal rat sympathetic preganglionic neurones. J. Physiol.,
399: 321–333.
Holets, V. and Elde, R. (1982) The differential distribution and
relationship of serotonergic and peptidergic fibers to sym-
pathoadrenal neurons in the intermediolateral cell column of
the rat: a combined retrograde axonal transport and
immunofluorescence study. Neuroscience, 7: 1155–1174.
Inokuchi, H., Yoshimura, M., Trzebski, A., Polosa, C. and
Nishi, S. (1992a) Fast inhibitory postsynaptic potentials and
responses to inhibitory amino acids of sympathetic pregangl-
ionic neurons in the adult cat. J. Auton. Nerv. Syst., 41:
53–60.
Inokuchi, H., Yoshimura, M., Yamada, S., Polosa, C. and
Nishi, S. (1992b) Fast excitatory postsynaptic potentials and
the responses to excitant amino acids of sympathetic
preganglionic neurons in the slice of the cat spinal cord.
Neuroscience, 46: 657–667.
Jansen, A.S.P., Wessendorf, M.W. and Loewy, A.D. (1995)
Transneuronal labeling of CNS neuropeptide and monoa-
mine neurons after pseudorabies virus injections into the
stellate ganglion. Brain Res., 683: 1–24.
Jensen, I., Llewellyn-Smith, I.J., Minson, J.B., Pilowsky, P.M.
and Chalmers, J.P. (1995) Serotonin inputs to rabbit sympa-
thetic preganglionic neurons projecting to the superior cer-
vical ganglion or adrenal medulla. J. Comp. Neurol., 353:
427–438.
Joshi, S., LeVatte, M.A., Dekaban, G.A. and Weaver, L.C.
(1995) Identification of spinal interneurons antecedent to ad-
renal sympathetic preganglionic neurons using trans-synaptic
transport of herpes simplex virus type 1. Neuroscience, 65:
893–903.
Kadzielawa, K. (1983) Inhibition of the activity of sympathetic
preganglionic neurones and neurones activated by visceral
afferents, by alpha-methylnoradrenaline and endogenous
catecholamines. Neuropharmacology, 22: 3–17.
Kolaj, M., Shefchyk, S.J. and Renaud, L.P. (1997) Two con-
ductances mediate thyrotropin-releasing-hormone-induced
depolarization of neonatal rat spinal preganglionic and lat-
eral horn neurons. J. Neurophysiol., 78: 1726–1729.
Krassioukov, A.V. and Weaver, L.C. (1996) Morphological
changes in sympathetic preganglionic neurons after spinal
cord injury in rats. Neuroscience, 70: 211–225.
Krenz, N.R., Meakin, S.O., Krassioukov, A.V. and Weaver,
L.C. (1999) Neutralizing intraspinal nerve growth factor
blocks autonomic dysreflexia caused by spinal cord injury.
J. Neurosci., 19: 7405–7414.
Krenz, N.R. and Weaver, L.C. (1998a) Changes in the mor-
phology of sympathetic preganglionic neurons parallel the
development of autonomic dysreflexia after spinal cord injury
in rats. Neurosci. Lett., 243: 61–64.
Krenz, N.R. and Weaver, L.C. (1998b) Sprouting of primary
afferent fibers after spinal cord transection in the rat.
Neuroscience, 85: 443–458.
Krukoff, T.L., Ciriello, J. and Calaresu, F.R. (1985) Segmental
distribution of peptide- and 5HT-like immunoreactivity in
nerve terminals and fibres of the thoracolumbar sympathetic
nuclei of the cat. J. Comp. Neurol., 240: 103–116.
Krupp, J., Bordey, A. and Feltz, P. (1997) Electrophysiological
evidence for multiple glycinergic inputs to neonatal rat sym-
pathetic preganglionic neurons in vitro. Eur. J. Neurosci., 9:
1711–1719.
Krupp, J. and Feltz, P. (1995) Excitatory postsynaptic currents
and glutamate receptors in neonatal rat sympathetic pregangl-
ionic neurons in vitro. J. Neurophysiol., 72(4): 1503–1512.
Lai, C.C., Wu, S.Y., Lin, H.H. and Dun, N.J. (1997) Excitatory
action of pituitary adenylate cyclase activating polypeptide
on rat sympathetic preganglionic neurons in vivo and in
vitro. Brain Res., 748: 189–194.
Lewis, D.I. and Coote, J.H. (1990a) Excitation and inhibition
of rat sympathetic preganglionic neurons by catecholamines.
Brain Res., 530: 229–234.
Lewis, D.I. and Coote, J.H. (1990b) The influence of 5-hydro-
xytryptamine agonists and antagonists on identified sympa-
thetic preganglionic neurons in the rat, in vivo. Br. J.
Pharmacol., 99: 667–672.
Llewellyn-Smith, I.J., Arnolda, L.F., Pilowsky, P.M.,
Chalmers, J.P. and Minson, J.B. (1998) GABA- and
glutamate-immunoreactive synapses on sympathetic pre-
ganglionic neurons projecting to the superior cervical gan-
glion. J. Auton. Nerv. Syst., 71: 96–110.
Llewellyn-Smith, I.J., Cassam, A.K., Krenz, N.R.,
Krassioukov, A.V., Minson, J.B., Pilowsky, P.M., Arnolda,
L.F., Chalmers, J.P. and Weaver, L.C. (1995a) Do sympa-
thetic preganglionic neurons receive synapses from intraspi-
nal catecholamine neurons? Soc. Neurosci. Abstr., 21: 1401.
Llewellyn-Smith, I.J., Cassam, A.K., Krenz, N.R., Krassiou-
kov, A.V. and Weaver, L.C. (1997) Glutamate- and GABA-
immunoreactive synapses on sympathetic preganglionic neu-
rons caudal to a spinal cord transection in rats. Neuroscience,
80: 1225–1235.

Llewellyn-Smith, I.J., DiCarlo, S.E., Collins, H.L. and Keast,
J.R. (2005) Enkephalin-immunoreactive interneurons exten-
sively innervate sympathetic preganglionic neurons regulat-
ing the pelvic viscera. J. Comp. Neurol., 488: 278–289.
Llewellyn-Smith, I.J., Minson, J.B., Pilowsky, P.M., Arnolda,
L.F. and Chalmers, J.P. (1995b) The one hundred percent
hypothesis: glutamate or GABA in synapses on sympathetic
preganglionic neurons. Clin. Exp. Hypertens., 17: 323–334.
Llewellyn-Smith, I.J., Minson, J.B., Pilowsky, P.M. and
Chalmers, J.P. (1991) There are few catecholamine- or ne-
uropeptide Y-containing synapses in the intermediolateral
cell column of rat thoracic spinal cord. Clin. Exp. Pharmacol.
Physiol., 18: 111–115.
Llewellyn-Smith, I.J., Phend, K.D., Minson, J.B., Pilowsky,
P.M. and Chalmers, J.P. (1992) Glutamate immunoreactive
synapses on retrogradely labelled sympathetic neurons in rat
thoracic spinal cord. Brain Res., 581: 67–80.
Llewellyn-Smith, I.J. and Weaver, L.C. (2001) Changes in
synaptic inputs to sympathetic preganglionic neurons after
spinal cord injury. J. Comp. Neurol., 435: 226–240.
Llewellyn-Smith, I.J. and Weaver, L.C. (2004) Interneuronal
inputs to sympathetic preganglionic neurons: evidence from
transected spinal cord. In: Dun N.J., Machado B.H. and
Pilowsky P.M. (Eds.), Neural Mechanisms of Cardiovascular
Regulation. Kluwer, pp. 265–283.
Loewy, A.D. and McKellar, S. (1981) Serotonergic projections
from the ventral medulla to the intermediolateral cell column
in the rat. Brain Res., 211: 146–152.
Ma, R.C. and Dun, N.J. (1985a) Norepinephrine depolarizes
lateral horn cells of neonatal rat spinal cord in vitro. Neuro-
sci. Lett., 60: 163–168.
Ma, R.C. and Dun, N.J. (1985b) Vasopressin depolarizes lat-
eral horn cells of the neonatal rat spinal cord in vitro. Brain
Res., 348: 36–43.
Ma, R.C. and Dun, N.J. (1986) Excitation of lateral horn neu-
rons of the neonatal rat spinal cord by 5-hydroxytryptamine.
Dev. Brain Res., 24: 89–98.
Maiorov, D.N., Krenz, N.R., Krassioukov, A.V. and Weaver,
L.C. (1997a) Role of spinal NMDA and AMPA receptors in
episodic hypertension in conscious spinal rats. Am. J.
Physiol., 273: H1266–H1274.
Maiorov, D.N., Weaver, L.C. and Krassioukov, A.V. (1997b)
Relationship between sympathetic activity and arterial pres-
sure in conscious spinal rats. Am. J. Physiol., 272:
H625–H631.
Marsh, D.R. and Weaver, L.C. (2004) Autonomic dysreflexia,
induced by noxious or innocuous stimulation, does not de-
pend on changes in dorsal horn substance P. J. Neurotrauma,
21: 817–828.
Milner, T.A., Morrison, S.F., Abate, C. and Reis, D.J. (1988)
Phenylethanolamine N-methyltransferase-containing termi-
nals synapse directly on sympathetic preganglionic neurons
in the rat. Brain Res., 448: 205–222.
Minson, J.B., Arnolda, L.F. and Llewellyn-Smith, I.J. (2002)
Neurochemistry of nerve fibers apposing sympathetic pre-
ganglionic neurons activated by sustained hypotension.
J. Comp. Neurol., 449: 307–318.
25
Minson, J.B., Llewellyn-Smith, I.J. and Arnolda, L.F. (2001)
Neuropeptide Y mRNA expression in interneurons in rat
spinal cord. Auton. Neurosci., 93: 14–20.
Minson, J.B., Llewellyn-Smith, I.J., Pilowsky, P.M. and
Chalmers, J.P. (1994) Bulbospinal neuropeptide Y-immuno-
reactive neurons in the rat: comparison with adrenaline-syn-
thesising neurons. J. Auton. Nerv. Syst., 47: 233–243.
Minson, J., Pilowsky, P., Llewellyn-Smith, I., Kaneko, T.,
Kapoor, V. and Chalmers, J. (1991) Glutamate in spinally
projecting neurons of the rostral ventral medulla. Brain Res.,
555: 326–331.
Miyazaki, T., Coote, J.H. and Dun, N.J. (1989) Excitatory and
inhibitory effects of epinephrine on neonatal rat sympathetic
preganglionic neurons in vitro. Brain Res., 497: 108–116.
Mo, N. and Dun, N.J. (1987a) Excitatory postsynaptic poten-
tials in neonatal rat sympathetic preganglionic neurons: pos-
sible mediation by NMDA receptors. Neurosci. Lett., 77:
327–332.
Mo, N. and Dun, N.J. (1987b) Is glycine an inhibitory trans-
mitter in rat lateral horn cells? Brain Res., 400: 139–144.
Morrison, S.F. (2001) Differential control of sympathetic out-
flow. Am. J. Physiol. Regul. Integr. Comp. Physiol., 281:
R683–R698.
Newton, B.W., Maley, B.E. and Hamill, R.W. (1986)
Immunohistochemical demonstration of serotonin neurons
in autonomic regions of the rat spinal cord. Brain Res., 376:
155–163.
Ondarza, A.B., Ye, Z. and Hulsebosch, C.E. (2003) Direct ev-
idence of primary afferent sprouting in distant segments fol-
lowing spinal cord injury in the rat: colocalization of GAP-43
and CGRP. Exp. Neurol., 184: 373–380.
Pickering, A.E., Spanswick, D. and Logan, S.D. (1994)
5-Hydoxytryptamine evokes depolarizations and membrane
potential oscillations in rat sympathetic preganglionic neu-
rones. J. Physiol., 480: 109–121.
Pilowsky, P., Llewellyn-Smith, I.J., Lipski, J. and Chalmers, J.
(1992) Substance P immunoreactive boutons form synapses
with feline sympathetic preganglionic neurons. J. Comp. Ne-
urol., 320: 121–135.
Romagnano, M.A., Braiman, J., Loomis, M. and Hamill, R.W.
(1987) Enkephalin fibers in autonomic nuclear regions: in-
traspinal vs. supraspinal origin. J. Comp. Neurol., 266:
319–331.
Sasek, C.A., Wessendorf, M.W. and Helke, C.J. (1990) Evi-
dence for co-existence of thyrotropin-releasing hormone,
substance P and serotonin in ventral medullary neurons that
project to the intermediolateral cell column in the rat. Ne-
uroscience, 35: 105–119.
Stornetta, R.L., Akey, P.J. and Guyenet, P.G. (1999) Location
and electrophysiological characterization of rostral medul-
lary adrenergic neurons that contain neuropeptide Y mRNA
in rat medulla. J. Comp. Neurol., 415: 482–500.
Stornetta, R.L., McQuiston, T.J. and Guyenet, P.G. (2004)
GABAergic and glycinergic presympathetic neurons of rat
medulla oblongata identified by retrograde transport of
pseudorabies virus and in situ hybridization. J. Comp Ne-
urol., 479: 257–270.
26
Stornetta, R.L., Sevigny, C.P., Schreihofer, A.M., Rosin, D.L.
and Guyenet, P.G. (2002) Vesicular glutamate transporter
DNPI/VGLUT2 is expressed by both C1 adrenergic and
nonaminergic presympathetic vasomotor neurons of the rat
medulla. J. Comp. Neurol., 444: 207–220.
Strack, A.M., Sawyer, W.B., Hughes, J.H., Platt, K.B. and
Loewy, A.D. (1989) A general pattern of CNS innervation of
the sympathetic outflow demonstrated by transneuronal
pseudorabies viral infections. Brain Res., 491: 156–162.
Strack, A.M., Sawyer, W.B., Marubio, L.M. and Loewy, A.D.
(1988) Spinal origin of sympathetic preganglionic neurons in
the rat. Brain Res., 455: 187–191.
Sved, A.F., Cano, G. and Card, J.P. (2001) Neuroanatomical
specificity of the circuits controlling sympathetic outflow to
different targets. Clin. Exp. Pharmacol. Physiol., 28:
115–119.
Tang, X., Neckel, N.D. and Schramm, L.P. (2004) Spinal in-
terneurons infected by renal injection of pseudorabies virus in
the rat. Brain Res., 1004: 1–7.
van den Top, M., Nolan, M.F., Lee, K., Richardson, P.J.,
Buijs, R.M., Davies, C.H. and Spanswick, D. (2003) Orexins
induce increased excitability and synchronisation of rat sym-
pathetic preganglionic neurones. J. Physiol., 549: 809–821.
Vera, P.L., Holets, V.R. and Miller, K.E. (1990) Ultrastructural
evidence of synaptic contacts between substance P-, enke-
phalin-, and serotonin-immunoreactive terminals and retro-
gradely labeled sympathetic preganglionic neurons in the rat:
a study using a double-peroxidase procedure. Synapse, 6:
221–229.
Weaver, L.C., Cassam, A.K., Krassioukov, A.V. and Lle-
wellyn-Smith, I.J. (1997) Changes in immunoreactivity for
growth associated protein-43 suggest reorganization of
synapses on spinal sympathetic neurons after cord transec-
tion. Neuroscience, 81: 535–551.
Weaver, L.C., Verghese, P., Bruce, J.C., Fehlings, M.G., Krenz,
N.R. and Marsh, D.R. (2001) Autonomic dysreflexia and
primary afferent sprouting after clip-compression injury of
the rat spinal cord. J. Neurotrauma, 18: 1107–1119.
Wong, S.T., Atkinson, B.A. and Weaver, L.C. (2000) Confocal
microscopic analysis reveals sprouting of primary afferent
fibres in rat dorsal horn after spinal cord injury. Neurosci.
Lett., 296: 65–68.
Yusof, A.P.M. and Coote, J.H. (1988) Excitatory and inhibi-
tory actions of intrathecally administered 5-hydroxytrypta-
mine on sympathetic nerve activity in the rat. J. Auton. Nerv.
Syst., 22: 229–236.
Zagon, A. and Smith, A.D. (1993) Monosynaptic projections
from the rostral ventrolateral medulla oblongata to identified
sympathetic preganglionic neurons. Neuroscience, 54:
729–743.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 2

Spinal sympathetic interneurons: Their identification

and roles after spinal cord injury

Lawrence P. Schramm

Departments of Biomedical Engineering and Neuroscience, The Johns Hopkins University School of Medicine, 606 Traylor
Building, 720 Rutland Avenue, Baltimore, MD 21205, USA

Abstract: Primary afferent neurons rarely, if ever, synapse on the sympathetic preganglionic neurons that
regulate the cardiovascular system, nor do sympathetic preganglionic neurons normally exhibit sponta-
neous activity in the absence of excitatory inputs. Therefore, after serious spinal cord injury ‘‘spinal
sympathetic interneurons’’ provide the sole excitatory and inhibitory inputs to sympathetic preganglionic
neurons. Few studies have addressed the anatomy and physiology of spinal sympathetic interneurons, to a
great extent because they are difficult to identify. Therefore, this chapter begins with descriptions of both
neurophysiological and neuroanatomical criteria for identifying spinal sympathetic interneurons, and it
discusses the advantages and disadvantages of each. Spinal sympathetic interneurons also have been little
studied because their importance in intact animals has been unknown, whereas the roles of direct pro-
jections from the brain to sympathetic preganglionic neurons are better known. This chapter presents
evidence that spinal sympathetic interneurons play only a minor role in sympathetic regulation when the
spinal cord is intact. However, they play an important role after spinal cord injury, both in generating
ongoing activity in sympathetic nerves and in mediating segmental and intersegmental sympathetic reflexes.
The spinal sympathetic interneurons that most directly influence the activity of sympathetic preganglionic
neurons after spinal cord injury are located close to their associated sympathetic preganglionic neurons,
and the inputs from distant segments that mediate multisegmental reflexes are relayed to sympathetic
preganglionic neurons multisynaptically via spinal sympathetic interneurons. Finally, spinal sympathetic
interneurons are more likely to be excited and less likely to be inhibited by both noxious and innocuous
somatic stimuli after chronic spinal transection. The onset of this hyperexcitability corresponds to mor-
phological changes in both sympathetic preganglionic neurons and primary afferents, and it may reflect the
pathophysiological processes that lead to autonomic dysreflexia and the hypertensive crises that may occur
with it in people after chronic spinal injury.

Introduction maker potentials, and under ordinary

Sympathetic preganglionic neurons are the final
neurons within the central nervous system that
regulate sympathetic output to nearly every tissue
and organ. Like somatic motoneurons, sympa-
thetic preganglionic neurons do not exhibit pace-
Corresponding author. Tel.: +410-955-3026;
Fax: +410-955-9826; E-mail: LSCHRAMM@bme.jhu.edu
circumstances their activity is determined by
synaptic inputs from the brain and spinal cord
(see Laskey and Polosa, 1988, for review). The
regulation of the activity of sympathetic pregangl-
ionic neurons by brainstem systems has been ex-
tensively investigated (Laskey and Polosa, 1988;
Cabot, 1996; Blessing, 1997). However, less is
known about the regulation of sympathetic activ-
ity after spinal cord injury, when inputs from the

DOI: 10.1016/S0079-6123(05)52002-8 27
28
brainstem are lost. Sympathetic activity after
spinal cord injury is enigmatic because it ranges
from abnormally low, leading to bouts of hypo-
tension, to abnormally high, leading to hyperten-
sive crises (Mathias and Frankel, 1992). One
characteristic upon which there appears to be lit-
tle disagreement is that few, if any, spinal primary
afferents synapse directly upon sympathetic pre-
ganglionic neurons (Laskey and Polosa, 1988).
Therefore, by definition, after a complete spinal
cord transection spinal interneurons convey all
spinal inputs to sympathetic preganglionic neu-
rons, whether these inputs are derived from pri-
mary afferents or from intraspinal sources of
ongoing sympathetic activity.
For the purposes of this chapter, I define spinal
interneurons as all spinal neurons other than (1)
somatic motoneurons and (2) autonomic pregangl-
ionic neurons. I define spinal sympathetic inter-
neurons as spinal interneurons with connections
that can directly or indirectly affect sympathetic
activity. Some spinal neurons that play no role in
sources of sympathetic activity in intact spinal
cords may participate in those sources after spinal
cord injury. These neurons may, themselves, be
under tonic descending inhibition when the spinal
cord is intact, or their connections to sympathetic
preganglionic neurons may be via other interneu-
rons that are tonically inhibited. Such neurons
would be classified as spinal sympathetic interneu-
rons after, but not before, spinal cord injury.
That spinal sympathetic interneurons have been
little studied is understandable for two reasons.
First, as discussed below, they are unique neither
in their neurotransmitters nor their morphology.
Thus, they are not readily identified. Second, Miller
et al. (2001) found very few spinal interneurons
with activities correlated with ongoing renal sym-
pathetic nerve activity in rats with intact spinal
cords. This observation suggests that spinal sym-
pathetic interneurons play a minor role in the reg-
ulation of sympathetic activity in animals with
intact spinal cords. Therefore, these interneurons
have not attracted attention in studies of normal,
autonomic regulation of the circulation. In recent
years, however, spinal sympathetic interneurons
have attracted more attention because they may
play important roles in autonomic dysfunction
after spinal cord injury and because they may play
positive roles in the recovery of autonomic function
or pathological roles in mediating autonomic dys-
function during spinal cord repair and regeneration.
The anatomy and physiology of spinal sympa-
thetic systems have been comprehensively re-
viewed (Laskey and Polosa, 1988; Cabot, 1996;
Weaver and Polosa, 1997). Therefore, this chapter
will concentrate on recent studies of the anatomy
and physiology of spinal sympathetic interneurons
in rats with intact, acutely transected, and chron-
ically transected spinal cords. I begin by reviewing
the methods used to identify and characterize spi-
nal sympathetic interneurons.
Spinal sympathetic interneurons are identified both
physiologically and anatomically
Ideally, we could identify each spinal sympathetic
interneuron, whether characterized physiologically
or anatomically, by tracing its axon to synapses
upon sympathetic preganglionic neurons. However,
this is possible under only special conditions, usu-
ally in vitro (see, for example, Deuchars et al.,
2001). In all other cases, the ‘‘sympathetic’’ nature
of spinal sympathetic interneurons must either be
inferred from their neurophysiological properties
or by tracing their connections to sympathetic
preganglionic neurons using specialized, trans-
synaptic, retrograde labeling methods.
Gebber and colleagues pioneered neurophysio-
logical identification of spinal sympathetic inter-
neurons by identifying spinal neurons with
discharge patterns that were correlated with the
discharge patterns in either pre- or postganglionic
sympathetic axons (Gebber and McCall, 1976;
Barman and Gebber, 1984). These investigators
cross-correlated the ongoing activity of single spi-
nal interneurons and the ongoing activity in sym-
pathetic nerves. Neurons with activities either
positively or negatively correlated with sympathet-
ic nerve activity were defined as spinal sympathetic
interneurons. This remains the only neurophysio-
logical method for identifying spinal sympathetic
interneurons (Chau et al., 1997, 2000; Miller et al.,
2001; Tang et al., 2003).
Neurophysiological identification of spinal sym-
pathetic interneurons has two drawbacks. First,
 29

how does one distinguish between spinal sympa-

thetic interneurons and sympathetic preganglionic
neurons? The activities of both types of neurons
could be correlated with sympathetic nerve activ-
ity. Second, how does one distinguish between
spinal sympathetic interneurons and other inter-
neurons which share inputs with sympathetic pre-
ganglionic neurons but which are not involved in
sympathetic processing?
Distinguishing between spinal sympathetic in-
terneurons and sympathetic preganglionic neurons
is the easier of these problems. As shown by
Gebber and McCall (1976), sympathetic pregangl-
ionic neurons rarely discharge at rates exceeding
20 Hz. Therefore, the minimum interspike interval
for sympathetic preganglionic neurons is approx-
imately 50 ms. The discharge patterns of spinal
interneurons, on the other hand, usually include
bursts of action potentials with interspike intervals
of 20 ms or less. Therefore, with the relatively mi-
nor risk of misidentifying some spinal sympathetic
interneurons with low discharge rates, spinal sym-
pathetic interneurons can be distinguished from
sympathetic preganglionic neurons by the presence
of short interspike intervals in their discharge pat-
terns. An additional criterion for some spinal sym- Fig. 1. Neurophysiological identification of spinal sympathetic
pathetic interneurons is their dorsal location in the interneurons by cross-correlation. Upper panel: renal sympa-
spinal cord. Sympathetic preganglionic neurons thetic nerve activity (RSNA, upper trace) and simultaneously
are never located within spinal laminae I–V. recorded occurrences of action potentials in a spinal interneu-
ron (lower trace). Lower panel: cross-correlation between a
Therefore, sympathetically correlated neurons lo- 10 min recording of spinal neuronal action potentials and si-
cated in the dorsal laminae of the spinal cord are multaneously recorded RSNA. From Krassioukov et al. (2002),
very likely to be spinal sympathetic interneurons. with permission.
Figure 1 illustrates the neurophysiological iden-
tification of a putative spinal sympathetic inter- these activities was calculated (Fig. 1, lower panel,
neuron. The ongoing activity of a spinal neuron dark trace). Zero time on the correlogram was the
was recorded at a depth of 300 mm from the dorsal instant at which the interneuron began a burst of
surface of the 10th thoracic (T10) spinal segment of activity. The sharp positive peak in the correlo-
an anesthetized rat, acutely spinally transected at gram, approximately 75 ms after the onset of the
the 3rd cervical segment (C3). The neuron was burst, indicated that bursts of renal sympathetic
identified as an interneuron both by its dorsal po- nerve activity regularly lagged the onset of bursts
sition and by the presence of bursts of activity with of activity of the interneuron by 75 ms. To gauge
interspike intervals of 10 ms. At the temporal res- the significance of the correlation, the interdis-
olution of Fig. 1, these bursts are visible as darker charge intervals of the interneuron’s ongoing
action potential indicators (upper panel, lower activity were shuffled 10 times to generate 10 ‘‘dum-
trace). Ongoing renal sympathetic nerve activity my’’ cross-correlations with renal sympathetic nerve
(Fig. 1, upper panel, upper trace) was recorded activity (Fig. 1, lower panel, lighter traces). The
simultaneously with the ongoing activity of the positive correlation between the interneuron’s actu-
interneuron, and the cross-correlation between al ongoing activity and ongoing renal sympathetic
30
nerve activity was so much larger than the enve-
lope of the 10 dummy correlations that the prob-
ability that this relationship could have occurred
by chance was very small.
A more difficult problem than distinguishing
spinal sympathetic interneurons from sympathetic
preganglionic neurons is distinguishing spinal
sympathetic interneurons from interneurons that
are only coincidently correlated with sympathetic
nerve activity. An example of such a coincidence
would be the case in which both sympathetic pre-
ganglionic neurons and interneurons were driven
by a common synaptic input. Indeed, this distinc-
tion cannot be made unambiguously using neuro-
physiological techniques. Nevertheless, confidence
in identifying spinal sympathetic interneurons is
possible when (1) bursts of ongoing or evoked ac-
tivity of a sympathetically correlated interneuron
usually lead ongoing or evoked bursts of sympa-
thetic nerve activity by an interval consistent with
the calculated conduction time from the interneu-
ron to the recording site on the sympathetic nerve
and (2) evoked excitatory and inhibitory responses
of interneurons to applied stimuli are correlated
with responses in sympathetic nerve activity to the
same stimuli.
Figure 1 illustrates a case in which the first of
these criteria was met. The 75 ms lag between
bursts of ongoing activity of the interneuron and
bursts of ongoing sympathetic nerve activity rep-
resents an aggregate conduction velocity of ap-
proximately 0.5 m/s, which is consistent with the
expected conduction velocity of the largely un-
myelinated axons of the renal sympathetic nerve.
Subsequently, this neuron also met the second cri-
terion. Pinch of the left flank, within the region of
the T10 dermatome, excited both the activity of
this neuron and renal sympathetic nerve activity,
whereas pinch of the left hip and left shoulder
caused decreases in both this neuron’s activity and
renal sympathetic nerve activity (data not shown).
Chau et al. (1997) found that the polarities
(direction in which firing frequency changed, up
or down) of somatically evoked responses of the
majority of interneurons with ongoing activities
positively correlated with renal sympathetic nerve
activity matched the polarities of simultaneously
evoked responses in renal sympathetic nerve
activity. The polarities of somatically evoked re-
sponses of uncorrelated interneurons were much
less likely to match those of simultaneous respons-
es in renal sympathetic nerve activity. Further-
more, the excitatory fields of uncorrelated neurons
were significantly larger than those of correlated
neurons, and they were often larger than the ex-
citatory fields for renal sympathetic nerve activity.
Excitatory fields are defined as the area of body
surface from which stimulation of sensory recep-
tors evoked excitation of the neuron.
Once identified neurophysiologically, spinal
sympathetic interneurons can be anatomically lo-
cated and morphologically characterized either by
intracellular labeling (Deuchars et al., 2001) or by
the juxtacellular labeling method (Pinault, 1996;
Schreihofer and Guyenet, 1997; Tang et al., 2003).
The juxtacellular method involves approaching the
soma or proximal dendrites of a spinal sympa-
thetic interneuron very closely and passing positive
current pulses into it through a biocytin-filled elec-
trode. The current apparently electroporates (gen-
erates temporary pores in) the neuron’s membrane
and carries biocytin into the cell. Biocytin rapidly
diffuses throughout the neuron’s soma and dend-
rites. Labeled neurons are identified and recon-
structed histologically after treatment with a
streptavidin-conjugated chromogen. Although this
method has been used to visualize spinal sympa-
thetic interneurons (Tang et al., 2003), it suffers
from two drawbacks. First, respiratory and vas-
cular movements of the spinal cord often prevent
one from approaching interneurons closely enough
to label them without injuring them. Second, al-
though the somas and dendrites of labeled spinal
sympathetic interneurons are well demonstrated
by juxtacellular labeling, axons are never observed.
Axons of spinal sympathetic interneurons can be
demonstrated by intracellular labeling. To date,
however, intracellular labeling of spinal sympa-
thetic interneurons has been accomplished only in
vitro (Deuchars et al., 2001).
Although neurophysiological studies permit
functional characterization of spinal sympathetic
interneurons, correlation methods, alone, cannot
unequivocally identify spinal sympathetic interneu-
rons. Spinal sympathetic interneurons can be more
definitively identified by retrograde, trans-synaptic

tracing from sympathetic preganglionic neurons.
In an ingenious series of experiments, Cabot and
colleagues (1994) simultaneously injected the beta
subunit of cholera toxin (cholera toxin B) and
wheat germ agglutinin into the superior cervical
ganglion of rats. Both the cholera toxin B and the
wheat germ agglutinin were transported from the
ganglion to the somas and dendrites of sympa-
thetic preganglionic neurons with synapses in that
ganglion. However, only the wheat germ aggluti-
nin was transported further in the retrograde di-
rection, across the synapses made by spinal
sympathetic interneurons on sympathetic pre-
ganglionic neurons, thereby labeling the spinal
sympathetic interneurons. Thus, sympathetic pre-
ganglionic neurons were identified by their com-
bined labeling with cholera toxin B and wheat
germ agglutinin. Spinal sympathetic interneurons
were identified by their labeling with wheat germ
agglutinin but not cholera toxin B. Although these
were landmark experiments, they were hampered
by faint labeling of spinal sympathetic interneu-
rons, due in large part to restricted, retrograde,
trans-synaptic transport of wheat germ agglutinin
from sympathetic preganglionic neurons.
More recently, spinal sympathetic interneurons
have been identified by the retrograde, trans-synap-
tic transport of herpes viruses (Strack et al., 1989a,
b; Schramm et al., 1993; Clarke et al., 1998; Tang
et al., 2004). Herpes simplex and pseudorabies vi-
rus are two herpes viruses that are rapidly taken up
by the axons of sympathetic postganglionic neu-
rons and by the axons of preganglionic neurons
projecting to the adrenal medulla. Virus is trans-
ported back to the somas of these neurons where it
replicates and moves trans-synaptically to the neu-
rons’ synaptic antecedents. Thus, virus taken up
from a peripheral organ or tissue by sympathetic
postganglionic neurons infects the sympathetic pre-
ganglionic neurons that synapse on those neurons.
Virus replicates in the sympathetic preganglionic
neurons, and spinal and brainstem interneurons
that synapse on infected sympathetic preganglionic
neurons are infected by further retrograde, trans-
synaptic movement of virus. Antibodies to the vi-
ruses are used to label infected neurons. The ap-
proximate number of synapses traversed by the
virus can be controlled by the interval between the
31
infection and the perfusion of the animal. This
interval usually ranges between 3 and 6 days. For
identification of spinal sympathetic interneurons,
infected rats are kept for approximately 72 h before
perfusion. Rats kept for this time manifest no
visible symptoms of the infection.
When virus is injected into the adrenal gland,
both preganglionic neurons projecting directly to
adrenal chromaffin cells and postganglionic neu-
rons projecting to both adrenal medullary and ad-
renal cortical blood vessels are infected. Therefore,
the spinal sympathetic interneurons infected by
injection of virus into the adrenal gland may be-
long to at least two classes of interneurons, neu-
rons involved in overall metabolic regulation and
neurons involved in the regulation of the adrenal
circulation. As discussed below, the distinction
between these classes of interneurons may be of
limited importance because it is likely that neither
play an important role in animals with intact spi-
nal cords. After spinal cord lesions, most stimuli
that activate one class of adrenal spinal sympa-
thetic interneurons are likely to activate both.
As in the case of the cholera toxin B and wheat
germ agglutinin experiments described above, viral
methods also require distinguishing between sym-
pathetic preganglionic neurons and spinal sympa-
thetic interneurons. Sympathetic preganglionic
neurons can be identified because, in addition to
being immunohistochemically labeled for the virus,
they also label positively for choline acetyl transf-
erase, a synthetic enzyme for acetyl choline found
in relatively few spinal neurons other than sympa-
thetic preganglionic neurons and somatic motoneu-
rons. Thus, spinal neurons that co-label for virus
and choline acetyl transferase can be identified as
sympathetic preganglionic neurons, and neurons
that are infected but do not co-label for choline
acetyl transferase can be identified as spinal sym-
pathetic interneurons. Sympathetic preganglionic
neurons and somatic motoneurons can be distin-
guished by their differential, dorsoventral locations.
An alternative method for identifying sympathetic
preganglionic neurons depends on their propensity
for transporting retrograde tracers from the circu-
lation to their somas and dendrites (Fig. 2). In this
method, a large quantity (8–12 mg/kg) of a con-
ventional retrograde tracer such as Fluorogolds
32

Fig. 2. Anatomical identification of interneurons using pseudorabies virus and Fluorogolds. Left panel: ultraviolet illumination.
Sympathetic preganglionic neuron (white arrow) identified by fluorescence of intraperitoneally injected Fluorogolds. Right panel:
under illumination for the chromogen used to identify pseudorabies virus, both the sympathetic preganglionic neuron (white arrow)
and a spinal sympathetic interneuron (gray arrow) are visible as gray neurons. The spinal sympathetic interneuron is definitively
identified by its absence under ultraviolet illumination. After Tang et al. (2004), with permission.

is injected either intraperitoneally or subcutaneously
(Anderson and Edwards, 1994). Approximately 1
week post-injection, most peripherally projecting
neurons (such as autonomic preganglionic neurons
and somatic motoneurons) are labeled with the
tracer and can be detected under ultraviolet illu-
mination. Although the labeling of somatic moto-
neurons is highly variable by tracers administered
intraperitoneally, the labeling of autonomic pre-
ganglionic neurons is more uniform. A potential
drawback of this method is that freshly adminis-
tered Fluorogolds appears to interfere with some
viral tracing methods (Strack and Loewy;
Schramm, unpublished data). In our hands, how-
ever, pseudorabies virus can be safely injected 1
week after treatment with this retrograde tracer.
The major drawback of the viral tracing meth-
ods is that infection by the virus may be capri-
cious. Within a population of identically treated,
virus-injected animals, some may not exhibit any
infection, some may exhibit infections that appear
highly specific (infecting only sympathetic pre-
ganglionic neurons and spinal sympathetic inter-
neurons) and some may exhibit infections that
destroy many neurons. A second drawback is that
the number of synapses retrogradely traversed by
the virus can only be estimated from the survival
time. Finally, the viral infection often initiates an
immune response that, itself, could alter the fur-
ther transport of the virus.
Spinal interneurons play a more important role in
generating sympathetic activity after spinal cord
lesions in rats
Although most investigators have found that ac-
tivity is reduced in sympathetic nerves of unanest-
hetized people (Wallin, 1986) and rats
(Krassioukov and Weaver, 1995; Randall et al.,
2005) after spinal cord transection, many investi-
gators report that detectable levels of ongoing ac-
tivity remain in some nerves. Therefore, spinal
sympathetic interneurons must provide ongoing
excitatory input to sympathetic preganglionic neu-
rons in the absence of pathways from the brain-
stem sympatho-excitatory systems. Although in
anesthetized, surgically prepared rats with acute
spinal transections, sympathetic activity is sub-
stantially reduced in some nerves, it is maintained
or even increased in others (Meckler and Weaver,
1985; Taylor and Schramm, 1987).
The observations of decreased activity in some
nerves are easily explained by the decrease in sup-
raspinal drive to some sympathetic preganglionic
neurons after spinal cord injury. Maintenance —
and even increases — in sympathetic activity after
spinal cord injury are less easily explained. Very
likely, sympathetic preganglionic neurons whose
activity was either not diminished or was increased
after spinal transection received little drive from
brainstem circuits before transection. Alternatively,

brainstem sources of activity for these neurons
were replaced by even more powerful intraspinal
sources after transection. In either case, it also is
likely that potentially excitatory spinal inputs to
these sympathetic preganglionic neurons were
under tonic inhibition from supraspinal systems.
Thus, I propose that spinal transection abolishes
descending excitation, either directly to sympa-
thetic preganglionic neurons or indirectly to spinal
sympathetic interneurons. However, it also abol-
ishes descending inhibition of spinal systems with
excitatory inputs to sympathetic preganglionic
neurons. Ruggiero et al. (1997a, b) provided clear
evidence that acute spinal transection releases the
activities of many dorsal horn and intermediate
zone neurons from inhibition. They found that
acute cervical spinal cord transection in anestheti-
zed rats and pigs significantly increased the
number of neurons expressing the c-fos gene in
many dorsal horn laminae and in lamina VII of the
thoracic spinal cord.
Based on these observations, Miller et al. (2001)
predicted that spinal neurons with ongoing activ-
ities correlated with renal sympathetic nerve ac-
tivity would be relatively rare in rats with intact
spinal cords because spinal circuits that might ex-
cite sympathetic preganglionic neurons would be
under tonic, supraspinal inhibition. As noted
above, this prediction was confirmed by their ob-
servation that the activities of only one-fifth as
many spinal interneurons were correlated with re-
nal sympathetic nerve activity in rats with intact
spinal cords as were correlated in rats with acutely
transected spinal cords.
The generation of ongoing sympathetic activity after
spinal transection is localized to a restricted number
of spinal segments
As described above, after acute spinal transection,
activity persists in some sympathetic nerves.
To what extent is this ongoing activity generated
locally, and to what extent does it represent activ-
ity common to the entire spinal cord? Chau et al.
(1997) searched the spinal cord from T2 to the 2nd
lumbar (L2) segment for interneurons with activ-
ities correlated to renal sympathetic nerve activity
in rats with acute spinal transections. Although the
33
ongoing activities of almost 50% of the interneu-
rons recorded at T10 were correlated to ongoing
renal sympathetic nerve activity, the activities of
only 16% of interneurons at T8 were correlated
with renal sympathetic nerve activity. The activi-
ties of no interneurons at T2, T13 or L2 were cor-
related with renal sympathetic nerve activity. In
unpublished studies (Chau and Schramm), this
exploration extended to C2 and L5 without detect-
ing additional interneurons correlated with renal
sympathetic nerve activity. Because anatomical
data indicate that the sympathetic preganglionic
neurons that are most likely to generate renal
sympathetic nerve activity lie in 8th through 12th
thoracic segments (Tang et al., 2004), these data
show that circuits in distant spinal segments play
little role in generating ongoing renal sympathetic
nerve activity. Whether a similar degree of longi-
tudinal specificity exists for cardiac and pelvic
sympathetic nerves remains to be determined.
Long propriospinal pathways affecting sympathetic
activity are multisynaptic
Although distant spinal segments appear to play
little or no role in generating ongoing sympathetic
activity in a given segment after spinal transection,
sympathetic reflexes can be evoked by stimulating
afferents to distant segments (see Weaver and
Polosa, 1997, for review). To what extent are the
sympathetic reflexes elicited from distant segments
mediated by monosynaptic projections to sympa-
thetic preganglionic neurons? Cabot et al. (1994)
noted that spinal sympathetic interneurons retro-
gradely labeled by injecting wheat germ agglutinin
into the superior cervical ganglion exhibited ‘‘a
strict segmental organization’’ with respect to their
associated sympathetic preganglionic neurons. In
other words, wheat germ agglutinin-labeled neu-
rons (spinal sympathetic interneurons) were not
found in segments that did not contain cholera
toxin B-labeled neurons (sympathetic preganglion-
ic neurons).
These observations were confirmed in the renal
sympathetic system using the retrograde transport
of pseudorabies virus. Tang et al. (2004) found
that between caudal cervical and caudal lumbar
segments, infected spinal sympathetic interneurons
34
were located only in segments of caudal thoracic
and rostral lumbar segments, the segments in
which infected sympathetic preganglionic neurons
were also located. The first infected thoracic
interneurons appear 68–72 h after injection of
pseudorabies virus into the kidney. This delay is
identical to that required to infect brainstem neu-
rons that have known, monosynaptic projections
to sympathetic preganglionic neurons. Apparently
in this model, the time required for retrograde
transport of pseudorabies virus from its uptake
at a synapse to the soma of the next neuron is brief
compared to the time necessary for enough
replication to occur for the virus to be visible
immunohistochemically in that newly infected
neuron. Similarly, the transport time is brief com-
pared to the time required for replication to in-
crease the intracellular concentration of virus for
retrograde infection of a neuron’s synaptic ante-
cedents. Because neurons as far rostral as the
paraventricular nucleus of the hypothalamus can
be infected in as little as 72 h, the absence of spinal
sympathetic interneurons in caudal cervical, ros-
tral thoracic and caudal lumbar spinal cord that
lack infected sympathetic preganglionic neurons
strongly suggests that long propriospinal inputs to
sympathetic preganglionic neurons infected from
renal injections are multisynaptic.
The majority of spinal sympathetic interneurons
projecting monosynaptically to sympathetic pre-
ganglionic neurons are located either among or
just dorsal to their functionally related popula-
tions of sympathetic preganglionic neurons. Not
only are the longitudinal distributions of spinal
sympathetic interneurons and their related sympa-
thetic preganglionic neurons similar, but the den-
sities of spinal sympathetic interneurons are
greatest in or near the spinal laminae that contain
their associated sympathetic preganglionic neu-
rons. Thus, Cabot et al. (1994) localized spinal
sympathetic interneurons to the sympathetic pre-
ganglionic neuron-rich lateral portion of lamina
VII and the reticulated (lateral) portion of lamina
V, just dorsal to the intermediolateral column.
Clarke et al. (1998) used the retrograde transport
of modified Herpes simplex virus to identify spinal
sympathetic interneurons that were presynaptic to
adrenal sympathetic preganglionic neurons. Infected
adrenal sympathetic preganglionic neurons were
located across the entire mediolateral span of lamina
VII, and spinal sympathetic interneurons were
similarly distributed, usually intercalated among
the sympathetic preganglionic neurons. Pseudorabies
virus injected into the kidney of the rat also
infected sympathetic preganglionic neurons locat-
ed across the entire intermediate zone of the spinal
cord between the lateral funiculus and lamina X
(Tang et al., 2004). The majority of spinal sympa-
thetic interneurons labeled in those experiments
were similarly distributed.
Although most anatomically identified spinal
sympathetic interneurons have been detected
among, or just dorsal to, populations of sympa-
thetic preganglionic neurons, small numbers iden-
tified after renal injections of pseudorabies virus
were located (in descending order of density) in
lamina IV, II, and I (Tang et al., 2004). Interest-
ingly, spinal sympathetic interneurons identified
by their positive cross-correlations with renal sym-
pathetic nerve activity were distributed somewhat
more widely than anatomically identified spinal
sympathetic interneurons, for instance in the me-
dial portions of laminae I, II, and III (Chau et al.,
2000; Tang et al., 2003). The wider distribution of
neurophysiologically identified spinal sympathetic
interneurons was not surprising. Anatomically
identified spinal sympathetic interneurons were
visualized using a relatively short, post-infection
survival time (72 h). As discussed above, during
that time, pseudorabies virus would have been un-
likely to have traversed more than the two
synapses between the renal sympathetic postgangl-
ionic neurons and the first spinal sympathetic in-
terneurons presynaptic to infected sympathetic
preganglionic neurons. Spinal sympathetic inter-
neurons identified by cross-correlation, on the
other hand, could have been located in spinal cir-
cuits many synapses removed from sympathetic
preganglionic neurons and could, therefore, be ex-
pected to be located more remotely.
The locations of spinal sympathetic interneu-
rons with respect to sympathetic preganglionic
neurons may be more important than their loca-
tions with respect to their inputs. Histological re-
construction of spinal sympathetic interneurons
(Deuchars et al., 2001; Tang et al., 2003) indicated

that the dendritic trees of these neurons often ex-
tended hundreds of microns in two, and sometimes
three, dimensions. Tang et al. (2004) concluded
that the dendrites of some individual spinal sym-
pathetic interneurons were so extensive that they
could receive not only primary afferent inputs but
inputs from a variety of descending or propriospi-
nal pathways as well.
Spinal sympathetic interneurons in rats are more
likely to be excited and less likely to be inhibited by
somatic stimuli after chronic spinal cord transection
Because the severity of autonomic dysreflexia
increases with time after spinal cord injury
(Krassioukov and Weaver, 1995; Krassioukov
et al., 2003), we have supplemented the studies of
rats with acutely transected spinal cords described
above with studies after chronic, T3, spinal tran-
section. In both chronically and acutely spinally
transected rats, Krassioukov et al. (2002) identified
spinal sympathetic interneurons in the T10 segment
by cross-correlation with renal sympathetic nerve
activity. They compared the responses in activity
of spinal sympathetic interneurons to somatic
stimulation in those two populations. To stand-
ardize stimulation sites, the left body wall was di-
vided into five regions, beginning at approximately
the T8 dermatome and ending at the left hip and
hindlimb (Fig. 3).
Two types of stimuli were delivered to these re-
gions, a 10-s pinch with toothed forceps (noxious)
and 10 s of brushing with a cotton applicator (in-
nocuous). Responses in the activities of spinal
sympathetic interneurons and renal sympathetic
nerve activity observed in acutely spinally trans-
ected rats by Krassioukov et al. (2002) corre-
sponded closely to those reported previously in
rats with acutely transected spinal cord (Chau
et al., 1997, 2000). Both noxious and innocuous
stimulation of somatic regions projecting to caudal
thoracic spinal cord (Fig. 3, regions 1–3), increased
the magnitudes of bursts in ongoing renal sympa-
thetic nerve activity. Responses in the activities of
spinal sympathetic interneurons were more varia-
ble. Nevertheless, the majority of T10 spinal sym-
pathetic interneurons were excited by stimulation
of regions 1–3. Noxious and innocuous stimulation
35
Fig. 3. Responses of spinal sympathetic interneurons to so-
matic stimulation 1 month after T3 spinal cord transection.
Upper panel: schematic drawing of the cutaneous regions from
which responses of spinal sympathetically correlated interneu-
rons were elicited. Lower panel: representative rate meter re-
sponses of a sympathetically correlated neuron to noxious (left)
and innocuous (right) stimulation of cutaneous regions 1–5 in a
rat chronically transected at T3. From Krassioukov et al.
(2002), with permission.
of somatic regions that project to caudal lumbar
spinal cord (Fig. 3, regions 4 and 5) decreased on-
going renal sympathetic nerve activity. The ongo-
ing activities of a majority of T10 spinal
sympathetic interneurons were inhibited by stim-
ulation of these regions. One month after spinal
cord transection, both noxious and innocuous
stimulation of regions 1, 3, and 5 were significantly
more likely to increase the activities of spinal sym-
pathetic interneurons than in the acutely trans-
ected state, and innocuous stimulation of regions 1
and 5 was less likely to decrease their activities.
Although autonomic dysreflexia may occur in
the acute stage of spinal cord injury (Krassioukov
et al., 2003), it is far more common in the chronic
stage in both humans (Mathias and Frankel, 1992)
and rats (Krassioukov and Weaver, 1995). In rats,
the onset of autonomic dysreflexia correlated well
with morphological changes in sympathetic pre-
ganglionic neurons (Krenz and Weaver, 1998b)
and with increases in sprouting of primary afferent
axons (Krenz and Weaver, 1998a; Wong et al.,
2000). Some of these axons appeared to synapse
on neurons appropriately positioned to be spinal
sympathetic interneurons (Wong et al., 2000). The
electrophysiological experiments described above
provided a neurophysiological correlation to both
36
the morphological changes in sympathetic pre-
ganglionic neurons and primary afferents exhibit-
ed in rats after spinal injury and to the increased
sympathetic reactivity to somatic stimuli experi-
enced by spinally injured patients.
Summary
Physiological data suggest that spinal sympathetic
interneurons play a minimal role in generating and
modulating renal sympathetic activity in rats with
intact spinal cords. Similar data suggest that spinal
sympathetic interneurons are released from nor-
mally occurring, descending, tonic inhibition after
spinal cord injury. Spinal sympathetic interneu-
rons then play a significant (although not a func-
tionally adaptive) role in both generating ongoing
sympathetic activity and participating in spinal
sympathetic reflexes. Most spinal sympathetic in-
terneurons are located close to their associated
sympathetic preganglionic neurons, and the ongo-
ing activity in individual sympathetic nerves ap-
pears to be generated within restricted numbers of
spinal segments. Although stimulation of afferents
to distant spinal segments can affect the activity of
sympathetic preganglionic neurons, multisynaptic
(rather than monosynaptic) pathways mediate
these effects. Finally, 1 month after spinal cord
transection at rostral thoracic level, spinal sympa-
thetic interneurons at T10 are more likely to be
excited by both innocuous and noxious stimula-
tion of more widespread cutaneous regions. Col-
lectively, these observations provide anatomical
and physiological substrates for both the genera-
tion of ongoing sympathetic activity and the in-
tense, dysfunctional, sympathetic responses to
both noxious and innocuous stimuli observed in
humans after spinal cord injury.
Acknowledgment
Research co-authored by Lawrence Schramm and
the preparation of this chapter were supported by
NIH grant HL16315. Critical editorial assistance
and advice were provided by Dr. Baohan Pan,
M.D., Ph.D. and Diana C. Schramm, M.A.
References
Anderson, C.R. and Edwards, S.L. (1994) Intraperitoneal injec-
tions of Fluorogold reliably labels all sympathetic pregangl-
ionic neurons in the rat. J. Neurosci. Methods, 53: 137–141.
Barman, S.M. and Gebber, G.L. (1984) Spinal interneurons
with sympathetic nerve-related activity. Am. J. Physiol., 247:
R761–R767.
Blessing, W.W. (1997) The Lower Brainstem and Bodily Home-
ostasis. Oxford University Press, New York.
Cabot, J.B. (1996) Some principles of the spinal organization of
the sympathetic preganglionic outflow. Prog. Brain Res., 107:
29–42.
Cabot, J.B., Alessi, V., Carroll, J. and Ligorio, M. (1994) Spinal
cord lamina V and lamina VII interneuronal projections in
sympathetic preganglionic neurons. J. Comp. Neurol., 347:
515–530.
Chau, D., Johns, D.G. and Schramm, L.P. (2000) Ongoing and
stimulus-evoked activity of sympathetically correlated neu-
rons in the intermediate zone and dorsal horn of acutely
spinalized rats. J. Neurophysiol., 83: 2699–2707.
Chau, D., Kim, N. and Schramm, L.P. (1997) Sympathetically
correlated activity of dorsal horn neurons in spinally
transected rats. J. Neurophysiol., 77: 2966–2974.
Clarke, H.A., Dekaban, G.A. and Weave, L.C. (1998) Identi-
fication of lamina V and VII interneurons presynaptic to
adrenal sympathetic preganglionic neurons in rats using a
recombinant herpes simplex virus type 1. Neuroscience, 85:
863–872.
Deuchars, S.A., Brooke, R.E., Frater, B. and Deuchars, J.
(2001) Properties of interneurones in the intermediolateral
cell column of the rat spinal cord: role of the potassium
channel subunit KV3. 1. Neuroscience,, 106: 433–446.
Gebber, G.L. and McCall, R.B. (1976) Identification and dis-
charge patterns of spinal sympathetic interneurons. Am. J.
Physiol., 231: 722–723.
Krassioukov, A.V. and Weaver, L.C. (1995) Episodic hyper-
tension due to autonomic dysreflexia in acute and chronic
spinal cord-injured rats. Am. J. Physiol., 268: H2077–H2083.
Krassioukov, A.V., Furlan, J.C. and Fehlings, M.G. (2003)
Autonomic dysreflexia in acute spinal cord injury: an under-
recognized clinical entity. J. Neurotrauma,, 20: 707–716.
Krassioukov, A.V., Johns, D.G. and Schramm, L.P. (2002)
Sensitivity of sympathetically correlated spinal interneurons,
renal sympathetic nerve activity, and arterial pressure to so-
matic and visceral stimuli after chronic spinal injury. J. Ne-
urotrauma,, 19: 1521–1529.
Krenz, N.R. and Weaver, L.C. (1998a) Sprouting of primary
afferent fibers after spinal cord transection in the rat. Ne-
uroscience, 85: 443–458.
Krenz, N.R. and Weaver, L.C. (1998b) Changes in the mor-
phology of sympathetic preganglionic neurons parallel the
development of autonomic dysreflexia after spinal cord injury
in rats. Neurosci. Lett., 243: 61–64.
Laskey, W. and Polosa, C. (1988) Characteristics of the sym-
pathetic preganglionic neuron and its synaptic input. Prog.
Neurobiol., 31: 41–87.

Mathias, C.J. and Frankel, H.L. (1992) Autonomic disturbanc-
es in spinal cord lesions. In: Bannister R. and Mathias C.J.
(Eds.), Autonomic Failure, A Textbook of Clinical Disorders
of the Autonomic Nervous System. Oxford University Press,
New York, pp. 839–881.
Meckler, R.L. and Weaver, L.C. (1985) Splenic, renal, and
cardiac nerves have unequal dependence upon tonic supra-
spinal inputs. Brain Res., 338: 123–135.
Miller, C.O., Johns, D.G. and Schramm, L.P. (2001) Spinal
interneurons play a minor role in generating ongoing renal
sympathetic nerve activity in spinally intact rats. Brain Res.,
918: 101–106.
Pinault, D. (1996) A novel single-cell staining procedure per-
formed in vivo under electrophysiological control: morpho-
functional features of juxtacellularly labeled thalamic cells
and other central neurons with biocytin or neurobiotin.
J. Neurosci. Meth., 65: 113–136.
Randall, D.C., Baldridg, B.R., Zimmerman, E.E, Carroll, J.J.,
Speakman, R.O., Brown, D.R., Taylor, R.F., Patwardhan,
A. and Burgess, D.E. (Oct. 21, 2004; Epub ahead of print)
Blood pressure power within frequency range around 0.4 Hz
in rat conforms to self-similar scaling following spinal cord
transection. Am. J. Physiol. Regul. Integr. Comp. Physiol.
Ruggiero, D.A., Sica, A.L., Anwar, M., Frasie, I., Gootman, N.
and Gootman, P.M. (1997a) Induction of c-fos gene expression
by spinal cord transection in Sus scrofa. Brain Res., 763: 21–29.
Ruggiero, D.A., Anwar, M., Kim, J., Sica, A.L., Gootman,
A.L. and Gootman, P.M. (1997b) Induction of c-fos gene
expression by spinal cord transection in the rat. Brain Res.,
763: 301–305.
Schramm, L.P., Strack, A.M., Platt, K.B. and Loewy, A.D.
(1993) Peripheral and central pathways regulating the kidney:
a study using pseudorabies virus. Brain Res., 616: 251–262.
37
Schreihofer, A.M. and Guyenet, P.G. (1997) Identification of
C1 presympathetic neurons in rat rostral ventrolateral me-
dulla by juxtacellular labeling in vivo. J. Comp. Neurol., 387:
524–536.
Strack, A.M., Sawyer, W.B., Platt, K.B. and Loewy, A.D.
(1989a) CNS cell groups regulating the sympathetic outflow
to adrenal gland as revealed by transneuronal cell body labe-
ling with pseudorabies virus. Brain Res., 491: 274–296.
Strack, A.M., Sawyer, W.B., Hughes, J.H., Platt, K.B. and
Loewy, A.D. (1989b) A general pattern of CNS innervation
of the sympathetic outflow demonstrated by transneuronal
pseudorabies viral infections. Brain Res., 491: 156–162.
Tang, X., Neckel, N.D. and Schramm, L.P. (2003) Locations
and morphologies of sympathetically correlated neurons in
the T10 spinal segment of the rat. Brain Res., 976: 185–193.
Tang, X., Neckel, N.D. and Schramm, L.P. (2004) Spinal in-
terneurons infected by renal injection of pseudorabies virus in
the rat. Brain Res., 1004: 1–7.
Taylor, R.F. and Schramm, L.P. (1987) Differential effects of
spinal transection on sympathetic nerve activities in rats. Am.
J. Physiol., 253: R611–R618.
Wallin, G. (1986) Abnormalities of sympathetic regulation after
cervical cord lesions. Acta Neurochir. Suppl. (Wien), 36:
123–124.
Weaver, L.C. and Polosa, C. (1997) Spinal cord circuits pro-
viding control of sympathetic preganglionic neurons. In: Jor-
dan D. (Ed.), The Autonomic Nervous System: Central
Nervous Control of Autonomic Function. Harwood Aca-
demic Press, Amsterdam, pp. 29–61.
Wong, S.T., Atkinson, B.A. and Weaver, L.C. (2000) Confocal
microscopic analysis reveals sprouting of primary afferent
fibres in rat dorsal horn after spinal cord injury. Neurosci.
Lett., 296: 65–68.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Published by Elsevier B.V.

CHAPTER 3

Which pathways must be spared in the injured human

spinal cord to retain cardiovascular control?

Andrei Krassioukov

International Collaboration on Repair Discoveries (ICORD) and School of Rehabilitation, University of British Columbia,
Vancouver, BC, Canada and Department of Physical Medicine and Rehabilitation, University of Western Ontario,
London, ON, Canada

Abstract: Cardiovascular abnormalities following spinal cord injury are attributed to autonomic instability
caused by a combination of changes occurring within the spinal cord, including loss of descending au-
tonomic control and plastic changes within spinal and peripheral circuits. Previous animal studies have
shown that localized disruption of the descending vasomotor pathways results in cardiovascular changes
similar to those observed following cord injury. However, the location of these pathways in humans is
uncertain. This chapter presents clinical and histopathological findings from individuals with spinal cord
injury that associates a common area of white matter destruction with severe cardiovascular symptoms.
These data provide evidence that descending vasomotor pathways in the human spinal cord project through
the dorsal aspects of the lateral funiculus.

Introduction cord-injured individuals (Piepmeier et al., 1985;

People with cervical or high thoracic spinal cord
injury face life-long abnormalities in systemic ar-
terial pressure control (Mathias and Frankel,
1992; Karlsson, 1999; Teasell et al., 2000). In gen-
eral, their basal systemic arterial pressure is lower
than normal and is complicated by orthostatic
intolerance (Mathias and Frankel, 1992; Cariga
et al., 2002). In addition, these cord-injured people
experience transient episodes of hypertension,
known as ‘‘autonomic dysreflexia’’ that are often
associated with disturbances in heart rate and
rhythm (Krassioukov et al., 2003; Clydon et al.,
2005). The severity of spinal cord injury varies
between individuals and impacts greatly upon
cardiovascular control. For example, hypotension
affects between 20% and 30% of all spinal
Corresponding author. Tel.: +603 822 2673 (off.);
+604 822 9305 (lab); Fax: +604 822 2924;
E-mail: krassioukov@icord.org
Lehmann et al., 1987; Atkinson and Atkinson, 1996).
Likewise, the severity of autonomic dysreflexia
correlates with completeness of spinal injury
as assessed by the American Spinal Injury
Association (ASIA) score: only 27% of incom-
plete quadriplegics present signs of dysreflexia in
comparison with 91% of complete quadriplegics
(Curt et al., 1997).
Determinations of plasma catecholamine levels
in cord-injured individuals, and other evidence,
suggest that a decrease in sympathetic neuronal
activity is the main cause of the hypotension and
postural intolerance (Figoni, 1984; Mathias, 1995;
Karlsson et al., 1998; Gao et al., 2002). The de-
creased sympathetic activity, in turn, presumably
results from damage to the spinal pathways that
carry facilitatory input from the lower brainstem
to the sympathetic preganglionic neurons. The de-
struction of these descending vasomotor pathways,
resulting in the loss of excitatory supraspinal input
to the spinal sympathetic preganglionic neurons,

DOI: 10.1016/S0079-6123(05)52003-X 39
40

is currently considered the major factor for the

persistent lack of sympathetic tone after spinal
cord injury (Mathias and Frankel, 1992; Atkinson
and Atkinson, 1996). Damage to the spinal path-
ways that carry inhibitory input from the lower
brainstem to the sympathetic preganglionic neu-
rons may have a role in autonomic dysreflexia, by
allowing exaggerated activity in the spinal reflex
circuits, caudal to the lesion, that connect spinal
afferent projections to preganglionic neurons. Fig. 1. Schematic diagrams of the possible localization of de-
scending vasomotor pathways within the spinal cord as previ-
Damage to the pathways from the lower brain- ously reported in experimental animals (A) and human
stem to sympathetic preganglionic neurons has a investigations (B). These pathways were localized within dif-
central role in generating the abnormal systemic ferent regions of the white matter in primates (Kerr and
arterial pressure control, typical of cervical or high Alexander, 1964) (arrow 1), in cats and rodents (Lebedev et al.,
thoracic spinal cord injury. Therefore, these path- 1986; Reis et al., 1988; Ruggiero et al., 1989) (arrow 2), and in
humans (Nathan and Smith, 1987) (arrow 3). (B) The boxed
ways are a high priority target for repair, regen- areas indicate the two potential localizations of descending
erative and neuroprotective treatment. Knowledge vasomotor pathways in man, which we examined in the present
of the localization of these pathways in the human investigation: the dorsal aspect of lateral funiculus (Area I), and
spinal cord is therefore essential. This chapter de- white matter adjacent to dorsolateral aspects of the intermedio-
scribes an approach to obtaining such knowledge. lateral cell column (Area II). (From Furlan et al. (2003), with
permission from J. Neurotrauma.)
Previous work in experimental animals, using
electrical stimulation or lesions, led to the conclu-
symptoms had the greatest damage in Area I. The
sion that the pathways for cardiovascular control
remainder of the cases, who had only minor car-
run in the dorsal aspect of the lateral funiculus of
diovascular symptoms, or no symptoms at all, had
the spinal cord white matter (Kerr and Alexander,
significantly less damage in this area. The degree
1964; Illert and Gabriel, 1972; Foreman and
of damage to Area II did not correlate well with
Wurster, 1973; Lebedev et al., 1986). Henceforth,
the extent of cardiovascular dysfunction in these
this area of the white matter will be referred to as
individuals.
Area I (Fig. 1). By contrast, a study in patients
On the basis of these data it may be concluded
undergoing limited cordotomy for the relief of
that, in humans, the pathways from the lower
chronic pain resistant to medical treatment, has
brainstem to the sympathetic preganglionic neu-
suggested that these pathways run in the white
rons run in the dorsal aspect of the lateral fun-
matter adjacent to the dorsolateral aspect of the
iculus of the white matter. Since this is the general
intermediolateral cell column (Nathan and Smith,
region where previous work on experimental an-
1987). Henceforth, this area of the white matter
imals (rat, cat, dog, others) had localized similar
will be referred to as Area II (Fig. 1).
pathways, this study further confirms the relevance
This chapter describes a retrospective study of
to humans of animal model studies of autonomic
cases of spinal cord injury from which detailed
dysfunctions after spinal cord injury.
clinical records and spinal cord specimens were
available. The cases with the most severe cardio-
vascular symptoms were identified. It was hypoth-
esized that this case subset would also have the Study groups
most severe damage to the pathways from lower
brainstem to the sympathetic preganglionic neu- We retrospectively reviewed the charts of the spi-
rons. The extent and severity of white matter nal cord injury cases included in this study and
damage was estimated using stains for myelin and collected data on age and gender, causes of spinal
for an axoplasmic marker. This study has shown cord injury, neurological assessment (including
that the group with the most severe cardiovascular severity and level of injury), cardiovascular
 41

parameters, and clinical history predating the
spinal cord injury, for example, pre-existing
cardiovascular disease. Detailed information on
cardiovascular parameters was collected during
the acute stage of injury in all individuals. We also
searched for evidence of episodes of autonomic
dysreflexia in these individuals. Heart rate and
blood pressure had been evaluated in each patient
hourly for the first 2 weeks and then every 2 h until
discharge. Daily averages were calculated from
all measurements available for each day in the
patients’ charts during a 5-week post-injury period.
These cases of cord injury all had a cervical injury
and were assigned to one of the two groups: cases
that developed severe cardiovascular dysfunction
during the acute post-injury period (Group 1), and
cases with no or minor cardiovascular dysfunction
in this period (Group 2). The control group
(Group 3) included five cases with intact central
nervous system. Neurological evaluation of the
severity of cord injury was conducted through as-
sessment of motor and sensory impairments ac-
cording to the ASIA scoring system (Maynard Jr.
et al., 1997). The ASIA Grade A represents the
most severe, complete injury with complete motor
and sensory impairment, and ASIA Grade D
characterizes minor, incomplete cord injury (mild
motor dysfunction and no sensory loss).
A total of seven cases with spinal cord injury
(two females and five males, aged 31–82 years with
a mean of 60.0 years), and five individuals with
intact central nervous system (two females and
three males, aged 30–73 years with a mean of 51.4
years) were analyzed. Individuals from the control
group were comparable to the cord injury group
with regard to age (P ¼ 0.42) and gender
(P ¼ 1.0). All spinal cord-injured individuals had
a cervical injury. Neurological evaluation using the
ASIA scale showed that individuals in Group 1
had a more severe cord injury (Table 1).
Cardiovascular parameters
There were significant differences in the cardio-
vascular parameters between individuals in
Groups 1 and 2. Severe hypotension, bradycardia,
and episodes of autonomic dysreflexia, which are
signs of disrupted supraspinal cardiovascular con-
trol, were prominent among the cases in Group 1.
Severe hypotension (neurogenic shock) in the early
post-injury period required the administration of
vasopressive agents to all individuals in Group 1.
Intravenous dopamine was administered, on aver-
age, for 774.1 days (1–19 days) in this group. In
contrast, only one individual in Group 2 required
infusion of dopamine for a period of 11 h. After

Table 1. Clinical and neurological data in spinal cord-injured individuals with severe (Group 1) or minor cardiovascular dysfunction
(Group 2) and in control cases (Group 3)

Groups Cases Gender Age (years) ASIA grade Level of SCI Cause of SCI Time from SCI to death

1 1 Female 31 A C 2,3 Diving accident 9 months

2 Male 66 B C 4,5 Fall 6 months
3 Female 43 A C 5,6 Motor vehicle accident 3.5 months
4 Male 66 A C 6,7 Spontaneous epidural hemorrhage 5 months
2 5 Male 65 C C 1,2 Motor vehicle accident 36 months
6 Male 67 A C 2,3 Fall 3.5 months
7 Male 82 B C 5,6 Spontaneous epidural hemorrhage 5 weeks
3 8 Male 30 N/A N/A N/A N/A
9 Male 53 N/A N/A N/A N/A
10 Female 37 N/A N/A N/A N/A
11 Male 73 N/A N/A N/A N/A
12 Female 64 N/A N/A N/A N/A

N/A, not applicable.

Adapted from Furlan et al. (2003), with permission from J. Neurotrauma.
42
the first 2 days, the cases in Group 2 had normal
arterial pressures, no bradycardia and no episodes
of autonomic dysreflexia. Although individuals in
Group 1 were treated with vasopressor agents in
the early stage of cord injury, their systolic and
diastolic blood pressures were significantly lower
than those of Group 2 during the first 4 weeks post
injury. Individuals in Group 1, and some of Group
2 also had bradycardia during the first 2 days after
injury. Then, the mean daily heart rate in Group 1
remained significantly lower than the heart rate in
Group 2 for the following 4 weeks after injury. By
the end of the 4th week, the cases in Group 1 had
recovered cardiovascular function and the two
groups did not differ with respect to heart rate and
blood pressure.
Three individuals in Group 1 (Cases 1, 3, and 4
of Table 1) developed episodes of autonomic
dysreflexia during their stay at the hospital. A
typical episode of dysreflexia occurred in Case 1 at
day 4. This was her first episode of dysreflexia.
During this episode her systolic and diastolic blood
pressures reached 180 and 100 mmHg, respectively,
from a resting mean pressure of 90 mmHg. This
episode was accompanied by an increased heart
rate to 80 beats per minute (from a resting rate of
55 beats per minute). Pounding headache, double
vision and anxiety were the major complaints
during this episode. Elevation of the head and
analgesic medication were effective in managing
this episode. No episodes of autonomic dysreflexia
were reported in individuals in Group 2. Pre-
existing systemic hypertension was established in
Case 4 (Group 1) and Case 5 (Group 2) prior to
spinal cord injury. Although daily recording of
cardiovascular parameters was not performed in
individuals from Group 3, we found no clinical
data to suggest cardiovascular abnormalities.
Histopathological findings
The spinal cord tissue from the cases included in
this study was fixed with 10% buffered formalin
for 2 weeks and paraffin embedded. In no case did
the postmortem interval exceed 24 h. In each cord
injury case, at least one segment caudal to the level
of injury (upper thoracic segments) was selected
for examination. In Group 3, the third thoracic
segment was examined. Two sets of alternate spi-
nal cord sections (5–8 mm) were obtained and
stained for: (1) general histology and myelin pres-
ervation using hematoxylin and eosin and luxol
fast blue; and (2) axonal preservation using
immunocytochemical staining for neurofilament
200 (Sigma, 1:200).
Sections were viewed with bright field illumina-
tion (Axioscope, Zeiss), and the extent of injury
and axonal degeneration was examined using the
Northern Eclipse imaging software (Version 6.0,
Empix Imaging Inc). In sections stained for myelin
with luxol fast blue, the total area of demyelina-
tion (Fig. 2, pink areas within the white matter)
was measured and presented as a fraction (percent
and standard error of the mean) of the total sur-
face area of the spinal cord section (including
white and gray matter). In sections stained for ne-
urofilament 200, axonal counts were conducted in
selected areas of white matter (see below).
Axonal preservation was examined within the
two areas that have been previously suggested to
contain descending vasomotor pathways: the dor-
sal aspects of the lateral funiculus (Area I in Fig.
1B), and the white matter adjacent to the dorso-
lateral portions of the intermediolateral column
(Area II in Fig. 1B) (Kerr and Alexander, 1964;
Foreman and Wurster, 1973; Lebedev et al., 1986;
Nathan and Smith, 1987). We also examined the
extent of axonal preservation within the dorsal
columns and the corticospinal tracts in each case
(Fig. 3). Sections stained for neurofilament 200
were examined under low magnification (
1.25)
and areas of interests were identified (Fig. 3). For
each section, using high magnification (
20) im-
ages, at least three representative fields from the
chosen areas were examined (Fig. 3a–3d). Finally,
the axonal counts were conducted using the
Northern Eclipse software. Axonal preservation
was expressed as the mean number of preserved
axons per 10,000 mm27SEM.
Cross-sectional analysis of spinal cord sections
from the high thoracic cord, caudal to the injury
site, revealed that the extent of white matter de-
generation was 24.6572.1% (range: 19.29–29.66%)
in Group 1, and 8.271.2% (range: 6.06–10.15%)
in Group 2 (Fig. 2). This showed that spinal cords
 43

Fig. 2. Myelin staining with Luxol Fast Blue of spinal cord sections from the high thoracic spinal cord. (A) Spinal cord injury case who
developed severe cardiovascular complications (Group 1); (B) spinal cord injury case with no significant cardiovascular dysfunction
(Group 2); and (C) individual with intact CNS (control case, Group 3). Calibration bar is 2 mm. A well-defined butterfly shaped area of
the gray matter is present in all sections. Myelin-containing white matter is stained blue. Areas of axonal degeneration and myelin loss
(pink areas within the white matter) are present in sections from Cases 3 and 7. (D) Average values of white matter degeneration
(expressed as a percent of total spinal cord area) in spinal cord sections from individuals in Group 1 were significantly greater than
those of individuals from Group 2. (From Furlan et al. (2003), with permission from J. Neurotrauma.)

from cases of Group 1 with severe cardiovascular
dysfunction after cord injury had more extensive
areas of white matter degeneration than spinal
cords from cases who had no or minor cardiovas-
cular dysfunction after injury (Po0.002).
Axons within the spinal cord were unequiv-
ocally identified in spinal cord sections using
immunohistochemistry for neurofilament 200 and
bright field microscopy (Fig. 3B). In control cases,
axons were evenly distributed throughout the white
matter. However, there was a striking difference in
axonal preservation in different regions of the spi-
nal cord from individuals with spinal cord injury.
The number of preserved axons per 10,000 mm2
within the corticospinal tract in Group 1 (2175)
was significantly lower than in Group 2 (88712;
Po0.001) and Group 3 (117714; Po0.001).
There was no significant difference between the
number of axons within the corticospinal tract of
individuals from Groups 2 and 3 (P ¼ 0.184).
The number of preserved axons per 10,000 mm2
within the dorsal column in Group 1 (100713),
Group 2 (131714), and Group 3 (136714) was
similar (P ¼ 0.167). There were fewer preserved
axons per 10,000 mm2 in Area I (dorsal aspects of
the lateral funiculus) in individuals from Group 1
(2075) than in individuals from Group 2 (52715;
P ¼ 0.029) and Group 3 (6573; Po0.001). Also,
the number of preserved axons within the Area I in
Group 2 was significantly less than in Group 3
(P ¼ 0.034).
The number of preserved axons per 10,000 mm2
within Area II (white matter adjacent to the
dorsolateral aspects of the intermediolateral cell
column) in Group 1 (5977) was significantly re-
duced in comparison with Group 2 (93713;
P ¼ 0.028) and Group 3 (109715; P ¼ 0.013).
There were no significant differences in axonal
counts within Area II between Groups 2 and 3
(P ¼ 0.357).
Discussion
Previous investigations have demonstrated that
hypotension, bradycardia, and autonomic dysre-
flexia occur more frequently in individuals with
44

Fig. 3. (A) Staining of axons by immunocytochemistry for neurofilament 200 (NF200) in low power (
1.25) photomicrograph of a
spinal cord from a cord-injured individual with severe cardiovascular dysfunction (Case 4). Calibration bar is 1 mm. The squares
indicate the areas of the spinal cord in which preserved axons were counted. The four areas examined in this study for axonal
preservation were the following: dorsal column (DC), Area I, lateral corticospinal tracts (CST), and Area II. (B) High magnification
(
20) of different areas of the spinal cord stained with NF200 from three representative cases, one from each of the groups. Brown-
stained dots represent cross-sections of spinal axons immunocytochemically identified with NF200. There was a significant axonal loss
within Area 1 (panel b-1) and the CST (panel c-1) in all individuals from Group 1. (From Furlan et al. (2003), with permission from
J. Neurotrauma.)

severe cervical spinal cord injury (Lehmann et al.,
1987; Mathias and Frankel, 1992; Noreau et al.,
2000; Silver, 2000). This study has demonstrated a
relationship between the location and severity of
pathology in the spinal cord and cardiovascular
dysfunction in human cases of spinal cord injury.
The histological analysis demonstrates that Group
1, with significant cardiovascular dysfunction, had
greater myelin and axon loss than the cases in
Group 2, who had insignificant cardiovascular
dysfunction. By these criteria, injury was more se-
vere in Group 1 than in Group 2. This matches the
higher ASIA grade for Group 1 than for Group 2,
and the difference in incidence and severity of car-
diovascular symptoms in the two groups. In addi-
tion, the histology shows that the axon loss was
not homogeneously distributed across the area of
the sections. The dorsal columns had very little
axon loss. Area I and the corticospinal tract
showed the greatest loss. Area II had an interme-
diate amount of loss.
We expected that the severe cardiovascular
symptoms of Group 1 would be associated with
a very large axon loss in the area traversed by the
descending vasomotor pathways. Area I had a loss
of 70% of axons in Group 1 cases, whereas this

area in Group 2 was decreased by 20%. In con-
trast, Group 1 lost 20% of axons in Area II vs. the
loss of 15% by Group 2, changes that were rela-
tively similar. Accordingly, the cardiovascular dys-
function in Group 1 and the lack of dysfunction in
Group 2 correlate best with the axonal losses in
Area I. Thus, Area I, the dorsal area of the dorso-
lateral funiculus, seems a more likely candidate
than the more ventral Area II as the site of the
cardiovascular pathways. This view is consistent
with previous work by Fehlings and Tator (1995)
demonstrating a relationship between loss of func-
tion and axonal loss after spinal cord injury (SCI).
They showed that inclined plane score varied log-
arithmically with number of axons at the injury
site, such that loss of greater than 50% of axons is
required for a significant drop in neurological
function.
These observations contradict the conclusions
reached by Nathan and Smith (1987) that, in
humans, the descending vasomotor pathways are
localized to the white matter adjacent to the
dorsolateral aspect of the intermediolateral cell
column (Area II of the present study). These
conclusions were based on the analysis of
postmortem spinal cord sections from patients
who underwent antero-lateral cordotomies for
control of intractable pain, and subsequently,
developed cardiovascular symptoms such as
hypotension and orthostatic intolerance. However,
careful analysis of their data showed that the
antero-lateral cordotomy resulted also in partial
destruction of the lateral funiculus (Area I of the
present study). In other words, the cases in the
report by Nathan and Smith (1987) had damage in
both areas examined in the present study. There-
fore, it cannot be excluded that the relevant
damage was to Area I.
Much of our present understanding of the
pathophysiology of central nervous system
(CNS) disorders, including spinal cord injury, is
based on extrapolations from animal models
(Krassioukov and Weaver, 1996; Krenz and
Weaver, 1998; Maiorov et al., 1998; Osborn
et al., 1989). Although considerable clinical data
are available on neurological function after human
spinal cord injury, only a limited number of stud-
ies have been directed to histopathological changes
45
within the human spinal cord (Bunge et al., 1993;
Hayes and Kakulas, 1997; Puckett et al., 1997;
Kakulas, 1999). Moreover, some histopathological
findings in humans are significantly different from
those observed in animal models of cord injury
(Puckett et al., 1997). Therefore, to extrapolate
information from animal models to human disor-
ders, it is essential to compare findings from an-
imal and human studies.
Using cervical electrical stimulation and selec-
tive lesions, numerous investigators have reported
that the descending vasomotor pathways are lo-
calized within extensive areas from ventral to dor-
sal in the peripheral aspects of the lateral funiculus
in cats and monkeys (Kerr and Alexander, 1964;
Illert and Gabriel, 1972; Foreman and Wurster,
1973). Barman and Wurster (1975) demonstrated
that the descending sympathetic pathways are
situated on the surface of the dorsolateral
funiculus and are organized in a dorsal-to-ventral
manner based on electrical stimulation in dogs.
Lebedev et al. (1986) carried out an electrophys-
iological study before and after the dorsolateral
funiculus transection showing that descending
vasomotor pathways are situated within the
dorsal parts of the lateral funiculus in cats, the
area which corresponds to Area I in our
investigation.
The knowledge of the area greatly responsible
for vasomotor control has significant relevance to
recovery from spinal cord injury. Area I is su-
perficial and easily accessible from the dorsolat-
eral surface of the spinal cord. It is conceivable
that topical treatments could be applied to this
region. Thus, an avenue for future studies is an
exploration of techniques for delivering treatment
to this region. An additional question, raised by
this study, is whether it is possible to distinguish
anatomically, within Area I, the region of faci-
litatory pathways responsible for the maintenance
of blood pressure and the orthostatic tolerance,
from the location of inhibitory pathways that
limit the spinal reflexes responsible for dysreflexia.
This investigation documents that anatomical
studies of cases of spinal cord injury can provide
crucial information that may assist with treatment
of the disabling cardiovascular consequences of
spinal cord injury.
46
Acknowledgments
This study was conducted with the support of a
Christopher Reeve Paralysis Foundation grant
(KB2-0003-1), a Cervical Spine Research Society
grant, support from the Canadian Syringomyelia
Network, and a grant from the Heart and Stroke
Foundation of Ontario (NA4951) awarded to
Dr. A. Krassioukov. Dr. J. Furlan (Toronto,
ON) was a postdoctoral fellow who conducted a
major part of the histopathological analysis. The
author also would like to acknowledge Dr. A.
Marcillo (Miami, FL), Mrs. Lorraine Yamamoto
(Burlington, ON), and Mrs. Lynda Rickards, R.N.
(Toronto, ON) for their assistance and support
during the project.
References
Atkinson, P.P. and Atkinson, J.L.D. (1996) Spinal shock.
Mayo Clin. Proc., 71: 384–389.
Barman, S.M. and Wurster, R.D. (1975) Visceromotor organ-
ization within descending spinal sympathetic pathways in the
dog. Circ. Res., 37: 209–214.
Bunge, R.P., Puckett, W.R., Becerra, J.L., Marcillo, A. and
Quencer, R.M. (1993) Observation on the pathology of hu-
man spinal cord injury. A review and classification of 22 new
cases with details from a case of chronic cord compression
with extensive focal demyelination. In: Seil F.J. (Ed.) Ad-
vances in Neurology, Vol. 59. Raven Press Ltd., New York.
Cariga, P., Catley, M., Mathias, C.J., Savic, G., Frankel, H.L.
and Ellaway, P.H. (2002) Organisation of the sympathetic
skin response in spinal cord injury. J. Neurol. Neurosurg.
Psychiat., 72(3): 356–360.
Clydon, V., Elliott, S., Sheel, W. and Krassioukov, A.V. (2005)
Incidence of cardiac dysrhythmias during sperm retrieval in
men with spinal cord injury. Spinal Cord (submitted).
Curt, A., Nitsche, B., Rodic, B., Schurch, B. and Dietz, V.
(1997) Assessment of autonomic dysreflexia in patients with
spinal cord injury. J. Neurol., Neurosurg. Psychiat., 62:
473–477.
Fehlings, M.G. and Tator, C.H. (1995) The relationships
among the severity of spinal cord injury, residual neurolog-
ical function, axon counts and counts of retrogradely labeled
neurons after experimental spinal cord injury. Exp. Neurol.,
132: 220–228.
Figoni, S.F. (1984) Cardiovascular and haemodynamic re-
sponses to tilting and to standing in tetraplegic patients.
Paraplegia, 22: 99–109.
Foreman, R.D. and Wurster, R.D. (1973) Localization and
functional characteristics of descending sympathetic spinal
pathways. Am. J. Physiol., 225(1): 212–217.
Furlan, J.C., Fehlings, M.G., Shannon, P., Norenberg, M.D.
and Krassioukov, A.V. (2003) Descending vasomotor path-
ways in humans: correlation between axonal preservation
and cardiovascular dysfunction after spinal cord injury.
J. Neurotrauma, 20(12): 1351–1363.
Gao, S.A., Ambring, A., Lambert, G. and Karlsson, A.K.
(2002) Autonomic control of the heart and renal vascular bed
during autonomic dysreflexia in high spinal cord injury. Clin.
Auton. Res., 12(6): 457–464.
Hayes, K.C. and Kakulas, B.A. (1997) Neuropathology of hu-
man spinal cord injury sustained in sports-related activities.
J. Neurotrauma, 14(4): 235–248.
Illert, M. and Gabriel, M. (1972) Descending pathways in the
cervical cord of cats affecting blood pressure and sympathetic
activity. Pflugers Arch., 335: 109–124.
Kakulas, B.A. (1999) A review of the neuropathology of human
spinal cord injury with emphasis on special features. J. Spinal
Cord Med., 22: 119–124.
Karlsson, A.K. (1999) Autonomic dysreflexia. Spinal Cord, 37:
383–391.
Karlsson, A.K., Friberg, P., Lonnroth, P., Sullivan, L. and
Elam, M. (1998) Regional sympathetic function in high
spinal cord injury during mental stress and autonomic
dysreflexia. Brain, 121: 1711–1719.
Kerr, F.W.L. and Alexander, S. (1964) Descending autonomic
pathways in the spinal cord. Arch. Neurol., 10: 249–261.
Krassioukov, A.V., Furlan, J.C. and Fehlings, M.G. (2003)
Autonomic dysreflexia in acute spinal cord injury: an under-
recognized clinical entity. J. Neurotrauma, 20(8): 707–716.
Krassioukov, A.V. and Weaver, L.C. (1996) Morphological
changes in sympathetic preganglionic neurons after spinal
cord injury in rats. Neuroscience, 70: 211–226.
Krenz, N.R. and Weaver, L.C. (1998) Sprouting of primary
afferent fibers after spinal cord transection in the rat. Ne-
uroscience, 85: 443–458.
Lebedev, V.P., Krasyukov(Krassioukov), A.V. and Nikitin,
S.A. (1986) Electrophysiological study of sympathoexcitatory
structures of the bulbar ventrolateral surface as related to
vasomotor regulation. Neuroscience, 17(1): 189–203.
Lehmann, K.G., Lane, J.G., Piepmeier, J.M. and Batsford,
W.P. (1987) Cardiovascular abnormalities accompanying
acute spinal cord injury in humans: incidence, time course
and severity. J. Am. Coll. Cardiol., 10(1): 46–52.
Maiorov, D.N., Fehlings, M.G. and Krassioukov, A.V. (1998)
Relationship between severity of spinal cord injury and ab-
normalities in neurogenic cardiovascular control in conscious
rats. J. Neurotrauma,, 15(5): 365–374.
Mathias, C.J. (1995) Orthostatic hypotension: causes, mecha-
nisms, and influencing factors. Neurology, 45(Suppl 5):
S6–S11.
Mathias, C.J. and Frankel, H.L. (1992) The cardiovascular
system in tetraplegia and paraplegia. In: Frankel H.L. (Ed.)
Handbook of Clinical Neurology, Vol. 17(25). Elsevier,
Amsterdam, pp. 435–456.
Maynard Jr., F.M., Bracken, M.B., Creasey, G., Ditunno Jr.,
J.F., Donovan, W.H., Ducker, T.B., Garber, S.L., Marino,
R.J., Stover, S.L., Tator, C.H., Waters, R.L., Wilberger, J.E.

and Young, W. (1997) International standards for neurolog-
ical and functional classification of spinal cord injury. Amer-
ican Spinal Injury Association. Spinal Cord, 35(5): 266–274.
Nathan, P.W. and Smith, M.C. (1987) The location of de-
scending fibres to sympathetic preganglionic vasomotor and
sudomotor neurons in man. J. Neurol., Neurosurg. Psychiat.,
50: 1253–1262.
Noreau, L., Proulx, P., Gagnon, L., Drolet, M. and Laramee,
M.T. (2000) Secondary impairments after spinal cord injury:
a population-based study. Am. J. Phys. Med. Rehabil., 79(6):
526–535.
Osborn, J.W., Taylor, R.F. and Schramm, L.P. (1989) Deter-
minants of arterial pressure after chronic spinal transection in
rats. Am. J. Physiol., 256: R666–R673.
Piepmeier, J.M., Lehmann, K.B. and Lane, J.G. (1985) Cardi-
ovascular instability following acute cervical spinal cord
trauma. Cent. Nerv. Syst. Trauma, 2(3): 153–160.
47
Puckett, W.R., Hiester, E.D., Norenberg, M.D., Marcillo, A.E.
and Bunge, R.P. (1997) The astroglial response to Wallerian
degeneration after spinal cord injury in humans. Exp. Ne-
urol., 148(2): 424–432.
Reis, D.J., Morrison, S. and Ruggiero, D.A. (1988) The C1 area
of the brainstem in tonic and reflex control of blood pressure.
Hypertension, 11: I8–I13.
Ruggiero, D.A., Cravo, S., Arango, V. and Reis, D.J. (1989)
Central control of the circulation by the rostral ventrolateral
reticular nucleus: anatomical substrates. Prog. Brain Res., 81:
49–79.
Silver, J.R. (2000) Early autonomic dysreflexia. Spinal Cord,
38: 229–233.
Teasell, R., Arnold, A.P., Krassioukov, A.V. and Delaney,
G.A. (2000) Cardiovascular consequences of loss of supra-
spinal control of the sympathetic nervous system following
spinal cord injuries. Arch. Phys. Med. Rehabil., 81: 506–516.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Published by Elsevier B.V.

CHAPTER 4

Disordered control of the urinary bladder after

human spinal cord injury: what are the problems?

Patrick J. Potter

Regional Spinal Cord Injury Rehabilitation Program, and Physical Medicine and Rehabilitation,
St. Joseph’s Health Center, The University of Western Ontario, London, ON, Canada

Abstract: Spinal cord injury has a profound impact on the storage and voiding functions of the urinary
bladder. Loss of autonomic and somatic control mechanisms leads to hypo- or hyperactivity of the bladder
wall and sphincters causing problems that range from incontinence to complete loss of the capacity to
empty the bladder. This chapter outlines the types of bladder dysfunction that occur after spinal cord
injury, their relative prevalence and current practices used to manage the problems. With all the inter-
ventions that are available, management of bladder function often still remains a compromise, as the
medications and physical interventions available may stimulate or block components of the voiding reflex,
but are often not fully restorative in this effort.

Introduction and autonomic functions usually do not recover

Urinary bladder control is dependent upon coor-
dinated interaction between the somatic and au-
tonomic nervous systems (Burns et al., 2001; de
Groat and Yoshimura, 2001). When the supraspi-
nal, coordinated control of these systems is lost,
the resulting bladder dysfunction is termed ‘‘ne-
urogenic bladder impairment’’ or a neurogenic
bladder. The presence of a neurogenic urinary
bladder is extremely common after spinal cord in-
jury (Waites et al., 1993; Cardenas and Hooton,
1995; Vines, 1996; Sapounzi-Krepia et al., 1998;
Stover et al., 1989; Chen et al., 1999; Wolfe et al.,
2002). In general, this dysfunction is not unique to
spinal cord injury as it can arise from disrupted
peripheral or central nervous system control. For
example, a neurogenic bladder can occur in dis-
eases such as multiple sclerosis (Anderson et al.,
1976). Although some neurological recovery may
occur following a spinal cord injury, motor, sensory
Corresponding author. Tel.:+519 685 4080;
Fax: +519 685 4081; E-mail: patrick.potter@sjhc.london.on.ca
completely (Menter and Hudson, 1995; Wolfe
et al., 2002; Potter et al., 2004). When only
partial resolution of the impairment in bladder
control occurs, then accommodation to the altered
physiological condition requires interventions that
range from pharmacological treatments to appli-
ances and mechanical or electrical devices, as pre-
sented in the following chapters. For previous
reviews of the neurogenic bladder after spinal cord
injury see Perkash (2004) and Burns et al. (2001).
Brief overview of normal bladder function
The urinary bladder is a fluid reservoir that nor-
mally empties completely in a well-controlled man-
ner. All of the clinical problems encountered after
spinal cord injury are manifestations of impairment
in these two basic functions of the bladder: storage
and emptying. The lower urinary tract is made up
of the bladder, internal sphincter, external sphinc-
ter and urethra. The bladder wall is composed of
smooth muscle, termed the detrusor muscle, and

DOI: 10.1016/S0079-6123(05)52004-1 51
52

has a base or trigone, a body and a neck. Para- spinal cord and in the bladder lead to malfunctions
sympathetic control of the detrusor muscle origi- in these reflexes and in the storage and voiding
nates from preganglionic axons in the pelvic nerve functions of the urinary bladder.
with cell bodies in the 2nd to 4th sacral (S) spinal
cord segments. Detrusor contraction is mediated
Clinical presentations of bladder dysfunction after
primarily by parasympathetic stimulation. Sympa-
spinal cord injury
thetic control of this muscle comes from the hypo-
gastric nerve that contains axons of preganglionic
Neurogenic bladder impairments depend greatly
neurons that are located in the 10th thoracic (T10)
upon the level and extent of central nervous system
to 2nd lumbar (L2) spinal cord segments. Sensory
injury. The lower levels of injury, such as at the level
innervation of the bladder wall travels to the S2–S4
of the conus medullaris, are more likely to result in
spinal segments via the pelvic nerve. The internal
a flaccid bladder. Thoracic and cervical level inju-
bladder sphincter is made of smooth muscle and is
ries commonly generate mixed pictures of detrusor
located at the junction of the bladder neck and
hyperactivity, sphincter spasticity and lack of coor-
urethra. This sphincter receives parasympathetic
dination between the detrusor and sphincters that is
and sympathetic innervation like that of the de-
termed dyssynergia. Because the urinary bladder is
trusor. However, in this muscle, sympathetic stim-
innervated bilaterally, hemi-cord impairments such
ulation causes contraction. The external sphincter
as the Brown Sequard Syndrome often do not result
is striated muscle that surrounds the urethra and is
in significant bladder dysfunction. Furthermore, in-
controlled by somatic innervation from the S2–S4
juries that spare the central portion of the cord re-
spinal segments that reach this sphincter via the
sult in relative sparing of bladder function.
pudendal nerve. Details of the anatomy and phar-
Issues of bladder dysfunction relate to four major
macology of these pathways are presented in later
problems: (1) inadequate or excessive detrusor
chapters and in reviews by de Groat and
function, (2) inadequate or excessive sphincter func-
Yoshimura (2001) and Burns et al. (2001).
tion, (3) dyssynergy between detrusor and sphincter
To promote the storage function of the urinary
actions and (4) impaired ability to sense the bladder
bladder, sympathetic innervation plays two key
(Lisenemayer and Oakley, 2003). Approaches to
roles. First, through a-adrenergic receptors, the
treatment can therefore be based on manipulation
neck and internal sphincter of the bladder are con-
of these functions. Often combinations of ap-
tracted to close the bladder outlet (Ek et al., 1977).
proaches are required and the type of bladder man-
Next, via b-adrenergic receptors, the body of the
agement may change through a cord-injured
bladder relaxes. Normally, filling of the bladder
person’s life. For example, during acute care im-
occurs with minimal increases in pressure. Voiding
mediately after injury, a Foley catheter is often in-
the bladder is a coordinated process that involves
serted to drain the flaccid bladder. As some degree
contraction of the detrusor muscle with concomi-
of continence develops with time after injury, this
tant relaxation of the striated muscle of the urethra
approach would likely change to intermittent cath-
and pelvic floor and relaxation of internal and ex-
eterization during rehabilitation. After discharge
ternal sphincters. This requires integrated control
from the hospital, some people who are able to re-
from pontine centers in the brain and sacral spinal
gain an active lifestyle, even including participation
neurons (see de Groat and Yoshimura, 2001).
in sports, rely on condom drainage into a leg bag
When the bladder volume increases, sensory input
that obviates the need for strictly timed procedures
from the bladder wall to the sacral spinal neurons
such as intermittent catheterization.
increases until the threshold for the micturition re-
flex is reached and reflex voiding can be initiated.
This process in the able-bodied person is under Inadequate detrusor function
voluntary control and is accomplished by well-
regulated and integrated autonomic and somatic Inadequate contraction of the detrusor muscle
reflexes. After spinal cord injury, changes in the is often associated with spinal cord injuries that

impair the distal conus medullaris region of the
spinal cord. These very low injuries mimic lower
motor neuron impairment such as found in
peripheral neuropathies, resulting in absent or sig-
nificantly decreased detrusor contraction. There-
fore the approach to remedy this problem is to
augment emptying. Such augmentation can be in
the form of cholinergic muscarinic receptor stim-
ulation such as the oral administration of bethane-
col. Bethanecol must be taken every 4–6 h, and a
bladder response occurs 30 min to 1 h after taking
the drug. Therefore, drug use has to be timed
to coordinate with bladder fullness. Mechanical
techniques used by some include increasing intra-
abdominal pressure with external mechanical pres-
sure such as the Credé maneuver that utilizes for-
ward flexion over the subject’s hand as it presses
into the abdomen to facilitate voiding. Often emp-
tying with this procedure is incomplete, and to
prevent urinary tract infections from occurring as
a consequence of the residual urine in the bladder,
the Credé maneuver may be combined with one
catheterization per day to empty the bladder com-
pletely. This procedure is not successful in people
with detrusor–sphincter dyssynergia as it also can
cause contraction of the sphincters, blocking the
outflow of urine (Chancellor et al., 1990). The
most common approach to the management of an
inadequate detrusor response is intermittent cath-
eterization. In some cases of significantly de-
creased detrusor function, spontaneous detrusor
contractions may occur that fail to empty the
bladder but are a cause of incontinence. In such
cases, an anticholinergic (anti-muscarinic) drug
such as oxybutin and intermittent catheterization
may be combined.
Excessive detrusor function
Increased detrusor tone or spasticity (detrusor
hyperreflexia) is part of the upper motor neuron
syndrome. In this situation, the detrusor muscle is
considered to be ‘‘unstable,’’ contracting at lower
bladder volumes and often producing excessive
intravesical pressures. Detrusor hyperreflexia of-
ten, but not always, occurs with thoracic and cer-
vical cord injuries, and the extent of hyperactivity
53
of the detrusor muscle is extremely variable be-
tween individuals. Clinical approaches to treat this
problem include anticholinergic (anti-muscarinic)
medication if excessive intravesical pressures pre-
vail. If the detrusor muscle cannot be relaxed ad-
equately with such medication to provide
continence between intermittent catherizations,
an indwelling catheter or attached device such as
a condom catheter is necessary. If, due to lack of
sensation, voiding cannot be managed effectively
or conveniently, an external appliance (condom
drainage) may be used in males. However similar
devices are notoriously difficult to maintain in fe-
males, resulting, instead, in the use of an indwell-
ing Foley or suprapubic (inserted through the
lower abdominal wall) catheter. Decreasing de-
trusor contractions may also be accomplished by
chemically blocking C-fiber bladder afferent ne-
urotransmission with intravesical vanilloids such
as capsaicin or resiniferatoxin or by intravesical
administration of anticholinergics. Another in-
travesical approach under investigation is injec-
tion of botulinum toxin into the detrusor muscle to
cause relaxation (Reitz et al., 2004). Intravesical
administration of medication is more invasive than
oral medications but does offer options when oral
anticholinergic drugs are not effective. When an
intravesical route is used for treatment, effects are
temporary and repeated treatments are necessary.
The ideal time frames for repeated intravesical
drug administration are not well established. More
invasive approaches for reducing detrusor hyper-
activity include denervation procedures such as
sacral rhizotomy, a procedure that must be viewed
cautiously as it is irreversible. To address reduced
bladder capacity due to detrusor hyperreflexia, the
bladder size and capacity may be increased by a
surgical augmentation cystoplasty using a piece
of bowel.
Inadequate sphincter function
Inadequate sphincter function, whether associated
with inadequate or excessive detrusor function,
results in incontinence. Sphincter tone can be en-
hanced by blocking muscarinic cholinergic recep-
tors or by stimulating b-adrenergic receptors. Of
54
these two possible approaches, blocking mu-
scarinic cholinergic receptors is usually the supe-
rior, although combination therapy may be
utilized. In this condition, the goal is to restore
the storage function of the bladder. Once this is
accomplished, if the person cannot initiate void-
ing, then emptying is done by intermittent cathe-
terization. If continence cannot be maintained
with drugs, then voiding into an appliance such as
a condom catheter is possible, and cholinergic
muscarinic receptor agonists such as bethanecol
can be utilized to facilitate voiding. Operative ap-
proaches to enhance the usefulness of the bladder
neck in maintaining continence include surgically
modifying the bladder neck or implantation of an
artificial sphincter. The most common approach
for dealing with inadequate sphincter tone is to
enhance the contraction pharmacologically and
utilize intermittent catheterization for bladder
emptying.
Excessive sphincter tone
a-2 receptor sympathetic adrenergic blockade is
the mainstay of pharmacological management of
the contracted bladder neck, to allow emptying in
the presence of excessive sphincter tone. Originally
developed to treat hypertension, this group of
medications has evolved, through several genera-
tions, to a family of drugs that can be taken once
per day and that have infrequent and less severe
side effects such as hypotension. Other approaches
to the spastic sphincter include sphincterotomy
and pudendal nerve section.
Detrusor-sphincter dyssynergia
After spinal cord injury, the clinical presentation
of a person can be an inability to empty the blad-
der either spontaneously or by self-initiated void-
ing. With these symptoms, urodynamic studies are
required to ascertain whether the impairment
stems from inadequate detrusor function, exces-
sive sphincter activity or dyssynergy between the
two muscle groups. Approaches to treating dyssy-
nergia usually involve decreasing bladder neck re-
sistance with a drug such as an a-2 receptor
adrenergic antagonist as discussed above, and ap-
plying a device for collection of urine such as a
condom catheter. Alternatively, detrusor contrac-
tion may be blocked pharmacologically and blad-
der emptying accomplished by intermittent
catheterization. Often these approaches are only
partially successful and a compromise in bladder
management is reached within the tolerance limits
for side effects of the medication. Either the person
voids more frequently, and experiences urgency, or
catheterizes more frequently. The most common
approach in males is to reduce sphincter tone and
apply an external device.
Impaired ability to sense the bladder
At this time, 4-aminopyridine is the only pharma-
cological agent that has been demonstrated to en-
hance electrical conduction in the spinal cord,
enhancing sensation of bladder contraction and
fullness in some individuals (Potter et al., 1998). In
the absence of such sensation, management must
be accomplished by systems that continuously
drain the urine such as condom drainage, indwell-
ing catheters (suprapubic or Foley) or diapers or
methods that employ timed, regular emptying such
as intermittent catheterization.
Infection can be a consequence of all
management systems
Infection is a problem, secondary to almost all
methods of managing the neurogenic bladder after
spinal cord injury (Bennett et al., 1995; Stover
et al., 1989; Esclarin De Ruz et al., 2000). As the
prevalence of resistant bacteria increases (Waites
et al., 2000; Siroky, 2002) commonly used, inex-
pensive antibiotics become ineffective. The fre-
quency of urinary tract infections may necessitate
antibiotic prophylaxis (Galloway, 1997; Waites
et al., 2001; Morton et al., 2002). Alternative prep-
arations such as cranberry juice that contribute to
maintenance of the integrity of the bladder urot-
helium as a barrier to bacteria then become more
important considerations for long-term prophy-
laxis (Reid et al., 2003). For the cord-injured per-
son, often the most sought after management

strategy is that which, in their mind, most mimics
‘‘normal function’’ (Jamil, 2001).
Incidence and prevalence of urinary bladder
dysfunction after cord injury
Most people, during the first days after spinal cord
injury, have evidence of a neurogenic bladder. In
people with incomplete injury, the majority of re-
covery of bladder function is evident in the first
6–9 months and improvement can continue for
2 years after injury. The negative consequences of
the neurogenic bladder to the health and quality of
life for cord-injured people are decreasing with
current improvements in management and under-
standing of the causes of the problems. In one of
the early papers on urological aspects of rehabil-
itation, Bors (1951), described up to 80% mortal-
ity of spinal cord injured soldiers in World War I,
before they were able to return to the United
States. By the time of World War II the survival
rate had increased to 88%. Bors attributed this
improvement to greater understanding of the
pathophysiology of the neurogenic bladder and
the advent of antibiotics. Mortality due to uro-
logical causes is now estimated to be o3% (Jamil,
2001). The current focus of modern rehabilitation
medicine and research is directed toward issues of
morbidity and not mortality. The prevalence of a
neurogenic bladder after spinal cord injury is high
(Anson and Shepard, 1996; Noreau et al., 2000). A
study by the Model Spinal Cord Injury Systems of
Care determined that 81% of persons with spinal
cord injury reported some degree of impaired
bladder function (McKinley et al., 1999). Even
more significant are the secondary sequelae, in-
cluding frequent urinary tract infections, pain sec-
ondary to urinary tract infection and pain
secondary to indwelling devices such as Foley
catheters.
In a survey of Spinal Cord Injured persons
(Wolfe et al., 2002; Potter et al., 2004) regarding
the long-term sequelae of spinal cord injury, uro-
logical problems had a high prevalence. The ‘‘neu-
rological impairment’’ of bladder function does
not appear to change with time, but time and ag-
ing result in secondary problems such as urethral
55
strictures, bladder diverticuli, chronic cystitis and
increased incidence of bladder cancer. These prev-
alence studies reveal that, although we are well
aware of the high incidence of neurogenic bladder,
we are still limited in our ability to manage its
consequences. For example, incomplete emptying
is associated with high residual urine volume,
which is a risk factor for incontinence and
infection (Shekelle et al., 1999; Trautner and
Darouiche, 2002). Recognizing that continence is
the first issue associated with a neurogenic bladder,
pain and infection are equally important long-term
sequelae (Post et al., 1998). Although continence
may be controlled with devices, sepsis, pain and
incontinence may result from recurrent urinary
tract infections.
Conclusion
The consequences of spinal cord injury to the
function of the urinary bladder are severe and play
a serious role in the health and well-being of the
cord-injured person for life. For the bladder to be
an effective reservoir, we fully utilize reflex con-
traction of the sphincter and reflex detrusor relax-
ation. For the bladder to empty, these processes
must be reversed. After minor impairments void-
ing may still be possible but with greater effort,
incontinence, incomplete emptying, increased fre-
quency, urgency or hesitancy. Given that we often
cannot fully reverse the effects of impaired neuro-
logical control of bladder function, even with the
extensive array of available medication, the most
effective approach to the management of ne-
urogenic bladder remains to find the best balance
between a person’s need for emptying their blad-
der, their tolerance for medication, assistive de-
vices and appliances, and the social consequences
of maintaining continence (Stover et al., 1989;
Cardenas and Hooton, 1995, Liguori et al., 1997;
Yavuzer et al., 2000; Boschen et al., 2003).
Although restoration of normal function is the
ultimate goal of research, development of superior
management methods is a high priority as well.
Newer generations of pharmacological agents are
being developed to provide better therapeutic re-
sponses with less side effects. Using strategies such
56
as administering medications directly into the
bladder can minimize side effects of drugs. Meth-
ods for intermittent catheterization have been im-
proved by devices such as hydrophilic catheters
that can be inserted with much less friction than
conventional catheters (Hedlund et al., 2001;
Vapnek et al., 2003). Probiotic treatment, the in-
travesicular administration of healthy bacteria to
prevent infection, may reduce the need for antibi-
otics. Research must be directed toward finding a
cure for the bladder dysfunction after cord injury,
but also must address the important objective of
providing better management of the neurogenic
bladder.
References
Anderson, J.T. and Bradley, W.E. (1976) Bladder and urethral
innervation in multiple sclerosis. Br. J. Urol., 48: 239–243.
Anson, C.A. and Shepherd, C. (1996) Incidence of secondary
complications in spinal cord injury. Int. J. Rehab. Res., 19:
55–66.
Bennett, C.J., Young, M.N. and Darrington, H. (1995) Differ-
ences in urinary tract infections in male and female spinal
cord injury patients on intermittent catheterization. Paraple-
gia, 33: 69–72.
Bors, E. (1951) Urologic aspects of rehabilitation in SCI.
JAMA, 146: 225–229.
Boschen, K.A., Tonack, M. and Gargaro, J. (2003) Long-term
adjustment and community reintegration following spinal
cord injury. Int. J. Rehab. Res., 6(3): 157–164.
Burns, A.S., Rivas, D.A. and Ditunno, J.F. (2001). The man-
agement of the neurogenic bladder and sexual dysfunction
after spinal cord injury, Spine, 26: S129–S136.
Cardenas, D.D. and Hooton, T.M. (1995) Urinary tract infec-
tion in persons with spinal cord injury. Arch. Phys. Med.
Rehab., 76: 272–280.
Chancellor, M.B., Kaplan, S.A. and Blaivas, J.G. (1990) De-
trusor-external sphincter dyssynergia. In: Block, G. and
Whelan, J. (Eds.), Ciba Foundation Symposium. Vol. 51,
Neurobiology of Continence. Wiley, Chichester, West Sus-
sex, pp. 195–213.
Chen, D., Apple Jr., D.F., Hudson, L.M. and Bode, R. (1999)
Medical complications during acute rehabilitation following
spinal cord injury—current experience of the Model Systems.
Arch. Phys. Med. Rehab., 80(11): 1397–1401.
de Groat, W.C. and Yoshimura, N. (2001) Pharmacology of the
lower urinary tract. Annu. Rev. Pharmacol. Toxicol., 41:
691–721.
Ek, A., Alm, P., Andersson, K.E. and Persson, C.G. (1977)
Adrenergic and cholinergic nerves of the human urethra and
urinary bladder: a histochemical study. Acta. Physiol.
Scand., 99(3): 345–352.
Esclarin De Ruz, A., Garcia Leoni, E. and Herruzo Cabrera, R.
(2000) Epidemiology and risk factors for urinary tract infec-
tion in patients with spinal cord injury. J. Urol., 164:
1285–1289.
Galloway, A. (1997) Prevention of urinary tract infection in
patients with spinal cord injury—a microbiological review.
Spinal Cord, 35: 198–204.
Hedlund, H., Hjelmais, K., Jonsson, O., Klarslov, P. and
Talja, M. (2001) Hydrophilic versus non-coated catheters for
intermittent catheterization. Scand. J. Urol. Nephrol., 35:
49–53.
Jamil, F. (2001) Towards a catheter free status in neurogenic
bladder dysfunction: a review of bladder management op-
tions in spinal cord injury. Spinal Cord, 39(7): 355–361.
Liguori, P.A., Cardenas, D.D. and Ullrich, P. (1997) Social and
functional variables associated with urinary tract infections
in persons with spinal cord injury. Arch. Phys. Med. Rehab.,
78: 156–160.
Lisenemayer, T.A. and Oakley, A. (2003) Accuracy of individ-
uals with spinal cord injury at predicting urinary tract infec-
tions based on their symptoms. J. Spinal Cord Med. Winter,
26(4): 352–357.
McKinley, W.O., Jackson, A.B., Cardenas, D.D. and De Vivo,
M.J. (1999) Long-term medical complications after traumatic
spinal cord injury: a regional model systems analysis. Arch.
Phys. Med. Rehab., 80(11): 1402–1410.
Menter, R.R. and Hudson, L.M. (1995) Effects of age at injury
and the aging process. In: Stover S.L., DeLisa J.A. and
Whiteneck G.G. (Eds.), Spinal Cord Injury Clinical Out-
comes from the Model Systems. Aspen Publishers, Gait-
hersburg, MD, pp. 272–288.
Morton, S.C., Shekelle, P.G., Adams, J.L., Bennett, C.,
Dobkin, B.H., Montgomerie, J. and Vickrey, B.G. (2002)
Antimicrobial prophylaxis for urinary tract infection in per-
sons with spinal cord dysfunction. Arch. Phys. Med. Rehab.,
83: 129–138.
Noreau, L., Proulx, P., Gagnon, L., Drolet, M. and Laramee,
M.T. (2000) Secondary impairments after spinal cord injury:
a population-based study. Am. J. Phys. Med. Rehab., 79:
526–535.
Perkash, I. (2004) Donald Munro Lecture 2003. Neurogenic
bladder: past, present and future. J. Spinal Cord Med., 27:
383–386.
Post, M.W., de Witte, L.P., van Asbeck, F.W., van Dijk, A.J.
and Schrijvers, A.J. (1998) Predictors of health status and life
satisfaction in spinal cord injury. Arch. Phys. Med. Rehab.,
79(4): 395–401.
Potter, P.J., Hayes, K.C., Segal, J.L., Hsieh, J.T.C., Brunnermann,
S.R., Delaney, G.A., Tierny, D.S. and Mason, D. (1998)
Randomized double-blind crossover trial of fampridine-SR
(sustained release 4-aminopyridine) in patients with incom-
plete spinal cord injury. J. Neurotrauma, 15(10): 837–849.
Potter, P.J., Wolfe, D.L., Burkell, T.A. and Hayes, K.C. (2004)
Challenges in educating individuals with spinal cord injury to
reduce secondary conditions. Topics in Spinal Cord Injury
Rehab., 10: 30–40.

Reid, G., Potter, P.J., Lam, D., Warren, D., Borrie, M. and
Hayes, K.C. (2003) Cranberry juice to reduce Biofilms and
infection in geriatric and spinal cord injured patients with
dysfunctional bladders. Natraceuticals & Food, 8: 23–28.
Reitz, A., Stohrer, M., Kramer, G., Delpopo, G., Chartier-
Kastler, E., Panneck, J., Burgdofer, H., Gocking, K.,
Madersbacher, H., Scchumacker, S., Richter, R., von Tobel,
J. and Schurch, B. (2004) European experience of 200 cases
treated with botulinum-A toxin injections into the detrusor
muscle for urinary incontinence due to neurogenic detrusor
overactivity. Eur. Urol., 4: 510–515.
Sapounzi-Krepia, D., Soumilas, A., Papadakis, N., Sapkas, G.,
Nomicos, J., Theodossopoulou, E. and Dimitriadou, A.
(1998) Post traumatic paraplegics living in Athens: the impact
of pressure sores and UTIs on everyday life activities. Spinal
Cord, 36: 432–437.
Shekelle, P.G., Morton, S.C., Adams, J.L., Clark, K.A.,
Pathak, M. and Vickrey, B.G. (1999) Systematic review of
risk factors for urinary tract infection in adults with spinal
cord dysfunction. J. Spinal Cord Med., 22: 258–272.
Siroky, M.B. (2002) Pathogenesis of bacteriuria and infection in
the spinal cord injured patient. Am. J. Med., 113(1A):
67S–79S.
Stover, S.L., Lloyd, L.K., Waites, K.B. and Jackson, A.B.
(1989) Urinary tract infection in spinal cord injury. Arch.
Phys. Med. Rehab., 70: 47–54.
Trautner, B.W. and Darouiche, R.O. (2002) Prevention of
urinary tract infection in patients with spinal cord injury.
J. Spinal Cord Med., 25: 277–283.
57
Vapnek, J.M., Maynard, F.M. and Kim, J. (2003) A prospec-
tive randomized trial of the LoFric hydrophilic coated cath-
eter versus conventional plastic catheter for clean
intermittent catheterization. J. Urol., 169(3): 994–998.
Vines, C.L. (1996) Identifying Secondary Conditions in Arkan-
sas with Spinal Cord Injuries. Arkansas Spinal Cord Com-
mission, Little Rock, AR.
Waites, K., Chen, Y., De Vivo, M.J., Canupp, K.C. and Moser,
S.A. (2000) Antimicrobial resistance in gram negative bacte-
ria isolated from the urinary tract in community-residing
persons with spinal cord injury. Arch. Phys. Med. Rehab.,
81: 764–769.
Waites, K.B., Canupp, K.C., Chen, Y., De Vivo, M.J. and
Moser, S.A. (2001) Bacteremia after spinal cord injury in
initial versus subsequent hospitalizations. J. Spinal Cord
Med., 24(2): 96–100.
Waites, K.B., Canupp, K.C. and De Vivo, M.J. (1993)
Epidemiology and risk factors for urinary tract infection
following spinal cord injury. Arch. Phys. Med. Rehab., 74(7):
691–695.
Wolfe, D.L., Potter, P.J., Jutai, J.W., Beggs, C., Colwell, S.,
Orenczuk, S., Regan, M.A., Tonack, M., Trujillo, S. and
Zarnowiecki, E. (2002) Long-term physical impact of spinal
cord injury: the role of rehabilitation in educating consumers
to reduce complications over the long-term. J. Spinal Cord
Med., 22(Suppl. 1): S50.
Yavuzer, G., Gok, H., Tuncer, S., Soygur, T., Arikan, N. and
Arasil, T. (2000) Compliance with bladder management in
spinal cord injury patients. Spinal Cord, 38(12): 762–765.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 5

Mechanisms underlying the recovery of lower urinary

tract function following spinal cord injury

William C. de Groat and Naoki Yoshimura

Departments of Pharmacology and Urology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA

Abstract: The lower urinary tract has two main functions, the storage and periodic expulsion of urine,
which are regulated by a complex neural control system in the brain and lumbosacral spinal cord. This
neural system coordinates the activity of two functional units in the lower urinary tract: (1) a reservoir (the
urinary bladder) and (2) an outlet (consisting of bladder neck, urethra and striated muscles of the pelvic
floor). During urine storage the outlet is closed and the bladder is quiescent, thereby maintaining a low
intravesical pressure over a wide range of bladder volumes. During micturition the outlet relaxes and the
bladder contracts to promote the release of urine. This reciprocal relationship between bladder and outlet is
generated by visceral reflex circuits, some of which are under voluntary control. Experimental studies in
animals indicate that the micturition reflex is mediated by a spinobulbospinal pathway passing through a
coordination center (the pontine micturition center) located in the rostral brainstem. This reflex pathway is
in turn modulated by higher centers in the cerebral cortex that are presumably involved in the voluntary
control of micturition. Spinal cord injury at cervical or thoracic levels disrupts voluntary control of voiding
as well as the normal reflex pathways that coordinate bladder and sphincter functions. Following spinal cord
injury, the bladder is initially areflexic but then becomes hyperreflexic due to the emergence of a spinal
micturition reflex pathway. Studies in animals indicate that the recovery of bladder function after spinal
cord injury is dependent in part on plasticity of bladder afferent pathways and the unmasking of reflexes
triggered by capsaicin-sensitive C-fiber bladder afferent neurons. The plasticity is associated with changes in
the properties of ion channels and electrical excitability of afferent neurons, and appears to be mediated in
part by neurotrophic factors released in the spinal cord and the peripheral target organs.

Introduction certain level of function even after elimination of

The functions of the lower urinary tract to store
and periodically release urine are dependent upon
neural circuits located in the brain, spinal cord and
peripheral ganglia (Barrington, 1925; Kuru, 1965;
de Groat et al., 1993; Morrison et al., 2002).
This dependence on central nervous control
distinguishes the lower urinary tract from many
other visceral structures (e.g., the gastrointestinal
tract and cardiovascular system) that maintain a
Corresponding author. Tel.: +412 648 9357;
Fax: +412 648 1945; E-mail: degroat@server.pharm.pitt.edu
extrinsic neural input.
The lower urinary tract is also unusual with
regard to its pattern of activity and the complexity
of its neural regulation. For example, the urinary
bladder has two principal modes of operation:
storage and elimination. Thus many of the neural
circuits exhibit switch-like or phasic patterns of
activity (de Groat, 1975) in contrast to tonic
patterns occurring in autonomic pathways to
cardiovascular organs. In addition, micturition is
under voluntary control and depends upon learned
behavior that develops during maturation of the
nervous system, whereas many other visceral

DOI: 10.1016/S0079-6123(05)52005-3 59
60
functions are regulated involuntarily. Micturition
also depends on the integration of autonomic and
somatic efferent mechanisms within the lumbosacral
spinal cord (Chancellor and Yoshimura, 2002;
Morrison et al., 2002). This is necessary during
urine storage and elimination to coordinate the
activity of visceral organs (the bladder and
urethra) with that of urethral striated muscles.
The dependence of lower urinary tract functions
on complex central neural networks renders these
functions susceptible to a variety of neurological
disorders (Torrens and Morrison, 1987; Chancellor
and Yoshimura, 2002). This chapter will review
studies in animals and humans that have provided
insights into the neural control of the lower
urinary tract and the disruption of this control
by spinal cord injury.
Anatomy and innervation
The storage and periodic elimination of urine are
regulated by the activity of two functional units in
the lower urinary tract: (1) a reservoir (the bladder)
and (2) an outlet (consisting of bladder neck,
urethra and striated muscles of the pelvic floor).
Under normal conditions, the urinary bladder and
outlet exhibit a reciprocal relationship. During
urine storage, the bladder neck and proximal
urethra are closed; and the bladder smooth muscle
is quiescent, allowing intravesical pressure to
remain low over a wide range of bladder volumes.
During voluntary micturition, the initial event is a
reduction of intraurethral pressure, which reflects
a relaxation of the pelvic floor and the periurethral
striated muscles, and an opening of the bladder
neck (Chancellor and Yoshimura, 2002). The
changes in the urethra are followed in a few
seconds by a detrusor contraction and a rise in
intravesical pressure that is maintained until the
bladder empties. Reflex inhibition of the smooth
and striated muscles of the urethra also contrib-
utes to the reduction of outlet resistance during
micturition. These changes are coordinated by three
sets of nerves emerging from the thoracolumbar
and sacral levels of the spinal cord: (1) sacral
parasympathetic (pelvic nerves), (2) sacral somatic
(pudendal nerves) and (3) thoracolumbar
sympathetic (hypogastric nerves and sympathetic
chain) (Fig. 1) (de Groat et al., 1993).
Parasympathetic pathways
The sacral, parasympathetic, efferent pathway
provides the major excitatory input to the blad-
der and consists of spinal preganglionic neurons
(Morgan et al., 1979) with cell bodies, in most of
the species studied, situated in the gray matter of
the sacral spinal segments. In some species, for
example, the rat, the cell bodies span the caudal
lumbar and rostral sacral segments. Regardless of
the actual location of the cell bodies, this group of
neurons will be called the ‘‘sacral parasympathetic
nucleus’’ in this chapter. These neurons send axons
to peripheral ganglion cells that, in turn, innervate
the bladder and urethral smooth muscle (Fig. 1).
Transmission in bladder ganglia is mediated by
acetylcholine that excites the ganglion cells by act-
ing on nicotinic cholinergic receptors, whereas
parasympathetic neuroeffector transmission in the
bladder is mediated by acetylcholine acting on
muscarinic receptors (de Groat and Yoshimura,
2001; Andersson and Arner, 2004). Both M2 and
M3 muscarinic receptor subtypes are expressed in
bladder smooth muscle; however, examination of
subtype-selective muscarinic receptor antagonists
and studies of muscarinic receptor knockout mice
have revealed that the M3 subtype is the principal
receptor involved in excitatory transmission.
In bladders of various animals, stimulation
of parasympathetic nerves also produces a
non-cholinergic contraction that is resistant to
atropine and other muscarinic receptor blocking
agents. Adenosine triphosphate (ATP) has been
identified as the excitatory transmitter mediating
the non-cholinergic contraction (Ralevic and
Burnstock, 1998; Burnstock, 2001). ATP excites
the bladder smooth muscle by acting on P2X
purinergic receptors that are ligand-gated ion
channels. Among the seven types of P2X recep-
tors that have been identified, P2X1 is the major
subtype expressed in the rat and also in the human
bladder smooth muscle. Although purinergic
excitatory transmission is not important in the
normal human bladder, it appears to be involved
 61

Fig. 1. Diagram showing the sympathetic, parasympathetic and somatic innervation of the urogenital tract of the male cat. Sym-
pathetic preganglionic pathways emerge from the lumbar spinal cord and pass to the sympathetic chain ganglia (SCG) and then via the
inferior splanchnic nerves (ISN) to the inferior mesenteric ganglia (IMG). Preganglionic and postganglionic sympathetic axons then
travel in the hypogastric nerve (HGN) to the pelvic plexus and the urogenital organs. Parasympathetic preganglionic axons that
originate in the sacral spinal cord pass in the pelvic nerve to ganglion cells in the pelvic plexus and to distal ganglia in the organs. Sacral
somatic pathways are contained in the pudendal nerve, which provides an innervation to the penis, the ischiocavernosus (IC),
bulbocavernosus (BC) and external urethral sphincter (EUS) muscles. The pudendal and pelvic nerves also receive postganglionic
axons from the caudal SCG. These three sets of nerves contain afferent axons from the lumbosacral dorsal root ganglia. Abbreviations:
U, ureter; PG, prostate gland; VD, vas deferens.

in bladders in patients with pathological condi- receptor antagonists or by desensitization of P2X

tions such as chronic urethral outlet obstruction or
interstitial cystitis (Palea et al., 1993; O’Reilly
et al., 2002).
Parasympathetic pathways to the urethra induce
relaxation during voiding. In various species the
relaxation is not affected by muscarinic antago-
nists and therefore is not mediated by acetylcho-
line. However inhibitors of nitric oxide synthase
block the relaxation in vivo during reflex voiding
or block the relaxation of urethral smooth muscle
strips induced in vitro by electrical stimulation of
intramural nerves indicating that nitric oxide is the
inhibitory transmitter involved in relaxation
(Burnett et al., 1997; Ho et al., 1999; Morrison
et al., 2002). In some species, neurally evoked con-
tractions of the urethra are reduced by muscarinic
purinergic receptors, indicating that acetylcholine
or ATP is involved in excitatory transmission to
urethral smooth muscle (Zoubek et al., 1993).
Sympathetic pathways
Sympathetic preganglionic pathways that arise
from the 11th thoracic (T11) to 2nd lumbar (L2)
spinal segments pass to the sympathetic chain
ganglia (SCG) and then to prevertebral ganglia in
the superior hypogastric and pelvic plexuses
(Fig. 1) and also to short adrenergic neurons in
the bladder and urethra (de Groat et al., 1993;
Delancey et al., 2002). Sympathetic postganglionic
nerves that release norepinephrine provide an
excitatory input to smooth muscle of the urethra
62
and bladder base, an inhibitory input to smooth
muscle in the bladder body as well as inhibitory
and facilitatory inputs to vesical parasympathetic
ganglia (Andersson, 1993; de Groat and Booth,
1993). Radioligand receptor binding studies
showed that a-adrenergic receptors are concen-
trated in the bladder base and proximal urethra,
whereas b-adrenergic receptors are most promi-
nent in the bladder body (Andersson, 1993). These
observations are consistent with pharmacological
studies showing that sympathetic nerve stimula-
tion or exogenous catecholamines produce b-
adrenergic receptor-mediated inhibition of the
body and a-adrenergic receptor-mediated contrac-
tion of the base, dome and urethra. Molecular and
contractility studies have shown that b3-adrenergic
receptors elicit inhibition and a1-adrenergic recep-
tors elicit contractions. The a1A-adrenergic recep-
tor subtype is most prominent in the normal
bladders but the a1D-subtype is upregulated in
bladders from patients with outlet obstruction,
raising the possibility that a1-adrenergic receptor
excitatory mechanisms in the bladder might con-
tribute to irritative lower urinary tract symptoms
in patients with obstruction (de Groat and
Yoshimura, 2001; Morrison et al., 2002).
Somatic pathways
The external urethral sphincter, which is composed
of striated muscle, receives a somatic cholinergic
innervation via the pudendal nerve from anterior
horn cells in the third and fourth sacral segments
(Fig. 1). Branches of the pudendal nerve and other
sacral somatic nerves also carry efferent impulses
to muscles of the pelvic floor and proprioceptive
afferent signals from these muscles as well as sen-
sory information from the urethra. Analysis of
urethral closure mechanisms in the female rat dur-
ing bladder distension-evoked and sneeze-induced
stress conditions revealed that the major rise in
urethral pressure occurred in the mid-urethra and
was mediated by efferent pathways in the puden-
dal nerve to the external urethral sphincter as well
as pathways in nerves to the iliococcygeus and
pubococcygeus muscles, but not by pathways in
the sympathetic or parasympathetic nerves (Kamo
et al., 2003, 2004). Studies of the biomechanical
properties of the intact female rat urethra in vitro
have confirmed the large contribution of striated
muscle activity and nicotinic receptor mechanisms
to the contractions of the mid-urethra (Jankowski
et al., 2004).
Afferent pathways
Afferent axons innervating the urinary tract are
present in the three sets of nerves that innervate
the lower urinary tract (Janig and Morrison, 1986;
de Groat et al., 1993; Bahns et al., 1998; Morrison
et al., 2002). The most important afferents for in-
itiating micturition are those passing through the
pelvic nerve to the sacral spinal cord. These affer-
ents are small myelinated (Ad) and unmyelinated
(C) fibers that convey information from receptors
in the bladder wall to second-order neurons in the
spinal cord. Ad bladder afferents in the cat re-
spond in a graded manner to passive distension as
well as active contraction of the bladder and ex-
hibit intravesical pressure thresholds in the range
of 5–15 mmHg, similar to the pressures at which
humans report the first sensation of bladder filling
(Chancellor and Yoshimura, 2002; Morrison et al.,
2002). These fibers also code for noxious stimuli in
the bladder. On the other hand, C-fiber bladder
afferents in the cat have high thresholds and com-
monly do not respond to even high levels of in-
travesical pressure (Habler et al., 1990). However,
activity in some of these afferents is unmasked or
enhanced by chemical irritation of the bladder
mucosa. These findings indicate that C-fiber affer-
ents in the cat have specialized functions, such as
the signaling of inflammatory or noxious events in
the lower urinary tract. Nociceptive and me-
chanoceptive information is also carried in the
hypogastric nerves to the thoracolumbar segments
of the spinal cord (Bahns et al., 1998).
In rats, A- and C-fiber bladder afferents are not
distinguishable on the basis of stimulus modality;
thus both types of afferents consist of mechano-
and chemo-sensitive populations (Sengupta and
Gebhart, 1994; Morrison et al., 1999; Shea et al.,
2000; Rong et al., 2002). C-fiber afferents that
respond only to bladder filling have also been

identified in the rat bladder and appear to be
volume receptors possibly sensitive to stretch of
the mucosa. C-fiber afferents are sensitive to the
neurotoxins, capsaicin and resiniferatoxin as well
as to other substances such as tachykinins, nitric
oxide, ATP, prostaglandins and neurotrophic fac-
tors released into the bladder by afferent nerves,
urothelial cells and inflammatory cells (Vizzard
et al., 1995; Lee et al., 2000; Chuang et al., 2001;
Yoshimura et al., 2001a; Morrison et al., 2002).
These substances can sensitize the afferent nerves
and change their response to mechanical stimuli.
The properties of lumbosacral dorsal root gan-
glion cells innervating the bladder, urethra and
external urethral sphincter in the rat have been
studied with patch clamp recording techniques in
combination with axonal tracing methods to
identify the different populations of neurons
(Yoshimura et al., 1996, 2001b, 2003; Yoshimura
and de Groat, 1997, 1999; Black et al., 2003).
Based on responsiveness to capsaicin it is estimated
that approximately 70% of bladder afferent
neurons in the rat are of the C-fiber type. These
neurons exhibit high threshold, tetrodotoxin-
resistant sodium channels and action potentials.
They show phasic firing (one to two spikes) in re-
sponse to prolonged depolarizing current pulses.
Approximately 90% of the bladder C-fiber afferent
neurons also are excited by ATP, which induces a
depolarization and firing by activating P2X3 or
P2X2/3 receptors (Zhong et al., 2003). These
neurons express isolectin-B4 binding, which is
commonly used as a marker for ATP responsive
sensory neurons. A-fiber afferent neurons are
resistant to capsaicin and ATP, and exhibit low
threshold tetrodotoxin-sensitive sodium channels
and action potentials and tonic firing (multiple
spikes) to depolarizing current pulses. C-fiber
bladder afferent neurons also express a slowly
decaying A-type K+ current that controls spike
threshold and firing frequency (Yoshimura et al.,
1996, 2003). Suppression of this K+ current induces
hyperexcitability of bladder afferent neurons. These
properties of dorsal root ganglion cells are
consistent with the different properties of A- and
C-fiber afferent receptors in the bladder.
Immunohistochemical studies have shown that a
large percentage of bladder afferent neurons
63
contain peptides: calcitonin gene-related peptide,
vasoactive intestinal polypeptide, pituitary-adenyl
cyclase activating polypeptide (PACAP), tachy-
kinins, galanin and opioid peptides (de Groat,
1987; Keast and de Groat, 1992; Maggi, 1993;
Morrison et al., 2002). Nerves containing these
peptides are common in the bladder, in the sub-
mucosal and epithelial layers, and around blood
vessels (de Groat and Yoshimura, 2001). In the
spinal cord, peptidergic nerve terminals have a
distribution very similar to the distribution of
pelvic nerve afferents labeled with horseradish
peroxidase (Kawatani et al., 1985; de Groat, 1987).
Peptidergic bladder afferent neurons in the rat also
express TrkA, a high-affinity receptor for nerve
growth factor (NGF) (McMahon et al., 1994) and
receptors for capsaicin (TRPV1) (Yoshimura
et al., 1996, 1998a, 2003) and tachykinins (NK-2
and NK-3 receptors) (Morrison et al., 2002;
Sculptoreanu and de Groat, 2003). Capsaicin, a
neurotoxin that can release peptides from afferent
terminals, produces inflammatory responses, in-
cluding plasma extravasation and vasodilatation,
when applied locally to the bladder in experimen-
tal animals (Maggi, 1993). These findings suggest
that the neuropeptides may be important trans-
mitters in the afferent pathways from the lower
urinary tract. Tachykinins may also act back on
afferent terminals in an auto-feedback manner
to modulate the excitability of the terminals
(Morrison et al., 2002; Sculptoreanu and de Groat,
2003).
Urothelial– afferent interactions
Recent studies have revealed that the urothelium,
which has been traditionally viewed as a passive
barrier at the bladder luminal surface (Lavelle
et al., 2000; Lewis, 2000), also has specialized
sensory and signaling properties that allow
urothelial cells to respond to their chemical and
physical environment and to engage in reciprocal
chemical communication with neighboring nerves
in the bladder wall (Ferguson et al., 1997; Birder
et al., 1998, 2001, 2002, 2003; Cockayne et al.,
2000). These properties include: (1) expression of
nicotinic, muscarinic, tachykinin, adrenergic and
capsaicin (TRPV1) receptors, (2) responsiveness to
64

transmitters released from sensory nerves, (3) close

physical association with afferent nerves and (4)
ability to release chemical mediators such as ATP
and nitric oxide that can regulate the activity of
adjacent nerves and thereby trigger local vascular
changes and/or reflex bladder contractions. The
role of ATP in urothelial–afferent communication
has attracted considerable attention because blad-
der distension releases ATP from the urothelium
and intravesical administration of ATP induces
bladder hyperactivity, an effect blocked by admin-
istration of P2X purinergic receptor antagonists
that suppress the excitatory action of ATP on
bladder afferent neurons (Morrison et al., 2002).
Mice in which the P2X3 receptor was knocked out
exhibited hypoactive bladder activity and ineffi-
cient voiding (Cockayne et al., 2000), suggesting
that activation of P2X3 receptors on bladder
afferent nerves by ATP released from the urothe-
lium is essential for normal bladder function. The
chapter by Birder (this volume) presents a more
detailed discussion of urothelial–afferent interac-
tions and changes in the urothelium after spinal
Fig. 2. Combined cystometrograms and sphincter electromyo-
cord injury. grams (EMG) comparing reflex voiding responses in an infant
(A) and in a paraplegic subject (C) with a voluntary voiding
Reflex mechanisms controlling the lower response in an able-bodied adult (B). The abscissa in all records
urinary tract represents bladder volume in milliliters and the ordinates rep-
resent bladder pressure in cm H2O and electrical activity of the
EMG recording. On the left side of each trace the arrows in-
The neural pathways controlling lower urinary dicate the start of a slow infusion of fluid into the bladder
tract function are organized as simple on–off (bladder filling). Vertical dashed lines indicate the start of
switching circuits that maintain a reciprocal rela- sphincter relaxation that precedes by a few seconds the bladder
tionship between the urinary bladder and urethral contraction in A and B. (B) Note that a voluntary cessation of
voiding (stop) is associated with an initial increase in sphincter
outlet (de Groat et al., 1981, 1993) (Fig. 2). The
EMG followed by a reciprocal relaxation of the bladder. A
principal reflex components of these switching resumption of voiding is again associated with sphincter relax-
circuits are listed in Table 1 and illustrated in ation and a delayed increase in bladder pressure. On the other
Fig. 3. Intravesical pressure measurements during hand, in the paraplegic subject (C) the reciprocal relationship
bladder filling in both humans and animals reveal between bladder and sphincter is abolished. During bladder
filling, transient uninhibited bladder contractions occur in as-
low and slowly increasing bladder pressures when
sociation with sphincter activity. Further filling leads to more
bladder volume is below the threshold for inducing prolonged and simultaneous contractions of the bladder and
voiding (Fig. 2). The accommodation of the blad- sphincter (bladder-sphincter dyssynergia). Loss of the recipro-
der to increasing volumes of urine is primarily a cal relationship between bladder and sphincter in paraplegic
passive phenomenon dependent upon the intrinsic people interferes with bladder emptying.
properties of the vesical smooth muscle and qui-
escence of the parasympathetic efferent pathway. urethra (Table 1, Fig. 3) (de Groat and Lalley,
In addition, in some species urine storage is also 1972; de Groat et al., 1981). During bladder filling
facilitated by sympathetic reflexes that mediate the activity of the sphincter electromyogram
an inhibition of bladder activity, closure of the (EMG) increases (Fig. 2), reflecting an increase
bladder neck and contraction of the proximal in efferent firing in the pudendal nerve and an

Table 1. Reflexes to the lower urinary tract
Afferent Pathway
Low-level vesical afferent activity (pelvic nerve)
High-level vesical afferent activity (pelvic nerve)
a
S-B-S reflex, spinobulbospinal reflex
increase in outlet resistance that contributes to the
maintenance of urinary continence.
The storage phase of the urinary bladder can be
switched to the voiding phase either involuntarily
(reflexly) or voluntarily (Fig. 2). The reflex switch
is readily demonstrated in the human infant
(Fig. 2A) or in the anesthetized animal when the
volume of urine exceeds the micturition threshold.
At this point, increased afferent firing from tension
receptors in the bladder reverses the pattern of
efferent outflow, producing firing in the sacral
parasympathetic pathways and inhibition of
sympathetic and somatic pathways. The expulsion
phase consists of an initial relaxation of the
urethral sphincter (Fig. 2) followed in a few
seconds by a contraction of the bladder, and an
increase in bladder pressure and flow of urine.
Relaxation of the urethral outlet is mediated by
activation of a parasympathetic reflex pathway to
the urethra that triggers the release of nitric oxide,
an inhibitory transmitter, as well as by removal of
adrenergic and somatic cholinergic excitatory
inputs to the urethra. Secondary reflexes elicited
by flow of urine through the urethra facilitate
bladder emptying.
The reflex circuitry controlling micturition
consists of four basic components: spinal efferent
neurons, spinal interneurons, primary afferent
neurons and neurons in the brain that modulate
spinal reflex pathways. Transneuronal virus
65
Efferent Pathway Central Pathway
Urine storage
1. External sphincter contraction (somatic nerves) Spinal reflexes
2. Internal sphincter contraction (sympathetic nerves)
3. Detrusor inhibition (sympathetic nerves)
4. Ganglionic inhibition (sympathetic nerves)
5. Sacral parasympathetic outflow inactive
Micturition
1. Inhibition of external sphincter activity S-B-S reflexa
2. Inhibition of sympathetic outflow S-B-S reflex
3. Activation of parasympathetic outflow to the bladder S-B-S reflex
4. Activation of parasympathetic outflow to the urethra Spinal reflex
tracing, measurements of gene expression and
patch-clamp recording in spinal cord slice prepa-
rations have recently provided many new insights
into the morphological and electrophysiological
properties of these reflex components. Neurotropic
viruses, such as pseudorabies virus, have been
particularly useful since they can be injected into a
target organ (urinary bladder, urethra and urethral
sphincter) and then move intraaxonally from the
periphery to the central nervous system, where
they replicate and then pass retrogradely across
synapses to infect second- and third-order neurons
in the neural pathways (Vizzard et al., 1995;
Nadelhaft and Vera, 1996; Sugaya et al., 1997).
Since pseudorabies virus can be transported across
many synapses, it could sequentially infect all of
the neurons that connect directly or indirectly to
the lower urinary tract (Fig. 4).
Anatomy of the spinal cord
The spinal cord gray matter is divided into three
general regions: (1) the dorsal horn that contains
interneurons that process sensory input, (2) the
ventral horn that contains motoneurons and (3)
the intermediate region, located between the dorsal
and ventral horns, that contains interneurons and
autonomic preganglionic neurons. These regions
are further subdivided into layers or laminae that
are numbered, starting with the superficial layer of
66

Fig. 3. Diagram showing neural circuits controlling continence and micturition. (A) Urine storage reflexes. During the storage of
urine, distention of the bladder produces low level vesical afferent firing, which in turn stimulates (1) the sympathetic outflow to the
bladder outlet (base and urethra) and (2) pudendal outflow to the external urethral sphincter. These responses occur by spinal reflex
pathways and represent guarding reflexes, which promote continence. Sympathetic firing also inhibits the detrusor muscle and mod-
ulates transmission in bladder ganglia. A region in the rostral pons (the pontine storage center) increases external urethral sphincter
activity. (B) Voiding reflexes. During elimination of urine, intense bladder afferent firing activates spinobulbospinal reflex pathways
passing through the pontine micturition center, which stimulate the parasympathetic outflow to the bladder and internal sphincter
smooth muscle and inhibit the sympathetic and pudendal outflow to the urethral outlet. Ascending afferent input from the spinal cord
may pass through relay neurons in the periaqueductal gray (PAG) before reaching the pontine micturition center.

the dorsal horn (lamina I) and extending to the
ventral horn (lamina IX), and the commissure
connecting the two sides of the spinal cord (lamina
X) (Fig. 5D).
Efferent pathways
Parasympathetic preganglionic neurons innervat-
ing the lower urinary tract are located in the
intermediolateral gray matter (laminae V–VII) in
the sacral (in the rat, also caudal lumbar) segments
of the spinal cord (Fig. 4) (Morgan et al., 1981; de
Groat et al., 1982; Araki and de Groat, 1997;
Miura et al., 2001a), whereas sympathetic pre-
ganglionic neurons are located in both medial
(lamina X) and lateral sites (laminae V–VII) in the
intermediate gray matter of the rostral lumbar
spinal cord. Parasympathetic preganglionic
 67

Fig. 4. Transneuronal virus tracing of the central pathways controlling the urinary bladder of the rat. Injection of pseudorabies virus
into the wall of the urinary bladder leads to retrograde transport of virus (dashed arrows) and sequential infection of postganglionic
neurons, preganglionic neurons and then various central neural circuits synaptically linked to the preganglionic neurons. Normal
synaptic connections are indicated by solid arrows. At long survival times virus can be detected with immunocytochemical techniques
in neurons at specific sites throughout the spinal cord and brain extending to the pontine micturition center in the pons (i.e.,
Barrington’s nucleus) and to the cerebral cortex. Other sites in the brain labeled by virus are: (1) the paraventricular nucleus (PVN),
medial preoptic area (MPOA) and periventricular nucleus (Peri V.N.) of the hypothalamus, (2) periaqueductal gray (PAG), (3) locus
coeruleus (LC) and subcoeruleus, (4) red nucleus, (5) medullary raphe nuclei and (6) the noradrenergic cell group designated A5. Sixth
lumbar (L6) spinal cord section showing on the left side the distribution of virus labeled parasympathetic preganglionic neurons (&)
and interneurons (K) in the region of the parasympathetic nucleus, the dorsal commissure (DCM) and the superficial laminae of the
dorsal horn (DH), 72 h after injection of the virus into the bladder. The right side shows the entire population of preganglionic neurons
(PGN)(&) labeled by axonal tracing with the fluorescent dye (Fluorogold), injected into the pelvic ganglia and the distribution of
virus-labeled bladder PGN (’). Composite diagram of neurons in 12 spinal sections (42 mm each).

neurons send dendrites to discrete regions of the
spinal cord including: (1) the lateral and dorsal
lateral funiculus, (2) lamina I on the lateral edge of
the dorsal horn, (3) the dorsal gray commissure
(lamina X) and (4) gray matter and lateral fun-
iculus ventral to the autonomic nucleus (Morgan
et al., 1993). As discussed below, this dendritic
structure very likely indicates the origin of impor-
tant synaptic inputs to these cells. Pudendal moto-
neurons innervating the external urethral sphincter
in the cat are located in the ventrolateral division
of Onuf’s nucleus and send dendritic projections
into (1) the lateral funiculus, (2) lamina X (3) in-
termediolateral gray matter and (4) rostrocaudally
within the nucleus (Thor et al., 1989; Beattie et al.,
1990; Sasaki, 1994). This dendritic distribution of
sphincter motoneurons is similar to that of sacral
preganglionic neurons indicating that these two
populations of neurons may receive synaptic
inputs from the same interneuronal sites and
fiber tracts in the spinal cord.
Afferent projections in the spinal cord
Afferent pathways from the lower urinary tract
project to discrete regions of the dorsal horn that
contain the interneurons as well as the soma and/
or dendrites of efferent neurons innervating the
lower urinary tract. Pelvic nerve afferent pathways
from the urinary bladder of the cat and rat project
68

Fig. 5. Comparison of the distribution of bladder afferent projections to the 6th lumbar (L6) spinal cord of the rat (A) with the
distribution of c-fos positive cells in the L6 spinal segment following chemical irritation of the lower urinary tract of the rat (B) and the
distribution of interneurons in the L6 spinal cord labeled by transneuronal transport of pseudorabies virus injected into the urinary
bladder (C). Afferents labeled by wheatgerm agglutinin-horseradish peroxidase (WGA-HRP) injected into the urinary bladder. C-fos
immunoreactivity is present in the nuclei of cells. DH, dorsal horn; SPN, sacral parasympathetic nucleus; CC central canal. (D)
Drawing shows the laminar organization of the cat spinal cord.

into Lissauer’s tract at the apex of the dorsal horn
and then pass rostrocaudally giving off collaterals
that extend laterally and medially through the su-
perficial layer of the dorsal horn (lamina I) into the
deeper layers (laminae V–VII and X) at the base of
the dorsal horn (Fig. 5A) (Morgan et al., 1981;
Steers et al., 1991). The lateral pathway, which is
the most prominent projection, terminates in the
region of the sacral parasympathetic nucleus
(SPN) and also sends some axons to the dorsal
commissure (Fig. 5A). Pudendal afferent pathways
from the urethra and urethral sphincter exhibit a
similar pattern of termination in the sacral spinal
cord (Thor et al., 1989). The overlap of bladder
and urethral afferents in the lateral dorsal horn
and dorsal commissure indicates these regions are
likely to be important sites of viscerosomatic in-
tegration and be involved in coordinating bladder
and sphincter activity.
Spinal interneurons
As shown in Fig. 5C, interneurons retrogradely
labeled by injection of pseudorabies virus into the
urinary bladder of the rat are located in the regions
of the spinal cord receiving afferent input from the
bladder (Nadelhaft and Vera, 1995; Sugaya et al.,
1997). Interneuronal locations also overlap in
many sites with the dendritic distribution of the
efferent neurons. A similar distribution of labeled
interneurons has been noted following injections
of virus into the urethra (Vizzard et al., 1995) or
the external urethral sphincter, indicating a prom-
inent overlap of the interneuronal pathways con-
trolling the various target organs of the lower
urinary tract.
The spinal neurons involved in processing af-
ferent input from the lower urinary tract have been
identified by the expression of the immediate early

gene, c-fos (Fig. 5B). In the rat, noxious or non-
noxious stimulation of the bladder and urethra
increases the levels of Fos protein, primarily in the
dorsal commissure, the superficial dorsal horn and
in the area of the sacral parasympathetic nucleus
(Fig. 5B) (Birder and de Groat, 1993; Birder et al.,
1999). Some of these interneurons send long pro-
jections to the brain, whereas others make local
connections in the spinal cord and participate in
segmental spinal reflexes.
Patch clamp recordings from parasympathetic
preganglionic neurons in the neonatal rat spinal
slice preparation have revealed that interneurons
located immediately dorsal and medial to the
parasympathetic nucleus make direct monosynap-
tic connections with the preganglionic neurons
(Araki, 1994; Araki and de Groat, 1996, 1997).
Microstimulation of interneurons in both loca-
tions elicits glutamatergic, N-methyl-D-aspartic
acid (NMDA) and non-NMDA excitatory postsy-
naptic currents in the preganglionic neurons. Stim-
ulation of a subpopulation of medial interneurons
elicits gamma-amino butyric acid (GABA)ergic
and glycinergic inhibitory postsynaptic currents.
Thus local interneurons are likely to play an im-
portant role in both excitatory and inhibitory re-
flex pathways controlling the preganglionic
outflow to the lower urinary tract. Glutamatergic
excitatory currents have also been elicited in pre-
ganglionic neurons by stimulation of the projec-
tions from lamina X and the lateral funiculus
(Miura et al., 2001a, 2003).
Pathways in the brain
The neurons in the brain that control the lower
urinary tract have been studied with a variety of
anatomical tracing techniques in several species
(Morrison et al., 2002). In the rat, transneuronal
virus tracing methods have identified many popu-
lations of neurons that are involved in the control
of bladder, urethra and the urethral sphincter
including: Barrington’s nucleus (the pontine mi-
cturition center), medullary raphe nucleus which
contains serotonergic neurons, the locus coeruleus
which contains noradrenergic neurons, per-
iaqueductal gray and the A5 noradrenergic cell
69
group (Fig. 4). Several regions in the hypothalamus
and the cerebral cortex also exhibited virus-infected
cells. Neurons in the cortex were located primarily
in the medial frontal cortex. Other anatomical
studies in which anterograde tracer substances
were injected into brain areas and then identified
in terminals in the spinal cord are consistent with
the virus tracing data (Morrison et al., 2002).
Organization of urine storage and voiding reflexes
Sympathetic storage pathway
The integrity of the sympathetic input to the lower
urinary tract is not essential for the performance of
micturition (Torrens and Morrison, 1986; de
Groat et al., 1993). However, physiologic experi-
ments in animals indicate that during bladder
filling, the sympathetic system does provide a tonic
inhibitory input to the bladder as well as an
excitatory input to the urethra. This sympathetic
input is physiologically significant since surgical
interruption or pharmacologic blockade of the
sympathetic innervation can reduce urethral
outflow resistance, reduce bladder capacity and
increase the frequency and amplitude of bladder
contractions recorded under constant volume
conditions.
Sympathetic reflex activity is elicited by a
sacrolumbar, intersegmental spinal reflex pathway
that is triggered by vesical afferent activity in the
pelvic nerves (Fig. 3A, Table 1) (de Groat and
Lalley, 1972). The reflex pathway is inhibited when
bladder pressure is raised to the threshold for
producing micturition. This inhibitory response is
abolished by transection of the spinal cord at the
lower thoracic level, indicating that it originates at
a supraspinal site, possibly the pontine micturition
center. Thus, the vesicosympathetic reflex repre-
sents a negative feedback mechanism, whereby an
increase in bladder pressure tends to increase in-
hibitory input to vesical ganglia and smooth mus-
cle, thus allowing the bladder to accommodate
large volumes (Fig. 3A). Increased sympathetic
excitatory input to the bladder base and urethra
would complement these mechanisms by increas-
ing outflow resistance.
70
Urethral sphincter storage pathway
Motoneurons innervating the striated muscles of
the urethral sphincter exhibit a tonic discharge
that increases during bladder filling. This activity
is mediated in part by low-level afferent input from
the bladder (Fig. 3A, Table 1). During micturition
the firing of sphincter motoneurons is inhibited.
This inhibition is dependent in part on supraspinal
mechanisms since it is not as prominent in chronic
spinal animals. Electrical stimulation of the pon-
tine micturition center induces sphincter relaxation
suggesting that bulbospinal pathways from the
pons may be responsible for maintaining the nor-
mal reciprocal relationship between bladder and
sphincter (Holstege et al., 1986; Shefchyk, 1989;
Mallory et al., 1991).
Spinobulbospinal parasympathetic micturition
pathway
Micturition is mediated by activation of the sacral
parasympathetic efferent pathway to the bladder
and the urethra as well as reciprocal inhibition of
the somatic pathway to the urethral sphincter
(Table 1, Fig. 3B). Studies in animals using brain
lesioning techniques revealed that neurons in the
brainstem at the level of the inferior colliculus (i.e.,
the pontine micturition center) have an essential
role in the control of the parasympathetic compo-
nent of micturition (Barrington, 1925; Kuru, 1965;
Torrens and Morrison, 1986; Mallory et al., 1991;
de Groat et al., 1993). Removal of areas of the
brain above the inferior colliculus by inter-
collicular decerebration usually facilitates micturit-
ion by elimination of inhibitory inputs from more
rostral centers (Yokoyama et al., 2000). However,
transections at any point below the colliculi abol-
ish micturition. Bilateral lesions in the rostral
pons, in the region of the pontine micturition cen-
ter in cats, also abolish micturition (Barrington,
1925), whereas electrical or chemical stimulation at
these sites triggers bladder contractions and mi-
cturition (Kuru, 1965; Sugaya et al., 1987; Kruse
et al., 1990; Mallory et al., 1991; Noto et al., 1991b).
These observations led to the concept of a spino-
bulbospinal micturition reflex pathway that passes
through the pontine micturition center (Fig. 3B).
The pathway functions as on–off switch (de Groat,
1975) that is activated by a critical level of afferent
activity arising from tension receptors in the blad-
der, and is in turn modulated by inhibitory and ex-
citatory influences from areas of the brain rostral to
the pons (e.g., diencephalon and cerebral cortex)
(Torrens and Morrison, 1986; de Groat et al., 1993).
In contrast to the reflex control of the bladder,
the parasympathetic control of the urethra in the
rat appears to be dependent on pathways organ-
ized in the spinal cord (Table 1) that are modulated
by input from the brain. Nitric oxide-mediated
relaxation of the urethra that occurs in response to
bladder distension is reduced but not eliminated by
acute transection of the spinal cord (Kakizaki
et al., 1997, Cheng et al., 1997). The reflex
relaxation of the urethral smooth muscle is very
prominent in chronic spinal-cord-transected rats.
Electrophysiological studies in cats and rats
have confirmed that the parasympathetic efferent
outflow to the urinary bladder is activated by a
long latency supraspinal reflex pathway (de Groat
et al., 1981, 1982, 1993; Mallory et al., 1989;
Cheng et al., 1999). In cats, recordings from sacral
parasympathetic preganglionic neurons innervat-
ing the urinary bladder show that reflex firing oc-
curs with a long latency (65–100 ms) following
stimulation of myelinated (Ad) vesical afferents in
the pelvic nerve (Fig. 6). Afferent stimulation also
evokes negative field potentials in the rostral pons
at latencies of 30–40 ms, whereas electrical stimu-
lation in the pons excites sacral preganglionic neu-
rons at latencies of 45–60 ms. The sum of the
latencies for the spinobulbar and bulbospinal
components of the reflex pathway approximates
the latency for the entire reflex. In cats, it is be-
lieved that the ascending afferent pathways from
the spinal cord project to a relay station in the
periaqueductal gray (PAG), which then connects
to the pontine micturition center (Fig. 3B) (Blok
and Holstege, 1994; Blok et al., 1995; Blok, 2002).
Pontine micturition center
Physiological and anatomical experiments have
provided substantial support for the concept that
 71

Fig. 6. Diagram showing the organization of the parasympathetic excitatory reflex pathway to the detrusor muscle. Scheme is based on
electrophysiological studies in cats. In animals with an intact spinal cord, micturition is initiated by a supraspinal reflex pathway
passing through a center in the brain stem. The pathway is triggered by myelinated afferents (A-d fibers), which are connected to the
tension receptors in the bladder wall. Injury to the spinal cord above the sacral segments interrupts the connections between the brain
and spinal autonomic centers and initially blocks micturition. However, over a period of several weeks following cord injury, a spinal
reflex mechanism emerges, which is triggered by unmyelinated vesical afferents (C-fibers); the A-fiber-afferent inputs are ineffective.
The C-fiber reflex pathway is usually weak or undetectable in animals with an intact nervous system. Stimulation of the C-fiber bladder
afferents by instillation of ice water into the bladder (cold stimulation) activates voiding responses in people with spinal cord injury.
Capsaicin (20–30 mg, subcutaneously) blocks the C-fiber reflex in chronic spinal cats, but does not block micturition reflexes in intact
cats. Intravesical capsaicin also suppresses detrusor hyper-reflexia and cold-evoked reflexes in people with neurogenic bladder dys-
function.

neuronal circuitry in the pontine micturition cen-
ter functions as a switch in the micturition reflex
pathway. The switch seems to regulate bladder
capacity and also coordinate the activity of the
bladder and external urethral sphincter. Electrical
or chemical stimulation in the pontine micturition
center of the rat, cat and dog induces: (1) a sup-
pression of urethral EMG, (2) firing of sacral
preganglionic neurons, (3) bladder contractions
and (4) release of urine (Torrens and Morrison,
1986; Mallory et al., 1989; de Groat et al., 1993;
Blok, 2002). On the other hand, microinjections of
putative inhibitory transmitters into the pontine
micturition center of the cat can increase the vol-
ume threshold for inducing micturition and in
high doses completely block reflex voiding, indi-
cating that synapses in this region are important
for regulating the set point for reflex voiding and
also are an essential link in the reflex pathway
(Mallory et al., 1991). Brain imaging studies in
humans using positron emission tomography
(PET) or functional magnetic resonance imaging
have identified increased neuronal activity in the
pontine micturition center and PAG during
72

voiding (Blok et al., 1997, 1998; Athwal et al., Yoshimura, 2001). Glutamic acid, which is the
1999, 2001). major excitatory transmitter in the central nervous
system, has an important role in the control of the
micturition reflex. Experiments in rats indicate
Suprapontine control of micturition
that glutamatergic transmission in the spinal cord
is essential for bladder and urethral reflexes and
The organization of suprapontine pathways con-
for the spinal processing of afferent input from the
trolling micturition is less well defined, despite the
bladder. Both NMDA and a-amino-3-hydroxy-
fact that there is a large body of literature dealing
5-methyl-4-isoxazoleproprionic acid (AMPA)/
with the responses of the lower urinary tract to
kainate receptors are involved in glutamatergic
lesions or electrical stimulation of the brain (de
transmission in the micturition reflex pathway
Groat et al., 1993). In brief, it appears that the
(Yoshiyama et al., 1993, 1994, 1995, 1997; Sugaya
voluntary control of micturition is dependent
and de Groat, 1994; Matsumoto et al., 1995a, b).
upon (1) connections between the frontal cortex
A study using spinal slice preparations from neo-
and the septal and the preoptic regions of the
natal rats also revealed that sacral preganglionic
hypothalamus and (2) connections between the
neurons directly receive glutamatergic excitatory
paracentral lobule and the brain stem and spinal
inputs through NMDA and AMPA/kainate re-
cord (Torrens and Morrison, 1986; de Groat et al.,
ceptors from spinal interneurons in the region of
1993). Lesions to these areas of cortex resulting
the sacral parasympathetic nucleus (Araki and de
from tumors, aneurysms or cerebrovascular
Groat, 1996). Thus, it is likely that glutamatergic
disease, appear to remove inhibitory control over
transmission is important at various sites in the
the anterior hypothalamic area that normally
micturition reflex pathway.
provides an excitatory input to micturition
The spinobulbospinal micturition reflex path-
centers in the brainstem (Yokoyama et al., 2000).
way controlling bladder and urethral reflexes is
Human PET scan studies have revealed that two
also modulated by various neurotransmitters such
cortical areas (the right dorsolateral prefrontal
as norepinephrine (Yoshimura et al., 1990a, b;
cortex and the anterior cingulate gyrus) were
Ishizuka et al., 1996), dopamine (Yoshimura et al.,
active during voiding (Blok et al., 1997, 1998;
1993, 1998b; Seki et al., 2001), 5-hydroxytrypta-
Blok, 2002). The hypothalamus including the
mine (Thor et al., 1990; Steers et al., 1992a; Espey
preoptic area as well as the pons and the PAG
and Downie, 1995; Danuser and Thor, 1996),
also showed activity in concert with voluntary
gamma aminobutyric acid (Steers et al., 1992b;
micturition. Other PET studies that examined the
Igawa et al., 1993; Araki, 1994), acetylcholine
changes in brain activity during filling of the
(Sugaya et al., 1987; Ishiura et al., 2001) and
bladder revealed that increased activity occurred in
neuropeptides, tachykinins, enkephalins (Booth
the PAG, the midline pons, the mid-cingulate
et al., 1985; Noto et al., 1991a), vasoactive intesti-
gyrus and bilaterally in the frontal lobes (Athwal
nal polypeptide (de Groat et al., 1990) and PACAP
et al., 1999, 2001; Matsuura et al., 2002). These
(Ishizuka et al., 1995), acting through different re-
results are consistent with the notion that the PAG
ceptor subtypes (de Groat and Yoshimura, 2001).
receives information about bladder fullness and
then relays this information to other brain areas
involved in the control of bladder storage.
Neurogenic dysfunction of the lower urinary tract

Supraspinal and spinal neurotransmitters Neurogenic disturbances of micturition can be

controlling micturition
Various neurotransmitters at the spinal and
supraspinal level are involved in regulation of
micturition and continence (de Groat and
classified into two general categories: failure to
store and failure to eliminate urine (Wein, 2002).
Problems with storage occur with differing degrees
of severity, ranging from reduced bladder capacity
and frequency of urination to urgency and

incontinence. A common finding is that disorders
affecting the brain, particularly suprapontine areas,
produce hyperactive or uninhibited bladders.
Cerebrovascular accidents, Parkinson’s disease,
tumors or demyelinating diseases are common
causes of this problem (Betts, 1999; Sakakibara
and Fowler, 1999; Wein, 2002).
Failure to eliminate urine occurs under various
conditions that interrupt the detrusor to detrusor
excitatory reflex pathway or that interfere with the
coordination between detrusor and sphincters
(Chancellor and Yoshimura, 2002; Wein, 2002).
Areflexic bladders can occur with (1) damage to
the pelvic nerve or the sacral spinal cord (lower
motor neuron lesions), (2) lesions of the afferent
pathways (e.g., diabetes, tabes dorsalis, pernicious
anemia, herniated intervertebral disc) or (3) the
acute stage of spinal cord injury rostral to the
sacral segments (an upper motor neuron lesion)
(Fam and Yalla, 1988; Chancellor and Blaivas,
1996; Yoshimura, 1999).
Spinal cord injury rostral to the lumbosacral level
The upper motoneuron type of spinal cord injury
initially leads to a phase of spinal shock that is
followed by a recovery phase during which neu-
rological changes emerge. During the period of
spinal shock immediately after spinal cord injury,
there is a flaccid paralysis and absence of reflex
activity below the level of lesion; thus the urinary
bladder becomes areflexic. However, activity of
striated and smooth muscle sphincters rapidly re-
covers after suprasacral injuries. Because sphincter
tone is present, urinary retention develops and
cord-injured patients have to be treated with in-
termittent or continuous catheterization to elimi-
nate urine from the urinary bladder. Following the
spinal shock phase, reflex detrusor activity reap-
pears after 2–12 weeks in most cases (Fam and
Yalla, 1988; Chancellor and Blaivas, 1996).
During the recovery phase, the detrusor devel-
ops involuntary reflex contractions (neurogenic
detrusor overactivity) in response to visceral stim-
uli such as bladder filling or suprapubic manual
compression. After the recovery period, these
involuntary bladder contractions become more
73
powerful and cause urinary incontinence. In addi-
tion, the bladder is usually only emptied partially
owing to development of simultaneous contrac-
tions of the bladder and the striated urethral
sphincter (detrusor–sphincter dyssynergia, Fig. 2C).
Inefficient voiding may also be due to unsustained
bladder contractions.
In normal micturition, activation of the pontine
micturition center simultaneously induces bladder
contractions and a suppression of sphincter activ-
ity. A complete suprasacral lesion interrupts this
coordination between the bladder and the striated
sphincter (Fig. 2C). Thus, cord-injured people with
chronic upper motoneuron lesions exhibit (1)
detrusor hyperreflexia, (2) coordinated relaxation
of urethral sphincter smooth muscle and (3)
dyssynergic contraction of urethral sphincter
striated muscle (detrusor–sphincter dyssynergia).
All patterns of detrusor hyperreflexia associated
with detrusor-sphincter dyssynergia lead to high
intravesical pressure and/or severe bladder
trabeculation with the formation of diverticula,
which often induce vesicoureteral reflux and dete-
rioration of the upper urinary tract (Fam and
Yalla, 1988; Chancellor and Blaivas, 1996).
People with spinal cord injury at the level of T6
or higher, often exhibit autonomic dysreflexia,
which is characterized by arterial pressor responses
induced by stimuli below the level of spinal cord
lesion such as bladder distension, fecal impaction
or bladder inflammation (Fam and Yalla, 1988).
These stimuli cause excitation of sympathetic
pathways and induce arteriolar vasoconstriction
and hypertension as well as piloerection and
sweating below the level of injury.
Spinal cord injury at or below the sacral level
In the lower motoneuron type of spinal cord
injury, complete lesions of the sacral cord or the
cauda equina usually result in flaccidity of the
bladder and its outlet. The bladder becomes
areflexic, thereby bladder compliance and bladder
capacity are increased. Pressures in the striated
urethral sphincter are decreased. When the lesion
extends to the thoracolumbar spinal cord, the
sympathetic outflow to the internal smooth muscle
74
sphincter of the urethra is also damaged, and the
bladder neck becomes incompetent in association
with hypoactive detrusor and striated sphincter
(Fam and Yalla, 1988; Chancellor and Blaivas,
1996; Yoshimura, 1999).
Changes in functions of bladder afferent pathways
after spinal cord injury
A slow recovery of lower urinary tract function
following spinal cord injury is also observed in
animals such as cats and rats (de Groat et al.,
1990; Kruse et al., 1993; de Groat, 1995). Com-
plete transection of the thoracic spinal cord ini-
tially produces detrusor areflexia in both species,
followed by the emergence of detrusor hype-
rreflexia with detrusor-sphincter dyssynergia.
Electrophysiological and pharmacological studies
have shown that the reflex pathways controlling
the lower urinary tract are markedly different be-
tween spinal intact and spinal-injured animals. In
spinal intact cats and rats, the micturition reflex is
mediated by a long-latency supraspinal reflex
pathway, passing through the pons, that is acti-
vated by myelinated Ad-fiber bladder afferents
(de Groat et al., 1981, 1993, 1998; Mallory et al.,
1989; Yoshimura, 1999). However, in chronic spi-
nal cats the afferent limb of the micturition reflex
consists of unmyelinated C-fiber afferents (Fig. 6).
It has also been demonstrated that in chronic spi-
nal cats (3–6 weeks after injury), subcutaneously
administered capsaicin, a C-fiber neurotoxin, com-
pletely blocked reflex bladder contractions induced
by bladder distention, whereas capsaicin had no
inhibitory effects on reflex bladder contractions in
spinal intact cats (de Groat et al., 1990; Cheng
et al., 1999). Thus, it is plausible that C-fiber
bladder afferents that usually do not respond to
bladder distention (i.e., silent C-fibers) (Habler
et al., 1990) become mechano-sensitive and initiate
automatic micturition after spinal cord injury.
Increased excitability of C-fiber afferents after
spinal cord injury has also been demonstrated in
rats in which detrusor hyperreflexia was identified
during cystometrograms as non-voiding bladder
contractions prior to micturition. The non-voiding
contractions were suppressed by capsaicin,
although voiding contractions in spinal cord-
injured rats were still triggered by capsaicin-
resistant Ad-fiber afferents (Cheng et al., 1995;
Cheng and de Groat, 2004). Capsaicin treatment
also eliminated detrusor-sphincter dyssynergia in
anesthetized chronic spinal rats. Clinical studies
demonstrated that C-fiber afferents innervating
the bladder are involved in detrusor hyperreflexia
and autonomic dysreflexia in spinal cord-injured
people (Geirsson et al., 1995; Igawa et al., 1996,
2003; Cruz et al., 1997) and detrusor hyperreflexia
in people with multiple sclerosis (Fowler et al.,
1992; Szallasi and Fowler, 2002). Taken together,
it is clear that the functional properties of C-fiber
afferents in the bladder are altered following spinal
cord injury, thereby inducing hyperreflexic bladder
activity in both humans and animals (Fig. 6).
Other evidence of a reorganization of C-
fiber-mediated reflex pathways in subjects with
suprasacral spinal cord injury was obtained in
studies of the cold stimulation-evoked voiding re-
flex. Instillation of cold water into the bladder in
these subjects induces reflex voiding (the Bors Ice
Water Test) (Bors and Comarr, 1971; Geirsson
et al., 1993, 1995). This reflex does not occur in
normal subjects, except for infants (Geirsson et al.,
1994). It has been shown in the cat that C-fiber
bladder afferents are responsible for cold-induced
bladder reflexes (Fall et al., 1990). Intravesical ad-
ministration of capsaicin to paraplegic people
blocks the cold-induced bladder reflexes, indicat-
ing that they are mediated by C-fiber afferents
(Geirsson et al., 1995, Igawa et al., 1996, 2003).
Studies in rats revealed that cold stimulation in-
duced detrusor-sphincter dyssynergia and that
capsaicin treatment prevented it (Cheng et al.,
1997). Thus, cold- and capsaicin-sensitive C-fiber
bladder afferents can evoke detrusor hyperreflexia
and detrusor-sphincter dyssynergia. These re-
sponses are facilitated after the elimination of
supraspinal controls by spinal cord injury (Fig. 6).
Changes in the firing properties of bladder afferent
neurons following spinal cord injury
The mechanisms for inducing hyperexcitability of
C-fiber bladder afferents were investigated by

whole-cell patch-clamp recording in dissociated
dorsal root ganglion neurons innervating the rat
urinary bladder (Yoshimura and de Groat, 1997).
Chronic spinal cord injury in rats produced hyper-
trophy of bladder afferent neurons as reflected by
an increase in cell diameter and cell input capac-
itance. This confirmed earlier findings that bladder
afferent neurons in the L6-S1 (1st sacral) dorsal
root ganglia undergo somal hypertrophy (45–50%
increase in cross-sectional area) in chronic spinal
cord-transected rats (Kruse et al., 1995). In addi-
tion to neuronal hypertrophy, bladder afferent
neurons in chronic spinal rats increased their ex-
citability. In contrast to the majority (approxi-
mately 70%) of bladder afferent neurons from
spinal cord intact rats that exhibited high thresh-
old tetrodotoxin-resistant humped action poten-
tials (Yoshimura et al., 1996), 60% of bladder
afferent neurons in chronic spinal cord-transected
rats exhibited tetrodotoxin-sensitive low-threshold
action potentials. The mean threshold for spike
activation in cord-transected rats ( 25.5 mV) was
21% lower than in intact animals ( 20 mV).
Plasticity in Na+ and K+ channels of bladder
afferent neurons following spinal cord injury
The alteration of electrophysiological properties in
bladder afferent neurons after spinal cord injury
was also reflected in changes in density of different
types of Na+ currents (Yoshimura and de Groat,
1997; Black et al., 2003). Consistent with the in-
crement in the proportion of neurons with tetrodo-
toxin-sensitive spikes, the number of bladder
afferent neurons that predominantly expressed
tetrodotoxin-sensitive Na+ currents (60–100% of
total Na+ currents) also increased. The density of
tetrodotoxin-sensitive Na+ currents in bladder af-
ferent neurons significantly increased from 32.1 to
80.6 pA/pF, while tetrodotoxin-resistant current
density decreased from 60.5 to 17.9 pA/pF fol-
lowing spinal cord injury. In addition, an increase
in tetrodotoxin-sensitive Na+ currents was detect-
ed in some bladder afferent neurons that still re-
tained a predominance of tetrodotoxin-resistant
currents (450% of total Na+ currents) after spi-
nal cord injury. These data indicate that spinal
cord injury induces a switch in expression of Na+
75
channels from the tetrodotoxin-resistant type to
the tetrodotoxin-sensitive type. Since tetrodotoxin-
sensitive Na+ currents have a lower threshold for
activation than tetrodotoxin-resistant currents, it
is reasonable to assume that these changes in ex-
pression of Na+ channels in bladder afferent neu-
rons after spinal cord injury contribute to a low
threshold for spike activation in these neurons.
In chronic spinal cord-transected rats, bladder
afferent neurons with tetrodotoxin-sensitive spikes
exhibited no apparent membrane potential relax-
ation when the neurons were gradually depolar-
ized by injecting inward currents. In these neurons,
voltage responses induced by current injections
were not altered by application of 4-aminopyri-
dine, a K+ channel blocker, although the neurons
with tetrodotoxin-resistant spikes still had 4-ami-
nopyridine-sensitive membrane potential relaxa-
tion during depolarization as found in spinal cord
intact rats. The phenomenon of membrane poten-
tial relaxation is due to slowly inactivating IA cur-
rents that can be elicited by depolarization from
the resting membrane potential (Yoshimura et al.,
1996; Yoshimura, 1999). Therefore it is likely that
following spinal cord injury A-type K+ channels
are suppressed in parallel with an increased ex-
pression of tetrodotoxin-sensitive Na+ currents,
thereby increasing excitability of bladder afferent
neurons. Since tetrodotoxin-resistant Na+ cur-
rents and IA currents are preferentially expressed
in small-sized C-fiber afferent neurons in spinal
cord intact rats (Gold et al., 1996a; Yoshimura
et al., 1996), the changes in these channels after
spinal cord injury must occur primarily in C-fiber
bladder afferent neurons and contribute to in-
creased cell excitability of these neurons.
Immunohistochemical studies using antibodies to
tetrodotoxin-resistant Na+ channel protein (Nav
1.8) revealed that tetrodotoxin-resistant Na+
channels are located not only in small-sized dor-
sal root ganglion cell bodies, but also in superficial
laminae of the dorsal horn of the spinal cord
(Novakovic et al., 1998; Yoshimura et al., 2001).
Thus changes occurring in afferent cell bodies fol-
lowing spinal cord injury may also occur at affer-
ent axons and/or receptors in the bladder and thus
contribute to the emergence of the C-fiber-
mediated spinal micturition reflex.
76
Fig. 7. Diagram of hypothetical mechanisms inducing lower
urinary tract dysfunction following spinal cord injury (SCI).
The subsequent events occurring after SCI are indicated by the
numbers 1–7. Injury to the spinal cord (1) causes detrusor-
sphincter dyssynergia (DSD) (2) leading to functional obstruc-
tion of urethra. Increased urethral resistance induces bladder
hypertrophy (3), resulting in increased levels of nerve growth
factor (NGF, 4) in the bladder smooth muscle. The nerve
growth factor (NGF) level in the spinal cord (5) is also in-
creased after SCI. Increased NGF in the bladder and spinal
cord is transported to bladder afferent pathways (6), followed
by hyperexcitability of bladder afferent pathways (7). Hyper-
excitability of bladder afferent pathways causes or enhances
neurogenic detrusor overactivity (8) and DSD (2).
Role of neurotrophic factors
Nerve growth factor (NGF) has been implicated as
a chemical mediator of pathology-induced changes
in C-fiber afferent nerve excitability and reflex
bladder activity (Yoshimura, 1999; Vizzard, 2000).
It has been demonstrated that chronic administra-
tion of NGF into the bladder of rats induced
bladder hyperactivity and increased the firing fre-
quency of dissociated bladder afferent neurons
(Yoshimura et al., 1999), and that the production
of neurotrophic factors including NGF increased
in the bladder after spinal cord injury (Vizzard,
2000, this volume). It has also been shown that the
bladder hyperactivity and hypertrophy of afferent
and efferent neurons innervating the hypertrophic
bladder in rats with partial urethral obstruction
was antagonized in part by systemic autoimmuni-
zation against NGF (Steers et al., 1996). Thus it
seems that target organ–neural interactions medi-
ated by neurotrophic factors such as NGF pro-
duced in the hypertrophied bladder muscle may
contribute to changes in C-fiber bladder afferent
pathways that underlie the detrusor hyperreflexia
and detrusor-sphincter dyssynergia after spinal
cord injury (Fig. 7).
In addition, we have recently found that in-
creased NGF in the spinal cord after spinal cord
injury is also responsible for inducing hyperexcit-
ability of C-fiber bladder afferent pathways, and
that intrathecal application of NGF antibodies,
which neutralized NGF in the spinal cord, sup-
pressed detrusor hyperreflexia and detrusor-
sphincter dyssynergia in spinal cord injured rats
(Seki et al., 2002, 2004a, b). Intrathecal adminis-
tration of NGF antibodies also reportedly blocked
autonomic dysreflexia in paraplegic rats (Krenz
et al., 1999). Thus, NGF and its receptors in the
bladder and/or the spinal cord are potential targets
for new therapies to suppress detrusor hype-
rreflexia and detrusor-sphincter dyssynergia after
spinal cord injury (Fig. 7).
Spinal mechanisms involving vasoactive intestinal
polypeptide (VIP) and pituitary adenylate cyclase
activating polypeptide (PACAP)
Vasoactive intestinal polypeptide (VIP) and
pituitary adenylate cyclase activating polypeptide
(PACAP) are contained in afferent neurons inner-
vating the urinary bladder of the cat and rat
(Kawatani et al., 1985; see Vizzard, this volume,
for references). In the cat sacral spinal cord, VIP
is present exclusively in C-fiber afferent axons
(Morgan et al., 1999), and is located in afferent ter-
minals projecting to the sacral parasympathetic nu-
cleus (Kawatani et al., 1985). In chronic spinal cats,
VIP-immunoreactivity is distributed over a wider
area of the lateral dorsal horn in the sacral spinal
cord, suggesting C-fiber afferent axonal sprouting
after spinal injury (Thor et al., 1986). In addition,
the effects of intrathecal administration of VIP are
changed. In normal cats, VIP inhibits the micturit-
ion reflex, whereas in paraplegic cats VIP facilitates
the micturition reflex (de Groat et al., 1990). These
findings suggest that the action of a putative C-fiber
afferent transmitter may underlie the emergence of
C-fiber bladder reflexes in the paraplegic cat.
In normal and spinal cord-injured rats, VIP and
PACAP, another member of the secretin/glucagon/
VIP peptide family, facilitated the micturition

reflex by actions on the spinal cord (Ishizuka et al.,
1995; Yoshiyama and de Groat, 1997; Vizzard,
this volume). Patch-clamp studies in the neonatal
rat spinal slice preparation revealed that PACAP
has a direct excitatory action on parasympathetic
preganglionic neurons due in part to blockade of
K+ channels and also has an indirect action med-
itated by activation of excitatory interneurons
(Miura et al., 2001b). PACAP increased the fre-
quency of spontaneous excitatory postsynaptic
potentials as well as spontaneous firing and de-
creased the threshold for action potential genera-
tion. PACAP also increased the number and
frequency of action potentials elicited by depolar-
izing current pulses. PACAP levels in bladder
afferent neurons and in spinal cord projections are
upregulated after spinal cord injury (Zvarova
et al., 2005; Vizzard, this volume). These findings
suggest that putative C-fiber afferent transmitters,
such as VIP in the cat or PACAP in the rat, may
underlie the emergence of C-fiber afferent evoked
bladder reflexes after spinal cord injury.
Spinal glutamatergic mechanisms
Glutamic acid plays an essential role as an exci-
tatory transmitter in the spinal reflex pathways
controlling bladder and external urethral sphincter
activity in both normal and spinal cord-injured
rats. A study of the effect of selective antagonists
revealed that NMDA and non-NMDA glut-
amatergic (AMPA) receptors are involved in the
supraspinal and spinal reflex pathways controlling
voiding, and that AMPA receptor mechanisms are
most important. Intrathecal or intravenous ad-
ministration of NMDA or AMPA antagonists in
urethane-anesthetized rats depressed bladder con-
tractions and electromyographic activity of the
external urethral sphincter (Yoshiyama et al.,
1993, 1995). In spinal cord injured-rats, external
urethral sphincter electromyographic activity was
more sensitive than bladder reflexes to glut-
amatergic antagonists (Yoshiyama et al., 1993,
1997), raising the possibility that the two reflex
pathways have different types of receptors. This
was confirmed with in situ hybridization tech-
niques which revealed that sacral parasympathetic
preganglionic neurons in the rat express high
77
mRNAs levels of GluR-A and GluR-B AMPA
receptor subunits and NR1, but not NR2 NMDA
receptor subunits (Shibata et al., 1999). On the
other hand, motoneurons in the urethral sphincter
nucleus express all four AMPA receptor subunits
(GluR-A, -B, -C and -D) in conjunction with
moderate amounts of NR2A and NR2B as well as
high levels of NR1 receptor subunits. It seems
likely that this difference in expression accounts
for the different sensitivity of bladder and sphinc-
ter reflexes to glutamatergic antagonists.
Spinal tachykinin mechanisms
Tachykinins, such as substance P and neurokinin
A, that are released at C-fiber afferent terminals
can act in the bladder wall to modulate the excit-
ability of afferent nerves and induce bladder con-
tractions (de Groat et al., 1993; Morrison et al.,
2002), and are also thought to be involved in the
spinal cord as mediators of excitatory transmission
between primary afferent nerves and second-order
spinal neurons that express NK1 receptors. Al-
though some studies have reported an increase in
bladder capacity and a decrease in maximal void-
ing pressure after intrathecal administration of
NK1 receptor antagonists in normal conscious or
anesthetized rats, other experiments have failed to
detect a change in bladder function after admin-
istration of these agents in normal rats (de Groat
and Yoshimura, 2001). On the other hand, a re-
duction in detrusor hyperreflexia in spinal cord-
injured rats has been noted after administration of
NK-1 and NK-2 tachykinin receptor antagonists
(Abdel-Gawad et al., 2001). The role of NK1 re-
ceptor-expressing spinal neurons in C-fiber-affer-
ent mediated bladder activity was also evaluated
by destroying these neurons in the L6-S1 spinal
cord by intrathecal administration of a ribosome-
inactivating toxin, saporin conjugated with a
specific NK1 receptor ligand that promotes the
binding and internalization of the toxin in NK1
receptor-expressing neurons. In treated animals,
NK1 receptor immunoreactivity was reduced in
lamina I of the spinal cord and the bladder
hyperactivity induced by intravesical instillation
of capsaicin was reduced, but cystometric param-
eters in awake rats were not changed (Seki et al.,
78
2005). Preliminary experiments in spinal cord-in-
jured rats indicate that detrusor hyperreflexia is
also reduced by saporin treatment (Seki et al.,
2004a). These studies raise the possibility that NK1
receptor antagonists might be useful clinically in
treating bladder dysfunction in people with spinal
cord injury.
Peripheral muscarinic mechanisms
Muscarinic receptor antagonists are the first line
therapy for detrusor hyperreflexia induced by spi-
nal cord injury because bladder contractions are
induced by activation of postjunctional muscarinic
receptors in the detrusor muscle by acetylcholine
released from parasympathetic postganglionic
nerves (Kim et al., 1997; Stohrer et al., 1999;
Chapple, 2000). Although the M2 receptor is the
predominant subtype in the bladder (approximate-
ly 80%), the contractions of the bladder are me-
diated by the M3 receptor subtype, which is
therefore considered the primary target of drug
therapy (Hegde and Eglen, 1999; Chapple, 2000).
However, other receptors may play a role in trig-
gering bladder activity after spinal cord injury.
M1 muscarinic receptors are located prejunc-
tionally on cholinergic nerve terminals in the blad-
der (Somogyi and de Groat, 1999). Activation of
these receptors facilitates ACh release during the
prolonged, high-frequency nerve firing that would
occur during voiding. This amplification mecha-
nism through pre-junctional M1 receptors is up-
regulated in spinal cord injured rats, suggesting
that detrusor hyperreflexia occurring after spinal
cord injury may be due in part to an enhancement
of ACh release from parasympathetic nerve ter-
minals (Somogyi et al., 1998; Somogyi and de
Groat, 1999). Activation of postjunctional M2 re-
ceptors also elicits a direct contraction of the rat
bladder smooth muscle after spinal cord injury
(Braverman et al., 1999). Thus, antimuscarinic
agents may act by multiple mechanisms to reduce
detrusor hyperreflexia.
Conclusions
The lower urinary tract has two main functions:
storage and periodic elimination of urine. These
functions are regulated by a complex neural con-
trol system located in the brain and spinal cord.
This control system performs like a simple switch-
ing circuit to maintain a reciprocal relationship
between the reservoir (bladder) and outlet compo-
nents (urethra and urethral sphincter) of the uri-
nary tract. Spinal cord injury disrupts voluntary
control and the normal reflex pathways that coor-
dinate bladder and sphincter function. Studies in
animals indicate that recovery of bladder function
following spinal cord injury is dependent upon the
reorganization of reflex pathways in both the pe-
ripheral and central nervous system. Part of this
reorganization may be influenced by neural-target
organ interactions that are mediated by neurotro-
phic factors released by the peripheral organs.
Acknowledgments
The authors’ research is supported by NIH
research grants (DK49430, DK 57267, P01 HD
39768).
References
Abdel-Gawad, M., Dion, S.B. and Elhilali, M.M. (2001)
Evidence of a peripheral role of neurokinins in detrusor
hyperreflexia: a further study of selective tachykinin antag-
onists in chronic spinal injured rats. J. Urol., 165: 1739–1744.
Andersson, K.E. (1993) Pharmacology of lower urinary tract
smooth muscle and penile erection tissues. Pharmacol. Rev.,
45: 253–308.
Andersson, K.E. and Arner, A. (2004) Urinary bladder con-
traction and relaxation: physiology and pathology. Physiol.
Rev., 84: 935–986.
Araki, I. (1994) Inhibitory postsynaptic currents and the effects
of GABA on visually identified sacral parasympathetic
preganglionic neurons in neonatal rats. J. Neurophysiol.,
72: 2903–2910.
Araki, I. and de Groat, W.C. (1996) Unitary excitatory synaptic
currents in preganglionic neurons mediated by two distinct
groups of interneurons in neonatal rat sacral parasympa-
thetic nucleus. J. Neurophysiol., 76: 215–226.
Araki, I. and de Groat, W.C. (1997) Developmental synaptic
depression underlying reorganization of visceral reflex path-
ways in the spinal cord. J. Neurosci., 17: 8402–8407.
Athwal, B.S., Berkley, K.J., Brennan, A., Craggs, M., Fowler,
C.J., Hussain, I., Sakakibara, R. and Frackowiak, R.S.J.
(1999) Brain activity associated with the urge to void and
bladder fill volume in normal men: preliminary data from a
PET study. BJU Int., 84: 148–149.

Athwal, B.S., Berkley, K.J., Hussain, I., Brennan, A., Craggs,
M., Sakakibara, R., Frackowiak, R.S.J. and Fowler, C.J.
(2001) Brain responses to changes in bladder volume and
urge to void in healthy men. Brain, 124(2): 369–377.
Bahns, E., Ernsberger, U., Janig, W. and Nelke, A. (1998)
Functional characteristics of lumbar visceral afferent fibers
from the urinary bladder and urethra in the cat. Eur. J.
Physiol., 407: 510.
Barrington, F.J.F. (1925) The effect of lesions of the hind-and
mid-brain on micturition in the cat. J. Exp. Physiol., 15:
81–102.
Beattie, B.S., Li, Q., Leedy, G. and Bresnahan, J.C. (1990)
Motoneurons innervating the external anal and urethral
sphincters of the female cat have different patterns of den-
dritic arborization. Neurosci. Lett., 111: 69–74.
Betts, C.D. (1999) Bladder and sexual function in multiple
sclerosis. In: Fowler C.J. (Ed.), Neurology of Bladder, Bowel,
and Sexual Dysfunction. Butterworth-Heinemann, Boston,
pp. 289–308.
Birder, L.A., Apodaca, G., de Groat, W.C. and Kanai, A.J.
(1998) Adrenergic- and capsaicin-evoked nitric oxide release
from urothelium and afferent nerves in urinary bladder. Am.
J. Physiol., 275: F226–F229.
Birder, L.A., Barrick, S.R., Roppolo, J.R., Kanai, A.J., de
Groat, W.C., Kiss, S. and Buffington, C.A. (2003) Feline
interstitial cystitis results in mechanical hypersensitivity and
altered ATP release from bladder urothelium. Am. J.
Physiol., 285: F423–F429.
Birder, L.A. and de Groat, W.C. (1993) Induction of c-fos gene
expression in spinal neurons of the rat by nociceptive and
non-nociceptive stimulation of the lower urinary tract. Am. J.
Physiol., 265: R643–R648.
Birder, L.A., Kanai, A.J., de Groat, W.C., Kiss, S., Nealen,
M.L., Burke, N.E., Dineley, K.E., Watkins, S., Reynolds, I.J.
and Caterina, M.J. (2001) Vanilloid receptor expression sug-
gests a sensory role for urinary bladder epithelial cells. Proc.
Natl. Acad. Sci., U. S. A., 98: 13396–13401.
Birder, L.A., Nakamura, Y., Kiss, S., Nealen, M.L., Barrick,
S., Kanai, A.J., Wang, E., Ruiz, G., de Groat, W.C.,
Apodaca, G., Watkins, S. and Caterina, M.J. (2002) Altered
urinary bladder function in mice lacking the vanilloid recep-
tor TRPV1. Nat. Neurosci., 5: 856–860.
Birder, L.A., Roppolo, J.R., Erickson, V.L. and de Groat,
W.C. (1999) Increased c-fos expression in spinal lumbosacral
projection and preganglionic neurons after irritation of the
lower urinary tract in the rat. Brain Res., 834: 55–65.
Black, J.A., Cummins, T.R., Yoshimura, N., de Groat, W.C.
and Waxman, S.G. (2003) Tetrodotoxin-resistant sodium
channels Nav 1.8/SNS and Nav 1.9/NaN in afferent neurons
innervating urinary bladder in control and spinal cord in-
jured rats. Brain Res., 963: 132–138.
Blok, B.F., De Weerd, H. and Holstege, G. (1995) Ultrastruc-
tural evidence for a paucity of projections from the
lumbosacral cord to the pontine micturition center or M-
region in the cat: a new concept for the organization of the
micturition reflex with the periaqueductal gray as central re-
lay. J. Comp. Neurol., 359: 300–309.
79
Blok, B.F. and Holstege, G. (1994) Direct projections from the
periaqueductal gray to the pontine micturition center
(M-region). An anterograde and retrograde tracing study in
the cat. Neurosci. Lett., 166: 93–96.
Blok, B.F., Sturms, L.M. and Holstege, G. (1998) Brain
activation during micturition in women. Brain,
121(Part 11): 2033–2042.
Blok, B.F., Willemsen, A.T. and Holstege, G. (1997) A PET
study on brain control of micturition in humans. Brain, 120:
111–121.
Blok, B.F.M. (2002) Brain control of the lower urinary tract.
Scand. J. Urol. Nephrol., 36(Suppl. 210): 11–15.
Booth, A.M., Hisamitsu, T., Kawatani, M. and de Groat, W.C.
(1985) Regulation of urinary bladder capacity by endogenous
opioid peptides. J. Urol., 133: 339–342.
Bors, E. and Comarr, A.E. (1971) Neurological Urology:
Physiology of Micturition, Its Neurological Disorders and
Sequelae. University Park Press, Baltimore.
Braverman, A.S., Legos, J., Young, W., Luthin, G. and
Ruggieri, M. (1999) M2 receptors in genito-urinary smooth
muscle pathology. Life Sci., 64: 429–436.
Burnett, A.L., Calvin, D.C., Chamness, S.L., Liu, J.X., Nelson,
R.J., Klein, S.L., Dawson, V.L., Dawson, T.M. and Snyder,
S.H. (1997) Urinary bladder-urethral sphincter dysfunction
in mice with targeted disruption of neuronal nitric oxide
synthase models idiopathic voiding disorders in humans.
Nat. Med., 3: 571–574.
Burnstock, G. (2001) Purinergic signaling in the lower urinary
tract. In: Abbracchio M.P. and Williams M. (Eds.), Handbook
of Experimental Pharmacology. Springer, Berlin, pp. 423–515.
Chancellor, M.B. and Blaivas, J.G. (1996) Spinal cord injury.
In: Chancellor M.B. and Blaivas J.G. (Eds.), Practical Ne-
uro-Urology. Genitourinary Complications in Neurologic
Disease. Butterworth-Heinemann, Stoneham, pp. 99–118.
Chancellor, M.B. and Yoshimura, N. (2002) Physiology and
pharmacology of the bladder and urethra. In: Walsh P.C.,
Retik A.B., Vaughn E.D. and Wein A.J. (Eds.), Campbell’s
Urology, Chapter 23. W.B. Saunders Co., Philadelphia,
pp. 831–886.
Chapple, C.R. (2000) Muscarinic receptor antagonists in the
treatment of overactive bladder. Urology, 55: 33–46.
Cheng, C.L., Chai, C.Y. and de Groat, W.C. (1997) Detrusor-
sphincter dyssynergia induced by cold stimulation of the
urinary bladder of rats. Am. J. Physiol., 272: R1271–R1282.
Cheng, C.L. and de Groat, W.C. (2004) The role of capsaicin-
sensitive afferent fibers in the lower urinary tract dysfunction
induced by chronic spinal cord injury in rats. Exp. Neurol.,
187: 445–454.
Cheng, C.L., Liu, J.C., Chang, S.Y., Ma, C.P. and de Groat,
W.C. (1999) Effect of capsaicin on the micturition reflex in
normal and chronic spinal cord-injured cats. Am. J. Physiol.,
277: R786–R794.
Cheng, C.L., Ma, C.P. and de Groat, W.C. (1995) Effect of
capsaicin on micturition and associated reflexes in chronic
spinal rats. Brain Res., 678: 40–48.
Chuang, Y.C., Fraser, M.O., Yu, Y., Beckel, J.M., Seki, S.,
Nakanishi, Y., Yokoyama, H., Chancellor, M.B.,
80
Yoshimura, N. and de Groat, W.C. (2001) Analysis of the
afferent limb of the vesicovascular reflex using neurotoxins,
resiniferatoxin and capsaicin. Am. J. Physiol., 281:
R1302–R1310.
Cockayne, D.A., Hamilton, S.G., Zhu, Q.M., Dunn, P.M.,
Zhong, Y., Novakovic, S., Malmberg, A.B., Cain, G.,
Berson, A., Kassotakis, L., Hedley, L., Lachnit, W.G.,
Burnstock, G., McMahon, S.B. and Ford, A.P. (2000)
Urinary bladder hyporeflexia and reduced pain-related
behaviour in P2X3-deficient mice. Nature, 407: 1011–1015.
Cruz, F., Guimaraes, M., Silva, C., Rio, M.E., Coimbra, A. and
Reis, M. (1997) Desensitization of bladder sensory fibers by
intravesical capsaicin has long lasting clinical and urody-
namic effects in patients with hyperactive or hypersensitive
bladder dysfunction. J. Urol., 157: 585–589.
Danuser, H. and Thor, K.B. (1996) Spinal 5-HT2 receptor-
mediated facilitation of pudendal nerve reflexes in the anaes-
thetized cat. Br. J. Pharmacol., 118: 150–154.
de Groat, W.C. (1975) Nervous control of the urinary bladder
of the cat. Brain Res., 87: 201–211.
de Groat, W.C. (1987) Neuropeptides in pelvic afferent path-
ways. Experientia, 43: 801–813.
de Groat, W.C. (1995) Mechanisms underlying the recovery of
lower urinary tract function following spinal cord injury.
Paraplegia, 33: 493–505.
de Groat, W.C., Araki, I., Vizzard, M.A., Yoshiyama, M.,
Yoshimura, N., Sugaya, K., Tai, C. and Roppolo, J.R. (1998)
Developmental and injury induced plasticity in the micturit-
ion reflex pathway. Behav. Brain Res., 92: 127–140.
de Groat, W.C. and Booth, A.M. (1993) Synaptic transmission
in pelvic ganglia. In: Maggi C.A. (Ed.), The Autonomic
Nervous System, Vol. 3, Nervous Control of the Urogenital
System, 1st Ed. Harwood Academic Publishers, London,
pp. 291–347.
de Groat, W.C., Booth, A.M., Milne, R.J. and Roppolo, J.R.
(1982) Parasympathetic preganglionic neurons in the sacral
spinal cord. J. Auton. Nerv. Syst., 5: 23–43.
de Groat, W.C., Booth, A.M. and Yoshimura, N. (1993)
Neurophysiology of micturition and its modification in an-
imal models of human disease. In: Maggi C.A. (Ed.), The
Autonomic Nervous System, Vol. 3, Chapter 8, Nervous
Control of the Urogenital System. Harwood Academic Pub-
lishers, London, UK, pp. 227–289.
de Groat, W.C., Kawatani, M., Hisamitsu, T., Cheng, C.L.,
Ma, C.P., Thor, K., Steers, W. and Roppolo, J.R. (1990)
Mechanisms underlying the recovery of urinary bladder
function following spinal cord injury. J. Auton. Nerv. Syst.,
30(Suppl.): S71–S77.
de Groat, W.C. and Lalley, P.M. (1972) Reflex firing in the
lumbar sympathetic outflow to activation of vesical afferent
fibres. J. Physiol. (Lond), 226: 289–309.
de Groat, W.C., Nadelhaft, I., Milne, R.J., Booth, A.M.,
Morgan, C. and Thor, K. (1981) Organization of the sacral
parasympathetic reflex pathways to the urinary bladder and
large intestine. J. Auton. Nerv. Syst., 3: 135–160.
de Groat, W.C. and Yoshimura, N. (2001) Pharmacology of the
lower urinary tract. Ann. Rev. Pharm. Toxicol., 41: 691–721.
Delancey, J., Gosling, J., Creed, K., Dixon, J., Delmas, V.,
Landon, D. and Norton, P. (2002) Gross anatomy and cell
biology of the lower urinary tract. In: Abrams P., Cardozo
L., Khoury S. and Wein A. (Eds.), Incontinence. Health
Publications, Ltd., Jersey, UK, pp. 17–82.
Espey, M.J. and Downie, J.W. (1995) Serotonergic modulation
of cat bladder function before and after spinal transection.
Eur. J. Pharmacol., 287: 173–177.
Fall, M., Lindstrom, S. and Mazieres, L. (1990) A bladder-to-
bladder cooling reflex in the cat. J. Physiol. (Lond.), 427:
281–300.
Fam, B. and Yalla, S.V. (1988) Vesicourethral dysfunction in
spinal cord injury and its management. Semin. Neurol., 8:
150–155.
Ferguson, D.R., Kennedy, I. and Burton, T.J. (1997) ATP is
released from rabbit urinary bladder cells by hydrostatic
pressure changes–a possible sensory mechanism? J. Physiol.
(Lond.), 505: 503.
Fowler, C.J., Jewkes, D., McDonald, W.I., Lynn, B. and de
Groat, W.C. (1992) Intravesical capsaicin for neurogenic
bladder dysfunction. Lancet, 339: 1239.
Geirsson, G., Fall, M. and Lindstrom, S. (1993) The ice-water
test–a simple and valuable supplement to routine cystometry.
Br. J. Urol., 71: 681–685.
Geirsson, G., Fall, M. and Sullivan, L. (1995) Clinical and
urodynamic effects of intravesical capsaicin treatment in pa-
tients with chronic traumatic spinal detrusor hyperreflexia. J.
Urol., 154: 1825–1829.
Geirsson, G., Lindstrom, S., Fall, M., Gladh, G., Hermansson,
G. and Hjalmas, K. (1994) Positive bladder cooling test
in neurologically normal young children. J. Urol., 151:
446–448.
Gold, M.S., Shuster, M.J. and Levine, J.D. (1996) Character-
ization of six voltage-gated K+ currents in adult rat sensory
neurons. J. Neurophysiol., 75: 2629–2646.
Habler, H.J., Janig, W. and Koltzenburg, M. (1990) Activation
of unmyelinated afferent fibres by mechanical stimuli and
inflammation of the urinary bladder in the cat. J. Physiol.
(Lond.), 425: 545–562.
Hegde, S.S. and Eglen, R.M. (1999) Muscarinic receptor sub-
types modulating smooth muscle contractility in the urinary
bladder. Life Sci., 64: 419–428.
Ho, K.M., Ny, L., McMurray, G., Andersson, K.E., Brading,
A.F. and Noble, J.G. (1999) Co-localization of carbon mon-
oxide and nitric oxide synthesizing enzymes in the human
urethral sphincter. J. Urol., 161: 1968–1972.
Holstege, G., Griffiths, D., deWall, H. and Dalm, E. (1986)
Anatomical and physiological observations on supraspinal
control of bladder and urethral sphincter muscles in cats. J.
Comp. Neurol., 250: 449.
Igawa, Y., Komiyama, I., Nishizawa, O. and Ogawa, A. (1996)
Intravesical capsaicin inhibits autonomic dysreflexia in pa-
tients with spinal cord injury. Neurourol. Urodyn., 15: 374.
Igawa, Y., Mattiasson, A. and Andersson, K.E. (1993) Effects
of GABA-receptor stimulation and blockade on micturition
in normal rats and rats with bladder outflow obstruction. J.
Urol., 150: 537–542.

Igawa, Y., Satoh, T., Mizusawa, H., Seki, S., Kato, H.,
Ishizuka, O. and Nishizawa, O. (2003) The role of capsaicin-
sensitive afferents in autonomic dysreflexia in patients with
spinal cord injury. BJU Int., 91: 637–641.
Ishiura, Y., Yoshiyama, M., Yokoyama, O., Namiki, M. and de
Groat, W.C. (2001) Central muscarinic mechanisms regulat-
ing voiding in rats. J. Pharmacol. Exp. Ther., 297: 933–939.
Ishizuka, O., Alm, P., Larsson, B., Mattiasson, A. and And-
ersson, K.E. (1995) Facilitatory effect of pituitary adenylate
cyclase activating polypeptide on micturition in normal, con-
scious rats. Neuroscience, 66: 1009–1014.
Ishizuka, O., Mattiasson, A. and Andersson, K.E. (1996) Role
of spinal and peripheral alpha 2 adrenoceptors in micturition
in normal conscious rats. J. Urol., 156: 1853–1857.
Janig, W. and Morrison, J.F. (1986) Functional properties of
spinal visceral afferents supplying abdominal and pelvic or-
gans, with special emphasis on visceral nociception. Prog.
Brain. Res., 67: 87–114.
Jankowski, R.J., Prantil, R.L., Fraser, M.O., Chancellor, M.B.,
de Groat, W.C., Huard, J. and Vorp, D.A. (2004) Develop-
ment of an experimental system for the study of urethral
biomechanical function. Am. J. Physiol., 286: F225–F232.
Kakizaki, H., Fraser, M.O. and de Groat, W.C. (1997) Reflex
pathways controlling urethral striated and smooth muscle
function in the male rat. Am. J. Physiol., 272: R1647–R1656.
Kamo, I., Cannon, T.W., Conway, D.A., Torimoto, K.,
Chancellor, M.B., de Groat, W.C. and Yoshimura, N.
(2004) The role of bladder-to urethral reflexes in urinary
continence mechanisms in rats. Am. J. Physiol., 287:
F434–F441.
Kamo, I., Torimoto, K., Chancellor, M.B., de Groat, W.C. and
Yoshimura, N. (2003) Urethral closure mechanisms under
sneeze-induced stress condition in rats: a new animal model
for evaluation of stress urinary incontinence. Am. J. Physiol.,
285: R356–R365.
Kawatani, M., Erdman, S. and de Groat, W.C. (1985) Vaso-
active intestinal polypeptide and substance P in afferent
pathways to the sacral spinal cord of the cat. J. Comp.
Neurol., 241: 327–347.
Keast, J.R. and de Groat, W.C. (1992) Segmental distribution
and peptide content of primary afferent neurons innervating
the urogenital organs and colon of male rats. J. Comp.
Neurol., 319: 615–623.
Kim, Y.H., Bird, E.T., Priebe, M. and Boone, T.B. (1997) The
role of oxybutynin in spinal cord injured patients with
indwelling catheters. J. Urol., 158: 2083–2086.
Krenz, N.R., Meakin, S.O., Krassioukov, A.V. and Weaver,
L.C. (1999) Neutralizing intraspinal nerve growth factor
blocks autonomic dysreflexia caused by spinal cord injury. J.
Neurosci., 19: 7405–7414.
Kruse, M.N., Belton, A.L. and de Groat, W.C. (1993) Changes
in bladder and external urethral sphincter function after
spinal cord injury in the rat. Am. J. Physiol., 264:
R1157–R1163.
Kruse, M.N., Bray, L.A. and de Groat, W.C. (1995) Influence
of spinal cord injury on the morphology of bladder afferent
and efferent neurons. J. Auton. Nerv. Syst., 54: 215–224.
81
Kruse, M.N., Noto, H., Roppolo, J.R. and de Groat, W.C.
(1990) Pontine control of the urinary bladder and external
urethral sphincter in the rat. Brain Res., 532: 182–190.
Kuru, M. (1965) Nervous control of micturition. Physiol. Rev.,
45: 425–494.
Lavelle, J.P., Meyers, S.A., Ruiz, W.G., Buffington, C.A.,
Zeidel, M.L. and Apodaca, G. (2000) Urothelial pathophys-
iological changes in feline interstitial cystitis: a human model.
Am. J. Physiol., 278: F540–F553.
Lee, H.Y., Bardini, M. and Burnstock, G. (2000) Distribution
of P2X receptors in the urinary bladder and the ureter of the
rat. J. Urol,, 163: 2002–2007.
Lewis, S.A. (2000) Everything you wanted to know about the
bladder epithelium but were afraid to ask. Am. J. Physiol.,
278: 867–874.
Maggi, C.A. (1993) The dual sensory and efferent functions of the
capsaicin-sensitive primary sensory neurons in the urinary blad-
der and urethra. In: Maggi C.A. (Ed.), The Autonomic Nervous
System, Vol. 3, Chapter 13, Nervous Control of the Urogenital
System. Harwood Academic Publishers, London, pp. 383–422.
Mallory, B., Steers, W.D. and de Groat, W.C. (1989) Electro-
physiological study of micturition reflexes in rats. Am. J.
Physiol., 257: R410–R421.
Mallory, B.S., Roppolo, J.R. and de Groat, W.C. (1991) Phar-
macological modulation of the pontine micturition center.
Brain Res., 546: 310–320.
Matsumoto, G., Hisamitsu, T. and de Groat, W.C. (1995a)
Non-NMDA glutamatergic excitatory transmission in the
descending limb of the spinobulbospinal micturition reflex
pathway of the rat. Brain Res., 693: 246–250.
Matsumoto, G., Hisamitsu, T. and de Groat, W.C. (1995b)
Role of glutamate and NMDA receptors in the descending
limb of the spinobulbospinal micturition reflex pathway of
the rat. Neurosci. Lett., 18: 58–61.
Matsuura, S., Kakizaki, H., Mitsui, T., Shiga, T., Tamaki, N.
and Koyanagi, T. (2002) Human brain region response to
distention or cold stimulation of the bladder: a positron
emission tomography study. J. Urol., 168: 2035–2039.
McMahon, S.B., Armanini, M.P., Ling, L.H. and Phillips, H.S.
(1994) Expression and coexpression of Trk receptors in sub-
populations of adult primary sensory neurons projecting to
identified peripheral targets. Neuron, 12: 1161–1171.
Miura, A., Kawatani, M. and de Groat, W.C. (2001a) Excita-
tory synaptic currents in lumbosacral parasympathetic pre-
ganglionic neurons elicited from the lateral funiculus. J.
Neurophysiol., 86: 1587–1593.
Miura, A., Kawatani, M. and de Groat, W.C. (2001b) Effect of
pituitary adenylate cyclase activating polypeptide on lumbo-
sacral preganglionic neurons in the neonatal rat spinal cord.
Brain Res., 895: 223–232.
Miura, A., Kawatani, M. and de Groat, W.C. (2003) Excitatory
synaptic currents in lumbosacral parasympathetic pregangl-
ionic neurons evoked by stimulation of the dorsal commis-
sure. J. Neurophysiol., 89: 382–389.
Morgan, C., Nadelhaft, I. and de Groat, W.C. (1979) Location
of bladder preganglionic neurons within the sacral parasym-
pathetic nucleus of the cat. Neurosci. Lett., 14: 189–194.
82
Morgan, C., Nadelhaft, I. and de Groat, W.C. (1981) The dis-
tribution of visceral primary afferents from the pelvic nerve
to Lissauer’s tract and the spinal gray matter and its rela-
tionship to the sacral parasympathetic nucleus. J. Comp.
Neurol., 201: 415–440.
Morgan, C.W., de Groat, W.C., Felkins, L.A. and Zhang, S.J.
(1993) Intracellular injection of neurobiotin or horseradish
peroxidase reveals separate types of preganglionic neurons in
the sacral parasympathetic nucleus of the cat. J. Comp. Ne-
urol., 331: 161–182.
Morgan, C.W., Ohara, P.T. and Scott, D.E. (1999) Vasoactive
intestinal polypeptide in sacral primary sensory pathways in
the cat. J. Comp. Neurol., 407: 381–394.
Morrison, J., Steers, W.D., Brading, A., Blok, B., Fry, C.,
de Groat, W.C., Kakizaki, H., Levin, R. and Thor, K. (2002)
Neurophysiology and neuropharmacology. In: Abrams P.,
Cardozo L., Khoury S. and Wein A. (Eds.), Incontinence.
Health Publications Ltd., Jersey, pp. 83–164.
Morrison, J., Wen, J. and Kibble, A. (1999) Activation of pelvic
afferent nerves from the rat bladder during filling. Scand.
J. Urol. Nephrol., 33(Suppl. 201): 73–75.
Nadelhaft, I. and Vera, P.L. (1996) Neurons in the rat brain
and spinal cord labeled after pseudorabies virus injected into
the external urethral sphincter. J. Comp. Neurol., 375:
502–517.
Noto, H., Roppolo, J.R., de Groat, W.C., Nishizawa, O.,
Sugaya, K. and Tsuchida, S. (1991a) Opioid modulation of
the micturition reflex at the level of the pontine micturition
center. Urol. Int., 47(Suppl. 1): 19–22.
Noto, H., Roppolo, J.R., Steers, W.D. and de Groat, W.C.
(1991b) Electrophysiological analysis of the ascending and
descending components of the micturition reflex pathway in
the rat. Brain Res., 549: 95–105.
Novakovic, S.D., Tzoumaka, E., McGivern, J.G., Haraguchi,
M., Sangameswaran, L., Gogas, K.R., Eglen, R.M. and
Hunter, J.C. (1998) Distribution of the tetrodotoxin-resistant
sodium channel PN3 in rat sensory neurons in normal and
neuropathic conditions. J. Neurosci., 18: 2174–2187.
O’Reilly, B.A., Kosaka, A.H., Knight, G.F., Chang, T.K.,
Ford, A.P., Rymer, J.M., Popert, R., Burnstock, G. and
McMahon, S.B. (2002) P2X receptors and their role in female
idiopathic detrusor instability. J. Urol., 167: 157–164.
Palea, S., Artibani, W., Ostardo, E., Trist, D.G. and Pietra, C.
(1993) Evidence for purinergic neurotransmission in the hu-
man urinary bladder affected by interstitial cystitis. J. Urol.,
150: 2007–2012.
Ralevic, V. and Burnstock, G. (1998) Receptors for purines and
pyrimidines. Physiol. Rev., 50: 413.
Rong, W., Spyer, K.M. and Burnstock, G. (2002) Activation
and sensitisation of low and high threshold afferent fibres
mediated by P2X receptors in the mouse urinary bladder.
J. Physiol., 541: 591–600.
Sakakibara, R. and Fowler, C.J. (1999) Cerebral control of
bladder, bowel and sexual function and effects of brain dis-
ease. In: Fowler C.J. (Ed.), Neurology of Bladder, Bowel and
Sexual Dysfunction. Butterworth-Heinemann, Boston,
pp. 229–244.
Sasaki, M. (1994) Morphological analysis of external urethral
and external anal sphincter motoneurones of cat. J. Comp.
Neurol., 349: 269–287.
Sculptoreanu, A. and de Groat, W.C. (2003) Protein kinase C is
involved in neurokinin receptor modulation of N- and L-type
Ca2+ channels in DRG neurons of the adult rat. J. Neuro-
physiol., 90: 21–31.
Seki, S., Erickson, K.A., Seki, M., Nishizawa, O., Chancellor,
M.B., de Groat, W.C. and Yoshimura, N. (2004a) Targeting
spinal neurokinin I receptor-expressing neurons for the treat-
ment of neurogenic detrusor overactivity in spinal cord
injury. J. Urol., 171: 142–143.
Seki, S., Erickson, K.E., Seki, M., Nishizawa, O., Igawa, Y.,
Ogawa, T., de Groat, W.C., Chancellor, M.B. and Yos-
himura, N. (2005) Elimination of lamina I spinal neurons
expressing neurokinin 1 receptor using [Sar9, Met (O2)11]
substance P-saporin reduces bladder overactivity and spinal
c-fos expression induced by intravesical instillation of caps-
aicin. Am. J. Physiol., 288: F466–F473.
Seki, S., Igawa, Y., Kaidoh, K., Ishizuka, O., Nishizawa, O.
and Andersson, K.E. (2001) Role of dopamine D1 and D2
receptors in the micturition reflex in conscious rats.
Neurourol Urodyn., 20: 105–113.
Seki, S., Sasaki, K., Fraser, M.O., Igawa, Y., Nishizawa, O.,
Chancellor, M.B., de Groat, W.C. and Yoshimura, N. (2002)
Immunoneutralization of nerve growth factor in lumbosacral
spinal cord reduces bladder hyperreflexia in spinal cord in-
jured rats. J. Urol., 168: 2269–2274.
Seki, S., Sasaki, K., Igawa, Y., Nishizawa, O., Chancellor,
M.B., de Groat, W.C. and Yoshimura, N. (2004b) Suppres-
sion of detrusor-sphincter dyssynergia by immunoneutrali-
zation of nerve growth factor in lumbosacral spinal cord in
spinal cord injury rats. J. Urol., 171: 478–482.
Sengupta, J.N. and Gebhart, G.F. (1994) Mechanosensitive
properties of pelvic nerve afferent fibers innervating the uri-
nary bladder of the rat. J. Neurophysiol., 72: 2420–2430.
Shea, V.K., Cai, R., Crepps, B., Mason, J.L. and Perl, E.R.
(2000) Sensory fibers of the pelvic nerve innervating the rat’s
urinary bladder. J. Neurophysiol., 84: 1924–1933.
Shefchyk, S.J. (1989) The effects of lumbosacral deafferentation
on pontine micturition centre-evoked voiding in the decere-
brate cat. Neurosci. Lett., 99: 175–180.
Shibata, T., Watanabe, M., Ichikawa, R., Inoue, Y. and
Koyanagi, T. (1999) Different expressions of a-amino-3-
hydroxy-5-methyl-4-isoxazole propionic acid and N-methyl-
D-aspartate receptor subunit mRNAs between visceomotor
and somatomotor neurons of the rat lumbosacral spinal cord.
J. Comp. Neurol., 404: 172–182.
Somogyi, G.T. and de Groat, W.C. (1999) Function, signal
transduction mechanisms and plasticity of presynaptic
muscarinic receptors in the urinary bladder. Life Sci., 64:
411–418.
Somogyi, G.T., Zernova, G.V., Yoshiyama, M., Yamamoto, T.
and de Groat, W.C. (1998) Frequency dependence of mu-
scarinic facilitation of transmitter release in urinary bladder
strips from neurally intact or chronic spinal cord transected
rats. Br. J. Pharmacol., 125: 241–246.

Steers, W.D., Albo, M. and van Asselt, E. (1992a) Effects on
serotonergic agonists on micturition and sexual function in
the rat. J. Drug. Dev. Res., 27: 361–375.
Steers, W.D., Ciambotti, J., Etzel, B., Erdman, S. and de Groat,
W.C. (1991) Alterations in afferent pathways from the
urinary bladder of the rat in response to partial urethral
obstruction. J. Comp. Neurol., 310: 401–410.
Steers, W.D., Creedon, D.J. and Tuttle, J.B. (1996) Immunity
to nerve growth factor prevents afferent plasticity following
urinary bladder hypertrophy. J. Urol., 155: 379–385.
Steers, W.D., Meythaler, J.M., Haworth, C., Herrell, D. and
Park, T.S. (1992b) Effects of acute bolus and chronic con-
tinuous intrathecal baclofen on genitourinary dysfunction
due to spinal cord pathology. J. Urol., 148: 1849–1855.
Stohrer, M., Madersbacher, H., Richter, R., Wehnert, J. and
Dreikorn, K. (1999) Efficacy and safety of propiverine in
SCI-patients suffering from detrusor hyperreflexia — a dou-
ble-blind, placebo-controlled clinical trial. Spinal Cord, 37:
196–200.
Sugaya, K. and de Groat, W.C. (1994) Effects of MK-801 and
CNQX, glutamate receptor antagonists, on bladder activity
in neonatal rats. Brain Res., 640: 1–10.
Sugaya, K., Matsuyama, K., Takakusaki, K. and Mori, S.
(1987) Electrical and chemical stimulations of the pontine
micturition center. Neurosci. Lett., 80: 197–201.
Sugaya, K., Roppolo, J.R., Yoshimura, N., Card, J.P. and de
Groat, W.C. (1997) The central neural pathways involved in
micturition in the neonatal rat as revealed by the injection of
pseudorabies virus into the urinary bladder. Neurosci. Lett.,
223: 197–200.
Szallasi, A. and Fowler, C.J. (2002) After a decade of
intravesical vanilloid therapy: still more questions than
answers. Lancet. Neurol., 1: 167–172.
Thor, K., Kawatani, M. and de Groat, W.C. (1986) Plasticity in
the reflex pathways to the lower urinary tract of the cat during
postnatal development and following spinal cord injury. In:
Goldberger M., Gorio A. and Murray M. (Eds.) Development
and Plasticity of the Mammalian Spinal Cord, Fidia Research
Series, Vol. III. Fidia Press, Padova, Italy, pp. 65–81.
Thor, K.B., Hisamitsu, T. and de Groat, W.C. (1990) Un-
masking of a neonatal somatovesical reflex in adult cats by
the serotonin autoreceptor agonist 5-methoxy-N,N-dimethyl-
tryptamine. Dev. Brain Res., 54: 35–42.
Thor, K.B., Morgan, C., Nadelhaft, I., Houston, M. and de
Groat, W.C. (1989) Organization of afferent and efferent
pathways in the pudendal nerve of the female cat. J. Comp.
Neurol., 288: 263–279.
Torrens, M. and Morrison, J. (1987) The Physiology of the
Lower Urinary Tract. Springer, New York, pp. 1–355.
Vizzard, M.A. (2000) Changes in urinary bladder neurotrophic
factor mRNA and NGF protein following urinary bladder
dysfunction. Exp. Neurol., 161: 273–284.
Vizzard, M.A., Erickson, V.L., Card, J.P., Roppolo, J.R. and
de Groat, W.C. (1995) Transneuronal labeling of neurons in
the adult rat brainstem and spinal cord after injection of
pseudorabies virus into the urethra. J. Comp. Neurol., 355:
629–640.
83
Wein, A.J. (2002) Pathophysiology and categorization of
voiding dysfunction. In: Walsh P.C., Retik A.B., Vaughn
E.D. and Wein A.J. (Eds.), Campbell’s Urology, Chapter 24.
W.B. Saunders Co., Philadelphia, pp. 887–899.
Yokoyama, O., Yoshiyama, M., Namiki, M. and de Groat,
W.C. (2000) Role of the forebrain in bladder overactivity
following cerebral infarction in the rat. Exp. Neurol., 163:
469–476.
Yoshimura, N. (1999) Bladder afferent pathway and spinal
cord injury: possible mechanisms inducing hyperreflexia of
the urinary bladder. Prog. Neurobiol., 57: 583–606.
Yoshimura, N., Bennet, N.E., Phelan, M.W. and de Groat,
W.C. (1999) Effects of chronic nerve growth factor treatment
on rat urinary bladder and bladder afferent neurons. Soc.
Neurosci. Abstr., 25: 1946.
Yoshimura, N. and de Groat, W.C. (1997) Plasticity of Na+
channels in afferent neurones innervating rat urinary bladder
following spinal cord injury. J. Physiol., 503: 269–276.
Yoshimura, N. and de Groat, W.C. (1999) Increased excitabil-
ity of afferent neurons innervating rat urinary bladder after
chronic bladder inflammation. J. Neurosci., 19: 4644–4653.
Yoshimura, N., Erdman, S.L., Snider, M.W. and de Groat,
W.C. (1998a) Effects of spinal cord injury on neurofilament
immunoreactivity and capsaicin sensitivity in rat dorsal root
ganglion neurons innervating the urinary bladder. Neurosci.,
83: 633–643.
Yoshimura, N., Mizuta, E., Kuno, S., Sasa, M. and Yoshida,
O. (1993) The dopamine D1 receptor agonist SKF 38393
suppresses detrusor hyperreflexia in the monkey with
parkinsonism induced by 1-methyl-4-phenyl-1,2,3,6-tetra-
hydropyridine (MPTP). Neuropharmacology, 32: 315–321.
Yoshimura, N., Mizuta, E., Yoshida, O. and Kuno, S. (1998b)
Therapeutic effects of dopamine D1/D2 receptor agonists on
detrusor hyperreflexia in 1-methyl-4-phenyl-1,2,3,6-tetra-
hydropyridine-lesioned parkinsonian monkeys. J. Pharma-
col. Exp. Ther., 286: 228–233.
Yoshimura, N., Sasa, M., Yoshida, O. and Takaori, S. (1990a)
Mediation of micturition reflex by central norepinephrine
from the locus coeruleus in the cat. J. Urol., 143: 840–843.
Yoshimura, N., Sasa, M., Yoshida, O. and Takaori, S. (1990b)
Alpha 1-adrenergic receptor-mediated excitation from the
locus coeruleus of the sacral parasympathetic preganglionic
neuron. Life Sci., 47: 789–797.
Yoshimura, N., Seki, S. and de Groat, W.C. (2001a) Nitric
oxide modulates Ca2+ channels in dorsal root ganglion
neurons innervating rat urinary bladder. J. Neurophysiol.,
86: 304–311.
Yoshimura, N., Seki, S., Erickson, K.A., Erickson, V.L.,
Chancellor, M.B. and de Groat, W.C. (2003) Histological
and electrical properties of rat dorsal root ganglion neurons
innervating the lower urinary tract. J. Neurosci., 23:
4355–4361.
Yoshimura, N., Seki, S., Novakovic, S.D., Tzoumaka, E.,
Erickson, V.L., Erickson, K.A., Chancellor, M.B. and de
Groat, W.C. (2001b) The role of the tetrodotoxin-resistant
sodium channel Nav1.8 (PN3/SNS) in a rat model of visceral
pain. J. Neurosci., 21: 8690–8696.
84
Yoshimura, N., White, G., Weight, F.F. and de Groat, W.C.
(1996) Different types of Na+ and A-type K+ currents in
dorsal root ganglion neurones innervating the rat urinary
bladder. J. Physiol. (Lond.), 494: 1–16.
Yoshiyama, M. and de Groat, W.C. (1997) Effects of in-
trathecal administration of pituitary adenylate cyclase acti-
vating polypeptide (PACAP) on the lower urinary tract in the
rat. Society for Neuroscience Abstracts, 23: 1523.
Yoshiyama, M., Roppolo, J.R. and de Groat, W.C. (1993) Ef-
fects of MK-801 on the micturition reflex in the rat — pos-
sible sites of action. J Pharmacol. Exp. Ther., 265: 844–850.
Yoshiyama, M., Roppolo, J.R. and de Groat, W.C. (1994) Al-
terations by urethane of glutamatergic control of micturition.
Eur. J. Pharmacol., 264: 417–425.
Yoshiyama, M., Roppolo, J.R. and de Groat, W.C. (1995) In-
teractions between NMDA and AMPA/kainate receptors in the
control of micturition in the rat. Eur. J. Pharmacol., 287: 73–78.
Yoshiyama, M., Roppolo, J.R. and de Groat, W.C. (1997)
Effects of LY215490, a competitive alpha-amino-3-hydroxy-
5-methylisoxazole-4-propionic acid (AMPA) receptor antag-
onist, on the micturition reflex in the rat. J. Pharmacol. Exp.
Ther., 280: 894–904.
Zhong, Y., Banning, A.S., Cockayne, D.A., Ford, A.P., Burn-
stock, G. and McMahon, S.B. (2003) Bladder and cutaneous
sensory neurons of the rat express different functional P2X
receptors. Neuroscience, 120: 667–675.
Zoubek, J.A., Somogyi, G.T. and de Groat, W.C. (1993) A
comparison of inhibitory effects of neuropeptide Y on rat
urinary bladder, urethra and vas deferens. Am. J. Physiol.,
265: R537–R543.
Zvarova, K., Dunleavy, J.D. and Vizzard, M. (2005) Changes
in pituitary adenylate cyclase activating polypeptide (PA-
CAP) expression in urinary bladder pathways after spinal
cord injury (SCI). Exp. Neurol., 192: 46–59.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 6

Spinal mechanisms contributing to urethral striated

sphincter control during continence and micturition:
‘‘How good things might go bad’’

Susan J. Shefchyk

Department of Physiology, Faculty of Medicine, University of Manitoba, Winnipeg, MB R3E 3J7, Canada

Abstract: The external urethral sphincter motoneurons in the sacral ventral horn control the striated
external urethral sphincter muscles that circle the urethra. Activity in these motoneurons and muscle
normally contribute to continence but during micturition, when urine must pass through the urethra, the
motoneurons and striated muscle must be silenced. Following injury to descending pathways in the spinal
cord, the ability to inhibit sphincter activity is disrupted or lost, resulting in bladder–sphincter dyssynergia
and functional obstruction of the urethra during voiding. This chapter will first review the various reflex
pathways and neuronal properties that contribute to continence, and which must be modulated during
micturition in the spinal intact animal. A discussion about how the dyssynergia seen with spinal cord injury
may be produced will then be presented.

Discussions about the neural control of the lower
urinary tract usually concentrate on the autonomic
control of the bladder and urethra smooth muscle
with less focus on the role of the striated urethral
sphincter muscle. However, the coordination
between autonomic and somatic motor systems
is fundamental to both continence and micturition.
While the striated sphincter is not under auto-
nomic control per se, the convergence and coor-
dination of various autonomic and somatic
sphincter reflex pathways unite the two systems
closely. For the purpose of this review, the discus-
sion will be limited to the spinal neural control
of the striated external urethral sphincter muscle.
Contraction of the sphincter muscle contributes
to continence by closing the urethra. In contrast,
sphincter muscle activity is suppressed during
micturition when the sacral parasympathetic
bladder preganglionic neurons are activated and
Corresponding author. Tel.: +204-789-3736;
Fax: +204-789-3930; E-mail: sjs@scrc.umanitoba.ca
the urethra is open to allow the passage of
urine from the contracting bladder. The sacral
ventral horn motoneurons innervating the striated
sphincter muscle fibres are thus subject to differ-
ential control during bladder filling and micturit-
ion. The manifestation of abnormal micturition
patterns involving both autonomic and sphincter
somatic systems creates major challenges in
clinical management, particularly in individuals
with spinal cord and supra-spinal disease or
damage.
We will see that the neural control of sphincter
motoneuron output is not unlike that of limb
motoneuron, including the interplay of descending
and segmental reflex pathways as well as major
contributions from the anatomical and electrical
properties of the motoneurons themselves. The
abnormal striated sphincter activity, commonly
referred to as sphincter dyssynergia (Andersen and
Bradley, 1976; Blaivas et al., 1981; Koyanagi et al.,
1982; Kruse et al., 1993; Pikov and Wrathall,
2001), occurs as a consequence of various spinal

DOI: 10.1016/S0079-6123(05)52006-5 85
86
cord and supra-spinal lesions. This review will
address the mechanism(s) contributing to the
recruitment of the sphincter motoneurons and
muscles during continence, the de-recruitment of
these motoneurons and muscle during micturition,
and the possible underlying factors responsible for
the abnormal sphincter motor patterns observed
following spinal cord injury. By first reviewing
what is known about the control of the urethral
sphincter reflex, a groundwork for a discussion
about how good things can go bad can begin.
Continence
In humans, as well as cats and rats, continence is
facilitated by closure of the urethra achieved via
contraction of the smooth muscle of the bladder
neck and urethra and activity of the striated
sphincter muscles surrounding the urethra
(reviewed in de Groat et al., 2001). In addition,
the muscles of the pelvic floor may also contribute
to continence. The source of excitation to the
sphincter motoneurons has been the subject of
investigation for over half a century. Much of the
focus has been on the segmental sensory inputs to
the sphincter motoneurons, a sensory input that
is largely excitatory and associated with perineal
and pudendal cutaneous, urethral (Bradley and
Teague, 1972, 1977; Mackel, 1979; Fedirchuk
et al., 1992; Shefchyk and Buss, 1998; Buss and
Shefchyk, 1999) and pelvic visceral (Garry et al.,
1959; McMahon et al., 1982) afferents. In contrast
to hindlimb motoneurons, there appears to be no
significant monosynaptic afferent input to sphinc-
ter motoneurons (Jankowska et al., 1978; Mackel,
1979; Fedirchuk et al., 1992; Buss and Shefchyk,
1999), and direct feedback from the muscle itself
is almost nonexistent (Chennells et al., 1960;
Lassmann, 1984). In addition, there is no electro-
physiological evidence in sphincter motoneurons
for any crossed or reciprocal inhibition as de-
scribed for other sacral motoneuron populations
(Jankowska et al., 1978). Sphincter motoneurons
have axon collaterals (Sasaki, 1994), but these
appear to terminate directly in the region of the
sphincter motor nucleus and do not project
medially to the region of recurrent inhibitory
interneurons such as Renshaw cells (Hultborn
et al., 1988; Alvarez et al., 1999). This lack of an
anatomical substrate for recurrent inhibition is
consistent with the absence of any evidence for
functional recurrent inhibition of the sphincter
motoneurons (Mackel, 1979). Although a variety
of anatomical and electrophysiological evidence
shows descending inputs from various supra-
spinal regions (Mackel, 1979; Nakagawa, 1980;
Holstege et al., 1987; Hermann et al., 1998), the
functional contributions of these descending path-
ways to the dynamic control of the sphincter
motoneurons is not clear. Figure 1 summarizes the
major inputs to the sphincter motoneurons.
The location and morphology of the ventral
horn motoneurons, sometimes referred to as
external urethral sphincter pudendal motoneu-
rons, innervating the external striated sphincter
muscle in humans, cats, rats and monkeys have
been described in some detail (Onuf, 1899;
Schroder, 1981; Roppolo et al., 1985; McKenna
and Nadelhaft, 1986; Beattie et al., 1990; Sasaki,
1994; Pullen et al., 1997). As mentioned earlier,
these motoneurons have axon collaterals that
project to the region of the cell somas (Sasaki,
1994). Studies have also shown that the dendrites
of these motoneurons form bundles that travel
rostral-caudally between clusters of sphincter
motoneuron somas (Beattie et al., 1990). There is
no evidence to date of gap junctions or coupling
between dendrites or somas.
The passive electrical membrane properties of
cat sphincter motoneurons have been described in
detail by several laboratories (Hochman et al.,
1991; Sasaki, 1991; Shimoda et al., 1992). Sphincter
motoneurons have a very high membrane input
resistance and low rheobase making them easily
recruited. They also display a depolarizing ‘‘sag’’
in the membrane response to hyperpolarization,
which can contribute to rebound excitation fol-
lowing hyperpolarization. Their action potentials
are followed by relatively short duration after
hyperpolarizations, suggesting that they are well
suited to fire tonically. These properties are con-
sistent with the need for a continuous motoneuron
output to drive the sphincter muscles tonically
during the long periods of bladder filling (Garry
et al., 1959; McMahon et al., 1982).

Fig. 1. A summary of the descending and segmental sensory systems described anatomically or physiologically as having inputs to the
external urethral sphincter motoneurons. The size of the arrow reflects the hypothesized functional strength of the input to the
motoneurons.
For over 20 years now non-linear membrane
responses to depolarizing current injection,
referred to as plateau potentials or bistable mem-
brane properties, have been described and studied
in spinal hindlimb motoneurons and interneurons
in a variety of vertebrates (Hultborn and Kiehn,
1992; Morisset and Nagy, 1999). Similar motor
responses have been documented in humans
(Collins et al., 2001). When expressed, these prop-
erties function to amplify or prolong a neuron’s
output in response to a brief excitatory input.
Furthermore, the expression of these properties
may be controlled by a balance of metabotropic
excitatory and inhibitory neuromodulatory sub-
stances. For instance, serotonin and noradrenalin
have been shown to facilitate plateau property
expression in ventral horn motoneurons
(Hounsgaard et al., 1988; Russo et al., 1998),
and activation of metabotropic glutamate type 1
receptors can facilitate their expression in deep
87
dorsal horn neurons (Russo et al., 1997; Morisset
and Nagy, 1999; Derjean et al., 2003) and sacral
preganglionic neurons (Derjean et al., 2005). Other
putative transmitter systems, including substance
P and acetylcholine, have also been implicated in
the facilitation of plateau properties (Delgado-
Lezama et al., 1997; Russo et al., 1997). On the
other hand, metabotropic GABAergic receptors
are thought to suppress plateaux and bistable
properties (Derjean et al., 2003).
More recently, for the first time in cat sphincter
motoneurons, the presence of similar active mem-
brane properties that are sensitive to neuromodu-
latory systems have been described (Paroschy and
Shefchyk, 2000). Paroschy and Shefchyk (2000)
reported that brief trains of perineal or pudendal
afferent stimulation produced sustained firing in
cat sphincter motoneurons. This type of sustained
response had been noted years ago by McMahon
et al. (1982). Using intracellular recordings,
88
Paroschy and Shefchyk (2000) showed that a brief
intracellular depolarizing current injection could
produce a train of action potentials characterized
by an accelerating firing rate, or a membrane de-
polarization, that persisted well beyond the period
of current injection. As with cat hindlimb moto-
neurons, the expression of these non-linear re-
sponses to depolarizing current injection was
facilitated by the intravenous administration of
serotonin precursors or noradrenalin. Paroschy
and Shefchyk (2000) concluded that at least some
proportion of sphincter motoneurons could alter
their excitability and firing characteristics (i.e.,
express non-linear responses manifested as persist-
ent firing or sustained membrane depolarization),
and that such properties could contribute to the
tonic activity of the sphincter motoneurons during
continence. This non-linear property may reduce
the need for a continuous synaptic excitation of
the sphincter motoneuron to maintain a given
muscle force because it could enhance and extend
any periodic excitatory actions of segmental or
descending synaptic inputs during bladder filling.
Micturition
During micturition, the external urethral sphincter
muscle relaxes and the urethra is opened allowing
urine to flow freely. When the sacral parasympa-
thetic bladder preganglionic neurons are recruited,
a parallel system is engaged to ensure that sphincter
motoneurons do not fire and that the sphincter
muscle activity ceases for a period of time during
the void. In humans (Andersen and Bradley, 1976;
Blaivas et al., 1981; Dyro and Yalla, 1986) and
cats (Barrington, 1914; Rampal and Mignard,
1979b; Sackman and Sims, 1990), there is normally
a complete silencing of sphincter muscle activity
during micturition while in the rat, the sphincter
displays a pattern of rhythmic bursting during the
bladder contraction (Kakizaki et al., 1997; Streng
et al., 2004). The mechanism responsible for this
rhythmic activity in the rat has not been identified,
but while it does not appear to be a detriment to
bladder emptying in the rat, such patterns in
humans or cats are considered pathological. This
review will focus on results from the cat model,
which for this purpose more closely mimics human
micturition.
It has been recognized that when the bladder
contracts during micturition and urine flow begins,
the urethra is unobstructed to facilitate the flow of
fluid out of the body. For many years, a variety of
reflex feedback loops recruited during the bladder
contraction and initial flow of urine into the ure-
thra (Barrington, 1914; Garry et al., 1959) were
conceived as responsible for the decrease in sphinc-
ter activity. However, it was not until examination
of the pattern of sphincter activity changes during
filling and micturition in the de-afferented animal
was done (Shefchyk, 1989) did it become evident
that a central micturition circuitry contributed to
the sphincter silencing in the absence of sensory
feedback. That is not to say that segmental reflexes
originating from the bladder and urethra do not
contribute to the sphincter silencing, but that a
central circuitry is also available.
Direct inhibition of sphincter motoneurons
during micturition
The initial question raised was whether the
absence of tonic or evoked sphincter activity in
the cat model during the void was due to a periodic
absence of excitation or an active inhibition of the
motoneurons by the micturition circuitry. Shimoda
et al. (1992) and Fedirchuk et al. (1993), using
intracellular recordings from sphincter motoneurons
during distension- and brainstem-evoked micturition
showed that the membrane of the urethra sphincter
motoneurons hyperpolarized during micturition
when the firing of sphincter motor axons, recorded
in the pudendal nerve, and the sphincter muscle
electromyogram were silenced. Furthermore,
during this hyperpolarization an increase in
membrane conductance could be detected and
intracellular injection of chloride ions could
reverse the hyperpolarization (Fedirchuk et al.,
1993). Together, these data lead to the suggestion
that an active inhibitory chloride conductance at
the motoneuron membrane was contributing to
the inhibition of the motoneurons (Fedirchuk
et al., 1993; Shefchyk, 1998). Shefchyk et al. (1998)
went on to show that in the cat, the sphincter

motoneuron inhibition was sensitive to the
glycinergic antagonist, strychnine, and that the
motoneuron somas and proximal dendrites were
immunopositive for gephyrin, a protein that has
been shown to be associated with glycinergic
receptors in the spinal cord (Fyffe et al., 1995).
Immunohistochemical evidence for GABAergic
terminals on cat sphincter motoneurons has also
been obtained (Ramirez-Leon et al., 1994), but a
functional examination of the GABAergic actions
directly on the motoneurons or sphincter activity
has not been reported. Unfortunately, in the rat
the sphincter system has not been subject to a
similar examination, and it is not known whether
the rhythmic sphincter muscle activity during
micturition is due to: (1) a periodic inhibition of
sphincter motoneurons during voiding, (2) the
expression of intrinsic oscillatory properties in the
motoneurons during voiding, or (3) a combination
of both tonic inhibition of some motoneurons as in
the cat and human, with only a subpopulation
phasically active during voiding.
The membrane hyperpolarization observed in
the cat sphincter motoneurons during micturition
functions to decrease the excitability of the cells by
moving the membrane potential away from the
firing threshold. As the expression of the non-
linear membrane properties described by Paroschy
and Shefchyk (2000) is voltage-dependent (thresh-
old around 43 mV), such membrane hyperpolar-
ization could function to turn off any expressed
non-linear responses (see Fig. 6 in Paroschy and
Shefchyk, 2000). The fact that perineal, pudendal
(Fedirchuk et al., 1994) and urethral afferent-
evoked (Buss and Shefchyk, 1999) excitatory
postsynaptic potentials in sphincter motoneurons
are decreased in amplitude during voiding, when
the membrane conductance of the sphincter moto-
neurons increases is consistent with a conductance
shunting effect to diminish the excitatory poten-
tials. Such postsynaptic inhibition in the spinal
cord is usually mediated by local interneurons.
Evidence for the location and identity of spinal
inhibitory interneurons that may be part of the
circuitry activated during micturition was provid-
ed by Blok et al. (1997), who reported a descend-
ing projection from the brainstem pontine
micturition center to a population of GABAergic
89
neurons in the dorsal commissure of the cat sacral
spinal cord. More recently, Sie et al. (2001)
extended this observation to show that some of
these spinal interneurons were glycinergic. Based
on the facts that these interneurons were positive
for GABA and glycine and that micro-stimulation
in the area of these neurons decreased intra-ure-
thral pressure (Blok et al., 1998), it was hypoth-
esized that these neurons mediated the inhibition
of the sphincter motoneurons during micturition.
Buss and Shefchyk (2003), using extracellular
recordings from single units in the dorsal commis-
sure and deep dorsal horn, identified a group of
neurons that were excited during reflex distension-
evoked or pontine micturition center-evoked void
responses, but not by bladder distension alone.
These neurons may be those described by Blok and
co-workers. In addition, the anatomical identifi-
cation within this region of a variety of neurons
linked to the sphincter motoneurons has been
described using viral tracing methods (Nadelhaft
and Vera, 1996; for review of sacral interneurons,
see Shefchyk, 2001).
Evidence for premotoneuronal inhibition
in excitatory reflex pathways to
sphincter motoneurons
Fedirchuk et al. (1994) observed that polysynaptic
excitatory postsynaptic potentials in hindlimb
motoneurons, within the first sacral segment, were
diminished in amplitude, or completely sup-
pressed, during voiding in the absence of a postsy-
naptic motoneuron membrane hyperpolarization.
This led to the hypothesis that the excitatory path-
ways from the perineal and pudendal afferents
may be gated out during micturition at sites pre-
motoneuronal, that is, at the segmental excitatory
interneurons mediating the excitation or at the
primary afferent terminals themselves.
The possibility that perineal and pudendal pri-
mary afferents were subjected to presynaptic inhi-
bition, or primary afferent depolarization, was
examined by monitoring the excitability of single
identified afferents in the dorsal horn of the sacral
spinal segments during micturition (Angel et al.,
1994; Buss and Shefchyk, 1999). Angel et al. (1994)
90
observed that while bladder distension in the ab-
sence of micturition did not produce excitability
changes in pudendal or perineal afferents, during
distension-evoked micturition reflexes, over one-
third of the afferents tested displayed an increase
in excitability during the void. It was concluded
that primary afferent depolarization could be de-
creasing transmitter release from these afferents
and diminishing the efficacy of the afferent exci-
tatory actions during voiding. Buss and Shefchyk
(1999) examined the perineal and pudendal affer-
ents separately from the urethral pudendal affer-
ents during micturition evoked by both distension
of the bladder and by electrical stimulation of the
pontine micturition center. The results obtained
during both types of evoked micturition were sim-
ilar: 50% of the dorsal penile/clitoral afferents un-
derwent primary afferent depolarization during
the void, while almost 60% of the urethral affer-
ents examined showed excitability increases during
voiding. Biphasic changes in excitability were not-
ed in 4/11 afferents, that is, they underwent pri-
mary afferent depolarization early during the void
then showed a decrease in excitability as the
sphincter activity returned following the void. It
was hypothesized that the spinal circuitry and in-
terneurons mediating presynaptic inhibition of the
perineal, pudendal and urethral sacral afferents
was organized in a way to recognize the different
functions of the perineal/pudendal versus urethral
afferents during the voiding cycle. Specifically,
Buss and Shefchyk (2003) proposed the existence
of at least two subgroups of interneurons recruited
during micturition to accomplish this selective
sensory modulation. The GABA-containing spinal
neurons that Blok and co-workers (Blok et al.,
1997,1998; Sie et al., 2001) described would be ex-
cellent candidates for mediating the presynaptic
inhibition of the primary afferents.
In addition to presynaptic inhibition of primary
afferents, it is also possible that the excitatory in-
terneurons interposed between the primary affer-
ents and the sphincter motoneurons could be the
targets of direct inhibitory inputs during micturit-
ion. Buss and Shefchyk (2003) found two popula-
tions of neurons in the sacral dorsal commissure of
the cat that responded to electrical stimulation
of perineal, pudendal and urethral afferents, but
differed in their activity during micturition. One
group, already mentioned earlier, were recruited
during micturition, and may be the inhibitory
neurons we are discussing. A second population,
which were excited by perineal and pudendal af-
ferents were inhibited during micturition and
could be the excitatory interneurons receiving in-
hibition during micturition (Buss and Shefchyk,
2003).
As summarized in Fig. 2, the central micturition
circuitry has access to a circuitry of spinal neurons
that coordinate various inhibitory pathways dur-
ing micturition. The need to ensure the suppres-
sion of sphincter motoneuron output must be
a high priority for the system as reflected in the
redundancy in the mechanisms responsible for this
inhibition. The relative contributions of each of
these inhibitory systems is not known at this time,
but it can by hypothesized that disruption of any
of these components may contribute to undesired
motoneuron hyperactivity and dyssynergia during
bladder emptying.
So, how might good things go bad?
Depending upon the extent of damage to the white
matter of the spinal cord, a variety of changes may
occur in lower urinary tract function. With supra-
sacral injuries, bladder and sphincter dyssynergia
is commonly encountered and has been character-
ized in humans (Andersen and Bradley, 1976), as
well as cat (Rampal and Mignard, 1976b) and ro-
dent (Pikov and Wrathall, 2001) models. Although
the details may vary somewhat, the common fea-
ture of the dyssynergia is the presence of activity in
the sphincter muscle during bladder contractions,
an activity that effectively limits the amount of
urine that can be expelled. Various attempts to
decrease the neural drive to the sphincter muscle,
including denervating the muscle physically, or
chemically with botulinum toxin (Smith et al.,
2002), are directed to the problem of too much
sphincter tone and reflex activity. But, such ap-
proaches are obviously limited and do not address
the basic issue, that is, the identity of the mech-
anisms contributing to the inappropriate sphincter
motoneuron activity.

Fig. 2. Summary of the inhibitory mechanisms thought to contribute to the suppression of sphincter motoneuron activity during
micturition in the spinal intact animal.
In humans and some of the animal models stud-
ied, the loss of descending pathways controlling
motor neurons, below the spinal cord lesion, results
in loss of voluntary striated muscle control. How-
ever, this is often accompanied by hyper-reflexia and
spasms in the ‘‘paralyzed’’ muscles. For limb motor
systems, the spasms and hyperactivity appear to in-
volve changes in the control of transmission through
excitatory segmental reflex pathways and adapta-
tions in the intrinsic properties of the motoneurons
themselves (for discussion, see Gorassini et al.,
2004). Even with incomplete lesions of the spinal
white matter, in particular the dorsolateral fun-
iculus, the release of a variety of segmental excita-
tory reflexes in motor systems below the level of the
lesion (Cavallari and Petersson, 1989) may contrib-
ute to increased reflex motor output. The striated
sphincter muscle appears to be no different from
limb muscles in terms of the increased tone and
reflex activity following cord injury. Not only is
there an apparent increase in excitability and output
of the sphincter motoneurons in general, but in
addition the mechanisms normally used by the
central micturition circuitry to inhibit sphincter
activity appear to be either absent or greatly
91
disrupted. If we consider the factors discussed so
far in this chapter, we can identify some potential
mechanisms for the abnormal sphincter activity
during micturition following spinal cord injury.
The loss of descending pathways that facilitate
spinal inhibitory circuits may be part of the prob-
lem. The loss of descending pathways that have
tonic inhibitory control over all excitatory seg-
mental reflex pathways, including those to the
sphincter motoneurons, might contribute to in-
creased reflex activation of the sphincter during
both continence and micturition. This increased
activity would demand even more powerful inhi-
bition to prevent the recruitment of sphincter
motoneurons during micturition. Furthermore, a
decrease in excitability of a subset of interneurons
specifically mediating the micturition-related
postsynaptic inhibition of the sphincter motoneu-
rons or of the excitatory interneurons interposed
between segmental afferents and the motoneurons,
could lead to the sphincter dyssynergia during mi-
cturition. We are addressing excitability changes in
the neurons responsible for postsynaptic inhibi-
tion, and these same concerns may be raised for
the spinal interneurons mediating presynaptic
92
inhibition of the primary afferents. It has been
documented that spinal cord presynaptic inhibito-
ry pathways are disrupted with loss of descending
systems (Carpenter et al., 1963; Hongo and
Jankowska, 1967), and this could contribute to
increased reflex transmission through many affer-
ent systems. This possibility has been addressed in
studies using baclofen, an agonist of spinal GABA
ergic systems, to treat spinal cord-injured subjects
with sphincter hyperactivity and dyssynergia
(Hachen and Krucker, 1977). Although the target
may be afferent excitatory pathways to the sphinc-
ter system, the clinical results revealed that this
approach was not selective and can also decrease
bladder afferent transmission to the point of com-
promising the bladder contractions (Steers, 1989).
Turning now to the possible effects of loss of
neuromodulators released from descending sys-
tems, we will consider the modulation of intrinsic
electrical properties and excitability changes in
spinal interneuron and sphincter motoneuron sys-
tems involved in continence and micturition. It has
been established in rat tail motoneurons that
changes in various membrane currents occur fol-
lowing lower lumbar spinal cord injury and that
these changes may contribute to an increased re-
cruitment of motoneurons and spastic motor ac-
tivity in the tail musculature (Li et al., 2004).
Gorassini and co-workers have described results
suggestive of similar mechanisms in limb moto-
neurons of spinal cord injured people (Gorassini
et al., 2004). With the loss of descending serotonergic
and adrenergic systems to the sphincter motoneu-
rons known to promote non-linear properties and
increased motoneuron output (Paroschy and
Shefchyk, 2000), one might predict decreased,
not increased sphincter reflex activity. However,
it is known that substances intrinsic to the spinal
cord may also facilitate the expression of similar
membrane currents and excitability changes. For
instance, acetylcholine can produce plateau poten-
tials (Svirskis and Hounsgaard, 1998) and a strong
intrinsic cholinergic source to the sphincter moto-
neurons may be directly from their own axon
collaterals (Sasaki, 1994). Furthermore, changes in
glutamatergic inputs and receptors following cord
injury (see Llewellyn-Smith et al., 1997), may
result in glutamatergic metabotropic receptors
contributing to excitability changes (Derjean
et al., 2003, 2005). These exciting possibilities
remain to be tested. Whether such systems can
enhance the expression of non-linear membrane
properties and/or membrane excitability in the
neurons remains to be tested in both spinal cord
intact and lesioned animal models. Furthermore,
we must look beyond the motoneurons themselves.
It is known that unidentified interneurons can ex-
press plasticity in their firing characteristics, in-
cluding membrane oscillations and plateau
potentials/enhanced firing (Derjean et al., 2003,
2005). It would be exciting to determine if only
excitatory interneurons have this capability or if
inhibitory spinal neurons also express such plas-
ticity and by what mechanisms. If inhibitory neu-
rons can, then we might develop strategies to
ensure their expression should it be diminished or
lost following spinal cord injury. Regardless of
whether we are considering motoneurons or inter-
neurons, the reality of the presence of plasticity of
firing pattern and excitability should now be care-
fully addressed in both intact and chronic lesion
models for the spinal neurons involved in lower
urinary tract function. We need to determine: (1) if
such properties are expressed in all spinal cell
types; (2) what ion channels and second messenger
systems mediate the plasticity; and (3) which
neuromodulators facilitate and inhibit the chan-
nels and systems involved in the expression of the
property. This knowledge may provide some ex-
citing new insights into selectivity and organiza-
tion of the systems and cellular mechanism.
In summary, the control of the external urethral
sphincter is not simple. The integration of various
segmental inputs, interneuron activity and moto-
neuron properties must be considered when ad-
dressing the changes in sphincter motor function
following spinal cord injury. Although the sphinc-
ter system differs in many ways from limb muscle
control, there are similarities that are significant
and we may do well to consider the well-developed
literature on limb reflex control and spinal cord
injury. The circuitry controlling the sphincter is
largely intrinsic to the spinal cord, and thus may
still be available for manipulation and participa-
tion in the normalization of lower urinary tract
function following spinal cord injury.

References
Alvarez, F.J., Dewey, D.E., McMillin, P. and Fyffe, R.E. (1999)
Distribution of cholinergic contacts on Renshaw cells in the
rat spinal cord: a light microscopic study. J. Physiol.,
515(Part 3): 787–797.
Andersen, J.T. and Bradley, W.E. (1976) The syndrome of
detrusor-sphincter dyssynergia. J. Urol., 116: 493–495.
Angel, M.J., Fyda, D., McCrea, D.A. and Shefchyk, S.J. (1994)
Primary afferent depolarization of cat pudendal afferents
during micturition and segmental afferent stimulation.
J. Physiol., 479: 451–461.
Barrington, F.J.F. (1914) The nervous mechanism of micturit-
ion. Q. J. Exp. Physiol., 8: 33–71.
Beattie, B.S., Li, Q., Leedy, G. and Bresnahan, J.C. (1990)
Motoneurons innervating the external anal and urethral
sphincters of the female cat have different patterns of
dendritic arborization. Neurosci. Lett., 111: 69–74.
Blaivas, J.G., Sinha, H.P., Zayed, A.A.H. and Labib, K.B.
(1981) Detrusor-external sphincter dyssynergia: A detailed
electromyographic study. J. Urol., 125: 545–548.
Blok, B.F.M., de Weerd, H. and Holstege, G. (1997) The pon-
tine micturition center projects to sacral cord GABA
immunoreactive neurons in the cat. Neurosci. Lett., 233:
109–112.
Blok, B.F.M., van Maarseveen, J.T.P.W. and Holstege, G.
(1998) Electrical stimulation of the sacral dorsal gray
commisure evoked relaxation of the external urethral
sphincter in the cat. Neurosci. Lett., 249: 68–70.
Bradley, W.E. and Teague, C.T. (1972) Electrophysiology of
pelvic and pudendal nerves in the cat. Exp. Neurol., 35:
378–393.
Bradley, W.E. and Teague, C.T. (1977) Synaptic events in
pudendal motoneurons of the cat. Exp. Neurol., 56: 237–240.
Buss, R.R. and Shefchyk, S.J. (1999) Excitability changes in
sacral afferents innervating the urethra, perineum and hind-
limb skin of the cat during micturition. J. Physiol., 514:
593–607.
Buss, R.R. and Shefchyk, S.J. (2003) Sacral dorsal horn neu-
rone activity during micturition in the cat. J. Physiol., 551:
387–396.
Carpenter, D., Lundberg, A. and Norsell, U. (1963) Primary
afferent depolarization evoked from the sensorimotor cortex.
Acta Physiol. Scand., 59: 126–142.
Cavallari, P. and Petersson, L.G. (1989) Tonic suppression of
reflex transmission in low spinal cats. Exp. Brain Res., 77:
201–212.
Chennells, M., Floyd, W.F. and Gould, R.P. (1960) Muscle
spindles in the external anal spincter of the cat. J. Physiol.,
151: 23P–25P.
Collins, D.F., Burke, D. and Gandevia, S.C. (2001) Large
involuntary forces consistent with plateau-like behavior of
human motoneurons. J. Neurosci., 21: 4059–4065.
De Groat, W.C., Fraser, M.O., Yoshiyama, M., Smerin, S.,
Tai, C., Chancellor, M.B., Yoshimura, N. and Roppolo, J.R.
(2001) Neural control of the urethra. Scand. J. Urol.
Nephrol. Suppl., 207: 35–43.
93
Delgado-Lezama, R., Perrier, J.-F., Nederfaard, S., Svirskis, C.
and Hounsgaard, J. (1997) Metabotropic synaptic regulation
of intrinsic response properties of turtle spinal motoneu-
rones. J. Physiol., 504: 97–102.
Derjean, D., Bertrand, S., Le Masson, G., Landry, M., Moris-
set, V. and Nagy, F. (2003) Dynamic balance of metabo-
tropic inputs causes dorsal horn neurons to switch functional
states. Nat. Neurosci., 6: 274–281.
Derjean, D., Bertrand, S., Nagy, F. and Shefchyk, S.J. (2005)
Plateau potentials and membrane oscillations in parasympa-
thetic preganglionic neurons and intermediolateral neurons in
the rat lumbo-sacral spinal cord. J. Physiol., 563.2: 583–596.
Dyro, F.M. and Yalla, S.V. (1986) Refractoriness of urethral
striated sphincter during voiding: studies with afferent pudendal
reflex arc stimulation in male subjects. J. Urol., 135: 732–736.
Fedirchuk, B., Downie, J.W. and Shefchyk, S.J. (1994) Reduc-
tion of perineal evoked excitatory postsynaptic potentials in
cat lumbar and sacral motoneurons during micturition.
J. Neurosci., 14: 6153–6159.
Fedirchuk, B., Hochman, S. and Shefchyk, S.J. (1992) An in-
tracellular study of perineal and hindlimb afferent inputs
onto sphincter motoneurons in the decerebrate cat. Exp.
Brain Res., 89: 511–516.
Fedirchuk, B. and Shefchyk, S.J. (1993) Membrane potential
changes in sphincter motoneurons during micturition in the
decerebrate cat. J. Neurosci., 13: 3090–3094.
Fedirchuk, B., Song, L., Downie, J.W. and Shefchyk, S.J.
(1992) Spinal distribution of extracellular field potentials
generated by electrical stimulation of pudendal and perineal
afferents in the cat. Exp. Brain Res., 89: 517–520.
Fyffe, R.E.W., Alvarez, F.J., Harrington, D. and Dewey, D.E.
(1995) Expression of glycine receptors by identified alpha and
gamma motoneurones. In: Taylor A., Gladden M.H. and
Durbaba R. (Eds.), Alpha and Gamma Motor Systems. Ple-
num Press, New York, pp. 421–428.
Garry, R.C., Roberts, T.D.M. and Todd, J.K. (1959) Reflexes
involving the external urethral sphincter in the cat.
J. Physiol., 149: 653–665.
Gorassini, M.A., Knash, M.E., Harvey, P.J., Bennett, D.J. and
Yang, J.F. (2004) Role of motoneurons in the generation of
muscle spasms after spinal cord injury. Brain., 127: 2247–2258.
Hachen, H.J. and Krucker, V. (1977) Clinical and laboratory
assessment of the efficacy of Baclofen (Lioresal(R)) on
urethral sphincter spacticity in patients with traumatic
paraplegia. Eur. Urol., 3: 237–240.
Hermann, G.E., Bresnahan, J.C., Holmes, G.M., Rogers, R.C.
and Beattie, M.S. (1998) Descending projections from the
nucleus raphe obscurus to pudendal motoneurons in the male
rat. J. Comp. Neurol., 397: 458–474.
Hochman, S., Fedirchuk, B. and Shefchyk, S.J. (1991)
Membrane electrical properties of external urethral and
external anal sphincter somatic motoneurons in the
decerebrate cat. Neurosci. Lett., 127: 87–90.
Holstege, G. and Tan, J. (1987) Supraspinal control of
motoneurons innervating the striated muscles of the pelvic
floor including urethral and anal sphincters in the cat. Brain,
110: 1323–1344.
94
Hongo, T. and Jankowska, E. (1967) Effects from the
sensorimotor cortex on the spinal cord in cats with transected
pyramids. Exp. Brain Res., 3: 117–134.
Hounsgaard, J., Hultborn, H., Jespersen, B. and Kiehn, O.
(1988) Bistability of a-motoneurones in the decerebrate cat
and in the acute spinal cat after intravenous 5-hydro-
xytryptophan. J. Physiol., 405: 345–367.
Hultborn, H. and Kiehn, O. (1992) Neuromodulation of ver-
tebrate motor neuron membrane properties. Curr. Opin.
Neurobiol., 2: 770–775.
Hultborn, H., Lipski, J., Mackel, R. and Wigstrom, H. (1988)
Distribution of recurrent inhibition within a motor nucleus.
I. Contribution from slow and fast motor units to the exci-
tation of Renshaw cells. Acta Physiol. Scand., 134: 347–361.
Jankowska, E., Padel, Y. and Zarzecki, P. (1978) Crossed
disynaptic inhibition of sacral motoneurones. J. Physiol.,
285: 425–444.
Kakizaki, H., Fraser, M.O. and de Groat, W.C. (1997) Reflex
pathways controlling urethral striated and smooth muscle
function in the male rat. Am. J. Physiol., 272: R1647–R1656.
Koyanagi, T., Arikado, K., Takamatsu, T. and Tsuji, I. (1982)
Experience with electromyography of the external urethral
sphincter in spinal cord injury patients. J. Urol., 127:
272–276.
Kruse, M.N., Belton, A.L. and de Groat, W.C. (1993) Changes
in bladder and external urethral sphincter function after
spinal cord injury in the rat. Am. J. Physiol., 264:
R1157–R1163.
Lassmann, G. (1984) Muscle spindles and encapsulated nerve
endings in external urethral sphincter. Acta Neuropathol.
(Berl.), 63: 344–346.
Li, Y., Li, X., Harvey, P.J. and Bennett, D.J. (2004) Effects of
baclofen on spinal reflexes and persistent inward currents in
motoneurons of chronic spinal rats with spasticity. J. Neuro-
physiol., 92: 2694–2703.
Llewellyn-Smith, I.J., Cassam, A.K., Krenz, N.R., Krassioukov,
A.V. and Weaver, L.C. (1997) Glutamate- and GABA-
immunoreactive synapses on sympathetic preganglionic neu-
rons caudal to a spinal cord transection in rats. Neuroscience,
80: 1225–1235.
Mackel, R. (1979) Segmental and descending control of the
external urethral and anal sphincters in the cat. J. Physiol.,
294: 105–122.
McKenna, K.E. and Nadelhaft, I. (1986) The organization of
the pudendal nerve in the -male and female rat. J. Comp.
Neurol., 248: 532–549.
McMahon, S.B., Morrison, J.F.B. and Spillane, K. (1982) An
electrophysiological study of somatic and visceral conver-
gence in the reflex control of the external sphincters.
J. Physiol., 328: 379–387.
Morisset, V. and Nagy, F. (1999) Ionic basis for plateau po-
tentials in deep dorsal horn neurons of the rat spinal cord.
J. Neurosci., 19: 7309–7316.
Nadelhaft, I. and Vera, P. (1996) Neurons in the rat brain and
spinal cord labeled after pseudorabies virus injected into the
external urethral sphincter. J. Comp. Neurol., 375: 502–517.
Nakagawa, S. (1980) Onuf’s nucleus of the sacral cord in a
South American monkey (Saimiri): its location and bilateral
cortical input from area 4. Brain Res., 191: 337–344.
Onuf, B. (1899) Notes on the arrangement and function of the
cell groups in the sacral region of the spinal cord. J. Nerv.
Ment. Dis., 26: 498–504.
Paroschy, K.L. and Shefchyk, S.J. (2000) Non-linear membrane
properties of sacral sphincter motoneurones in the decere-
brate cat. J. Physiol., 523: 741–753.
Pikov, V. and Wrathall, J.R. (2001) Coordination of the blad-
der detrusor and the external urethral sphincter in a rat
model of spinal cord injury: effect of injury severity.
J. Neurosci., 21: 559–569.
Pullen, A.H., Tucker, D. and Martin, J.E. (1997) Morpholog-
ical and morphometric characterisation of Onuf’s nucleus in
the spinal cord in man. J. Anat., 191: 201–213.
Ramirez-Leon, V., Ulfhake, B., Arvidsson, U., Verhofstad,
A.A.J., Visser, T.J. and Hokfelt, T. (1994) Serotoninergic,
peptidergic and GABAergic innervation of the ventrolateral
and dorsolateral motor nuclei in the cat s1/s2 segments: an
immunofluorescence study. J. Chem. Neuroanat., 7: 87–103.
Roppolo, J.R., Nadelhaft, I. and de Groat, W.C. (1985) The
organization of pudendal motoneurons and primary afferent
projections in the spinal cord of the Rhesus monkey revealed
by horseradish peroxidase. J. Comp. Neurol., 234: 475–488.
Russo, R.E., Nagy, F. and Hounsgaard, J. (1997) Modulation
of plateau properties in dorsal horn neurones in a slice prep-
aration of the turtle spinal cord. J. Physiol., 499: 459–474.
Russo, R.E., Nagy, F. and Hounsgaard, J. (1998) Inhibitory
control of plateau properties in dorsal horn neurones in the
turtle spinal cord in vitro. J. Physiol., 506: 795–808.
Sackman, J.E. and Sims, M.H. (1990) Electromyographic eval-
uation of the external urethral sphincter during cystometry in
male cats. Am. J. Vet. Res., 51: 1237–1241.
Sasaki, M. (1991) Membrane properties of external urethral
and external anal sphincter motoneurones in the cat.
J. Physiol., 440: 345–366.
Sasaki, M. (1994) Morphological analysis of external urethral
and external anal sphincter motoneurones of cat. J. Comp.
Neurol., 349: 269–287.
Schroder, H.D. (1981) Onuf’s nucleus X: a morphological study
of a human spinal nucleus. Anat. Embryol., 162: 443–453.
Shefchyk, S.J. (1989) The effects of lumbosacral deafferentation
on pontine micturition center-evoked voiding in the decere-
brate cat. Neurosci. Lett., 99: 175–180.
Shefchyk, S.J. (1998) Modulation of excitatory perineal reflexes
and sacral striated sphincter motoneurons during micturition
in the cat. In: Rudomin P., Romo R. and Mendell L.M.
(Eds.), Presynaptic Inhibition and Neural Control. Oxford
University Press, Oxford, pp. 398–406.
Shefchyk, S.J. (2001) Sacral spinal interneurones and the con-
trol of urinary bladder and urethral striated sphincter muscle
function. J. Physiol., 533: 57–63.
Shefchyk, S.J. and Buss, R.R. (1998) Urethral pudendal affer-
ent-evoked bladder and sphincter reflexes in decerebrate and
acute spinal cats. Neurosci. Lett., 244: 137–140.

Shefchyk, S.J., Espey, M.J., Carr, P., Nance, D., Sawchuk, M.
and Buss, R. (1998) Evidence for a strychnine-sensitive
mechanism and glycine receptors involved in the control of
urethral sphincter activity during micturition in the cat. Exp.
Brain Res., 119: 297–306.
Shimoda, N., Takakusaki, K., Nishizawa, O., Tsuchida, S. and
Mori, S. (1992) The changes in the activity of pudendal
motoneurons in relation to reflex micturition evoked in
decerebrate cats. Neurosci. Lett., 135: 175–178.
Sie, J.A., Blok, B.F., de Weerd, H. and Holstege, G. (2001)
Ultrastructural evidence for direct projections from the pon-
tine micturition center to glycine-immunoreactive neurons in
the sacral dorsal gray commissure in the cat. J. Comp.
Neurol., 429: 631–637.
95
Smith, C.P., Somogyi, T. and Chancellor, M.B. (2002) Emerg-
ing role of botulinum toxin in the treatment of neurogenic
and non-neurogenic voiding dysfunction. Curr. Urol. Rep.,
3: 382–387.
Steers, W.D. (1989) Intrathecal baclofen administration —
insights into the regulation of genitourinary function and
therapy. J. Urol., 141: 142.
Streng, T., Santti, R., Andersson, K.-E. and Talo, A. (2004)
The role of the rhabdosphincter in female rat voiding. BJU
Int., 94: 138–142.
Svirskis, G. and Hounsgaard, J. (1998) Transmitter regulation
of plateau properties in turtle motoneurons. J. Neuro-
physiol., 79: 45–50.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 7

Neurochemical plasticity and the role of

neurotrophic factors in bladder reflex pathways after
spinal cord injury

Margaret A. Vizzard

Departments of Neurology and Anatomy and Neurobiology, University of Vermont College of Medicine, Burlington,
VT 05405, USA

Abstract: Transection of the spinal cord that interrupts the spinobulbospinal micturition reflex pathway,
abolishes voluntary voiding and initially produces an areflexic bladder with complete urinary retention.
However, depending upon the species, reflex bladder activity slowly recovers over the course of weeks or
months. In chronic spinal animals, reflex mechanisms in the lumbosacral spinal cord are capable of du-
plicating many of the functions performed by reflex pathways in animals with an intact spinal cord and can
induce bladder hyperreflexia. However, the bladder does not empty efficiently due to a loss of blad-
der–sphincter coordination (bladder–sphincter dyssynergia). In contrast to normal animals in which the
sphincter relaxes during voiding, animals with a spinal cord injury exhibit sphincter contractions during
voiding, an increase in urethral outlet resistance, urinary retention, bladder hyperreflexia, bladder over-
distension, and an increase in bladder afferent cell size. Changes in electrophysiological or neurochemical
properties of bladder afferent cells in the dorsal root ganglia and of spinal pathways could contribute to the
emergence of the spinal micturition reflex, bladder hyperreflexia and changes in the pharmacologic re-
sponses of reflex pathways in the lumbosacral spinal cord after spinal cord injury. Urinary bladder hype-
rreflexia after spinal cord injury may reflect a change in the balance of neuroactive compounds in bladder
reflex pathways. This review will detail: (1) changes in the neurochemical phenotype of bladder afferent
neurons and of spinal neurons mediating micturition reflexes after spinal cord injury, with an emphasis on
three neuroactive compounds, neuronal nitric oxide synthase (nNOS), galanin, and pituitary adenylate
cyclase activating polypeptide (PACAP); (2) possible functional consequences on bladder reflexes of
changes in spinal cord neurochemistry after spinal cord injury, and (3) the potential role of neurotrophic
factors expressed in the urinary bladder or spinal cord after spinal cord injury in mediating these neuro-
chemical changes.

Neural control of micturition
The storage and periodic elimination of urine re-
quires a complex neural control system that coor-
dinates the activities of a variety of effector organs
including the smooth muscle of the urinary bladder
Corresponding author. Tel.: +802-656-3209;
Fax: +802-656-8704; E-mail: margaret.vizzard@uvm.edu
and the smooth and striated muscle of the urethral
sphincters (Kuru, 1965; Klück, 1980; de Groat and
Steers, 1990) (Fig. 1). Three neural pathways reg-
ulate the lower urinary tract (Fig. 1): (1) sacral
parasympathetic (pelvic) nerves provide excitatory
input to the bladder; (2) thoracolumbar sympathetic
(hypogastric) nerves provide an inhibitory input to
the bladder and an excitatory input to the bladder
neck and urethra; and (3) sacral somatic (pudendal)

DOI: 10.1016/S0079-6123(05)52007-7 97
98

Fig. 1. Neuroanatomy of micturition reflex circuitry. Postganglionic neurons that innervate the urinary bladder may be located in
intramural ganglia within the detrusor wall or in some species (e.g., rattus norvegicus), postganglionic neurons are located in pelvic
ganglia located in close proximity to the urinary bladder. From Fernandez (2002) copyright 2002 with permission.

nerves which innervate the striated muscles of the
sphincters and pelvic floor (Kuru, 1965; Klück,
1980; de Groat and Steers, 1990; Middleton and
Keast, 2004). Each of these sets of nerves contains
afferent (sensory) as well as efferent (motor) axons
(Morrison, 1987; Lincoln and Burnstock, 1993).
The central neural pathways controlling the
lower urinary tract exhibit ‘‘all-or-none’’ or
‘‘switch-like’’ characteristics reflecting the storage
and elimination functions of the lower urinary tract
(de Groat and Kruse, 1993; de Groat et al., 1993;
de Groat et al., 1997) (Fig. 2). During urine storage,
somatic and sympathetic pathways to the sphinc-
ters and sympathetic inhibitory inputs to the blad-
der are tonically active, whereas parasympathetic
pathways are inactive (Kuru, 1965; de Groat and
Kruse, 1993; de Groat et al., 1993; de Groat
et al., 1997). During reflex or voluntary micturition,
the activity patterns are reversed such that para-
sympathetic pathways are excited and somatosym-
pathetic pathways are inhibited thereby promoting
urine flow (Middleton and Keast, 2004) (Fig. 2).
The lower urinary tract reflex mechanisms, or-
ganized at the level of the lumbosacral spinal cord,
are modulated predominantly by supraspinal con-
trols (Kuru, 1965; de Groat, 1975; de Groat and
Kruse, 1993; de Groat et al., 1993; de Groat et al.,
1997; Middleton and Keast, 2004). These mecha-
nisms can be summarized as follows: (1) storage re-
flexes (parasympathetic and somatic) are organized
 99

Fig. 2. Diagram illustrating the switch-like properties of the micturition reflexes. During urine storage, a low level of bladder afferent
activity activates efferent input to the external urethral sphincter (sympathetic and somatic nerves). A high level of afferent activity
induced by urinary bladder distention activates the switching circuit in the central nervous system resulting in the activation of efferent
pathways to the urinary bladder detrusor muscle (parasympathetic nerves), inhibition of efferent outflow to the sphincter and urine
outflow. From de Groat (1993) copyright 1993 with permission.

at the spinal level; (2) elimination reflexes (para-
sympathetic) are organized at a supraspinal site in
the pons; and (3) spinal storage reflexes are mod-
ulated by inputs from the rostral pons.
Spinal voiding reflexes after spinal cord injury
Spinal cord injury above the lumbosacral spinal cord
(upper motoneuron injury) alters the coordination
between urinary bladder and sphincter and
chronically impairs micturition in humans, and in
experimental animals (Kuru, 1965; de Groat et al.,
1993). Spinal cord injury produces an initial period
of urinary bladder areflexia that persists for several
weeks to months depending on the species. This
period is followed by the emergence of a micturition
reflex at the spinal level and the appearance of spon-
taneous, involuntary bladder contractions (de Groat
et al., 1993). However, the micturition reflex in
chronically spinalized animals is characterized by si-
multaneous bladder and external urethral sphincter
contractions (bladder–sphincter dyssynergia) leading
to inefficient bladder emptying, large residual urine
volumes and bladder hypertrophy (de Groat et al.,
1993). It has been suggested that bladder–sphincter
dyssynergia results from the loss of brainstem
coordination mechanisms (de Groat et al., 1993).
Emergence of automatic micturition after spinal
cord injury may be dependent on multiple factors,
including: (1) elimination of bulbospinal inhibitory
pathways; (2) strengthening of existing synapses, or
formation of new synaptic connections due to ax-
onal sprouting in the spinal bladder reflex path-
ways; (3) changes in synthesis, release, or action of
neurotransmitters in bladder reflex pathways;
(4) alteration in afferent input from peripheral or-
gans; or (5) central (spinal cord) and peripheral
(urinary bladder) changes in the expression of
neurotrophic factors, which, in turn, would act by
their influence on factors 2–5. A number of labo-
ratories have demonstrated electrical (Yoshimura,
1999) and organization changes (Vizzard, 2000b)
in the central and peripheral components of the
micturition reflex pathways after spinal cord injury
that may underlie the emergence of the spinal mi-
cturition reflex with associated bladder dysfunction.
The changes that occur in spinal voiding reflexes
after spinal cord injury appear to be similar in
humans and experimental animals and are begin-
ning to provide important insights into a variety of
neurogenic disorders of the lower urinary tract
(de Groat and Kruse, 1993; de Groat et al., 1993).
A major breakthrough has been the recognition
that C-fiber bladder afferents can reflexly trigger
100
bladder hyperactivity (de Groat et al., 1981, 1990,
1993). In cats with transected spinal cords, the
properties of C-fiber bladder afferents are altered,
so that they become mechanosensitive and re-
spond to bladder distension (de Groat et al., 1981,
1990, 1993). In chronic spinal cord injury, C-fiber
afferent evoked bladder reflexes emerge. However,
in cats with an intact spinal cord, myelinated (A-q)
afferents activate the micturition reflex (de Groat
and Ryall, 1969; de Groat, 1975; de Groat et al.,
1993). de Groat and colleagues (de Groat et al.,
1990, 1993; de Groat and Kruse, 1993) have dem-
onstrated that systemically administered capsaicin,
a C-fiber neurotoxin, blocked bladder hype-
rreflexia in the chronic paraplegic cat but was
without effect in cats with intact spinal cords. In
the rat, both the spinal and supraspinal micturit-
ion reflexes are activated by capsaicin-resistant Aq
afferents (Mallory et al., 1989); by contrast, caps-
aicin-sensitive afferents do appear to modulate
micturition under certain conditions (Maggi, 1991,
1993). In rats, C-fiber bladder afferents are not
necessary for eliciting bladder reflexes after spinal
cord injury but do contribute to the appearance of
non-voiding bladder contractions (i.e., increases in
bladder pressure not associated with release of
urine) after spinal cord injury (Cheng et al., 1995).
The mechanisms underlying the emergence of the
C-fiber evoked reflex are unknown. However, recent
experiments have begun to examine changes in the
electrical properties of afferent neurons innervating
the urinary bladder of the adult rat before and after
spinal cord injury (Yoshimura, 1999). These studies
suggest an ionic mechanism underlying the relative
inexcitability of C-fiber bladder afferents in normal
animals and the increased excitability of these af-
ferents after spinal cord injury (Yoshimura, 1999).
Changes in the electrophysiological properties of
bladder afferent neurons after spinal cord injury
may occur concomitantly with the changes in ne-
urochemical properties reviewed here.
Neurochemistry and morphology of afferent and
spinal pathways to the urogenital tract
Axonal tracing and immunocytochemical tech-
niques (de Groat et al., 1981, 1986; Donovan et al.,
1983; Steers et al., 1991a; de Groat and Kruse,
1993; Vizzard et al., 1995b; de Groat et al., 1997)
have been used to examine the properties of uro-
genital afferent pathways in the lumbosacral spinal
cord (L6–S1). Anterograde transganglionic tracing
using horseradish peroxidase or wheat germ ag-
glutinin-conjugated horseradish peroxidase re-
vealed that bladder afferents in rats pass through
the dorsal roots into Lissauer’s tract at the apex of
the dorsal horn and then give off collaterals which
extend ventromedially and ventrolaterally along
the superficial layers of the dorsal horn to the
dorsal commissure and to the area of the sacral
parasympathetic nucleus (laminae V–VII), which
contains preganglionic parasympathetic neurons.
These afferents do not extend into the center of the
dorsal horn or into the ventral horn. The most
prominent pathway is located in lamina I on the
lateral edge of the dorsal horn in a region termed
the lateral collateral pathway of Lissauer’s tract.
Afferents from the uterine cervix, and from the
urethra and external urethral sphincter, also
project heavily into the lateral collateral pathway.
In contrast, afferents from the clitoris or penis
project almost exclusively to the dorsal commis-
sure. Thus, it has been concluded that excretory
reflexes depend on spinal processing in the regions
of the lateral collateral pathway and sacral
parasympathetic nucleus, whereas sexual reflexes
are processed in the dorsal commissure.
The neuroactive compounds in the afferent
pathways from the urogenital tract have been
examined using histochemical techniques (Donovan
et al., 1983; de Groat et al., 1986; Keast and de
Groat, 1992; Vizzard et al., 1993b, c, 1994b;
Vizzard and de Groat, 1996). Bladder afferents
contain a variety of neuropeptides, including
calcitonin gene-related peptide, substance P,
vasoactive intestinal polypeptide, cholecystokinin,
and enkephalins (Donovan et al., 1983; de Groat
et al., 1986; Vizzard, 2000d, 2001). Multiple
neuropeptides are present in the same cells sug-
gesting that transmission at afferent terminals in
the spinal cord and in the target organs is likely to
be complex and involve multiple neurotransmitters.
Our recent studies (Zvarova et al., 2004, 2005)
have also demonstrated that bladder afferent cells
express pituitary adenylate cyclase-activating
polypeptide (PACAP) and galanin and that this

expression is increased after chronic cyclophosph-
amide-induced cystitis or spinal cord injury (see
below). With the exception of calcitonin gene-
related peptide, all of these substances are predom-
inantly expressed in small (presumably C-fiber)
afferents (Donovan et al., 1983; de Groat et al.,
1986; Keast and de Groat, 1992; Vizzard et al.,
1993b, c, 1994b; Vizzard and de Groat, 1996).
Neurochemical plasticity in bladder afferent cells in
dorsal root ganglia after spinal cord injury
The neuroactive compounds in the afferent path-
ways from the lower urinary tract and those in the
central pathways (interneurons, preganglionic neu-
rons) exhibit either excitatory or inhibitory ac-
tions. Pathological conditions can alter the known
balance of these neuroactive compounds either in
the periphery or in the central pathways, conceiv-
ably shifting the balance to a hyper- or hypo-active
state. Urinary bladder hyperreflexia, after spinal
cord injury (upper motoneuron injury), may reflect
this change in the balance of neuroactive com-
pounds in bladder reflex pathways.
Nitric oxide
Neuronal nitric oxide synthase
Previous histochemical and pharmacological stud-
ies have raised the possibility that nitric oxide (NO)
is a transmitter in autonomic reflex pathways.
In the rat, nicotinamide adenine dinucleotide
phosphate diaphorase (NADPH-d) activity (a pre-
sumed indicator of the presence of neuronal nitric
oxide synthase) (Dawson et al., 1991; Hope et al.,
1991) and neuronal nitric oxide synthase (nNOS)
immunoreactivity (IR) have been identified in
sympathetic (Valtshanoff et al., 1992) and para-
sympathetic preganglionic neurons (Vizzard et al.,
1993c, 1995a; Saito et al., 1994) in the spinal cord
and in some parasympathetic postganglionic neu-
rons in the peripheral ganglia. In cats, NADPH-d
activity is present in sympathetic preganglionic
neurons but not in parasympathetic preganglionic
neurons (Vizzard et al., 1994c). NADPH-d activity
is also present in a large percentage of visceral
afferent neurons in dorsal root ganglia at various
levels of the spinal cord of the rat (Aimi et al.,
101
1991; Vizzard et al., 1994a). In both the rat and
cat, NADPH-d is present in a prominent afferent
bundle projecting from Lissauer’s tract to the
region of the parasympathetic nucleus (Vizzard
et al., 1993a, c, 1994a, 1995a). This afferent path-
way closely resembles the central projections of the
afferent neurons innervating the pelvic viscera
(Steers et al., 1991a). In the cat, the NADPH-d
afferent pathways closely resemble the vasoactive
intestinal polypeptide-containing afferent projec-
tions to the sacral spinal cord (Basbaum and
Glazer, 1983; Honda et al., 1983; Kawatani et al.,
1985).
Although NADPH-d was present in the primary
afferent neurons (Aimi et al., 1991; Vizzard et al.,
1993b, 1994b) and in their central projections in
the rat (Vizzard et al., 1993a, c) and the cat
(Vizzard et al., 1994a, c), nNOS-IR was not iden-
tified (Vizzard et al., 1994c, 1995a). These data
indicate that in pelvic afferent neurons in normal
rats, NADPH-d is not a marker for nNOS and
that NO is not a transmitter, or that immunore-
activity for nNOS is not as sensitive as NADPH-d
histochemistry. A similar situation exists for so-
matic afferent neurons in the adjacent lumbar
(L4–L5) dorsal root ganglia that contain relatively
large numbers of NADPH-d positive neurons
(Ruda et al., 1994) but few nNOS-IR neurons in
rats (Verge et al., 1992; Zhang et al., 1993). However,
following sciatic nerve injury (Verge et al., 1992;
Zhang et al., 1993), pelvic nerve injury (Vizzard
et al., 1995a), or chronic bladder irritation
(Vizzard and de Groat, 1996) the levels of nNOS
protein or nNOS mRNA are markedly increased
in the lumbosacral dorsal root ganglia. Increased
NADPH-d staining or nNOS-IR has also
been noted following axotomy in motoneurons
(Wu et al., 1994) and in some parasympathetic
preganglionic neurons (Vizzard et al., 1993a, 1995a).
Thus, expression of nNOS is plastic and neurons
that do not normally synthesize nNOS can synthe-
size the protein after injury or chemical stimulation.
The possibility that nNOS participates in the
pathophysiology of spinal cord injury has recently
been investigated (Guizar-Sahagun et al., 1996;
Sharma et al., 1996). Changes in the expression of
nNOS or NADPH-d activity in the myenteric
plexus or thoracic spinal cord after spinal cord
102
injury have been demonstrated (Guizar-Sahagun
et al., 1996). Acute (1 day) spinal cord injury re-
sulted in increased numbers of NADPH-d positive
cell bodies in the myenteric plexus. However, no
increase in NADPH-d positive cell bodies in the
myenteric plexus was observed after chronic (10
weeks) injury (Guizar-Sahagun et al., 1996). Focal
trauma to the dorsal horn of the thoracic
(T10–T11) spinal cord of the rat significantly in-
creased the numbers of nNOS-IR spinal neurons
in the perifocal T9 and T12 segments of the spinal
cord (Sharma et al., 1996). Topical application of
nNOS antiserum, 2 min after injury, prevented the
upregulation of nNOS-IR (Sharma et al., 1996).
These studies have raised the possibility that: (1)
nNOS participates in the pathogenesis of second-
ary spinal damage after spinal cord injury and (2)
changes in nNOS expression in the gastrointestinal
tract may be relevant to its reduced motility after
spinal cord injury.
nNOS expression in lower urinary tract pathways
after spinal cord transection at the 8th thoracic
segment
Following spinal cord injury, nNOS-IR fibers were
detected along the lateral edge of the dorsal horn
extending from Lissauer’s tract to the region of the
sacral parasympathetic nucleus in the lumbar (L6)
and sacral (S1) spinal segments (Vizzard, 1997).
These fibers were not present in the adjacent spinal
segments (L4, L5, or S2) nor were they present at
the rostral lumbar (L1–L2) spinal levels before or
after spinal cord injury. The nNOS-IR fiber stain-
ing in the lateral collateral pathway was not
present in every transverse section suggesting that
the nNOS-IR fibers may occur intermittently
along the rostral–caudal axis as noted for viscer-
al afferent projections labeled with wheat germ
agglutinin horseradish peroxidase or horseradish
peroxidase (Steers et al., 1991a). The general lo-
cation of the nNOS-IR fibers in lamina I and their
selective segmental distribution are very similar to
the central projections of the visceral afferents in
the pelvic nerve, designated the lateral collateral
pathway of Lissauer’s tract (de Groat et al., 1986).
Following spinal cord injury, the numbers
of nNOS-IR cells in the region of the sacral
parasympathetic nucleus in the L6–S1 spinal seg-
ments were significantly increased. In the L6–S1
sacral parasympathetic nucleus of animals with
intact spinal cords, an average of 5.270.4 cell
profiles/section (L6) and 4.870.6 cell profiles/sec-
tion (S1) were nNOS-IR, whereas in animals with
spinal cord injury, the L6–S1 sacral parasympa-
thetic nucleus had an average of 9.370.9 cell pro-
files/section (L6) and 10.370.4 cell profiles/section
(S1) that were nNOS-IR (Vizzard, 1997). In con-
trast, no difference in the numbers of nNOS-IR
cells in the region of the intermediolateral cell nu-
cleus in the L1–L2 segments was detected after
spinal cord injury (8 cell profiles/section in L1
and 6 cell profiles/section in L2) (Vizzard, 1997).
Following complete spinal cord transection, the
number of nNOS-IR neurons increased approxi-
mately 30–70-fold in L6 and S1 dorsal root ganglia
and 2–4-fold in L1 and L2 dorsal root ganglia.
However, there was not a significant change in the
numbers of nNOS-IR cells in the L5 dorsal root
ganglia (Vizzard, 1997). After spinal cord injury,
the increase in nNOS-IR in the L6 and S1 dorsal
root ganglia was highly significant (pp0.001;
12–19 nNOS-IR cell profiles/section) as was the
increase in nNOS-IR in the L1 and L2 dorsal root
ganglia (p p 0.001; 15–40 cell profiles/section)
(Vizzard, 1997). Following spinal cord injury,
nNOS-IR was not restricted to the small and the
medium sized dorsal root ganglia neurons as noted
in control animals. After spinal cord injury,
nNOS-IR was present in the medium (25–30 mm)
and in the large (430 mm) sized dorsal root
ganglion neurons although occasional small
dorsal root ganglion cells still exhibited nNOS-
IR (Vizzard, 1997).
To determine if the increase in nNOS-IR in
dorsal root ganglion neurons was occurring after
spinal cord injury in the urinary bladder afferent
neurons, fluorogold was injected into the urinary
bladder of spinal cord injured animals 5–7 days
prior to euthanasia. In the L6 and S1 dorsal root
ganglia of spinal cord injured animals, an average
of 31.373.0 cell profiles/section and 17.571.4 cell
profiles/section, respectively, were fluorogold-labe-
led after injection of dye into the bladder (Vizzard,
1997). In these ganglia, an average of 41.277.8%
(L6) and 36.370.9% (S1), respectively, of

fluorogold-labeled bladder afferent neurons were
nNOS-IR (Vizzard, 1997). In contrast, no fluoro-
gold-labeled bladder afferent neurons were nNOS-
IR in spinal cord intact animals. At rostral lumbar
levels, a higher percentage of bladder afferents
normally express nNOS-IR in comparison to
lumbosacral dorsal root ganglia (5%). Follow-
ing spinal cord injury, a significantly greater per-
centage of dye-labeled bladder afferents in the L1
and L2 dorsal root ganglia exhibited nNOS–IR
compared to spinal cord intact animals and the
percentage of fluorogold-labeled cells that were
nNOS-IR (20–55%) was similar to the percent-
age of fluorogold nNOS-IR cells in the L6–S1
dorsal root ganglia (Vizzard, 1997).
Role of nitric oxide in lower urinary tract pathways
after spinal cord injury
The function of the NO formed by the enhanced
expression of nNOS in bladder afferent cells and
lumbosacral preganglionic neurons following
spinal cord injury is uncertain. However, the role
of NO in the lower urinary tract after spinal cord
injury may be similar to its suggested role follow-
ing chronic chemical irritation/inflammation of the
urinary bladder (Kakizaki and de Groat, 1996;
Vizzard and de Groat, 1996). Although NO does
not appear to be involved in the normal micturit-
ion reflex in the rat (Rice, 1995; Kakizaki and
de Groat, 1996; Vizzard and de Groat, 1996) NO
does appear to play a role in the facilitation of the
micturition reflex by noxious chemical irritation of
the bladder (Rice, 1995; Kakizaki and de Groat,
1996). Bladder hyperreflexia induced by either
acetic acid (0.1%) (Kakizaki and de Groat, 1996)
or turpentine (Rice, 1995) was partially antago-
nized by intrathecal spinal cord administration of
NOS inhibitors. This suggests that NO is involved
at the spinal level in the facilitation of the mi-
cturition reflex by nociceptive bladder afferents.
The spinal micturition reflex pathway can also
produce bladder hyperreflexia in paraplegic and
urethral obstructed animals. Thus, NO may also
be involved in the facilitation of the spinal
micturition reflex following spinal cord injury.
Increased expression of nNOS in lumbosacral
preganglionic neurons could also contribute to
103
this hyperreflexia at the spinal cord level. The
present results indicate that NO may have addi-
tional functions in bladder primary afferent path-
ways following spinal cord injury. It is possible
that NO may play a role in the sensitization of the
bladder afferents or in changes in the central
processing of afferent input that could contribute
to the pathologically induced alterations in lower
urinary tract function (de Groat et al., 1993;
de Groat and Kruse, 1993).
Pituitary adenylate cyclase activating
polypeptide (PACAP)
PACAP belongs to the vasoactive intestinal
polypeptide/secretin/ glucagon family of bioactive
peptides, and was isolated from the hypothalamus
based on its stimulation of anterior pituitary
adenylyl cyclase activity (Arimura, 1998). Two
p-amidated forms of PACAP arise from alterna-
tive post-translational processing; PACAP38 has
38 amino acid residues [proPACAP(131–168)],
while PACAP27 corresponds to the N-terminus
of PACAP38 [proPACAP(131–157)]. PACAP27
exhibits 68% homology with vasoactive intestinal
polypeptide (Kimura et al., 1990). The relative
levels of the two forms are tissue-specific, although
PACAP38 predominates in most tissues (Arimura,
1998). The rat PACAP precursor protein consists
of 175 amino acid residues (Kimura et al., 1990;
Arimura, 1998; Braas et al., 1998); PACAP38 is
identical among mammalian species, suggesting
similar physiologically important roles for this
peptide. These peptides are abundantly expressed
and have diverse functions as regulators, signaling
modulators, and trophic factors in the nervous and
endocrine systems (Arimura, 1998).
PACAP expression in lower urinary tract pathways
after spinal cord transection at the 8th thoracic
segment
In rats with an intact spinal cord, PACAP is
expressed in nerve fibers in the superficial laminae
of the dorsal horn and dorsal commissure in all
thoracic, lumbar, and sacral segments examined.
Some PACAP staining in the intact spinal cord
was unique to specific levels with PACAP staining
104
being present in the lateral horn in L1–L2, and
L6–S1 spinal segments.
At 6 weeks after spinal cord injury, PACAP–IR
increased in several regions in the rostral lumbar
L1–L2 spinal cord compared to that in rats with
intact spinal cords. The density of PACAP–IR
increased in the superficial laminae (I–II) of the
dorsal horn having a denser distribution through-
out the entire medial (3.0-fold increase) to the lat-
eral (7.0-fold increase) extent of the laminae
(Zvarova et al., 2005). Increased (17.0-fold in-
crease) PACAP–IR fiber staining was also present,
after spinal cord injury, in a small fiber bundle
extending laterally from Lissauer’s tract (LT) in
lamina I into the dorsolateral funiculus (Zvarova
et al., 2005). No dramatic changes in PACAP–IR
were observed in the region of the intermediolat-
eral nucleus following spinal cord injury.
PACAP–IR was unchanged in the L4–L5
segments after spinal cord injury in every region
examined: dorsal horn, dorsal commissure, or
lateral horn regions. In contrast, significant
changes in PACAP–IR were detected in the
L6–S1 spinal cord after spinal cord injury (Fig.
3B,D). In the L6 spinal segment, PACAP–IR was
dramatically increased in the dorsal horn (1.4–7.4-
fold increase), dorsal commissure (11.7-fold in-
crease), sacral parasympathetic nucleus (15.0-fold
increase), and lateral collateral pathway (17.0-fold
increase) (Fig. 3E) (Zvarova et al., 2005). Changes
in PACAP–IR in the S1 segment were comparable
to those in the L6 segment after spinal cord injury
(Fig. 3D) (Zvarova et al., 2005). In some trans-
verse sections of the L6–S1 spinal cord, PACA-
P–IR axons in the lateral collateral pathway
terminated at the base of the dorsal horn (Fig. 3B)
whereas in others, they extended medially toward
the central canal in distinct bundles through
laminae V–VII (Fig. 3D) (Zvarova et al., 2005).
In contrast to PACAP–IR in the spinal cord,
PACAP–IR in the dorsal root ganglia (L1–S1) was
expressed by neuronal cell bodies and fibers
throughout each dorsal root ganglion examined.
In control animals, PACAP–IR was present in
small numbers of cells in the L1–S1 dorsal root
ganglia. The number of PACAP–IR cells among
the dorsal root ganglia examined was comparable
(range 20–24 PACAP–IR cell profiles/section)
(Zvarova et al., 2005). At various times after
spinal cord injury (48 h to 6 weeks), PACAP–IR
was significantly (pp0.001) increased in the rostral
lumbar (L1–L2) and lumbosacral (L6–S1) dorsal
root ganglia (Zvarova et al., 2005). Both small
(16.873.5 mm) and medium (24.072.0 mm) sized
dorsal root ganglion cells expressed PACAP–IR in
animals with intact or injured spinal cords.
PACAP–IR was occasionally observed in larger
(X30 mm) sized dorsal root ganglion cells. No
change in numbers of cells expressing PACAP–IR
was observed in the L4–L5 dorsal root ganglia at
any time after spinal cord injury.
To determine if the increase in PACAP–IR in
the lumbosacral dorsal root ganglia neurons after
cord injury was occurring in bladder afferent cells,
Fast Blue (FB) was injected into the urinary blad-
der to label bladder afferent cells retrogradely in
the L1, L2, L6, S1 dorsal root ganglia (Fig.
4A–4C) (Zvarova et al., 2005). In animals with
intact spinal cords, approximately 45% of bladder
afferent cells in the L6–S1 dorsal root ganglia ex-
hibited PACAP–IR (Fig. 4D). A similar percent-
age (40%) of bladder afferent cells in rostral
lumbar dorsal root ganglia (L1–L2) of control an-
imals also exhibited PACAP–IR. After spinal
cord injury (6 weeks), the percentage of bladder
afferent cells exhibiting PACAP–IR significantly
(p p 0.001) increased in the L6 (88.872.2%) and
S1 dorsal root ganglia (80.272.5%) and in the
L1–L2 dorsal root ganglia (L1, 74.873.5%; L2,
69.573.2%) (Fig. 4D) (Zvarova et al., 2005).
Increases in the percentage of bladder afferent cells
expressing PACAP–IR after spinal cord injury were
observed at the earliest time point after spinal cord
injury (48 h) and were maintained up to 6 weeks
after spinal cord injury with little variation with
time after injury (Fig. 4D) (Zvarova et al., 2005).
PACAP neuronal functions in the lower urinary
tract
PACAP have diverse functions in the endocrine,
nervous, gastrointestinal, and cardiovascular sys-
tems (Braas and May, 1996; Arimura, 1998). High
levels of PACAP and vasoactive intestinal poly-
peptide expression have been identified in many
CNS neurons and in sensory and autonomic gan-
glia (Sundler et al., 1996; Brandenburg et al., 1997;
Fig. 3. Fluorescence photomicrographs showing PACAP–IR in the dorsolateral quadrant of L6 (A,B) and S1 (C,D) spinal segments in control animals (A, C) and after
spinal cord injury (SCI, 6 weeks, B, D). Increased density of PACAP–IR was observed in the medial to lateral extent of the superficial laminae (I–II) of the dorsal horn
(DH) following SCI (A vs. B). Increased PACAP–IR was present in a fiber bundle (B) extending from Lissauer’s tract in lamina I along the lateral edge of the DH to the
region of the sacral parasympathetic nucleus (SPN) (lateral collateral pathway of Lissauer, LCP). Although this fiber bundle was present in control tissue sections, the
staining was less intense (C) and was less frequently observed in transverse sections compared to that after SCI. Faint PACAP–IR was present in the region of the SPN in
control sections (A,C) and was increased after SCI (B,D). Some PACAP–IR fibers in the LCP appeared to terminate in the region of the SPN, whereas others projected
medially toward the central canal (D, arrows). CC, central canal; DCM, dorsal commissure. Calibration bar represents 125 mm. (E). Histogram summarizing changes in
PACAP staining density in specific regions of the L6 spinal cord segment after spinal cord injury (SCI, 6 weeks). The spinal cord inset depicts the areas analyzed: medial
dorsal horn (MDH), lateral dorsal horn (LDH), lateral collateral pathway of Lissauer (LCP), sacral parasympathetic nucleus (SPN), dorsal commissure (DCM), and
ventral horn (VH). The density of PACAP–IR was significantly increased in the LDH, MDH, SPN, DCM, and LCP of the L6 spinal cord segment. Similar changes were
observed in the S1 spinal cord segment. *pp0.001. Reprinted from Zvarova et al. (2005) copyright 2005 with permission from Elsevier.

105
 106
Fig. 4. Effect of spinal cord injury on PACAP expression by dorsal root ganglion cells. PACAP–IR in the L6 dorsal root ganglion (DRG) after spinal cord injury (SCI)
(A, B, C). (A) Fast Blue (FB) labeled bladder afferent cells in a L6 DRG section after SCI. (B) Same L6 DRG section shown in (A) immunostained for PACAP–IR.
PACAP–IR was primarily located in small and medium sized DRG cells. Bladder afferent cells expressing PACAP–IR are indicated by white arrows (A, B). (C) Merged
image of panels (A, B) with FB cells pseudocolored blue and PACAP–IR cells pseudocolored red. FB cells (presumptive bladder afferents) expressing PACAP–IR appear
pinkish-purple (white arrows). Some PACAP–IR cells do not show FB (red cells, yellow arrows). (D) After SCI, a significantly greater percentage (85%) of FB-labeled
bladder afferent cells expressed PACAP–IR at all time points examined; however, not all bladder afferent cells expressed PACAP after SCI. In addition, not all
PACAP–IR in the L6 DRG is accounted for by bladder afferent cells (B,C, yellow arrows). Calibration bar represents 40 mm in (A, B, C). *pp0.001. Reprinted from
Zvarova et al. (2005) copyright 2005 with permission from Elsevier.

Arimura, 1998; Braas et al., 1998). In neurons,
PACAP facilitates calcium ion flux, induces mem-
brane depolarization, increases spike frequency,
activates potently adenylyl cyclase and phospholi-
pase C signaling, and stimulates neurotransmitter
secretion (May and Braas, 1995; Braas and May,
1996, 1999; May et al., 1998; Beaudet et al., 2000).
Several immunocytochemical studies using a spe-
cific monoclonal antibody against PACAP
demonstrate widespread PACAP–IR in nerve
fibers along the rat urinary tract including the
bladder smooth muscle, suburothelial plexuses,
and blood vessels (Fahrenkrug and Hannibal,
1998). Neonatal capsaicin (C-fiber neurotoxin)
treatment significantly reduced this distribution
in adults (Fahrenkrug and Hannibal, 1998) sug-
gesting that the majority of the fibers are derived
from small sensory neurons. Dorsal root ganglia
with high PACAP expression demonstrate dra-
matic neurochemical plasticity during altered
physiological states (Vizzard, 2000d; Zvarova
et al., 2005). Many studies have demonstrated
changes in PACAP expression in sensory neurons
following nerve injury (i.e., axotomy) (Zhang
et al., 1996; Larsen et al., 1997). A limited number
of studies have examined PACAP expression fol-
lowing the induction of inflammatory states. Pre-
vious studies have suggested an association of
PACAP expression with inflammation in sensory
neurons following either somatic (hindpaw) or oc-
ular inflammation (Wang et al., 1996; Zhang et al.,
1998). Our laboratory was the first to demonstrate
an up-regulation of PACAP levels in bladder
afferent cells and spinal cord projections follow-
ing cyclophosphamide-induced cystitis (Vizzard,
2000d) or spinal cord injury (Zvarova et al., 2005).
Cyclophosphamide-induced cystitis is character-
ized by an increased frequency of voiding in awake
rats and by urinary bladder overactivity in anest-
hetized rats (Lecci et al., 1994). Spinal cord injury
rostral to the lumbosacral spinal cord results in
bladder hyperreflexia and bladder–sphincter dys-
synergia (Kruse et al., 1993; Vizzard, 1997, 2000a,
b). As studies have shown that PACAP facilitates
spontaneous bladder contractions in control ani-
mals (Ishizuka et al., 1995), the observed increase
in PACAP expression in bladder afferent cells and
spinal cord projections during cystitis (Vizzard,
107
2000d) and spinal cord injury (Zvarova et al.,
2005) may represent a principal component of
bladder hyperreflexia, by increasing excitability of
sensory neurons in the bladder reflex arc.
Galanin
Previous studies have demonstrated that galanin
has a potent neuromodulatory action on the isolat-
ed human detrusor muscle where galanin suppresses
the cholinergic component of the response to elec-
tric field stimulation (Maggi et al., 1987). Thus, an
inhibitory action for galanin on neurotransmitter
release has been suggested in smooth muscle tissues
and may also pertain to the urinary bladder (Maggi
et al., 1987). Galanin-IR was expressed in identical
spinal cord regions in animals with intact or injured
spinal cords. However, the intensity and the overall
distribution of the staining were increased in specific
spinal cord segments and regions after cord injury
(Zvarova et al., 2004). Increases in galanin expres-
sion in bladder afferent cells in the dorsal root gan-
glia may therefore act to oppose the actions of
PACAP and nNOS, whereas a decrease would re-
inforce these actions in micturition reflex pathways
after spinal cord injury (Vizzard, 1997; Vizzard
et al., 2003). Significant changes in galanin expres-
sion were found after spinal cord injury in specific
regions of the L1, L2, L4, and S1 spinal segments,
suggesting a modulatory role for galanin in spinal
micturition reflex pathways.
Galanin expression in lower urinary tract pathways
after spinal cord transection at the 8th thoracic
segment
Galanin-IR was expressed in identical spinal cord
regions in animals with intact or injured spinal
cords. However, the intensity and the overall dis-
tribution of the staining were increased in specific
spinal cord segments and regions after cord injury
(Zvarova et al., 2004). In the intact spinal cord,
galanin-IR was present in nerve fibers but not in
neuronal cell bodies as in the injured spinal cord.
In the L1 spinal cord segments, the density of
galanin-IR was significantly decreased (pp0.001)
in the superficial laminae (I–II) (2-fold) of the
dorsal horn and in the lateral collateral pathway
108
following spinal cord injury. In contrast, in the L2
spinal segments the only changes in galanin-IR were
increases in the intermediolateral nucleus (Zvarova
et al., 2004). In the L4–L6 segments galanin-IR was
increased in the dorsal commissure region of the L4
segment after cord injury. In the S1 spinal segment,
galanin-IR was increased in the dorsal commissure
(1.2-fold), lateral collateral pathway (1.4-fold), and
sacral parasympathetic nucleus (1.4-fold).
In contrast to galanin-IR in the spinal cord,
galanin-IR in the dorsal root ganglia (L1–S1) was
expressed consistently by both neuronal cell bodies
and fibers (Fig. 5A1, B1). In control animals,
galanin-IR was present in modest numbers of cells
in the L1–S1 dorsal root ganglia (Fig. 5A1, B1, C)
(Zvarova et al., 2004). The number of galanin-IR
cells among the dorsal root ganglia examined was
comparable (range 12–29 galanin-IR cell profiles/
section). At 6 weeks after spinal cord injury,
galanin-IR was significantly (pp0.001) increased
in the rostral lumbar (L1) and sacral (S1) dorsal
root ganglia (Fig. 5A). Both small (17.574.2 mm)
and medium (23.573.5 mm) sized dorsal root
ganglion cells expressed galanin-IR in control
animals following spinal cord injury. No change in
numbers of cells expressing galanin-IR was
observed in the L2, L4–L6 dorsal root ganglia
following spinal cord injury (Fig. 5C) (Zvarova
et al., 2004). We examined galanin-IR after acute
spinal cord injury (o1 week) to determine if we
had missed an earlier increase in galanin expres-
sion that might have returned to control levels by 6
weeks after spinal cord injury. The number of
galanin-IR cells in the dorsal root ganglia after
acute spinal cord injury (o1 week) was not
different from control (Fig. 5C).
To determine if galanin-IR was expressed in
bladder afferent cells, F B dye was injected into the
urinary bladder to retrogradely label bladder
afferent cells in the L1, L2, L6, S1 dorsal root
ganglia (Fig. 5A2, B2). In control animals,
approximately 1.5% of bladder afferent cells in
the L1, L2, L6, or S1 dorsal root ganglia exhibited
galanin-IR (Fig. 5B2, 5D) (Zvarova et al., 2004).
Following spinal cord injury (6 weeks), the
percentage of bladder afferent cells exhibiting
galanin-IR significantly increased in the L1–L2,
L6, and S1 dorsal root ganglia (Fig. 5D).
Role of neurotrophic factors in neuronal
plasticity and lower urinary tract dysfunction after
spinal cord injury
Neurotrophic factors
A potential mechanism underlying the neurochem-
ical changes (Vizzard and de Groat, 1996; Vizzard,
2000b, c, d) in bladder afferent neurons after spinal
cord injury (described above) may involve ne-
urotrophic factors expressed in the urinary bladder
or spinal cord or changes in neural activity
(Vizzard, 2000a). The concept of trophic interac-
tions between nerve cells and their targets is clearly
demonstrated during embryonic or postnatal
development (Oppenheim et al., 1991). Recent
experiments from several laboratories have
demonstrated the influence of target organ–neuron
interactions in the adult animal (Steers and
de Groat, 1988; Steers et al., 1991a, b; Tuttle and
Steers, 1992; Tuttle et al., 1994; Vizzard, 2000a).
A large number of studies have demonstrated
that pathological changes in a target organ after
spinal cord injury can alter the neurochemical
(Vizzard, 1997; Yoshimura, 1999; Vizzard et al.,
2003), electrical (Yoshimura, 1999), and organiza-
tional (Vizzard, 2000b) properties of micturition
reflex pathways. A possible mechanism underlying
these changes may involve neurotrophic factors or
neural activity arising in the bladder. Previous
experiments have demonstrated target organ to
neuron interactions in the adult animal (Steers and
de Groat, 1988; Steers et al., 1991a, b,1996; Tuttle
et al., 1994; Zvara et al., 2002). Furthermore,
a recent study from this laboratory has demon-
strated changes in mRNA or protein expression
of neurotrophic factors in the urinary bladder after
complete spinal cord injury, including nerve
growth factor, brain-derived neurotrophic factor,
glial-derived neurotrophic factor, neurotrophin-3
and -4 (Vizzard, 2000a). Both acute and chronic
spinal cord injury (4–6 weeks) resulted in signifi-
cant increases in nerve growth factor, brain-derived
neurotrophic factor, glial-derived neurotrophic
factor, neurotrophin-3 and -4 transcript expres-
sion as well as in increased nerve growth factor
protein expression in urinary bladder (Vizzard,
2000a).
 109

Fig. 5. (A, B) Bladder afferent cells in lumbosacral dorsal root ganglia (DRG) express galanin (Gal)-immunoreactivity (IR). (A1, B1).
Fluorescence photographs of Gal-IR cells (arrows and arrowheads) in the L1 DRG from control (A1) and spinal cord injured (SCI)
(B1) rats. Bladder afferent cells in the DRG were labeled by retrograde transport of Fast Blue (FB; A2, B2). Some bladder afferent cells
express Gal-IR before (A1, A2, arrowheads) and after SCI (B1, B2, arrowheads). Note that not all bladder afferent cells express Gal-
IR (B2, arrow) and not all Gal-IR cells are bladder afferent cells (A1, B1; arrows). Calibration bar represents 100 mm. (C) Histogram
depicting the number of Gal-IR DRG cells per section in DRG examined in control (spinal intact) and SCI rats 6 weeks or less than 1
week (o week) after SCI. No changes in the numbers of Gal-IR cells in DRG examined were observed less than 1 week after SCI. In
contrast, significant increases in the numbers of Gal-IR cells in the L1 and S1 DRG were observed 6 weeks after SCI. (D) Histogram
depicting the percentage of bladder afferent cells in the DRG expressing Gal-IR in control or SCI rats. Six weeks after SCI, the
percentage of bladder afferent cells expressing Gal-IR significantly increased in all DRG examined. *pp0.001. Reprinted from
Zvarova et al. (2004) copyright 2004 with permission.

It has also been reported that nerve growth fac-
tor levels increase in the transected spinal cord
(Krenz and Weaver, 2000; Brown et al., 2004) after
spinal cord injury. Our work (Zvarova et al., 2004)
also demonstrated a significant increase in nerve
growth factor content in spinal segments immedi-
ately rostral to the T8 transection site (Fig. 6A).
However, some of the spinal segments caudal to
the transection site (T9–T10; T13–L1; L6–S1)
exhibit decreased nerve growth factor protein
110

Fig. 6. (A). Changes in total spinal cord nerve growth factor (NGF) as detected with a NGF ELISA (Enzyme Linked Immunosorbant
Assay) after spinal cord injury (SCI) (o 1 week or 6 weeks). The line drawing at the top represents the spinal cord and the vertical line
indicates the position of the spinal cord transection at (T8). A significant increase in total NGF in the T7–T8 spinal segments was
present 6 weeks after SCI. Significant decreases in total NGF in the T9–T10, T13–L1, and L6–S1 spinal segments were present after
acute or chronic SCI compared to control values. *pp0.001. (B) Changes in total spinal cord brain-derived neurotrophic factor
(BDNF) as detected with a BDNF ELISA after SCI o1 week or 6 weeks. The line drawing at the top represents the spinal cord and the
vertical line indicates the position of the spinal cord transection at (T8). A significant increase in total BDNF from the T7–T8, T9–T10,
T13–L1, and L6–S1 spinal segments were detected after acute (o1 week) or chronic (6 weeks) SCI. Significant increases in spinal cord
BDNF were also seen acutely after SCI in the T11–T12 and L4–L5 spinal segments but not after chronic SCI. *pp0.001. Reprinted
from Zvarova et al. (2004) copyright 2004 with permission.

content with acute or chronic spinal cord injury
(Fig. 6A). In contrast, brain-derived neurotrophic
factor protein content significantly increased in the
majority of spinal segments examined (Fig. 6B).
No spinal segments exhibited a decrease in brain-
derived neurotrophic factor protein content after
spinal cord injury (Fig. 6B). Thus, bladder afferent
neurons may have at least two potential sources of
increased brain-derived neurotrophic factor fol-
lowing spinal cord injury: (1) central terminals in
the spinal cord and (2) peripheral terminals in the
urinary bladder (Vizzard, 2000a). The results of
these studies may focus new attention on the po-
tential role of brain-derived neurotrophic factor in
micturition reflex plasticity after spinal cord injury.
A model of nerve growth factor-dependent
sensory consequences of tissue damage and in-
flammation in the somatic system has been pro-
posed (Lewin and Mendell, 1993). This proposed
scheme has been modified for this review to dem-
onstrate how an excess of nerve growth factor or
another neurotrophic factor in the urinary bladder
or spinal cord could alter lower urinary tract
pathways after spinal cord injury, and is shown in
Fig. 7. In addition to peripheral afferent changes
after spinal cord injury, it is clear that central reflex
mechanisms are also changed.
Fig. 7. Proposed involvement of urinary bladder or spinal cord neurotrophic factors in plasticity of micturition reflexes after spinal
cord injury (SCI).
111
At sites of tissue injury, inflammation or target
organ hypertrophy, cytokines and growth factors
are up-regulated and this can result in the up-
regulation of nerve growth factor (Lewin and
Mendell, 1993; Woolf et al., 1997) (Fig. 7). Nerve
growth factor activates TrkA receptors on axon
terminals in the urinary bladder or spinal cord re-
sulting in internalization and retrograde transport
of activated TrkA (Kuruvilla et al., 2004) to affer-
ent cells in the dorsal root ganglia. Excess nerve
growth factor within the dorsal root ganglia may
induce increased production of neuropeptides
(i.e., substance P, calcitonin gene-related peptide
and PACAP) in sensory neurons (Gary and
Hargreaves, 1992; Woolf et al., 1997) (Fig. 7). An
increase in the levels of neuroactive compounds
(e.g., enkephalin (Lewin and Mendell, 1993),
dynorphin (Ruda et al., 1988), calcitonin gene-
related peptide (Gary and Hargreaves, 1992; Woolf
et al., 1997; Vizzard, 2001), substance P (Ruda
et al., 1988; Gary and Hargreaves, 1992; Lewin and
Mendell, 1993; Vizzard, 2001), neuropeptide Y
(Lewin and Mendell, 1993), nNOS (Vizzard et al.,
1995a; Vizzard and de Groat, 1996; Vizzard, 1997),
and PACAP (Jongsma et al., 2000; Vizzard, 2000d)
following noxious peripheral stimulation, cyclophos-
phamide-induced cystitis (Vizzard, 2000c, d, 2001;
112
Vizzard and de Groat, 1996), or spinal cord injury
(Vizzard and de Groat, 1996; Vizzard, 2000b) has
also been demonstrated in dorsal root ganglion cells
as well as in spinal cord neurons. Furthermore, in-
travesical administration of exogenous nerve growth
factor in animals may facilitate afferent firing and
induce bladder hyperreflexia that is blocked by anti-
nerve growth factor treatment (Dmitrieva
et al., 1997). Over-expression of nerve growth fac-
tor in bladder smooth muscle in spontaneously hy-
pertensive rats leads to bladder hyperinnervation
and bladder overactivity (Clemow et al., 1998).
Immunoneutralization of nerve growth factor in the
lumbosacral spinal cord suppresses detrusor–sphinc-
ter dyssynergia in spinal cord-injured rats (Seki et
al., 2004). Thus, changes in the expression of ne-
urotrophic factors after spinal cord injury may result
in changes in the neurochemical phenotype of blad-
der afferent cells in dorsal root ganglia as well as
contribute to urinary bladder dysfunction (Fig. 7).
Conclusions
Spinal cord injury rostral to the lumbosacral spinal
cord (upper motoneuron injury) results in dramatic
changes in the neurochemistry of peripheral and
central micturition reflex pathways. Changes in the
neurochemical properties of these neurons after spi-
nal cord injury may be mediated by pathological
changes in the target organ (i.e., urinary bladder) or
spinal cord and contribute to urinary bladder hype-
rreflexia. Neurotrophic factors expressed in the
hypertrophied urinary bladder or spinal cord after
spinal cord injury are likely to contribute to this
neurochemical plasticity. Distinguishing between
beneficial or detrimental effects of these neuroactive
compounds and neurotrophic factors in the context
of micturition reflexes or regenerative responses will
be a challenge but is essential to understanding the
effects of therapies directed at blocking the effects of
neuroactive compounds or neurotrophic factors.
Acknowledgments
This work was supported by NIH grants
DK051369, NS040796, DK065989, DK060481.
Gratitude is expressed to Dr. William C. de Groat,
University of Pittsburgh, for his tremendous
mentorship and support. Gratitude is also ex-
pressed to Dr. Victor May and Dr. Karen Braas,
University of Vermont, for encouraging me to
explore the role of PACAP in lower urinary tract
reflexes. Many former and current members of
my laboratory have contributed to the studies
discussed in this review; including Dr. Li-ya Qiao,
Dr. Katarina Zvarova, Mr. Dana J. Dunleavy,
Ms. Elaine Murray, and Ms. Susan Malley.
References
Aimi, Y., Fujimura, M., Vincent, S.R. and Kimura, H. (1991)
Localization of NADPH-diaphorase containing neurons in
sensory ganglia of the rat. J. Comp. Neurol., 306: 382–392.
Arimura, A. (1998) Perspectives on pituitary adenylate cyclase
activating polypeptide (PACAP) in the neuroendocrine, en-
docrine, and nervous systems. Jpn. J. Physiol., 48: 301–331.
Basbaum, A.I. and Glazer, E.J. (1983) Immunoreactive vaso-
active intestinal polypeptide is concentrated in the sacral spi-
nal cord: a possible marker for pelvic visceral afferent fibers.
Somatosens. Res., 1: 69–82.
Beaudet, M.M., Parsons, R.L., Braas, K.M. and May, V.
(2000) Mechanisms mediating pituitary adenylate cyclase-ac-
tivating polypeptide depolarization of rat sympathetic neu-
rons. J. Neurosci., 20: 7353–7361.
Braas, K.M. and May, V. (1996) Pituitary adenylate cyclase-ac-
tivating polypeptides, PACAP-38 and PACAP-27, regulation of
sympathetic neuron catecholamine, and neuropeptide Y expres-
sion through activation of type I PACAP/VIP receptor iso-
forms. Ann. N. Y. Acad. Sci., 805: 204–216 discussion 217–218.
Braas, K.M. and May, V. (1999) Pituitary adenylate cyclase-
activating polypeptides directly stimulate sympathetic neuron
neuropeptide Y release through PAC(1) receptor isoform
activation of specific intracellular signaling pathways. J. Biol.
Chem., 274: 27702–27710.
Braas, K.M., May, V., Harakall, S.A., Hardwick, J.C. and
Parsons, R.L. (1998) Pituitary adenylate cyclase-activating
polypeptide expression and modulation of neuronal excitabil-
ity in guinea pig cardiac ganglia. J. Neurosci., 18: 9766–9779.
Brandenburg, C.A., May, V. and Braas, K.M. (1997) Identi-
fication of endogenous sympathetic neuron pituitary adeny-
late cyclase-activating polypeptide (PACAP): depolarization
regulates production and secretion through induction of
multiple propeptide transcripts. J. Neurosci., 17: 4045–4055.
Brown, A., Ricci, M.J. and Weaver, L.C. (2004) NGF message
and protein distribution in the injured rat spinal cord. Exp.
Neurol., 188: 115–127.
Cheng, C.L., Ma, C.P. and de Groat, W.C. (1995) Effect of
capsaicin on micturition and associated reflexes in chronic
spinal rats. Brain Res., 678: 40–48.
Clemow, D.B., Steers, W.D., McCarty, R. and Tuttle, J.B.
(1998) Altered regulation of bladder nerve growth factor and
neurally mediated hyperactive voiding. Am. J. Physiol., 44:
R1279–R1286.

Dawson, T.M., Bredt, D.S., Fotuhi, M., Hwang, P.M. and
Snyder, S.H. (1991) Nitric oxide synthase and neuronal
NADPH diaphorase are identical in brain and peripheral
tissues. Proc. Natl. Acad. Sci., 88: 7797–7801.
de Groat, W.C. (1975) Nervous control of the urinary bladder
of the cat. Brain Res., 87: 201–211.
de Groat, W.C. (1993) Anatomy and physiology of the lower
urinary tract. Urol. Clin. North Am., 20: 383–401.
de Groat, W.C., Booth, A.M. and Yoshimura, N. (1993)
Neurophysiology of micturition and its modification in
animal models of human disease. In: Maggi C.A. (Ed.) The
Autonomic Nervous System, Vol. 3. Harwood Academic
Publishers, London, pp. 227–290.
de Groat, W.C., Kawatani, M., Hisamisu, T., Cheng, C.-L.,
Ma, C.P., Thor, K., Steers, W. and Roppolo, J.R. (1990)
Mechanisms underlying the recovery of urinary bladder
function following spinal cord injury. J. Auton. Nerv. Syst.,
30: S71–S78.
de Groat, W.C., Kawatani, M., Hisamitsu, T., Booth, A.M.,
Roppolo, J.R., Thor, K., Tuttle, P. and Nage, J. (1986)
Neural control of micturition: the role of neuropeptides.
J. Auton. Nerv. Syst., 9(Suppl): 369–387.
de Groat, W.C. and Kruse, M.N. (1993) Central processing
and morphological plasticity in lumbosacral afferent path-
ways from the lower urinary tract. In: Mayer E.A. and
Raybould H.E. (Eds.) Basic and Clinical Aspects of
Chronic Abdominal Pain. Pain Research and Clinical
Management, Vol. 9. Elsevier Science Publishers, Amster-
dam, pp. 219–236.
de Groat, W.C., Kruse, M.N., Vizzard, M.A., Cheng, C.-L.,
Araki, I. and Yoshimura, N. (1997) Modification of urinary
bladder function after spinal cord injury. In: Seil F.J. (Ed.)
Advances in Neurology, Vol. 72. Raven Press, New York,
pp. 347–364.
de Groat, W.C., Nadelhaft, I., Milne, R.J., Booth, A.M.,
Morgan, C. and Thor, K. (1981) Organization of the sacral
parasympathetic reflex pathways to the urinary bladder and
large intestine. J. Auton. Nerv. Syst., 3: 135–160.
de Groat, W.C. and Ryall, R.W. (1969) Reflexes to sacral
parasympathetic neurones concerned with micturition in the
cat. J. Physiol. (London), 200: 87–108.
de Groat, W.C. and Steers, W.D. (1990) Autonomic regulation
of the urinary bladder and sex organs. In: Loewy A.D. and
Spyer K.M. (Eds.), Central Regulation of Autonomic Func-
tions. Oxford University Press, London, pp. 310–333.
Dmitrieva, N., Shelton, D., Rice, A.S.C. and McMahon, S.B.
(1997) The role of nerve growth factor in a model of visceral
inflammation. Neuroscience, 78: 449–459.
Donovan, M.K., Winternitz, S.R. and Wyss, J.M. (1983) An
analysis of the sensory innervation of the urinary system of
the rat. Brain Res. Bull., 11: 321–324.
Fahrenkrug, J. and Hannibal, J. (1998) Pituitary adenylate
cyclase activating polypeptide immunoreactivity in capsaicin-
sensitive nerve fibres supplying the rat urinary tract. Neuro-
science, 83: 1261–1272.
Fernandez, O. (2002) Mechanisms and current treatments of uro-
genital dysfunction in multiple sclerosis. J. Neurol., 249: 1–8.
113
Gary, M.B. and Hargreaves, K.M. (1992) Enhanced release of
immunoreactive CGRP and substance P from spinal dorsal
horn slices occurs during carrageenan inflammation. Brain
Res., 582: 139–142.
Guizar-Sahagun, G., Garcia-Lopez, P., Espitia, A.L., Mendez,
S., Castaneda-Hernandez, G., Madrazo, I. and Franco-
Bourland, R.E. (1996) Histochemical evidence of the in-
creased expression of nicotinamide adenine dinucleotide
phosphate-dependent diaphorase in neurons of the myenter-
ic plexus after acute spinal cord injury in adult rats. Neurosci.
Lett., 206: 185–188.
Honda, C.N., Rethelyi, M. and Petrusz, P. (1983) Preferential
immunohistochemical localization of vasoactive intestinal po-
lypeptide (VIP) in the sacral spinal cord of the cat: light and
electron microscopic observations. J. Neurosci., 3: 2183–2196.
Hope, B.T., Michael, G.J., Jnigge, K.M. and Vincent, S.R.
(1991) Neuronal NADPH diaphorase is a nitric oxide synt-
hase. Proc. Natl. Acad. Sci., 88: 2811–2814.
Ishizuka, O., Alm, P., Larsson, B., Mattiasson, A. and
Andersson, K.E. (1995) Facilitatory effect of pituitary adeny-
late cyclase-activating polypeptide on micturition in normal,
conscious rats. Neuroscience, 66: 1009–1014.
Jongsma, H., Danielsen, N., Sundler, F. and Kanje, M. (2000)
Alteration of PACAP distribution and PACAP receptor
binding in the rat sensory nervous system following sciatic
nerve transection. Brain Res., 853: 186–196.
Kakizaki, H. and de Groat, W.C. (1996) Role of nitric oxide in
the facilitation of the micturition reflex by bladder irritation.
J. Urol., 155: 355–360.
Kawatani, M., Lowe, I., Nadelhaft, I., Morgan, C. and
de Groat, W.C. (1985) Vasoactive intestinal polypeptide
and substance P in primary afferent pathways to the sacral
spinal cord of the cat. J. Comp. Neurol., 241: 311–316.
Keast, J.R. and de Groat, W.C. (1992) Segmental distribution
and peptide content of primary afferent neurons innervating
the urogenital organs and colon of male rats. J. Comp.
Neurol., 319(4): 615–623.
Kimura, C., Ohkubo, S., Ogi, K., Hosoya, M., Itoh, Y., Onda,
H., Miyata, A., Jiang, L., Dahl, R.R., Stibbs, H.H., Arimura,
A. and Fujino, M. (1990) A novel peptide which stimulates
adenylate cyclase: molecular cloning and characterization
of the ovine and human cDNAs. Biochem. Biophys. Res.
Commun., 166: 81–89.
Klück, P. (1980) The autonomic innervation of the human uri-
nary bladder neck and urethra: a histochemical study. Anat.
Rec., 198: 439–447.
Krenz, N.R. and Weaver, L.C. (2000) Nerve growth factor in
glia and inflammatory cells of the injured rat spinal cord.
J. Neurochem., 74: 730–739.
Kruse, M.N., Bray, L.A. and de Groat, W.C. (1993) Influence
of spinal cord injury on the morphology of bladder afferent
and efferent neurons. J. Auton. Nerv. Syst., 54: 215–224.
Kuru, M. (1965) Nervous control of micturition. Physiol. Rev.,
45: 425–494.
Kuruvilla, R., Zweifel, L.S., Glebova, N.O., Lonze, B.E., Valdez,
G., Ye, H. and Ginty, D.D. (2004) A neurotrophin signaling
cascade coordinates sympathetic neuron development through
114
differential control of TrkA trafficking and retrograde signa-
ling. Cell, 118: 243–255.
Larsen, J.O., Hannibal, J., Knudsen, S.M. and Fahrenkrug, J.
(1997) Expression of pituitary adenylate cyclase-activating
polypeptide (PACAP) in the mesencephalic trigeminal
nucleus of the rat after transection of the masseteric nerve.
Mol. Brain Res., 46: 109–117.
Lecci, A., Giulani, S., Santiciolo, P. and Maggi, C.A. (1994)
Involvement of spinal tachykinin NK1 and NK2 receptors in
detrusor hyperreflexia during chemical cystitis in anaesthe-
tized rats. Eur. J. Pharm., 259: 129–135.
Lewin, G.R. and Mendell, L. (1993) Nerve growth factor and
nociception. Trends Neurosci., 16: 353–359.
Lincoln, J. and Burnstock, G. (1993) Autonomic innervation of
the urinary bladder and urethra. In: Maggi C.A. (Ed.) The
Autonomic Nervous System, Vol. 3. Harwood Academic
Publishers, London, pp. 33–68.
Maggi, C.A. (1991) The role of neuropeptides in the regulation of
the micturition reflex: An update. Gen. Pharmacol., 22: 1–24.
Maggi, C.A. (1993) The dual, sensory and efferent function of
the capsaicin-sensitive primary sensory nerves in the bladder
and urethra. In: Maggi C.A. (Ed.) The Autonomic Nervous
System, Vol. 3. Nervous Control of the Urogenital System.
Harwood Academic Publishers, London.
Maggi, C.A., Santicioli, P., Patacchini, R., Turini, D., Barbanti,
G., Beneforti, P., Giuliani, S. and Meli, A. (1987) Galanin: a
potent modulator of excitatory neurotransmission in the hu-
man urinary bladder. Eur. J. Pharmacol., 143: 135–137.
Mallory, B., Steers, W.D. and de Groat, W.C. (1989) Electro-
physiological study of micturition reflexes in rats. Am.
J. Physiol., 257: R410–R421.
May, V., Beaudet, M.M., Parsons, R.L., Hardwick, J.C.,
Gauthier, E.A., Durda, J.P. and Braas, K.M. (1998) Mech-
anisms of pituitary adenylate cyclase activating polypeptide
(PACAP)-induced depolarization of sympathetic superior
cervical ganglion (SCG) neurons. Ann. N. Y. Acad. Sci., 865:
164–175.
May, V. and Braas, K.M. (1995) Pituitary adenylate cyclase-
activating polypeptide (PACAP) regulation of sympathetic
neuron neuropeptide Y and catecholamine expression.
J. Neurochem., 65: 978–987.
Middleton, J.W. and Keast, J.R. (2004) Artificial autonomic
reflexes: using functional electrical stimulation to mimic
bladder reflexes after injury or disease. Auton. Neurosci.,
113: 3–15.
Morrison, J.F.B. (1987) Reflex control of the lower urinary
tract. In: Torrens M. and Morrison J.F.B. (Eds.), The Phys-
iology of the Lower Urinary Tract. Springer, New York.
Oppenheim, R.W., Prevette, D., Qin-Wei, Y., Collins, R. and
MacDonald, J. (1991) Control of embryonic motoneuron
survival in vivo by ciliary neurotrophic factor. Science, 251:
1616–1618.
Rice, A.S.C. (1995) Topical spinal administration of a nitric
oxide synthase inhibitor prevents the hyper-reflexia associat-
ed with a rat model of persistent visceral pain. Neurosci.
Lett., 187: 111–114.
Ruda, M.A., Besse, D., Inagaki, S., DeLeon, M. and Ren, K.
(1994) Nitric oxide expression and regulation in the dorsal
root ganglion and spinal cord. In: Chiueh C.C., Gilbert D.L.
and Colton C.A. (Eds.), The Neurobiology of NO and OH.
New York Academy of Sciences, New York, pp. 181–200.
Ruda, M.A., Iadarola, M.J., Cohen, L.V. and Young II, W.S.
(1988) In situ hybridization, histochemistry and
immunocytochemistry reveal an increase in spinal dynorphin
biosynthesis in a rat model of peripheral inflammation and
hyperalgesia. Proc. Natl. Acad. Sci., 85: 622–626.
Saito, S., Kidd, G.J., Trapp, B.D., Dawson, T.M., Bredt, D.S.,
Wilson, D.A., Traystman, R.J., Snyder, S.H. and Hanley,
D.F. (1994) Rat spinal cord neurons contain nitric oxide
synthase. Neuroscience, 59: 447–456.
Seki, S., Sasaki, K., Igawa, Y., Nishizawa, O., Chancellor,
M.B., De Groat, W.C. and Yoshimura, N. (2004) Suppres-
sion of detrusor–sphincter dyssynergia by immunoneutrali-
zation of nerve growth factor in lumbosacral spinal cord in
spinal cord injured rats. J. Urol., 171: 478–482.
Sharma, H.S., Westman, J., Olsson, Y. and Alm, P. (1996)
Involvement of nitric oxide in acute spinal cord injury: an
immunocytochemical study using light and electron
microscopy in the rat. Neurosci. Res., 24: 373–384.
Steers, W.D., Ciambotti, J., Etzel, B., Erdman, S. and de Groat,
W.C. (1991a) Alterations in afferent pathways from the uri-
nary bladder of the rat in response to partial urethral
obstruction. J. Comp. Neurol., 310: 1–10.
Steers, W.D., Creedon, D.J. and Tuttle, J.B. (1996) Immunity
to nerve growth factor prevents afferent plasticity following
urinary bladder hypertrophy. J. Urol., 155: 379–385.
Steers, W.D. and de Groat, W.C. (1988) Effect of bladder outlet
obstruction on micturition reflex pathways in the rat.
J. Urol., 140: 864–871.
Steers, W.D., Kolbeck, S., Creedon, D. and Tuttle, J.B. (1991b)
Nerve growth factor in the urinary bladder of the adult reg-
ulates neuronal form and function. J. Clin. Invest., 88:
1709–1715.
Sundler, F., Ekblad, E., Hannibal, J., Moller, K., Zhang, Y.Z.,
Mulder, H., Elsas, T., Grunditz, T., Danielsen, N., Fahrenkrug,
J. and Uddman, R. (1996) Pituitary adenylate cyclase-acti-
vating peptide in sensory and autonomic ganglia: localization
and regulation. Ann. N.Y. Acad. Sci., 805: 410–426.
Tuttle, J.B. and Steers, W.D. (1992) Nerve growth factor re-
sponsiveness of cultured pelvic ganglion neurons from the
adult rat. Brain Res., 588: 29–40.
Tuttle, J.B., Steers, W.D., Albo, M. and Nataluk, E. (1994)
Neural input regulates tissue NGF and growth of the adult
rat urinary bladder. J. Auton. Nerv. Syst., 49: 147–158.
Valtshanoff, J.G., Weinberg, R.J. and Rustioni, A. (1992)
NADPH diaphorase in the spinal cord of rats. J. Comp.
Neurol., 321: 209–222.
Verge, V.M.K., Xu, Z., Xu, X.J., Wiesenfeld-Hallin, Z. and
Hökfelt, T. (1992) Marked increase in nitric oxide synthase
mRNA in dorsal root ganglia after peripheral axotomy: in
situ hybridization and functional studies. Proc. Natl. Acad.
Sci., 89: 11618–11621.

Vizzard, M.A. (1997) Increased expression of neuronal nitric
oxide synthase in bladder afferent and spinal neurons
following spinal cord injury. Dev. Neurosci., 19: 232–246.
Vizzard, M.A. (2000a) Changes in urinary bladder neurotro-
phic factor mRNA and NGF protein following urinary blad-
der dysfunction. Exp. Neurol., 161: 273–284.
Vizzard, M.A. (2000b) Increased expression of spinal cord Fos
protein induced by bladder stimulation after spinal cord
injury. Am. J. Physiol., 279: R295–R305.
Vizzard, M.A. (2000c) Increased expression of spinal Fos
protein in lower urinary tract pathways induced by bladder
distension following chronic cystitis. Am. J. Physiol., 279:
R295–R305.
Vizzard, M.A. (2000d) Up-regulation of pituitary adenylate
cyclase-activating polypeptide in urinary bladder pathways
after chronic cystitis. J. Comp. Neurol., 420: 335–348.
Vizzard, M.A. (2001) Alterations in neuropeptide expression in
lumbosacral bladder pathways following chronic cystitis.
J. Chem. Neuroanat., 21: 125–138.
Vizzard, M.A. and de Groat, W.C. (1996) Increased expression
of neuronal nitric oxide synthase (NOS) in bladder afferent
pathways following chronic bladder irritation. J. Comp.
Neurol., 370: 191–202.
Vizzard, M.A., Erdman, S.L. and de Groat, W.C. (1993a) The
effect of rhizotomy on NADPH diaphorase staining in the
lumbar spinal cord of the rat. Brain Res., 607: 349–353.
Vizzard, M.A., Erdman, S.L. and de Groat, W.C. (1993b)
Localization of NADPH-diaphorase in bladder afferent and
postganglionic efferent neurons of the rat. J. Auton. Nerv.
Syst., 44: 85–90.
Vizzard, M.A., Erdman, S.L. and de Groat, W.C. (1993c)
Localization of NADPH-diaphorase in pelvic afferent and
efferent pathways of the rat. Neurosci Lett., 152: 72–76.
Vizzard, M.A., Erdman, S.L. and de Groat, W.C. (1995a) In-
creased expression of neuronal nitric oxide synthase (NOS) in
visceral neurons after nerve injury. J. Neurosci., 15: 4033–4045.
Vizzard, M.A., Erdman, S.L., Erickson, V.L., Stewart, R.J.,
Roppolo, J.R. and de Groat, W.C. (1994a) Localization of
NADPH-diaphorase in the lumbosacral spinal cord and
dorsal root ganglia of the cat. J. Comp. Neurol., 339: 62–75.
Vizzard, M.A., Erdman, S.L., Förstermann, U. and de Groat,
W.C. (1994b) Differential distribution of nitric oxide synt-
hase in neural pathways to the urogenital organs (urethra,
penis, urinary bladder) of the rat. Brain Res., 646: 279–291.
Vizzard, M.A., Erdman, S.L., Roppolo, J.R., Förstermann, U.
and de Groat, W.C. (1994c) Differential localization of neu-
ronal nitric oxide synthase immunoreactivity and NADPH-
diaphorase activity in the cat spinal cord. Cell Tiss. Res., 278:
299–309.
115
Vizzard, M.A., Erickson, V.L., Card, J.P., Roppolo, J.R. and
de Groat, W.C. (1995b) Transneuronal labeling of neurons in
the adult rat brainstem and spinal cord after injection of
pseudorabies virus into the urethra. J Comp. Neurol., 355:
629–640.
Vizzard, M.A., Qiao, L. and Zvara, P. (2003) Neurotrophic
mechanisms in bladder overactivity after spinal cord injury.
Auton. Neurosci. Basic Clin., 106: 14.
Wang, Z.-Y., Alm, P. and Hakanson, R. (1996) PACAP occurs
in sensory nerve fibers and participates in ocular inflamma-
tion in the rabbit. Ann. N.Y. Acad. Sci., 805: 779–783.
Woolf, C.J., Allchorne, A., Safieh-Garabedian, B. and Poole, S.
(1997) Cytokines, nerve growth factor and inflammatory
hyperalgesia: the contribution of tumour necrosis factor. Br.
J. Pharm., 121: 417–424.
Wu, W., Liuzzi, F.J., Schinco, F.P., Depto, A.S., Li, Y., Mong,
J.A., Dawson, T.M. and Snyder, S.H. (1994) Neuronal nitric
oxide synthase is induced in spinal neurons by traumatic in-
jury. Neuroscience, 61: 719–726.
Yoshimura, N. (1999) Bladder afferent pathways and spinal
cord injury: possible mechanisms inducing hyperreflexia of
the urinary bladder. Prog. Neurobiol., 57: 583–606.
Zhang, X., Verge, V., Wiesenfeld-Hallin, Z., Ju, G., Bredt,
D.S., Snyder, S.H. and Hökfelt, T. (1993) Nitric oxide
synthase-like immunoreactivity in lumbar dorsal root ganglia
and spinal cord of rat and monkey and effect of peripheral
axotomy. J. Comp. Neurol., 335: 563–575.
Zhang, Y.Z., Danielsen, N., Sundler, F. and Mulder, H. (1998)
Pituitary adenylate cyclase-activating peptide is upregulated
in sensory neurons by inflammation. Neuroreport, 9:
2833–2836.
Zhang, Y.-Z., Hannibal, J., Zhao, Q., Moller, K., Danielsen,
N., Fahrenkrug, J. and Sundler, F. (1996) Pituitary adenylate
cyclase activating peptide expression in the rat dorsal root
ganglia: up-regulation after peripheral nerve injury. Neuro-
science, 74: 1099–1110.
Zvara, P., Kliment, J., DeRoss, A.L., Irwin, B.H., Malley, S.E.,
Plante, M.K. and Vizzard, M.A. (2002) Differential expres-
sion of bladder neurotrophic factor mRNA in male and fe-
male rats after bladder outflow obstruction. J. Urol., 168:
2682–2688.
Zvarova, K., Dunleavy, J.D. and Vizzard, M.A. (2005) Chang-
es in pituitary adenylate cyclase activating polypeptide
(PACAP) expression in urinary bladder pathways after spi-
nal cord injury (SCI). Exp. Neurol., 192(1): 46–59.
Zvarova, K., Murray, E. and Vizzard, M.A. (2004) Changes in
galanin immunoreactivity in rat lumbosacral spinal cord and
dorsal root ganglia after spinal cord injury. J. Comp. Ne-
urol., 475: 590–603.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 8

Effect of injury severity on lower urinary tract

function after experimental spinal cord injury

Jean R. Wrathall and Gregory S. Emch

Department of Neuroscience, Georgetown University Medical Center, TRB EP04, Washington, DC 20057, USA

Abstract: Lower urinary tract dysfunction is a serious burden for patients following spinal cord injury.
Patients are usually limited to treatment with urinary drainage catheters, which can lead to repeated urinary
tract infections and lower quality of life. Most of the information previously obtained regarding lower
urinary tract function after spinal cord injury has been in completely transected animals. After thoracic
transection in the rat, plasticity of local lumbosacral spinal circuitry establishes a ‘‘reflex bladder,’’ which
results in partial recovery of micturition, albeit with reduced voiding efficiency. Since at least half of cord-
injured patients exhibit neurologically incomplete injury, rat models of clinically relevant incomplete con-
tusion injury have been developed. With respect to lower urinary tract function, recent anatomical and
physiological studies have been performed after incomplete thoracic contusion injury. The results show
greater recovery of lower urinary tract function that varies inversely with the severity of the initial trauma
and is positively correlated with time after injury. Recovery, as measured by coordination of the bladder
with the external urethral sphincter, occurs between 1 and 4 weeks after spinal cord injury. It is associated
with normalization of: serotonin immunoreactivity and glutamate receptor subunit mRNA expression in
the dorsolateral nucleus that innervates the external urethral sphincter muscle, the response to glut-
amatergic pharmacological probes administered at the lumbosacral spinal cord level, and c-Fos activation
patterns in the lumbar spinal cord. Understanding the mechanisms involved in this recovery will provide a
basis for enhancing lower urinary tract function in patients after incomplete spinal cord injury.

Introduction often with resistant organisms (Trautner and

Spinal cord injury results in lower urinary tract
dysfunction that contributes to patients’ morbidity
and mortality and profoundly limits their quality
of life. Currently, there are few options to treat
lower urinary tract dysfunction after spinal cord
injury. Most patients are limited to the use of uri-
nary drainage catheters transiently or permanent-
ly. The altered voiding dynamics, repeated use of
catheters, and frequent exposure to antibiotic
agents predispose individuals with spinal cord in-
jury to recurrent episodes of urinary tract infection
Corresponding author. Tel.: +202-687-1196;
Fax: +202-687-0617; E-mail: wrathalj@georgetown.edu
Darouiche, 2002; Garcia Leoni and Esclarin De
Ruz, 2003). Recently, there have been advances in
the treatment of other types of lower urinary tract
disorders, such as stress incontinence (Thor, 2003).
These have been based on understanding the ne-
uroanatomical pathways and transmitter systems
involved in lower urinary tract function. With re-
spect to spinal cord injury, detailed studies of the
changes that occur after complete spinal cord
transection have been performed and are reviewed
elsewhere in this book. However, at least half of all
spinal cord injury patients have incomplete injuries
(Bracken et al., 1990), and much less is known
about the alterations in lower urinary tract func-
tion after incomplete spinal cord injury. Recent

DOI: 10.1016/S0079-6123(05)52008-9 117

118
studies from our laboratory, reviewed below, dem-
onstrate that the effect of incomplete spinal cord
injury on lower urinary tract function depends
upon the severity of the injury. Under certain con-
ditions considerable recovery of lower urinary
tract function can occur (Pikov and Wrathall,
2001). Understanding the mechanisms involved in
this recovery may provide a basis for new thera-
peutic approaches to enhance lower urinary tract
function after clinical spinal cord injury.
Normal Control of lower urinary tract function and
effect of spinal cord transection
The bladder and external urethral sphincter com-
prise a coordinated visceral system that has been
well studied both in normal and spinal transected
animals (Tiseo and Yaksh, 1990; de Groat, 1995;
Morrison, 1997; de Groat et al., 1998). Urine stor-
age and voiding are the two main functions of the
lower urinary tract. Urine storage and release from
the bladder depend on sympathetic and parasym-
pathetic innervation, respectively. The pre-gang-
lionic sympathetic neurons are located in the rostral
segments of the lumbar spinal cord and the pre-
ganglionic parasympathetic neurons in the rostral
portion of the sacral spinal cord (de Groat, 1995).
The urethral sphincter is responsible for outlet re-
sistance during storage. Micturition in rats is pro-
duced by coordinated contraction of the smooth
muscle of the urinary bladder and contraction of the
striated muscle of the external urethral sphincter
intermittently, in 4–6 Hz frequency bursts, which
appears to facilitate voiding (Mersdorf et al., 1993).
If external urethral sphincter activity is prevented by
neuromuscular blockade, efficient voiding is abol-
ished (Maggi et al., 1986a; Mersdorf et al., 1993;
Kakizaki et al., 1997). As shown in Fig. 1 effective
voiding, therefore, requires interaction of spinal au-
tonomic reflexes with supraspinal micturition-con-
trolling centers to coordinate urine expulsion from
the bladder with low-frequency intermittent con-
tractions of the somatically innervated external ure-
thral sphincter muscle (Holstege et al., 1986;
Holstege and Tan, 1987; Kakizaki et al., 1997).
Among the brainstem centers involved in the
control of micturition, the pontine micturition
center (Barrington’s nucleus) in the dorsolateral
Pontine micturition
centers
CRF GLU
Bladder
detrusor SPN
5-HT GLU
EUS
DL
Fig. 1. Spinal cord and brainstem control of micturition in the
rat. Neurons innervating the bladder detrusor reside in the
sacral parasympathetic nucleus (SPN) at spinal segments
L6–S1, while those innervating the external urethral sphincter
(EUS) muscle are part of the dorsolateral nucleus (DL) in the
spinal cord at L6/S1. Afferents from the bladder convey pres-
sure information to the pontine micturition centers, which
stimulate the sacral parasympathetic nucleus to cause bladder
contraction and the dorsolateral nucleus to activate, intermit-
tently, the external urethral sphincter in a coordinated fashion.
Both afferent and efferent pathways are glutamatergic (GLU).
Serotonin (5-HT) and corticotrophin releasing hormone (CRF)
are also involved in descending motor control pathways.
tegmentum of the pons and the periaqueductal
gray are considered to be important in triggering
the initiation of voiding (Kruse et al., 1990, 1991).
The external urethral sphincter-controlling motor
neurons, located in the dorsolateral nucleus of
L6-S1 (Schroder, 1980), receive direct and indirect
(via spinal interneurons) supraspinal projections
mostly from the pontine micturition center
(Vizzard et al., 1995; Nadelhaft and Vera, 1996;
Marson, 1997), the D-region just ventral to
Barrington’s nucleus (Ding et al., 1995), and the
ventrolateral pontine periaqueductal gray (Marson,
1997; Ding et al., 1998; Matsuura et al., 1998).
Other brainstem nuclei with identified connec-
tions to bladder and external urethral sphincter
pathways are the raphe magnus, raphe pallidus,
parapyramidal medullary reticular formation,
subcoeruleus pars alpha, locus coeruleus, and the
A5 and A7 nuclei (Vizzard et al., 1995; Marson,
1997). Of these, cells in the raphe nuclei and

nucleus paragigantocellularis in the medullary re-
ticular formation produce serotonin (5-hydroxy-
tryptamine, 5-HT) (Marson, 1997). Thus, 5-HT is
a marker of direct supraspinal projections to the
dorsolateral nucleus (Ramirez-Leon et al., 1994;
Tang et al., 1998). Because of the importance of
these descending control pathways, a lesion of the
spinal cord above the lumbar level would be ex-
pected to leave spinal micturition reflexes intact
and compromise lower urinary tract function by
affecting bladder–external urethral sphincter sy-
nergy of activation (Chancellor et al., 1990).
Previous studies of complete spinal cord transec-
tion in the cat (de Groat, 1990), and in the rat
(Kruse et al., 1993; Mimata et al., 1993), have
shown that lower urinary tract function varies with
time after injury. Initially, during the phase of spinal
shock, the bladder is areflexic and urinary retention
occurs (Hassouna et al., 1984). During this stage,
the bladder becomes tonically overdistended and
noncompliant. Then the spinal reflex activity reap-
pears (Tiseo and Yaksh, 1990), in an exaggerated
(spastic) mode with hyperreflexic bladder detrusor
muscle contractions (Osborn et al., 1990). This is
believed to be due to a lack of supraspinal inhibition
and/or an increase of afferent signaling (Cheng
et al., 1995) as well as plasticity of afferents (Kruse
et al., 1995) resulting from the enlarged bladder.
The external urethral sphincter, which in uninjured
animals is under supraspinal control and works in
synergy with the detrusor muscle, becomes contin-
uously active and therefore dyssynergic with the
emerging automatic bladder contraction reflex
(Schalow et al., 1995). Thus, although spinal cir-
cuits alone are capable of establishing automatic
bladder control after transection (de Groat et al.,
1998), detrusor–external urethral sphincter coordi-
nation that is mediated via a spino-bulbo-spinal re-
flex (Holstege et al., 1986; de Groat, 1990) does not
recover after complete spinal cord transection.
Incomplete spinal cord injury and lower urinary
tract function
Large-scale clinical trials demonstrate that at least
half of cord-injured people have neurologically
incomplete injury (Bracken et al., 1990). Further,
a significant proportion of those classified as
119
‘‘complete’’ may have some residual connections
with supraspinal control centers. For example,
with respect to lower urinary tract function, 15%
of people classified as complete can consciously
detect bladder filling and/or electrical stimulation
indicating the persistence of afferent connections
with the cerebral cortex (Wyndaele, 1991). Because
of the importance of incomplete spinal cord injury,
rat models of clinically relevant incomplete con-
tusion injury have been developed and character-
ized (e.g., Wrathall et al., 1985; Gruner, 1992). In
these models, spinal cord injury is produced by the
impact of a weight onto the exposed dura after a
laminectomy, usually at a mid-thoracic location.
The impact produces mechanical destruction of
tissue and hemorrhage that is maximal in the cen-
tral gray matter and the white matter just above it
in the dorsal funiculus (Noble and Wrathall,
1989). However, a peripheral rim of white matter
is spared, the thickness of which depends on the
severity of the impact. A 10 g weight impacting
onto the exposed dura at thoracic level T8 will
spare nearly a complete peripheral rim if dropped
from 12.5 or 25 mm, but only residual white matter
in the most ventrolateral region remains after a
50 mm impact, as shown in Fig. 2. However, the
spared white matter is far from normal, as there is
a preferential loss of the larger axons from even
the most peripheral regions and the myelination
and glial microenvironment remain abnormal
chronically for at least 2 months after spinal cord
injury (Wrathall et al., 1998; Rosenberg et al.,
2005). Thus, there is complete loss of long de-
scending and ascending axons in some regions of
the spinal cord at the injury epicenter and partial
loss in other regions of white matter.
There is now considerable data documenting re-
covery of hind limb sensory and motor function
after experimental contusion of the thoracic spinal
cord (Wrathall, 1994). The extent of recovery is
inversely related to the severity of the initial trauma
and, for the most part, positively correlated with
time after injury. The recovery phase following in-
itial hind limb areflexia and complete paralysis is
characterized by the return of segmental reflexes in
a modified state and increasingly effective use of the
hind limbs in coordinated movements that are
known to mediate postural control, swimming
120

Fig. 2. White matter sparing at the spinal cord injury epicenter. Top, bottom left: Photomicrographs of representative sections through
the lesion epicenter from one rat of each of the three T8 spinal cord injury (SCI) groups and from an uninjured control stained with
eriochrome cyanine to label myelin. The dorsal, lateral, and ventral funicular white matter of the normal spinal cord is heavily stained,
whereas little myelin staining is seen in the gray matter. The cross-sectional profiles of the injured spinal cords are reduced in diameter.
The center of the injured cords contains cavities and an abnormal loose network of cells, but no myelin staining is apparent.
A peripheral rim of residual white matter is seen. Myelin staining is present but reduced compared with normal white matter, consistent
with the chronic hypo-myelination of residual axons. Bottom right: The average areas of myelinated white matter from the ventral and
lateral funicular zones at the lesion epicenter in the injury groups. SCI height indicates height from which weight is dropped, i.e.,
severity of injury. * Indicates a significant difference from the 12.5 mm weight drop group, based on p.o.0.05 in Tukey’s post hoc test
after ANOVA. Scale bar, 250 mm. Taken from Pikov and Wrathall (2001) with permission.

movements and elements of locomotion. This re-
covery phase plateaus at 3–4 weeks in the adult rat
with the standard tests revealing no significant ad-
ditional recovery between 4 and 8 weeks (Noble
and Wrathall, 1989) or even by 6 months (J.R.
Wrathall, unpublished). With respect to open field
locomotion after an incomplete contusion, rats
show a stereotypical pattern of recovery of loco-
motion consisting of early, intermediate, and late
stages (Basso et al., 1995). During the first stage
there is increasing joint movement in the hind
limbs. In the second, rats become capable of plan-
tar stepping and bearing weight on their hind limbs.
In the last phase of recovery, exhibited by the least
severely injured rats, there is consistent coordinated
weight-bearing locomotion with increasingly nor-
mal positioning of the paws, the trunk, and the tail.
The difference in function between these spinal
cord-injured animals at a few days after injury and
4 weeks later is remarkable — from almost com-
plete paralysis to quite effective, albeit still abnor-
mal, use of their hind limbs. Thus, considerable
natural recovery of hind limb sensory-motor func-
tion occurs in rats that retain only 10–20% of spi-
nal cord tissue at the injury epicenter.
These spinal cord-injured rats also demonstrate
abnormalities of lower urinary tract function and
are initially unable to urinate. A ‘‘reflex bladder’’
develops with time as seen after spinal cord tran-
section. However, depending on the severity of the
lesion, coordinated function of the bladder and the
external urethral sphincter may also recover after
incomplete spinal cord injury (Pikov et al., 1998;
Pikov, 2000; Pikov and Wrathall, 2001, 2002).

Evaluating recovery of lower urinary tract function
after experimental contusion spinal cord injury
Our initial interest in recovery of lower urinary
tract function stemmed from the observation that
the length of time required for manual expression
of the bladder after contusion injury appeared to
be reduced with acute treatments that enhanced
white matter sparing and recovery of hind limb
function. Groups of rats treated with the gluta-
mate receptor antagonist NBQX demonstrated a
dose-related sparing of white matter at the injury
epicenter and a dose-related decrease in the
number of days required for them to acquire a
reflex bladder (Wrathall et al., 1994). Similarly,
when a sodium channel blocker was used to reduce
axonal loss, the number of days required to attain
a reflex bladder after spinal cord injury was re-
duced (Teng and Wrathall, 1997).
Evaluating the volume of urine expressed over
time after three different severities of thoracic con-
tusion injury revealed an initial increase in expressed
volume followed by a decrease, as the spinal bladder
reflex was established (Pikov and Wrathall, 2001).
However, more severe spinal cord injury, associated
with greater loss of white matter at the injury site, as
shown in Fig. 2, was reflected in a slower establish-
ment of spontaneous reflex voiding (Fig. 3).
To study recovery of lower urinary tract func-
tion further, we used a urodynamic procedure that
allows a rapid collection of data over a large
number of voiding cycles (Maggi et al., 1986b). As
shown in Fig. 4, bladder intravesical pressure was
recorded with a transurethral bladder catheter
(polyethylene-50) during continuous perfusion
with warm saline (0.22 ml/min). During the blad-
der detrusor contractions, fluid was released by
flowing around the catheter in the urethra. The
signal from the pressure transducer was amplified,
sampled at 1 kHz and acquired on-line using Bio-
Bench 1.0 software (National Instruments, Austin,
TX). For electromyography, two fine (50 mm)
epoxy-coated platinum–iridium wire electrodes
were placed percutaneously in the sphincter area
of the urethra to record external urethral sphincter
electrical activity. The electromyographic activity
was pre-amplified, sampled at 1 kHz, and acquired
on-line simultaneously with intravesical pressure.
121
Fig. 3. Time course of recovery of spontaneous voiding. The
bars on the curve show the urine volumes (71 standard error of
the mean) manually expressed every 12 h from the urinary
bladder, plotted against time after spinal cord injury (SCI). The
volume collected increased between days 0 and 4, presumably
due to an increase in urine production during the immediate
post-operative period. The expressed volume then decreased
after day 4, presumably due to recovery of spontaneous void-
ing. After day 4, progressively less urine was expressed from the
bladders of the rats with the (milder) 12.5 mm weight drop cord
injury, indicating a faster recovery of spontaneous voiding in
this group than in the other two more severely injured groups.
Taken from Pikov and Wrathall (2001) with permission.
In initial studies average values of external ure-
thral sphincter spiking activity during bladder fill-
ing and voiding were calculated, and threshold (at
the initiation of contraction) and maximal in-
travesical pressures during voiding were measured
for each voiding cycle over a 20-min period in each
of the animals (Pikov et al., 1998). The most useful
measure was found to be the change in external
urethral sphincter spiking activity calculated from
the raw electromyography data. This was obtained
by counting the number of peaks above the base-
line at 100 ms intervals with a custom-written peak
detection macro in Microsoft Excel (Pikov and
Wrathall, 2001). The change in external urethral
sphincter spiking activity was measured during
bladder filling and emptying as illustrated in
Fig. 5. Although the catheter and electrodes were
inserted under anesthesia, the rats were then al-
lowed to recover so that the urodynamic record-
ings were done on awake restrained animals
122

Fig. 4. Experimental design for urodynamic recordings. A transurethral bladder catheter (polyethylene-50) is implanted in an anest-
hetized rat. After recovery from anesthesia, urodynamic recordings are performed under light anesthesia. The catheter is connected to
an infusion pump and a pressure transducer. Bladder intravesical pressure is recorded during continuous perfusion with room
temperature saline (CMG, cystometrogram). During bladder contractions, fluid is released by flowing around the catheter in the
urethra. For electromyography (EMG), two fine platinum wire electrodes are placed percutaneously in the sphincter area of the urethra
to record external urethral sphincter (EUS) electrical activity. The EMG activity is amplified, sampled at 1 kHz, and acquired on-line
simultaneously with intravesical pressure. Taken from Pikov et al. (1998) with permission.

because anesthesia markedly reduces the efficiency
of voiding (Yoshiyama et al., 1994, 1999).
Comparing urodynamic measures of normal rats
to those after thoracic contusion or complete tran-
section injury, we found that rats after both tran-
section and contusion spinal cord injury showed
evidence of reflexive bladder contractions in week 2
after injury, but only the contused groups demon-
strated some recovery of coincidental activation of
the external urethral sphincter (Pikov et al., 1998).
Bladder weight was measured, showing a six-fold
increase at week 1, an eight-fold increase at week 2
after incomplete contusion, and an 11.6-fold in-
crease in cord-transected animals. Voided volume,
or the amount that is released from the bladder in
each contraction, was found to decrease in both
contused and transected animals. Voiding efficiency
(volume voided/capacity
100), due to increase in
bladder capacity and decrease in voided volume was
very low (2.8–3.5%) in both groups of cord-injured
animals as compared to uninjured controls (58%).
The inter-contraction interval (time between de-
trusor contractions) was also lower in the spinal cord
injury groups (40–70 vs. 115 s in control animals).
In order to investigate anatomical evidence of
supraspinal involvement in lower urinary tract func-
tion, pseudorabies virus was injected into the blad-
der wall in normal and spinal cord-injured animals.
The rats were allowed to survive long enough for
transneuronal tracing to the brainstem. As shown in
Fig. 6, labeling was present in the pontine micturit-
ion center and in the periaqueductal gray. As might
be expected, there was much lower labeling in the
contused cord-injured animal as compared to unin-
jured control, illustrating an anatomical basis for
the reduced supraspinal control of lower urinary
tract function in rats after injury.
Effect of injury severity on chronic lower urinary
tract function after incomplete spinal cord injury
More information was obtained from an injury
dose–response study in which groups of rats were
subjected to spinal cord injury with the widely
used Multicenter Animal Spinal Cord Injury Study
(MASCIS) injury device (Gruner, 1992) and pro-
duced by the impact of a 10 g weight dropped from
Fig. 5. Urodynamic analysis of detrusor–external urethral sphincter coordination in representative uninjured animals (A, D, G and J), and animals with 12.5 mm weight
drop (B, E, H and K) and 50 mm weight drop (C, F, I and L) spinal cord injury (SCI) at 8 weeks. (A–C) Bladder intravesical pressure (IVP) recordings during one voiding
cycle. Solid horizontal line in A and B indicates the duration of stream-like voiding, and the dashed line in C indicates the drop-by-drop voiding. (D–F) external urethral
sphincter EMG recordings, showing activation of external urethral sphincter EMG in relation to the voiding cycle in uninjured and 12.5 mm SCI animals but not in the
50 mm SCI animal. (G–I) Power spectrum analysis of external urethral sphincter EMG activity as a function of time. A broad band of frequencies (5–40 Hz) shows an
increased power during the voiding phase in uninjured and 12.5 mm cord-injured animals but not in the 50 mm cord-injured animal. (J–L) Peak detection analysis of the
external urethral sphincter spiking activity (ESA). Peaks were detected in 1 ds (ds ¼ 101 s) intervals. An increase in spiking activity occurred at the time corresponding to
the voiding phase in uninjured and 12.5 mm cord-injured animals, but there was no change in the level of spiking activity in the 50 mm cord-injured animal. Taken from
Pikov and Wrathall (2001) with permission.

123
124

Fig. 6. Pseudorabies virus tracing from the bladder. (A and B) Micrographs of Pseudorabies virus-labeled neurons in the pontine
micturition center of a normal rat (A) and a rat on day 12 after a 10 g
25 mm weight drop contusion (B). The large neurons of the
mesencephalic nucleus of the fifth cranial nerve on the left of each field demonstrate some nonspecific staining. Bar ¼ 100 mm. (C–F)
Schematized images of Pseudorabies virus retrograde transneuronal labeling from the bladder. Dorsolateral tegmentum (C and D) and
periaqueductal gray (E and F) in uninjured (C and E) and SCI animals (D and F). Abbreviations: Mes5-mesencephalic trigeminal
nucleus, PAG-periaqueductal gray, SCP-superior cerebellar peduncle. Taken from Pikov et al. (1998) with permission.

a height of 12.5, 25 or 50 mm onto the dura after a
laminectomy at T8 (Pikov and Wrathall, 2001).
Table 1 shows a comparison of bladder weights,
volumes and pressure (intravesical pressure) at 8
weeks after injury. Compared to uninjured con-
trols, the 25 and 50 mm groups demonstrated sig-
nificantly greater bladder weight and volume
chronically. The most severely injured (50 mm)
group also had significantly decreased maximal
bladder pressure during urodynamic evaluation. In
contrast the 12.5 mm group, although showing
similar tendencies, was not significantly different
from controls in these parameters of lower urinary
tract function.
As shown in Fig. 7A, the 12.5 mm spinal cord-
injured group recovered bladder–external urethral
sphincter coordination (as measured by the change
in external urethral sphincter spiking activity) by 8
 125

Table 1. Changes in lower urinary tract parameters 8 weeks after spinal cord injury

Parameter Spinal cord injury severity (weight drop in mm)

0 (n ¼ 6) 12.5 (n ¼ 7) 25 (n ¼ 4) 50 (n ¼ 7)

Bladder weight (g) 0.1170.01 0.2370.09 0.37*70.13 0.41*70.18

Bladder volume (ml) 3.871.5 9.176.0 14.9*76.2 26.1**78.1
Intravesical pressure amplitude (mmHg) 25.477.7 19.477.6 14.173.0 13.0*75.1

The number of animals in each group is shown within parentheses. Urodynamic evaluation was performed at 8 weeks after SCI or laminectomy. The
bladder was then weighed after blot-drying, and its length and width were measured to calculate volume. The amplitude of the bladder pressure was
calculated as the difference between the maximal intravesicular pressure during voiding and the pressure just before voiding was initiated. Mean
values7standard error are presented for each measurement. *po0.05, indicates significantly different from values in control rats (0 weight drop). **,
significantly different from both the control and 12.5 group. Data from Pikov and Wrathall (2001).

Fig. 7. Recovery of detrusor–external urethral sphincter coordination at 8 weeks after spinal cord injury (SCI) depends upon injury
severity (A), and is correlated to white matter sparing at the epicenter (B), to the amount of serotonin immunoreactivity (5HT-IR) in
the dorsolateral nucleus (C), and to CRF immunoreactivity (CRF-IR) in the sacral parasympathetic nucleus (SPN; D). In panel A, SCI
height indicates distance of weight drop, i.e., severity of spinal cord injury. Bars represent means and standard errors. Vertical axis is
change in external urethral sphincter spiking activity (sESA) during voiding. N ¼ 6, 7, 5 and 7 for the 0, 12.5, 25, and 50 mm groups,
respectively (ds ¼ 101 s). Correlation coefficients are shown at the top left corner of B–D. In A, symbols indicate a significant
difference from the uninjured group (*) or from both the uninjured and 12.5 mm groups (**). (po0.001). Taken from Pikov and
Wrathall (2001) with permission. Data from Pikov and Wrathall (2001).

weeks to an extent statistically indistinguishable
from uninjured controls, whereas the 25 and 50 mm
spinal cord-injured groups did not. Furthermore,
at 8 weeks there was a significant correlation be-
tween the degree of recovery of lower urinary tract
function in terms of bladder–external urethral
sphincter coordination and the initial impact as
well as chronic white matter sparing at the injury
epicenter (Fig. 7B). Correlation with spared de-
scending control pathways from the brainstem was
indicated by quantification of the relative immuno-
reactivity for 5-HT associated with the dorsolateral
nucleus motor neurons that innervate the external
urethral sphincter (Fig. 7C), and corticotrophin
126
Table 2. Expression of NMDA (NR1, NR2A, NR2B) and AMPA (GluR1, GluR2, GluR3, GluR4) subunit mRNA in dorsolateral
nucleus motoneurons at 8 weeks after spinal cord injury
Subunit Spinal cord injury severity (weight drop in mm)
0 ðn ¼ 6Þ
NR1 39.478.7
NR2A 6.871.5
NR2B 1.670.4
GluR1 1.370.2
GluR2 4.671.1
GluR3 3.272.1
GluR4 2.470.6
The data are presented as the number of grains per square micrometer of cell area (mean 7 standard error). Significant difference from the uninjured
group is indicated by asterisk (*) and bold font and is based on po0.05 in Tukey’s post hoc test after ANOVA.
releasing factor immunoreactivity associated with
the sacral parasympathetic nucleus that innervates
the bladder detrusor muscle (Fig. 7D).
Glutamate receptors are utilized in spinal circuits
controlling the detrusor and external urethral
sphincter (Matsumoto et al., 1995a, b; Iwabuchi,
1997), and thus changes in properties of these re-
ceptors may be involved in altered lower urinary
tract function after spinal cord injury. Comparing
groups of rats that did or did not recover blad-
der–external urethral sphincter coordination at 8
weeks after spinal cord injury showed significant
differences in the expression of mRNAs for gluta-
mate receptors in the dorsolateral nucleus motoneu-
rons that innervate the external urethral sphincter
(Table 2). The 25 and 50 mm groups that did not
recover normal external urethral sphincter spiking
activity also exhibited abnormally high expression of
NR2A and GluR2, as determined by in situ hybrid-
ization autoradiography. Assuming these changes in
chronic mRNA levels produce altered functional
glutamate receptors, these alterations may be related
to the aberrant hyperactivity of these motoneurons
as in the spontaneous spastic activity seen chroni-
cally in lumbosacral somatic motoneurons after
spinal cord injury (Hiersemenzel et al., 2000; Little
et al., 2000). In contrast the 12.5 mm group that did
recover normal external urethral sphincter spiking
activity expressed normal levels of the receptor sub-
unit mRNAs in the dorsolateral nucleus motoneu-
rons that innervate the external urethral sphincter.
Recently, we investigated, by immunohistoche-
mical identification, the proto-oncogene product
12.5 ðn ¼ 7Þ 25 ðn ¼ 3Þ 50 ðn ¼ 5Þ
43.776.0 43.176.4 45.175.3
9.972.7 12.4*74.1 12.3*72.9
2.370.5 2.270.5 2.270.2
1.670.5 1.670.4 2.070.6
6.372.2 10.5*71.6 8.2*72.6
5.871.3 4.971.6 5.972.6
2.970.6 2.070.3 2.370.2
c-Fos, which is expressed in neurons after intense
or prolonged activation (Rinaman et al., 1993), in
response to bladder filling in normal and spinal
cord-injured rats at 8 weeks after injury (Emch
et al., 2003). Rats that had not recovered normal
levels of external urethral sphincter spiking activity
(25 mm injury group) demonstrated extensive
c-Fos activation (Fig. 8), as previously reported
for rats after complete transection (Vizzard, 2000).
In contrast, the mild (12.5 mm) group that recov-
ered bladder–external urethral sphincter coordina-
tion had a c-Fos activation pattern at 8 weeks that
was similar to uninjured controls. Thus, aspects of
recovered lower urinary tract function after this
severity of spinal cord injury include normal blad-
der–external urethral sphincter coordination, nor-
mal expression of glutamate receptor mRNAs in
dorsolateral nucleus neurons and a normal local
segmental response to bladder stimulation as in-
dicated by c-Fos expression.
What occurs during recovery of bladder–external
urethral sphincter coordination after mild
contusion injury?
In order to study the mechanisms that may be in-
volved in this recovery of lower urinary tract func-
tion we compared animals after mild spinal cord
injury at 5 days after injury (non-recovered; sub-
acute), when the bladder reflex can be detected
without any coordinated activation of the external
urethral sphincter, to animals at 8 weeks (recovered;
 127

A B

L6 spinal cord
55 +
50
Fos count/10 um section

45 +
40
35
30
25 +
*
20
15
10
5
0
C SPN DCM MDH LDH 12.5mm 8wk 25mm 8wk

Fig. 8. After spinal cord injury c-Fos immunoreactivity in the L6 spinal cord segment is altered in animals with abnormal lower
urinary tract function and returns to normal when animals recover detrusor–external urethral sphincter coordination. The bladders of
lightly anesthetized rats were catheterized and room temperature saline was perfused continuously for 2 h to stimulate the voiding
reflex. Spinal cords were sectioned and c-Fos immunohistochemistry was performed with methods adapted from Emch et al. (2001). In
animals with abnormal detrusor–external urethral sphincter coordination as measured in our EMG preparation, the pattern of c-Fos
expression was altered in 4 areas of the L6 spinal cord: the sacral parasympathetic nucleus (SPN), the dorsal gray commissure (DCM),
the medial dorsal horn (MDH), and the lateral dorsal horn (LDH). Increases in c-Fos expression normalized by 8 weeks after injury in
animals that recovered lower urinary tract function, i.e. the mild injury group (10 g
12.5 mm weight drop). The pattern of c-Fos
expression in animals that did not recover lower urinary tract function, i.e. the moderate injury group (10 g
25 mm weight drop), did
not normalize. (A and B) Sections of the L6 DCM tissue stained for c-Fos-IR in a control rat (A) and in a rat 8 weeks after the injury
produced by a 10 g weight dropped 25 mm (B). Note the large number of immunoreactive cells in B. (C) Histograms of c-Fos counts in
the four L6 areas listed above. Open bars indicate controls. Gray and black bars show counts in the same regions in rats injured with
the 12.5 and 50 mm weight drop, respectively. +, significantly different from both control and the 12.5 mm injury group. *Significantly
different from control only.

chronic), when the external urethral sphincter
spiking activity is indistinguishable from uninjured
controls (Fig. 9). Bladder weight at both subacute
and chronic time points was higher than in unin-
jured animals, and there was no difference in blad-
der weight between these two time points. The
change in bladder pressure during contraction
(change in intravesical pressure) was decreased by
about 30% at the earlier time point (subacute) with
some recovery by 8 weeks (chronic), when the
change in intravesical pressure was not significantly
different from that in uninjured animals. The in-
crease and decrease in intravesical pressure during
detrusor contraction and relaxation, respectively,
occurred more slowly in injured than in uninjured
animals. The external urethral sphincter spiking
128
Fig. 9. Time course of recovery of detrusor and external ure-
thral sphincter muscle function after mild spinal cord injury.
The increase in intravesical pressure (dIVP, top panel) and in
external urethral sphincter spiking activity (dESA, bottom pan-
el) during voiding was measured in uninjured rats and in rats at
5 days (subacute) and 8 weeks (chronic) after spinal cord injury
produced by dropping at 10 g weight 12.5 mm unto the dura at
T8 (ds ¼ 101 s). Symbols indicate a significant difference from
the uninjured group (*) or from both the uninjured and chronic
groups (**). Adapted from Pikov and Wrathall (2002) with
permission.
activity was significantly inhibited at the subacute
time point (5 days) and recovered by 8 weeks after
this mild spinal cord injury (Fig. 9).
In examining the dorsolateral nucleus motoneu-
rons we found that 5-HT immunoreactivity at 5
days is significantly below that of uninjured con-
trols, and, by 8 weeks, it has recovered to normal
levels. This suggests that sprouting of raphespinal
fibers spared by the injury may be involved in the
recovery of external urethral sphincter spiking ac-
tivity shown in Fig. 9. In contrast, corticotrophin
releasing factor associated with the sacral para-
sympathetic nucleus was about 60% of the normal
at both 5 days and 8 weeks, consistent with the
ability of a reflex bladder to be established in the
absence of supraspinal connections.
To test the hypothesis that altered glutamate
receptor function is involved with altered lower
urinary tract function after spinal cord injury, the
intravesical pressure and external urethral sphinc-
ter spiking activity were measured during urody-
namic evaluation in the presence of the NMDA
receptor antagonist 3(2-carboxypiperazin-4-yl)-
propyl-1-phosphonic acid (CPP) and the (R,S)-2-
amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic
acid (AMPA) receptor antagonist 2,3-Dihydro-
xy-6-nitro-7-sulphamoylbenzo(f)-quinoxaline 6-
Nitro-7-sulphamoylbenzo(f)-quinoxaline-2,3-dione
(NBQX) (Pikov and Wrathall, 2002). The drugs
were given intrathecally to the lumbosacral spinal
cord with increasing dosages administered during
the urodynamic procedures. The range of doses
was chosen so as not to affect the detrusor itself.
We found that external urethral sphincter spik-
ing activity was mildly affected by low doses of
either drug and was dramatically inhibited by high
doses. The mean values of intravesical pressure
change were unaffected by either drug except for
high doses of CPP in the uninjured and chronic
groups. In contrast, the external urethral sphincter
spiking activity was decreased with each drug in a
dose-dependent manner, with maximal inhibition
of external urethral sphincter spiking activity
(60–70%) seen at the highest doses used. The in-
hibitory dose (ID)50 values for individual animals
were used to calculate the average ID50 values
for NBQX and CPP. There was no effect of
injury on the ID50 of NBQX between experimental
groups. In contrast, the subacute group exhibited
a 50% lower ID50 for CPP than uninjured
controls. By 8 weeks, this effect was no longer
evident.
Our finding of changes only in NMDA receptor
function is consistent with evidence suggesting that
NMDA and non-NMDA receptors are part of
parallel, but functionally separate, synaptic circuits
that are important in micturition (Yoshiyama
et al., 1995). An intriguing hypothesis to explain
our results is an alteration in NMDA receptor sub-
unit composition during recovery of lower urinary
tract function. CPP, a potent NMDA receptor an-
tagonist, has different affinities to NMDA recep-
tors depending upon the NR2 subunit present as
part of the receptor complex, ranking in the order
 129

NR2A4NR2B4NR2D while the affinity to
glutamate is in the opposite order (Monaghan
et al., 1998). Thus, increased sensitivity to CPP
could be due to a shift toward a higher proportion
of NR2A at 5 days after spinal cord injury. If such
receptor composition shifts are confirmed and are
typical after spinal cord injury, this information
may serve as the basis for novel pharmacological
strategies to enhance functional recovery in the
subacute period after spinal cord injury.
Subsequent studies have confirmed that at 5 days
after a mild spinal cord injury, NMDA receptors
on dorsolateral nucleus motoneurons may contain
a higher proportion of NR2A than after recovery

Fig. 10. (Upper panel) Photomicrograph depicting Neutral Red-stained dorsolateral (DL) nucleus motoneuronal cell bodies with
overlying NR2A mRNA grains. Uninjured animals exhibit normal levels of mRNA, whereas 5 days after a MASCIS mild contusion
injury (12.5 mm weight drop), the number of grains are increased. Scale bar ¼ 5 mm. (Lower panel) Quantitative comparison of in situ
hybridization net grain counts for the NMDA subunits NR1, NR2A, NR2B, and the AMPA subunit GluR2 in the DL nucleus. Data
are expressed as net grain counts/mm2 cell body. NR2A and GluR2 net grain counts from injured animals are significantly higher than
uninjured controls, po0.01, whereas those for NR1 and NR2B do not differ significantly between injured and uninjured animals.
Symbols represent data from individual rats; the bar represents the mean value for the group.
130

at 8 weeks. Slides containing the dorsolateral nu- 22.5

cleus from animals at 5 days after mild MASCIS 20.0
contusion injury and from uninjured control ani- 17.5
mals were hybridized with 35S-ATP labeled anti- 15.0
sense oligonucleotides to the NMDA subunit

dESA
12.5
mRNA for NR1, NR2A, and NR2B and the
AMPA subunit mRNA for GluR2. The density of
10.0 ∗
7.5
the grains over the dorsolateral nucleus neurons 5.0 ∗ ∗
was then calculated as a ratio of grains/mm2 and 2.5 ∗
was corrected for background by subtracting the
0.0
density of grains from adjacent areas devoid of tis-

k
s

k
8w
ol

1w

1w

4w

4w

8w
sue. The results (Fig. 10) indicate that the mRNAs

tr
on

m
m

m
m

m
m
for GluR2 and NR2A are abnormally expressed

.5

25

.5

25

.5

25
12

12

(upregulated) at 5 days as found chronically in
more severely injured groups as shown in Table 2.
GluR2 is generally found as part of functional
AMPA receptors but the presence of NR2A in
NMDA receptors is associated with altered recep-
tor sensitivity as mentioned above. Our current
data suggest that altered sensitivity of dorsolateral
nucleus motoneurons may be a general occurrence
in the subacute period after spinal cord injury but is
normalized if and when recovery of bladder–exter-
nal urethral sphincter coordination occurs.
Recently, we have undertaken studies to further
define the time of recovery of lower urinary tract
function after mild spinal cord injury as measured
by the normalization of external urethral sphincter
spiking activity. We have found that complete re-
covery is consistently found by 4 weeks after spinal
cord injury, as shown in Fig. 11. Further, this func-
tional recovery is mirrored by recovery of 5-HT
immunoreactivity of the dorsolateral nucleus moto-
neurons (Fig. 12). Are these correlations indications
of a functional connection? Future studies will fo-
cus on this question, further testing the temporal
relationship between recovery of external urethral
sphincter spiking activity and 5-HT immunoreac-
tivity and extending the studies to pharmacological
evaluation of the functional role of the serotonin
system in recovery of lower urinary tract function.
The serotonin system has been shown to alter
the micturition reflex in both normal rats and
those after a complete spinal cord transection.
Administration of 8-OH-DPAT (a selective
5-HT1A agonist) facilitates the voiding reflex re-
gardless of whether drug administration is intra-
peritoneal, intracerebroventricular or intrathecal
12
Fig. 11. Recovery of detrusor–external urethral sphincter co-
ordination occurs by 4 weeks after mild spinal cord injury. At 1
week after cord injury both mild (12.5 mm) and moderate
(25 mm) injury groups exhibit changes in external urethral
sphincter spiking activity (dESA) during bladder contraction
that are significantly decreased from uninjured controls. By 4
weeks, animals in the mild group exhibit dESA that have re-
covered to control levels. *po0.001, ANOVA with Tukey’s
post hoc testing.
to the lumbar spinal cord (Lecci et al., 1992).
Administration of 5-HT antagonists, such as
WAY-100635 increases bladder capacity and in-
hibits the voiding reflex (Kakizaki et al., 1997;
Testa et al., 1999). In fact, modulation of the se-
rotonergic influence on micturition is under clin-
ical investigation for application in stress-induced
urinary incontinence. The drug duloxetine, an in-
hibitor of both serotonin and norepinephrine re-
uptake appears especially promising for reducing
stress-induced urinary incontinence by facilitating
external urethral sphincter activity (Thor, 2003).
We speculate that pharmacological support of
serotonergic neurotransmission after incomplete
spinal cord injury may modulate the extent of ex-
ternal urethral sphincter activation and thus nor-
malize the bladder–sphincter dyssynergia for
efficient voiding.
Conclusion
On the basis of our studies, we postulate that spinal
cord injury initially affects the lower urinary tract
 131

Fig. 12. 5-HT immunoreactivity in the dorsolateral (DL) nucleus recovers by 4 weeks after mild SCI. 5-HT pixel density was quantified
in the DL nucleus using METAMORPHs software in controls (A), at 1 week (B), and at 4 weeks (C) after mild and moderate spinal
cord injury. # indicates DL motor neurons in A–C. (D) 5-HT pixel density is significantly decreased in both injury groups at 1 week
post-injury (*po0.001), but normalizes by 4 weeks in the mild group. Bars represent mean+SEM of 5-HT pixel density with N ¼ 6 for
controls and 5 for each injury group. ANOVA with Tukey’s post hoc testing.

function similarly after a wide range of severities of
thoracic spinal cord injury ranging from a mild
contusion through a complete surgical transection.
The loss of normal descending control inhibits
normal micturition requiring manual bladder
expression in the rat or catheterization for a
cord-injured person. The bladder responds by
enlarging, and in transection models this is known
to initiate the afferent plasticity and subsequent
interneuronal alterations that results in the devel-
opment of a reflex bladder (Kruse et al., 1993;
de Groat, 1995), as described elsewhere in this
volume. We postulate that the same mechanisms
occur after incomplete contusion injury in the first
week(s) after spinal cord injury. However, the
sparing of key long tract axons then allows a sec-
ond form of recovery to occur, which is only seen
with incomplete spinal cord injury. Changes occur
132
in some spared descending control pathways for
lower urinary tract function, such as the raphespi-
nal system, that allow simultaneous activation of
the external urethral sphincter with bladder con-
traction. With time this coordination improves
and some of the initial changes associated with loss
of lower urinary tract function normalize, such as
altered glutamate receptor subunit expression on
dorsolateral nucleus motoneurons, 5-HT immuno-
reactivity in the dorsolateral nucleus, and perhaps
increased c-Fos activation patterns.
A challenge for future studies will be identifying
the key causal factors in this secondary recovery
phase that produces bladder–external urethral
sphincter coordination. With this information,
we may learn how to increase the efficiency of
voiding and thus reduce the long-term deleterious
effects of abnormal lower urinary tract function
after incomplete spinal cord injury.
Acknowledgments
Studies reviewed in this chapter were supported by
NIH PO1 NS28130, RO1 NS 35647, and RO1
NS37733.
References
Basso, D.M., Beattie, M.S. and Bresnahan, J.C. (1995) A sen-
sitive and reliable locomotor rating scale for open field testing
in rats. J. Neurotrauma, 12: 1–21.
Bracken, M.B., Shepard, M.J., Collins, W.F., Holford, T.R.,
Young, W., Baskin, D.S., Eisenberg, H.M., Flamm, E.,
Leo-Summers, L., Maroon, J., et al. (1990) A randomized,
controlled trial of methylprednisolone or naloxone in the
treatment of acute spinal-cord injury. Results of the Second
National Acute Spinal Cord Injury Study. N. Engl. J. Med.,
322: 1405–1411.
Chancellor, M.B., Kaplan, S.A. and Blaivas, J.G. (1990) De-
trusor–external sphincter dyssynergia. Ciba Found. Symp.,
151: 195–206 discussion 207–113.
Cheng, C.L., Ma, C.P. and de Groat, W.C. (1995) Effect of
capsaicin on micturition and associated reflexes in chronic
spinal rats. Brain Res., 678: 40–48.
de Groat, W.C. (1990) Central neural control of the lower uri-
nary tract. Ciba Found. Symp., 151: 27–44 discussion 44–56.
de Groat, W.C. (1995) Mechanisms underlying the recovery of
lower urinary tract function following spinal cord injury.
Paraplegia, 33: 493–505.
de Groat, W.C., Araki, I., Vizzard, M.A., Yoshiyama, M.,
Yoshimura, N., Sugaya, K., Tai, C. and Roppolo, J.R. (1998)
Developmental and injury induced plasticity in the micturit-
ion reflex pathway. Behav. Brain Res., 92: 127–140.
Ding, Y.Q., Takada, M., Tokuno, H. and Mizuno, N. (1995)
Direct projections from the dorsolateral pontine tegmentum
to pudendal motoneurons innervating the external urethral
sphincter muscle in the rat. J. Comp. Neurol., 357: 318–330.
Ding, Y.Q., Wang, D., Nie, H., Guan, Z.L., Lu, B.Z. and Li,
J.S. (1998) Direct projections from the periaqueductal gray to
pontine micturition center neurons projecting to the lumbo-
sacral cord segments: an electron microscopic study in the
rat. Neurosci. Lett., 242: 97–100.
Emch, G., Lund, IV., Gandy, V., Lytle, J.M. and Wrathall, J.R.
(2003) Glutamate receptor plasticity during recovery of lower
urinary tract function after spinal cord injury. Soc. Neurosci.
Abs., 954: 914.
Emch, G.S., Hermann, G.E. and Rogers, R.C. (2001) TNF-
alpha-induced c-Fos generation in the nucleus of the solitary
tract is blocked by NBQX and MK-801. Am. J. Physiol.
Regul. Integr. Comp. Physiol., 281: R1394–R1400.
Garcia, Leoni, M.E. and Esclarin De Ruz, A. (2003) Manage-
ment of urinary tract infection in patients with spinal cord
injuries. Clin. Microbiol. Infect., 9: 780–785.
Gruner, J.A. (1992) A monitored contusion model of spinal
cord injury in the rat. J. Neurotrauma, 9: 123–126 discussion
126–128.
Hassouna, M., Galeano, C., Abdel-Rahman, M. and Elhilali,
M. (1984) Vesicourethral motility following acute spinal cord
transection in the cat. J. Urol., 131: 370–373.
Hiersemenzel, L.P., Curt, A. and Dietz, V. (2000) From spinal
shock to spasticity: neuronal adaptations to a spinal cord
injury. Neurology, 54: 1574–1582.
Holstege, G., Griffiths, D., de Wall, H. and Dalm, E. (1986)
Anatomical and physiological observations on supraspinal
control of bladder and urethral sphincter muscles in the cat.
J. Comp. Neurol., 250: 449–461.
Holstege, G. and Tan, J. (1987) Supraspinal control of moto-
neurons innervating the striated muscles of the pelvic floor
including urethral and anal sphincters in the cat. Brain,
110(Part 5): 1323–1344.
Iwabuchi, N. (1997) Sacral glutamatergic transmission in the
descending limb of the micturition reflex in the cat. Fukuoka
Igaku Zasshi., 88: 30–38.
Kakizaki, H., Fraser, M.O. and De Groat, W.C. (1997) Reflex
pathways controlling urethral striated and smooth muscle
function in the male rat. Am. J. Physiol., 272: R1647–R1656.
Kruse, M.N., Belton, A.L. and de Groat, W.C. (1993) Changes
in bladder and external urethral sphincter function after spi-
nal cord injury in the rat. Am. J. Physiol., 264: R1157–R1163.
Kruse, M.N., Bray, L.A. and de Groat, W.C. (1995) Influence of
spinal cord injury on the morphology of bladder afferent and
efferent neurons. J. Auton. Nervous System, 54: 215–224.
Kruse, M.N., Mallory, B.S., Noto, H., Roppolo, J.R. and de
Groat, W.C. (1991) Properties of the descending limb of the
spinobulbospinal micturition reflex pathway in the cat. Brain
Res., 556: 6–12.

Kruse, M.N., Noto, H., Roppolo, J.R. and de Groat, W.C.
(1990) Pontine control of the urinary bladder and external
urethral sphincter in the rat. Brain Res., 532: 182–190.
Lecci, A., Giuliani, S., Santicioli, P. and Maggi, C.A. (1992)
Involvement of 5-hydroxytryptamine 1A receptors in the
modulation of micturition reflexes in the anesthetized rat.
J. Pharmacol. Exp. Ther., 262: 181–189.
Little, J.W., Burns, S.P., James, J.J. and Stiens, S.A. (2000)
Neurologic recovery and neurologic decline after spinal cord
injury. Phys. Med. Rehabil. Clin. N. Am., 11: 73–89.
Maggi, C.A., Giuliani, S., Santicioli, P. and Meli, A. (1986a)
Analysis of factors involved in determining urinary bladder
voiding cycle in urethan-anesthetized rats. Am. J. Physiol.,
251: R250–R257.
Maggi, C.A., Santicioli, P. and Meli, A. (1986b) The nonstop
transvesical cystometrogram in urethane-anesthetized rats: a
simple procedure for quantitative studies on the various
phases of urinary bladder voiding cycle. J. Pharmacol. Meth.,
15: 157–167.
Marson, L. (1997) Identification of central nervous system neu-
rons that innervate the bladder body, bladder base, or ex-
ternal urethral sphincter of female rats: a transneuronal
tracing study using pseudorabies virus. J. Comp. Neurol.,
389: 584–602.
Matsumoto, G., Hisamitsu, T. and de Groat, W.C. (1995a)
Non-NMDA glutamatergic excitatory transmission in the
descending limb of the spinobulbospinal micturition reflex
pathway of the rat. Brain Res., 693: 246–250.
Matsumoto, G., Hisamitsu, T. and de Groat, W.C. (1995b)
Role of glutamate and NMDA receptors in the descending
limb of the spinobulbospinal micturition reflex pathway of
the rat. Neurosci. Lett., 183: 58–61.
Matsuura, S., Allen, G.V. and Downie, J.W. (1998) Volume-
evoked micturition reflex is mediated by the ventrolateral
periaqueductal gray in anesthetized rats. Am. J. Physiol., 275:
R2049–R2055.
Mersdorf, A., Schmidt, R.A. and Tanagho, E.A. (1993) Uro-
dynamic evaluation and electrical and pharmacologic neuro-
stimulation. The rat model. Urol. Res., 21: 199–209.
Mimata, H., Satoh, F., Tanigawa, T., Nomura, Y. and Ogata,
J. (1993) Changes of rat urinary bladder during acute phase
of spinal cord injury. Urol. Int., 51: 89–93.
Monaghan, D.T., Andaloro, V.J. and Skifter, D.A. (1998) Mo-
lecular determinants of NMDA receptor pharmacological
diversity. Prog. Brain Res., 116: 171–190.
Morrison, J. (1997) Central nervous control of the bladder. In:
Jordan D. (Ed.), Central Nervous Control of Autonomic
Function. Harwood Academic Publishers, Amsterdam,
pp. 129–149.
Nadelhaft, I. and Vera, P.L. (1996) Neurons in the rat brain
and spinal cord labeled after pseudorabies virus injected into
the external urethral sphincter. J. Comp. Neurol., 375:
502–517.
Noble, L.J. and Wrathall, J.R. (1989) Correlative analyses
of lesion development and functional status after graded
spinal cord contusive injuries in the rat. Exp. Neurol., 103:
34–40.
133
Osborn, J.W., Taylor, R.F. and Schramm, L.P. (1990) Chronic
cervical spinal cord injury and autonomic hyperreflexia in
rats. Am. J. Physiol., 258: R169–R174.
Pikov, V. (2000) Recovery of the detrusor–external urethral
sphincter coordination after incomplete spinal cord injury.
Cell Biol., Georgetown University, Washington, DC.
Pikov, V., Gillis, R.A., Jasmin, L. and Wrathall, J.R. (1998)
Assessment of lower urinary tract functional deficit in rats
with contusive spinal cord injury. J. Neurotrauma, 15:
375–386.
Pikov, V. and Wrathall, J.R. (2001) Coordination of the blad-
der detrusor and the external urethral sphincter in a rat
model of spinal cord injury: effect of injury severity. J. Ne-
urosci., 21: 559–569.
Pikov, V. and Wrathall, J.R. (2002) Altered glutamate receptor
function during recovery of bladder detrusor–external ure-
thral sphincter coordination in a rat model of spinal cord
injury. J. Pharmacol. Exp. Ther., 300: 421–427.
Ramirez-Leon, V., Ulfhake, B., Arvidsson, U., Verhofstad,
A.A., Visser, T.J. and Hokfelt, T. (1994) Serotoninergic,
peptidergic and GABAergic innervation of the ventrolateral
and dorsolateral motor nuclei in the cat S1/S2 segments:
an immunofluorescence study. J. Chem. Neuroanat., 7:
87–103.
Rinaman, L., Verbalis, J.G., Stricker, E.M. and Hoffman, G.E.
(1993) Distribution and neurochemical phenotypes of caudal
medullary neurons activated to express cFos following pe-
ripheral administration of cholecystokinin. J. Comp. Neurol.,
338: 475–490.
Rosenberg, L.J., Zai, L.J. and Wrathall, J.R. (2005) Chronic
alterations in the cellular composition of spinal cord white
matter following contusion injury. Glia, 49: 107–120.
Schalow, G., Bersch, U., Michel, D. and Koch, H.G. (1995)
Detrusor–sphincteric dyssynergia in humans with spinal cord
lesions may be caused by a loss of stable phase relations
between and within oscillatory firing neuronal networks of
the sacral micturition center. J. Auton. Nervous System, 52:
181–202.
Schroder, H.D. (1980) Organization of the motoneurons in-
nervating the pelvic muscles of the male rat. J. Comp. Ne-
urol., 192: 567–587.
Tang, Y., Rampin, O., Calas, A., Facchinetti, P. and Giuliano,
F. (1998) Oxytocinergic and serotonergic innervation of
identified lumbosacral nuclei controlling penile erection in
the male rat. Neuroscience, 82: 241–254.
Teng, Y.D. and Wrathall, J.R. (1997) Local blockade of sodium
channels by tetrodotoxin ameliorates tissue loss and long-
term functional deficits resulting from experimental spinal
cord injury. J. Neurosci., 17: 4359–4366.
Testa, R., Guarneri, L., Poggesi, E., Angelico, P., Velasco, C.,
Ibba, M., Cilia, A., Motta, G., Riva, C. and Leonardi, A.
(1999) Effect of several 5-hydroxytryptamine(1A) receptor
ligands on the micturition reflex in rats: comparison with
WAY 100635. J. Pharmacol. Exp. Ther., 290: 1258–1269.
Thor, K.B. (2003) Serotonin and norepinephrine involvement in
efferent pathways to the urethral rhabdosphincter: implica-
tions for treating stress urinary incontinence. Urology, 62: 3–9.
134
Tiseo, P.J. and Yaksh, T.L. (1990) The spinal pharmacology of
urinary function: studies on urinary continence in the un-
anaesthetized rat. Ciba Found. Symp., 151: 91–104 discus-
sion 104–109.
Trautner, B.W. and Darouiche, R.O. (2002) Prevention of uri-
nary tract infection in patients with spinal cord injury.
J. Spinal Cord Med., 25: 277–283.
Vizzard, M.A. (2000) Increased expression of spinal cord Fos
protein induced by bladder stimulation after spinal cord in-
jury. Am. J. Physiol. Regul. Integr. Comp. Physiol., 279:
R295–R305.
Vizzard, M.A., Erickson, V.L., Card, J.P., Roppolo, J.R. and de
Groat, W.C. (1995) Transneuronal labeling of neurons in the
adult rat brainstem and spinal cord after injection of pseudo-
rabies virus into the urethra. J. Comp. Neurol., 355: 629–640.
Wrathall, J.R. (1994) Behavioral analysis in experimental spinal
cord trauma. In: Salzman S. and Faden A. (Eds.), Neuro-
biology of Central Nervous System Trauma. Oxford Uni-
versity Press, NY, pp. 79–85.
Wrathall, J.R., Choiniere, D. and Teng, Y.D. (1994) Dose-de-
pendent reduction of tissue loss and functional impairment
after spinal cord trauma with the AMPA/kainate antagonist
NBQX. J. Neurosci., 14: 6598–6607.
Wrathall, J.R., Li, W. and Hudson, L.D. (1998) Myelin gene
expression after experimental contusive spinal cord injury.
J. Neurosci., 18: 8780–8793.
Wrathall, J.R., Pettegrew, R.K. and Harvey, F. (1985) Spinal
cord contusion in the rat: production of graded, reproduc-
ible, injury groups. Exp. Neurol., 88: 108–122.
Wyndaele, J.J. (1991) Investigation of the afferent nerves of the
lower urinary tract in patients with ‘complete’ and ‘incom-
plete’ spinal cord injury. Paraplegia, 29: 490–494.
Yoshiyama, M., Nezu, F.M., Yokoyama, O., de Groat, W.C.
and Chancellor, M.B. (1999) Changes in micturition after
spinal cord injury in conscious rats. Urology, 54: 929–933.
Yoshiyama, M., Roppolo, J.R. and de Groat, W.C. (1994) Al-
teration by urethane of glutamatergic control of micturition.
Eur. J. Pharmacol., 264: 417–425.
Yoshiyama, M., Roppolo, J.R. and de Groat, W.C. (1995) In-
teractions between NMDA and AMPA/kainate receptors in
the control of micturition in the rat. Eur. J. Pharmacol., 287:
73–78.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 9

Role of the urothelium in urinary bladder

dysfunction following spinal cord injury

Lori A. Birder

Departments of Medicine and Pharmacology, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261, USA

Abstract: A consequence of spinal cord injury is a change in bladder reflex pathways resulting in the
emergence of detrusor hyperreflexia and increased activity of the urethral sphincter. A basis for some of
these alterations could be changes in the environment of bladder sensory nerve endings at the target organ.
Recent evidence suggests that the urothelium (the lining of the urinary bladder) plays a prominent role in
modulating bladder sensory nerve ending excitability. It is conceivable that factors and processes affecting
the plasticity of bladder neurons after spinal cord injury may be partly due to changes occurring in the
urothelium. Although the urothelium has classically been thought of as a passive barrier to ions/solutes, a
number of novel properties have been recently attributed to these cells. Our work and that of others clearly
demonstrates that the urothelium exhibits both ‘‘sensor’’ (expression of sensor molecules or response to
thermal, mechanical and chemical stimuli) as well as ‘‘transducer’’ (release of factors/transmitters) prop-
erties. Taken together, these and other findings discussed in this chapter suggest a sensory function for the
urothelium and that alterations in urothelial properties may contribute to afferent abnormalities following
spinal cord injury.

The urothelium: an effective barrier against solutes
and pathogens
The bladder urothelium is a specialized lining of
the urinary tract, extending from the renal pelvis
to the urethra. The urothelium is composed of at
least three layers: a basal cell layer attached to a
basement membrane, an intermediate layer and a
superficial apical layer with large (diameters of
25–250 mm) hexagonal ‘‘umbrella’’ cells (Lewis,
2000; Acharya et al., 2004). It has been reported
for some species that both the umbrella and per-
haps intermediate cells may have projections to the
basement membrane (Martin, 1972; Hicks, 1975;
Apodaca, 2004). The basal cells, which are thought
to be precursors for other cell types, normally ex-
hibit a low (3–6 months) turnover rate; however,
Corresponding author. Tel.: +412-383-7368;
Fax: +412-648-7197; E-mail: lbirder@pitt.edu
accelerated proliferation can occur in pathology.
For example, using a model which creates a selec-
tive injury of apical urothelial cells (protamine
sulfate), it has been shown that, in response to
injury, the urothelium undergoes both functional
and structural changes in order to restore the bar-
rier (Lavelle et al., 2002).
The umbrella cells function as a barrier against
most substances found in urine thus protecting the
underlying tissues (Negrete et al., 1996; Zeidel,
1996; Lewis, 2000; Apodaca, 2004). When this
function is compromised during injury or inflam-
mation, it can result in the passage of toxic sub-
stances into the underlying tissue (neural/muscle
layers) resulting in urgency, frequency and dysuria.
The superficial or umbrella cells play a prominent
role in maintaining this barrier, and exhibit a
number of unique properties including specialized
membrane lipids, asymmetric unit membrane par-
ticles and a plasmalemma with stiff plaques (Lewis,

DOI: 10.1016/S0079-6123(05)52009-0 135

136
2000; Hu et al., 2002; Apodaca, 2004). These cells
are also interconnected with extensive junctional
complexes which include cytoskeletal elements,
and cytoplasmic and transmembrane proteins,
some of which play a role in cell–cell adhesion
(Lewis, 2000; Acharya et al., 2004; Apodaca,
2004). This ‘‘water-tight’’ function of the apical
cell membrane is partly due to the above-mentioned
specialized lipid molecules and uroplakin proteins,
a major protein component of the apical cell
membrane, which reduce the permeability of
the urothelium to small molecules (water, urea,
protons). Tight junction complexes are thought to
reduce the movement of ions and solutes between
cells (Tammela et al., 1993; Lewis, 2000; Wang
et al., 2003; Apodaca, 2004).
Sensor and transducer functions of urothelium
Urothelial cells exhibit a number of properties
similar to neurons (nociceptors/mechanorecep-
tors): both types of cells use diverse signal-trans-
duction mechanisms to detect physiological
stimuli. Examples of ‘‘sensor molecules’’ (i.e., re-
ceptors/ion channels) associated with neurons that
have been thus far identified in urothelium include
receptors for: bradykinin (Chopra et al., 2005),
neurotrophins (trkA and p75) (Wolf-Johnston
et al., 2004), purines (P2X and P2Y) (Lee et al.,
2000; Birder et al., 2004a; Sun and Chai, 2004;
Tempest et al., 2004), norepinephrine (a- and b-)
(Birder et al., 1998, 2002b), acetylcholine (mu-
scarinic and nicotinic) (Beckel et al., 2002, 2004;
Chess-Williams, 2002). Other sensor molecules
identified in urothelial cells are protease-activated
receptors (D’Andrea et al., 2003), mechanosensi-
tive Na+ channels (Lewis and Hanrahan, 1985;
Wellner and Isenberg, 1993; Smith et al., 1998;
Carattino et al., 2005) and a number of transient
receptor potential (TRP) channels (TRPV1,
TRPV2, TRPV4, TRPM8) (Birder et al., 2001,
2002a; Barrick et al., 2003; Stein et al., 2004).
The ability of urothelial cells to express ‘‘sensor
molecules’’ and release chemical mediators (nitric
oxide, adenosine triphosphate (ATP), acetylcho-
line, substance P, prostaglandins) suggests that
these cells exhibit specialized sensory and signaling
properties that could allow reciprocal communi-
cation with neighboring urothelial cells as well
as nerve fibers or other cell types (i.e., immune,
myofibroblasts, inflammatory) in the bladder wall.
Recent studies have shown that afferent as well as
autonomic axons are located in close proximity to
the urothelium (Birder et al., 2001; Beckel et al.,
2004). Peptide- and TRPV1-immunoreactive nerve
fibers have been found localized throughout the
urinary bladder musculature and in a plexus
beneath, and extending into, the urothelium. This
suggests that the release of a number of mediators
from the urothelium could alter bladder nerve
excitability and, in turn, release of mediators from
nearby bladder nerves may also impact urothelial
function. In support of this idea is evidence that
ATP (released from urothelial cells during stretch)
can activate a population of suburothelial bladder
afferents expressing P2X3 purine receptors (Ferguson
et al., 1997; Burnstock, 2001), signaling changes in
bladder fullness and pain (Vlaskaovaka et al.,
2001). Accordingly, P2X3 null mice exhibit urinary
bladder hyporeflexia, suggesting that this receptor
and neural–epithelial interactions are necessary for
normal bladder function (Cockayne et al., 2000).
Thus, the activation of bladder nerves and
urothelial cells could modulate urinary bladder
function directly or indirectly via the release
of chemical factors in the urothelial layer. This
type of regulation may be similar to the epithelial-
dependent secretion of chemical factors in airway
epithelium thought to modulate submucosal
nerves and bronchial smooth muscle tone
(Homolya et al., 2000; Jallat-Daloz et al., 2001).
The cellular mechanism(s) by which stretch
evokes the release of ATP from epithelial, end-
othelial or other cell types is unclear. One hypoth-
esis is that cellular distension causes intracellular
vesicles rich in ATP to fuse with the urothelial cell
membrane promoting ATP release and an increase
in umbrella cell surface area during bladder filling.
This mechanism would allow the bladder to ex-
pand its epithelial surface area as urine accumu-
lates. Consistent with this possibility, mechanical
distension of the excised bladder has been shown
to trigger an increase in the membrane capacitance
of the apical urothelial surface (Truschel et al.,
2002).

Impact of spinal cord injury on urothelial cell barrier
function and morphology
Spinal cord injury (transection of the spinal cord
in rats at level T8-9) resulted in changes in both
urothelial morphology and barrier function
(Apodaca et al., 2003). At 24 h post spinal cord
injury, there is a decrease in transepithelial resist-
ance and an increase in permeability to water and
urea. The urothelium also exhibits a number of
regions which lack umbrella cells (Figs. 1E and F).
The alterations in ultrastructure and function
worsen within a few days following spinal cord
injury, when significant disruptions in the urothe-
lium are observed (Figs. 1G and H) correlating
with decreased transepithelial resistance and in-
creased permeabilities. Although some of these
changes could be the downstream consequence of
barrier disruption, some of the alterations could be
due to cord injury-induced urinary retention and
bladder overdistension. The bladders were not
cannulated to avoid catheter-induced injury/in-
flammation, but were manually expressed several
times a day. Following recovery of the spinal reflex
pathways and emergence of automatic micturition
14–28 days after spinal cord injury, barrier func-
tion was re-established, although the morphology
of the urothelium was altered and the superficial
urothelial cells were smaller (Figs. 1I–L).
Examination of urothelial morphology and
function at earlier time points (1–2 h after tran-
section) revealed significant changes in urothelial
morphology including areas of urothelium which
lacked apical cells (Figs. 1C and D). These findings
correlated with decreased transepithelial resist-
ance, which suggests disruption of the tight junc-
tions and cell–cell contact. In contrast, only minor
alterations in urea and water permeabilities oc-
curred at this time. The reason the permeabilities
did not change is unknown. One possibility is the
underlying cells still provide an adequate barrier
during the acute phase of spinal cord injury.
A number of mechanisms may contribute to
these acute urothelial changes including increased
autonomic activity (i.e., release of catecholamines
from stimulated efferent nerves) following spinal
cord injury. We have shown that pretreatment with
the ganglionic blocking agent, hexamethonium,
137
prevents the ‘‘acute’’ cord injury-induced disrup-
tions in both epithelial morphology and barrier
function (Figs. 2A–D). These and other findings
suggest the involvement of the autonomic nervous
system in the acute effects of cord injury on
the bladder urothelium. Studies have also shown
that the release of stress hormones (i.e., nor-
epinephrine) can disrupt the urothelial tight junc-
tions with loss of urothelial cells (Veranic and
Jezernik, 2000). Thus, it is possible that transec-
tion of the spinal cord could induce the release of
‘‘stress-hormones’’ which could contribute to the
changes observed in the urothelium at early time
points following spinal cord injury. We have found
that intravesical administration of norepinephrine
significantly altered epithelial permeability
(decrease in transepithelial resistance) compared
to controls (Birder et al., unpublished results).
Although the mechanism for these changes is
under investigation, one possibility is adrenergic-
induced release of a soluble factor such as nitric
oxide from urothelial cells (Birder et al., 2002b),
which in excess levels can alter the barrier function
of the urothelium (Truschel et al., 2002). Alterna-
tively, cord injury-dependent neurotransmitter re-
lease from efferent nerves could stimulate mast cell
release of histamine, bradykinin, prostaglandins,
leukotrienes and proteases, all of which could
stimulate urothelial cells and contribute to tissue
damage and inflammation. In addition to the re-
lease of catecholamines from bladder efferent
nerves, other modulators released from immune
cells might also play a role in the loss of barrier
function after spinal cord injury.
Reports suggest that capsaicin-sensitive nerves
may contribute to mucosal protection following
injury or inflammation (Abdel-Salam et al., 1999).
The neurotoxin capsaicin was therefore used to
evaluate the involvement of capsaicin-sensitive
bladder afferents in changes in mucosa ultrastruc-
ture and permeability after acute spinal cord injury
(Figs. 2E–H). In these studies, capsaicin-pretreat-
ment did not prevent functional changes but en-
hanced the susceptibility of the mucosa to injury
by decreasing transepithelial resistance compared
to untreated cord-injured animals or capsaicin-
treated controls. It has been suggested that the
effect of capsaicin may be due to alterations in
138

Fig. 1. Scanning electron micrograph (SEM) of urothelial images taken from control and at various time points following spinal cord
injury (SCI). (A and B), Images taken from sham-treated animals; or 2 h (C and D), 24 h (E and F), 3 days (G and H), 14 days (I and J)
or 28 days (K and L) following SCI. Panels on the right depict a higher magnification view of inset regions shown on the left. (With
permission from Apodaca et al., Am. J. Physiol., 2003.) Arrows in F indicate small shrunken cells associated with epithelium.
 139

Fig. 2. Scanning electron micrograph (SEM) of images depicting effects of either hexamethonium (50 mM) or capsaicin (100 mg/kg s.c.,
4 days prior) on acute changes in urothelial ultrastructure 2 h following spinal cord injury (SCI). (A and B) pretreatment with
hexamethonium before SCI; (C and D) hexamethonium pretreatment prior to sham surgery; (E and F) capsaicin pretreatment (4 days
prior) then SCI; (G and H) capsaicin pretreatment then sham surgery. Panels on the right depict a higher magnification view of inset
regions shown on the left. (With permission from Apodaca et al., Am. J. Physiol., 2003.)

substance P/calcitonin gene-related peptide con-
tent in capsaicin-sensitive nerves (Szolcsanyi and
Bartho, 2001). Although we did not detect a
significant difference in the surface architecture
between treated animals and controls, dilation of
extracellular spaces or alterations in tight junctions
leading to increased permeability may not be
evident using the employed morphological
approaches.
Thus, in addition to affecting the bladder de-
trusor muscle and its innervation (de Groat, 1995),
spinal cord injury also leads to a rapid disruption
140
of urothelial barrier function. This is evident by a
loss of cell–cell interactions, decreased transepi-
thelial resistance and increased water and urea
permeabilities. Our results indicate that release of
neurotransmitters by bladder efferent nerves is at
least partially responsible for the disruption of the
urothelium. Although the neurotransmitters and/
or inflammatory mediators that disrupt barrier
function are not well characterized, our observa-
tions indicate that bladder nerves play an important
role in regulating and maintaining barrier function.
Impact of spinal cord injury on urothelial cell sensor
and transducer properties
Sensitization of urothelial cells and afferents can
be triggered by various mediators (nerve growth
factor, ATP, nitric oxide, prostaglandins) released
by both neuronal and non-neuronal cells (urothe-
lial cells, fibroblasts, mast cells) located near the
luminal surface of the bladder. Increased endog-
enous levels of nerve growth factor and/or
urothelial receptors for nerve growth factor (p75;
trkA) have been detected in the target organ
(smooth muscle and urothelium) in a number of
bladder pathologies (Steers et al., 1991; Vizzard,
2000; Jallat-Daloz et al., 2001; Wolf-Johnston
et al., 2003; Wolf-Johnston et al., 2004). More-
over, altered nerve growth factor levels (even in the
absence of inflammation) have been linked to
changes in the properties of afferent pathways
(Kornblum and Johnson, 1982; Dmitrieva and
McMahon, 1996; Lamb et al., 2004). Thus, nerve
growth factor may play a significant role in en-
hancing the sensitivity of a number of ‘‘sensor
molecules’’ within both the urothelium and senso-
ry neurons. These findings suggest that targeting
nerve growth factor and/or nerve growth factor
signaling mechanisms may provide important in-
sight into new therapies for urinary bladder dys-
function caused by inflammation or injury.
Another important component of the injury/in-
flammatory response is ATP release from various
cell types including the urothelium, which can in-
itiate painful sensations by exciting purinergic
(P2X) receptors on sensory fibers (Cockayne
et al., 2000; Burnstock, 2001). Recently, it has
been shown in sensory nerves that ATP can po-
tentiate the response of vanilloid receptors (caps-
aicin, protons and moderate heat act on vanilloid
receptors) by lowering the threshold for responses
to these stimuli (Tominaga et al., 2001). This rep-
resents a novel mechanism through which the large
amounts of ATP released from damaged or sen-
sitized cells in response to injury or inflammation
may trigger increased excitability of afferent
nerves. These findings have clinical significance
and suggest that alterations in afferents or epithe-
lial cells in pelvic viscera may contribute to the
sensory abnormalities in a number of pelvic dis-
orders, including interstitial cystitis, a chronic
painful condition of the urinary bladder (Nickel,
2003). In a comparable disease in cats, termed fe-
line interstitial cystitis (Buffington et al., 1999) we
reported alterations in stretch-evoked release of
urothelium-derived ATP (Birder et al., 2003), con-
sistent with the augmented release of ATP from
urothelial cells from some patients with interstitial
cystitis (Sun et al., 2001). We have recently found
similar results in chronic spinal cord-injured cats
(Birder et al., unpublished results), in which the
augmented stretch-evoked ATP release from urothe-
lial cells may contribute to bladder hyperreflexia.
The urothelium maintains a tight barrier to ion/
solute flux and augmented release of mediators
such as nitric oxide from urothelial cells and/or
nearby bladder nerves may play a role in the
maintenance and regulation of this urothelial bar-
rier function. Our previous studies have demon-
strated an upregulation in inducible nitric oxide
synthase and elevated basal levels of nitric oxide
measured in the bladder mucosa in cats with in-
terstitial cystitis (Birder et al., 2005). Similar find-
ings have been obtained clinically, as some patients
with classic interstitial cystitis also demonstrate
elevated release of nitric oxide (Hosseini et al.,
2004). As increased nitric oxide has been linked to
cellular damage and alterations in epithelial bar-
rier function (Salzman et al., 1995; Arkovitz et al.,
1997), and it has been demonstrated that feline
interstitial cystitis is accompanied by changes in
bladder permeability and urothelial ultrastructure
(Lavelle et al., 2000), we examined whether nitric
oxide levels might also be altered in rats following
spinal cord injury, which results in changes in both

urothelial morphology and function (Apodaca
et al., 2003). In urothelium from normal rats,
basal nitric oxide release (measured from the urinary
bladder mucosal surface) remained undetectable
(o10 nM nitric oxide release). However, after
chronic spinal cord injury, we detected elevated
levels of basal nitric oxide release (200–500 nM)
recorded from the mucosal surface of the urinary
bladder (Truschel et al., 2001).
To evaluate the impact of elevated mucosal ni-
tric oxide levels on epithelial function, we exam-
ined the effects of prolonged exposure to high
concentrations (2.5–5 mM) nitric oxide donors
(S-nitroso-N-acetyl penicillamine or sodium nit-
roprusside) on cultured urothelial cells. It has been
previously shown that cultured urothelial cells ex-
hibit properties similar to native urothelium
(Truschel et al., 1999). Normally, these urothelial
cultures exhibit a high transepithelial resistance
and low urea and water permeabilities. However,
the administration of high concentrations of nitric
oxide resulted in a significant decrease in transep-
ithelial resistance (90% decrease as well as 3–5-
fold increase in permeability to water and urea), as
compared to controls (Truschel et al., 2001). This
response was reversible upon washout of the nitric
oxide donor. Similar findings were obtained using
excised urinary bladder from rodents, in which
application of high concentrations of nitric oxide
also decreased transepithelial resistance by 60%
compared to untreated control (Truschel et al.,
2001). Although their mechanism is unknown,
these effects are reminiscent of similar observa-
tions in epithelia of other organs (lung, gut) in
which excess production of nitric oxide has been
linked to changes in epithelial integrity (Ding
et al., 2004; Han et al., 2004).
Disruption of epithelial integrity in some blad-
der pathologies may also be due to substances such
as antiproliferative factor, which has been shown
to be secreted by bladder epithelial cells from in-
terstitial cystitis patients and can inhibit epithelial
proliferation thereby adversely affecting barrier
function (Keay et al., 1999, 2004). Uropathogenic
Escherichia coli can also bind to uroplakin pro-
teins present on the apical surface of superficial
umbrella cells (Schilling and Hultgren, 2002). This
is thought to be an initial step leading to a cascade
141
of events that are thought to be part of symptoms
associated with urinary tract infections.
Taken together, modification of the urothelium
and/or loss of epithelial integrity in a number of
bladder pathologies could result in passage of tox-
ic/irritating urinary constituents through the epi-
thelium leading to changes in the properties of
sensory pathways.
Therapeutic options for spinal cord injury that could
target the urothelium
An emerging body of evidence indicates that
urothelial cells exhibit ‘‘polymodal’’ properties,
i.e., can be activated by chemical, thermal or me-
chanical stimuli, and that their activation can po-
tentially evoke the release of a myriad of
transmitters which can impact afferent activity
and ultimately bladder function. While the urot-
helium has been historically viewed as primarily a
‘‘barrier,’’ it is becoming increasingly clear that it
is a responsive structure capable of detecting phys-
iological and chemical stimuli, and of releasing a
number of signaling molecules. The following is a
summary of various therapies, most given intrave-
sically, which are traditionally thought to target
bladder nerves. It is conceivable that a number of
these treatments could also target urothelial re-
ceptors and/or release mechanisms.
Intravesical vanilloid compounds
One example of a urothelial ‘‘neuronal-like’’ sen-
sor molecule is the TRP channel TRPV1, known
to play an important role in nociception and in
urinary bladder function (Szallasi, 2001). It is well
established that the painful sensations induced by
capsaicin, the pungent substance in hot peppers,
are caused by stimulation of vanilloid receptor-1
(TRPV1), an ion channel protein which is activat-
ed by vanilloid compounds such as capsaicin,
moderate heat and protons (Caterina et al., 1997;
Caterina, 2001). TRPV1 is highly expressed in
urinary bladder unmyelinated axons (C-fiber) that
detect bladder distension or the presence of irritant
chemicals (Chancellor and de Groat, 1999).
Intravesical instillation of vanilloid compounds
142

Fig. 3. (A) Confocal image of basal cells depicting TRPV1-immunoreactivity (cy-3, red) and cytokeratin-17, a marker for these cells
(Fluorscein isothiocyanate or FITC, green) immunoreactivity. (B) Confocal image of urinary bladder urothelium reveals TRPV1-
positive (cy-3, red) nerve fibers located in close proximity to basal urothelial cells (FITC, green). Punctate TRPV1 staining in urothelial
cells was electronically subtracted to facilitate imaging of the TRPV1-IR nerve fiber. (With permission from Birder et al., Proc. Natl.
Acad. Sci. USA, 2001.)

such as capsaicin or resiniferatoxin, which leads to
desensitization of bladder nerves, has been shown
to improve voiding efficiency significantly in cord-
injured animals as well as in patients with detrusor
hyperactivity (Szallasi and Fowler, 2002; Kim
et al., 2003).
One of the more remarkable findings in our own
studies is that TRPV1 is not only expressed by
afferent nerves that form close contact with
urothelial cells but also by the urothelial cells
themselves (Fig. 3) (Birder et al., 2001). Further,
TRPV1 receptor expression correlates with sensi-
tivity to vanilloid compounds, as exogenous ap-
plication of capsaicin or resiniferatoxin increases
intracellular calcium and evokes the release of
transmitters (nitric oxide, ATP) in cultured ur-
othelial cells. These responses are dependent upon
TRPV1 expression (Birder et al., 2001, 2002a). In
neurons, TRPV1 is thought to integrate/amplify
the response to various stimuli and thus plays an
essential role in the development of inflammation-
induced hyperalgesia. Thus, it seems likely that
urothelial-TRPV1 might participate in a similar
manner, in the detection of irritant stimuli follow-
ing bladder inflammation or infection.
While anatomically normal, TRPV1 null mice ex-
hibited a number of alterations in bladder function
including a reduction of in vitro, stretch-evoked
ATP release and membrane capacitance as well as
a decrease in hypotonic or stretch-evoked ATP
release from cultured TRPV1 null urothelial cells.
Thus, the functional significance of these receptors
in the bladder extends beyond pain sensation to
include participation in normal bladder function.
These receptors are also essential for normal me-
chanically evoked, purinergic signaling by the
urothelium. In addition to the known effects on
bladder nerves, intravesical use of vanilloids could
also target TRPV1 on urothelial cells, where
persistent activation of urothelial TRPV1 might
lead to receptor desensitization or depletion of
urothelial-derived transmitters.
Antimuscarinic drugs
Antimuscarinic drugs are widely regarded as a
standard treatment in patients with neurogenic
lower urinary tract dysfunction (Andersson and
Yoshida, 2003). By targeting muscarinic receptors
on bladder smooth muscle, these agents prevent
receptor stimulation by acetylcholine released
from bladder efferent nerves and promote in-
creased bladder capacity. However, these drugs are
thought to be effective during bladder storage when
parasympathetic nerves are silent. Since various
stimuli have been shown to release acetylcholine

from urothelial cells (Andersson and Yoshida,
2003; Beckel et al., 2004), it is postulated that this
release from non-neural stores (i.e., urothelium)
could also contribute to detrusor overactivity
(Andersson and Yoshida, 2003). Thus, the effec-
tiveness of some of these agents may be partly due
to targeting urothelial receptors and/or release
mechanisms. Although muscarinic receptor sub-
types have been detected on urothelial cells
(Hawthorn et al., 2000; Beckel et al., 2004), a role
for these receptors in bladder function has not yet
been established. Taken together, these data sug-
gest that the bladder urothelium may be an addi-
tional source of acetylcholine that influences
bladder contractility by modulating smooth mus-
cle tone and afferent activity.
Botulinum toxin
Recent studies have demonstrated that in-
tradetrusor injection of botulinum neurotoxin type
A (Botox) is an effective therapy in a number of
lower urinary tract disturbances including the
severe incontinence due to neurogenic detrusor
overactivity of spinal cord injury patients (Harper
et al., 2004; Reitz and Schurch, 2004). Following
injection, the toxin binds to bladder cholinergic
nerve terminals and cleaves the protein, SNAP25,
necessary for exocytosis and release of acetylcho-
line (Harper et al., 2004). In patients with spinal
cord injury, this treatment can lead to a significant
reduction in episodes of incontinence and an in-
crease in maximum bladder capacity.
There is evidence that Botox can suppress the
release of a number of mediators (acetylcholine,
ATP and neuropeptides) from both neural and
non-neural cells (Morris et al., 2001). Recent stud-
ies have demonstrated that this agent can effec-
tively reduce both chemically and mechanically
evoked ATP release from cultured urothelial cells
(Barrick et al., 2004). Alterations in the release of
ATP or other transmitters from the urothelium
could have a profound impact on neural excita-
bility. Taken together, these findings suggest that
targeting neural and non-neural release mechanisms
may be effective for the treatment of bladder
hyperreactivity in spinal cord injury.
143
Diagnostic test for spinal cord injury: the ice
water test
It has been reported that intravesical instillation of
cold solutions can unmask the presence of de-
trusor reflex activity in people with spinal cord
injury (Balmaseda et al., 1988). This ‘‘bladder
cooling reflex’’ is thought to be due to activation of
a subset of cold-sensitive C-fiber-type bladder
afferents, which are sensitive to both cold temper-
atures and menthol (Jiang et al., 2002).
Some TRP ion channels can be activated by a
wide range of temperatures as well as by natural
products (capsaicin, menthol), which can elicit
sensations of hot or cold (Patapoutian et al.,
2003). In contrast to TRPV1, which is a detector
of warm temperatures, TRPM8 has been shown to
be activated by cold temperatures as well as by
cooling agents (menthol) and is expressed in a
subset of sensory neurons (Clapham, 2003). Both
of these TRP channels are also expressed in blad-
der urothelium, suggesting that the urothelium can
express a range of thermoreceptors underlying
both ‘‘cold’’ and ‘‘heat’’ stimuli. While the func-
tional role of these thermosensitive channels in
urothelium remains to be clarified, it seems likely
that a primary role for these proteins may be
to recognize noxious stimuli in the bladder. For
example, noxious cold solutions instilled into the
urinary bladder could, via stimulation of urothelial
TRP channels, augment the release of urothelial-
derived mediators, thereby altering bladder affer-
ent excitability. Further studies are needed to
elucidate fully the role of these TRP channels in
urothelium and influence on bladder function.
Conclusion
There is considerable interest in the putative role
of urothelial receptors/ion channels and release
mechanisms in bladder function. By targeting
various urothelial sensor molecules and/or modu-
lating the release of transmitters/inflammatory me-
diators, it may be possible to modulate afferent
activity and prevent the disruption of the urothe-
lium that accompanies spinal cord injury and oth-
er bladder conditions. These results highlight the
144
need for additional studies in order to establish the
physiological relevance of these urothelial targets.
Acknowledgments
I thank Drs. A. Kanai and W. C. de Groat for
critical comments and suggestions during prepa-
ration of this chapter. The electron micrographs
were prepared by W. Giovani Ruiz. This work was
supported by grants to Lori A. Birder from the
NIH (RO1-DK-54824 and RO1-DK-57284).
References
Abdel-Salam, O.M., Debrecini, A., Mozsik, G. and Szolcsanyi, J.
(1999) Capsaicin-sensitive afferent sensory nerves in modu-
lating gastric mucosal defense against noxious agents.
J. Physiol. Paris, 93: 443–454.
Acharya, P., Beckel, J., Ruiz, W.G., Wang, E., Rojas, R.,
Birder, L. and Apodaca, G. (2004) Distribution of the tight
junction proteins ZO-1, occludin, and claudin-4, -8, and -12
in bladder epithelium. Am. J. Physiol., 287: F305–F318.
Andersson, K.E. and Yoshida, M. (2003) Antimuscarinics and
the overactive detrusor-which is the main mechanism of ac-
tion? Eur. Urol., 43: 1–5.
Apodaca, G. (2004) The uroepithelium: not just a passive bar-
rier. Traffic, 5: 1–12.
Apodaca, G., Kiss, S., Ruiz, W.G., Meyers, S., Zeidel, M.L.
and Birder, L. (2003) Disruption of bladder epithelium bar-
rier function after spinal cord injury. Am. J. Physiol., 284:
F966–F976.
Arkovitz, M.S., Wispe, J.R., Garcia, V.F. and Szabo, C. (1997)
Selective inhibition of the inducible isoform of nitric oxide
synthase prevents pulmonary transvascular flux during acute
endotoxemia. J. Pediatr. Surg., 31: 1009–1015.
Balmaseda, M.T., Reynolds, H.T. and Gordon, C. (1988) The
value of the ice water test in the management of the neuro-
genic bladder. Am. J. Physiol. Med. Rehabil., 67: 225–227.
Barrick, S.R., de Groat, W.C. and Birder, L.A. (2004)
Regulation of chemical and mechanical-evoked ATP release
from urinary bladder urothelium by botulinum toxin type A.
Soc. Neurosci. Abstr., 26: 541.5.
Barrick, S.R., Lee, H., Meyers, S., Caterina, M.J., Kanai, A.J.,
Zeidel, M.L., Chopra, B., de Groat, W.C. and Birder, L.
(2003) Expression and function of TRPV4 in urinary bladder
urothelium. Soc. Neurosci. Abstr., 26: 608.6.
Beckel, J., Barrick, S.R., Keast, J.R., Meyers, S., Kanai, A.J., de
Groat, W.C., Zeidel, M.L. and Birder, L. (2004) Expression
and function of urothelial muscarinic receptors and interac-
tions with bladder nerves. Soc. Neurosci. Abstr., 26: 846.23.
Beckel, J., Birder, L., Kiss, S., Kanai, A.J., Lee, S.J.,
Yoshiyama, M. and de Groat, W.C. (2002) Expression of
nicotinic acetylcholine receptors in rat urothelium. Soc.
Neurosci. Abstr., 25: 538.10.
Birder, L., Apodaca, G., de Groat, W.C. and Kanai, A.J. (1998)
Adrenergic- and capsaicin-evoked nitric oxide release from
urothelium and afferent nerves in urinary bladder. Am. J.
Physiol., 275: F226–F229.
Birder, L., Barrick, S.R., Roppolo, J.R., Kanai, A.J., de Groat,
W.C., Kiss, S. and Buffington, C.A. (2003) Feline interstitial
cystitis results in mechanical hypersensitivity and altered
ATP release from bladder urothelium. Am. J. Physiol., 285:
F423–F429.
Birder, L., Kanai, A.J., de Groat, W.C., Kiss, S., Nealen, M.L.,
Burke, N.E., Dineley, K.E., Watkins, S., Reynolds, I.J. and
Caterina, M.J. (2001) Vanilloid receptor expression suggests
a sensory role for urinary bladder epithelial cells. Proc. Natl.
Acad. Sci. USA, 98: 13396–13401.
Birder, L., Nakamura, Y., Kiss, S., Nealen, M.L., Barrick,
S.R., Kanai, A.J., Wang, E., Ruiz, W.G., de Groat, W.C.,
Apodaca, G., Watkins, S. and Caterina, M.J. (2002a) Altered
urinary bladder function in mice lacking the vanilloid
receptor TRPV1. Nat. Neurosci., 5: 856–860.
Birder, L., Nealen, M.L., Kiss, S., de Groat, W.C., Caterina,
M.J., Wang, E., Apodaca, G. and Kanai, A.J. (2002b)
Beta-adrenoceptor agonists stimulate endothelial nitric oxide
synthase in rat urinary bladder urothelial cells. J. Neurosci.,
22: 8063–8070.
Birder, L., Ruan, H.Z., Chopra, B., Xiang, Z., Barrick, S.R.,
Buffington, C.A., Roppolo, J.R., Ford, A.P., de Groat, W.C.
and Burnstock, G. (2004a) Alterations in P2X and P2Y
purinergic receptor expression in urinary bladder from
normal cats and cats with interstitial cystitis. Am. J. Physiol.,
287: F1084–F1091.
Birder, L., Wolf-Johnston, A.S., Buffington, C.A., Roppolo,
J.R., de Groat, W.C. and Kanai, A.J. (2005) Altered
inducible nitric oxide synthase expression and nitric oxide
production in urinary bladder from cats with feline interstitial
cystitis. J. Urol., 173: 625–629.
Buffington, C.A., Chew, D.J. and Woodworth, B.E. (1999)
Feline interstitial cystitis. J. Am. Vet. Med. Assoc., 215:
682–687.
Burnstock, G. (2001) Purine-mediated signalling in pain
and visceral perception. Trends Pharmacol. Sci., 22: 182–188.
Carattino, M.D., Sheng, S. and Kleyman, T.R. (2005) Muta-
tions in the pore region modify epithelial sodium channel
gating by shear stress. J. Biol. Chem., 280: 4393–4401.
Caterina, M.J. (2001) The vanilloid receptor: a molecular
gateway to the pain pathway. Ann. Rev. Neurosci., 24:
602–607.
Caterina, M.J., Schumacher, M.A., Tominaga, M., Rosen,
T.A., Levine, J.D. and Julius, D. (1997) The capsaicin
receptor: a heat activated ion channel in the pain pathway.
Nature, 389: 816–824.
Chancellor, M.B. and de Groat, W.C. (1999) Intravesical caps-
aicin and resiniferatoxin therapy: spicing up the ways to treat
the overactive bladder. J. Urol., 162: 3–11.
Chess-Williams, R. (2002) Muscarinic receptors of the urinary
bladder: detrusor, urothelial and prejunctional. Auton.
Autacoid. Pharmacol., 22: 133–145.

Chopra, B., Barrick, S.R., Meyers, S., Beckel, J., Zeidel, M.L.,
Ford, A.P., de Groat, W.C. and Birder, L. (2005) Expression
and function of bradykinin B1/B2 receptors in normal and in-
flamed rat urinary bladder urothelium. J. Physiol., 562: 859–871.
Clapham, D.E. (2003) TRP channels as cellular sensors. Nat.
Rev., 426: 517–524.
Cockayne, D.A., Hamilton, S.G., Zhu, Q.M., Dunn, P.M.,
Zhong, Y., Novakovic, S., Malmberg, A.B., Cain, G., Berson,
A., Kassotakis, L., Hedley, L., Lachnit, W.G., Burnstock,
G., McMahon, S.B. and Ford, A.P.D.W. (2000) Urinary
bladder hyporeflexia and reduced pain-related behaviour in
P2X(3)-deficient mice. Nature, 407: 1011–1015.
D’Andrea, M.R., Saban, M.R., Nguyen, N.B., Andrade-Gordon,
P. and Saban, R. (2003) Expression of protease-activated
receptor-1, -2, -3 and -5 in control and experimentally
inflamed mouse bladder. Am. J. Pathol., 162: 907–923.
de Groat, W.C. (1995) Mechanisms underlying the recovery of
lower urinary tract function following spinal cord injury.
Paraplegia, 33: 493–505.
Ding, L.A., Li, J.S., Li, Y.S., Zhu, N.T., Lui, F.N. and Tan, L.
(2004) Intestinal barrier damage caused by trauma and lip-
opolysaccharide. W. J. Gastroentero., 10: 2373–2378.
Dmitrieva, N. and McMahon, S.B. (1996) Sensitization of
visceral afferents by nerve growth factor in the adult rat.
Pain, 66: 87–97.
Ferguson, D.R., Kennedy, I. and Burton, T.J. (1997) ATP is
released from rabbit urinary bladder epithelial cells by hy-
drostatic pressure changes — a possible sensory mechanism?
J. Physiol., 505(2): 503–511.
Han, X., Fink, M.P., Uchiyama, T., Yang, R. and Delude, R.L.
(2004) Increased iNOS activity is essential for pulmonary
epithelial tight junction dysfunction in endotoxemic mice.
Am. J. Physiol., 286: L259–L267.
Harper, M., Fowler, C.J. and Dasgupta, P. (2004) Botulinum
toxin and its applications in the lower urinary tract. BJU Int.,
93: 702–706.
Hawthorn, M.H., Chapple, C.R., Cock, M. and Chess-
Williams, R. (2000) Urothelium-derived inhibitory factor(s)
influences on detrusor muscle contractility in vitro. Br. J.
Pharmacol., 129: 416–419.
Hicks, M. (1975) The mammalian urinary bladder: an accom-
modating organ. Biol. Rev., 50: 215–246.
Homolya, L., Steinberg, T.H. and Boucher, R.C. (2000) Cell to
cell communication in response to mechanical stress via
bilateral release of ATP and UTP in polarized epithelia.
J. Cell Biol., 150: 1349–1360.
Hosseini, A., Ehren, I. and Wiklund, N.P. (2004) Nitric oxide as
an objective marker for evaluation of treatment response in
patients with classic interstitial cystitis. J. Urol., 172: 2261–2265.
Hu, P., Meyers, S., Liang, F.X., Deng, F.M., Kachar, B., Zeidel,
M.L. and Sun, T.T. (2002) Role of membrane proteins in
permeability barrier function: uroplakin ablation elevates
urothelial permeability. Am. J. Physiol., 283: F1200–F1207.
Jallat-Daloz, I., Cognard, J.L., Badet, J.M. and Regnard, J.
(2001) Neural-epithelial cell interplay: in vitro evidence that
vagal mediators increase PGE2 production by human nasal
epithelial cells. Allergy Asthma Proc., 22: 17–23.
145
Jiang, C.H., Mazieres, L. and Lindstrom, S. (2002) Cold- and
menthol-sensitive C afferents of cat urinary bladder. J.
Physiol., 543(1): 211–220.
Keay, S., Warren, J.W., Zhang, C.O., Tu, L.M., Gordon, D.A.
and Whitmore, K.E. (1999) Antiproliferative activity is
present in bladder but not renal pelvic urine from intersti-
tial cystitis patients. J. Urol., 162: 1487–1489.
Keay, S.K., Szekely, Z., Conrads, T.P., Veenstra, T.D., Barchi,
J.J., Zhang, C.O., Koch, K.R. and Michejda, C.J. (2004) An
antiproliferative factor from interstitial cystitis patients is a
frizzled 8 protein-related sialoglycopeptide. Proc. Natl. Acad.
Sci. USA, 101: 11803–11808.
Kim, J.H., Rivas, D.A., Shenot, P.J., Green, B., Kennelly, M.,
Erickson, J.R., O’Leary, M., Yoshimura, N. and Chancellor,
M.B. (2003) Intravesical resiniferatoxin for refractory
detrusor hyperreflexia: a multicenter, blinded, randomized,
placebo-controlled trial. J. Spinal Cord Med., 26: 358–363.
Kornblum, H.I. and Johnson, E.M. (1982) Time and dose
dependencies of effects of nerve growth factor on sympa-
thetic and sensory neurons in neonatal rats. Brain Res.,
234: 41–51.
Lamb, K., Gebhart, G.F. and Bielefeldt, K. (2004) Increased
nerve growth factor expression triggers bladder overactivity.
J. Pain, 5: 150–156.
Lavelle, J., Meyers, S., Ramage, R., Bastacky, S., Doty, D.,
Apodaca, G. and Zeidel, M.L. (2002) Bladder permeability
barrier: recovery from selective injury of surface epithelial
cells. Am. J. Physiol., 283: F242–F253.
Lavelle, J., Meyers, S.A., Ruiz, W.G., Buffington, C.A., Zeidel,
M.L. and Apodaca, G. (2000) Urothelial pathophysiological
changes in feline interstitial cystitis: a human model. Am. J.
Physiol., 278: F540–F553.
Lee, H.Y., Bardini, M. and Burnstock, G. (2000) Distribution
of P2X receptors in the urinary bladder and the ureter of the
rat. J. Urol., 163: 2002–2007.
Lewis, S.A. (2000) Everything you wanted to know about the
bladder epithelium but were afraid to ask. Am. J. Physiol.,
278: F867–F874.
Lewis, S.A. and Hanrahan, J.W. (1985) Apical and basolateral
membrane ionic channels in rabbit urinary bladder epitheli-
um. Pflugers Arch., 405: S83–S88.
Martin, B.F. (1972) Cell replacement and differentiation in
transitional epithelium: a histological and autoradiographic
study of the guinea-pig bladder and ureter. J. Anat., 112:
433–455.
Morris, J.L., Jobling, P. and Gibbins, I.L. (2001) Differential
inhibition by botulinum neurotoxin A of cotransmitters re-
leased from autonomic vasodilator neurons. Am. J. Physiol.,
281: H2124–H2127.
Negrete, H.O., Lavelle, J., Berg, J., Lewis, S.A. and Zeidel,
M.L. (1996) Permeability properties of the intact mammalian
bladder epithelium. Am. J. Physiol., 271: F886–F894.
Nickel, J.C. (2003) Interstitial cystitis — an elusive clinical
target? J. Urol., 170: 816–817.
Patapoutian, A., Peier, A.P., Story, G.M. and Viswanath, V.
(2003) ThermoTRPs and beyond: mechanisms of tempera-
ture sensation. Nat. Rev. Neurosci., 4: 529–539.
146
Reitz, A. and Schurch, B. (2004) Intravesical therapy options for
neurogenic detrusor overactivity. Spinal Cord, 42: 267–272.
Salzman, A.L., Menconi, M.J., Unno, N., Ezzell, R.M., Casey,
D.M., Gonzalez, P.K. and Fink, M.P. (1995) Nitric oxide
dilates tight junctions and depletes ATP in cultured Caco-
2BBe intestinal epithelial monolayers. Am. J. Physiol., 268:
G361–G373.
Schilling, J. and Hultgren, S. (2002) Recent advances into the
pathogenesis of recurrent urinary tract infections: the bladder
as a reservoir for uropathogenic Escherichia coli. Int. J.
Antimicrob. Agents, 19: 457–460.
Smith, P.R., Mackler, S.A., Weiser, P.C., Brooker, D.R., Ahn,
Y.J., Harte, B.J., McNulty, K.A. and Kleyman, T.R. (1998)
Expression and localization of epithelial sodium channel in
mammalian urinary bladder. Am. J. Physiol., 274: F91–F96.
Steers, W.D., Ciambotti, J., Etzel, B., Erdman, S. and de Groat,
W.C. (1991) Alterations in afferent pathways from the
urinary bladder of the rat in response to partial urethral
obstruction. J. Comp. Biol., 310: 401–410.
Stein, R.J., Santos, S., Nagatomi, J., Hayashi, Y., Minnery,
B.S., Xavier, M., Patel, A.S., Nelson, J.B., Futrell, W.J.,
Yoshimura, N., Chancellor, M.B. and deMiguel, F. (2004)
Cool (TRPM8) and hot (TRPV1) receptors in the bladder
and male genital tract. J. Urol., 172: 1175–1178.
Sun, Y. and Chai, T.C. (2004) Up-regulation of P2X3 receptor
during stretch of bladder urothelial cells from patients with
interstitial cystitis. J. Urol., 171: 448–452.
Sun, Y., Keay, S., DeDeyne, P.G. and Chai, T.C. (2001)
Augmented stretch activated adenosine triphosphate release
from bladder uroepithelial cells in patients with interstitial
cystitis. J. Urol., 166: 1951–1956.
Szallasi, A. (2001) Vanilloid receptor ligands: hopes and
realities for the future. Drugs Aging, 18: 561–573.
Szallasi, A. and Fowler, C.J. (2002) After a decade of
intravesical vanilloid therapy: still more questions than
answers. Lancet Neurol., 1: 167–172.
Szolcsanyi, J. and Bartho, L. (2001) Capsaicin-sensitive affer-
ents and their role in protection: an update. J. Physiol. Paris,
95: 181–188.
Tammela, T., Wein, A.J., Monson, F.C. and Levin, R.M.
(1993) Urothelial permeability of the isolated whole bladder.
Neurourol. Urodynam., 12: 39–47.
Tempest, H.V., Dixon, A.K., Turner, W.H., Elneil, S., Sellers,
L.A. and Ferguson, D.R. (2004) P2X and P2Y receptor
expression in human bladder urothelium and changes in
interstitial cystitis. BJU Int., 93: 1344–1388.
Tominaga, M., Wada, M. and Masu, M. (2001) Potentiation
of capsaicin receptor activation by metabotropic ATP
receptors: a possible mechanism for ATP-evoked pain and
hypersensitivity. Proc. Natl. Acad. Sci. USA, 29: 820–825.
Truschel, S.T., Ruiz, W.G., Kanai, A.J., Kiss, S., Meyers, S.,
Apodaca, G., de Groat, W.C., Maggi, C.A., Zeidel, M.L. and
Birder, L. (2001) Involvement of nitric oxide (NO) in bladder
afferent and urothelial abnormalities following chronic spinal
cord injury. Soc. Neurosci. Abstr., 24: 842.9.
Truschel, S.T., Ruiz, W.G., Shulman, T., Pilewski, J., Sun,
T.T., Zeidel, M.L. and Apodaca, G. (1999) Primary uroep-
ithelial cultures. A model system to analyze umbrella cell
barrier function. J. Biol. Chem., 274: 15020–15029.
Truschel, S.T., Wang, E., Ruiz, W.G., Leung, S.M., Rojas, R.,
Lavelle, J., Zeidel, M.L., Stoffer, D. and Apodaca, G. (2002)
Stretch-regulated exocytosis/endocytosis in bladder umbrella
cells. Mol. Biol. Cell, 13: 830–846.
Veranic, P. and Jezernik, K. (2000) The response of junctional
complexes to induced desquamation in mouse bladder urot-
helium. Biol. Cell., 92: 105–113.
Vizzard, M.A. (2000) Changes in urinary bladder neurotrophic
factor mRNA and NGF protein following urinary bladder
dysfunction. Exp. Neurol., 161: 273–284.
Vlaskaovaka, M., Kasakov, L., Rong, W.F., Bodin, R., Bardini,
M., Cockayne, D.A., Ford, A.P.D.W. and Burnstock, G.
(2001) P2X3 knockout mice reveal a major sensory role for
urothelially released ATP. J. Neurosci., 21: 5670–5677.
Wang, E., Lee, J.M., Johnson, J.P., Kleyman, T., Bridges, R.
and Apodaca, G. (2003) Hydrostatic pressure-regulated ion
transport in bladder uroepithelium. Am. J. Physiol., 285:
F651–F663.
Wellner, M.C. and Isenberg, G. (1993) Properties of stretch-
activated channels in myocytes from the guinea pig urinary
bladder. J. Physiol., 466: 213–227.
Wolf-Johnston, A.S., Beckel, J.M., Barrick, S.R., Kanai, A.J.,
de Groat, W.C. and Birder, L. (2004) Induction of urinary
cystitis results in increased expression of nerve growth factor
and its receptors. Soc. Neurosci. Abstr., 27: 742.16.
Wolf-Johnston, A.S., Buffington, C.A., Roppolo, J.R., Tai, C.,
Kanai, A.J., de Groat, W.C. and Birder, L. (2003) Increased
NGF and TRPV1 expression in urinary bladder and sensory
neurons from cats with feline interstitial cystitis. Soc.
Neurosci. Abstr., 27: 608.4.
Zeidel, M.L. (1996) Low permeabilities of apical membranes of
barrier epithelia: what makes water-tight membranes water-
tight? Am. J. Physiol., 271: F243–F245.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 10

Plasticity in the injured spinal cord: can we use it

to advantage to reestablish effective bladder voiding
and continence?

Natasha D.T. Zinck1, and John W. Downie1,2

1
Department of Pharmacology, Faculty of Medicine, Dalhousie University, 5850 College St., Halifax, NS B3H 1X5,
Canada
2
Department of Urology, Faculty of Medicine, Dalhousie University, 5850 College St., Halifax, NS B3H 1X5, Canada

Abstract: Micturition is coordinated at the level of the spinal cord and the brainstem. Spinal cord injury
therefore directly interrupts spinal neuronal pathways to the brainstem and results in bladder areflexia.
Some time after injury, however, dyssynergic bladder and sphincter function emerges. The changes me-
diating the appearance of bladder function after spinal cord injury are currently unknown. Primary afferent
neurons have been shown to sprout in response to spinal cord injury. Sprouting primary afferents have been
linked to the pathophysiology of centrally manifested disorders, such as autonomic dysreflexia and ne-
uropathic pain. It is proposed that sprouting of bladder primary afferents contributes to disordered bladder
functioning after spinal cord injury. During development of the central nervous system, the levels of specific
neuronal growth-promoting and guidance molecules are high. After spinal cord injury, some of these
molecules are upregulated in the bladder and spinal cord, suggesting that axonal outgrowth is occurring.
Sprouting in lumbosacral spinal cord is likely not restricted to neurons involved in the micturition reflex.
Furthermore, sprouting of some afferents may be contributing to bladder function after injury, whereas
sprouting of others might be hindering emergence of function. Thus selective manipulation of sprouting
targeting afferents that are contributing to emergence of bladder function after injury is critical. Further
research regarding the role that neuronal sprouting plays in the emergence of bladder function may con-
tribute to improved treatment of bladder dyssynergia after spinal cord injury.

Introduction neurons exist within an environment rich in mole-

Plasticity within the spinal cord
Axonal outgrowth and pathfinding are limited
within the adult central nervous system (CNS).
During development, however, axonal outgrowth
and pathfinding are critical for a properly wired,
and thus, functioning nervous system (Crowley
et al., 1994; Maier et al., 1999). Developing
Corresponding author. Tel.: +(902) 494-3459;
Fax: +(902) 494-1388; E-mail: nzinck@dal.ca
cules that are important for axonal guidance,
outgrowth and targeting (Drescher et al., 1997;
Dickson and Senti, 2002). These molecules act by
either attracting or repelling the leading edge of
growing axons (the growth cone), thus directing
developing axons to their appropriate targets
(Gallo and Letourneau, 2004; Gordon-Weeks,
2004). It has been proposed that there is a de-
crease or absence in growth-promoting molecules
once adulthood is reached, and that this is why the
adult CNS is unable to promote neuronal plastic-
ity. For this reason, much research has focused on

DOI: 10.1016/S0079-6123(05)52010-7 147

148
reproducing the molecular environment that is
present during development with the hopes of aid-
ing neuronal repair and regrowth after CNS injury.
On the other hand, CNS plasticity may also have
negative consequences. There is increasing evidence
that upregulation of key elements involved in aid-
ing axonal outgrowth contributes to the pathogen-
esis of centrally manifested disorders. These include
neuropathic pain (Theodosiou et al., 1999) as well
as autonomic dysreflexia (Krenz et al., 1999). Spi-
nal cord injury, depending on its severity, affects
many centrally mediated visceral functions, includ-
ing bladder function. Normal micturition requires
supraspinal integration at the level of the pontine
micturition center (Barrington, 1915). Suprasacral
spinal cord injury disconnects the parasympathetic
spinal outflow from the pons rendering the bladder
areflexic. Bladder function is novel in that micturit-
ion emerges some time after spinal cord injury
without neuronal integration at the level of the
brainstem (Yoshiyama et al., 1999). However, post
spinal cord injury, voiding is dysfunctional due to
lack of coordination between the detrusor muscle
and the external urethral sphincter, clinically
termed detrusor–sphincter dyssynergia (Kruse
et al., 1993). For spinal cord-injured patients, this
results in hyperactive inefficient bladder function.
Treatments for these patients focus on alleviating
hyperactive bladder dysfunction pharmacologically
via non-selective anti-muscarinic agents (Pannek
et al., 2000), reducing dyssynergy surgically
through external sphincterotomy (Reynard et al.,
2003) and generating voiding on demand by sacral
anterior root stimulation (Schumacher et al., 1999).
Spinal cord injury has been shown to induce
neuronal sprouting within the spinal cord. Plastic-
ity of neurons within the micturition reflex circuit
is a possible mechanism by which bladder function
emerges after spinal cord injury. Continued re-
search regarding the role that neuronal sprouting
plays in bladder function after spinal cord injury
will aid in the development of treatment methods
which target the cause of bladder dysfunction after
injury rather than its symptoms. In this chapter,
we will discuss the evidence for neuronal plasticity
occurring in the micturition reflex path and how
this may mediate the emergence of bladder dys-
function after spinal cord injury.
Axonal remodeling occurs normally within the
developing micturition reflex
The micturition reflex is subject to extensive re-
modeling during development. For approximately
the first 3 weeks of life, micturition in rats and cats
is evoked by stimulation of somatic perineal af-
ferents, via licking of the perineum by the mother
(Maggi et al., 1986; Thor et al., 1989). The effe-
rent limb, through the pelvic nerve, stimulates
detrusor contraction, thereby facilitating bladder
emptying. This somatic-bladder reflex is a spinal
reflex that becomes progressively weaker through-
out postnatal life to a point where there appears to
be a switch from spinally mediated to supraspin-
ally mediated micturition (de Groat et al., 1998).
Micturition in these animals, as in adults, is in
response to bladder stretch, having both afferent
and efferent limbs in the pelvic nerve. This switch
is thought to be accompanied by major changes in
neuronal circuits used to elicit micturition.
Developing nervous systems undergo a great
deal of synaptic strengthening and neuronal re-
finement. One process thought to play a major role
during development of the micturition reflex is
synaptic competition. The numbers of synapses on
spinal neurons derived from spinal and supraspinal
sources change throughout development (Fig. 1).
It is thought that synaptic input to the para-
sympathetic preganglionic nucleus from supraspi-
nal centers increases during postnatal development
and out competes sacral interneurons for the same
target (de Groat, 2002). This may explain the loss
of spinally mediated micturition and the emer-
gence of supraspinally mediated micturition dur-
ing development.
Coincidently, spinal cord injury in adult animals
triggers a switch back to micturition mediated via
a spinal reflex (Kakizaki and de Groat, 1997;
Shefchyk and Buss, 1998) thereby making synaptic
competition an hypothesis not only for the emer-
gence of brain stem-mediated micturition during
development, but also for the emergence of spinal
reflex micturition after spinal cord injury.
If the alternative neuronal pathways for eliciting
bladder contraction already exist in the spinal
cord, why does it take so long for a spinal reflex to
emerge after spinal cord injury? Assuming that

Fig. 1. Synaptic competition may explain the switch from spi-
nal to brainstem control of micturition during development. It
is proposed that sacral preganglionic neurons (PGN) in neon-
ates receive input from sacral interneurons (INT) as well as
from descending fibers from the brain stem. In young neonates
a spinal reflex elicits micturition because sacral interneurons
outcompete any descending modulation by the brainstem. As
the neonate ages the number of synapses on preganglionic neu-
rons from sacral interneurons decreases, and the number from
the brain stem increases. The strong synaptic input from
the brainstem is thought to underlie supraspinally controlled
micturition in the older neonate and throughout adulthood.
From de Groat (2002).
spinal shock accounts for bladder areflexia during
the first few days after injury, reflex mediated
micturition should occur soon after resolution of
areflexia. However, micturition does not emerge
in rats until around 2 weeks after spinal cord
149
transection (Yoshiyama et al., 1999) and in humans,
depending on the severity of the injury, it may not
emerge for months (Weld and Dmochowski, 2000).
This delay in emergence of the bladder function
implies that other changes associated with neurons
in the micturition reflex, both phenotypic and
anatomical, may be involved in emergence of
bladder function after spinal cord injury.
Bladder primary afferents and spinal cord injury
General characteristics
Retrograde tracers such as cholera toxin B subunit
and horseradish peroxidase have proven useful for
illustrating the pattern of bladder primary afferent
termination in the spinal cord (Morgan et al.,
1981; Nadelhaft and Booth, 1984; Wang et al.,
1998). Bladder primary afferents enter Lissauer’s
tract from which two major tracts form the path-
ways of entry into the gray matter of the dorsal
horn. The lateral collateral pathway extends from
superficial dorsal horn (lamina I and II) through
the dorsolateral funiculus and terminates densely
in the area of the parasympathetic preganglionic
nucleus. A few terminations are also seen to enter
the dorsal gray commissure from this pathway.
A second and less dense pathway, the medial col-
lateral pathway, sends fibers from the dorsomedial
border of the dorsal horn in laminae I and II
to lamina X. Terminating fibers from this pathway
can also be seen in medial laminae V and VI.
Bladder primary afferents exhibit a periodical
rostrocaudal termination pattern within the
spinal cord.
Bladder primary afferents consist of small un-
myelinated C-fibers and thinly myelinated A-delta
fibers. These subtypes can be divided further into
peptidergic and non-peptidergic groups. Pep-
tidergic fibers are characterized by the presence
of calcitonin gene-related peptide and substance P.
Non-peptidergic primary afferents contain fluoride
resistant acid phosphatase activity and are able to
bind the plant isolectin B4 (IB4) (Stucky and
Lewin, 1999). Regionally, peptidergic afferents
terminate in lamina I as well as lamina II outer,
whereas non-peptidergic afferents terminate in
150
lamina II inner. 86% of all bladder primary affer-
ents innervating the bladder body are peptidergic
fibers (Yoshimura et al., 2003). The greatest pro-
portion of non-peptidergic afferent innervation is
to the distal urethra where close to 30% of sensory
innervation is from IB4 positive fibers.
Further characterization of these two groups
reveals that peptidergic and non-peptidergic fibers
possess different trophic factor receptors and
therefore respond to distinct neurotrophic factors.
Peptidergic primary afferents contain the trkA nerve
growth factor receptor while non-peptidergics
do not contain trk receptors but contain glial
derived neurotrophic factor receptors Ret and
GRFalpha1.
Phenotypic changes
After spinal cord injury, bladder primary afferents
may undergo changes in phenotype as well as a
sprouting response. In the cat, under normal
conditions, bladder sensory information travels to
the spinal cord via mechano-sensitive A-delta
fibers (Janig and Morrison, 1986). After spinal
cord injury, sensory information is conveyed by
what were once mechano-insensitive C-fibers
(de Groat et al., 1990). This increase in C-fiber
afferent excitability may be mediated by a decrease
in tetrodotoxin-resistant and an increase in
tetrodotoxin-sensitive Na+ channel expression
(Yoshimura and de Groat, 1997; Waxman et al.,
1999). A decrease in Nav 1.8 channels, a subtype of
tetrodotoxin-resistant Na+ channels, is associated
with bursting behavior of cerebellar Purkinje
neurons (Renganathan et al., 2003) as well as
spontaneous activity of dorsal root ganglion
cells (Renganathan et al., 2001). Decreased ion
conductance across A-type potassium channels is
also thought to contribute to increased afferent
excitability (Sculptoreanu et al., 2004).
Although overall afferent number remains the
same, there is an increase in proportion of
myelinatated fibers (Yoshimura et al., 1998) as
well as dorsal root ganglion cell body size (Yu
et al., 2003) among bladder primary afferents after
spinal cord injury. Other injury-induced changes
to bladder afferent phenotype include changes in
the density and distribution of trk receptors
thought to play a role in growth-promoting signa-
ling (Qiao and Vizzard, 2002) among L6/S1 dorsal
root ganglion cells.
Sprouting and the factors implicated in this response
As well as phenotypic changes, anatomical reor-
ganization of bladder primary afferents may con-
tribute to emergence of bladder function after
spinal cord injury.
Neurotrophic factors
Neuronal plasticity that follows spinal cord injury
is thought to be regulated largely by neurotrophic
factors. If neuronal plasticity is involved in the
emergence of bladder function after spinal cord
injury, then changes in trophic factor levels at both
organ and spinal levels are likely to be involved.
Nerve growth factor provides trophic support to
the majority of bladder primary afferents and
chemically increases the sensitivity of primary affe-
rents (Lamb et al., 2004). Several studies have in-
vestigated the role that nerve growth factor may
play in bladder function after spinal cord injury.
Increases in spinal nerve growth factor have
been linked to the development of disorders such
as neuropathic pain and autonomic dysreflexia. In
both disorders, nerve growth factor is increased
predominantly within small diameter primary affe-
rent neurons. Neuronal nerve growth factor is in-
creased in lumbar and sacral (L6/S1) dorsal root
ganglia and L6 spinal cord after thoracic spinal
cord injury (Seki et al., 2002). Schwann cells,
astrocytes and other microglia also upregulate
their expression of nerve growth factor after spinal
cord injury (Krenz and Weaver, 2000). In models
of neuropathic pain, expression of nerve growth
factor by glial cells sensitizes primary afferent
nociceptors leading to hyperalgesia and allodynia.
The spinal cord is not the only source of nerve
growth factor. In fact, the bladder has increased
levels of nerve growth factor mRNA acutely after
spinal cord injury. Upregulation of nerve growth
factor protein and increased expression of its re-
ceptor, trkA, also occur in the bladder and dorsal
root ganglia 6 weeks post injury (Vizzard, 2000;
 151

Qiao and Vizzard, 2002), well after the emergence
of voiding function in the rat. Thus bladder-
derived nerve growth factor may facilitate changes
that occur in bladder function at chronic time-
points after spinal cord injury but not the initial
emergence of bladder function. Nerve growth fac-
tor acts on peptidergic C-fibers and has been
shown to induce the expression and secretion of
calcitonin gene-related peptide (Bowles et al.,
2004). Increased calcitonin gene-related peptide
distribution within the L6/S1 spinal cord segments
has been demonstrated in spinal cord injury rats at
timepoints before the emergence of bladder func-
tion (Fig. 2) (N. Zinck, V. Rafuse and J. Downie,
unpublished data). Furthermore, administration
of antibodies against nerve growth factor after
spinal cord injury leads to a decrease in density
and distribution of these peptidergic fibers in the
dorsal horn (Christensen and Hulsebosch, 1997).
Therefore increases in L6/S1 spinal calcitonin
gene-related peptide may be the consequence of
peptidergic fiber sprouting elicited by increased
nerve growth factor. At both acute (3 days)
(N. Zinck, V. Rafuse, J. Downie, unpublished
data) and chronic (6 weeks) (Vizzard and Boyle,
1999) time-points post spinal cord injury growth
associated protein-43 is increased within L6/S1 spi-
nal cord segments, suggesting synaptic remodeling

Fig. 2. Calcitonin gene-related peptide-immunoreactive primary afferents sprout in rat lumbosacral spinal cord after spinal cord
injury. Inset shows the approximate level of the longitudinal sections. Increases in density and distribution of fiber terminations is seen
in rats 8 days post spinal cord transection at T10 when compared to non-injured (control) rats. Arrowhead indicates filling in of gaps
between primary afferent termination bundles. SPN, sacral parasympathetic nucleus; DCG, dorsal commissural nucleus. (N.D.T.
Zinck, V.F. Rafuse, J.W. Downie, unpublished.)
152
is occurring both before and after emergence of
bladder function.
Although previous literature suggested that
non-peptidergic afferents do not sprout in re-
sponse to injury (Belyantseva and Lewin, 1999),
recent studies report increased fiber branching of
IB4 positive neurons within the dorsal root ganglia
after spinal nerve transection (Li and Zhou, 2001).
Bladder mRNA levels of glial-derived neurotro-
phic factor are also increased after spinal cord in-
jury in the rat (Vizzard, 2000) although direct
studies of IB4 positive fiber density or distribution
in L6/S1 spinal cord segments after spinal cord
injury have not been reported. Furthermore, un-
like nerve growth factor, no studies have been
conducted that selectively target non-peptidergic
afferents via application of glial-derived neurotro-
phic factor or antibodies to glial-derived neurotro-
phic factor in lumbosacral spinal cord making it
difficult to determine what role this neurotrophin
plays in the micturition reflex.
Brain-derived neurotrophic factor is a member
of the nerve growth factor family and is also
associated with small to medium diameter
peptidergic primary afferents that terminate in
lamina I and II of the dorsal horn (Luo et al.,
2001). It is upregulated and synthesized in response
to peripheral inflammation (Apfel et al., 1996). In
fact, increases in nerve growth factor have been
linked to increased expression of brain-derived
neurotrophic factor in trkA-containing sensory
afferents and to heightened pain sensitivity associ-
ated with neuropathic pain (Obata et al., 2003).
Increased expression of brain-derived neurotrophic
factor occurs throughout the spinal cord, including
L6/S1, after complete thoracic spinal cord injury.
Upregulation of trk B within bladder-specific
dorsal root ganglion neurons also occurs after
spinal cord injury (Qiao and Vizzard, 2002).
Unlike other neurotrophic factors studied,
increased spinal brain-derived neurotrophic factor
expression occurs both acutely and chronically
after spinal cord injury. In fact, in the lumbosacral
cord, brain-derived neurotrophic factor upregula-
tion precedes that of nerve growth factor (Zvarova
et al., 2004), implying that in spinal cord injury,
the increased brain-derived neurotrophic factor
expression in primary afferents does not require
the presence of nerve growth factor. Because nerve
growth factor-stimulated brain-derived neurotro-
phic factor upregulation is associated with inflam-
matory processes in the peripheral nervous system,
central inflammatory processes after spinal cord
injury may not be contributing to increases in brain-
derived neurotrophic factor at the spinal level.
Cellular adhesion molecules
A critical aspect of neuronal outgrowth is appro-
priate synapse formation, facilitated largely by
changes in adhesivity of growth cones. Several
factors have been implicated in mediating adhe-
sion of individual axon terminals to neighboring
cells or to the extracellular matrix. Three major
players in the cell adhesion molecule family are
N-cadherin, the Ig cell adhesion molecules, L1 and
neural cell adhesion molecule (NCAM) (Kiryushko
et al., 2004). Homophilic binding of these members
is necessary for proper adhesion and synapse
formation throughout the development of the
central nervous system.
NCAM levels are greatly upregulated at critical
timepoints in neuronal development (Bruses et al.,
2002). Alternative splicing of the same gene results
in three different isoforms of NCAM that have
been implicated in the creation of neuronal
networks during development and after injury in
the adult nervous system (Kiss and Muller, 2001).
Named by their molecular weights, NCAM-180
and -140 span the cell membrane whereas NCAM-
120 has only an extracellular domain. NCAM can
be separated from other members of the cell
adhesion molecule family by their ability to bind
homopolymers of the carbohydrate polysialic acid.
Polysialic acid-bound NCAM facilitates a decrease
in homophilic NCAM binding and thus a decrease
in cellular adhesion and an increase in axonal
defasciculation (Tang and Landmesser, 1993;
Monnier et al., 2001). Polysialic acid has been
shown to increase after central (Bonfanti et al.,
1996) and peripheral nerve (Franz et al., 2005)
injury. Increases in polysialic acid have been asso-
ciated with regeneration of injured neurons of the
hippocampus (Aubert et al., 1998). During the first
few weeks after birth neonatal rats use a spinal
pathway to elicit micturition. Approximately three
weeks after birth bladder emptying is controlled by

supraspinal neurons. Polysialic acid-NCAM levels
are elevated in L6/S1 spinal cord segments of
postnatal day 6 (P6) rats (Fig. 3) (N. Zinck, V.
Rafuse, J. Downie, unpublished data) when com-
pared to adult control, as suggested by the level of
smearing in lane one (P6) compared to lane two
(control). This is a time when rat pups still use a
spinal reflex to elicit micturition implying that po-
lysialic acid may have a role in the plasticity that
accompanies the transition from spinal to supra-
spinal micturition in rats.
NCAM has recently been shown to interact with
brain-derived neurotrophic factor and glial-de-
rived neurotrophic factor. NCAM, depending on
the isoform, signals through two main pathways,
the Fyn/FAK pathway or through the fibroblast
growth factor receptor, both of which converge
upstream of the mitogen-activated protein kinase
kinase MEK (Kiryushko et al., 2004). Signaling
via the fibroblast growth factor receptor also in-
duces release of intracellular calcium stores, which
has been shown to mediate axonal outgrowth by a
variety of mechanisms, including by growth asso-
ciated protein-43 dependent phosphorylation and
Fig. 3. Neural cell adhesion molecule (NCAM) in neonatal
spinal cord is highly polysialated. Illustration shows a western
blot of NCAM isoforms in rat L6/S1 spinal cord. Arrowheads
point out the level of three NCAM isoforms (NCAM-120,
NCAM-140, NCAM-180). The level of polysialation can be
inferred by the smearing of the NCAM bands due to variable
polysialic acid (PSA) binding to NCAM. At post-natal day 6
(P6) rats contain all three isoforms as well as showing smearing
between these bands. Adult rats (control) have weak to no ex-
pression of NCAM-180 and little PSA expression as indicated
by a lack of smearing between bands. (N.D.T. Zinck, V.F.
Rafuse, J.W. Downie, unpublished.)
153
activation. NCAM is not only unique in its ability
to bind polysialic acid, but has been shown to act
as a co-receptor with GFRa1, a glycosylpho-
sphatidylinositol-linked receptor for glial-derived
neurotrophic factor (Zhou et al., 2003). Glial-de-
rived neurotrophic factor binding to this receptor
complex has been associated with stimulating ax-
onal growth in vitro (Paratcha et al., 2003). Inter-
action of brain-derived neurotrophic factor with
its receptor trkB may be, in part, mediated by the
presence of polysialic acid-NCAM, as selective
enzymatic removal of polysialic acid decreases
trkB phosphorylation (Vutskits et al., 2001). Be-
cause both polysialic acid and NCAM have been
shown to interact with neurotrophic factors that
are upregulated in lumbosacral spinal cord after
spinal cord injury (Zvarova et al., 2004), polysialic
acid-NCAM may play a significant role in the
emergence of the spinal micturition reflex.
Inhibitors of neuronal outgrowth after injury
The central nervous system contains several
known inhibitors of neuronal sprouting. These
molecules are expressed to stop neuronal out-
growth and aberrant synapse formation at the end
of development and are often upregulated after
nerve injury. Injury to the central nervous system
causes the formation of a glial scar that over ex-
presses both myelin-dependent and -independent
inhibitors of neuronal extension thereby impeding
neuronal sprouting and regeneration. Myelin-de-
pendent inhibitors of sprouting include Nogo-A,
myelin associated glycoprotein and oligodendro-
cyte myelin glycoprotein (Grados-Munro and
Fournier, 2003) whereas myelin-independent inhi-
bitors of sprouting are members of the proteo-
glycan family including chondroitin sulfate
proteoglycans (Bovolenta and Fernaud-Espinosa,
2000). Important signaling pathways of key mol-
ecules that inhibit sprouting will be discussed in
the section ‘‘Manipulation of neuronal sprouting’’.
Do interneurons play a role in bladder function after
spinal cord injury?
There are many neurons in the lumbosacral spinal
cord that respond to bladder afferent stimulation
154
(McMahon and Morrison, 1982; Honda, 1985;
Coonan and Downie, 1999). Some may participate
in ascending transmission of bladder-related ac-
tivity, either as part of the micturition reflex path-
way or in pain-related pathways (McMahon and
Morrison, 1982; Milne et al., 1982; Ding et al.,
1994). Also, activation of spinal interneurons is an
important component of efficient micturition both
for bladder contraction and sphincter inhibition
(Shefchyk, 2001). These interneuron pools have
not been well localized. However, it appears from
virus tracing and immediate early gene expression
studies that the dorsal gray commissure and the
region of the parasympathetic preganglionic nu-
cleus are important locations of bladder and
sphincter-related interneurons (Nadelhaft et al.,
1992; Nadelhaft and Vera, 1996; Marson, 1997;
Grill et al., 1998; Vera and Nadelhaft, 2000).
There is some suggestion of synaptic reorgani-
zation in the lumbosacral spinal cord after spinal
cord injury in rats (Yu et al., 2003). On the other
hand, evidence for change in the pelvic afferent
terminal arbor in the spinal cord is lacking (Kruse
et al., 1995). There is evidence for reorganization
of motor neuronal pathways after spinal cord in-
jury to facilitate hind limb locomotion (Grasso
et al., 2004). Thus one possibility is that spinal
interneuron reorganization may underlie the emer-
gence of bladder activity after spinal cord injury.
A second possibility is that emergence of blad-
der function after spinal cord injury is a matter of
developing access to an existing spinal circuit sub-
serving micturition. The existence of such a circuit,
analogous to the spinal pattern generator for lo-
comotion, is implied by the finding that micturit-
ion can be evoked by perineal or urethral nerve
stimulation in some circumstances (Shefchyk and
Buss, 1998; Boggs et al., 2004). One problem with
this circuit is that it appears not to be activated by
bladder distension and thus coordinated micturit-
ion never emerges in spinal cord-injured cats. It is
possible that sprouting of urethral or perineal af-
ferents is hindering the use of this circuit at later
time points post spinal cord injury. Thus encour-
aging the sprouting of bladder afferents to target
the spinal micturition circuit while suppressing the
perineal inputs to the circuit may be an appropri-
ate approach to restoration of function.
It is likely that spinal interneurons undergo
physiological and structural changes after spinal
cord injury and that these changes are contributing
to emergence of micturition after injury. However,
until precise methods are developed to study these
interneurons specifically, the contribution of spi-
nal bladder interneurons to bladder function after
spinal cord injury will remain elusive.
Repairing the injured spinal cord to improve
bladder function
Cellular implants
Stem cells have been touted as a major aid in the
treatment of many diseases, including Parkinson’s
disease, diabetes and amyotrophic lateral sclerosis.
Because stem cells have the ability to differentiate
into various tissue types, they also have been tested
for a role in repairing the damaged spinal cord af-
ter injury. Stem cells injected into severely injured
rat spinal cord have shown cellular differentiation,
resulting in increased axonal regeneration and sig-
nificant improvement in motor function below the
site of injury (McDonald et al., 1999). Few studies,
however, have addressed the impact of stem cell
implantation on visceral function after spinal cord
injury. Injections of neural stem cells in the injured
rat spinal cord has improved lower urinary tract
function by increasing voiding efficiency, although
has not improved bladder–sphincter dyssynergia
(Mitsui et al., 2003).
Implantation of nerve grafts with cells genetically
engineered to secrete growth factors has become a
common animal model to treat spinal cord injury.
The ability of the peripheral nervous system to
regenerate after nerve injury is far greater than that
of the CNS. Because of this ability of peripheral
nerves to regenerate, nerve grafts implanted in the
injured spinal cord are often made from peripheral
nerve tissue or peripheral nervous system specific
cell types. For example, implantation of a Schwann
cell graft, secreting trkB activating neurotrophins,
brain-derived neurotrophic factor and neurotro-
phin-3, into a severely contused rat spinal cord has
provided restoration of bladder function (Sakamoto
et al., 2002). Not every growth factor will have
positive effects on bladder function. For example,

implantation of peripheral nerve grafts secreting
acidic fibroblast growth factor in a patient with a
complete hemisection of the thoracic spinal cord
improved motor recovery although bladder function
was unchanged (Cheng et al., 2004). Because differ-
ent motor and visceral systems are reliant on specific
growth factors, it is likely that if we are to signif-
icantly improve micturition by the addition of
neuronal growth factors, then we need first to
establish which trophic factors are pertinent to the
micturition reflex.
Olfactory ensheathing cells, as the name implies,
ensheath fibers as they travel from the olfactory
bulb to the peripheral nervous system. These cells
are thought to provide a permissive environment
from the CNS to the peripheral nervous system by
remaining in contact with the olfactory nerve and
inhibiting astrocytes from blocking entry into the
CNS. Injection of olfactory ensheathing cells into
deep dorsal horn regions after transection of dorsal
roots promotes afferent reentry into the dorsal spi-
nal cord and also facilitates the emergence of blad-
der function (Pascual et al., 2002). Although
deafferentation is not a model of spinal cord inju-
ry, these findings provide justification for further
investigation on the effects that olfactory ensheath-
ing cells may have on spinally mediated bladder
function. Recently, many investigators studying
bladder incontinence are turning to gene therapy
as a possible way to treat urological dysfunction.
This technique involves the use of a viral vector that
encodes a particular gene of interest and is injected
into the bladder wall. Transport of the vector to the
bladder afferent neurons in dorsal root ganglia as
well as to the spinal cord is accomplished due to
retrograde virion transmission. This technique has
previously been used to transport molecules like
nerve growth factor as well as pre-proenkephalin to
bladder primary afferents to treat bladder afferent
neuropathy associated with diabetes (Goins et al.,
2001; Yoshimura et al., 2001). This may also be an
interesting avenue of research for treating bladder
dysfunction after spinal cord injury.
Manipulation of neuronal sprouting
Although it has been implied that neurons con-
trolling the micturition reflex sprout after spinal
155
cord injury, no definitive studies have been con-
ducted. The last decade has provided a wealth of
knowledge with respect to specific extracellular
signaling pathways involved in neurite outgrowth
and extension. With a molecular knowledge of key
elements responsible for axon regeneration, new
targets have been uncovered for experimental ma-
nipulation. The function of the lower urinary tract
could benefit from such investigation, providing
new avenues of research that have the potential to
answer questions regarding the neuronal plasticity
of the micturition reflex after spinal cord injury.
Neurotrophic factors, especially nerve growth fac-
tor, have been shown to play a major role in bladder
function after spinal cord injury, thereby making
possible methods for interfering with signaling path-
ways of these molecules invaluable. Intrathecal ad-
ministration of immuno-neutralizing nerve growth
factor antibody into the L6/S1 spinal cord segments
results in a marked decrease in the number of non-
voiding bladder contractions, as well as increasing
voiding efficiency in rats with complete spinal cord
injury (Fig. 4). Electromyographic recording of
external urethral sphincter muscles in spinal rats
treated with this antibody also shows a decrease in
detrusor–sphincter dyssynergia (Seki et al., 2004).
These results are very similar to the effects of
subcutaneous administration of capsaicin to spinal
cord-injured rats. Because capsaicin is selectively
neurotoxic to C-fibers, upregulation of nerve growth
factor is likely to occur in small diameter primary
afferents (Cheng and de Groat, 2004).
Saporin-tagged IB4 injected intrathecally at the
level of L6/S1 in a recent experiment caused selec-
tive reduction of IB4-positive (non-peptidergic) af-
ferents in L6 and a decrease in bladder overactivity
in response to inflammation (Vulchanova et al.,
2001; Nishiguchi et al., 2004). It was not clear
which visceral population of IB4-positive afferents
were destroyed after application of the cytotoxin
saporin in this experiment, so the role of bladder
specific non-peptidergic fibers in detrusor overacti-
vity after inflammation is unclear. Injections of
saporin-tagged IB4 in the bladder wall or pelvic
nerve would aid in clarifying these results. Recent-
ly, saporin-tagged IB4 that had been injected into
the sciatic nerve was detected in the spinal cord
indicating that these more selective methods are
156

Fig. 4. Antagonism of nerve growth factor (NGF) normalizes aberrant bladder function after spinal cord injury. Bladder function is
abnormal 10 days after a mid-thoracic spinal cord injury and the administration of vehicle i.t. (A). However, 14 days after i.t.
administration of 10 ug NGF-Ab the non-voiding contractions are fewer and smaller (B). Arrows indicate voiding episodes. The
micturition pattern in panel B closely resembles that seen in uninjured rats. Infusion of saline ¼ 0.04 ml/min. From Seki et al. (2002)
with permission.

possible (Vulchanova et al., 2001; Tarpley et al.,
2004). The role that IB4 positive bladder afferents
play in bladder overactivity associated with spinal
cord injury, however, has not yet been investigated.
As described earlier, the peripheral nervous
system has a far greater ability to regenerate after
neuronal assault than the CNS. The most studied
of all myelin-associated inhibitors is Nogo. The
receptor for Nogo, NgR, forms a receptor complex
with the low-affinity neurotrophin receptor p75.
Myelin-associated glycoprotein and oligodendro-
cyte myelin glycoprotein, along with Nogo,
mediate their inhibitory actions on neuronal out-
growth through the NgR-p75 receptor complex
(Wang et al., 2002). The antagonist to the Nogo
receptor, NEP (1–40), and exogenous addition of
neurotrophins (Cai et al., 1999) have both signi-
ficantly improved axon re-growth after nerve injury
(GrandPre et al., 2002). p75 receptor knockouts
have shown similar results (Wang et al., 2002). An
overview of signaling pathways that are critical in
regulating neuronal extension is presented in Fig. 5.
During development, myelin-associated glyco-
protein appears to play a key role in the switch
from neurite extension to arrest. Cyclic adenosine
monophosphate (cAMP) levels appear to be critical
in this switch. During development, cAMP levels
are elevated to a point at which myelin-associated
glycoprotein signaling is overridden. When cAMP
levels decline, myelin-associated glycoprotein is
then able to promote growth cone collapse, ending
axonal outgrowth during development. Exogenous
elevation of cAMP (Neumann et al., 2002) or
vaccination against myelin (Huang et al., 1999) in
models of spinal cord injury has improved neuronal
regenerative ability as well as increased functional
recovery. In addition, inhibiting protein kinase A, a
downstream regulator of cAMP, decreases the
ability of peripheral nerve grafts to extend axons
into host spinal cord (Cai et al., 2001).
Signaling through NgR-p75 activates other
downstream effectors important in signaling inhi-
bition of neuronal outgrowth. A small group of
GTPases, known as Rho GTPases, are activated in
response to Nogo- and myelin-associated glyco-
protein binding to the p75 receptor. It is becoming
increasingly evident that the low-affinity receptor
for all neurotrophins, p75, has a major role in reg-
ulation of neuronal extension and survival. Con-
version of RhoGDP to its active GTP-bound state is
thought to be mediated by a number of GTPase-
activating proteins, guanine nucleotide exchange
factors, and guanine nucleotide dissociation inhib-
itor (GDI). Rho-GDI interacts directly with
 157

Fig. 5. Major signaling pathways of myelin associated inhibitors of neuronal outgrowth. Myelin associated glycoprotein (MAG),
Nogo and oligodendrocyte myelin glycoprotein signal through the Nogo receptor (NgR)-p75 receptor complex to activate Rho and
Rho-kinase. Neurotrophic factors (NT) stimulate neuronal regeneration through increases in cAMP levels. MAG signaling is able to
inhibit cAMP induced regeneration via signaling through an inhibitory G protein. Arrest of axonal remodeling during development is
likely mediated by MAG signaling. From Grados-Munro and Fournier (2003) with permission.

the intracellular domain of the p75 receptor
(Yamashita and Tohyama, 2003) and thus is
thought to play a key role in Rho signaling. In its
active state, RhoGTP acts to stabilize actin polym-
erization, thereby facilitating growth cone collapse
and inhibiting axonal sprouting through activation
of several effector molecules (Bishop and Hall,
2000). One of the most studied Rho effector mol-
ecules involved in actin reorganization is a serine
threonine kinase known as Rho-kinase. Rho-
kinase regulates axonal outgrowth via two main
pathways. Phosphorylation of myosin light chain
as well as the myosin binding subunit of myosin
light chain phosphatase induces actomyosin as-
sembly and growth cone collapse. Inhibition of
Rho-kinase signaling with the specific Rho-kinase
inhibitor, Y–27632 results in increased axonal ex-
tension and growth cone formation within dorsal
root ganglia neurons (Borisoff et al., 2003).
Most research directed at restoring function
after spinal cord injury seeks to enhance survival
and extension of injured and uninjured neurons.
Although neuronal sprouting and synapse creation
is essential for reestablishing damaged connections
158
between neurons, it is critical that new synapses
are functionally relevant. Increasing synaptogene-
sis has the potential to do harm as well as good as
seen in neuropathic pain and autonomic dysre-
flexia. Upregulation of nerve growth factor has
been linked to bladder hyperactivity after spinal
cord injury (Seki et al., 2002). Thus, if plasticity is
mediating emergence of micturition after spinal
cord injury, then perhaps the goal for intervention
in this process should be to maximize functional
outcome by selectively targeting sprouting to pro-
duce efficient micturiton and avoid unwanted side
effects. Several molecules in the CNS (e.g. netrin,
semaphorins and slits) are implicated in mecha-
nisms in guidance of growth cones to appropriate
targets and formation of appropriate synapses (for
review see Dickson and Senti, 2002) However,
specific targeting of sprouting neurons to produce
efficient, non-dyssynergic function after spinal
cord injury is currently not feasible because the
intricate molecular interactions involved are not
well understood. The most practical strategy, for
the present, may be to reduce neuronal sprouting
in lumbosacral spinal cord in an attempt to alle-
viate bladder hyperreflexia.
Current technology has provided ways to ma-
nipulate the level of polysialic acid-neural cell ad-
hesion molecule thereby creating possible methods
to explore sprouting responses of bladder neurons
after spinal cord injury. NCAM knockout mice
have a nervous system completely stripped of
NCAM and polysialic acid (Cremer et al., 1994),
thereby having severe consequences in synapto-
genesis-driven processes like long term potentiat-
ion (Muller et al., 2000; Bukalo et al., 2004). In
addition, a specific enzyme endoneuraminidase has
been used to cleave the polysialic acid moiety from
NCAM. In vivo injections of endoneuraminidase
delays collateral sprouting and decreases branching
of corticospinal axons in rat pups (Daston et al.,
1996), as well as decreases axonal plasticity in the
arcuate nucleus of adult rats (Hoyk et al., 2001).
Can neuronal plasticity play a part in restoring
bladder function after spinal cord injury?
This chapter has outlined several plastic changes
that neurons involved in the micturition reflex may
undergo. Some of these changes may play a role in
the emergence of bladder function after spinal cord
injury. However, uncontrolled plasticity does not
appear to provide functional outcomes that are
favorable to a spinal cord-injured person. Great
advances have been made in unraveling the mo-
lecular mechanisms of neuronal plasticity. These
findings have already contributed to a better
understanding of how bladder afferents respond
to spinal cord injury and how bladder function
might be dictated by that response. Because
bladder activity after spinal cord injury is
dysfunctional, specifically targeting bladder affer-
ent outgrowth may serve as a potential therapy for
restoring ‘‘normal’’ bladder voiding and conti-
nence. Although this presents a difficult task,
research will be facilitated by the plethora of new
molecular techniques and the creation of knockout
and transgenic animals. In conclusion, plasticity of
bladder primary afferents after spinal cord injury is
associated with bladder dysfunction, and therefore
this dysfunction may be considered to be an
unwanted consequence of plasticity. We suggest
that, with more research, uncontrolled plasticity
may be directed toward the more positive outcome
of normal bladder function after spinal cord injury.
References
Apfel, S.C., Wright, D.E., Wiideman, A.M., Dormia, C., Snider,
W.D. and Kessler, J.A. (1996) Nerve growth factor regulates
the expression of brain-derived neurotrophic factor mRNA in
the peripheral nervous system. Mol. Cell. Neurosci., 7: 134–142.
Aubert, J., Ridet, J., Schachner, M., Rougon, G. and Gage, F.
(1998) Expression of L1 and PSA during sprouting and re-
generation in adult hippocampal formation. J. Comp. Ne-
urol., 399: 1–19.
Barrington, F. (1915) The nervous mechanism of micturition.
Quart. J. Exp. Physiol., 8: 33–71.
Belyantseva, I.A. and Lewin, G.R. (1999) Stability and plasticity
of primary afferent projections following nerve regeneration
and central degeneration. Eur. J. Neurosci., 11: 457–468.
Bishop, A.L. and Hall, A. (2000) Rho GTPases and their ef-
fector proteins. Biochem. J., 348(Part 2): 241–255.
Boggs, J.W., Wenzel, B.J., Gustafson, K.J., and Grill, W.M.
(2004) A spinal micturition reflex mediated by afferents in the
deep perineal nerve. J. Neurophysiol., in press.
Bonfanti, L., Merighi, A. and Theodosis, D.T. (1996) Dorsal
rhizotomy induces transient expression of the highly sialylated
isoform of the neural cell adhesion molecule in neurons and
astrocytes of the adult rat spinal cord. Neuroscience, 74:
619–623.

Borisoff, J.F., Chan, C.C., Hiebert, G.W., Oschipok, L.,
Robertson, G.S., Zamboni, R., Steeves, J.D. and Tetzlaff,
W. (2003) Suppression of Rho-kinase activity promotes
axonal growth on inhibitory CNS substrates. Mol. Cell.
Neurosci., 22: 405–416.
Bovolenta, P. and Fernaud-Espinosa, I. (2000) Nervous system
proteoglycans as modulators of neurite outgrowth. Prog.
Neurobiol., 61: 113–132.
Bowles, W.R., Sabino, M., Harding-Rose, C. and Hargreaves,
K.M. (2004) Nerve growth factor treatment enhances release
of immunoreactive calcitonin gene-related peptide but not
substance P from spinal dorsal horn slices in rats. Neurosci.
Lett., 363: 239–242.
Bruses, J.L., Chauvet, N., Rubio, M.E. and Rutishauser, U.
(2002) Polysialic acid and the formation of oculomotor
synapses on chick ciliary neurons. J. Comp. Neurol., 446:
244–256.
Bukalo, O., Fentrop, N., Lee, A.Y., Salmen, B., Law, J.W.,
Wotjak, C.T., Schweizer, M., Dityatev, A. and Schachner,
M. (2004) Conditional ablation of the neural cell adhesion
molecule reduces precision of spatial learning, long-term po-
tentiation, and depression in the CA1 subfield of mouse hip-
pocampus. J. Neurosci., 24: 1565–1577.
Cai, D., Qiu, J., Cao, Z., McAtee, M., Bregman, B.S. and
Filbin, M.T. (2001) Neuronal cyclic AMP controls the de-
velopmental loss in ability of axons to regenerate. J. Neuro-
sci., 21: 4731–4739.
Cai, D., Shen, Y., De Bellard, M., Tang, S. and Filbin, M.T.
(1999) Prior exposure to neurotrophins blocks inhibition of
axonal regeneration by MAG and myelin via a cAMP-de-
pendent mechanism. Neuron, 22: 89–101.
Cheng, C.L. and de Groat, W.C. (2004) The role of capsaicin-
sensitive afferent fibers in the lower urinary tract dysfunction
induced by chronic spinal cord injury in rats. Exp. Neurol.,
187: 445–454.
Cheng, H., Liao, K.K., Liao, S.F., Chuang, T.Y. and
Shih, Y.H. (2004) Spinal cord repair with acidic fibroblast
growth factor as a treatment for a patient with chronic par-
aplegia. Spine, 29: E284–E288.
Christensen, M.D. and Hulsebosch, C.E. (1997) Spinal cord
injury and anti-NGF treatment results in changes in CGRP
density and distribution in the dorsal horn in the rat. Exp.
Neurol., 147: 463–475.
Coonan, E.M., Downie, J.W. and Du, H.-J. (1999) Sacral spi-
nal cord neurons responsive to bladder pelvic and perineal
inputs in cats. Neurosci. Lett., 260: 137–140.
Cremer, H., Lange, R., Christoph, A., Plomann, M., Vopper,
G., Roes, J., Brown, R., Baldwin, S., Kraemer, P. and Scheff,
S. (1994) Inactivation of the N-CAM gene in mice results in
size reduction of the olfactory bulb and deficits in spatial
learning. Nature, 367: 455–459.
Crowley, C., Spencer, S.D., Nishimura, M.C., Chen, K.S.,
Pitts-Meek, S., Armanini, M.P., Ling, L.H., MacMahon,
S.B., Shelton, D.L. and Levinson, A.D. (1994) Mice lacking
nerve growth factor display perinatal loss of sensory and
sympathetic neurons yet develop basal forebrain cholinergic
neurons. Cell, 76: 1001–1011.
159
Daston, M.M., Bastmeyer, M., Rutishauser, U. and O’Leary,
D.D. (1996) Spatially restricted increase in polysialic acid
enhances corticospinal axon branching related to target rec-
ognition and innervation. J. Neurosci., 16: 5488–5497.
de Groat, W. (2002) Plasticity of bladder reflex pathways dur-
ing postnatal development. Physiol. Behav., 77: 689–692.
de Groat, W., Kawatani, M., Hisamitsu, T., Cheng, C., Ma, C.,
Thor, K., Steers, W. and Roppolo, J. (1990) Mechanisms
underlying the recovery of urinary bladder function following
spinal cord injury. J. Auton. Nerv. Sys., 30: S71–S78.
de Groat, W.C., Araki, I., Vizzard, M.A., Yoshiyama, M.,
Yoshimura, N., Sugaya, K., Tai, C. and Roppolo, J.R. (1998)
Developmental and injury induced plasticity in the micturit-
ion reflex pathway. Behav. Brain Res., 92: 127–140.
Dickson, B.J. and Senti, K.A. (2002) Axon guidance: growth
cones make an unexpected turn. Curr. Biol., 12: R218–R220.
Ding, Y.-Q., Qin, B.-Z., Li, J.-S. and Mizuno, N. (1994) In-
duction of c-fos-like protein in the spinoparabrachial tract-
neurons locating within the sacral parasympathetic nucleus in
the rat. Brain Res., 659: 283–286.
Drescher, U., Faissner, A., Klein, R., Rathjen, F.G. and
Sturmer, C. (1997) Axonal growth and pathfinding: from
phenomena to molecules. Cell. Tissue Res., 290: 187–188.
Franz, C., Rutishauser, U. and Rafuse, V. (2005) Polysialyated
neural cell adhesion molecule (PSA-NCAM) is necessary for
selective targeting of regenerating motor neurons. J. Neuro-
sci., 25: 2081–2091.
Gallo, G. and Letourneau, P.C. (2004) Regulation of growth cone
actin filaments by guidance cues. J. Neurobiol., 58: 92–102.
Goins, W.F., Yoshimura, N., Phelan, M.W., Yokoyama, T.,
Fraser, M.O., Ozawa, H., Bennett, N.J., de Groat, W.C.,
Glorioso, J.C. and Chancellor, M.B. (2001) Herpes simplex
virus mediated nerve growth factor expression in bladder and
afferent neurons: potential treatment for diabetic bladder
dysfunction. J. Urol., 165: 1748–1754.
Gordon-Weeks, P.R. (2004) Microtubules and growth cone
function. J. Neurobiol., 58: 70–83.
Grados-Munro, E.M. and Fournier, A.E. (2003) Myelin-asso-
ciated inhibitors of axon regeneration. J. Neurosci. Res., 74:
479–485.
GrandPre, T., Li, S. and Strittmatter, S.M. (2002) Nogo-66
receptor antagonist peptide promotes axonal regeneration.
Nature, 417: 547–551.
Grasso, R., Ivanenko, Y.P., Zago, M., Molinari, M.,
Scivoletto, G., Castellano, V., Macellari, V. and Lacquaniti,
F. (2004) Distributed plasticity of locomotor pattern gener-
ators in spinal cord injured patients. Brain, 127: 1019–1034.
Grill, W., Wang, B., Hadziefendie, S. and Haxhiu, M. (1998)
Identification of the spinal neural network involved in coordi-
nation of micturition in the male cat. Brain Res., 796: 150–160.
Honda, C.N. (1985) Visceral and somatic afferent convergence
onto neurons near the central canal in the sacral spinal cord
of the cat. J. Neurophysiol., 53: 1059–1078.
Hoyk, Z., Parducz, A. and Theodosis, D.T. (2001) The highly
sialylated isoform of the neural cell adhesion molecule is re-
quired for estradiol-induced morphological synaptic plasticity
in the adult arcuate nucleus. Eur. J. Neurosci., 13: 649–656.
160
Huang, D.W., McKerracher, L., Braun, P.E. and David, S. (1999)
A therapeutic vaccine approach to stimulate axon regeneration
in the adult mammalian spinal cord. Neuron, 24: 639–647.
Janig, W. and Morrison, J.F. (1986) Functional properties of
spinal visceral afferents supplying abdominal and pelvic or-
gans, with special emphasis on visceral nociception. Prog.
Brain Res., 67: 87–114.
Kakizaki, H. and de Groat, W.C. (1997) Reorganization of
somato-urethral reflexes following spinal cord injury in the
rat. J. Urol., 158: 1562–1567.
Kiryushko, D., Berezin, V. and Bock, E. (2004) Regulators of
neurite outgrowth: role of cell adhesion molecules. Ann. NY
Acad. Sci., 1014: 140–154.
Kiss, J.Z. and Muller, D. (2001) Contribution of the neural cell
adhesion molecule to neuronal and synaptic plasticity. Rev.
Neurosci., 12: 297–310.
Krenz, N.R., Meakin, S.O., Krassioukov, A.V. and Weaver,
L.C. (1999) Neutralizing intraspinal nerve growth factor
blocks autonomic dysreflexia caused by spinal cord injury.
J. Neurosci., 19: 7405–7414.
Krenz, N.R. and Weaver, L.C. (2000) Nerve growth factor in
glia and inflammatory cells of the injured rat spinal cord.
J. Neurochem., 74: 730–739.
Kruse, M.N., Belton, A.L. and de Groat, W.C. (1993) Changes
in bladder and externalurethral sphincter function after spi-
nal cord injury in the rat. Am. J. Physiol., 264: R1157–R1163.
Kruse, M.N., Bray, L.A. and de Groat, W.C. (1995) Influence
of spinal cord injury on the morphology of bladder afferent
and efferent neurons. J. Auton. Nerv. Syst., 54: 215–224.
Lamb, K., Gebhart, G.F. and Bielefeldt, K. (2004) Increased
nerve growth factor expression triggers bladder overactivity.
J. Pain., 5: 150–156.
Li, L. and Zhou, X.F. (2001) Pericellular Griffonia simplicifolia
I isolectin B4-binding ring structures in the dorsal root gan-
glia following peripheral nerve injury in rats. J. Comp.
Neurol., 439: 259–274.
Luo, X.G., Rush, R.A. and Zhou, X.F. (2001) Ultrastructural
localization of brain-derived neurotrophic factor in rat pri-
mary sensory neurons. Neurosci. Res., 39: 377–384.
Maggi, C.A., Santicioli, P. and Meli, A. (1986) Postnatal de-
velopment of micturition reflex in rats. Am. J. Physiol., 250:
R926–R931.
Maier, D.L., Mani, S., Donovan, S.L., Soppet, D., Tessarollo,
L., McCasland, J.S. and Meiri, K.F. (1999) Disrupted cor-
tical map and absence of cortical barrels in growth-associated
protein (GAP)-43 knockout mice. Proc. Natl. Acad. Sci.
USA, 96: 9397–9402.
Marson, L. (1997) Identification of central nervous system neu-
rons that innervate the bladder body, bladder base, or ex-
ternal urethral sphincter of female rats: a transneuronal
tracing study using pseudorabies virus. J. Comp. Neurol.,
389: 584–602.
McDonald, J.W., Liu, X.Z., Qu, Y., Liu, S., Mickey, S.K.,
Turetsky, D., Gottlieb, D.I. and Choi, D.W. (1999)
Transplanted embryonic stem cells survive, differentiate and
promote recovery in injured rat spinal cord. Nat. Med., 5:
1410–1412.
McMahon, S. and Morrison, J. (1982) Two groups of spinal
interneurons that respond to stimulation of the viscera of the
cat. J. Physiol., 322: 21–34.
McMahon, S.B. and Morrison, J.F.B. (1982) Spinal neurones
with long projections activated from the abdominal viscera of
the cat. J. Physiol., 322: 1–20.
Milne, R.J., Foreman, R.D. and Willis, W.D. (1982) Responses
of primate spinothalamic neurons located in the sacral
intermediomedial gray (Stilling’s nucleus) to proprioceptive
input from the tail. Brain Res., 234: 227–236.
Mitsui, T., Kakizaki, H., Tanaka, H., Shibata, T., Matsuoka, I.
and Koyanagi, T. (2003) Immortalized neural stem cells
transplanted into the injured spinal cord promote recovery of
voiding function in the rat. J. Urol., 170: 1421–1425.
Monnier, P.P., Beck, S.G., Bolz, J. and Henke-Fahle, S. (2001)
The polysialic acid moiety of the neural cell adhesion
molecule is involved in intraretinal guidance of retinal
ganglion cell axons. Dev. Biol., 229: 1–14.
Morgan, C., Nadelhaft, I. and de Groat, W.C. (1981) The dis-
tribution of visceral primary afferents from the pelvic nerve
to Lissauer’s tract and the spinal gray matter and its rela-
tionship to the sacral parasympathetic nucleus. J. Comp.
Neurol., 201: 415–440.
Muller, D., Djebbara-Hannas, Z., Jourdain, P., Vutskits, L.,
Durbec, P., Rougon, G. and Kiss, J.Z. (2000) Brain-derived
neurotrophic factor restores long-term potentiation in poly-
sialic acid-neural cell adhesion molecule-deficient hippocam-
pus. Proc. Natl. Acad. Sci. USA, 97: 4315–4320.
Nadelhaft, I. and Booth, A.M. (1984) The location and
morphology of preganglionic neurons and the distribution
of visceral afferents from the rat pelvic nerve: a horseradish
peroxidase study. J. Comp. Neurol., 226: 238–245.
Nadelhaft, I. and Vera, P.L. (1996) Neurons in the rat brain and
spinal cord labeled after pseudorabies virus injected into the
external urethral sphincter. J. Comp. Neurol., 375: 502–517.
Nadelhaft, I., Vera, P.L., Card, J.P. and Miselis, R.R. (1992)
Central nervous system neurons labelled following the injec-
tion of pseudorabies virus into the rat urinary bladder.
Neurosci. Lett., 143: 271–274.
Neumann, S., Bradke, F., Tessier-Lavigne, M. and Basbaum,
A.I. (2002) Regeneration of sensory axons within the injured
spinal cord induced by intraganglionic cAMP elevation.
Neuron, 34: 885–893.
Nishiguchi, J., Sasaki, K., Seki, S., Chancellor, M.B., Erickson,
K.A., de Groat, W.C., Kumon, H. and Yoshimura, N. (2004)
Effects of isolectin B4-conjugated saporin, a targeting cyto-
toxin, on bladder overactivity induced by bladder irritation.
Eur. J. Neurosci., 20: 474–482.
Obata, K., Yamanaka, H., Dai, Y., Tachibana, T., Fukuoka,
T., Tokunaga, A., Yoshikawa, H. and Noguchi, K. (2003)
Differential activation of extracellular signal-regulated pro-
tein kinase in primary afferent neurons regulates brain-de-
rived neurotrophic factor expression after peripheral
inflammation and nerve injury. J. Neurosci., 23: 4117–4126.
Pannek, J., Sommerfeld, H.J., Botel, U. and Senge, T. (2000)
Combined intravesical and oral oxybutynin chloride in adult
patients with spinal cord injury. Urology, 55: 358–362.

Paratcha, G., Ledda, F. and Ibanez, C.F. (2003) The neural cell
adhesion molecule NCAM is an alternative signaling recep-
tor for GDNF family ligands. Cell, 113: 867–879.
Pascual, J.I., Gudino-Cabrera, G., Insausti, R. and Nieto-
Sampedro, M. (2002) Spinal implants of olfactory ensheath-
ing cells promote axon regeneration and bladder activity after
bilateral lumbosacral dorsal rhizotomy in the adult rat.
J. Urol., 167: 1522–1526.
Qiao, L. and Vizzard, M.A. (2002) Up-regulation of tyrosine
kinase (Trka, Trkb) receptor expression and phosphorylation
in lumbosacral dorsal root ganglia after chronic spinal cord
(T8-T10) injury. J. Comp. Neurol., 449: 217–230.
Renganathan, M., Cummins, T.R. and Waxman, S.G. (2001)
Contribution of Na(v)1.8 sodium channels to action potential
electrogenesis in DRG neurons. J. Neurophysiol., 86:
629–640.
Renganathan, M., Gelderblom, M., Black, J.A. and Waxman,
S.G. (2003) Expression of Nav1.8 sodium channels perturbs
the firing patterns of cerebellar Purkinje cells. Brain Res.,
959: 235–242.
Reynard, J.M., Vass, J., Sullivan, M.E. and Mamas, M. (2003)
Sphincterotomy and the treatment of detrusor-sphincter
dyssynergia: current status, future prospects. Spinal Cord,
41: 1–11.
Sakamoto, K., Uvelius, B., Khan, T. and Damaser, M.S. (2002)
Preliminary study of a genetically engineered spinal cord im-
plant on urinary bladder after experimental spinal cord injury
in rats. J. Rehabil. Res. Dev., 39: 347–357.
Schumacher, S., Bross, S., Scheepe, J.R., Alken, P. and June-
mann, K.P. (1999) Sacral anterior root stimulation and pos-
terior rhizotomy in spastic neuropathic bladder. Restor.
Neurol. Neurosci., 14: 195–199.
Sculptoreanu, A., Yoshimura, N. and de Groat, W.
(2004) KW-7158 [(2S)-(+)-3,3,3-trifluoro-2-hydroxy-2-me-
thyl-N-(5,5,10-trioxo-4,10-dihydrothieno[3,2-c][1]benzothiepin-
9-yl)propanamide] enhances A-type K+ currents in neurons
of the dorsal root ganglion of the adult rat. J. Pharmacol.
Exp. Ther., 310: 159–168.
Seki, S., Sasaki, K., Fraser, M.O., Igawa, Y., Nishizawa, O.,
Chancellor, M.B., de Groat, W.C. and Yoshimura, N. (2002)
Immunoneutralization of nerve growth factor in lumbosacral
spinal cord reduces bladder hyperreflexia in spinal cord in-
jured rats. J. Urol., 168: 2269–2274.
Seki, S., Sasaki, K., Igawa, Y., Nishizawa, O., Chancellor,
M.B., De Groat, W.C. and Yoshimura, N. (2004) Suppres-
sion of detrusor-sphincter dyssynergia by immunoneutrali-
zation of nerve growth factor in lumbosacral spinal cord in
spinal cord injured rats. J. Urol., 171: 478–482.
Shefchyk, S. and Buss, R. (1998) Urethral pudendal afferent-
evoked bladder and sphincter reflexes in decerabrate and
acute spinal cats. Neurosci. Lett., 244: 137–140.
Shefchyk, S.J. (2001) Sacral spinal interneurones and the con-
trol of urinary bladder and urethral striated sphincter muscle
function. J. Physiol., 533: 57–63.
Shefchyk, S.J. and Buss, R.R. (1998) Urethral pudendal affer-
ent-evoked bladder and sphincter reflexes in decerebrate and
acute spinal cats. Neurosci. Lett., 244: 137–140.
161
Stucky, C.L. and Lewin, G.R. (1999) Isolectin B(4)-positive and
-negative nociceptors are functionally distinct. J. Neurosci.,
19: 6497–6505.
Tang, J. and Landmesser, L. (1993) Reduction of intramuscular
nerve branching and synaptogenesis is correlated with de-
creased motoneuron survival. J. Neurosci., 13: 3095–3103.
Tarpley, J.W., Kohler, M.G. and Martin, W.J. (2004) The be-
havioral and neuroanatomical effects of IB(4)-saporin treat-
ment in rat models of nociceptive and neuropathic pain.
Brain Res., 1029: 65–76.
Theodosiou, M., Rush, R.A., Zhou, X.F., Hu, D., Walker, J.S.
and Tracey, D.J. (1999) Hyperalgesia due to nerve damage:
role of nerve growth factor. Pain, 81: 245–255.
Thor, K.B., Blais, D.P. and de Groat, W.C. (1989) Behavioral
analysis of the postnatal development of micturition in kit-
tens. Brain Res. Dev. Brain Res., 46: 137–144.
Vera, P.L. and Nadelhaft, I. (2000) Anatomical evidence for
two spinal ‘afferent-interneuron-efferent’ reflex pathways in-
volved in micturition in the rat: a ‘pelvic nerve’ reflex path-
way and a ‘sacrolumbar intersegmental’ reflex pathway.
Brain Res., 883: 107–118.
Vizzard, M.A. (2000) Changes in urinary bladder neurotrophic
factor mRNA and NGF protein following urinary bladder
dysfunction. Exp. Neurol., 161: 273–284.
Vizzard, M.A. and Boyle, M.M. (1999) Increased expression of
growth-associated protein (GAP-43) in lower urinary tract
pathways following cyclophosphamide (CYP)-induced cysti-
tis. Brain Res., 844: 174–187.
Vulchanova, L., Olson, T.H., Stone, L.S., Riedl, M.S., Elde, R.
and Honda, C.N. (2001) Cytotoxic targeting of isolectin IB4-
binding sensory neurons. Neuroscience, 108: 143–155.
Vutskits, L., Djebbara-Hannas, Z., Zhang, H., Paccaud, J.P.,
Durbec, P., Rougon, G., Muller, D. and Kiss, J.Z. (2001) PSA-
NCAM modulates BDNF-dependent survival and differentia-
tion of cortical neurons. Eur. J. Neurosci,, 13: 1391–1402.
Wang, H.F., Shortland, P., Park, M.J. and Grant, G. (1998)
Retrograde and transganglionic transport of horseradish
peroxidase-conjugated cholera toxin B subunit, wheatgerm
agglutinin and isolectin B4 from Griffonia simplicifolia I in
primary afferent neurons innervating the rat urinary bladder.
Neuroscience, 87: 275–288.
Wang, K.C., Kim, J.A., Sivasankaran, R., Segal, R. and He, Z.
(2002) P75 interacts with the Nogo receptor as a co-receptor
for Nogo, MAG and OMgp. Nature, 420: 74–78.
Waxman, S.G., Cummins, T.R., Dib-Hajj, S., Fjell, J. and
Black, J.A. (1999) Sodium channels, excitability of primary
sensory neurons, and the molecular basis of pain. Muscle
Nerve, 22: 1177–1187.
Weld, K.J. and Dmochowski, R.R. (2000) Association of level
of injury and bladder behavior in patients with post-trau-
matic spinal cord injury. Urology, 55: 490–494.
Yamashita, T. and Tohyama, M. (2003) The p75 receptor acts
as a displacement factor that releases Rho from Rho-GDI.
Nat. Neurosci., 6: 461–467.
Yoshimura, N. and de Groat, W.C. (1997) Plasticity of Na+
channels in afferent neurones innervating rat urinary bladder
following spinal cord injury. J. Physiol., 503(Part 2): 269–276.
162
Yoshimura, N., Erdman, S.L., Snider, M.W. and de Groat,
W.C. (1998) Effects of spinal cord injury on neurofilament
immunoreactivity and capsaicin sensitivity in rat dorsal root
ganglion neurons innervating the urinary bladder. Neurosci-
ence, 83: 633–643.
Yoshimura, N., Franks, M.E., Sasaki, K., Goins, W.F., Goss,
J., Yokoyama, T., Fraser, M.O., Seki, S., Fink, J., Glorioso,
J., de Groat, W.C. and Chancellor, M.B. (2001) Gene ther-
apy of bladder pain with herpes simplex virus (HSV) vectors
expressing preproenkephalin (PPE). Urology, 57: 116.
Yoshimura, N., Seki, S., Erickson, K.A., Erickson, V.L., Chan-
cellor, M.B. and de Groat, W.C. (2003) Histological and
electrical properties of rat dorsal root ganglion neurons in-
nervating the lower urinary tract. J. Neurosci., 23: 4355–4361.
Yoshiyama, M., Nezu, F.M., Yokoyama, O., de Groat, W.C.
and Chancellor, M.B. (1999) Changes in micturition after
spinal cord injury in conscious rats. Urology, 54: 929–933.
Yu, X., Xu, L., Zhang, X.D. and Cui, F.Z. (2003) Effect of
spinal cord injury on urinary bladder spinal neural pathway:
a retrograde transneuronal tracing study with pseudorabies
virus. Urology, 62: 755–759.
Zhou, F.Q., Zhong, J. and Snider, W.D. (2003) Extra-
cellular crosstalk: when GDNF meets N-CAM. Cell,
113: 814–815.
Zvarova, K., Murray, E. and Vizzard, M.A. (2004) Changes in
galanin immunoreactivity in rat lumbosacral spinal cord and
dorsal root ganglia after spinal cord injury. J. Comp.
Neurol., 475: 590–603.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 11

Control of urinary bladder function with devices:

successes and failures

Robert A. Gaunt and Arthur Prochazka

Department of Biomedical Engineering and Center for Neuroscience, University of Alberta, 507 HMRC, Edmonton,
AB T6G 2S2, Canada

Abstract: The management of urinary tract dysfunction is crucial for the health and well-being of people
with spinal cord injury. Devices, specifically catheters, play an important role in the daily regime of bladder
management for most people with spinal cord injury. However, the high incidence of complications as-
sociated with the use of catheters, and the fact that the spinal segments involved in lower urinary tract
control remain intact in most cord-injured people, continue to motivate research into devices that could
harness the nervous system to provide greater control over lower urinary tract function. Mechanical devices
discussed in this review include catheters, artificial urethral sphincters, urethral stents and intraurethral
pumps. Additionally, many attempts to restore control of the lower urinary tract with electrical stimulation
have been made. Stimulation sites have included: inside the bladder, bladder wall, thigh, pelvic floor, dorsal
penile nerve, pelvic nerve, tibial nerve, sacral roots, sacral nerves and spinal cord. Catheters and sacral root
stimulators are two techniques whose efficacy is well established. Some approaches have proven less
successful and others are still in the development stage. Modifications to sacral root stimulation including
posterior root stimulation, anodal blockade and high-frequency blockade as well as new techniques in-
cluding intraspinal microstimulation, urethral afferent stimulation and injectable microstimulators are also
discussed. No single device has yet restored the control and function of the lower urinary tract to the pre-
injury state, but new techniques are bringing this possibility closer to reality.

Introduction spinal cord injury and account for the second

People with spinal cord injury face many chal-
lenging medical problems. Inadequate post-injury
management of lower urinary tract dysfunction
can lead to many complications including renal
failure. This used to be the leading cause of death
after spinal cord injury, but has dropped to fourth
position in recent decades (Frankel et al., 1998)
with improved treatment methods (Jamil, 2001).
However, complications of the genitourinary sys-
tem, primarily urinary tract infections, are the
most common cause of rehospitalization after
Corresponding author. Tel.: +780-492-3783;
Fax: +780-495-1617; E-mail: arthur.prochazka@ualberta.ca
highest number of bed-days for readmitted pa-
tients (Savic et al., 2000; Middleton et al., 2004).
While management of lower urinary tract dys-
function with devices, primarily catheters, has re-
duced mortality after spinal cord injury, the high
incidence of complications is largely due to the
limited success that these devices or other treat-
ment modalities have had in restoring normal
function to the neurogenic bladder. In addition to
these clinical considerations, effective management
of lower urinary tract dysfunction is generally
outranked in its importance to patients only by the
desire for hand function in people with quadriple-
gia and sex function in people with paraplegia
(Anderson, 2004). These factors provide an

DOI: 10.1016/S0079-6123(05)52011-9 163

164
impetus to develop improved methods of manag-
ing lower urinary tract dysfunction after spinal
cord injury.
Lower urinary tract control
The lower urinary tract has two functions: storing
urine (continence) and voiding urine (micturition).
The lower urinary tract is innervated by the so-
matic nervous system and both the sympathetic
and parasympathetic branches of the autonomic
nervous system. Efferent parasympathetic inner-
vation of the detrusor, the muscular layer of the
bladder, arises from preganglionic neurons in the
sacral (S) parasympathetic nucleus in spinal seg-
ments S2–S4. The preganglionic neurons send ax-
ons via the pelvic nerve to the pelvic plexus where
they synapse with ganglionic neurons. Afferent in-
nervation of the bladder is also primarily via the
pelvic nerve. Efferent somatic innervation of the
external urethral sphincter arises from motoneu-
rons in Onuf’s nucleus in spinal segments S1–S3.
These efferent axons as well as the afferents of the
external urethral sphincter and urethra travel via
the pudendal nerve. As the bladder fills during the
storage phase, stretch-sensitive mechanoreceptors
in the bladder wall transmit a sense of fullness to
both spinal and supraspinal centers. Once the
decision to void is reached, the external urethral
sphincter is voluntarily relaxed and parasympa-
thetic activity causes detrusor contractions. This
synergistic activity, coordinated by the pontine mi-
cturition center, results in micturition (Barrington,
1921, 1925). More details on the anatomy and
physiology of the lower urinary tract can be found
in de Groat (1993) and de Groat et al. (2001).
After spinal cord injury, supraspinal coordina-
tion from the pontine micturition center is lost
leading to lower urinary tract dysfunction. Sacral
spinal cord or cauda equina lesions generally lead
to an areflexive bladder and sphincter paralysis.
Suprasacral lesions, however, spare sacral spinal
reflexes, and after a period of shock reflexive blad-
der contractions often occur at low bladder vol-
umes. This condition, called detrusor hyperreflexia
or neurogenic detrusor overactivity (reviewed in
Yoshimura, 1999), is often accompanied by reflexive
co-contractions of the external urethral sphincter.
This combination, termed detrusor–sphincter dys-
synergia (Andersen and Bradley, 1976; Blaivas et al.,
1981), leads to incontinence, inefficient voiding
with high residual volumes and high intravesical
pressures which in turn leads to ureteric reflux and
upper urinary tract deterioration.
Why devices?
Any treatment for lower urinary tract dysfunction
after spinal cord injury should create a bladder
capable of storing large volumes of urine at low
pressure, prevent incontinent episodes and allow
periodic evacuation of urine at low pressure. Sur-
gical treatments, such as bladder augmentation
using a section of intestine, ameliorate the problem
of hyperreflexia and low storage volume (Hollander
and Diokno, 1993), while sphincterotomies
(cutting into the external urethral sphincter) im-
prove detrusor–sphincter dyssynergia (Reynard
et al., 2003). Anticholinergic medications are fre-
quently used to relax the hyperreflexive bladder
but have undesirable side effects including a dry
mouth and blurred vision (Wein, 1998). These
treatments address the symptoms of the ne-
urogenic bladder so that storage and evacuation
of urine is achieved without upper urinary tract
damage, but they do not address the fundamental
loss of control associated with spinal cord injury.
Devices present attractive alternatives to the
management of lower urinary tract dysfunction
after spinal cord injury as they attempt, at least
partly, to restore the control of the neurogenic
bladder. Additionally, while devices are locally in-
vasive to varying degrees, they do not generally
cause systemic complications as do pharmacolog-
ical treatments. Surgical procedures such as the
ones described above are usually irreversible and
subsequently limit patients to a specified course of
treatment while possibly excluding new tech-
niques. The devices described in this review, and
those under development do not generally cause
irreversible changes and therefore do not prevent
patients from taking advantage of improved treat-
ments in the future.

Many review articles have been published that
focus on devices for bladder control (Schmidt,
1983; Talalla et al., 1987; Lee, 1997; Rijkhoff et al.,
1997b; Grill et al., 2001; Groen and Bosch, 2001;
Jamil, 2001; Jezernik et al., 2002; Van Kerrebroeck,
2002; Middleton and Keast, 2004; Rijkhoff, 2004b;
van Balken et al., 2004), so no attempt will be
made to provide detailed descriptions of each of
these methods here. Rather, we will summarize the
methods that have been devised over the years for
device-based management of the neurogenic blad-
der secondary to spinal cord injury and summarize
current research on those devices and methods
that are likely to affect the field in the future. Ad-
ditionally, we will attempt to identify the reasons
that many methods and devices have ultimately
been unsuccessful, sometimes in spite of good clin-
ical results. Finally, we will summarize the prob-
lems that we feel should be addressed to improve
the effectiveness and adoption of devices in the
management of the neurogenic bladder. Both me-
chanical and electrical devices will be described as
they have met with different levels of success and
failure and have the potential to offer solutions to
a variety of problems faced by people with spinal
cord injury.
Mechanical devices for control of the lower
urinary tract
Catheters
The use of catheters to manage urinary retention
dates back to ancient Egypt (reviewed in Nacey
and Delahunt, 1993). During World War I, up
to 80% of patients with spinal cord injury died
shortly after injury due to complications arising
from the neurogenic bladder (Kennedy, 1946).
However, improved management of the lower
urinary tract using catheters during World War II
(Kennedy, 1946) and especially Guttmann’s
technique of sterile intermittent catheterization
(Guttmann and Frankel, 1966) helped reduce this
figure significantly. Sterile intermittent catheteriza-
tion was eventually modified to non-sterile clean in-
termittent catheterization for reasons of practicality
165
(Comarr, 1972; Lapides et al., 1972), and this
technique, along with generally improved medical
care, has caused urinary tract dysfunction to fall
from the primary cause of death (22%) for pa-
tients injured between 1943 and 1972 to the fourth
most common cause of death (9%) for patients
injured between 1973 and 1990 (Frankel et al.,
1998).
Chronically indwelling urethral and suprapubic
catheters, condom catheters and clean intermittent
catheterization are common forms of catheteriza-
tion currently used in the management of spinal
cord injury patients. Each method has its own ad-
vantages and disadvantages (reviewed in Selzman
and Hampel, 1993), but clean intermittent cathe-
terization is the form of bladder management
least likely to lead to complications (Weld and
Dmochowski, 2000). Clean intermittent catheter-
ization is generally the most prescribed form
of bladder management at hospital discharge
(Cardenas et al., 1995), and although a number
of reports suggest that there is a trend for some
people to switch to other methods (Cardenas et al.,
1995; Weld and Dmochowski, 2000), a more re-
cent study suggests that this trend may be revers-
ing (Hansen et al., 2004). However, only 30% of
cord-injured people using clean intermittent cath-
eterization remain free of urinary tract infections.
Clean intermittent catheterization requires good
hand function, preventing people with tetraplegia
and impaired hand function as well as some people
with paraplegia from performing this procedure
themselves (Selzman and Hampel, 1993; Dahlberg
et al., 2004).
The critical role of catheter technology and
techniques in the management of lower urinary
tract dysfunction after spinal cord injury cannot be
overstated. The simplicity and clinical efficacy of
catheters in increasing life expectancy in people
with cord injury make them arguably the single
most important device for these people. However,
the high incidence of urinary tract infections and
other complications associated with catheter
use presents a continuous burden on patients and
the medical system. This, and the desire of people
with spinal cord injury for improved methods
(Anderson, 2004), is a motivation for new device
development.
166
Artificial sphincters
The concept of an artificial urethral sphincter was
first proposed by Foley (1947) to treat urinary in-
continence. The artificial urethral sphincter devel-
oped by Scott, Bradley and Timm (Scott et al.,
1974; Timm et al., 1974) has developed into the
commercially available AMS 800 artificial sphinc-
ter (American Medical Systems, Minnetonka,
MN, USA) (reviewed in Hajivassiliou, 1998). The
AMS 800 uses a pump to deflate a cuff placed
around the bladder neck or urethra by transferring
fluid to a pressure-regulated reservoir. The cuff re-
inflates automatically over a period of several
minutes. Of reported studies using the AMS 800
including 2606 subjects, 73% achieved full conti-
nence, 14% experienced device failure, 4.5% ex-
perienced infections and 11.7% experienced
urethral erosion from excessive pressure placed
on the urethra by the cuff (Hajivassiliou, 1998).
Artificial urethral sphincters are primarily used
to treat patients with post-prostatectomy inconti-
nence, but have been successful in managing
incontinence with other etiologies as well (Petrou
et al., 2000). It was originally suggested that
detrusor hyperreflexia was a contraindication for
artificial urethral sphincter implantation as high
intravesicular pressures may cause deflation of
the pressure-regulated cuff (Scott et al., 1974).
However, artificial urethral sphincters have been
implanted in spinal cord injury patients with an
overall success rate of 70% (Light and Scott, 1983),
though device removal due to infections was high
(24%). Currently, artificial urethral sphincters are
not commonly used to manage incontinence after
spinal cord injury, but can be useful in people with
lesions leading to a flaccid bladder and sphincter.
Urethral stents
Urethral stents were first developed to treat ure-
thral strictures, but shortly after, their use in spinal
cord injury patients with detrusor–sphincter dys-
synergia leading to hydronephrosis and vesicoure-
teric reflux was described (Shaw et al., 1990).
Urethral stents were proposed as an alternative to
sphincterotomies, the primary surgical treatment
for patients with detrusor–sphincter dyssynergia.
Sphincterotomies are generally irreversible and
can cause hemorrhage, erectile dysfunction, blad-
der neck stenosis or stricture (reviewed in Reynard
et al., 2003). Urethral stents are inserted into the
urethra and mechanically hold the external ure-
thral sphincter open. After sphincterotomy, or im-
plantation of a urethral stent, most cord-injured
people must wear a collection device such as a
condom catheter as the continence mechanism of
the urethra is defeated.
Several different urethral stent designs have
been tested in various trials including the
UroLumes (American Medical Systems, Minne-
tonka, MN, USA) (Chancellor et al., 1999b),
Memokaths (Doctors & Engineers A/S Ltd.,
Kvistgaard, Denmark) (Low and McRae, 1998;
Hamid et al., 2003), Memotherms (Bard Corp.,
Covington, GA, USA) (Juan Garcia et al., 1999)
and Ultraflexs (Boston Scientific Corp., Natick,
MA, USA) (Chartier-Kastler et al., 2000). The
UroLume, Memotherm and Ultraflex are flexible
wire mesh tubes while the Memokath is a helically
wound wire. The devices are inserted into the ure-
thra and positioned in the region of the external
urethral sphincter where the wire becomes largely
covered by urothelium over time. The UroLume is
the best studied of these devices and has similar
results, in terms of urodynamic parameters and
incidence of urinary tract infection to sphincter-
otomies, but requires less hospitalization and is
potentially reversible (Chancellor et al., 1999a). A
5-year multi-center trial of the UroLume in 160
cord-injured subjects showed that the treatment
was successful in 84%, while 15% required ex-
plantation. Complications such as device migra-
tion were most common in the first 3 months
(Chancellor et al., 1999b). Although explantation
of the stent was possible, it has presented a variety
of challenges (Chancellor et al., 1999b; Wilson et al.,
2002). The Memokath was found to be suitable
for short-term implantation only as most devices
fail within 2 years (Hamid et al., 2003) and com-
plications including migration, autonomic dysre-
flexia and stone formation on the stent can occur
(Low and McRae, 1998). However, explantation
of this device is much simpler than the UroLume
due to its helical design and thermosensitive
 167

material which, when cooled with saline, becomes

soft and uncoils, making this device useful for
acute management of detrusor–sphincter dyssy-
nergia (Hamid et al., 2003).
Urethral stents represent a clinically successful
device for management of detrusor–sphincter
dyssynergia in people with spinal cord injury. Al-
though stents do not restore normal control of the
sphincter, their efficacy, simplicity and potential
reversibility makes them an attractive option for
people who would otherwise receive an irreversible
sphincterotomy (Chancellor et al., 1999b).

Intraurethral pump

In 1997, Nativ et al. (1997) described a device in-

corporating a miniature valve and pump that
could be inserted into the urethra to control both
continence and voiding in women. The In-FlowTM
intraurethral pump (SRS Medical Systems, Inc.,
Billerica, MA, USA) is designed to manage chron-
ic urinary retention caused by an atonic bladder or
urethral dysfunction. The device secures itself in
the urethra by means of flexible fins that open in
the bladder and a flange at the external urethral
meatus (see Fig. 1). The device is controlled by a
remote activator that is placed over the pubic area
and is magnetically coupled to the pump. Once
activated, the turbine actively pumps urine out of Fig. 1. The In-FlowTM intraurethral pump. (A) Photograph
the bladder at a rate of 6–12 ml/s until the bladder showing the unfolded petals that secure the device in the ure-
is empty. The device is easily inserted by a phy- thra and prevent migration. (B) Diagram showing the place-
sician and can be removed by the patients if they ment of the device in the urethra. Adapted from Madjar et al.
(1999) and Schurch et al. (1999).
wish. The device is designed to be replaced every
month, but successful usage to an average of 90
days has been reported, at which time the device
can become fouled by salt deposits (Madjar et al., reported success rates of 50% with average follow-
1999). up times of 3 and 7.6 months respectively. Most of
In a study of 18 women with spinal cord injury those patients that adopted this device for long-
and hyporeflexive bladders, only six continued to term usage were previously dependent on clean
use the device at follow-up (mean 9.6 months) intermittent catheterization and preferred the con-
(Schurch et al., 1999). Discomfort, incontinence, venience of this device. Intraurethral pumps are
urinary tract infections, technical failures, urethral very interesting from a technical viewpoint and
dilation and the possibility of long-term urethral further investigation with clearer indications for
damage were cited as reasons why this device was use, such as complete spinal cord injury, atonic
unsuitable for chronic use. Studies in 60 (Mazouni bladder and previous dependence on clean inter-
et al., 2004) and 92 (Madjar et al., 1999) patients mittent catheterization may improve the success
with voiding dysfunction from various etiologies rate among people with spinal cord injury.
168
Electrical stimulation devices for control of the
lower urinary tract
While mechanical devices are necessarily limited to
treating symptoms of the neurogenic bladder, elec-
trical stimulation techniques allow devices to be
created that can exert control over spared muscles
and their neural control systems. Electrical cur-
rent, passed between two electrodes, can be used to
generate action potentials in surviving neurons in
the spinal cord or peripheral nerve below the le-
sion in spinal cord injury patients. These artificial-
ly generated action potentials can lead directly to
muscular contraction or they can modulate the
activity of neuronal networks and reflex pathways
(termed neuromodulation).
The discussion below of devices and techniques
that have been developed to control the lower
urinary tract is organized by the location of stim-
ulation electrodes rather than by the neurophys-
iological mechanisms on which the devices operate
or by their intended function. Five primary loca-
tions can be identified where electrical stimulation
electrodes can be placed: on or in the bladder, on
the skin, peripheral nerve, sacral roots and in the
spinal cord itself. Figure 2 shows the various stim-
ulation locations for devices discussed throughout
this review.
Electrical stimulation of the bladder
Intravesical stimulation
Intravesical electrical stimulation was the first
attempt at treating bladder dysfunction using elec-
trical stimulation. In 1878, M.H. Saxtorph de-
scribed a technique in which stimulation between a
catheter-mounted electrode, passed into the blad-
der to act as the cathode (see Fig. 2A), and a sup-
rapubically placed indifferent electrode, was used
to treat urinary retention caused by an underactive
bladder (reviewed in Madersbacher, 1990). In-
travesical electrical stimulation is essentially a ne-
uromodulation therapy intended to reinforce the
weak functioning of existing neural micturition
pathways by stimulating mechanoreceptors in the
bladder wall to facilitate reflex bladder contrac-
tions and improve sensation. Electrical stimulation
is below the threshold required to elicit bladder
contractions directly via stimulation of the efferent
portion of the pelvic nerve or of the detrusor
myocytes themselves. Acute studies in rats and
cats have confirmed the hypothesis that in-
travesical electrical stimulation acts by stimulat-
ing stretch-sensitive mechanoreceptors in the wall
of the bladder that reflexively cause contractions
of the bladder (Ebner et al., 1992).
Few reports of intravesical electrical stimulation
studies in people with spinal cord injury exist, but
one dealing specifically with subjects with incom-
plete spinal cord injury reported improvements in
bladder sensation, detrusor contraction and resid-
ual volumes in almost all subjects (Madersbacher
et al., 1982). A retrospective study on the effec-
tiveness of intravesical electrical stimulation for
people with spinal cord injury by the same author
indicated that one third of the subjects experienced
improvements in sensation, detrusor contractility
and voluntary control. This occurred only in in-
dividuals with preserved pain sensation in the
S2–S4 dermatomes (Madersbacher, 1990). This
would seem to be the only predictor of the efficacy
of this therapy. Additionally, patients require
many hours of treatment before the effectiveness
of intravesical electrical stimulation can begin to
be evaluated and the positive results reported by
some investigators (Kaplan, 2000) have not been
repeatable by others (Decter, 2000). While in-
travesical electrical stimulation has been used to
treat patients with spinal cord injury, recent stud-
ies have focused on children with underactive
bladders (Gladh et al., 2003). Intravesical electrical
stimulation seems ultimately unattractive as a clin-
ical technique to improve micturition in people
with spinal cord injury as it only seems to work in
some with incomplete spinal cord injury, requires
long treatments before effectiveness can be evalu-
ated and the results have not been repeatable
among investigators.
Bladder wall stimulation
Electrical stimulation of the exterior surface of the
bladder (see Fig. 2B) was first studied in the early
1950s (Boyce et al., 1964). This marked the begin-
ning of the development of electrical stimulation
 169

Fig. 2. Electrode locations for controlling the lower urinary tract. The locations are numbered primarily by the order in which they are
discussed in the text. The location for a posterior rhizotomy is also indicated. (A) Intravesical, (B) bladder wall, (C) Thigh, (D) pelvic
floor, (E) dorsal penile nerve, (F) tibial nerve, (G) pelvic nerve, (H) intradural sacral anterior root, (I) extradural mixed sacral root, (J)
intradural sacral posterior root, (K) sacral nerve, (L) spinal cord, (M) intraurethral, (N) pudendal nerve, (O) sacrum.

devices to elicit voiding directly, in response to the
high morbidity and mortality associated with cath-
eterization (Bradley et al., 1962). Several groups
developed implanted stimulators inductively
coupled to external transmitters with variations
in the design, placement and number of electrodes
(Bradley et al., 1962; Hald et al., 1967; Stenberg
et al., 1967; Susset and Boctor, 1967; Merrill and
Conway, 1974; Magasi and Simon, 1986). Initial
animal experiments demonstrated that dogs with
spinal cord transections were able to void regularly
using the implanted stimulators without requiring
additional procedures (Bradley et al., 1962, 1963;
Kantrowitz and Schamaun, 1963). However, re-
sults in spinal cord injury patients implanted with
these stimulators were much less successful (Bradley
et al., 1963; Hald et al., 1967; Stenberg et al., 1967;
Susset and Boctor, 1967; Merrill and Conway,
1974). The primary reason that these people were
unable to void was that stimulation currents high
enough to generate useful bladder contractions
spread to surrounding structures causing co-
activation of the external urethral sphincter and
pelvic floor musculature. It was noted that the
canine bladder is primarily an abdominal organ,
whereas the human bladder is a pelvic organ and is
in close proximity to the pelvic floor musculature,
increasing pelvic floor susceptibility to contraction
by current spread (Bradley et al., 1963).
Because of this problem, experimental and clinical
work was then directed toward obtaining sufficient
contraction of the bladder while limiting current
spread. Tape electrodes and more powerful stimula-
tors successfully elicited micturition, but infection
and technical failures prevented evaluation of their
long-term effect (Bradley et al., 1963). Experience
with the Avco stimulator, in which individual wires
were embedded into the bladder wall, were also
hampered by activation of urethral and pelvic floor
musculature (Hald et al., 1967; Stenberg et al.,
170

1967). Another stimulator design, the Mentor

bladder stimulator, used two helical wire electrodes
sewn into the bladder wall. This was successful in
two of five people with upper motoneuron lesions,
but required subarachnoid injections of phenol
to abolish electrically induced detrusor–sphincter
dyssynergia that otherwise prevented micturition
(Merrill and Conway, 1974). Susset and Boctor
(1967) reported a successful implant that incorpo-
rated eight disc electrodes around the dome of
the bladder in a person with a complete lower
motoneuron lesion. These investigators considered
upper motoneuron lesions to be a contraindication
for implantation of these systems due to the un-
wanted activation of sphincter and pelvic floor mus-
cles. The most successful report of bladder wall
stimulation was made by Magasi and Simon (1986)
in which 29 of 32 subjects with neurogenic bladder
paralysis attained complete voiding with eight disc
electrodes implanted around the bladder (see
Fig. 3). However, the concomitant sphincter acti-
vation reported by most investigators, lead and
electrode breakage, receiver malfunction, bladder
perforation and pain caused failure in most human
studies. With the success of sacral root stimulation
(see below) for restoring micturition in people with
upper motoneuron lesions, and the multiple diffi-
culties in achieving successful clinical results with
bladder wall stimulation, recent work in this area
has focused on people with lower motoneuron
lesions who cannot benefit from sacral root stim-
ulation (Walter et al., 1999).

Transcutaneous electrical stimulation

Thigh stimulation
In 1986 it was reported that electrical stimulation
through surface electrodes over the thigh muscles
(see Fig. 2C) could cause changes in the urody-
namic parameters of spinal cord injury patients
(Wheeler et al., 1986). Stimulation was applied
through bilateral quadriceps surface electrodes on a
daily basis for 4–8 weeks. Some people exhibited
persistent increases in bladder capacity and/or re-
ductions in bladder pressure, while others experi-
enced the opposite result. Another study, examining
hamstring and quadriceps stimulation to reduce
Fig. 3. The bladder wall stimulator used by Magasi and Simon
(1986). (A) The intended positioning of electrodes on the blad-
der. (B) Actual positions of electrodes around the bladder in
one female subject. Reprinted from Magasi and Simon (1986)
with permission from S. Karger AG, Basel.
spasticity in cord-injured people, noted that 16 of
32 subjects became continent (Shindo and Jones,
1987) perhaps indicating a suppression of detrusor
hyperreflexia. A more recent study examining uro-
dynamic changes in response to thigh muscle stim-
ulation showed that 8 of 14 subjects, including one
person with spinal cord injury and neurogenic de-
trusor overactivity, increased their bladder volumes

by 450% (Okada et al., 1998). However, no meth-
ods of identifying those people likely to respond
positively to treatment exist. None of these studies
noted any adverse side effects from the treatment.
The effects of electrical stimulation of the thigh
muscles on the bladder may be mediated by limb
afferents known to inhibit bladder contractions to
prevent leakage during physical activity (Fall and
Lindstrom, 1991), although other mechanisms
have been proposed (Okada et al., 1998). Carry-
over, observed with thigh stimulation, has also
been observed with other electrical stimulation
techniques (Fall and Lindstrom, 1991), and may
be at least partially explained by mechanisms such
as those proposed by Vodovnik (1981). Despite the
simplicity of this approach, efficacy in some people
and lack of adverse side effects, few studies of
stimulation of the thigh muscles have been report-
ed, and this technique does not appear to be widely
used in practice. This is likely because many pa-
tients show no improvement, and those that may
cannot be identified prior to treatment. Addition-
ally, treatment requires a significant time invest-
ment and most people can achieve effective
suppression of hyperreflexive bladder contractions
with anticholinergic medications.
Pelvic floor maximal functional electrical
stimulation
On the basis of a previous observation, Moore and
Schofield (1967) decided to test the effectiveness of
electrically induced maximal contraction of the pel-
vic floor musculature (see Fig. 2D) to treat female
patients with stress incontinence. Some people re-
ported being cured after a single session and more
reported a reduction in symptoms. Maximal func-
tional electrical stimulation may involve the use of
surface, vaginal, anal, penile, percutaneous or a
combination of such electrodes to stimulate the
pelvic floor musculature and pudendal nerve at the
maximum tolerable threshold for subjects. This
treatment can lead to long-lasting bladder inhibi-
tion in subjects with non-neurogenic bladder
overactivity (Fall and Lindstrom, 1991). While
maximal functional electrical stimulation is used
in some clinical settings for treating incontinence
in many patient groups (Geirsson and Fall, 1997),
171
results in people with spinal cord injury for sup-
pressing hyperreflexive bladder contractions are
mixed (reviewed in Previnaire et al., 1998). Given
the side effects of maximal functional electrical
stimulation, including physical discomfort in people
with incomplete spinal cord injury, possibly limiting
the stimulation current to non-therapeutic levels
(Previnaire et al., 1998), as well as psychological
discomfort (van Balken et al., 2004), anticholinergic
medications are often a more practical method to
manage neurogenic detrusor overactivity.
Dorsal penile nerve stimulation
Stimulation of the dorsal penile nerve or clitoral
nerve can inhibit detrusor activity. These nerves
form the most superficial branch of the pudendal
nerve and are therefore easily accessible. Detrusor
inhibition by this means was first demonstrated sci-
entifically using mechanical stimulation (penile
squeeze) to suppress ongoing bladder contractions
(Kondo et al., 1982). This effect has also been
demonstrated using electrical stimulation with bi-
polar surface electrodes placed on the penis
(Nakamura and Sakurai, 1984) (see Fig. 2E). Giv-
en the success of this simple technique in inhibiting
bladder contractions, its potential in treating sub-
jects with detrusor hyperreflexia secondary to spinal
cord injury has been examined by several groups. In
one study of six spinal cord injury subjects with
complete and incomplete cervical and thoracic le-
sions, inhibition of detrusor contractions during
bladder filling was demonstrated in all subjects
(Wheeler et al., 1992). Continuous stimulation at
five pulses per second was sufficient to increase the
volume at which reflexive bladder contractions first
occurred during a cystometrogram by an average of
76% (range 26% to 150%) without side effects.
Continuous stimulation of the genital nerves, how-
ever, may pose practical challenges for the design of
a neuroprosthesis and would preclude measurement
of bladder activity using peripheral nerve recording
techniques (Jezernik et al., 2000). Kirkham et al.
(2001) and Dalmose et al. (2003) therefore exam-
ined whether conditional stimulation of the dorsal
penile nerve was sufficient to effect clinically useful
inhibition of the detrusor. Stimulation lasting 1 min
was initiated by a rise in bladder pressure of 10 cm
172
H2O during a cystometrogram and successfully in-
hibited bladder contractions while increasing blad-
der capacity by 144% (7127%) in all six spinal
cord injury subjects studied (Kirkham et al., 2001).
The effects of conditional stimulation on detrusor
inhibition are robust among both male (Kirkham
et al., 2001; Dalmose et al., 2003) and female
(Dalmose et al., 2003) patients with a wide range of
injury levels. In addition to inhibiting hyperreflexive
bladder contractions, dorsal penile nerve stimula-
tion can reduce blood pressure in people with high
level spinal cord injuries (Lee et al., 2003). This may
reduce the risks associated with autonomic dysre-
flexia often triggered by a full bladder or bowel.
Dorsal penile nerve stimulation is an active area
of neuroprosthesis development due to the relative
simplicity of the technique and its reliability and
efficacy in people with spinal cord injury. Two re-
cent reports describe attempts at moving dorsal
penile nerve stimulation from the laboratory to
clinical use. Lee and Creasey (2002) describe the
application of a surface stimulation system to a
person with an incomplete cervical (C)6 injury
who experienced episodes of incontinence after
sensing his full bladder, but before he was able to
catheterize himself. Figure 4 shows the effect that
conditional dorsal penile nerve stimulation had on
bladder contractions in this subject. During home
use for 3 weeks, this man applied stimulation when
he sensed his bladder was full, allowing him time
to perform successful catheterization. He contin-
ued to use the system after the trial was over be-
cause of its success and his confidence in it.
Additionally, Fjorback et al. (2003) developed a
portable device that measured bladder pressure
and automatically stimulated the dorsal penile
nerve to inhibit bladder contractions. This device
used a catheter to measure bladder pressure, and
as such was impractical clinically, but did serve to
demonstrate the feasibility of a closed-loop system
to treat neurogenic detrusor overactivity.
Stimulation of peripheral nerve
Tibial nerve stimulation
In 1983, a report of investigations on nonhuman
primates demonstrated that the amount of current
required to cause detrusor inhibition with bipolar
percutaneous anal sphincter stimulation could be
reduced by changing the cathode to a surface elec-
trode positioned over the posterior tibial nerve
(McGuire et al., 1983). In the same study, similar
results were observed with percutaneous tibial
nerve stimulation alone (see Fig. 2F). This target
was chosen as it is the acupuncture point used to
inhibit bladder contractions in Chinese medicine.
This technique was successful in improving conti-
nence in 19 of 22 subjects, including four with
spinal cord injury, although bladder contractions
returned immediately once stimulation ceased.
More recent work with tibial nerve stimulation
includes evaluation of the commercially available
Urgent PC device (CystoMedix, Andover, MN,
USA), formerly the Urosurge SANS device given
FDA approval in 2000 (Govier et al., 2001;
van Balken et al., 2001; Vandoninck et al., 2003).
Encouraging results were reported in patients with
overactive bladders although no spinal cord injury
subjects were included in these studies. Two small-
scale studies evaluating tibial nerve stimulation in
subjects with spinal cord injury also reported re-
ductions in incontinence caused by neurogenic de-
trusor overactivity, although the results in the
subset of people with spinal cord injury are not
stated in one study (Amarenco et al., 2003) and the
other is a report from a single patient (Andrews
and Reynard, 2003). Given the technical simplicity
of this technique and its potential to suppress ne-
urogenic detrusor overactivity, further experi-
ments to determine efficacy in larger groups of
spinal cord injury subjects would help determine if
this technique should be pursued.
Pelvic nerve stimulation
Stimulation of the nerve supply to the bladder
presents some potential advantages over bladder
wall stimulation for people with spinal cord injury.
Stimulation of the pelvic nerve (see Fig. 2G), which
contains the preganglionic parasympathetic fibers
innervating the detrusor, should cause contraction
of the entire detrusor at a much lower current than
bladder wall stimulation (Hald, 1969). Pelvic nerve
stimulation was shown to elicit bladder contrac-
tions in dogs, but co-activation of the sphincters
prevented good micturition, especially in male dogs
 173

Fig. 4. Bladder pressure recording with and without stimulation during provocation of the bladder with rapid infusions of saline.
Reproducible reflexive bladder contractions were caused by rapid infusions of 60 ml of saline (C and D) and were abolished by
withdrawal of the saline. Further provocations (E and F) resulted in small reflexive bladder contractions that were immediately
abolished by stimulation of the dorsal penile nerve. During F and G, bladder pressure increases are hyperreflexive contractions caused
by the high volume. The two stimulation periods shown by open arrows at the onset and termination of G indicate patient-initiated
stimulation in response to sensation of bladder fullness. These suppressions of reflexive bladder contractions were better than the
previous ones. Without stimulation, bladder contractions are not suppressed (H). Reprinted from Lee and Creasey (2002) with
permission from the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Reha-
bilitation.

(Holmquist and Olin, 1968a, b). Sphincter activa-
tion was likely a reflex caused by stimulation of
bladder afferents in the pelvic nerve. In addition,
chronic stimulation was found to cause fibrosis of
the pelvic nerve leading to reductions in bladder
response over time (Hald, 1969). Application in
humans was also frustrating because of the struc-
ture of the pelvic nerve. Whereas a distinct pelvic
nerve exists in the cat and dog, parasympathetic
innervation of the human bladder is distributed
from the pelvic plexus arising from the pelvic
nerves shortly after their exit from the sacral fora-
mina (Wozniak and Skowronska, 1967), making
placement of electrodes very difficult (Susset and
Boctor, 1967; Hald, 1969). Despite this, brief re-
ports describing mixed results with pelvic nerve
stimulation in humans appeared in the 1970s
(Burghele, 1973; Kaeckenbeeck, 1979 cited in
Rijkhoff et al., 1997b), but no further reports ex-
ist to our knowledge.
Stimulation of sacral roots and nerves
Sacral root stimulation
The first electrical stimulation technique to devel-
op into a commercially available device for blad-
der emptying in cord-injured people began with
the work of Brindley (1977) as well as that of
Tanagho and Schmidt’s group (Heine et al., 1977;
Schmidt et al., 1979). These investigators reported
positive results using a sacral anterior root
stimulator to elicit voiding in spinalized animals.
This was followed by successful outcomes in hu-
mans with spinal cord injury (Brindley et al., 1982;
Tanagho et al., 1989). Brindley’s device was
commercialized as the Finetech–Brindley Bladder
System (Finetech Medical Ltd., Welwyn Garden
City, UK) and has been implanted in over 2500
people, in some cases for over 20 years (Rijkhoff,
2004b). This system has been described and
reviewed in detail in a number of articles, so only
a summary of the device will be presented here
(Brindley, 1977; Brindley et al., 1982; Brindley
et al., 1986; Creasey, 1993; Egon et al., 1998).
The two prerequisites for implantation of sacral
anterior root stimulators are intact parasympathetic
preganglionic neurons and a detrusor that is able to
contract (Creasey, 1993). Electrodes can either be
implanted intradurally (see Fig. 2H) on the S2–S4
anterior roots (Brindley et al., 1982) or extradurally
(see Fig. 2I) on the mixed sacral roots within the
spinal canal (Sauerwein et al., 1990; Lee, 1997). In
either case, the procedure is usually combined with
sacral posterior rhizotomy to abolish hyperreflexive
bladder and sphincter contractions, autonomic
dysreflexia triggered by bladder fullness and pain
in patients with incomplete lesions (Creasey, 1993;
174
Fig. 5. The components of the Finetech–Brindley sacral ante-
rior root stimulator. (A) Intradural electrodes and leads. (B)
Extradural electrodes and leads. (C) 2 and 3 channel implan-
table receiver blocks. These components are implanted subcu-
taneously and connect to the electrode leads. (D) The external
components of the device including the stimulator and trans-
mission block that is placed on the skin over the receiver block.
Adapted from Egon et al. (1998).
Brindley, 1994). After electrode placement, leads
are tunneled subcutaneously to an implantable re-
ceiver, activated by an external controller through a
radiofrequency link (Brindley et al., 1982). Figure 5
shows the components of this system. After im-
plantation and posterior rhizotomy, the majority of
patients are continent, have increased bladder ca-
pacity, are able to void using their stimulator with
residual volumes o30 ml and are freed from cath-
eter usage leading to a great reduction in urinary
tract infections (Van Kerrebroeck et al., 1993). Ad-
ditionally, patients have reported beneficial stimula-
tor-driven erections and defecation (Brindley et al.,
1986; Van Kerrebroeck et al., 1993; Egon et al.,
1998).
One limitation of this device is that electrical
stimulation of the sacral roots, in addition to pro-
ducing sustained increases in bladder pressure, ac-
tivates the external urethral sphincter due to the
presence of both small diameter parasympathetic
preganglionic fibers and large diameter somatic fib-
ers in the sacral anterior roots (Brindley, 1977).
Since large fibers have lower thresholds of electrical
stimulation, excitation of the parasympathetic pre-
ganglionic fibers is accompanied by excitation of
the somatic fibers, leading to external urethral
sphincter contraction and urethral occlusion. The
Finetech–Brindley sacral anterior root stimulator
circumvents this problem by utilizing the difference
in the relaxation time of the detrusor and the
sphincter (Brindley et al., 1982). A train of electrical
stimuli is applied for 3–9 s, allowing bladder pres-
sure to rise behind the closed sphincter. Upon ces-
sation of stimulation, the striated sphincter relaxes
quickly while bladder pressure is transiently main-
tained allowing post-stimulus voiding (Brindley
et al., 1982) (see Fig. 6). Despite the supranormal
bladder pressures that occur with this technique, no
evidence of vesicoureteric reflux or hydronephrosis
has been found (Creasey, 1993). Sacral roots also
contain fibers innervating the musculature of the
legs, and leg movement during stimulation can be
cumbersome to some patients.
Given the proven benefits to people and the
large number of other neuroprosthetic devices
implanted in patients (Rijkhoff, 2004b), one might
ask why more Finetech–Brindley sacral anterior
root stimulators have not been implanted. This is
likely due in part to the unwillingness of people to
undergo the irreversible posterior rhizotomy,
which, in addition to its very great benefits, abol-
ishes reflex erection, defecation and micturition as
well as any remaining perineal sensation. Implan-
tation of this system is also technically demanding

Fig. 6. Urine flow rate in a subject voiding his bladder using the
post-stimulus voiding technique. During stimulation, the flow
rate drops to nearly zero as contraction of the sphincter oc-
cludes the urethra. Reprinted with permission from Brindley
et al. (1982).
and attempts to market the device in the United
States as the VocareTM Bladder System by Ne-
uroControl Corp. (Cleveland, OH, USA) were
ultimately unsuccessful for commercial and regu-
latory reasons (Hall, 2003), despite evidence that
long-term use could realize a reduction in costs for
management of lower urinary tract dysfunction
(Creasey and Dahlberg, 2001). Commercialization
in the USA has recommenced through NDI Med-
ical (Cleveland, OH, USA).
Sacral nerve neuromodulation
In the early 1980s, Tanagho and Schmidt began
implanting extradural sacral root stimulators in
patients (Tanagho and Schmidt, 1988; Tanagho
et al., 1989). It was found that continence could be
controlled by low-frequency, low-amplitude stim-
ulation to maintain sphincter contraction without
concomitant detrusor contraction (Tanagho and
Schmidt, 1988). As contractions of the detrusor
are inhibited by contractions of the sphincter, it
was also noted that stimulation causing sphincter
contraction could inhibit detrusor activity leading
to improvements in continence in a range of ne-
urogenic and non-neurogenic bladder conditions
(Tanagho and Schmidt, 1988; Tanagho et al.,
1989). This neuromodulatory aspect of these im-
plants was pursued and techniques for accessing
175
the sacral nerves through the sacral foramina were
developed (Schmidt et al., 1990) making the im-
plant procedure faster and less invasive than spinal
implantation of extradural electrodes. Since this
time, sacral nerve neuromodulation for treating
non-neurogenic bladder dysfunction including in-
continence, urgency–frequency and urinary reten-
tion have been well studied (Bosch and Groen,
2000; Siegel et al., 2000). A sacral nerve stimulator
based on this work has been commercialized
by Medtronic as the InterStims (Medtronic,
Minneapolis, MN, USA) and implanted in more
than 10,000 people (Rijkhoff, 2004b). The InterS-
tim consists of a battery-powered implantable
stimulator connected to a single quadripolar elec-
trode usually inserted through the S3 sacral for-
amen to lie next to the S3 spinal nerve (see Fig. 2K
and Fig. 7). Sacral neuromodulation is only effec-
tive in a subset of patients with the above-men-
tioned bladder dysfunctions, so all patients are
initially evaluated with a percutaneous electrode
connected to an external stimulator to assess their
response to this treatment before permanent im-
plantation (Bosch and Groen, 2000; Siegel et al.,
2000). Other reviews provide more detail about
this technology and its history (Schmidt, 1988;
Groen and Bosch, 2001; Jezernik et al., 2002; Van
Kerrebroeck, 2002; Middleton and Keast, 2004;
Rijkhoff, 2004b; van Balken et al., 2004).
With the effectiveness of sacral nerve neuromod-
ulation, a number of investigators have examined
this modality for treating neurogenic detrusor over-
activity in small-scale studies of subjects with spinal
cord injury. Improvements in incontinence and in-
creases in maximal cystometric capacity have been
demonstrated in subjects with incomplete spinal
cord injury (Ishigooka et al., 1998; Chartier-Kastler
et al., 2001; Hohenfellner et al., 2001). However, S3
sacral nerve neuromodulation in subjects with
complete spinal cord injury has been generally less
effective (Chartier-Kastler et al., 2001) or had no
effect at all (Hohenfellner et al., 2001; Schurch
et al., 2003) leading to the suggestion that intact
spinobulbospinal pathways contribute to the suc-
cess of sacral neuromodulation (Schurch et al.,
2003). On the other hand, stimulation of the mixed
S2 root extradurally using the Finetech–Brindley
stimulator without posterior rhizotomy, has
176
Fig. 7. Insertion of the quadripolar electrode through the S3
sacral foramen to lie next to the S3 nerve. This electrode in-
sertion can be performed percutaneously to test the acute re-
sponse of a patient to neuromodulation. Reprinted with the
permission of Medtronic, Inc. r 2005.
successfully suppressed hyperreflexive bladder con-
tractions in people with complete spinal cord injury
and neurogenic detrusor overactivity (Kirkham
et al., 2002). The differing results observed using
the S2 and S3 spinal nerves for neuromodulation
in subjects with complete spinal cord injury may
indicate a fundamental difference in the neural
pathways being excited in these two cases: namely
that S3 neuromodulation has a larger supraspinal
component than S2 neuromodulation.
These results suggest that the success of neuro-
modulatory techniques for people with spinal cord
injury may depend on the completeness of the in-
jury as well as the specific location of the elec-
trodes. Given the commercial availability of the
InterStim and the straightforward and well-estab-
lished evaluation techniques, sacral neuromodula-
tion will likely continue to be investigated for its
usefulness in treating cord-injured people with ne-
urogenic detrusor overactivity.
NeoPraxis Praxis/Minax
In 1983, a multi-channel implantable neuropros-
thesis targeting sacral roots and nerves, based up-
on existing cochlear stimulation technology from
Cochlear Ltd. (Lane Cove, NSW, Australia), was
proposed (reviewed in Davis et al., 2001). After
successful implantation of a 22 channel system for
standing (Davis et al., 1994, 1997) a functional
electrical stimulator (FES 24-A; NeoPraxis Pty.
Ltd., Lane Cove, NSW, Australia) that included
electrodes intended to restore bladder function was
implanted in a thoracic (T)10 paraplegic (Davis
et al., 1999). Three pairs of electrodes intended to
elicit bladder contractions were inserted bilaterally
through the sacral foramina targeting the S2–S4
spinal nerves. An epidural electrode, intended to
suppress detrusor hyperreflexia, was implanted on
the conus medullaris to obviate the need for dorsal
rhizotomies (Davis et al., 2001). Increases in blad-
der pressure and some voiding was reported with
intermittent stimulation, but few data were pre-
sented (Davis et al., 1999, 2001). Two additional
subjects were implanted with similar systems, but
bladder contractions could not be elicited (Smith
et al., 2002; Benda et al., 2003). However, in these
subjects, external urethral sphincter activity caused
by low-frequency stimulation of the sacral nerves
was reduced by selective high-frequency blockade
of the large somatic fibers (Shaker et al., 1998;
Benda et al., 2003). A conference report intro-
duced the Minax system, a subset of the Praxis
FES 24 stimulator specifically for bladder control
(Houdayer et al., 2002), but no further reports
have been published on either the Minax or Praxis
systems and the company NeoPraxis appears to
have become inactive.
Stimulation of the spinal cord
During the late 1960s and early 1970s, experiments
were performed by Nashold and Friedman in an-
imals (Nashold et al., 1971) and humans (Nashold
et al., 1972) to test the efficacy of deep stimulation
of the spinal cord (see Fig. 2L) to restore micturit-
ion after spinal cord injury. These experiments
were conducted based on earlier studies demon-
strating that electrical stimulation of the cut sacral
spinal cord could elicit bladder contractions
(Stewart, 1899). It was proposed that electrical
stimulation of the presumed sacral micturition
center (Kuru, 1965) would be capable of causing
coordinated micturition after spinal cord injury.
Two electrodes, forming a bipolar pair, were im-
planted in the intermediolateral gray matter of the

sacral spinal cord (Nashold et al., 1971; Friedman
et al., 1972). Optimal rostrocaudal electrode place-
ment was determined by monitoring the change in
bladder pressure during stimulation of the dorsal
surface of the sacral spinal cord. The location
causing the greatest increase in bladder pressure
was selected as the location for implantation of the
penetrating electrodes. Friedman et al. (1972) re-
ported that six of 11 animals with intact spinal
cords and five of nine animals with transected spi-
nal cords voided during acute experimentation.
Additionally, in chronic experiments, stimulation
in five of six animals with intact spinal cords and
five of 10 animals with transected spinal cords
produced voiding. In animals where voiding did
not occur, bladder pressures elicited by electrical
stimulation were low, possibly indicating poor
electrode placement. While some reduction in pel-
vic floor electromyogram, suggesting coordinated
sphincter inhibition, was occasionally noted, stim-
ulus spread resulting in activation of the sphincter
was also observed (Friedman et al., 1972). Studies
conducted by another group, utilizing a variety of
electrode designs and stimulation parameters, con-
cluded that sphincter motoneurons in the spinal
cord were always stimulated with the sacral para-
sympathetic nucleus but that post-stimulus void-
ing could be used successfully to empty the bladder
(Jonas et al., 1975; Jonas and Tanagho, 1975).
Based on these results, 27 patients, 17 in the
USA, nine in France and one in Sweden, were
implanted with penetrating spinal cord electrodes
(see Fig. 8) beginning in 1970 (Nashold et al.,
1981). These patients are believed to be the only
people in the world implanted with electrodes tar-
geting intraspinal structures. The implanted device
consisted of two electrodes, connected to a subcu-
taneous radiofrequency receiver that could be ac-
tivated by a handheld stimulator placed over the
skin. The report on the initial four patients (three
male, one female) with electrodes implanted at the
S1 level eventually showed good voiding in three
patients (Nashold et al., 1972). Unlike some of the
animal work, electrical stimulation did not gener-
ally cause concomitant relaxation of the urethra in
these patients. Rather, a spastic external urethral
sphincter prevented micturition in the male subjects
even though large increases in bladder pressure
177
Fig. 8. Diagram showing the location of the penetrating elec-
trodes implanted into the spinal cord. Reprinted with permis-
sion from Nashold et al. (1972).
were achieved. To overcome this, partial transure-
thral sphincterotomies were performed that al-
lowed voiding in two of the male patients without
causing incontinence. The female subject was able
to void with an intact sphincter. This general pat-
tern was reported in most patients in a 10-year
review of the technique (Nashold et al., 1981).
Good voiding was achieved in 10 of 13 females
subjects, but in only 5 of 14 male subjects. Clinical
results from Duke University, where the technique
was developed, reported success in six of seven fe-
male subjects and four of seven male subjects
(Nashold et al., 1981). Of the male subjects, two
voided successfully after bladder neck resections or
partial sphincterotomies but two others failed to
void even after these procedures were performed.
However, two male subjects did achieve concom-
itant relaxation of the urethra leading to complete
bladder evacuation without sphincterotomy, sug-
gesting that at least in some patients, spinal cord
stimulation can elicit coordinated voiding (Grimes
et al., 1975). Ultimately, 60% of the subjects ob-
tained clinically good micturition with low residual
volumes, reductions in urinary tract infections,
increases in bladder capacity and freedom from
catheterization. Reductions in spasticity, as well
as erections in some male patients, and defecation
in some female patients were reported. However,
178
autonomic and motor responses including sweat-
ing and lower limb movement often accompanied
stimulation of the spinal cord (Nashold et al.,
1981).
Despite these reasonably good clinical results,
no further implants using this procedure were per-
formed. There were several reasons for the aban-
donment of this approach. One reason is that the
procedure, which involves highly invasive surgery,
was unsuccessful in 40% of the subjects (Nashold
et al., 1981) presumably due to ineffective elec-
trode placement. Since neither coordinated mi-
cturition nor selective stimulation of the bladder
were achieved in most patients, this procedure
offered no advantages over Brindley’s sacral ante-
rior root stimulator system (1982) but added the
complication of a more unpredictable electrode
placement compared to sacral root electrode im-
plantation. The inability to achieve stimulation of
the bladder without concomitant sphincter activity
was likely due to stimulus spread and the close
proximity of the sacral parasympathetic nucleus
and Onuf’s nucleus (Kuru, 1965; de Araujo et al.,
1982).
Future devices for electrical control of the bladder
Modifications of sacral root stimulators
The Finetech–Brindley sacral anterior root
stimulator has proven to be the only commercial-
ly successful electrical stimulation device to restore
voiding in people with spinal cord injury. Despite
its proven efficacy (Van Kerrebroeck et al., 1993;
Brindley, 1994) and low risk of complications and
technical failures (Brindley, 1995), two issues exist,
that if overcome, could help improve the function
and acceptance of this device. The posterior
rhizotomy that is performed in conjunction with
the implantation of the Finetech–Brindley sacral
anterior root stimulator eliminates neurogenic de-
trusor overactivity, detrusor–sphincter dyssynergia
and autonomic dysreflexia triggered by bladder
afferents, but irreversibly eliminates reflex erec-
tions, reflex defecation, reflex micturition and any
remaining perineal sensation. The second issue is
that both detrusor and sphincter efferent fibers are
activated during stimulation, resulting in the func-
tional, but non-physiological, post-stimulus void-
ing pattern associated with this implant. Rather
than abandoning stimulation of sacral roots be-
cause of these limitations however, several tech-
niques are being developed to address these issues.
Three promising techniques are discussed here.
Additional techniques to reduce or eliminate stim-
ulation-induced sphincter contractions are re-
viewed by Rijkhoff et al. (1997b).
Sacral posterior and anterior root stimulation
One of the primary benefits of posterior rhizotomy
in spinal cord injury is the abolition of neurogenic
detrusor overactivity allowing low-pressure stor-
age of urine and an increase in bladder capacity.
Since inhibition of reflexive bladder contractions
has been demonstrated in some people with spinal
cord injury using sacral nerve neuromodulation
techniques (Ishigooka et al., 1998; Chartier-
Kastler et al., 2001; Hohenfellner et al., 2001), ne-
uromodulation of sacral roots was investigated in
five subjects with complete spinal cord injury and
neurogenic detrusor overactivity who underwent
implantation of a Finetech–Brindley system with-
out the usual posterior rhizotomy (Kirkham et al.,
2002). Electrodes were implanted extradurally on
the mixed sacral roots in four subjects and in-
tradurally on anterior, posterior (see Fig. 2J) and
mixed roots in the remaining subject. Since pos-
terior rhizotomy was not performed, this system
was referred to as a ‘‘Sacral Posterior and Anterior
Root Stimulator’’ (Kirkham et al., 2002). In the
three subjects that exhibited neurogenic detrusor
overactivity after implantation, stimulation with
small pulse widths successfully inhibited hyperre-
flexive bladder contractions and increased bladder
capacity to a level similar to that obtained with
anticholinergic medication. However, intermittent
stimulation at larger pulse widths to induce void-
ing was unsuccessful, in spite of large increases in
bladder pressure, because of detrusor–sphincter
dyssynergia between stimulation periods. Less
than 50% of the bladder volume was voided.
The sacral posterior and anterior root stimula-
tor system is an important next step in the devel-
opment of sacral root stimulation in that

neurogenic detrusor overactivity, previously
abolished by posterior rhizotomy can be eliminat-
ed by posterior root neuromodulation. However,
detrusor–sphincter dyssynergia, also previously
abolished by posterior rhizotomy, prevents blad-
der voiding in this system. Clinical use of the sacral
posterior and anterior root stimulator system
will require additional techniques to inhibit
detrusor–sphincter dyssynergia without posterior
rhizotomy to allow effective voiding.
Selective anodal block
Anodal blocking offers a method to block action
potential propagation in large fibers, effectively
allowing selective stimulation of the small para-
sympathetic fibers innervating the bladder. Selec-
tive stimulation of these fibers would reduce the
concomitant contraction of the sphincter and low-
er limbs that presently occurs with the Fine-
tech–Brindley sacral anterior root stimulator. This
could restore a more physiological voiding pattern.
Anodal blocking takes advantage of the fact that
axons are hyperpolarized under the anode, reduc-
ing their excitability. Since larger diameter axons
have a lower threshold for electrical activation,
they can be selectively hyperpolarized (Accornero
et al., 1977; Rijkhoff et al., 1994a). Stimulation at
the cathode excites both large somatic fibers and
the smaller parasympathetic fibers, but action po-
tential propagation in the somatic fibers is blocked
at the hyperpolarized portion of membrane, al-
lowing selective transmission in the parasympa-
thetic fibers (see Fig. 9). Brindley and Craggs
(1980) successfully tested this technique in animals
to achieve selective parasympathetic fiber stimula-
tion with a sacral anterior root stimulator, but it
did not work well enough in humans for regular
use (Brindley et al., 1982). More recently, anodal
blocking has been examined in modeling studies to
determine stimulation parameters (Fang and
Mortimer, 1991; Rijkhoff et al., 1994a). Addition-
ally, both animal studies (Fang and Mortimer,
1991; Koldewijn et al., 1994; Rijkhoff et al., 1994b;
Grunewald et al., 1998) and intraoperative hu-
mans studies (Rijkhoff et al., 1997a, 1998) of
anodal blocking have demonstrated large decreas-
es in sphincter and leg activity while producing
bladder contractions.
179
Fig. 9. Electrical current under the anode hyperpolarizes the
membrane of large fibers at lower stimulation amplitudes than
smaller fibers. This section of hyperpolarized membrane blocks
the transmission of action potentials generated at the cathode
preventing activation of the sphincter while allowing activation
of the bladder. Action potentials propagate and are blocked in
the large fibers in both directions. The tripolar cuff design
minimizes current spread and I1 and I2 are the two independent
current sources sharing a common cathode. Reprinted with
permission from Rijkhoff et al. (1997a).
Several issues remain with anodal blocking
however. Anodal blocking waveforms being ex-
amined currently are monophasic and require
pulse widths of approximately 600 ms in humans
(Rijkhoff et al., 1998). Depending on the currents
required, long duration pulses can lead to irre-
versible electrochemical reactions at the electrodes
and eventual nerve damage (McCreery et al.,
1990). However, methods to increase safety by re-
ducing the charge per phase used to achieve anodal
block are being examined (Vuckovic and Rijkhoff,
2004). Additionally, implantable stimulators capa-
ble of producing the waveforms generally required
for anodal blocking do not exist, although they are
under development (Bugbee et al., 2001; Rijkhoff,
2004a). Although anodal blocking allows selective
activation of the bladder, posterior rhizotomies
may still be required to allow voiding. In one an-
imal study, anodal blocking without posterior
rhizotomy never resulted in voiding because of
reflexes that increased intraurethral pressure. In
the same experiment however, continuous voiding
was achieved once a posterior rhizotomy was per-
formed (Grunewald et al., 1998). In another ani-
mal study, complete voiding was achieved using
anodal blocking without posterior rhizotomy
(Koldewijn et al., 1994). Even if posterior rhizo-
tomy is required to allow voiding when using
anodal blocking, this technique may allow a more
physiological continuous voiding pattern and may
180

reduce unwanted leg movements that currently

occur with sacral anterior root stimulators.

High-frequency blockade
Another technique to prevent external urethral
sphincter activation with sacral anterior root
stimulators uses high-frequency stimulation to
block action potential propagation in somatic
fibers. Sawan, Elhilali and colleagues have suc-
cessfully stimulated mixed sacral roots with high-
frequency (600 Hz), low-amplitude pulses to block
urethral sphincter efferent activity while superim-
posing low-frequency, high-amplitude pulses to
activate parasympathetic preganglionic efferents
that in turn generate bladder contractions
(Shaker et al., 1998; Abdel-Gawad et al., 2001) (see
Fig. 10). High-frequency stimulation has been
shown to block action potential propagation by
hyperpolarizing axons and maintaining them in
their refractory period (Solomonow et al., 1983).
High-frequency, low-amplitude stimulation hyper-
polarizes large fibers but the stimulation amplitude
is not high enough to affect the smaller parasym-
pathetic fibers. In a study of 12 chronically
implanted dogs, all voided with o20% residual
urine and seven of the group voided with o10%
residual urine (Abdel-Gawad et al., 2001). This
system represents a potential advantage over the Fig. 10. High-frequency block of large fibers allowing selective
current Finetech–Brindley sacral anterior root activation of the bladder by sacral ventral root stimulation.
stimulator where the bladder and sphincter contract Low-frequency pulses lead to activation of large and small
simultaneously and voiding occurs post-stimulus. fibers, while high-frequency pulses block action potential prop-
agation in large fibers selectively. (A) High-frequency, low-am-
This device, including a proposed method of plitude pulses superimposed on low-frequency, high-amplitude
neuromodulation to inhibit the hyperreflexive pulses. LFA – low-frequency amplitude, LFP – low-frequency
bladder by stimulation of the sacral nerves, has period, LFW – low-frequency pulse width, HFA – high-fre-
been patented (Sawan and Elhilali, 2002). While quency amplitude, HFP – high-frequency period, HFW – high-
this group has not published results regarding the frequency pulse width. (B) Difference between low-frequency
only stimulation and combined low- and high-frequency stim-
proposed neuromodulatory action of their device, ulation. Intraurethral pressure and sphincter EMG were reduced
proof of principle has been demonstrated in cord- while bladder pressure was maintained when the selective stim-
injured people with a similar device (Kirkham ulation (high-frequency, low-amplitude) waveform was utilized.
et al., 2002). Neuromodulation could remove the Reprinted from Boyer et al. (2000) with permission from the
necessity for posterior rhizotomy in sacral root IEEE.
stimulators, especially since successful voiding
without posterior rhizotomy has been achieved et al., 2003). Although no reports in humans exist
with this stimulation paradigm, at least in animals from Sawan and Elhilali’s group, Victhom Human
(Abdel-Gawad et al., 2001). A brief report from Bionics (Saint-Augustin-de-Desmaures, Quebec,
another group using the same stimulation paradigm Canada) has licensed the technology and is con-
suggests that it is also effective in humans (Benda tinuing development of the device.

Intraspinal microstimulation
Since Nashold and Friedman’s original work
on spinal cord stimulation to evoke micturition
(Nashold et al., 1971; Nashold et al., 1972), interest
has persisted in this technique. In these experi-
ments, electrodes were on the order of 0.3–0.4 mm
in diameter with 0.5–1.0 mm long exposed tips
(Nashold et al., 1971; Jonas et al., 1975) leading to
geometric electrode surface areas of 0.5–1.4 mm2.
Stimulus spread from these comparatively large
electrodes between the adjacent sacral parasympa-
thetic nucleus and Onuf’s nucleus (Kuru, 1965;
de Araujo et al., 1982) was the likely cause of ob-
served concomitant bladder and sphincter contrac-
tions (Nashold et al., 1972). Electrode development
has improved the ability to selectively stimulate
specific regions within the spinal cord (Prochazka
et al., 1976; Mushahwar et al., 2000; McCreery
et al., 2004). Currently, electrode arrays for chronic
intraspinal microstimulation use microwire elec-
trodes (see Fig. 11A) or silicon substrate micro-
electrodes manufactured using photolithographic
processes (see Fig. 11B). Microwire-based elec-
trodes use iridium and platinum–iridium alloy
wires 20–30 mm in diameter with 20–100 mm long
exposed tips resulting in geometric electrode sur-
face areas of 1600–10,000 mm2 (Mushahwar et al.,
2000). Silicon substrate microelectrodes used for
chronic intraspinal microstimulation implants can
have multiple stimulation sites at various depths
per penetrating shank with electrode surface areas
around 2000 mm2 (McCreery et al., 2004). These
electrode surface areas are 50–300 times smaller
than the smallest electrodes used in the first spinal
cord stimulation experiments (Nashold et al., 1971;
Jonas et al., 1975) and allow selective stimulation
of the sacral parasympathetic nucleus without con-
comitant sphincter activation (Carter et al., 1995;
Grill et al., 1999). Although electrode arrays can
cause inflammatory reactions, glial scarring and
neural death around the implantation site, this can
be minimized (McCreery et al., 2004) and would
presumably be just as safe or safer in long-term
use as the larger electrodes implanted in humans
(Nashold et al., 1981).
If intraspinal microstimulation is to be clinically
useful, clear advantages over sacral root stimulation
181
must be offered, especially since an intraspinal
microstimulation implant would likely be at least as
difficult to perform as a sacral anterior root stimula-
tor implant. One advantage of intraspinal micro-
stimulation is that bladder contractions can be
evoked without concomitant sphincter contractions
(Carter et al., 1995; Grill et al., 1999). However,
more importantly, intraspinal microstimulation
allows the possibility of activating sacral intern-
euronal networks that produce coordinated mi-
cturition or some part thereof (Nashold et al., 1971;
Grimes et al., 1975; Grill, 2000).
Networks of interneurons responding to pelvic
and pudendal afferents and receiving projections
from the pontine micturition center exist in various
regions around the central canal, in the dorsal gray
commissure and in the intermediolateral cell column
of the sacral spinal cord (reviewed in de Groat et al.,
1996; Shefchyk, 2001) and are active during mi-
cturition (Grill et al., 1998; Buss and Shefchyk,
2003). One group of interneurons, located in the
dorsal gray commissure, are of particular interest in
relation to inhibition of the external urethral sphinc-
ter as the interneurons contain inhibitory neuro-
transmitter, receive direct projections from the
pontine micturition center and are believed to
project to Onuf’s nucleus (Blok et al., 1997; Sie
et al., 2001). Since Onuf’s nucleus does not receive
inhibitory projections from supraspinal centers, in-
hibitory interneurons in the dorsal gray commissure
may mediate voluntary relaxation of the sphincter
(Blok, 2002). This view is supported by the finding
that electrical stimulation in the dorsal gray com-
missure produced active and sustained decreases in
urethral pressure in spinally intact cats (Blok et al.,
1998; McCreery et al., 2004). Additionally, some
voiding can occur when electrodes around the cen-
tral canal are stimulated (Grill et al., 1999). The
ability to actively inhibit urethral activity is not cur-
rently possible with sacral root stimulation. These
results demonstrate the potential of intraspinal mi-
crostimulation to achieve coordinated micturition
through bladder excitation and sphincter inhibition.
Current research on intraspinal microstimulation
is therefore focused on simultaneous stimulation of
the sacral parasympathetic nucleus to produce blad-
der contractions and the dorsal gray commissure
to actively relax the urethra (see Fig. 11C). This
182

Fig. 11. Electrode designs and target locations (in the cat) for intraspinal microstimulation. (A) Microwire electrode array with
electrode tips placed in protective tubing. (B) Multi-site penetrating silicon electrode array. (C) Electrode targets in the cat sacral spinal
cord. Electrical stimulation in the dorsal gray commissure, which contains interneurons with inhibitory projections to sphincter
motoneurons, can elicit relaxation of the external urethral sphincter. Electrical stimulation of sacral parasympathetic nucleus, which
contains bladder preganglionic neurons, can elicit sustained increases in bladder pressure. Adapted from Prochazka et al. (2002a) and
McCreery et al. (2004).

approach has produced sustained high-pressure
bladder contractions, coordinated increases in blad-
der pressure and decreases in urethral pressure and
occasional incomplete voiding (Prochazka et al.,
2003b), but so far it has proven unreliable. One
possible reason is that in addition to the interneu-
rons in the dorsal gray commissure that inhibit
sphincter motoneurons, there are other interneurons
in this same region that have been shown to decrease
their firing rate during micturition, and may be part
of pathways with excitatory connections to sphincter
motoneurons (Buss and Shefchyk, 2003). If this is
the case, electrical stimulation in parts of the dorsal
gray commissure may, in fact, activate more excita-
tory interneurons than inhibitory interneurons, and
thereby cause contraction, rather than relaxation of
the external urethral sphincter. It is also unknown
whether intraspinal microstimulation can affect the
activity of the smooth muscle internal urethral
sphincter, which is the primary mechanism to main-
tain continence until the bladder is very full or at
high pressure. These results were obtained in an
awake animal, suggesting that previous results
in anesthetized animals may also hold true in the
absence of anesthesia. Additionally, electrodes tar-
geting the sacral parasympathetic nucleus produced
similar increases in bladder pressure using the same
stimulation parameters, before and after complete
spinal cord transection. However, bladder pressure
increases produced by stimulation in the sacral para-
sympathetic nucleus and urethral pressure decreases
produced by stimulation in the dorsal gray commis-
sure may not be sufficient to produce micturition. In
some cases, simultaneous intraurethral pressure re-
cordings in the vicinity of the external urethral
sphincter and bladder pressure recordings, have in-
dicated that voiding should occur (Prochazka et al.,
2003a) (see Fig. 12), but once the urethral pressure
catheter was removed to unblock the urethra, stim-
ulation through the same electrodes did not elicit
voiding. In this case, high pressure in the bladder
neck or distal urethra may have prevented voiding.
Urethral afferent stimulation
Electrical stimulation of afferent branches of the
pudendal nerve, specifically the dorsal penile

Fig. 12. Bladder and urethral pressure changes in response to
electrical stimulation through three electrodes: two targeting the
sacral parasympathetic nucleus and one targeting the dorsal
gray commissure. During stimulation, immediate increases in
bladder pressure and decreases in urethral pressure were
achieved. The pressures measured in the bladder and urethra
became essentially equal (within the calibration error of the
transducers) during stimulation indicating that voiding might
occur in the absence of the urethral catheter. However, once the
catheter was removed to unblock the urethra, stimulation
through the same electrodes did not induce voiding. Adapted
from Prochazka et al. (2003a).
nerve, has been shown to inhibit hyperreflexive
bladder contractions occurring after spinal cord
injury. However, electrical stimulation of urethral
afferents, also forming part of the pudendal nerve,
has been shown to elicit bladder contractions
as well as relaxation of the sphincter. These spinal
reflexes were first described by Barrington in
spinal cord-transected cats (Barrington, 1914,
1941), and more recently have been investigated
in cord-transected cats (Shefchyk and Buss, 1998;
Gustafson et al., 2003) and humans with spinal
cord injury (Gustafson et al., 2003, 2004). These
reflexes are presumed to facilitate voiding by pos-
itive feedback from afferents sensitive to urethral
dilation (Shafik et al., 2003a, b). In humans, ure-
thral afferents were electrically stimulated using a
catheter-mounted electrode (Gustafson et al.,
2003) passed into the urethra (see Fig. 2M). In
subjects with complete spinal cord injury, bladder
contractions reaching 70 cm H2O as well as void-
ing could be achieved if the bladder volume was
above a threshold value (Gustafson et al., 2004).
Below this threshold, bladder contractions could
not be generated and voiding was therefore
183
incomplete. Despite the fact that only incomplete
voiding has been demonstrated, this stimulation
technique is very intriguing from the perspective of
neuroprosthesis device development. So far, it is
the only way in which bladder contractions can be
generated without invasive implantations of elec-
trodes targeting the bladder, sacral roots or spinal
cord itself. However, even more interesting is the
apparent ability of urethral afferent stimulation to
evoke coordinated micturition after complete spi-
nal cord injury (Shefchyk and Buss, 1998).
Microstimulators
Implanted neuroprostheses discussed in this review
have all used the same basic set of components. An
implanted stimulator, either battery powered or
inductively coupled to an external power source, is
implanted subcutaneously in a convenient location
such as the abdominal or chest region and long
leads connect the electrodes to the stimulator. This
design can lead to time-consuming surgical proce-
dures as well as technical failures including lead
breakage and connector failure (Brindley, 1995).
Microstimulators such as the BION (Advanced
Bionics, Valencia, CA) provide an alternative ap-
proach. BIONs are self-contained, injectable mi-
crostimulators that are programmed and powered
by inductive coupling. A single external transmit-
ter coil can communicate with up to 256 BIONs,
each of which can deliver current controlled pulses
with a pulse width range of 4–512 ms and ampli-
tude range of 0–30 mA up to frequencies of 50
pulses per second in a package 2 mm in diameter
and 16 mm long (Loeb et al., 2001) (see Fig. 13).
A more recent version of the BION includes a
lithium-ion battery to power the device during
normal use, but is programmed and recharged us-
ing the external coil. This device is larger (3.3 mm
diameter, 27 mm long) (Groen et al., 2004), but
allows subjects to be free of the external coil and
associated hardware during daily activities, at least
in applications not requiring phasic control.
Two studies have evaluated the use of BIONs
for treating bladder dysfunction using the induc-
tively powered (Buller et al., 2002) and battery-
powered (Groen et al., 2004) systems. Although
184
Fig. 13. A glass encapsulated BION. Adapted from Loeb et al.
(2001).
final reports have yet to be published, both studies
have shown positive results using pudendal nerve
(see Fig. 2N) stimulation to treat overactive blad-
der incontinence. BIONs have not been clinically
tested in spinal cord injury patients for bladder
control, but a number of stimulation techniques
presented in this review could be adapted to use
BIONs.
MiniatURO
The miniatURO (Biocontrol Medical Ltd.,
Yahud, Israel) is a closed-loop implantable
electrical stimulation system for treating mixed
urinary incontinence. The device consists of an
intra-abdominal pressure sensor, pelvic floor stim-
ulation electrode and a stimulator with integrated
control system. Stimulation is triggered by in-
creases in intra-abdominal pressure. This device
has been tested in short-term settings where ab-
dominal pressure was measured rectally, stimulat-
ing electrodes were introduced percutaneously and
the stimulator was worn externally. All subjects in
a group of 16 with stress incontinence responded
positively and became dry or had reduced episodes
of incontinence (Nissenkorn et al., 2004). While
this device has not been tested in cord-injured
people, there is reason to believe that this kind of
device might be effective in the spinal cord injury
population (Fjorback et al., 2003).
Transcutaneous magnetic stimulation
Transcutaneous magnetic stimulation of the nerv-
ous system is a noninvasive method of activating
neural tissue and has therefore become a useful
technique for human experimentation and clinical
diagnostics. Two studies have examined the ability
of magnetic stimulation to exert control over lower
urinary tract function in spinal cord injury subjects
but have arrived at very different results (Lin et al.,
1997; Bycroft et al., 2004). In the first study it was
concluded that repetitive magnetic stimulation
(15–30 Hz) of the sacral nerves, achieved by plac-
ing the stimulation coil over the sacrum (see Fig.
2O), caused direct activation of parasympathetic
preganglionic neurons innervating the bladder
leading to increases in bladder pressure. Fatigue
of the external urethral sphincter and intermittent
stimulation to allow post-stimulus voiding were
the two methods postulated to explain the voiding
that occurred, even though many subjects had re-
ceived sphincterotomies (Lin et al., 1997). The
more recent study suggests that magnetic stimula-
tion of the sacral nerves causes direct inhibition of
bladder contractions in cord-injured people with
neurogenic detrusor overactivity. It also stated
that the previously described bladder contractions
were rebounds occurring at the end of stimulation
as a result of releasing the detrusor from direct
inhibition (Bycroft et al., 2004).
Whatever the effect of magnetic stimulation on
bladder function in cord-injured people, this tech-
nique holds promise for a device-based therapy for
bladder control. Current magnetic stimulation
systems are large and would not be suitable for
chronic use. However, the potential benefits of
this completely noninvasive technique, and its
demonstrated efficacy for treating neurogenic de-
trusor overactivity and/or eliciting partial voiding,
deserve further investigation.
Successes and failures of devices for bladder control
A wide variety of techniques and devices have been
developed to manage the significant problems as-
sociated with lower urinary tract dysfunction after
spinal cord injury. While devices such as the Fine-
tech–Brindley sacral anterior root stimulator can
truly be said to exert control over the lower uri-
nary tract, other devices, including catheters mere-
ly manage the symptoms associated with lower
urinary tract dysfunction after spinal cord injury.
However, the ultimate goal of any treatment
modality, including pharmacological and surgical

Table 1. Summary of mechanical devices used for lower urinary tract management after spinal cord injury
Device Efficacy Advantages
Catheter Very good Simple, inexpensive
Artificial sphincter Mixed Proven design
Urethral stent Good Reversible alternative to sphincterotomy
Intraurethral pump Good Simple implant
methods, is to create a bladder capable of storing
large volumes of urine at low pressure, while pre-
venting incontinent episodes and allowing periodic
evacuation of that urine at low pressure. If a
treatment achieves its intended function, then the
merits of one technique versus another depend on
issues such as adverse side effects, procedural re-
versibility, device cost, ease of use and ease of im-
plantation.
Device efficacy, advantages and disadvantages
Two classes of devices have been described in this
review: mechanical devices and electrical stimula-
tion devices. The mechanical devices discussed
range from the very simple to the very complex
and address voiding dysfunction, incontinence and
detrusor–sphincter dyssynergia. A brief summary
of the efficacy, main advantages and disadvantages
and current status of these devices is presented in
Table 1. Several general comments about the use
of mechanical devices can be made. Catheters and
the In-Flow intraurethral pump do not require
implantation. They are also commercially availa-
ble, allowing easy adoption by clinicians. Artificial
urethral sphincters and urethral stents do require
surgery, but this is relatively simple. However,
none of these devices offer the potential to restore
lower urinary tract function to the pre-injury state.
There are also significant side effects due to the
chronic presence of foreign materials in the lower
urinary tract. Finally, no mechanical device is
able to affect neurogenic detrusor overactivity, so
pharmacological or surgical treatments must be
used to establish a high-volume, low-pressure
bladder.
Electrical stimulation devices, while not cura-
tive, offer the unique potential to restore normal
function to the lower urinary tract after spinal
185
Disadvantages Current status
Urinary tract infections, inconvenient Clinical
Infection, urethral erosion Limited use
Not for long-term implantation Some clinical
Regular replacement, discomfort Investigational
cord injury. Many techniques have been presented
in this review that vary significantly in their ability
to establish a high-volume, low-pressure bladder
and to produce low-pressure voiding. Improving
continence, storage volume and storage pressure
by inhibiting hyperreflexive bladder contractions
has been the specific aim of a number of devices.
A brief summary of the efficacy, main advantages
and disadvantages and current status of these
devices is presented in Table 2, while a similar
summary for devices intended to evoke bladder
emptying is presented in Table 3. Of the electrical
stimulation techniques intended to inhibit hyper-
reflexive bladder contractions, both dorsal penile
nerve stimulation and tibial nerve stimulation have
shown promising results. However, these methods
require additional study with the use of more
practical devices that are suitable for chronic
use. Bladder wall, pelvic nerve and spinal
cord stimulation are all methods to elicit voiding
that have been tested in humans with spinal cord
injury and subsequently abandoned. However,
bladder wall stimulation and spinal cord stimula-
tion continue to be explored in the laboratory
setting.
Clinical success of devices in managing dysfunction
after spinal cord injury
Catheters and urethral stents, as well as dorsal pe-
nile nerve, tibial nerve, spinal cord and sacral root
stimulation have all been shown to be effective in
managing various aspects of lower urinary tract
dysfunction after spinal cord injury. However, the
success of these devices from a clinical perspective
varies significantly. The clinical success of catheters
remains foremost. Of the other effective techniques,
only the Finetech–Brindley sacral anterior root
stimulator could be considered a proven clinical
186

Table 2. Summary of electrical stimulation devices used to control continence after spinal cord injury

Device Efficacy Advantages Disadvantages Current status

Pelvic floor maximal Mixed Long-lasting Physical and Limited use

functional electrical improvements psychological
stimulation discomfort
Inconsistent results
Thigh stimulation Mixed Non-invasive No predictors for Limited use
success
Long-lasting Time-consuming
improvements treatment
Dorsal penile nerve Promising Non-invasive Not well studied Investigational
surface stimulation
Tibial nerve stimulation Promising Simple treatment Permanent device Investigational
procedure required
Not well studied
Sacral neuro- Mixed Minimally invasive May not work in Investigational
modulation complete injuries
Proven design and Inconsistent results, Not
implant procedure well studied
Commercially available

success with over 2500 systems implanted (Rijkhoff,
2004b). This system is successful because it is the
only device that reliably evokes complete bladder
evacuation, significantly reducing the complications
associated with catheterization. When combined
with sacral posterior rhizotomy, as it nearly always
is, the Finetech–Brindley sacral anterior root
stimulator provides a complete system for lower
urinary tract control after spinal cord injury. How-
ever, the consequences of irreversible rhizotomy,
the technically demanding implant procedure and
commercial and regulatory issues have limited the
availability of this device as well as the acceptance
of it by patients and clinicians. With this in mind,
given the large number of people that could benefit
from such a device, and a generally increasing clin-
ical acceptance of neuroprostheses in general, the
Finetech–Brindley sacral anterior root stimulator
has had a limited impact on the spinal cord injury
population throughout most of the world.
The future of devices for bladder control
Progress is being made on regeneration of the spi-
nal cord, but a complete biological cure for spinal
cord injury is unlikely to be developed in the near
future (Fawcett, 2002). In light of this, and the fact
that eventual regenerative therapies will likely be
combined with neuroprostheses to maximize func-
tional recovery (Prochazka et al., 2002b), new and
improved devices are required to restore control of
the lower urinary tract after spinal cord injury.
The feasibility and efficacy of a number of me-
chanical and electrical devices for treating lower
urinary tract dysfunction has been clearly demon-
strated in spinal cord injury patients. However, no
clinical device can be said to have solved the
problem of bladder control as low-pressure phys-
iological voiding can not yet be produced and no
device has successfully incorporated methods to
produce both voiding and suppression of ne-
urogenic detrusor overactivity. While future im-
proved biomaterials will undoubtedly reduce some
of the side effects associated with the use of
mechanical devices (Beiko et al., 2004), these are
unlikely to achieve complete control over the lower
urinary tract. Electrical stimulation devices on the
other hand have demonstrated the ability to achieve
significant control over neurogenic detrusor
overactivity and voiding. Adaptations of the Fine-
tech–Brindley sacral anterior root stimulator, in-
cluding posterior root stimulation and anodal
blocking or high-frequency blockade of somatic fib-
ers, offer the possibility of inhibiting hyperreflexive
bladder contractions and producing physiological
 187

Table 3. Summary of electrical stimulation devices used to control voiding after spinal cord injury

Device Efficacy Advantage Disadvantages Current status

Intravesical electrical Mixed Long-lasting results Ineffective in many Limited use

stimulation patients
Can improve voluntary No predictors for
voiding in incomplete success
injuries
Time-consuming
treatment
Bladder wall Mixed Simple surgical Sphincter and pelvic Abandoned except for
stimulation approach floor contraction lower motoneuron
lesions
Electrode–bladder
interface failure
Pelvic nerve stimulation Poor Sphincter and pelvic Abandoned
floor contraction
Difficult surgical
approach
Nerve unsuitable for
stimulation
Sacral root stimulation Very good Voiding without Posterior rhizotomy Clinical
catheterization required
Low residual volume Demanding surgical
implant
Proven design and Unphysiological
implant procedure voiding
Long-term usage with
few failures
Commercially available
Spinal cord stimulation Good Voiding without Sphincter contraction Originally abandoned,
catheterization currently investigational
Low residual volume Sphincterotomy
required in males
Long-term usage Difficult electrode
placement 40% failure
rate

voiding without the currently requisite posterior
rhizotomy. Combining such methods may signifi-
cantly improve future electrical stimulation devices
for bladder control.
Other electrical stimulation techniques currently
under investigation also show promise. Intraspinal
microstimulation has the potential to utilize remain-
ing spinal cord networks and can achieve selective
stimulation of the bladder as well as active inhibi-
tion of the sphincter, albeit not reliably enough to
produce consistent voiding. Intraspinal microstim-
ulation research has yet to address the problem of
neurogenic detrusor overactivity and surgical im-
plantation of intraspinal microstimulation devices
in their current form would be even more difficult
than implantation of sacral root stimulators. Unless
simpler electrode configurations are developed, this
may limit clinical use of intraspinal microstimula-
tion even if it can be shown to be reliable and ef-
fective in animals. Stimulation of the dorsal penile
nerve and urethral afferents are among the most
interesting techniques currently being investigated
to restore control of the lower urinary tract from a
neuroprosthetic device development perspective.
They offer the potential of a minimally invasive
implant to suppress neurogenic detrusor overactiv-
ity and produce voiding, perhaps using micro-
stimulators. If these techniques are to have an
188
impact on the clinical management of lower urinary
tract dysfunction after spinal cord injury, they must
provide clear improvements in treatment over what
can be currently accomplished with catheters, phar-
macological therapies and surgical intervention.
An ideal device for controlling lower urinary
tract function after spinal cord injury can be pos-
tulated based on the material presented in this
review. Above all, the device must suppress ne-
urogenic detrusor overactivity and allow user-
controlled, continuous, low-pressure voiding with
a low residual volume. The device should not re-
quire additional pharmacological or surgical pro-
cedures to operate successfully and would ideally
require only minimally invasive surgery. Although
not discussed in this review, an implant to restore
lower urinary tract function should ultimately
provide means to restore bowel and sex function as
well. Finally, any procedure must be completely
reversible so that people are not committed to a
particular device for the rest of their lives and are
able to take advantage of future devices.
Conclusions
In this review we have discussed devices and tech-
niques aimed at restoring normal function to the
lower urinary tract after spinal cord injury. Some of
these devices and techniques have ultimately proven
to be unsuccessful for reasons such as insufficient
efficacy, unacceptable treatment procedures, side ef-
fects and technical failures. Other devices, most no-
tably catheters and sacral root stimulation, have
been successful. Over 20 years of experience with
sacral root stimulation to restore voiding clearly in-
dicates the advantages that people with spinal cord
injury can expect to achieve with neuroprostheses.
However, detrusor–sphincter dyssynergia must be
overcome before neuroprostheses can gain more
widespread use. Electrical stimulation techniques to
suppress neurogenic detrusor overactivity, allowing
efficient bladder filling and storage, have been dem-
onstrated in some cord-injured people. But relatively
few studies have been conducted and many details
regarding these techniques have yet to be elucidated.
These, or other similar techniques, must be success-
ful if a neuroprosthesis for complete control of lower
urinary tract function is to be developed. Perhaps the
most difficult and crucial problem is achieving re-
laxation of the sphincter to allow physiological void-
ing. It has been said that ‘‘The key to control of the
bladder lies in control of the sphincter’’ (Schmidt,
1986). While this comment was originally made in
reference to the ability of sphincter activity to mod-
ulate bladder contractility, it is equally true that the
ability to control the sphincter and eliminate un-
wanted activity remains the final piece of the puzzle
that must be put in place for bladder control neuro-
prostheses to be completely effective. A number of
techniques are being investigated to accomplish this,
but no single one has completely achieved this goal.
In conclusion, while many devices and techniques
have been tried and ultimately abandoned, the suc-
cesses clearly show the immense benefits that can be
achieved with the use of devices in the control of the
dysfunctional bladder
Acknowledgments
The authors would like to gratefully acknowledge
Dr. Jonathan Norton for his helpful comments in
the preparation of this article and Jan Kowalczewski
for preparing Fig. 2. The work on bladder control
devices in Arthur Prochazka’s laboratory is sup-
ported by NIH-NINDS contract N01-NS-2-2342.
R.G. is supported by the Alberta Heritage Foun-
dation for Medical Research.
References
Abdel-Gawad, M., Boyer, S., Sawan, M. and Elhilali, M.M.
(2001) Reduction of bladder outlet resistance by selective
stimulation of the ventral sacral root using high frequency
blockade: a chronic study in spinal cord transected dogs.
J. Urol., 166: 728–733.
Accornero, N., Bini, G., Lenzi, G.L. and Manfredi, M. (1977)
Selective activation of peripheral nerve fibre groups of
different diameter by triangular shaped stimulus pulses.
J. Physiol., 273: 539–560.
Amarenco, G., Ismael, S.S., Even-Schneider, A., Raibaut, P.,
Demaille-Wlodyka, S., Parratte, B. and Kerdraon, J. (2003)
Urodynamic effect of acute transcutaneous posterior tibial
nerve stimulation in overactive bladder. J. Urol., 169:
2210–2215.
Andersen, J.T. and Bradley, W.E. (1976) The syndrome of de-
trusor–sphincter dyssynergia. J. Urol., 116: 493–495.

Anderson, K.D. (2004) Targeting recovery: priorities of the
spinal cord-injured population. J. Neurotrauma, 21:
1371–1383.
Andrews, B.J. and Reynard, J.M. (2003) Transcutaneous pos-
terior tibial nerve stimulation for treatment of detrusor
hyperreflexia in spinal cord injury. J. Urol., 170: 926.
Barrington, F.J.F. (1914) The nervous mechanism of micturit-
ion. Q. J. Exp. Physiol., 8: 33–71.
Barrington, F.J.F. (1921) The relation of the hind-brain to mi-
cturition. Brain, 4: 23–53.
Barrington, F.J.F. (1925) The effect of lesions on hind- and
mid-brain on micturition in the cat. Q. J. Exp. Physiol., 15:
81–102.
Barrington, F.J.F. (1941) The component reflexes of micturit-
ion in the cat. Part III. Brain, 64: 239–243.
Beiko, D.T., Knudsen, B.E., Watterson, J.D., Cadieux, P.A.,
Reid, G. and Denstedt, J.D. (2004) Urinary tract biomate-
rials. J. Urol., 171: 2438–2444.
Benda, B.J., Houdayer, T., Creasey, G., Betz, R.R., Smith,
B.T., Johnston, T.E. and Davis, R. (2003) The Praxis FES
system and bladder/bowel management in patients with
spinal cord injury. In: Meadows P. (Ed.), IFESS Conference
Proceedings, Brisbane, Australia.
Blaivas, J.G., Sinha, H.P., Zayed, A.A. and Labib, K.B. (1981)
Detrusor–external sphincter dyssynergia: a detailed elect-
romyographic study. J. Urol., 125: 545–548.
Blok, B.F. (2002) Central pathways controlling micturition and
urinary continence. Urology, 59: 13–17.
Blok, B.F., de Weerd, H. and Holstege, G. (1997) The pontine
micturition center projects to sacral cord GABA immunore-
active neurons in the cat. Neurosci. Lett., 233: 109–112.
Blok, B.F., van Maarseveen, J.T. and Holstege, G. (1998) Elec-
trical stimulation of the sacral dorsal gray commissure evokes
relaxation of the external urethral sphincter in the cat. Ne-
urosci. Lett., 249: 68–70.
Bosch, J.L. and Groen, J. (2000) Sacral nerve neuromodulation
in the treatment of patients with refractory motor urge in-
continence: long-term results of a prospective longitudinal
study. J. Urol., 163: 1219–1222.
Boyce, W.H., Lathem, J.E. and Hunt, L.D. (1964) Research
related to the development of an artificial electrical stimula-
tor for the paralyzed human bladder: a review. J. Urol., 91:
41–51.
Boyer, S., Sawan, M., Abdel-Gawad, M., Robin, S. and
Elhilali, M.M. (2000) Implantable selective stimulator to im-
prove bladder voiding: design and chronic experiments in
dogs. IEEE Trans. Rehab. Eng., 8: 464–470.
Bradley, W.E., Chou, S.N. and French, L.A. (1963) Further
experience with the radio transmitter receiver unit for the
neurogenic bladder. J. Neurosurg., 20: 953–960.
Bradley, W.E., Wittmers, L.E., Chou, S.N. and French, L.A.
(1962) Use of a radio transmitter receiver unit for the treat-
ment of neurogenic bladder. A preliminary report. J. Ne-
urosurg., 19: 782–786.
Brindley, G.S. (1977) An implant to empty the bladder or
close the urethra. J. Neurol. Neurosurg. Psychiat., 40:
358–369.
189
Brindley, G.S. (1994) The first 500 patients with sacral anterior
root stimulator implants: general description. Paraplegia, 32:
795–805.
Brindley, G.S. (1995) The first 500 sacral anterior root stimula-
tors: implant failures and their repair. Paraplegia, 33: 5–9.
Brindley, G.S. and Craggs, M.D. (1980) A technique for anod-
ally blocking large nerve fibres through chronically implanted
electrodes. J. Neurol. Neurosurg. Psychiat., 43: 1083–1090.
Brindley, G.S., Polkey, C.E. and Rushton, D.N. (1982) Sacral
anterior root stimulators for bladder control in paraplegia.
Paraplegia, 20: 365–381.
Brindley, G.S., Polkey, C.E., Rushton, D.N. and Cardozo, L.
(1986) Sacral anterior root stimulators for bladder control in
paraplegia: the first 50 cases. J. Neurol. Neurosurg. Psychiat.,
49: 1104–1114.
Bugbee, M., Donaldson, N.N., Lickel, A., Rijkhoff, N.J. and
Taylor, J. (2001) An implant for chronic selective stimulation
of nerves. Med. Eng. Phys., 23: 29–36.
Buller, J., Cundiff, G., Noel, K., Van Rooyen, J., Leffler, K.,
Ellerkman, M. and Bent, A. (2002) RF BION(TM): an in-
jectable microstimulator for the treatment of overactive blad-
der disorders in adult females. Eur. Urol. Suppl., 1: 40.
Burghele, T. (1973) Electrostimulation of the neurogenic uri-
nary bladder. In: Lutzeyer W. and Melchior H. (Eds.), Uro-
dynamics: Upper and Lower Urinary Tract. Springer, Berlin,
pp. 319–322.
Buss, R.R. and Shefchyk, S.J. (2003) Sacral dorsal horn neurone
activity during micturition in the cat. J. Physiol., 551: 387–396.
Bycroft, J.A., Craggs, M.D., Sheriff, M., Knight, S. and Shah,
P.J. (2004) Does magnetic stimulation of sacral nerve roots
cause contraction or suppression of the bladder? Neurourol.
Urodyn., 23: 241–245.
Cardenas, D.D., Farrell-Roberts, L., Sipski, M.L. and Rubner,
D. (1995) Management of gastrointestinal, genitourinary,
and sexual function. In: Stover S.L., DeLisa J.A. and
Whiteneck G.G. (Eds.), Spinal Cord Injury: Clinical Out-
comes from the Model Systems. Aspen Publishers Inc., Gait-
hersburg, pp. 120–144.
Carter, R.R., McCreery, D.B., Woodford, B.J., Bullara, L.A.
and Agnew, W.F. (1995) Micturition control by microstim-
ulation of the sacral spinal cord of the cat: acute studies.
Rehab. Eng., IEEE Trans., 3: 206–214.
Chancellor, M.B., Bennett, C., Simoneau, A.R., Finocchiaro,
M.V., Kline, C., Bennett, J.K., Foote, J.E., Green, B.G.,
Martin, S.H., Killoran, R.W., Crewalk, J.A. and Rivas, D.A.
(1999a) Sphincteric stent versus external sphincterotomy in
spinal cord injured men: prospective randomized multicenter
trial. J. Urol., 161: 1893–1898.
Chancellor, M.B., Gajewski, J., Ackman, C.F., Appell, R.A.,
Bennett, J., Binard, J., Boone, T.B., Chetner, M.P., Crewalk,
J.A., Defalco, A., Foote, J., Green, B., Juma, S., Jung, S.Y.,
Linsenmeyer, T.A., MacMillan, R., Mayo, M., Ozawa, H.,
Roehrborn, C.G., Shenot, P.J., Stone, A., Vazquez, A.,
Killorin, W. and Rivas, D.A. (1999b) Long-term followup of
the North American multicenter UroLume trial for the treat-
ment of external detrusor–sphincter dyssynergia. J. Urol.,
161: 1545–1550.
190
Chartier-Kastler, E.J., Denys, P., Chancellor, M.B., Haertig,
A., Bussel, B. and Richard, F. (2001) Urodynamic monitor-
ing during percutaneous sacral nerve neurostimulation in
patients with neurogenic detrusor hyperreflexia. Neurourol.
Urodyn., 20: 61–71.
Chartier-Kastler, E.J., Thomas, L., Bussel, B., Chancellor,
M.B., Richard, F. and Denys, P. (2000) A urethral stent for
the treatment of detrusor-striated sphincter dyssynergia. BJU
Int., 86: 52–57.
Comarr, A.E. (1972) Intermittent catheterization for the trau-
matic cord bladder patient. J. Urol., 108: 79–81.
Creasey, G.H. (1993) Electrical stimulation of sacral roots for
micturition after spinal cord injury. Urol. Clin. North Am.,
20: 505–515.
Creasey, G.H. and Dahlberg, J.E. (2001) Economic conse-
quences of an implanted neuroprosthesis for bladder and
bowel management. Arch. Phys. Med. Rehab., 82:
1520–1525.
Dahlberg, A., Perttila, I., Wuokko, E. and Ala-Opas, M. (2004)
Bladder management in persons with spinal cord lesion. Spi-
nal Cord, 42: 694–698.
Dalmose, A.L., Rijkhoff, N.J., Kirkeby, H.J., Nohr, M.,
Sinkjaer, T. and Djurhuus, J.C. (2003) Conditional stimula-
tion of the dorsal penile/clitoral nerve may increase cysto-
metric capacity in patients with spinal cord injury.
Neurourol. Urodyn., 22: 130–137.
Davis, R., Houdayer, T., Andrews, B. and Barriskill, A. (1999)
Paraplegia: prolonged standing using closed-loop functional
electrical stimulation and Andrews ankle-foot orthosis. Artif.
Organs, 23: 418–420.
Davis, R., Houdayer, T., Andrews, B., Emmons, S. and
Patrick, J. (1997) Paraplegia: prolonged closed-loop stand-
ing with implanted nucleus FES-22 stimulator and Andrews’
foot-ankle orthosis. Stereotact. Funct. Neurosurg., 69:
281–287.
Davis, R., MacFarland, W.C. and Emmons, S.E. (1994) Initial
results of the nucleus FES-22-implanted system for limb
movement in paraplegia. Stereotact. Funct. Neurosurg., 63:
192–197.
Davis, R., Patrick, J. and Barriskill, A. (2001) Development of
functional electrical stimulators utilizing cochlear implant
technology. Med. Eng. Phys., 23: 61–68.
de Araujo, C.G., Schmidt, R.A. and Tanagho, E.A. (1982)
Neural pathways to lower urinary tract identified by retro-
grade axonal transport of horseradish peroxidase. Urology,
19: 290–295.
de Groat, W.C. (1993) Anatomy and physiology of the lower
urinary tract. Urol. Clin. North Am., 20: 383–401.
de Groat, W.C., Fraser, M.O., Yoshiyama, M., Smerin, S., Tai,
C., Chancellor, M.B., Yoshimura, N. and Roppolo, J.R.
(2001) Neural control of the urethra. Scand. J. Urol. Ne-
phrol. Suppl. 207: 35–43; discussion 106–125.
de Groat, W.C., Vizzard, M.A., Araki, I. and Roppolo, J. (1996)
Spinal interneurons and preganglionic neurons in sacral au-
tonomic reflex pathways. Prog. Brain Res., 107: 97–111.
Decter, R.M. (2000) Intravesical electrical stimulation of the
bladder: con. Urology, 56: 5–8.
Ebner, A., Jiang, C. and Lindstrom, S. (1992) Intravesical elec-
trical stimulation — an experimental analysis of the mech-
anism of action. J. Urol., 148: 920–924.
Egon, G., Barat, M., Colombel, P., Visentin, C., Isambert, J.L.
and Guerin, J. (1998) Implantation of anterior sacral root
stimulators combined with posterior sacral rhizotomy in spi-
nal injury patients. World J. Urol., 16: 342–349.
Fall, M. and Lindstrom, S. (1991) Electrical stimulation. A
physiologic approach to the treatment of urinary inconti-
nence. Urol. Clin. North Am., 18: 393–407.
Fang, Z.P. and Mortimer, J.T. (1991) Selective activation of
small motor axons by quasi-trapezoidal current pulses. IEEE
Trans. Biomed. Eng., 38: 168–174.
Fawcett, J. (2002) Repair of spinal cord injuries: where are we,
where are we going? Spinal Cord, 40: 615–623.
Fjorback, M.V., Hansen, J., Dalmose, A.L., Rijkhoff, N.J.M.
and Sinkjaer, T. (2003) A portable device for experimental
treatment of neurogenic detrusor overactivity. Neuromodu-
lation, 6: 158–165.
Foley, F.E.B. (1947) An artificial sphincter: a new device and
operation for control of enuresis and urinary incontinence.
J. Urol., 58: 250–259.
Frankel, H.L., Coll, J.R., Charlifue, S.W., Whiteneck, G.G.,
Gardner, B.P., Jamous, M.A., Krishnan, K.R., Nuseibeh, I.,
Savic, G. and Sett, P. (1998) Long-term survival in spinal cord
injury: a fifty year investigation. Spinal Cord, 36: 266–274.
Friedman, H., Nashold Jr., B.S. and Senechal, P. (1972) Spinal
cord stimulation and bladder function in normal and para-
plegic animals. J. Neurosurg., 36: 430–437.
Geirsson, G. and Fall, M. (1997) Maximal functional electrical
stimulation in routine practice. Neurourol. Urodyn., 16:
559–565.
Gladh, G., Mattsson, S. and Lindstrom, S. (2003) In-
travesical electrical stimulation in the treatment of micturit-
ion dysfunction in children. Neurourol. Urodyn., 22:
233–242.
Govier, F.E., Litwiller, S., Nitti, V., Kreder Jr., K.J. and
Rosenblatt, P. (2001) Percutaneous afferent neuromodulation
for the refractory overactive bladder: results of a multicenter
study. J. Urol., 165: 1193–1198.
Grill, W.M. (2000) Electrical activation of spinal neural circuits:
application to motor-system neural prostheses. Neuromod-
ulation, 3: 97–106.
Grill, W.M., Bhadra, N. and Wang, B. (1999) Bladder and
urethral pressures evoked by microstimulation of the sacral
spinal cord in cats. Brain Res., 836: 19–30.
Grill, W.M., Craggs, M.D., Foreman, R.D., Ludlow, C.L. and
Buller, J.L. (2001) Emerging clinical applications of electrical
stimulation: opportunities for restoration of function. J. Re-
hab. Res. Dev., 38: 641–653.
Grill, W.M., Wang, B., Hadziefendic, S. and Haxhiu, M.A.
(1998) Identification of the spinal neural network involved in
coordination of micturition in the male cat. Brain Res., 796:
150–160.
Grimes, J.H., Nashold, B.S. and Anderson, E.E. (1975) Clinical
application of electronic bladder stimulation in paraplegics.
J. Urol., 113: 338–340.

Groen, J., Amiel, C. and Bosch, R. (2004) Chronic puden-
dal nerve neuromodulation using an implantable mini-
stimulator in women with idiopathic refractory detrusor
overactivity incontinence. International Continence Society,
Paris.
Groen, J. and Bosch, J.L. (2001) Neuromodulation techniques
in the treatment of the overactive bladder. BJU Int., 87:
723–731.
Grunewald, V., Bhadra, N., Creasey, G.H. and Mortimer, J.T.
(1998) Functional conditions of micturition induced by
selective sacral anterior root stimulation: experimental
results in a canine animal model. World J. Urol., 16:
329–336.
Gustafson, K.J., Creasey, G.H. and Grill, W.M. (2003) A cath-
eter based method to activate urethral sensory nerve fibers.
J. Urol., 170: 126–129.
Gustafson, K.J., Creasey, G.H. and Grill, W.M. (2004) A ure-
thral afferent mediated excitatory bladder reflex exists in hu-
mans. Neurosci. Lett., 360: 9–12.
Guttmann, L. and Frankel, H. (1966) The value of intermittent
catheterisation in the early management of traumatic para-
plegia and tetraplegia. Paraplegia, 4: 63–84.
Hajivassiliou, C.A. (1998) The development and evolution of
artificial urethral sphincters. J. Med. Eng. Technol., 22:
154–159.
Hald, T. (1969) Neurogenic dysfunction of the urinary bladder.
An experimental and clinical study with special reference to
the ability of electrical stimulation to establish voluntary mi-
cturition. Dan. Med. Bull., 16(Suppl 5): 1+.
Hald, T., Meier, W., Khalili, A., Agrawal, G., Benton, J.G. and
Kantrowitz, A. (1967) Clinical experience with a radio-linked
bladder stimulator. J. Urol., 97: 73–78.
Hall, S.W. (2003) Commercializing neuroprostheses: the busi-
ness of putting the brain back in business. Molecular Biology.
Princeton University, Princeton, p. 185.
Hamid, R., Arya, M., Wood, S., Patel, H.R. and Shah, P.J.
(2003) The use of the Memokath stent in the treatment of
detrusor sphincter dyssynergia in spinal cord injury patients:
a single-centre seven-year experience. Eur. Urol., 43:
539–543.
Hansen, R.B., Biering-Sorensen, F. and Kristensen, J.K. (2004)
Bladder emptying over a period of 10–45 years after a trau-
matic spinal cord injury. Spinal Cord, 42: 631–637.
Heine, J.P., Schmidt, R.A. and Tanagho, E.A. (1977) Intraspi-
nal sacral root stimulation for controlled micturition. Invest.
Urol., 15: 78–82.
Hohenfellner, M., Humke, J., Hampel, C., Dahms, S., Matzel,
K., Roth, S., Thuroff, J.W. and Schultz-Lampel, D. (2001)
Chronic sacral neuromodulation for treatment of neurogenic
bladder dysfunction: long-term results with unilateral im-
plants. Urology, 58: 887–892.
Hollander, J.B. and Diokno, A.C. (1993) Urinary diversion and
reconstruction in the patient with spinal cord injury. Urol.
Clin. North Am., 20: 465–474.
Holmquist, B. and Olin, T. (1968a) Electromicturition in female
dogs at pelvic nerve stimulation. An urethrocystographic
study. Scand. J. Urol. Nephrol., 2: 129–135.
191
Holmquist, B. and Olin, T. (1968b) Electromicturition in male
dogs at pelvic nerve stimulation. An urethrocystographic
study. Scand. J. Urol. Nephrol., 2: 115–127.
Houdayer, T., Barriskill, A., Davis, R., Milijasevic, Z. and
Cousins, R. (2002) The Praxis and Minax FES systems for
functional restoration in SCI. Proceedings of the IEEE 28th
Annual Northeast Bioengineering Conference, Philadelphia,
PA, pp. 189–190.
Ishigooka, M., Suzuki, Y., Hashimoto, T., Sasagawa, I., Nakada,
T. and Handa, Y. (1998) A new technique for sacral nerve
stimulation: a percutaneous method for urinary incontinence
caused by spinal cord injury. Br. J. Urol., 81: 315–318.
Jamil, F. (2001) Towards a catheter free status in neurogenic
bladder dysfunction: a review of bladder management op-
tions in spinal cord injury (SCI). Spinal Cord, 39: 355–361.
Jezernik, S., Craggs, M., Grill, W.M., Creasey, G. and Rijkhoff,
N.J. (2002) Electrical stimulation for the treatment of bladder
dysfunction: current status and future possibilities. Neurol.
Res., 24: 413–430.
Jezernik, S., Wen, J.G., Rijkhoff, N.J., Djurhuus, J.C. and
Sinkjaer, T. (2000) Analysis of bladder related nerve cuff
electrode recordings from preganglionic pelvic nerve and
sacral roots in pigs. J. Urol., 163: 1309–1314.
Jonas, U., Heine, J.P. and Tanagho, E.A. (1975) Studies on
feasibility of urinary bladder evacuation by direct spinal cord
stimulation. I. Parameters of most effective stimulation. In-
vest. Urol., 13: 142–150.
Jonas, U. and Tanagho, E.A. (1975) Studies on feasibility of
urinary bladder evacuation by direct spinal cord stimulation
II Poststimulus voiding: A way to overcome outflow resist-
ance. Invest. Urol., 13: 151–153.
Juan Garcia, F.J., Salvador, S., Montoto, A., Lion, S., Balvis,
B., Rodriguez, A., Fernandez, M. and Sanchez, J. (1999) In-
traurethral stent prosthesis in spinal cord injured patients
with sphincter dyssynergia. Spinal Cord, 37: 54–57.
Kantrowitz, A. and Schamaun, M. (1963) Paraplegic dogs: uri-
nary bladder evacuation with direct electric stimulation. Sci-
ence, 139: 115–116.
Kaplan, W.E. (2000) Intravesical electrical stimulation of the
bladder: pro. Urology, 56: 2–4.
Kennedy, R.H. (1946) The new viewpoint toward spinal cord
injuries. Ann. Surg., 124: 1057–1065.
Kirkham, A.P., Knight, S.L., Craggs, M.D., Casey, A.T. and
Shah, P.J. (2002) Neuromodulation through sacral nerve
roots 2 to 4 with a Finetech–Brindley sacral posterior and
anterior root stimulator. Spinal Cord, 40: 272–281.
Kirkham, A.P., Shah, N.C., Knight, S.L., Shah, P.J. and Crag-
gs, M.D. (2001) The acute effects of continuous and condi-
tional neuromodulation on the bladder in spinal cord injury.
Spinal Cord, 39: 420–428.
Koldewijn, E.L., Rijkhoff, N.J., van Kerrebroeck, E.V., De-
bruyne, F.M. and Wijkstra, H. (1994) Selective sacral root
stimulation for bladder control: acute experiments in an an-
imal model. J. Urol., 151: 1674–1679.
Kondo, A., Otani, T. and Takita, T. (1982) Suppression of
bladder instability by penile squeeze. Br. J. Urol., 54:
360–362.
192
Kuru, M. (1965) Nervous control of micturition. Physiol. Rev.,
45: 425–494.
Lapides, J., Diokno, A.C., Silber, S.J. and Lowe, B.S. (1972)
Clean, intermittent self-catheterization in the treatment of
urinary tract disease. J. Urol., 107: 458–461.
Lee, T. (1997) Electrical stimulation techniques to restore blad-
der function. Neurosurg. Quart., 7: 264–272.
Lee, Y.H. and Creasey, G.H. (2002) Self-controlled dorsal pe-
nile nerve stimulation to inhibit bladder hyperreflexia in in-
complete spinal cord injury: a case report. Arch. Phys. Med.
Rehab., 83: 273–277.
Lee, Y.H., Creasey, G.H., Lim, H., Song, J., Song, K. and Kim,
J. (2003) Detrusor and blood pressure responses to dorsal
penile nerve stimulation during hyperreflexic contraction of
the bladder in patients with cervical cord injury. Arch. Phys.
Med. Rehab., 84: 136–140.
Light, J.K. and Scott, F.B. (1983) Use of the artificial urinary
sphincter in spinal cord injury patients. J. Urol., 130:
1127–1129.
Lin, V.W., Wolfe, V., Frost, F.S. and Perkash, I. (1997) Mi-
cturition by functional magnetic stimulation. J. Spinal Cord
Med., 20: 218–226.
Loeb, G.E., Peck, R.A., Moore, W.H. and Hood, K. (2001)
BION system for distributed neural prosthetic interfaces.
Med. Eng. Phys., 23: 9–18.
Low, A.I. and McRae, P.J. (1998) Use of the Memokath for
detrusor–sphincter dyssynergia after spinal cord injury — a
cautionary tale. Spinal Cord, 36: 39–44.
Madersbacher, H. (1990) Intravesical electrical stimulation for
the rehabilitation of the neuropathic bladder. Paraplegia, 28:
349–352.
Madersbacher, H., Pauer, W. and Reiner, E. (1982) Rehabil-
itation of micturition by transurethral electrostimulation of
the bladder in patients with incomplete spinal cord lesions.
Paraplegia, 20: 191–195.
Madjar, S., Sabo, E., Halachmi, S., Wald, M., Issaq, E.,
Moskovitz, B., Beyar, M. and Nativ, O. (1999) A remote
controlled intraurethral insert for artificial voiding: a new
concept for treating women with voiding dysfunction.
J. Urol., 161: 895–898.
Magasi, P. and Simon, Z. (1986) Electrical stimulation of the
bladder and gravidity. Urol. Int., 41: 241–245.
Mazouni, C., Karsenty, G., Bladou, F. and Serment, G. (2004)
Urethral device in women with chronic urinary retention: an
alternative to self-catheterization? Eur. J. Obstet. Gynecol.
Reprod. Biol., 115: 80–84.
McCreery, D., Pikov, V., Lossinsky, A., Bullara, L. and
Agnew, W. (2004) Arrays for chronic functional microstim-
ulation of the lumbosacral spinal cord. IEEE Trans. Neural.
Syst. Rehab. Eng., 12: 195–207.
McCreery, D.B., Agnew, W.F., Yuen, T.G. and Bullara, L.
(1990) Charge density and charge per phase as cofactors in
neural injury induced by electrical stimulation. IEEE Trans.
Biomed. Eng., 37: 996–1001.
McGuire, E.J., Zhang, S.C., Horwinski, E.R. and Lytton, B.
(1983) Treatment of motor and sensory detrusor instability
by electrical stimulation. J. Urol., 129: 78–79.
Merrill, D.C. and Conway, C.J. (1974) Clinical experience with
the Mentor bladder stimulation. I. Patients with upper motor
neuron lesions. J. Urol., 112: 52–56.
Middleton, J.W. and Keast, J.R. (2004) Artificial autonomic
reflexes: using functional electrical stimulation to mimic blad-
der reflexes after injury or disease. Auton Neurosci, 113: 3–15.
Middleton, J.W., Lim, K., Taylor, L., Soden, R. and Rutkowski,
S. (2004) Patterns of morbidity and rehospitalisation follow-
ing spinal cord injury. Spinal Cord, 42: 359–367.
Moore, T. and Schofield, P.F. (1967) Treatment of stress in-
continence by maximum perineal electrical stimulation. Br.
Med. J., 3: 150–151.
Mushahwar, V.K., Collins, D.F. and Prochazka, A. (2000)
Spinal cord microstimulation generates functional limb
movements in chronically implanted cats. Exp. Neurol.,
163: 422–429.
Nacey, J. and Delahunt, B. (1993) The evolution and develop-
ment of the urinary catheter. Aust. N. Z. J. Surg., 63:
815–819.
Nakamura, M. and Sakurai, T. (1984) Bladder inhibition by
penile electrical stimulation. Br. J. Urol., 56: 413–415.
Nashold Jr., B.S., Friedman, H. and Boyarsky, S. (1971) Elec-
trical activation of micturition by spinal cord stimulation.
J. Surg. Res., 11: 144–147.
Nashold Jr., B.S., Friedman, H., Glenn, J.F., Grimes, J.H.,
Barry, W.F. and Avery, R. (1972) Electromicturition in par-
aplegia. Implantation of a spinal neuroprosthesis. Arch.
Surg., 104: 195–202.
Nashold Jr., B.S., Friedman, H. and Grimes, J. (1981) Electrical
stimulation of the conus medullaris to control the bladder in
the paraplegic patient. A 10-year review. Appl. Neurophysiol.,
44: 225–232.
Nativ, O., Moskowitz, B., Issaq, E., Condrea, A., Kastin, A.,
Halachmi, S., Burbara, J., Madjar, S. and Beyar, M. (1997) A
new intraurethral sphincter prosthesis with a self contained
urinary pump. ASAIO J., 43: 197–203.
Nissenkorn, I., Shalev, M., Radziszewski, P., Dobronski, P.,
Borkowski, A. and De Jong, P.R. (2004) Patient-adjusted
intermittent electrostimulation for treating stress and urge
urinary incontinence. BJU Int., 94: 105–109.
Okada, N., Igawa, Y., Ogawa, A. and Nishizawa, O. (1998)
Transcutaneous electrical stimulation of thigh muscles in the
treatment of detrusor overactivity. Br. J. Urol., 81: 560–564.
Petrou, S.P., Elliott, D.S. and Barrett, D.M. (2000) Artificial
urethral sphincter for incontinence. Urology, 56: 353–359.
Previnaire, J.G., Soler, J.M. and Perrigot, M. (1998) Is there a
place for pudendal nerve maximal electrical stimulation for
the treatment of detrusor hyperreflexia in spinal cord injury
patients? Spinal Cord, 36: 100–103.
Prochazka, A., Gaunt, R.G., Mushahwar, V., Downie, J.W.
and Shefchyk, S.J. (2003a) Functional microstimulation of
the lumbosacral spinal cord: Quarterly Report #6. NIH-NI-
NDS contract N01-NS-2-2342.
Prochazka, A., Mushahwar, V., Downie, J.W. and Shefchyk,
S.J. (2002a) Functional microstimulation of the lumbosacral
spinal cord: Quarterly Report #1. NIH-NINDS contract
N01-NS-2-2342.

Prochazka, A., Mushahwar, V., Downie, J.W., Shefchyk, S.J.
and Gaunt, R.G. (2003b) Functional microstimulation of the
lumbosacral spinal cord: Quarterly Report #5. NIH-NINDS
contract N01-NS-2-2342.
Prochazka, A., Mushahwar, V. and Yakovenko, S. (2002b)
Activation and coordination of spinal motoneuron pools af-
ter spinal cord injury. Prog. Brain Res., 137: 109–124.
Prochazka, A., Westerman, R.A. and Ziccone, S.P. (1976) Dis-
charges of single hindlimb afferents in the freely moving cat.
J. Neurophysiol., 39: 1090–1104.
Reynard, J.M., Vass, J., Sullivan, M.E. and Mamas, M. (2003)
Sphincterotomy and the treatment of detrusor–sphincter
dyssynergia: current status, future prospects. Spinal Cord, 41:
1–11.
Rijkhoff, N.J. (2004a) Emerging FES applications for control
of the urinary bladder. In: Horch K.W. and Dhillon
G.S. (Eds.) Neuroprosthetics: Theory and Practice, Vol. 2.
World Scientific Publishing Co. Ltd, Singapore, pp.
1054–1066.
Rijkhoff, N.J. (2004b) Neuroprostheses to treat neurogenic
bladder dysfunction: current status and future perspectives.
Childs Nerv. Syst., 20: 75–86.
Rijkhoff, N.J., Holsheimer, J., Koldewijn, E.L., Struijk, J.J.,
van Kerrebroeck, P.E., Debruyne, F.M. and Wijkstra, H.
(1994a) Selective stimulation of sacral nerve roots for bladder
control: a study by computer modeling. IEEE Trans. Bio-
med. Eng., 41: 413–424.
Rijkhoff, N.J., Koldewijn, E.L., Van Kerrebroeck, P.E.,
Debruyne, F.M. and Wijkstra, H. (1994b) Acute animal stud-
ies on the use of anodal block to reduce urethral resistance in
sacral root stimulation. IEEE Trans. Rehab. Eng., 2: 92–99.
Rijkhoff, N.J., Wijkstra, H., van Kerrebroeck, P.E. and
Debruyne, F.M. (1997a) Selective detrusor activation by
electrical sacral nerve root stimulation in spinal cord injury.
J. Urol., 157: 1504–1508.
Rijkhoff, N.J., Wijkstra, H., van Kerrebroeck, P.E. and
Debruyne, F.M. (1997b) Urinary bladder control by electri-
cal stimulation: review of electrical stimulation techniques in
spinal cord injury. Neurourol. Urodyn., 16: 39–53.
Rijkhoff, N.J., Wijkstra, H., van Kerrebroeck, P.E. and
Debruyne, F.M. (1998) Selective detrusor activation by
sacral ventral nerve-root stimulation: results of intraoperative
testing in humans during implantation of a Finetech–Brindley
system. World J. Urol., 16: 337–341.
Sauerwein, D., Ingunza, W., Fischer, J., Madersbacher, H.,
Polkey, C.E., Brindley, G.S., Colombel, P. and Teddy, P.
(1990) Extradural implantation of sacral anterior root
stimulators. J. Neurol. Neurosurg. Psychiat., 53: 681–684.
Savic, G., Short, D.J., Weitzenkamp, D., Charlifue, S. and
Gardner, B.P. (2000) Hospital readmissions in people with
chronic spinal cord injury. Spinal Cord, 38: 371–377.
Sawan, M. and Elhilali, M.M. (2002) Electronic stimulator im-
plant for modulating and synchronizing bladder and sphinc-
ter function. US Pat. 6,393,323. McGill University, Gestion
Univalor.
Schmidt, R.A. (1983) Neural prostheses and bladder control.
IEEE Eng. Med. Biol. Mag., 2: 31–36.
193
Schmidt, R.A. (1986) Advances in genitourinary neurostimula-
tion. Neurosurgery, 19: 1041–1044.
Schmidt, R.A. (1988) Applications of neurostimulation in urol-
ogy. Neurourol. Urodyn., 7: 585–592.
Schmidt, R.A., Bruschini, H. and Tanagho, E.A. (1979) Uri-
nary bladder and sphincter responses to stimulation of dorsal
and ventral sacral roots. Invest. Urol., 16: 300–304.
Schmidt, R.A., Senn, E. and Tanagho, E.A. (1990) Functional
evaluation of sacral nerve root integrity. Report of a tech-
nique. Urology, 35: 388–392.
Schurch, B., Reilly, I., Reitz, A. and Curt, A. (2003) Electro-
physiological recordings during the peripheral nerve evalua-
tion (PNE) test in complete spinal cord injury patients.
World J. Urol., 20: 319–322.
Schurch, B., Suter, S. and Dubs, M. (1999) Intraurethral
sphincter prosthesis to treat hyporeflexic bladders in women:
does it work? BJU Int., 84: 789–794.
Scott, F.B., Bradley, W.E. and Timm, G.W. (1974) Treatment
of urinary incontinence by an implantable prosthetic urinary
sphincter. J. Urol., 112: 75–80.
Selzman, A.A. and Hampel, N. (1993) Urologic complications
of spinal cord injury. Urol. Clin. North Am., 20: 453–464.
Shafik, A., el-Sibai, O. and Ahmed, I. (2003a) Effect of urethral
dilation on vesical motor activity: identification of the
urethrovesical reflex and its role in voiding. J. Urol., 169:
1017–1019.
Shafik, A., Shafik, A.A., El-Sibai, O. and Ahmed, I. (2003b)
Role of positive urethrovesical feedback in vesical evacua-
tion. The concept of a second micturition reflex: the
urethrovesical reflex. World J. Urol., 21: 167–170.
Shaker, H.S., Tu, L.M., Robin, S., Arabi, K., Hassouna, M.,
Sawan, M. and Elhilali, M.M. (1998) Reduction of bladder
outlet resistance by selective sacral root stimulation using
high-frequency blockade in dogs: an acute study. J. Urol.,
160: 901–907.
Shaw, P.J., Milroy, E.J., Timoney, A.G., el Din, A. and
Mitchell, N. (1990) Permanent external striated sphincter
stents in patients with spinal injuries. Br. J. Urol., 66: 297–302.
Shefchyk, S.J. (2001) Sacral spinal interneurones and the con-
trol of urinary bladder and urethral striated sphincter muscle
function. J. Physiol., 533: 57–63.
Shefchyk, S.J. and Buss, R.R. (1998) Urethral pudendal affer-
ent-evoked bladder and sphincter reflexes in decerebrate and
acute spinal cats. Neurosci. Lett., 244: 137–140.
Shindo, N. and Jones, R. (1987) Reciprocal patterned electrical
stimulation of the lower limbs in severe spasticity. Physio-
therapy, 73: 579–582.
Sie, J.A., Blok, B.F., de Weerd, H. and Holstege, G. (2001)
Ultrastructural evidence for direct projections from the pon-
tine micturition center to glycine-immunoreactive neurons in
the sacral dorsal gray commissure in the cat. J. Comp. Ne-
urol., 429: 631–637.
Siegel, S.W., Catanzaro, F., Dijkema, H.E., Elhilali, M.M.,
Fowler, C.J., Gajewski, J.B., Hassouna, M.M., Janknegt,
R.A., Jonas, U., van Kerrebroeck, P.E., Lycklama a
Nijeholt, A.A., Oleson, K.A. and Schmidt, R.A. (2000)
Long-term results of a multicenter study on sacral nerve
194
stimulation for treatment of urinary urge incontinence,
urgency-frequency, and retention. Urology, 56: 87–91.
Smith, B.T., Betz, R.R., Johnston, T., Benda, B., Mulcahey,
M.J., Davis, R. and Barriskill, A. (2002) Initial experiences
with the Praxis multi-functional implantable FES system for
individuals with spinal cord injury. FESnet http:
//fesnet.eng.gla.ac.uk/conference/x_smithb.pdf.PDF.
Solomonow, M., Eldred, E., Lyman, J. and Foster, J. (1983)
Control of muscle contractile force through indirect high-
frequency stimulation. Am. J. Phys. Med., 62: 71–82.
Stenberg, C.C., Burnette, H.W. and Bunts, R.C. (1967) Elec-
trical stimulation of human neurogenic bladders: experience
with 4 patients. J. Urol., 97: 79–84.
Stewart, C.C. (1899) On the course of impulses to and from the
cat’s bladder. Am. J. Physiol., 2: 182–202.
Susset, J.G. and Boctor, Z.N. (1967) Implantable electrical
vesical stimulator: clinical experience. J. Urol., 98: 673–678.
Talalla, A., Bloom, J.W. and Quang, N. (1987) FES for blad-
der: direct or indirect means? Pacing Clin. Electrophysiol.,
10: 240–245.
Tanagho, E.A. and Schmidt, R.A. (1988) Electrical stimulation
in the clinical management of the neurogenic bladder.
J. Urol., 140: 1331–1339.
Tanagho, E.A., Schmidt, R.A. and Orvis, B.R. (1989) Neural
stimulation for control of voiding dysfunction: a preliminary
report in 22 patients with serious neuropathic voiding disor-
ders. J. Urol., 142: 340–345.
Timm, G.W., Bradley, W.E. and Scott, F.B. (1974) Experi-
mental evaluation of an implantable externally controllable
urinary sphincter. Invest. Urol., 11: 326–330.
van Balken, M.R., Vandoninck, V., Gisolf, K.W., Vergunst, H.,
Kiemeney, L.A., Debruyne, F.M. and Bemelmans, B.L. (2001)
Posterior tibial nerve stimulation as neuromodulative treat-
ment of lower urinary tract dysfunction. J. Urol., 166: 914–918.
van Balken, M.R., Vergunst, H. and Bemelmans, B.L. (2004)
The use of electrical devices for the treatment of bladder
dysfunction: a review of methods. J. Urol., 172: 846–851.
Van Kerrebroeck, P.E. (2002) Neuromodulation and other
electrostimulatory techniques. Scand. J. Urol. Nephrol., Sup-
pl. 210: 82–86.
Van Kerrebroeck, P.E., Koldewijn, E.L. and Debruyne, F.M.
(1993) Worldwide experience with the Finetech–Brindley
sacral anterior root stimulator. Neurourol. Urodyn., 12:
497–503.
Vandoninck, V., van Balken, M.R., Finazzi Agro, E., Petta, F.,
Micali, F., Heesakkers, J.P., Debruyne, F.M., Kiemeney,
L.A. and Bemelmans, B.L. (2003) Percutaneous tibial nerve
stimulation in the treatment of overactive bladder: urody-
namic data. Neurourol. Urodyn., 22: 227–232.
Vodovnik, L. (1981) Therapeutic effects of functional electrical
stimulation of extremities. Med. Biol. Eng. Comput., 19:
470–478.
Vuckovic, A. and Rijkhoff, N.J. (2004) Different pulse shapes
for selective large fibre block in sacral nerve roots using a
technique of anodal block: an experimental study. Med. Biol.
Eng. Comput., 42: 817–824.
Walter, J.S., Wheeler, J.S., Cai, W., King, W.W. and Wurster,
R.D. (1999) Evaluation of a suture electrode for direct blad-
der stimulation in a lower motor neuron lesioned animal
model. IEEE Trans. Rehab. Eng., 7: 159–166.
Wein, A.J. (1998) Pharmacologic options for the overactive
bladder. Urology, 51: 43–47.
Weld, K.J. and Dmochowski, R.R. (2000) Effect of bladder
management on urological complications in spinal cord in-
jured patients. J. Urol., 163: 768–772.
Wheeler Jr., J.S., Robinson, C.J., Culkin, D.J. and Bolan, J.M.
(1986) The effect of thigh muscle reconditioning by electrical
stimulation on urodynamic activity in SCI patients. J. Am.
Paraplegia Soc., 9: 16–23.
Wheeler Jr., J.S., Walter, J.S. and Zaszczurynski, P.J. (1992)
Bladder inhibition by penile nerve stimulation in spinal cord
injury patients. J. Urol., 147: 100–103.
Wilson, T.S., Lemack, G.E. and Dmochowski, R.R. (2002)
UroLume stents: lessons learned. J. Urol., 167: 2477–2480.
Wozniak, W. and Skowronska, U. (1967) Comparative anat-
omy of pelvic plexus in cat, dog, rabbit, macaque and man.
Anat. Anz., 120: 457–473.
Yoshimura, N. (1999) Bladder afferent pathway and spinal
cord injury: possible mechanisms inducing hyperreflexia of
the urinary bladder. Prog. Neurobiol., 57: 583–606.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 12

Novel repair strategies to restore bladder function

following cauda equina/conus medullaris injuries

Thao X. Hoang and Leif A. Havton

Department of Neurology, David Geffen School of Medicine at University of California Los Angeles, 710 Westwood Plaza,
Los Angeles, CA 90095-1769, USA

Abstract: Trauma to the thoracolumbar junction or lumbosacral spine may result in a conus medullaris or
cauda equina syndrome. In both conditions, symptoms typically include paraparesis or paraplegia, sensory
impairment, pain, as well as bladder, bowel, and sexual dysfunctions. We present in this review a series of
neural repair strategies that have been developed to address the unique features and challenges of subjects
with a conus medullaris or cauda equina syndrome. We address, in particular, neural repair strategies that
may have a translational research potential to restore bladder function. Recent animal injury models have
suggested that a progressive retrograde death of both autonomic and motor neurons may contribute to the
neurological deficits in subjects with conus medullaris and cauda equina injuries. For subjects with acute
injuries, we present novel strategies to promote neuroprotection, axonal regeneration, and functional
reinnervation of the lower urinary tract. For subjects with chronic injuries, we discuss new approaches to
replace lost autonomic and motor neurons. A brief discussion on a variety of outcome measures that may
be suitable to evaluate the function of the lower urinary tract in rodent neural repair models is also
provided.

Introduction dysfunction. Bladder dysfunction may include loss

In humans, trauma to the spine at the thoraco-
lumbar junction or to the lumbosacral spine may
result in a conus medullaris or cauda equina syn-
drome. According to the International Standards
for Neurological and Functional Classification of
Spinal Cord Injury, a conus medullaris syndrome
consists of an injury to both the sacral cord
(conus) and lumbar nerve roots within the spinal
canal, while a cauda equina syndrome consists of
an injury to the lumbosacral nerve roots below the
tip of the conus medullaris (Maynard et al., 1997).
Clinically, injury to the conus medullaris and ca-
uda equina may cause paraparesis or paraplegia,
sensory disturbance, as well as bladder and bowel
Corresponding author. Tel.: +310-206-6500;
Fax: +310-794-9486; E-mail: lhavton@mednet.ucla.edu
of voluntary micturition, urethral closure impair-
ment, and an areflexic bladder (Pavlakis et al.,
1983; Fowler et al., 1984; Light et al., 1993;
Fowler, 1999).
Occasionally, sacral segments may show pre-
served reflexes, e.g., the micturition and bulbocav-
ernosus reflexes. Such reflex sparing may be
encountered when the injury affects the most ros-
tral part of the sacral cord with sparing of the most
caudal portion of the sacral spinal cord. In de-
tailed neurologic evaluations of patients with
conus medullaris and cauda equina lesions, find-
ings typically include an absent or substantially
diminished bulbocavernosus reflex, a bladder de-
trusor areflexia on cystometrogram examinations,
and neuropathic perineal electromyographic
changes indicated by polyphasic waves, fibrilla-
tion potentials, and positive sharp waves (Pavlakis

DOI: 10.1016/S0079-6123(05)52012-0 195

196
et al., 1983). However, a conventional urodynamic
evaluation alone is unable to distinguish between a
pure conus medullaris lesion and a conus lesion
with a concomitant cauda equina involvement
(Beric and Light, 1992).
Conus medullaris and cauda equina injuries are
complex lesions, which may present with a com-
bination of pathologies, including intramedullary
damage as well as with stretching, tearing, crush,
laceration, and avulsion of proximal and distal
portions of nerve roots as a result of, e.g., trau-
matic falls, motor vehicle accidents, acts of vio-
lence, herniated discs, and sports injuries (Fig. 1).
The relative proportion of lesions to the central
Fig. 1. Magnetic resonance imaging (MRI) study of the lower
thoracic and upper lumbar spine of a young adult male fol-
lowing an acute traumatic injury from a motor vehicle accident.
The MRI scan shows a severe dislocation at the T12–L1 ver-
tebral level with a posterior displacement of the lumbar column,
rupture of the T12–L1 intervertebral disc, and an L1 vertebral
fracture. There is a near obliteration of the spinal canal with
crush injury to the conus medullaris. SC ¼ spinal cord.
and peripheral portions of the nervous system may
vary between patients with a conus medullaris
syndrome. While the literature on postmortem
pathology of traumatic conus medullaris and
cauda equina injuries is scant, certain types of
pathologies may be more prevalent in association
with certain types of trauma. For instance,
lumbosacral root avulsion injuries may develop
more frequently in association with high-energy
trauma, e.g., high velocity motor vehicle accidents
(Moschilla et al., 2001; Hans et al., 2004; Lang
et al., 2004; Monga et al., 2004). However, animal
models to study conus medullaris and cauda
equina injuries have been relatively sparse.
Effects of axonal lesions on efferent spinal
cord neurons
Injuries to the conus medullaris and lumbosacral
nerve roots may cause central and/or peripheral
lesions to the axons of efferent spinal cord moto-
neurons and preganglionic parasympathetic neu-
rons, as well as to the primary afferents located in
the dorsal roots. Factors such as species, age,
location, and severity of injury may influence the
retrograde response of an axotomized neuron. In
particular, the location of the axonal lesion greatly
affects the fate of preganglionic parasympathetic
neurons and motoneurons. In rats, cats, and
primates, a peripheral nerve transection, spinal
nerve transection or limb amputation resulted in
no or limited motoneuron loss (Carlson et al.,
1979; Johnson et al., 1991; Vanden Noven et al.,
1993; Piehl et al., 1995; Anneser et al., 2000; Wu
and Kaas, 2000; Ma et al., 2001). Similarly,
preganglionic parasympathetic neurons showed
no detectable loss at 2–4 weeks after a peripheral
nerve injury performed by a surgical removal of
the major pelvic ganglion (Vizzard et al., 1995).
In contrast, a cervical or lumbar ventral root
avulsion injury, alone or combined with a seg-
mental dorsal root lesion, resulted in a progressive
and marked retrograde death of motoneurons in
the adult rat (Wu, 1993; Koliatsos et al., 1994;
Novikov et al., 1995, 2000; Kishino et al., 1997;
Martin et al., 1999) and the adult cat (Holmberg
and Kellerth, 2000). A more proximal lesion, such

as an intramedullary transection of the ventral
funiculus, also induced a retrograde death of many
motoneurons in the adult cat (Lindå et al., 1992).
Less information is available on the response of
autonomic efferent neurons to lesions to the cen-
tral and most proximal portions of their axons.
A model to study cauda equina/conus medullaris
injuries in the rat
We have recently developed a conus medullaris
and cauda equina injury model in the adult rat to
study the retrograde effects of a proximal lumbo-
sacral ventral root lesion on axotomized pregangl-
ionic parasympathetic neurons and motoneurons
with pelvic targets, including the external urethral
and anal sphincters (Hoang et al., 2003). In this
model, we performed a unilateral ventral root av-
ulsion injury of the lumbar (L) and sacral (S), L5,
L6, S1, and S2 segments (Fig. 2). These segmental
levels were selected to include primarily the L6 and
S1 segments, which contain the preganglionic
Fig. 2. Schematic drawing of the unilateral lumbosacral ventral
root avulsion injury model. Four consecutive lumbosacral ven-
tral roots (L5–S2) are avulsed at the ventral root exit from the
spinal cord surface. Note that the dorsal roots remain intact.
197
parasympathetic neurons of the intermediolateral
nucleus and the somatic motoneurons of the
dorsolateral and dorsomedial nuclei innervating
the external urethral and anal sphincters, respec-
tively (Schrøder, 1980). We demonstrated that an
L5–S2 ventral root avulsion injury causes a pro-
gressive and parallel death of axotomized pre-
ganglionic parasympathetic neurons and
motoneurons (Fig. 3; Hoang et al., 2003).
Already at 1 week after the lesion, there was a
significant loss of 12% of the preganglionic para-
sympathetic neurons and 17% of the motoneu-
rons. The number of undetectable neurons
gradually increased to 78% of the preganglionic
parasympathetic neurons and 84% of the moto-
neurons by 6 weeks postoperatively.
Fig. 3. Immunohistochemical light stable detection of prelabe-
led Fluorogold (FG, Fluorochrome, Denver, CO) shows close
to a 50% loss of the preganglionic parasympathetic neurons
(PPN) and motoneurons (MN) on the injured side compared to
the contralateral non-lesioned side at 2 weeks after an L5–S2
ventral root avulsion injury in the adult rat (a–d). Rats were
injected intraperitoneally with 400 ml of 0.5% FG at 1 week
prior to the ventral root avulsion injury to label all pregangl-
ionic parasympathetic neurons and motoneurons. Following an
intravascular perfusion with a 4% paraformaldehyde solution,
40 mm spinal cord frozen sections of the L6 and S1 segments
were processed for immunohistochemistry using an antibody
against FG (Fluorochrome, Denver, CO) to enhance detection
of the fluorescent retrograde tracer. Scale bar ¼ 50 mm for a–d.
198
In our lumbosacral ventral root avulsion model,
we detected nuclear chromatin condensation,
apoptotic bodies, and the activation of caspase-3
in the cytoplasm and nuclei of both pregangl-
ionic parasympathetic neurons and motoneurons,
suggesting that apoptosis contributes to the retro-
grade cell death of both preganglionic parasym-
pathetic neurons and motoneurons (Fig. 4; Hoang
et al., 2003). Our findings are in agreement with
other studies that show evidence for apoptotic cell
death mechanisms following a sciatic nerve injury
in the adult rat. For instance, axotomized moto-
neurons demonstrated fragmentation of nuclear
DNA, early accumulation of single-strand breaks
(Martin et al., 1999; Liu and Martin, 2001), acti-
vation of caspase-3, nuclear accumulation of the
apoptosis-inducing factor p53, and increased
detection of apoptosis-associated Bax (Martin
and Liu, 2002). However, a ventral root avulsion
injury at the cervical level in adult mice induced
ultrastructural changes in motoneurons, including
disruption, perturbation, and depletion of organ-
elles, suggestive of a necrotic process (Li et al.,
1998). Interestingly, nuclear pyknosis, a finding
suggestive of apoptosis, was also seen in lesioned
motoneurons at later stages (Li et al., 1998).
Therefore, it seems possible that both apoptotic
and necrotic mechanisms may be involved in
the retrograde cell death of efferent spinal cord
Fig. 4. Immunohistochemistry for light stable detection of ac-
tivated caspase-3 (Promega, Madison, WI) demonstrates pres-
ence of the activated caspase-3 in a subset of the remaining
parasympathetic preganglionic neurons and motoneurons on
the lesioned side at 2 weeks after the avulsion injury (arrows, a,
b). High magnification micrograph demonstrates activated ca-
spase-3 being located primarily within the soma (b). Scale
bar ¼ 300 mm for a; 50 mm for b.
neurons following proximal lesions to their axons.
The mechanisms for cell death may also be age-
dependent, as treatment with caspase inhibitors
protected axotomy-induced motoneuron death in
neonatal rats but not in adult rats (Chan et al.,
2001).
Expression of nitric oxide synthase in axotomized
spinal cord neurons
A ventral root transection injury that does not lead
to any significant motoneuron loss, does not in-
duce any nitric oxide synthase expression in axo-
tomized rat motoneurons (Wu, 1993). However, a
ventral root avulsion injury induces the expression
of nitric oxide synthase in axotomized motoneu-
rons, which are destined to undergo retrograde
degeneration and death (Wu, 1993; Novikov et al.,
1995). An additional association between the in-
jury-induced expression of nitric oxide synthase
and motoneuron death was demonstrated in a
pharmacological study, which showed that treat-
ment with nitroarginine, a specific nitric oxide
synthase inhibitor, reduces motoneuron loss fol-
lowing a spinal nerve root lesion in the rat (Wu
and Li, 1993).
In contrast to motoneurons innervating the up-
per and lower extremities, about half of the normal
population of preganglionic parasympathetic neu-
rons stain for nicotinamide adenine dinucleotide
phosphate hydrogen (NADPH) diaphorase or
express neuronal nitric oxide synthase in the rat
conus medullaris (Vizzard et al., 1993a; Saito et al.,
1994). A marked increase in the number of
NADPH diaphorase positive preganglionic para-
sympathetic neurons was detected at 1–3 weeks
after a combined ventral and dorsal rhizotomy
(Vizzard et al., 1993b). The number of pregangl-
ionic parasympathetic neurons expressing neuron-
al nitric oxide synthase also increased in the
absence of any detectable cell death, at 2–4 weeks
following a surgical resection of the major pelvic
ganglion (Vizzard et al., 1995). However, the
expression of neuronal nitric oxide synthase
in preganglionic parasympathetic neurons under-
going a progressive cell death from a lumbosacral
ventral root avulsion is not yet known. Because

of the above differences between motoneurons
and preganglionic parasympathetic neurons in
neuronal nitric oxide synthase expression under
normal conditions, and in response to injury, we
speculate that the role of nitric oxide synthase
in preganglionic parasympathetic neurons and
motoneurons may differ.
Acute surgical interventions to protect neurons from
cell death
Previous investigations have demonstrated that a
surgical implantation of peripheral nervous tissue
into the brain or spinal cord may induce regener-
ation of central axons into the implanted graft
(Richardson et al., 1980; David and Aguayo, 1981;
Benfey and Aguayo, 1982; Campbell et al., 1992;
Anderson et al., 1998). It has been recognized that
the regeneration of central axons into the implant-
ed peripheral nerve graft was increased when the
implantation was combined with an axonal injury
(Richardson and Issa, 1984).
Attempts to reduce a ventral root avulsion-
induced death of motoneurons using implanted
grafts have had variable results. A surgical im-
plantation of a peripheral nervous system graft
into the cervical spinal cord of rats following an
avulsion injury of the C7 spinal roots resulted in
an inhibition of the injury-induced expression of
nitric oxide synthase in axotomized motoneuron
and an increased survival of the motoneurons (Wu
et al., 1994). A similar inhibition of nitric oxide
synthase and protection of motoneurons against a
cervical spinal root avulsion-induced retrograde
death has also been described following an acute
implantation of the avulsed ventral root into the
spinal cord in adult rats (Chai et al., 2000). On the
other hand, Novikova et al. (1997) reported that
neither a peripheral nerve graft nor a combina-
tion of the graft and an embryonic spinal cord
implanted into the dorsolateral funiculus of the
injured lumbar spinal cord could prevent the
retrograde motoneuron degeneration, and death
induced by ventral root avulsion in the adult rat. A
recent study investigating the therapeutic window
of opportunity for nerve graft implants demon-
strated that lesioned motoneurons retain their
199
ability to regenerate axons into a peripheral nerve
graft implanted into the spinal cord for up to 3
weeks, after a cervical spinal nerve avulsion injury
in the adult rat (Wu et al., 2004).
Changes in levels of neurotrophins and their
receptors after injury
It has been assumed that the implanted avulsed
nerve roots may produce factors that promote
neuroprotection. Previous studies have shown that
following a peripheral nerve injury, the distal nerve
segment demonstrates a proliferation of Schwann
cells and an increased production of growth fac-
tors, including glial-derived neurotrophic factor,
nerve growth factor, brain-derived neurotrophic
factor, and neurotrophin-4/5 (Salzer and Bunge,
1980; Meyer et al., 1992; Funakoshi et al., 1993;
Naveilhan et al., 1997). It is therefore possible that
the neuroprotective effect of the root implant may
be exerted by either one such neurotrophic factor
or by a combination of growth factors. Spinal cord
neurons may demonstrate injury-induced changes
in their gene expression for trophic factor recep-
tors. Following a ventral root avulsion in the adult
rat, the axotomized motoneurons showed an
increase in their mRNA expression for glial-
derived neurotrophic factor receptors, such as gly-
cosyl-phosphatidylinositol-linked protein receptor
(GFRa-1) and tyrosine kinase receptor c-Retp-
roto-oncogene (c-RET), and the leukemia inhibi-
tory factor receptor, but a decrease in mRNA
expression for tyrosine kinase B (trkB) and trkC
receptors (Hammarberg et al., 2000). From this
perspective it is of interest to note that treatments
using glial-derived neurotrophic factor are neuro-
protective for axotomized motoneurons (Li et al.,
1995; Watabe et al., 2000; Natsume et al., 2002).
Also, treatment with brain-derived neurotrophic
factor, a trkB agonist, following a ventral root
avulsion injury blocked the expression of nitric
oxide synthase in axotomized motoneurons
(Novikov et al., 1997) and reduced the retrograde
motoneuron death (Kishino et al., 1997; Novikov
et al., 1997). In contrast, ciliary neurotrophic fac-
tor had no detectable protective effect on axo-
tomized motoneurons after a ventral root avulsion
200
injury in adult rats (Novikov et al., 1995).
Although extensive information is available on
the gene expression of growth factor receptors in
motoneurons and on the effects of treatment
with a wide variety of neurotrophic factors on
motoneurons, corresponding information for au-
tonomic neurons, specifically preganglionic para-
sympathetic neurons innervating pelvic organs, is
presently very sparse or lacking.
Functional reinnervation of denervated targets using
surgical interventions
A ventral root avulsion injury followed by an
acute surgical implantation of the avulsed ventral
root into the spinal cord results in regenerative
growth of axons by motoneurons, innervation of
the implanted ventral root, and functional rein-
nervation of skeletal muscle in rats (Carlstedt
et al., 1986; Smith and Kodama, 1991; Chai et al.,
2000; Gu et al., 2004), cats (Cullheim et al., 1989;
Hoffmann et al., 1996), and primates (Carlstedt
et al., 1993; Hallin et al., 1999).
The above encouraging studies have also been
translated into the clinic for the repair of avulsed
nerve roots in humans with brachial plexus injuries
(Carlstedt and Norén, 1995; Carlstedt et al., 1995,
2000). A review of the first human cases suggested
that an early intervention with a short lag time
between the injury and restorative surgery is an
important factor for reestablishment of muscle
function, primarily in the most proximal upper
extremity muscle groups (Carlstedt et al., 2000).
In a recent study, the therapeutic opportunities
and clinical outcomes were investigated in 10 pa-
tients after lumbosacral plexus injuries and surgi-
cal interventions (Lang et al., 2004). Most patients
had suffered a severe traumatic injury with a pelvic
fracture and a traction injury to the lumbosacral
plexus. Clinically, lower extremity weakness and
dysfunction were present. In this series of cases,
one of four defined surgical strategies was applied
for each patient. Specifically, an intradural repair
of ventral roots with nerve grafts was performed
when an intradural lesion with both stumps was
present, an intradural–extradural repair with nerve
grafts was performed when an intradural lesion
with proximal stumps only was present, nerve
transfers using a femoral or intercostal nerve were
performed when an intradural lesion with distal
stumps was present, and an intrapelvic repair with
nerve grafts was performed when an intrapelvic
lesion was present. The study concluded that sub-
jects with lesions to the proximal nerve roots and
cauda equina may recover basic lower extremity
functions of unsupported standing and walking
following a surgical reconstruction of spinal
nerves, repair of ventral roots, and nerve transfers
(Lang et al., 2004). No subjects with bladder and
bowel dysfunctions were included in the series,
however.
Surgical attempts involving peripheral nerve or
nerve root grafting techniques to promote a func-
tional reinnervation of the lower urinary tract fol-
lowing a spinal cord injury, have been sparse in the
literature. However, an interesting proof-of-prin-
ciple demonstration was provided in a cat model,
where a skin–CNS–bladder reflex pathway for mi-
cturition was constructed following a spinal cord
injury by performing a microanastomosis of the
L7–S1 ventral root, while leaving the dorsal root
intact (Xiao et al., 1999). The cross-wired somato-
autonomic bladder reflex was effective in initiating
bladder contractions and coordinated voiding in
cats with an intact spinal cord, as well as inducing
bladder contractions after an acute spinal cord
transection injury (Xiao et al., 1999). These results
are of particular interest for future possible
lumbosacral ventral root repair interventions, in
part because the findings show that somatic moto-
neurons can innervate a parasympathetic nerve
and functionally innervate the bladder.
Strategies to treat chronic conus medullaris and
cauda equina injuries
Because of the progressive and parallel retrograde
cell death of both preganglionic parasympathetic
neurons and motoneurons that follows lesions to
their axons at the ventral root exit zone (Hoang
et al., 2003), neuroprotective strategies and at-
tempts to induce axonal regeneration in lesioned
efferent spinal cord neurons are likely to have a
specific time window of therapeutic opportunity.

Therefore, subjects with a chronic conus medulla-
ris syndrome face a different set of challenges, as
the lesioned efferent neurons in the conus medul-
laris may have become largely depleted. In this
setting, alternative treatment strategies need to be
considered. Given the injury-induced neuronal
degeneration and death of preganglionic parasym-
pathetic neurons and motoneurons, a cell replace-
ment approach appears attractive.
Much interest has been generated for the po-
tential use of pluripotent and multipotent stem
cells in the treatment of neurological disorders,
including spinal cord injury. Multipotent neural
stem cells have been isolated from both the fetal
and adult human nervous system (Johansson et al.,
1999; Vescovi et al., 1999; Villa et al., 2000; Palmer
et al., 2001; Westerlund et al., 2003). Although
neural stem cells have demonstrated their ability to
differentiate into both neurons and glial cells in
vitro, most stem cells are restricted to a glial
lineage and very few stem cells become neurons
when transplanted into the spinal cord (McDonald
et al., 1999; Cao et al., 2001). For some neural
repair strategies, the apparent preferred develop-
ment of neural stem cells into a glial lineage may
be a good feature, for instance when remyelination
of lesioned ascending and descending white matter
tracts is desired. However, the limited yield of
neurons derived from transplanted neural stem
cells into the spinal cord presents a potential
problem for purposes of replacing degenerated
and lost spinal cord neurons in subjects with a
conus medullaris syndrome.
Interestingly, recent studies have demonstrated
that an in vitro priming procedure of fetal human
neural stem cells can markedly increase the per-
centage of stem cells that differentiate into neurons
when transplanted into the spinal cord (Wu et al.,
2002). A large portion of the primed cells, which
were grafted into the spinal cord, demonstrated
immunoreactivity for choline acetyltransferase
(Wu et al., 2002). The ability to influence the
differentiation of grafted cells into choline acetylt-
ransferase-immunoreactive neurons in the spinal
cord is of particular interest when considering cell
replacement strategies for the treatment of conus
medullaris and cauda equina injuries, as both
preganglionic parasympathetic neurons and
201
motoneurons normally exhibit a cholinergic
phenotype (Barber et al., 1984). It is likely that a
cell replacement strategy for the treatment of a
chronic conus medullaris and cauda equina injury
will also need to be combined with other thera-
peutic interventions, especially with a peripheral
nerve or root grafting procedure to provide a con-
duit for the extension of central axons of grafted
cells into the peripheral nervous system and for
ultimate functional reinnervation of pelvic targets,
e.g., the lower urinary tract.
Assessments of lower urinary tract function
following acute and chronic treatment interventions
In future animal studies on neural repair following
a traumatic injury to the conus medullaris and/or
cauda equina, it is important to be able to assess
the effects of the treatment intervention. In rodent
studies, a variety of functional and end-organ as-
sessment methods are available and used to eval-
uate the lower urinary tract under normal and
pathological conditions. These methods include
bladder expressions to assess for urinary retention,
behavioral studies on voiding patterns, urodynam-
ic studies, metabolic cage assessments, as well as
postmortem bladder weight and size measure-
ments (Maggi et al., 1986; Chancellor et al.,
1994; Pikov et al., 1998; Kerns et al., 2000; Pikov
and Wrathall, 2001).
Conclusions
Conus medullaris and cauda equina injuries com-
monly result in severe neurological impairments,
including paralysis, sensory disturbance, pain, as
well as bladder, bowel, and sexual dysfunctions. In
recent years, a more comprehensive understanding
of the underlying neurobiology and pathophysiology
has been emerging. A post-traumatic progressive
loss of efferent somatic and autonomic neurons is
an important effect of conus medullaris and cauda
equina injuries in experimental models. Early
interventions to decrease the retrograde death of
axotomized neurons are likely to remain important
in future translational research efforts. For acute
and subacute injuries, surgical strategies, where
202
peripheral nerve grafts or lesioned nerve roots are
implanted into the lumbosacral spinal cord, may
also be applied in select subjects to promote func-
tional reinnervation of peripheral targets, includ-
ing the lower urinary tract. A combination of
medical and surgical treatments to increase neuro-
protection, axonal regeneration and peripheral
target reinnervation may also be considered. In
chronic injuries, cell replacement strategies, rather
than neuroprotective approaches, may become
more viable options, as many efferent spinal cord
neurons may already have undergone retrograde
cell death. However, cell replacement strategies
may still need to be combined with e.g., a periph-
eral nerve or nerve root grafting procedure to fa-
cilitate and promote functional reinnervation of
peripheral targets.
Acknowledgments
Supported by research grants from NIH
(NS042719), The Paralysis Project of America,
The State of California Roman Reed Bill, and
SCORE.
References
Anderson, P.N., Campbell, G., Zhang, Y. and Lieberman, A.R.
(1998) Cellular and molecular correlates of the regeneration
of adult mammalian CNS axons into peripheral nerve grafts.
Prog. Brain Res., 117: 211–232.
Anneser, J.M.H., Berhele, A., Borasio, G.D., Castro-Lopez,
J.M., Zieglgänsberger, W. and Tölle, T.R. (2000) Axotomy of
the sciatic nerve differentially affects expression of metabo-
tropic glutamate receptor mRNA in adult motoneurons.
Brain Res., 868: 215–221.
Barber, R.P., Phelps, P.E., Houser, C.R., Crawford, G.D.,
Salvaterra, P.M. and Vaughn, J.E. (1984) The morphology
and distribution of neurons containing choline acetyltransf-
erase in the adult rat spinal cord: an immunohistochemical
study. J. Comp. Neurol., 229: 329–346.
Benfey, M. and Aguayo, A.J. (1982) Extensive elongation of
axons from rat brain into peripheral nerve grafts. Nature,
296: 150–152.
Beric, A. and Light, J.K. (1992) Detrusor function with lesions
of the conus medullaris. J. Urol., 148: 104–106.
Campbell, G., Lieberman, A.R., Anderson, P.N. and
Turmaine, M. (1992) Regeneration of adult rat CNS axons
into peripheral nerve autografts: ultrastructural studies of the
early stages of axonal sprouting and regenerative axonal
growth. J. Neurocytol., 21(11): 755–787.
Cao, Q.L., Zhang, Y.P., Howard, R.M., Walters, W.M.,
Tsoulfas, P. and Whittemore, S.R. (2001) Pluripotent stem
cells engrafted into the normal or lesioned adult rat spinal
cord are restricted to a glial lineage. Exp. Neurol., 167:
48–58.
Carlson, J., Lais, A.C. and Dyck, P.J. (1979) Axonal atrophy
from permanent peripheral axotomy in adult cat. J. Ne-
uropathol Exp. Neurol., 38(6): 579–585.
Carlstedt, T., Anand, P., Hallin, R.G., Misra, P.V., Norén, G.
and Seferlis, T. (2000) Spinal root repair and reimplantation
of avulsed ventral roots into the spinal cord after brachial
plexus injury. J. Neurosurg., 93: 237–247.
Carlstedt, T., Grane, P., Hallin, R.G. and Norén, G. (1995)
Return of function after spinal cord implantation of avulsed
roots. Lancet, 346: 1323–1325.
Carlstedt, T., Lindå, H., Cullheim, S. and Risling, M. (1986)
Reinnervation of hindlimb muscles after ventral root avuls-
ion and implantation in the lumbar spinal cord in the adult
rat. Acta Physiol. Scand., 128: 645–646.
Carlstedt, T. and Norén, G. (1995) Repair of ruptured
spinal nerve roots in a brachial plexus lesion. Case report.
J. Neurosurg., 82(4): 661–663.
Carlstedt, T.P., Hallin, R.G., Hedström, K.G. and Nilsson-
Remahl, I.A.M. (1993) Functional recovery in primates with
brachial plexus injury after spinal cord implantation of avu-
lsed ventral roots. J. Neurol. Neurosurg. Psychiat., 56:
649–654.
Chai, H., Wu, W., So, K.-F. and Yip, H.K. (2000) Survival and
regeneration of motoneurons in adult rats by reimplantation
of ventral root following spinal root avulsion. Neuroreport,
11: 1249–1252.
Chan, Y.M., Wu, W., Yip, H.K., So, K.F. and Oppenheim,
R.W. (2001) Caspase inhibitors promote the survival of avu-
lsed spinal motoneurons in neonatal rats. Neuroreport, 12:
541–545.
Chancellor, M.B., Rivas, D.A., Huang, B., Kelly, G. and
Salzman, S.K. (1994) Micturition patterns after spinal
trauma as a measure of autonomic functional recovery.
J. Urol., 151(1): 250–254.
Cullheim, S., Carlstedt, T., Lindå, H., Risling, M. and Ulfhake,
B. (1989) Motoneurons reinnervate skeletal muscle after ven-
tral root implantation into spinal cord of the cat. Neurosci-
ence, 29: 725–734.
David, S. and Aguayo, A.J. (1981) Axonal elongation into
peripheral nervous system ‘‘bridges’’ after central nervous
system injury in rats. Science, 214: 931–933.
Fowler, C.J. (1999) Neurological disorders of micturition and
their treatment. Brain, 122: 1213–1231.
Fowler, C.J., Kirby, R.S., Harrison, M.J., Milroy, E.J. and
Turner-Warwick, R. (1984) Individual motor unit analysis in
the diagnosis of disorders of urethral sphincter innervation.
J. Neurol. Neurosurg. Psychiat., 47(6): 637–641.
Funakoshi, H., Frisén, J., Barbany, G., Timmusk, T.,
Zachrisson, O., Verge, V.M. and Person, H. (1993)
Differential expression of mRNA for neurotrophins and
their receptors after axotomy of the sciatic nerve. J. Cell
Biol., 123: 455–465.

Gu, H.-Y., Chai, H., Zhang, J.-Y., Yao, Z.-B., Zhou, L.-H.,
Wong, W.-M., Bruce, I. and Wu, W.-T. (2004) Survival,
regeneration and functional recovery of motoneurons in
adult rats by implantation of ventral root following spinal
root avulsion. Eur. J. Neurosci., 19: 2123–2131.
Hallin, R.G., Carlstedt, T., Nilsson-Remahl, I. and Risling, M.
(1999) Spinal cord implantation of avulsed ventral roots in
primates; correlation between restored motor function and
morphology. Exp. Brain Res., 124: 304–310.
Hammarberg, H., Piehl, F., Risling, M. and Cullheim, S. (2000)
Differential regulation of trophic factor receptor mRNAs in
spinal motoneuons after sciatic nerve transection and ventral
root avulsion in the rat. J. Comp. Neurol., 426: 587–601.
Hans, F.J., Reinges, M.H. and Krings, T. (2004) Lumbar nerve
root avulsion following balanced fast field echo MRI.
Neuroradiology, 46: 144–147.
Hoang, T.X., Nieto, J.H., Tillakaratne, N.J.K. and Havton,
L.A. (2003) Autonomic and motor neuron death is progres-
sive and parallel in a lumbosacral ventral root avulsion model
of cauda equine injury. J. Comp. Neurol., 426: 587–601.
Hoffmann, C.F.E., Marani, E., Van Dijk, J.G., Kamp, W.V.D.
and Thoemer, R.T.W.M. (1996) Reinnervation of avulsed
and reimplanted ventral rootlets in the cervical spinal cord of
the cat. J. Neurosurg., 84: 234–243.
Holmberg, P. and Kellerth, J.O. (2000) Do synaptic rearrange-
ments underlie compensatory reflex enhancement in spinal
motoneurons after partial cell loss? Synapse, 38(4): 384–391.
Johnson, I.P., Sears, T.A. and Hunter, A.S. (1991) Retrograde
response to axotomy of motoneurons in the thoracic spinal
cord of the aging cat. Neurobiol. Aging, 12(2): 151–160.
Johansson, C.B., Svensson, M., Wallstedt, L., Janson, A.M.
and Frisén, J. (1999) Neural stem cells in the adult human
brain. Exp. Cell Res., 253: 733–736.
Kerns, J.M., Damaser, M.S., Kane, J.M., Sakamoto, K.,
Benson, J.T., Shott, S. and Brubaker, L. (2000) Effects of
pudendal nerve injury in the female rat. Neurourol. Urodyn.,
19(1): 53–69.
Kishino, A., Ishige, Y., Tatsuno, T., Nakayama, C. and
Noguchi, H. (1997) BDNF prevents and reverses adult rat
motor neuron degeneration and induces axonal outgrowth.
Exp. Neurol., 144(2): 273–286.
Koliatsos, V.E., Price, W.L., Pardo, C.A. and Price, D.L.
(1994) Ventral root avulsion: an experimental model of death
of adult motor neurons. J. Comp. Neurol., 342(1): 35–44.
Lang, E.M., Borges, J. and Carlstedt, T. (2004) Surgical treat-
ment of lumbosacral plexus injuries. J. Neurosurg. (Spine 1),
1: 64–71.
Li, L., Houenou, L.J., Wu, W., Lei, M., Prevette, D.M. and
Oppenheim, R.W. (1998) Characterization of spinal moto-
neuron degeneration following different types of peripheral
nerve injury in neonatal and adult mice. J. Comp. Neurol.,
396: 158–168.
Li, L., Wu, W., Lin, L.F., Lei, M., Oppenheim, R.W. and
Houenou, L.J. (1995) Rescue of adult mouse motoneurons
from injury-induced cell death by glial cell line-derived
neurotrophic factor. Proc. Natl. Acad. Sci., 92(21):
9771–9775.
203
Light, J.K., Beric, A. and Petronic, I. (1993) Detrusor
dysfunction with lesions of the cauda equina, with special
emphasis on the bladder neck. J. Urol., 149(3): 539–542.
Lindå, H., Cullheim, S. and Risling, M. (1992) A light and
electron microscopic study of intracellularly HRP-labeled
lumbar motoneurons after intramedullary axotomy in the
adult cat. J. Comp. Neurol., 318(2): 188–208.
Liu, Z. and Martin, L.J. (2001) Motor neurons rapidly accu-
mulate DNA single-strand breaks after in vitro exposure to
nitric oxide and peroxynitrite and in vivo axotomy. J. Comp.
Neurol., 432: 35–60.
Ma, J., Novikov, L.N., Wiberg, M. and Kellerth, J.-O. (2001)
Delayed loss of spinal motoneurons after peripheral nerve
injury in adult rats: a quantitative morphological study. Exp.
Brain Res., 139: 216–223.
Maggi, C.A., Santicioli, P. and Meli, A. (1986) The nonstop
transvesical cystometrogram in urethane-anesthetized rats: a
simple procedure for quantitative studies on the various
phases of urinary bladder voiding cycle. J. Pharmacol. Meth-
ods, 15(2): 157–167.
Martin, L.J., Kaiser, A. and Price, A.C. (1999) Motor neuron
degeneration after sciatic nerve avulsion in adult rat evolves
with oxidative stress and is apoptosis. J. Neurobiol., 40:
185–201.
Martin, L.J. and Liu, Z. (2002) Injury-induced spinal
motor neuron apoptosis preceeded by DNA single-strand
breaks and is p53- and Bax-dependent. J. Neurobiol.,
50: 181–197.
Maynard, F.M., Braken, M.B., Creasey, G., Ditunno, J.F.,
Donovan, W.H., Ducker, T.B., Garber, S.L., Marino, R.J.,
Stover, S.L., Tator, C.H., Waters, R.L., Wilberger, J.E. and
Young, W. (1997) International Standards for Neurological
and Functional Classification of Spinal Cord Injury. Spinal
Cord, 35: 266–274.
McDonald, J.W., Liu, X.Z., Qu, Y., Liu, S., Mickey, S.K.,
Turetsky, D., Gottlieb, D.I. and Choi, D.W. (1999) Trans-
planted embryonic stem cells survive, differentiate and pro-
mote recovery in injured rat spinal cord. Nat. Med., 5:
1410–1412.
Meyer, M., Matsuoka, I., Wetmore, C., Olson, L. and Thoenen,
H. (1992) Enhanced synthesis of brain-derived neurotrophic
factor in the lesioned peripheral nerve: different mechanisms
are responsible for the regulation of BDNF and NGF
mRNA. J. Cell Biol., 119: 45–54.
Monga, P., Ahmed, A., Gupta, G.R. and Rao, K.S. (2004)
Traumatic lumbar nerve root avulsion: evaluation using
electrodiagnostic studies and magnetic resonance myelo-
graphy. J. Trauma, 56: 182–184.
Moschilla, G., Song, S. and Chakera, T. (2001) Post-traumatic
lumber nerve avulsion. Australas. Radiol., 45: 281–284.
Natsume, A., Mata, M., Wolfe, D., Oligino, T., Goss, J.,
Huang, S., Glorioso, J. and Fink, D.J. (2002) Bcl-2 and
GDNF delivered by HSV-mediated gene transfer after spinal
root avulsion provide a synergistic effect. J. Neurotrauma,
19(1): 61–68.
Naveilhan, P., ElShamy, W.M. and Ernfors, P. (1997) Differ-
ential regulation of mRNAs for GDNF and its receptors Ret
204
and GDNFR alpha after sciatic nerve lesion in the mouse.
Eur. J. Neurosci., 9(7): 1450–1460.
Novikov, L., Novikova, L. and Kellerth, J.-O. (1995) Brain-
derived neurotrophic factor promotes survival and blocks
nitric oxide synthase expression in adult rat spinal motoneu-
rons after ventral root avulsion. Neurosci. Lett., 200: 45–48.
Novikov, L., Novikova, L. and Kellerth, J.-O. (1997) Brain-
derived neurotrophic factor promotes axonal regeneration
and long-term survival of adult rat spinal motoneurons in
vivo. Neuroscience, 79: 765–774.
Novikov, L.N., Novikova, L.N., Holmberg, P. and Kellerth,
J.-O. (2000) Exogenous brain-derived neurotrophic factor
regulates the synaptic composition of axonally lesioned and
normal adult rat motoneurons. Neuroscience, 100(1): 171–181.
Novikova, L., Novikov, L. and Kellerth, J.-O. (1997) Effects of
neurotransplants and BDNF on the survival and regenera-
tion of injured adult spinal motoneurons. Eur. J. Neurosci.,
9: 2774–2777.
Palmer, T.D., Schwartz, P.H., Taupin, P., Kaspar, B., Stein,
S.A. and Gage, F.H. (2001) Progenitor cells from human
brain after death. Nature, 411: 42–43.
Pavlakis, A.J., Siroky, M.B., Goldstein, I. and Krane, R.J.
(1983) Neurourologic findings in conus medullaris and cauda
equina injury. Arch. Neurol., 40: 570–573.
Piehl, F., Tabar, G. and Cullheim, S. (1995) Expression of
NMDA receptor mRNAs in rat motoneurons is down-reg-
ulated after axotomy. Eur. J. Neurosci., 7: 2101–2110.
Pikov, V., Gillis, A., Jasmin, L. and Wrathall, J.R. (1998) As-
sessment of lower urinary tract functional deficit in rats with
contusive spinal cord injury. J. Neurotrauma, 15(5): 375–386.
Pikov, V. and Wrathall, J.R. (2001) Coordination of the
bladder detrusor and the external urethral sphincter in a rat
model of spinal cord injury: effect of injury severity.
J. Neurosci., 21(2): 559–569.
Richardson, P.M. and Issa, V.M. (1984) Peripheral injury
enhances central regeneration of primary sensory neurons.
Nature, 309: 791–793.
Richardson, P.M., McGuinness, U.M. and Aguayo, A.J. (1980)
Axons from CNS neurons regenerate into PNS grafts.
Nature, 284(5753): 264–265.
Saito, S., Kidd, G.J., Trapp, B.D., Dawson, T.M., Bredt, D.S.,
Wilson, D.A., Traystman, R.J., Snyder, S.H. and Hanley,
D.F. (1994) Rat spinal cord neurons contain nitric oxide
synthase. Neuroscience, 59(2): 447–456.
Salzer, J.L. and Bunge, R.P. (1980) Studies of Schwann cell
proliferation I. An analysis in tissue culture of proliferation
in development, Wallerian degeneration, and direct injury.
J. Cell Biol., 84: 739–752.
Schrøder, H.D. (1980) Organization of the motoneurons
innervating the pelvic muscles of the male rat. J. Comp.
Neurol., 192(3): 567–587.
Smith, K.J. and Kodama, R.T. (1991) Reinnervation of den-
ervated skeletal muscle by central neurons regenerating via
ventral roots implanted into the spinal cord. Brain Res., 551:
221–229.
Vanden Noven, S., Wallace, N., Muccio, D., Turtz, A. and
Pinter, M.J. (1993) Adult spinal motoneurons remain viable
despite prolonged absence of functional synaptic contact with
muscle. Exp. Neurol., 123: 147–156.
Vescovi, A.L., Parati, E.A., Gritti, A., Poulin, P., Ferrario, M.,
Wanke, E., Frolichstal-Schoeller, P., Cova, L., Arcellana-
Panlilio, M., Colombo, A. and Galli, R. (1999) Isolation and
cloning of multipotential stem cells from the embryonic hu-
man CNS and establishment of transplantable human neural
stem cell lines by epigenetic stimulation. Exp. Neurol., 56(1):
71–83.
Villa, A., Snyder, E.Y., Vescovi, A. and Martinez-Serrano, A.
(2000) Establishment and properties of a growth factor-
dependent, perpetual neural stem cell line from the human
CNS. Exp. Neurol., 161: 67–84.
Vizzard, M.A., Erdman, S.L. and de Groat, W.C. (1993a)
Localization of NADPH-diaphorase in pelvic afferent and
efferent pathways of the rat. Neurosci. Lett., 152: 72–76.
Vizzard, M.A., Erdman, S.L. and de Groat, W.C. (1993b) The
effect of rhizotomy on NADPH diaphorase staining in the
lumbar spinal cord of the rat. Brain Res., 607: 349–353.
Vizzard, M.A., Erdman, S.L. and de Groat, W.C. (1995) In-
creased expression of neuronal nitric oxide synthase (NOS) in
visceral neurons after nerve injury. J. Neurosci., 15: 4033–4045.
Watabe, K., Ohashi, T., Sakamoto, T., Kawazoe, Y.,
Takeshima, T., Oyanagi, K., Inoue, K., Eto, Y. and
Kim, S.U. (2000) Rescue of lesioned adult rat spinal moto-
neurons by adenoviral gene transfer of glial cell line-derived
neurotrophic factor. J. Neurosci. Res., 60(4): 511–519.
Westerlund, U., Moe, M.C., Varghese, M., Berg-Johnsen, J.,
Ohlsson, M., Langmoen, I.A. and Svensson, M. (2003) Stem
cells from the adult human brain develop into functional
neurons in culture. Exp. Cell Res., 289: 378–383.
Wu, C.W. and Kaas, J.H. (2000) Spinal cord atrophy and re-
organization of motoneuron connections following long-
standing limb loss in primates. Neuron, 28: 967–978.
Wu, P., Tarasenko, Y.I., Gu, Y., Huang, L.-Y.M., Coggeshall,
R.E. and Yu, Y. (2002) Region-specific generation of
cholinergic neurons from fetal human neural stem cells graft-
ed in adult rat. Nat. Neurosci., 5: 1271–1278.
Wu, W. (1993) Expression of nitric-oxide synthase (NOS) in
injured CNS neurons as shown by NADPH diaphorase his-
tochemistry. Exp. Neurol., 120: 153–159.
Wu, W., Chai, H., Zhang, J., Gu, H., Xie, Y. and Zhou, L.
(2004) Delayed implantation of a peripheral nerve graft re-
duces motoneurons survival but does not affect regeneration
following spinal root avulsion in adult rats. J. Neurotrauma,
21: 1050–1058.
Wu, W., Han, K., Li, L. and Schinco, F.P. (1994) Implantation
of PNS graft inhibits the induction of neuronal nitric oxide
synthase and enhances the survival of spinal motoneurons
following root avulsion. Exp. Neurol., 129: 335–339.
Wu, W. and Li, L. (1993) Inhibition of nitric oxide synthase
reduces motoneuron death due to spinal root avulsion.
Neurosci. Lett., 153(2): 121–124.
Xiao, C.-G., de Groat, W.C., Godec, C.J., Dai, C. and Xiao, Q.
(1999) ‘‘Skin–CNS–bladder’’ reflex pathway for micturition
after spinal cord injury and its underlying mechanisms.
J. Urol., 162: 936–942.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 13

Pelvic somato-visceral reflexes after spinal cord

injury: measures of functional loss and
partial preservation

Michael D. Craggs

Centre for Spinal Research, Functional Assessment and Restoration, London Spinal Cord Injuries Unit, Royal National
Orthopaedic Hospital NHS Trust, Brockley Hill, Stanmore, Middlesex HA7 4LP, UK

Abstract: For people with spinal cord injuries, the impact of bladder, bowel and sexual problems on quality
of life and lost opportunities can be devastating. Supra-sacral spinal lesions can cause incontinence by
interrupting those pathways that normally coordinate the function of the bladder, bowel and sphincters.
From a scientific perspective, neural control of the pelvic organs is one of the most intriguing in the body,
involving both somatic and autonomic pathways participating in an exquisitely fine integration of lumbo-
sacral reflexes. This chapter aims to review briefly those aspects of neural control of the pelvic organs that
are amenable to neurophysiological examination in man. More specifically, it will focus in greater detail on
the interactions of somatic and autonomic lumbo-sacral pathways responsible for coordinating the bladder
and sphincters. Where appropriate, it will make comparisons with those controlling the bowel. It will
describe how measurement of pelvic floor and sphincter reflexes can be used to assess the modulatory
effects of sacral autonomic pathways on sacral somatic reflexes and vice versa including the so-called
‘‘guarding reflex’’ and vesical inhibitory reflexes. Aberrant activity following spinal cord injury (SCI), such
as bladder hyperreflexia and sphincter dyssynergia, will be discussed in relation to these reflexes. The effects
of volitional modulation of pelvic floor reflexes in people with both complete and incomplete lesions will be
described. Finally, the chapter will address the possible utility of neurophysiological measures for com-
plementing the established neurological classification and the assessment of somatic sensory-motor im-
pairment in SCI.

Introduction Supra-sacral spinal lesions cause incontinence

For people with spinal cord injuries, the impact of
bladder, bowel and sexual problems on quality of
life and lost opportunities can be devastating. The
social stigma of urinary and fecal incontinence
make these problems some of the most seriously
debilitating conditions needing long-term care and
management in spinal cord injury (SCI).
Corresponding author. Tel./fax: +44 (0)20 8909 5343 (Hos-
pital), +44 (0)20 7679 9379 (University);
E-mail: michael.craggs@ucl.ac.uk
by interrupting those pathways that normally co-
ordinate the bladder, bowel and sphincter func-
tions. Such lesions are conventionally termed
upper motor neuron lesions. Lesions of the sacral
cord or nerve roots, for example, cauda equina
(lumbar-sacral nerve roots) damage or disease, are
conventionally termed lower motor neuron lesions
as they disconnect bladder, part of the bowel and
their sphincters from their source of innervation in
the spinal cord. Depending on their completeness,
these lesions can result in total loss of reflex con-
trol of the bladder, bowel and sphincters, again

DOI: 10.1016/S0079-6123(05)52013-2 205

206
leading to serious problems of storage and void-
ing.
From a scientific perspective, neural control of
the pelvic organs is one of the most intriguing in
the body, involving both somatic and autonomic
pathways participating in an exquisitely fine inte-
gration of lumbo-sacral reflexes. Volition can play
a significant part in this modulation. In adult man,
segmental reflexes are under fine tuning by mod-
ulatory pathways from various levels of the brain,
from brain stem to cerebral cortex. For the lower
urinary tract, precise coordination between the
bladder detrusor muscle, urethral smooth muscles
and voluntary muscle of the urethral sphincter is
essential for maintaining continence over long pe-
riods of time and voiding efficiently when neces-
sary.
Through a comprehensive knowledge of the
physiological and neurophysiological mechanisms
controlling the pelvic organs, we can hope to de-
velop more reliable and sensitive diagnostic tools
to assess new interventions for treating pelvic dys-
function.
This chapter aims to review briefly those aspects
of neural control of the pelvic organs that are
amenable to neurophysiological examination in
man. More specifically, it will focus in greater de-
tail on the interactions of somatic and autonomic
lumbo-sacral pathways responsible for coordinat-
ing the bladder and sphincters, but make compar-
isons, where appropriate, with those controlling
the bowel. It will examine the modulatory effects
of sacral autonomic pathways on sacral somatic
reflexes and vice versa. It will compare these inter-
actions using neurophysiological measures follow-
ing SCI and show the effects of volitional
modulation in people with both complete and in-
complete lesions. Finally, it will discuss the possi-
ble utility of such neurophysiological measures for
complementing the established neurological clas-
sification and the assessment of somatic sensory-
motor impairment in SCI.
Neural control of the pelvic viscera and pelvic floor
Normal micturition and defecation are initiated
voluntarily at an appropriate and socially conven-
ient time when sensory signals indicate fullness in
the bladder or rectum, respectively. Although the
sensations of fullness and the strong desire to mi-
cturate or defecate are sometimes difficult to sup-
press, micturition and defecation can be suspended
for some length of time, and continence main-
tained even when the bladder and rectum are very
full. The mechanisms of control in man are
brought about through a balance of modulatory
influences at many levels in the neuraxis from
brain to the sacral (S) cord and pelvic plexus.
Peripheral innervation
The bladder and colo-rectum, together with their
respective sphincters, show many similarities in
muscular organization and peripheral extrinsic
nerve supply. However, their functions are very
different and depend on their unique neuro-ana-
tomical specialization. The colo-rectum has a well-
developed enteric (intrinsic) nervous system that
coordinates motility of its smooth muscle, secre-
tion and absorption and is modulated by extrinsic
nerves. The bladder, in contrast, is controlled al-
most exclusively by extrinsic nerves so that con-
traction of its smooth muscle, the detrusor, is a
coordinated event for efficient emptying. The de-
trusor remains quiescent at all other times while
the bladder is filling.
The bowel
The intrinsic nerve supply to the colo-rectum is
derived from the myenteric and submucosal plex-
uses. These plexuses have a complex structural or-
ganization and involve the interactions of many
different neurotransmitter pathways, including the
modulating influences of the extrinsic nerve supply
with its sympathetic and parasympathetic nerves
(Burnstock, 1990). The plexuses in the rectum and
distal colon are dense and irregular compared to
those in the more proximal colon. Sacral para-
sympathetic innervation of the rectum and distal
colon is clearly an important aspect of the extrinsic
control of defecation. Outlet function is also very
specialized (Mathers, 1992). The ano-rectum is
able to distinguish solid, liquid and gas content by
a complex sampling mechanism and to facilitate

selective passage of gas when appropriate, whereas
urethro-vesical muscle function is very well coor-
dinated, so that even the smallest leakage of urine
is prevented at inappropriate times.
The bladder
The bladder detrusor smooth muscle and smooth
muscle of the distal rectal wall receive parasympa-
thetic innervation from preganglionic neurons in
the intermediolateral column of the S2–S4 sacral
segments of the spinal cord via post-ganglionic
neurons originating in ganglia of the pelvic plexus
or in the bladder wall itself (Burnstock, 1990). In-
teracting at the peripheral ganglia are sympathetic
efferent fibers of the hypogastric nerve plexus (in-
ferior mesenteric ganglia). The sympathetic path-
way has its origin in the intermediolateral column
of the 9th thoracic (T) to 2nd lumbar (L) segments
of the spinal cord. The preganglionic axons synapse
first in the pre-vertebral ganglia or lumbosacral
sympathetic chain ganglia (De Groat, 1997).
The pelvic floor
Muscles of the pelvic floor, including the peri-ure-
thral and peri-anal muscles (for example, levator
ani and pubo-rectalis) and urethral and anal stri-
ated sphincters, are innervated from the sacral
cord. However, only the specialized sphincter mus-
cles receive their peripheral innervation via the
pudendal nerves. All of these somatic neural path-
ways arise from motor neurons in Onuf’s nucleus,
a specialized group of anterior horn cells spanning
the second, third and fourth sacral spinal seg-
ments. Unlike the anal canal, the urethral wall
contains a predominantly slow-twitch striated
muscle sphincter known as the rhabdosphincter.
This sphincter is innervated by small motor neu-
rons in Onuf’s nucleus that is situated close to the
intermediolateral neurons of the preganglionic
parasympathetic pathway to the bladder detrusor
muscle. This close anatomic relationship has func-
tional significance for close somato-visceral inte-
gration in the lower urinary tract (Mundy and
Thomas, 1994). Interestingly, unlike the external
anal sphincter, the rhabdosphincter is said to have
207
no muscle spindles and, is therefore quite distinct
from all other striated muscles of the pelvic floor
(Schrøder and Reske-Nielsen, 1983). The other
muscles of the pelvic floor are innervated by
branches of the sacral anterior roots. Of these
muscles, the levator ani group comprises a mixture
of slow-twitch and fast-twitch fibers that function-
ally provide sustained tone and phasic anti-stress
contractions respectively to give peri-urethral and
peri-anal occlusion.
The sensory pathways
Pelvic visceral nerves are the main sensory path-
ways from the bladder and rectum, but some sen-
sory information is also conveyed in the
sympathetic hypogastric nerves to the thoraco-
lumbar spinal cord, especially from the colorec-
tum, bladder neck and proximal urethra (Jänig
and Morrison, 1986). Sensory information from
the distal urethra, anal canal and perineum is car-
ried almost exclusively by the pudendal nerves.
Essential to proper coordination of the bladder,
bowel and sphincters is the reflex circuitry of the
lumbo-sacral spinal segments. Modulation is by
supra-spinal pathways from the brain-stem and
diencephalon together with volitional pathways
from the cerebral cortex (Craggs and Vaizey,
1999).
Coordination of the lower urinary tract
The coordinated function of the urinary bladder
and urethral sphincters depends on the complete
integrity of central and peripheral neural circuitry
in a complex control system located in the brain
and spinal cord. These functions have been inves-
tigated extensively in recent times through a com-
bination of urodynamics, neurophysiology and
sophisticated imaging techniques (Blok and Will-
emsen, 1997).
In mature adults, this integrated control system
is believed to act like a switching circuit (De
Groat, 1997) but under close cognitive and voli-
tional supervision (Blok and Willemsen, 1997).
Ascending afferent activity from the lower urinary
tract is routed both to those parts of the brain-
detecting sensation and to the pontine micturition
208
center through relay neurons in the peri-aqueduct-
al gray of the brain stem. This center appears to
maintain an appropriate reciprocal relationship
between lumbo-sacral reflexes so as to coordinate
the bladder and the urethral sphincters during
storage and voiding (Yoshimura et al., 2004).
Storage reflexes
During bladder filling, a number of lumbo-sacral
reflex pathways appear to be active, ensuring com-
petent urethral closure together with detrusor in-
hibition to maintain continence. As the bladder
slowly fills and stores urine, any tendency for
spontaneous reflex contractions of the detrusor
smooth muscle in the bladder wall, due to rising
pelvic afferent nerve activity, is inhibited by a
combination of sympathetically mediated reflexes
suppressing detrusor activity and contracting the
smooth muscle of the bladder neck and proximal
urethra (Yoshimura et al., 2004). As the bladder
volume increases, the striated urethral sphincter
also contracts automatically via a pelvic afferent to
pudendal efferent reflex to prevent leakage, a
mechanism sometimes known as the ‘‘guarding
reflex’’ (Park et al., 1997). Evidence for this mech-
anism is seen as a build-up in the sphincter elect-
romyogram (EMG) during bladder filling. Such
pudendal activity probably further inhibits any
tendency for aberrant parasympathetic activity
(Craggs and MacFarlane, 1999). When the bladder
reaches its maximum capacity and there is a strong
desire to void, then voluntary control of the stri-
ated urethral sphincter and other pelvic floor mus-
cles can come into play to preserve continence
until a suitable place for micturition is found.
Voiding reflexes
Voiding, heralded by an abrupt cessation of the
striated sphincter EMG, is brought about by four
synergistic reflexes (Yoshimura et al., 2004). Two
of these cause relaxation of the bladder neck
smooth muscle and striated urethral sphincter, re-
spectively before a third reciprocal action gives a
powerful detrusor contraction, sustained by a
fourth, a urethral to bladder facilitatory reflex,
enabling urine to be expelled quickly and efficient-
ly from the bladder until empty. The detrusor
contraction relies on the descending drive from the
pontine medial nucleus to the parasympathetic
motoneurons in the sacral cord being turned ‘‘on’’
(Blok et al., 1997), while activity from the pontine
lateral nucleus modulating the ‘‘guarding reflex’’ is
switched off.
Coordination of the bowel
Although we have considerable knowledge about
the control of the urinary bladder and its sphinc-
ters, much less is known about the precise role of
spinal reflex mechanisms and central modulatory
effects on bowel function. It is the enteric nervous
system, both in the colon and rectum, which plays
a major role in the coordination of these parts of
the bowel. Unlike the bladder, the rectum is only a
very temporary reservoir for the feces propelled
into it by peristalsis from the colon. Furthermore,
the anal canal has a very distinct internal sphincter
comprising smooth muscle, which provides conti-
nence, but which relaxes automatically when the
rectum fills. This recto-anal response is not a spinal
reflex, but appears to depend on a local nitric ox-
ide-mediated mechanism (Burleigh, 1992). Howev-
er, whereas the enteric nervous system plays the
major role in coordinating ano-rectal function, the
extrinsic autonomic influences of the sacral para-
sympathetic pathway through the pelvic nerves and
lumbar sympathetic pathways via the inferior mes-
enteric plexus and hypogastric nerves, together
with their central connections via the brain stem,
do play an important part in the modulation of
these intrinsic effects in the normal bowel (Craggs
and Vaizey, 1999). We do know that experimen-
tally, direct sympathetic stimulation decreases con-
tractility and motility of the colorectum and
constriction of the internal anal sphincter, where-
as parasympathetic stimulation has the opposite
effects (see Brading and Ramalingam, this volume).
Storage and voiding reflexes
The maintenance of fecal continence is a product
of stool consistency, colo-rectal activity and the

synchronous relationship between the external and
internal anal sphincters (Engel et al., 1995). The
internal anal sphincter maintains a constant tone
and prevents stool leakage during everyday activ-
ities. The external anal sphincter acts like an
‘‘emergency brake’’ being used voluntarily and
maximally when there is a feeling of impending
defecation, thus allowing time to organize a suit-
able situation for toiletting. Interestingly, we can
observe an anal sphincter ‘‘guarding reflex’’ during
distension of the rectum (Chung et al., 2004) par-
alleling that seen during filling of the bladder.
The guarding reflex
Electromyographic evidence for a guarding reflex
can be seen during the normal urodynamic test,
that is, as the bladder reaches its full capacity (end
fill volume, EFV) the striated sphincter EMG sig-
nal reaches its maximum peak-to-peak amplitude.
To reiterate, the normal guarding reflex seems to
function to automatically enhance sphincter clo-
sure, and in the process, helps to inhibit unwanted
activity of the bladder, and perhaps also the bowel,
so helping to prevent incontinence probably
through contraction-activated pudendal afferents.
It may be that, part of a so-called guarding re-
flex could also involve smooth muscle responses in
the bladder neck, proximal urethra and internal
anal sphincter. However, whereas in striated mus-
cle, electromyographic signals are proven to be
real and related to muscle action potentials, in
smooth muscle, this relationship is much more
controversial because of the presence of large
movement artifact (Craggs, 1998). Therefore, at
this time we can only resort to measuring the slow
pressure changes associated with autonomic func-
tions of the urethra and anal canal.
As described previously, the peripheral afferent
limb of the guarding reflex is conveyed by sensory
afferent fibers in the pelvic nerves projecting from
the bladder or bowel to the sacral cord, whereas
the efferent limb is via the pudendal motor nerves
to the pelvic floor and sphincters from Onuf’s nu-
cleus in the sacral ventral cord. Interestingly, re-
cent neurophysiological testing of the pathways of
a similar reflex (bladder-anal) has revealed a long
209
latency of around 90 ms (Basinski et al., 2003)
suggesting that the guarding reflex probably in-
volves many interneurons in its arc and could even
engage a supra-spinal pathway (Park et al., 1997).
Neurophysiological measures of sacral reflexes
Routine clinical neurophysiological testing of pel-
vic floor or sphincter muscles can be done by di-
rectly examining various sacral reflexes involving
the bulbocavernosus, ischiocavernosus or sphinc-
ter muscles. Diagnostically, examination of these
reflexes by concentric needle electromyography or,
less specifically, with surface electrodes, is advo-
cated for all people presenting with sacral dys-
function (Fowler et al., 2002). For convenience,
the pudendo-anal reflex (PAR) has proved to be
one of the easiest and most useful sacral somatic
reflexes to test (Fig. 1). The normal pudendal-anal
reflex can be easily facilitated by voluntary con-
tractions of the pelvic floor (Fig. 1C).
Pudendal urethral and anal reflexes
Interestingly, the pudendal-anal reflex often mir-
rors the pudendo-urethral reflex and can be used
as its surrogate (Podnar and Vodusek, 2001). It is
known that the motoneurones for both the anal
and urethral sphincters are sited close together in
Onuf’s nucleus in humans, which is in contrast to
most other non-primate animals (Blok, 2002). Op-
timal recruitment of the pudendal-anal reflex is
achieved by using a condition-test paradigm to
stimulate the dorsal penile or clitoral nerves (ex-
clusively pudendal sensory nerves) with pairs of
electrical pulses through skin electrodes. The reflex
is recorded as a compound motor-evoked poten-
tial of the sphincter. By using a pressure sensor
with integral ring electrodes to record the pressure
and surface-evoked motor potentials simultane-
ously in the anal canal (Fig. 1A), it has been shown
that the best recruitment of Onuf’s motoneurones
is achieved when the double-pulse interval is be-
tween 2–4 ms (Fig. 1B) (Rodi and Vodusek, 1995).
210

A C
Pudendo -Anal

100 µV
Reflex
V

Anal
Sphincter

250 V
10 cmH2 O
Pressure

probe
Electrical Stimulation
Anal

(+ voluntary sphincter contraction)

Of the Pudendal
Afferent Nerves
DPN Stimulation
0.3s (Dorsal Penile Nerve) Paired pulses 100ms
ms
2ms pulse interval
Paired pulses 200s
200 s
@ 15mA (2 x PAR Threshold)

B
n = 10 The effect of paired pulse
0 ms separation on EAS amplitude
1 ms
Pudendal Anal Reflex

150
EMG Amplitude (V)
EMG Amplitude µV
V

2 ms
100µV
100 V

4 ms 100
8 ms
16 ms 50
32 ms
64 ms 0.00
0
Paired Pulse Separation (msec)
0 0.05 0.10 s Paired-pulse interval ms

Fig. 1. The Pudendo-Anal Reflex (PAR). A. Stimulation and recording setup for studies of this reflex. Pudendal afferent fibers in the
dorsal penile (or dorsal clitoral) nerves are stimulated to evoke the reflex. B. Paired-pulse stimulation at a 2–4 ms interval elicits an
optimal-compound motor-evoked potential response from the anal sphincter. C. The response is further facilitated by voluntary
contraction of the sphincter muscles.

Pelvo-pudendal reflex integration Aberrant somato-visceral reflexes following SCI

By combining pudendal-anal reflex measurement Damage to the spino-bulbo-spinal pathways (con-

with urodynamics (or proctodynamics), it is pos- necting the lumbar-sacral segments with the brain-
sible to demonstrate facilitatory effects of the pel- stem) in SCI, whether complete or incomplete, can
vic afferent activity from the bladder (or rectum) cause serious disruption to coordination of the
on the sphincters. In a recent study on bladder and pelvic organs and sphincters leading to un-inhib-
bowel filling, it was shown that modulation of the ited pelvic (parasympathetic) reflexes. In the blad-
reflex, reflecting the effects of a guarding reflex, der, this results in what is commonly termed
increased for both organs at end fill volume (EFV) detrusor hyperreflexia or neurogenic detrusor
(Chung et al., 2004). As expected, the pudendal- over-activity (Fig. 3). However, it is not clear
anal reflex was suppressed during bladder empty- whether over-activity is a regular feature seen in
ing in healthy subjects (Fig. 2). This latter finding the rectum or descending colon following SCI; in
confirmed an earlier study (Dyro and Yalla, 1986) fact, the common occurrence of constipation
and correlated well with the standard electromyo- might indicate otherwise although drugs used to
graphic findings seen with filling and voiding suppress the overactive bladder may be relevant in
cystometrograms. this context.
 211

Fig. 2. Modulation of the normal PAR during cystometry. For these studies, the PAR can be conveniently used as a surrogate for the
pudendo-urethral reflex (see explanation in the text). A. During filling cystometry, the sphincter EMG and PAR increase progressively
as EFV is reached, demonstrating the presence of a ‘‘guarding reflex’’. During voiding, the PAR and sphincter EMG are markedly
suppressed, consistent with the need for unobstructed bladder emptying. B. A graphical presentation of the responses of peak-to-peak
PAR in five healthy control subjects (mean 7 SD) at EFV and during voiding. For comparison across the group, the responses for
each subject were normalized to the size of the PAR when the bladder was empty.

Detrusor hyperreflexia is often exacerbated by Efferent effects

uncoordinated viscero-somatic reflexes giving im-
paired vesico-urethral function (detrusor–sphinc-
ter dyssynergia), leading to obstructed voiding,
high bladder pressures and incontinence (Has-
souna et al., 2004). If left untreated, it is these high
pressures that can lead to renal failure. The usual
management for the neurogenic bladder is a com-
bination of anti-cholinergic drugs to suppress the
over-activity and clean intermittent catheterization
to empty the bladder.
Afferent effects
For the lower urinary tract, the effect of a com-
plete spinal cord lesion is to block ascending af-
ferent activity from the bladder or urethra,
reaching the brain-stem peri-aqueductal gray dur-
ing filling and prevent proper pontine coordina-
tion of the bladder and sphincters. In fact, all
sensations from the lower urinary tract, bowel,
pelvic floor and sphincters would be absent as a
result of such a lesion and this would include all
other proprioceptive pathways normally linking
the pelvic floor more directly with the somato-
sensory cortex. The loss of pelvic sensation is,
perhaps, one of the most disabling features of SCI.
By cutting off the descending excitatory modulat-
ing influences from the lateral pontine nucleus to
Onuf’s nucleus in the sacral cord, we could expect
poor or absent sphincter guarding reflexes and the
loss of reflex inhibition of detrusor activity during
the filling phase (Park et al., 1997). Conversely, if
proper activation of the medial pontine nucleus
were dysfunctional, then the drive necessary for
bladder emptying would also be missing.
It appears that a combination of these effects
and the loss of voluntary descending control of the
sphincters and facilitation of the guarding reflex
could be to disinhibit visceral Ad afferent activity,
allowing unfettered sacral hyperreflexia to take
over reducing bladder capacity. In addition, the
loss of any descending inhibitory pathways from
the brain stem, normally keeping reflexes mediated
by bladder C-fiber afferents in check, could also
make the bladder more sensitive to distension,
again tending to drive the bladder toward lower
capacity (Yoshimura, 1999) [see De Groat and
Yoshimura, this volume]. The exact mechanism of
C-fiber hyper-excitability after SCI is not entirely
clear. It could be due to a change in the mechanical
sensitivity of the sensory endings, but more likely
it involves reorganization in the spinal neural
 212
A B C
Detrusor Hyperreflexia

<F>

DRG Detrusor Pressure

Urine Flow

Pelvic Intraurethral Pressure

nerves
Pudendal
nerves Sphincter EMG Sphincter Dyssynergia

Fig. 3. Reflexes of the lower urinary tract after SCI. A. Aberrant pelvic reflexes causing detrusor hyperreflexia and detrusor–external sphincter dyssynergia. B. Cystometry
in a cord-injured person with a complete injury, showing the high detrusor pressures generated during sphincter dyssynergia measured by electromyography. C. A
simultaneous X-ray frame showing the moment of detrusor–sphincter dysynergia.

circuitry (Shefchyk, 2002). Perhaps there is prolif-
eration of sacral spinal connections of these affer-
ent pathways, brought about by new occupation
of vacant excitatory post-synaptic receptor sites in
the sacral segmental pelvic reflex pathway. The
vacancy might follow degeneration of the excita-
tory pathway from the medial pontine nucleus,
which under normal bladder filling conditions
would be silent but comes into play to activate
the detrusor for voiding. Speculatively, these re-
ceptor sites in the sacral cord could be taken over
by collateral sprouting of the bladder afferents.
Inhibitory effects
Whatever the precise mechanisms generating de-
trusor hyperreflexia, it has been known for many
years from studies in experimental animal models
and clinically that pudendal afferent stimulation
can inhibit bladder reflexes (Lindstrom et al., 1983;
Fall et al., 1991). For example, in recent studies it
has been shown that bladder activity can be pro-
foundly suppressed by magnetic stimulation of the
pudendal afferents in the spinal roots (Sheriff
et al., 1996) or by electrical stimulation of the
dorsal penile nerves in healthy volunteers, in sub-
jects with idiopathic detrusor over-activity as well
as in subjects with complete spinal cord lesions
(Shah et al., 1998) (Fig. 4B). This suggests that
reflexes involving somato-visceral inhibitory inter-
actions are present at the local sacral segmental
level whether or not the spinal cord is intact
(Fig. 4A) (Craggs and MacFarlane, 1999). More
importantly for people with spinal cord injuries,
such stimulation (popularly referred to as neuro-
modulation) not only inhibits detrusor hype-
rreflexia but also increases bladder capacity
significantly (Kirkham et al., 2001), which could
provide a useful alternative treatment to drugs.
The guarding reflex after SCI
Synthesis of the guarding reflex appears to require
involvement of the pontine micturition centre as
well as the integrity of supra-sacral pathways. The
guarding reflex is said to be absent in over 85% of
cord-injured people with a complete (American
213
Spinal Injuries Association, ASIA A) spinal cord
lesion, but it is often present but variable in people
with incomplete lesions (Siroky and Krane, 1982).
Therefore, it is believed that the partial preserva-
tion or loss of the guarding reflex, in response to
bladder filling, would be an indication of the com-
pleteness of spinal injury.
The bladder guarding reflex
Recent studies of neurourological function (Balasu-
bramaniam et al., 2004) have confirmed that at
bladder EFV (defined as the volume at which void-
ing or neurogenic detrusor overactivity, NDO, oc-
curs), the guarding reflex, measured by the
modulation of the PAR (see Fig. 2), is absent or
very weak in most people with a neurologically de-
fined complete supra-sacral spinal cord lesion
(Fig. 5). In people with incomplete lesions (ASIA
B-D), the guarding reflex is often preserved but very
variable. It has also been demonstrated that a weak
guarding reflex is often associated with low bladder
capacity, perhaps as a result of weak pudendal in-
hibition on aberrant detrusor contractions.
In contrast to healthy volunteers who have a
very suppressed level of PAR during voiding, most
people with a complete or incomplete supra-sacral
lesion have an exaggerated pudendal reflex as ex-
pected during vesico-sphincter dyssynergia (Dyro
and Yalla, 1986; Sethi et al., 1989).
The bowel guarding reflex
A recent addition to this study has interestingly
discovered that whilst an analogous bowel ‘‘guard-
ing reflex’’ during rectal filling exists, it appears to
be present irrespective of SCI (Chung et al., 2004).
Perhaps this is a reflection of the greater depend-
ence on automatic control in bowel function; the
guarding reflex requiring only sacral segmental
control. The external anal sphincter is the auto-
matic braking system in the proper coordination
of the bowel and sphincters for continence, but in
proper bowel function, continence is a differential
process depending on whether the rectal content is
gas, solid or liquid. In man, it is social circum-
stances and voluntary control of the sphincters
 A B
140

Pressure
cm H2O
Bladder
0

Stimulation

C
Neuromodulation
200

Bladder Capacity
Mean % Change
Electrical
Stimulation
of Pudendal
Afferent
100
Pathways
at the Dorsal
Penile Nerves
0
Serial Cystometrograms

Fig. 4. Controlling detrusor hyperreflexia by non-invasive neuromodulation through pudendal afferent pathways. (A) By stimulating
the dorsal penile (or clitoral) nerves with electrical pulses between 10–20 per second and above twice the threshold for the pudendo-
anal reflex it is possible to profoundly suppress detrusor hyperreflexia. (B) The upper trace shows the effect of continuous stimulation
of the dorsal penile nerves on the bladder pressure rise associated with a detrusor hyperreflexia contraction provoked at the middle
arrow. Control hyperreflexic contractions provoked at the other arrows can be seen before and after stimulation. Again this response is
flanked by control provocations. (C) Repeated cystometrograms with continuous neuromodulation (shaded area) demonstrating
significant increases in bladder volume when compared to control fills. Following stimulation the bladder takes some time to restore to
its smaller capacity probably as a result of stretching of the bladder wall during the period of neuromodulation.

A B
End Fill Volume ASIA E – Non SCI ASIA A – Complete SCI
Voiding or Overactivity ASIA B-D – Incomplete SCI
1.0
(standardised to an empty
normalised PAR +/-SD

0.6
*
Mean peak-peak PAREFV

* 0.5
0.5
(mV mean +/-SD)

0.4

** 0.3
0.0 0.2
bladder=0)

0.1
0
-0.5
-0.1 200 400 600 800
-0.2
-1.0 -0.3
Non-SCI Incomplete Complete Mean Bladder EFV (mL mean+/-SD)
Subjects SCI SCI
Subjects Subjects

Fig. 5. Changes in the bladder guarding reflex after SCI. For an explanation of the relationship between the measured pudendo-anal
reflex during bladder filling and the guarding reflex refer to Fig. 2 Panel A shows the changes in the peak-to-peak amplitude of the
normalized PAR at EFV and voiding (or NDO in the case of the two SCI groups). Subjects with a complete spinal lesion have little or
no guarding reflex (white bars) compared to subjects without SCI, whereas subjects with incomplete injuries have a very variable reflex.
This variability probably reflects the wide range of neurological impairment in incomplete subjects. In subjects with a complete injury,
the PAR change (striped bars) reflects the presences of sphincter dyssynergia. B. Subjects with a complete lesion present with the
smallest guarding reflex and bladder capacity suggesting that there is little reflex inhibition of the detrusor (that is, neurogenic detrusor
over-activity occurs at low bladder volumes) (Balasubramaniam et al., 2004).
 215

that determine the final emergency brake for con- their pelvic floor and sphincter reflexes including
tinence. the guarding reflex.
For definitions of neurological grading see ap-
pendix [American Spinal Injuries Association/In-
Volitional effects on sphincter reflexes after SCI ternational Medical Society of Paraplegia (ASIA/
IMSOP), 1996] adapted from an original classifi-
Although most of the neural circuits involved in cation scheme by Frankel et al. (1969).
the normal control of the bladder and bowel are
autonomic, continence is very much a function of
Neurophysiological measures of volition on
volitional control. Voluntary contraction of the
pelvic sphincters
pelvic floor muscles plays an important role in
normal continence mechanisms particularly during
Measuring the effects of volition on the pelvic
postponement of voiding. When the bladder or
sphincters in some cord-injured people, for exam-
bowel reach their near-maximal capacity at EFV,
ple, those with a putative complete injury (as de-
not only does contraction of the sphincters prevent
termined by the ASIA classification), is difficult.
inadvertent leaking, but such contractions proba-
To overcome these difficulties, a recent study
bly inhibit the parasympathetic reflex pathways
(Balasubramaniam et al., 2005) has tested subjects
within the spinal cord to suppress premature void-
by presenting them with a visual signal instructing
ing contractions. This mechanism could operate
them to make a voluntary squeeze of their anal
directly via descending cortico-spinal pathways
sphincter (or attempt to do so) and hold it for the
leading to inhibition of visceral reflexes. Indirectly,
duration of the signal (Fig. 6). Clearly, for those
there could be facilitation of the motor component
with a complete lesion, there was no internal sen-
of the guarding reflex that in turn activates pu-
sory feedback, but by making their anal sphincter
dendal afferents to suppress the visceral reflexes.
EMG audible, they could hear whether they were
Such a mechanism would be similar to that envis-
successful at the task. The EMG was integrated to
aged for electrical neuromodulation (Craggs and
determine the mean peak amplitude. During the
McFarlane, 1999). Interestingly, the sensations of
attempted contraction, the computer controlling
a full bladder and the strong desire to void can
the experiment elicited a pudendal-anal reflex. The
also be suppressed by voluntary contractions of
test was repeated ten times and averaged puden-
the pelvic floor, or again, by therapeutic neuro-
dal-anal reflexes and integrated EMGs were de-
modulation (Oliver et al., 2003).
termined for each subject with an empty bladder
It remains to be determined whether people with
and rectum.
an incomplete SCI and some volitional control
The results compared non-SCI subjects with co-
over their pelvic floor can also voluntarily inhibit
horts of complete and incomplete chronic spinal
their detrusor hyperreflexia and suppress the desire
cord injuries. The study concluded that both the
to void.
pudendal-anal reflex and integrated EMG corre-
lated reasonably well with ASIA grading taken at
the same time, however, it was the pudendal-anal
Neurological grading and pelvic floor reflexes
reflex measurement that correlated more system-
atically with the neurological assessment of injury.
In complete supra-sacral SCI (ASIA A), we would
expect all volitional effects and sensations related
to pelvic function to be lost. In addition, we might Sensitivity of sacral reflex measurement and
also expect a very poor or absent guarding reflex neurological assessment in SCI
during bladder filling (see Fig. 5). On the other
hand, in people with incomplete lesions (ASIA B- Routine assessment of patients with SCI currently
D), the picture would be much more variable with relies on subjective clinical measurement using the
some preservation of voluntary modulation of ASIA classification and Impairment Score. It is
216

A Bladder Empty

PAR
EMG

PAR 50µV
+ Squeeze
150ms
EMG 500µV
+ Squeeze
3s
Integrated
EMG

Squeeze

B Integrated EMG C PAR Response

400
% Change in Integrated

% Change in peak to

200
PAR peak to peak

300
amplitude
amplitude
EMG peak

200
100

100

0
0
Non-SCI Incomplet Complete Non-SCI Incomplet Complete
Subjects e SCI SCI Subjects e SCI SCI
Subjects Subjects Subjects Subjects

Fig. 6. Neurophysiological measures of volition on the pelvic sphincters. A. During voluntary contractions of the pelvic floor muscles
and sphincters (squeeze) the PAR is facilitated. B and C show the change in anal sphincter EMG and PAR amplitude, respectively,
during voluntary anal squeeze (Balasubramaniam et al., 2005).

not always consistently rated and neither does it
assess autonomic function. However, recent so-
mato-sensory (Krassioukov et al., 1999) and so-
mato-motor (Smith et al., 2000) testing are
introducing more objective neurophysiological
measures into the evaluation of people with in-
complete SCI. As described here, there is little
doubt that objective measures of pelvic function
through neurophysiological testing of somato-vis-
ceral reflexes could also help to evaluate auto-
nomic function; there appears to be a good
correlation with the standard neurological assess-
ment of injury. It remains to be seen how sensitive
the tests will be for assessing the outcome of ther-
apeutic interventions for functional restoration.
Conclusions
Regardless of the approach for functional resto-
ration, whether by promoting natural recovery
with neuro-protection, tapping into cortical plas-
ticity (Belci et al., 2004), implanting novel devices
to stimulate residual neural pathways (Grill et al.,
2001; Craggs, 2004), or ultimately developing a
‘‘cure’’ through functional neural repair (Ramer
 217

et al., 2005), there is going to be a need to assess level and includes the
carefully the benefits through more sensitive meas- sacral segments S4–S5.
ures of autonomic neural function. C Incomplete Motor function is
By using a combination of urodynamic and preserved below the
proctodynamic techniques, ideally with simultane- neurological level, and
ous video-imaging, together with the sort of ob- more than half of key
jective neurophysiological measures described in muscles below the
this chapter, we can look forward to developing neurological level have
reliable protocols for sensitive assessment of au- a muscle grade less
tonomic function of the pelvic organs. Such proto- than 3.
cols (together with those for assessing other D Incomplete Motor function is
autonomic physiological disturbances in SCI) will preserved below the
help us to characterize all autonomic dysfunction neurological level, and
more objectively. These can then be combined with at least half of key
the other well-established neurological and neuro- muscles below the
physiological assessments to give a much more neurological level have
comprehensive picture of SCI and the effect of in- a muscle grade of 3 or
terventions. The ultimate goal for all those con- more.
cerned with autonomic function of the bladder and E Normal Motor and sensory
bowel will be to develop a robust impairment function are normal.
grading and scoring system for pelvic dysfunction.

Acknowledgments
References
I am grateful to Vernie Balasubramaniam for al-
ASIA/IMSOP. (1996) International Standards for Neurological
lowing me to use some of the data obtained in her and Functional Classification of Spinal Cord Injury-Revised.
research studies funded by the International Spinal American Spinal Injuries Association, Chicago, IL.
Research Trust, UK. Also, I thank Joe Wood- Balasubramaniam, A.V., Chung, E.A.L., Bycroft, J.A., Knight,
house and Alex Chung, two of my Clinical Re- S.L., Gall, A., Middleton, F.R.I., Fowler, C.J. and Craggs,
M.D. (2005). Modulation of Pelvic Floor Reflexes by Voli-
search Fellows, for reading the various draft
tion in Spinal Cord Injury. Proceedings of the Urological
manuscripts and making some very helpful com- Research Society, Royal College of Surgeons, London.
ments and suggestions whilst writing this chapter. Balasubramaniam, V., Bycroft, J., Wood, S., Middleton,
F.R.I., Fowler, C.J. and Craggs, M.D. (2004) Urinary guard-
ing reflex: Aberrant in spinal cord injury? Brit. J. Urol. Int.,
93(Suppl 4): 4.
Appendix Basinski, C., Fuller, E., Brizendine, E.G. and Benson, J.T.
(2003) Bladder-anal reflex. Neurourol. Urodyn., 22: 683–686.
Belci, M., Catley, M., Husain, M., Frankel, H.L. and Davey,
American Spinal Injuries Association Classifica-
N.J. (2004) Magnetic brain stimulation can improve clinical
tion Grades and Impairment Scale (1996). outcome in incomplete spinal cord injured patients. Spinal
Cord, 42: 417–419.
Grade Completeness Description
Blok, B.F. (2002) Central pathways controlling micturition and
A Complete No sensory or motor urinary continence. Urology, 59(Suppl 1): 13–17.
function preserved in Blok, B.F.F., Willemsen, A.T.M. and Hostege, G. (1997) A
the sacral segments PET study on brain control of micturition in humans. Brain,
S4–S5. 120: 111–121.
Burleigh, D.E. (1992) Pharmacology of the internal anal
B Incomplete Sensory but not motor
sphincter. In: Henry M.M. and Swash M. (Eds.), Colopr-
function is preserved octology and the Pelvic Floor. Butterworth-Heinemann,
below the neurological Oxford, pp. 37–53.
218
Burnstock, G. (1990) Innervation of the bladder and bowel. In:
Block G. and Whelan J. (Eds.) Neurobiology of Inconti-
nence, Ciba Foundation Symposium, Vol. 151. Wiley, Chich-
ester, pp. 2–26.
Chung, E.A.L., Balasubramaniam, A.V., Middleton, F., Craggs,
M.D. and Emmanuel, A.V. (2004) Relationship between
bladder and bowel filling and the guarding reflex mechanism.
Brit. J. Urol. Int., 94: 1176.
Craggs, M.D. (1998) The elusive electromyogram: Fact vs ar-
tifact. Editorial comment. Neurourol. Urodyn., 17: 80–82.
Craggs, M.D. (2004) Future prospects for complete restoration
of bladder function by neuroprostheses. In: Corcos J. and
Schick E. (Eds.), The Textbook of Neurogenic Bladder,
Adults and Children. Martin Dunitz, London, pp. 625–635
Ch. 56.
Craggs, M.D. and MacFarlane, J. (1999) Neuromodulation of
the lower urinary tract. Exp. Physiol., 84: 149–160.
Craggs, M.D. and Vaizey, C.J. (1999) Neurophysiology of the
bladder and bowel. In: Fowler C.J. (Ed.), Neurology of
Bowel, Bladder and Sex, (Blue Books of Practical Neurolo-
gy). Butterworth-Heinemann, Oxford, pp. 19–32.
De Groat, W.C. (1997) Central nervous system control of mi-
cturition. In: O’Donnell P. (Ed.), Urinary Incontinence. Mo-
sby, St. Louis, pp. 33–47.
Dyro, F.M. and Yalla, S.V. (1986) Refractoriness of urethral
striated sphincter during voiding: Studies with afferent pu-
dendal reflex arc stimulation in male subjects. J. Urol., 135:
732–736.
Engel, A.F., Kamm, M.A., Bartram, C.I. and Nicholls, R.J.
(1995) Relationship of symptoms in faecal incontinence to
specific sphincter abnormalities. Int. J. Colorectal Dis., 10:
152–155.
Fall, M. and Lindstrom, S. (1991) Electrical stimulation. A
physiologic approach to the treatment of urinary inconti-
nence. Urol. Clin. North Am., 18: 393–407.
Fowler, C.J., Benson, J.T., Craggs, M.D., Vodusek, D.B.,
Yang, C.C. and Podnar, S. (2002) Clinical neurophysiology.
In: Abrams P., Cardozo L., Khoury S. and Wein A. (Eds.),
Incontinence (International Consultation on Incontinence
ICS/WHO). Health Publication Limited, Plymouth,
pp. 391–424.
Frankel, H.D.H., Hancock, D.O., Hyslop, G., Melzak, J.,
Michaelis, L.S., Ungar, G.H., Vernon, J.D.S. and Walsh,
J.J. (1969) The value of postural reduction in the initial
management of closed injuries of the spine with paraplegia
and tetraplegia. Paraplegia, 7: 179–192.
Grill, W.M., Craggs, M.D., Foreman, R.D., Ludlow, C.L. and
Buller, J.R. (2001) Emerging clinical applications of electrical
stimulation: Opportunities for restoration of function. J. Re-
habil. Res. Dev., 38(6): 641–653.
Hassouna, M., Elmayergi, N. and Abdelhady, M. (2004)
Pathophysiology of spinal cord injury. In: Corcos J. and
Schick E. (Eds.), Textbook of the Neurogenic Bladder. Mar-
tin-Dunitz, London, pp. 178–191 Ch. 14.
Jänig, W. and Morrison, J.F.B. (1986) Functional properties of
spinal visceral afferents supplying abdominal and pelvic or-
gans with special emphasis on visceral nociception. In:
Cervero F. and Morrison J.F.B. (Eds.) Visceral Sensation,
Progress in Brain Research, Vol. 69. Elsevier, Amsterdam,
pp. 87–114.
Kirkham, A.P.S., Shah, N.C., Knight, S.L., Shah, P.J.R. and
Craggs, M.D. (2001) The acute effects of continuous and
conditional neuromodulation on the bladder in spinal cord
injury. Spinal Cord, 39: 420–428.
Krassioukov, A., Wolfe, D.L., Hsieh, J.T., Hayes, K.C. and
Durham, C.E. (1999) Quantitative sensory testing in patients
with incomplete spinal cord injury. Arch. Phys. Med. Re-
habil., 80: 1258–1263.
Lindstrom, S., Fall, M., Carlsson, C.A. and Erlandson, B.E.
(1983) The neurophysiological basis of bladder inhibition in
response to intravaginal electrical ‘‘stimulation’’. J. Urol.,
129: 405–410.
Mathers, S. (1992) Neural control of the pelvic sphincters. In:
Henry M.M. and Swash M. (Eds.), Coloproctology and the
Pelvic Floor. Butterworth-Heinemann, Oxford, pp. 61–71.
Mundy, A.R. and Thomas, P.J. (1994) Clinical physiology
of the bladder, urethra, and pelvic floor. In: Mundy A.R.,
Stephenson T.P. and Wein A.J. (Eds.), Urodynamics –
Principles, Practice, and Application. Churchill-Livingstone,
Edinburgh, pp. 15–27.
Oliver, S.E., Fowler, C.J., Mundy, A.R. and Craggs, M.D.
(2003) Measuring the sensations of urge and bladder filling
during cystometry in urge incontinence and the effects of
neuromodulation. Neurourol. Urodynam., 33: 7–16.
Park, L.M., Bloom, D.A. and McGuire, E.J. (1997) The guard-
ing reflex revisited. Brit. J. Urol., 80: 940–945.
Podnar, S. and Vodusek, D.B. (2001) Protocol for clinical ne-
urophysiologic examination of the pelvic floor. Neurourol.
Urodynam., 20: 669–682.
Ramer, L.M., Ramer, M.S. and Steeves, J.D. (2005) Setting the
stage for functional repair of spinal cord injuries: A cast of
thousands. Spinal Cord, March 2005, 43(3): 134–161.
Rodi, Z. and Vodusek, D.B. (1995) The sacral reflex studies:
Single versus double pulse electrical stimulation. Neurourol.
Urodynam., 14: 496.
Schrøder, H.D. and Reske-Nielsen, E. (1983) Fiber S types in
the striated urethral and anal sphincters. Acta Neuropathol.,
60: 278–282.
Sethi, R.K., Bauer, S.B., Dyro, F.M. and Krarup, C. (1989)
Modulation of the bulbocavernosus reflex during voiding:
Loss of inhibition in upper motor neuron lesions. Muscle
Nerve, 12: 892–897.
Shah, N., Edhem, I., Knight, S.L., Shah, P.J.R. and Craggs,
M.D. (1998) Acute suppression of provoked detrusor hype-
rreflexia by electrical stimulation of the dorsal penile nerve.
Eur. Urol., 33(Suppl 1): 60.
Shefchyk, S.J. (2002) Spinal cord neural organization control-
ling the urinary bladder and striated sphincter. Prog. Brain
Res., 137: 71–82.
Sheriff, M.K.M., Shah, P.J.R., Fowler, C.J., Mundy, A.R. and
Craggs, M.D. (1996) Neuromodulation of detrusor hyper-
reflexia by functional magnetic stimulation of the sacral
roots. Brit. J. Urol., 78: 39–46.

Siroky, M.B. and Krane, R.J. (1982) Neurologic aspects of de-
trusor-sphincter dyssynergia, with reference to the guarding
reflex. J. Urol., 127: 953–957.
Smith, H.C., Savic, G., Frankel, H.L., Ellaway, P.H., Maskill,
D.W., Jamous, M.A. and Davey, N.J. (2000) Corticospinal
function studied over time following incomplete spinal cord
injury. Spinal Cord, 38: 292–300.
219
Yoshimura, N. (1999) Bladder afferent pathway and spinal
cord injury: Possible mechanisms including hyperreflexia of
the urinary bladder. Prog. Neurobiol., 57: 583–606.
Yoshimura, N., Seki, S. and Chancellor, M.B. (2004) Integrated
physiology of the lower urinary tract. In: Corcos J. and
Schick E. (Eds.), Textbook of the Neurogenic Bladder. Mar-
tin-Dunitz, London, pp. 73–87 Ch. 6.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 14

The clinical problems in cardiovascular control

following spinal cord injury: an overview

Andrei Krassioukov1,2,3,4, and Victoria E. Claydon1

1
International Collaboration on Repair Discoveries (ICORD), University of British Columbia, Vancouver, BC V6T 1Z4,
Canada
2
Division of Physical Medicine and Rehabilitation, University of British Columbia, Vancouver, BC V6T 1Z4, Canada
3
School of Rehabilitation, University of British Columbia, Vancouver, BC V6T 1Z4, Canada
4
Department of Medicine, University of British Columbia, Vancouver, BC V6T 1Z4, Canada

Abstract: On a daily basis, individuals with cervical and upper thoracic spinal cord injury face the challenge
of managing their unstable blood pressure, which frequently results in persistent hypotension and/or
episodes of uncontrolled hypertension. This chapter will focus on the clinical issues related to abnormal
cardiovascular control in individuals with spinal cord injury, which include neurogenic shock, autonomic
dysreflexia and orthostatic hypotension. Blood pressure control depends upon tonic activation of sym-
pathetic preganglionic neurons by descending input from the supraspinal structures (Calaresu and Yardley,
1988). Following spinal cord injury, these pathways are disrupted, and thus spinal circuits are solely
responsible for the generation of sympathetic activity (Osborn et al., 1989; Maiorov et al., 1997). This
results in a variety of cardiovascular abnormalities that have been well documented in human studies, as
well as in animal models (Osborn et al., 1990; Mathias and Frankel, 1992a, b; Krassioukov and Weaver,
1995; Maiorov et al., 1997, 1998; Teasell et al., 2000). However, the recognition and management of these
cardiovascular dysfunctions following spinal cord injury represent challenging clinical issues. Moreover,
cardiovascular disorders in the acute and chronic stages of spinal cord injury are among the most common
causes of death in individuals with spinal cord injury (DeVivo et al., 1999).

Pathophysiology of cardiovascular dysfunction after
spinal cord injury
The first description of one of the most common
autonomic disturbances in individuals with spinal
cord injury, known as autonomic dysreflexia, ap-
peared in 1860 (Hilton, 1860). However, it was not
until 1947 that Guttman and Whitteridge de-
scribed the possible mechanisms responsible for
the development of this condition (Guttman and
Whitteridge, 1947). Both clinical observations and
experimental animal data have contributed signif-
Corresponding author. Tel.: +001 604 822 2673;
Fax: +001 604 822 2924; E-mail: krassioukov@icord.org
icantly to our present understanding of the patho-
physiology of autonomic dysreflexia and abnor-
mal cardiovascular control after spinal cord
injury (Osborn, Taylor and Schramm, 1989, 1990;
Mathias and Frankel, 1992a; Krassioukov and
Weaver, 1995; Maiorov, Weaver and Krassioukov,
1997; Maiorov, Fehlings and Krassioukov, 1998;
Krassioukov et al., 2002). Recently, numerous
elements within autonomic circuits were identified
that could contribute to abnormal cardiovascular
control after spinal cord injury: (a) descending
vasomotor (sympathoexcitatory) pathways, (b)
sympathetic preganglionic neurons, (c) spinal in-
terneurons, (d) spinal afferents, and (e) peripheral
neurovascular mechanisms. Spinal cord injury

DOI: 10.1016/S0079-6123(05)52014-4 223

224
disrupts the descending sympathoexcitatory path-
ways provided by medullary neurons located with-
in the rostroventrolateral medulla. These neurons
provide tonic input to the sympathetic pregangl-
ionic neurons located within the lateral horns of
the spinal grey matter of the thoracic and upper
lumbar spinal segments (T1–L2). Disruption of
these descending cardiovascular pathways results
in at least four phenomena: initial sympathetic
hypoactivity; alterations in the morphology of
sympathetic preganglionic neurons; plastic chang-
es within the spinal circuits (including sprouting
and the potential formation of inappropriate
synaptic connections); and the development of
changes in sympathetic neurovascular transmission
and smooth muscle responsiveness (Krassioukov
et al., 1999; Teasell et al., 2000; Yeoh et al., 2004).
The initial sympathetic hypoactivity results in low
resting blood pressure, loss of blood pressure
homeostasis and disturbed reflex control (Mathias
and Frankel, 1992b). However, with time follow-
ing spinal cord injury, the loss of descending in-
hibitory pathways and plastic changes within the
spinal cord, coupled with peripheral neurovascular
changes, are likely to predispose to episodes of
extreme hypertension, associated with autonomic
dysreflexia, that frequently develop in both acute
and chronic stages of spinal injury.
The acute post-injury period and neurogenic shock
Acute spinal cord injury in humans, especially at
the cervical level, results in severe hypotension and
persistent bradycardia that are common compo-
nents of the phenomenon known as neurogenic
shock (Atkinson and Atkinson, 1996). This event
is more profound and long lasting in humans after
spinal cord injury than in experimental animals.
Moreover, clinical observations strongly suggest
that the extent to which prolonged and severe hy-
potension requiring vasopressive therapy occurs is
associated with the severity of the spinal cord in-
jury and cervical or thoracic location of the injury,
and can last up to 5 weeks after injury (Mathias
and Frankel, 1992b; Atkinson and Atkinson, 1996;
Vale et al., 1997; Nacimiento and Noth, 1999;
Hadley et al., 2002a). In one study, Glenn et al.
reported that severe hypotension was present in all
31 tetraplegic subjects assessed with severe spinal
cord injury, half of whom required pressor therapy
in order to maintain arterial blood pressure (Glenn
and Bergman, 1997). In addition to the pro-
nounced hypotension described, many patients
with acute spinal cord injury experience severe ab-
normalities in heart rate. Bradycardia was report-
ed in 64–77% of patients with cervical spinal cord
injury during the acute post-injury stage, and was
more severe and frequent within the first 5 weeks
after injury (Piepmeier et al., 1985; Winslow et al.,
1986; Lehmann et al., 1987). In contrast, when the
injury is in the mid-thoracic spinal cord, leaving
cardiac sympathetic neurons under brainstem con-
trol, and vagal and sympathetic influences more in
balance, bradycardia is a less severe problem.
Furlan et al. (2003) and colleagues reported that
the hypotension and bradycardia observed initially
after injury persisted in the individuals with more
severe injury of the descending cardiovascular
autonomic pathways. Moreover, all individuals
in this group required vasopressor therapy in order
to maintain systolic arterial blood pressure above
90 mmHg. In contrast, individuals with less severe
injury to the descending cardiovascular pathways
tended to show higher levels of blood pressure and
heart rate, although minor and short-term hypo-
tension and low heart rates were occasionally
observed.
In addition to neurogenic shock, the acute phase
of spinal cord injury is also associated with ‘‘spinal
shock’’ (Nacimiento and Noth, 1999; Ditunno
et al., 2004). Some authors use these terms inter-
changeably, however, it is important to recognize
that these are two clinically important and distinct
conditions. Neurogenic shock is characterized
by changes occurring in blood pressure control
following spinal cord injury, whereas spinal shock
is characterized by a marked reduction or aboli-
tion of sensory, motor or reflex function of the
spinal cord below the level of injury (Ditunno
et al., 2004). Clinically, spinal shock in humans
can persist for days to weeks, with a mean dura-
tion of between 4 and 6 weeks after the injury.
Traditional views of the clinical course of the re-
covery of spinal shock were related to the emer-
gence of certain groups of reflexes. For example,

some considered that spinal shock had ended when
the appearance of initial reflexes, such as the
bulbocavernosus reflex occurred in the first few
days after spinal cord injury, others with the
recovery of deep tendon reflexes at 2 weeks post
injury, while some groups classified the end of
spinal shock as when the bladder voiding reflexes
recover after approximately 2 months. For further
details, see Ditunno et al. (2004).
Management of the acute period of spinal
cord injury
Unfortunately, there are no prospective controlled
studies on the effects of hypotension upon the
outcome following acute spinal cord injury in hu-
mans. However, the occurrence of hypotension in
the acute period following traumatic cervical or
upper thoracic spinal cord injury or severe head
injury has been shown to be associated with worse
outcomes (King et al., 2000). Recent guidelines on
arterial blood pressure management after acute
spinal cord injury (Hadley, 2002b) recommend
maintenance of mean arterial blood pressure at a
level of at least 85–90 mmHg for the first week
after acute spinal cord injury to ensure adequate
spinal cord perfusion. Prompt treatment of hypo-
tension with restoration of blood volume (using
both colloid and albumin) and vasopressive ther-
apy with dopamine, and the a-adrenergic agonist,
phenylephrine were recommended to maintain the
mean blood pressure.
Autonomic dysreflexia
Individuals with a cervical or high thoracic spinal
cord injury face life-long abnormalities of blood
pressure control (Mathias and Frankel, 1992a;
Teasell et al., 2000). In general, the resting arterial
blood pressure in these individuals is lower than
in able-bodied subjects, often with disabling
episodes of orthostatic hypotension. However,
life-threatening episodes of autonomic dysreflexia,
characterized by extreme hypertension accompa-
nied by a pounding headache, slow heart rate and
upper body flushing may also occur, where systolic
blood pressure can reach up to 300 mmHg
225
(Mathias and Frankel, 1992a; DeVivo et al.,
1999). Untreated episodes of autonomic dysreflex-
ia may have serious consequences, including in-
tracranial hemorrhage, retinal detachments,
seizures and death (Yarkony et al., 1986; Pine
et al., 1991; Eltorai et al., 1992). Different noxious
and non-noxious stimuli such as bowel and blad-
der distension, spasticity and pressure sores may
provoke the sudden increases in arterial blood
pressure of autonomic dysreflexia (Teasell et al.,
2000). Dysreflexia is three times more prevalent in
quadriplegics with complete injury, in comparison
to those with incomplete injury (Curt et al., 1997).
Finally, autonomic dysreflexia, even in quadriple-
gics, is not always severe, and may be character-
ized only by sweating and piloerection, or may
even be asymptomatic (Kirshblum et al., 2002).
These cardiovascular abnormalities are attrib-
uted to autonomic instability, caused by changes
occurring within the spinal autonomic circuits in
both the acute and chronic stages following spinal
cord injury (Mathias and Frankel, 1992a; Teasell
et al., 2000). The destruction of the descending
vasomotor pathways results in the loss of excita-
tory supraspinal input to the sympathetic pre-
ganglionic neurons, and is currently considered as
the major factor underlying the persistent arterial
hypotension and lack of sympathetic tone seen af-
ter high spinal cord injury (Furlan et al., 2003).
It is important to note, however, that although
autonomic dysreflexia occurs more often in the
chronic stage of spinal cord injury at or above the
6th thoracic segment, there is clinical evidence of
early episodes of autonomic dysreflexia in the first
days and weeks after the injury (Silver, 2000;
Krassioukov et al., 2003). In fact, it seems likely
that autonomic dysreflexia is under-recognized in
the acute phase of spinal cord injury. We recently
conducted a study in order to determine the inci-
dence and clinical associations of early autonomic
dysreflexia in individuals with acute traumatic spi-
nal cord injury (Krassioukov et al., 2003). Among
58 patients with acute traumatic spinal cord injury,
three individuals showed early autonomic dysre-
flexia and all the three had complete cervical
tetraplegia. The trigger mechanisms for autonomic
dysreflexia were somatic pain, fecal impaction
and abdominal distension. The earliest episode of
226
autonomic dysreflexia occurred on the 4th day post
injury. Patients with severe cervical spinal cord
injury are particularly susceptible to early onset of
autonomic dysreflexia (Krassioukov et al., 2003).
Management of autonomic dysreflexia
In patients with spinal cord injury, appropriate
bladder and bowel routines, in addition to the
prevention of pressure sore development, are the
most effective measures for prevention of auto-
nomic dysreflexia. However, for each individual,
the identification and elimination of specific trig-
gers for autonomic dysreflexia should also be em-
ployed to manage and prevent episodes of
autonomic dysreflexia (Mathias and Frankel,
1992a; Teasell et al., 2000). The initial manage-
ment of an episode of autonomic dysreflexia
should involve placing the patient in an upright
position in order to provoke an orthostatic reduc-
tion in blood pressure, and the loosening of any
tight clothing. Throughout the episode, the blood
pressure should be checked at 5 min intervals. It is
then necessary to search for and eliminate the pre-
cipitating stimulus, which is most commonly (in
85% of cases) related to either bladder distention
or bowel impaction (Mathias and Frankel, 1992a;
Teasell et al., 2000). The use of antihypertensive
drugs should be considered as a last resort, but
may be necessary if the blood pressure remains
elevated after following the aforementioned steps.
Orthostatic hypotension
Low arterial blood pressure is a problem in both
acute and chronic high-level spinal cord injured
patients. Indeed, Mathias and colleagues noted
that there was an inverse linear relationship
between the level of spinal cord injury and resting
blood pressure (Mathias and Frankel, 1992a;
Mathias, 1995). This lower resting blood pressure
is thought to be secondary to a reduction in sym-
pathetic nervous activity below the level of the
spinal cord injury.
In addition to the low resting blood pressure,
many individuals with high spinal cord injury also
experience a further drop in blood pressure in the
upright posture (orthostatic hypotension), partic-
ularly in the acute phase of injury (Mathias, 1995;
Cariga et al., 2002). The symptoms of orthostatic
hypotension in spinal cord injured individuals are
similar to those seen in other populations experi-
encing orthostatic hypotension, and are associated
with cerebral hypoperfusion (Cleophas et al.,
1986). The symptoms commonly include fatigue
or weakness, light-headedness, dizziness, blurred
vision, dyspnea and restlessness (Frisbie and
Steele, 1997; Sclater and Alagiakrishnan, 2004).
In one study (Illman et al., 2000), 41.1% of spinal
cord injured individuals who developed orthostatic
hypotension were asymptomatic despite significant
blood pressure falls. Concerning the incidence and
prevalence of orthostatic hypotension in this pop-
ulation, orthostatic maneuvers performed during
physiotherapy and mobilization are reported to
induce blood pressure changes, diagnostic of
orthostatic hypotension, in 74% of spinal cord
injured individuals, suggesting that orthostatic
hypotension is a common phenomenon among
the spinal cord injured population (Illman et al.,
2000).
Orthostatic hypotension following spinal cord
injury appears to be related to excessive pooling of
blood in the viscera and dependent extremities,
presumably due in part to the absence or low level
of efferent sympathetic nervous activity and to the
loss of the reflex vasoconstrictor effect of arterial
baroreceptor unloading by the sympathetic pre-
ganglionic neurons below the lesion (Mathias,
1995). This is likely to be compounded by the loss
of lower extremity muscle function that is known
to be important in counteracting venous pooling in
the upright position. The resultant excessive ve-
nous pooling in the lower extremities, and reduced
blood volume in the intrathoracic veins leads to a
reduced pressure in the large veins draining into
the atria of the heart (Jacobsen et al., 1992; Faghri
et al., 2001). This in turn results in a decrease in
ventricular end-diastolic filling pressure and stroke
volume (Ten Harkel et al., 1994), leading to a
decrease in cardiac output and arterial pressure.
Tachycardia may occur as a consequence of
reduced cardiac parasympathetic (vagal) activity,
reflexly induced by unloading the arterial bar-
oreceptors, but this response is not usually

sufficient to compensate for the decreased stroke
volume, and blood pressure remains low.
It is likely, however, that there are additional
factors that may predispose spinal cord injured
individuals to orthostatic hypotension. Individuals
with spinal cord injury are reported to have im-
paired baroreflex function (Wecht et al., 2003),
smaller plasma volumes due to hyponatremia
(Frisbie and Steele, 1997), and possible cardiovas-
cular deconditioning, at least in the early period
following spinal cord injury, due to prolonged pe-
riods of bed rest (Vaziri, 2003). Any combination
of these additional factors following spinal cord
injury would be likely to further increase the like-
lihood and severity of orthostatic hypotension.
Orthostatic hypotension is usually primarily as-
sociated with the acute phase of spinal cord injury
and improves over time. Although the reasons for
this improvement have not been clearly estab-
lished, the view that orthostatic hypotension is
only a temporary problem for spinal cord injured
individuals may not be entirely accurate. Indeed,
there is some evidence to suggest that, in some
cord-injured people, these troublesome episodes of
orthostatic hypotension can persist for many
years, and may even become worse with time
(Frisbie and Steele, 1997). It may be, however, that
with time, spinal cord injured individuals become
more tolerant to these hypotensive episodes. There
is some evidence to suggest that people with spinal
cord injury are able to tolerate profound hypo-
tension without symptoms (Illman et al., 2000),
and that this is related to alterations in cerebral
autoregulation such that cerebral blood flow
(Gonzalez et al., 1991) and/or oxygenation
(Houtman et al., 2000) are maintained, even in
the face of low cerebral perfusion pressures.
Management of orthostatic hypotension
Recognition of the symptoms of orthostatic
hypotension (lightheadedness, dizziness, blurred
vision, fatigue and others) is crucial in the man-
agement of individuals with spinal cord injury. In
emergency situations, it is imperative to return the
patient to a horizontal position and elevate the
lower extremities in order to prevent prolonged
227
cerebral ischemia. The major goals of non-phar-
macologic therapy for orthostatic hypotension are
the expansion of blood volume and avoidance of
excessive venous pooling (Oldenburg et al., 2002).
Shannon et al. reported that drinking water prior
to meals may reduce the risk of developing post-
prandial orthostatic hypotension (Shannon et al.,
2002). Plasma volume expansion by increased salt
intake is also recommended, particularly in the
mornings, at which time individuals with spinal
cord injury are reported to be more prone to hy-
potension (Oldenburg et al., 2002). It is also rec-
ommended to sleep with the head of the bed raised
by 15–301, which increases plasma volume, possi-
bly by activation of the renin–angiotensin–aldos-
terone system, and thus reducing the excretion of
sodium during the night (Oldenburg et al., 2002).
Another commonly used measure to prevent or-
thostatic hypotension in individuals with spinal
cord injury is the use of abdominal and lower ex-
tremity compression bandages to prevent pooling
of the blood in these regions (Tanaka et al., 1997).
Pharmacological treatment for orthostatic
hypotension may be used as an alternative or
adjunct, but is usually only considered when
non-pharmacological management fails (Mukand
et al., 2001; Oldenburg et al., 2002; Nieshoff
et al., 2004). The physiological targets for phar-
macological therapy are essentially the same as for
non-pharmacological therapy: expansion of the
circulating blood volume and prevention of
venous pooling (Oldenburg et al., 2002). One of
the most commonly used agents is fludrocortisone,
a mineralocorticoid that leads to sodium and
water retention (Groomes and Huang, 1991;
Barber et al., 2000), thus causing expansion of
the blood volume, and might also increase
a-adrenoceptor sensitivity. Side effects of this
treatment include hypokalemia and excessive wa-
ter retention, which limits its use in the elderly and
in individuals with congestive heart failure. The
most commonly prescribed agent for the manage-
ment of orthostatic hypotension following after
spinal cord injury is midodrine (Barber et al., 2000;
Mukand et al., 2001; Sclater and Alagiakrishnan,
2004). Midodrine is rapidly absorbed after oral
administration with peak serum levels occurring in
approximately 30 min. In the body it is metabolized
228
to desglymidodrine, an active a-1 adrenoreceptor
agonist. Side effects of midodrine include pruritus,
piloerection and urinary retention (Mukand et al.,
2001). Patients with autonomic dysreflexia could
be very sensitive to catecholamines and midodrine
as a result of changes in peripheral vascular
neurotransmission and vascular responsiveness.
Therefore, treatment should be started with low
doses, and gradually titrated to the effective dose
(Mukand et al., 2001).
In conclusion, cardiovascular dysfunction is a
serious problem after spinal cord injury that varies
with the severity and location of injury and with
the time after injury. The degree of cardiovascular
dysfunction after spinal cord injury correlates well
with the severity of injury to the spinal motor and
sensory pathways. Currently, treatments for
cardiovascular problems after cord injury can
address only the adverse symptoms. The unique
pathways in the spinal cord and organization of
spinal circuits controlling cardiovascular function
must be understood more completely in humans to
permit specific targeting of cardiovascular
dysfunctions by future strategies for protection
and repair of the injured cord.
References
Atkinson, P.P. and Atkinson, J.L.D. (1996) Spinal shock.
Mayo Clin. Proc., 71: 384–389.
Barber, D.B., Rogers, S.J., Fredrickson, M.D. and Able, A.C.
(2000) Midodrine hydrochloride and the treatment of ortho-
static hypotension in tetraplegia: two cases and a review of
the literature. Spinal Cord, 38(2): 109–111 Calaresu, F.R.
and Yardley, C.P. (1988) Medullary basal sympathetic tone.
Ann. Rev. Physiol., 50: 511–524.
Cariga, P., Ahmed, S., Mathias, C.J. and Gardner, B.P. (2002)
The prevalence and association of neck (coat-hanger) pain
and orthostatic (postural) hypotension in human spinal cord
injury. Spinal Cord, 40(2): 77–82.
Cleophas, T.J.M., Kauw, F.H.W., Bijl, C., Meijers, J. and
Stapper, G. (1986) Effects of beta-adrenergic-receptor ago-
nists and antagonists in diabetics with symptoms of postural
hypotension — a double-blind, placebo-controlled study.
ANGIA, 37(11): 855–862.
Curt, A.A., Nitsche, B., Rodic, B., Schurch, B. and Dietz, V.
(1997) Assessment of autonomic dysreflexia in patients with
spinal cord injury. J. Neurol. Neurosurg. Psychiatry., 62:
473–477.
DeVivo, M.J., Krause, J.S. and Lammertse, D.P. (1999) Recent
trends in mortality and causes of death among persons with
spinal cord injury. Arch. Phys. Med. Rehabil., 80(11):
1411–1419.
Ditunno, J.F., Little, J.W., Tessler, A. and Burns, A.S. (2004)
Spinal shock revisited: a four-phase model. Spinal Cord,
42(7): 383–395.
Eltorai, I., Kim, R., Vulpe, M., Kasravi, H. and Ho, W. (1992)
Fatal cerebral hemorrhage due to autonomic dysreflexia in a
tetraplegic patient: case report and review. Paraplegia, 30:
355–360.
Faghri, P.D., Yount, J.P., Pesce, W.J., Seetharama, S. and
Votto, J.J. (2001) Circulatory hypokinesis and functional
electric stimulation during standing in persons with spinal
cord injury. Arch. Phys. Med. Rehabil., 82(11): 1587–1595.
Frisbie, J.H. and Steele, D.J.R. (1997) Postural hypotension
and abnormalities of salt and water metabolism in myelopa-
thy patients. Spinal Cord, 35(5): 303–307.
Furlan, J.C., Fehlings, M.G., Shannon, P., Norenberg, M.D.
and Krassioukov, A.V. (2003) Descending vasomotor path-
ways in humans: correlation between axonal preservation
and cardiovascular dysfunction after spinal cord injury.
J. Neurotrauma, 20(12): 1351–1363.
Glenn, M.B. and Bergman, S.B. (1997) Cardiovascular changes
following spinal cord injury. Topics in Spinal Cord Injury
Rehabil., 2(4): 47–53.
Gonzalez, F., Chang, J.Y., Banovac, K., Messina, D.,
Martinez-Arizala, A. and Kelle, R.E. (1991) Autoregulation
of cerebral blood flow in patients with orthostatic hypoten-
sion after spinal cord injury. Paraplegia, 29: 1–7.
Groomes, T.E. and Huang, C.T. (1991) Orthostatic hypoten-
sion after spinal cord injury: treatment with fludrocortisone
and ergotamine. Arch. Phys. Med. Rehabil., 72(1): 56–58.
Guttman, F.L. and Whitteridge, D. (1947) Effects of bladder
distention on autonomic mechanisms after spinal cord inju-
ries. Brain, 70: 361–404.
Hadley, M. (2002a) Blood pressure management after acute
spinal cord injury. Neurosurgery, 50(3): S58–S62.
Hadley, M.N., Walters, B.C., Grabb, P.A., Oyesiku, N.M.,
Przybylski, G.J., Resnick, D.K. and Rylen, T.C. (2002b)
Guidelines for the management of acute cervical spine and
spinal cord injuries. Neurosurgery, 50(Suppl 3): S1–S199.
Hilton, J. (1860) Pain and therapeutic influence of mechanical
and physiological rest in accidents and surgical diseases.
Lancet, 2: 401–404.
Houtman, S., Colier, W.N., Oeseburg, B. and Hopman, M.T.
(2000) Systemic circulation and cerebral oxygenation during
head-up tilt in spinal cord injured individuals. Spinal Cord,
38: 158–163.
Illman, A., Stiller, K. and Williams, M. (2000) The prevalence
of orthostatic hypotension during physiotherapy treatment in
patients with an acute spinal cord injury. Spinal Cord, 38(12):
741–747.
Jacobsen, T.N., Nielsen, H.V., Kassis, E. and Amtorp, O.
(1992) Subcutaneous and skeletal-muscle vascular-responses
in human limbs to lower-body negative-pressure. Acta
Physiol. Scand., 144(3): 247–252.
King, B.S., Gupta, R. and Narayan, R.K. (2000) The early
assessment and intensive care unit management of patients

with severe traumatic brain and spinal cord injuries. Surg.
Clin. North Am., 80(3): 855–870.
Kirshblum, S.C., House, J.G. and O’Connor, K.C. (2002) Silent
autonomic dysreflexia during a routine bowel program in
persons with traumatic spinal cord injury: a preliminary
study. Arch. Phys. Med. Rehabil., 83: 1774–1776.
Krassioukov, A.V., Bunge, R.P., Puckett, W.R. and Bygrave,
M.A. (1999) The changes in human spinal cord sympathetic
preganglionic neurons after spinal cord injury. Spinal Cord,
37: 6–13.
Krassioukov, A.V., Furlan, J.C. and Fehlings, M.G. (2003)
Autonomic dysreflexia in acute spinal cord injury: an under-
recognized clinical entity. J. Neurotrauma, 20(8): 707–716.
Krassioukov, A.V., Johns, D.G. and Schramm, L.P. (2002)
Sensitivity of sympathetically correlated spinal interneurons,
renal sympathetic nerve activity, and arterial pressure to so-
matic and visceral stimuli after chronic spinal injury. J. Ne-
urotrauma, 19(12): 1521–1529.
Krassioukov, A.V. and Weaver, L.C. (1995) Episodic hyper-
tension due to autonomic dysreflexia in acute and chronic
spinal cord-injured rats. Am. J. Physiol., 268: H2077–H2083.
Lehmann, K.G., Lane, J.G., Piepmeier, J.M. and Batsford,
W.P. (1987) Cardiovascular abnormalities accompanying
acute spinal cord injury in humans: incidence, time course
and severity. J. Am. Coll. Cardiol., 10(1): 46–52.
Maiorov, D.N., Fehlings, M.G. and Krassioukov, A.V. (1998)
Relationship between severity of spinal cord injury and ab-
normalities in neurogenic cardiovascular control in conscious
rats. J. Neurotrauma, 15(5): 365–374.
Maiorov, D.N., Weaver, L.C. and Krassioukov, A.V. (1997)
Relationship between sympathetic activity and arterial pres-
sure in conscious spinal rats. Am. J. Physiol., 272:
H625–H631.
Mathias, C.J. (1995) Orthostatic hypotension: causes, mecha-
nisms, and influencing factors. Neurology, 45(Suppl 5):
S6–S11.
Mathias, C.J. and Frankel, H.L. (1992a) Autonomic distur-
bances in spinal cord lesions. In: Bannister R. and Mathias
C.J. (Eds.) Autonomic Failure, a Textbook of Clinical Dis-
orders of the Autonomic Nervous System, Vol. 3(43). Oxford
Medical Publications, Oxford, UK, pp. 839–881.
Mathias, C.J. and Frankel, H.L. (1992b) The cardiovascular
system in tetraplegia and paraplegia. In: Frankel H.L. (Ed.)
Handbook of Clinical Neurology, Vol. 17(25). Elsevier Sci-
ence Publishers, B.V., Amsterdam, pp. 435–456.
Mukand, J., Karlin, L., Barrs, K. and Lublin, P. (2001) Mid-
odrine for the management of orthostatic hypotension in
patients with spinal cord injury: a case report. Arch. Phys.
Med. Rehabil., 82(5): 694–696.
Nacimiento, W. and Noth, J. (1999) What, if anything, is spinal
shock? Arch. Neurol., 56(8): 1033–1035.
Nieshoff, E.C., Birk, T.J., Birk, C.A., Hinderer, S.R. and
Yavuzer, G. (2004) Double-blinded, placebo-controlled trial
of midodrine for exercise performance enhancement in
Tetraplegia: a pilot study. J. Spinal Cord Med., 27(3):
219–225.
229
Oldenburg, O., Kribben, A., Baumgart, D., Philipp, T., Erbel,
R. and Cohen, M.V. (2002) Treatment of orthostatic hypo-
tension. Curr. Opin. Pharmacol., 2(6): 740–747.
Osborn, J.W., Taylor, R.F. and Schramm, L.P. (1989) Deter-
minants of arterial pressure after chronic spinal transection in
rats. Am. J. Physiol., 256: R666–R673.
Osborn, J.W., Taylor, R.F. and Schramm, L.P. (1990) Chronic
cervical spinal cord injury and autonomic hyperreflexia in
rats. Am. J. Physiol., 258: R169–R174.
Piepmeier, J.M., Lehmann, K.B. and Lane, J.G. (1985) Cardi-
ovascular instability following acute cervical spinal cord
trauma. Central Nervous System Trauma, 2(3): 153–160.
Pine, Z.M., Miller, S.D. and Alonso, J.A. (1991) Atrial fibril-
lation associated with autonomic dysreflexia. Am. J. Phys.
Med. Rehabil., 70(5): 271–273.
Sclater, A. and Alagiakrishnan, K. (2004) Orthostatic hypo-
tension — a primary care primer for assessment and treat-
ment. Geriatrics, 59(8): 22–27.
Shannon, J.R., Diedrich, A., Biaggioni, I., Tank, J., Robertson,
R.M., Robertson, D. and Jordan, J. (2002) Water drinking as
a treatment for orthostatic syndromes. Am. J. Med., 112(5):
355–360.
Silver, J.R. (2000) Early autonomic dysreflexia. Spinal Cord,
38: 229–233.
Tanaka, H., Yamaguchi, H. and Tamai, H. (1997) Treatment of
orthostatic intolerance with inflatable abdominal band. Lan-
cet, 349(9046): 175.
Teasell, R., Arnold, A.P., Krassioukov, A.V. and Delaney,
G.A. (2000) Cardiovascular consequences of loss of supra-
spinal control of the sympathetic nervous system following
spinal cord injuries. Arch. Phys. Med. Rehabil., 81: 506–516.
Ten Harkel, A.D., van Lieshout, J.J. and Wieling, W. (1994)
Effects of leg muscle pumping and tensing on orthostatic
arterial pressure: a study in normal subjects and patients with
autonomic failure. Clin. Sci., 87(5): 553–558.
Vale, F.L., Burns, J., Jackson, A.B. and Hadley, M.N. (1997)
Combined medical and surgical treatment after acute spinal
cord injury: results of a prospective pilot study to assess the
merits of aggressive medical resuscitation and blood pressure
management. J. Neurosurg., 87(2): 239–246.
Vaziri, N.D. (2003) Nitric oxide in microgravity — induced
orthostatic intolerance: relevance to spinal cord injury.
J. Spinal Cord Med., 26(1): 5–11.
Wecht, J.M., De Meersman, R.E., Weir, J.P., Spungen, A.M.
and Bauman, W.A. (2003) Cardiac autonomic responses to
progressive head-up tilt in individuals with paraplegia. Clin.
Auton. Res., 13(6): 433–438.
Winslow, E.B., Lesch, M., Talano, J.V. and Meyer Jr., P.R.
(1986) Spinal cord injuries associated with cardiopulmonary
complications. Spine, 11(8): 809–812.
Yarkony, G.M., Katz, R.T. and Wu, Y. (1986) Seizures sec-
ondary to autonomic dysreflexia. Arch. Phys. Med. Rehabil.,
67: 834–835.
Yeoh, M., McLachlan, E.M. and Brock, J.A. (2004) Tail arteries
from chronically spinalized rats have potentiated responses to
nerve stimulation in vitro. J. Physiol., 556: 545–555.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 15

Orthostatic hypotension and paroxysmal

hypertension in humans with high spinal cord injury

Christopher J. Mathias1,2,

1
Neurovascular Medicine Unit, Faculty of Medicine, Imperial College London at St Mary’s Hospital, London W2 1NY, UK
2
Autonomic Unit, National Hospital for Neurology and Neurosurgery, Queen Square, and Institute of Neurology,
University College London, London, UK

Abstract: The spinal cord is essential for normal autonomic nervous system regulation of the cardiovas-
cular system as the preganglionic neurons controlling the heart and blood vessels originate in the tho-
racolumbar spinal segments. The site and extent of a spinal cord injury determine the degree of autonomic
involvement in cardiovascular dysfunction after the injury. After complete cervical cord lesions the entire
sympathetic outflow is separated from cerebral control; this may cause orthostatic hypotension. Commonly
after traumatic injuries to the spinal cord, one or more segments are totally destroyed. However, the distal
portion of the spinal cord often retains function and activation of spinal cord reflexes working independ-
ently of the brain can result in paroxysmal hypertension. This chapter will focus on orthostatic hypotension
and paroxysmal hypertension in cord-injured people with lesions affecting the cervical and upper thoracic
spinal cord. Conditions promoting these abnormalities in blood pressure will be elaborated. Possible
mechanisms for the hypo- and hypertension will be discussed, as will strategies for managing these prob-
lems.

Introduction are totally destroyed. However, the distal portion

The spinal cord is essential for normal functioning
of the autonomic nervous system, as the entire
sympathetic outflow (from the 1st thoracic (T1) to
upper lumbar (L2/3)) and a proportion of the
parasympathetic outflow (the sacral parasympa-
thetic) travel in the spinal cord before they reach
their target organs. In spinal cord injuries, there-
fore, there are varying degrees of autonomic in-
volvement, depending upon the site and extent of
the lesion. In complete tetraplegics with cervical
cord lesions the entire sympathetic outflow is sep-
arated from cerebral control; this may cause or-
thostatic hypotension. Commonly after traumatic
injuries to the spinal cord, one or more segments
Corresponding author. Tel.: +0207 886 1468;
Fax: +0207 996 1540; E-mail: c.mathias@imperial.ac.uk
of the spinal cord often retains function and ac-
tivation of spinal cord reflexes working independ-
ently of the brain can result in paroxysmal
hypertension. This chapter will focus on ortho-
static hypotension and paroxysmal hypertension
in cord-injured people with lesions affecting the
cervical and upper thoracic spinal cord.
Basal blood pressure
In recently injured tetraplegics the basal supine
level of blood pressure usually is lower than nor-
mal (mean arterial pressure: 57 mmHg in tetra-
plegics and 82 mmHg in normal subjects; Mathias
et al., 1979a). It is dependent upon a number of
factors, including complicating trauma and drug
therapy. The lower levels of blood pressure appear

DOI: 10.1016/S0079-6123(05)52015-6 231

232
secondary to diminution in sympathetic nervous
activity (Stjernberg et al., 1986). It is unlikely that
skeletal muscle paralysis alone is the explanation,
as tetraplegics due to poliomyelitis often have
normal or even higher levels of blood pressure.
In recently injured tetraplegics the basal heart
rate is usually below 100 beats/min, unlike low
spinal cord injuries in whom the heart rate often is
higher. This probably is due to a reduction in
neural and hormonal sympathetic chronotropic
influences in high lesions. The efferent parasym-
pathetic cardiac pathways, however, are intact and
the absence of sympathetic activation may predis-
pose susceptible patients to vagal over-activity.
This may result in bradycardia and cardiac arrest,
as may occur during tracheal stimulation (Frankel
et al., 1975; Mathias, 1976)
In the chronic stage, the supine blood pressure
in people with high lesions is lower than in normal
subjects (Frankel et al., 1972). Non-invasive am-
bulatory 24 h recordings confirm loss of nocturnal
circadian fall in blood pressure, which occurs in
normal subjects (Nitsche et al., 1996). Basal levels
of plasma noradrenaline and adrenaline in tetra-
plegics are low, consistent with a diminished
peripheral sympathetic activity, caused by absence
of tonic supraspinal drive. This has been con-
firmed by measuring skin and muscle sympathetic
nerve activity using microneurography (Stjernberg
et al., 1986). Subjects with spinal cord injury,
however, are prone to renal damage and renal
failure, as complications of lower urinary tract
dysfunction (see also Karlsson, this volume) may
account for sustained hypertension in some.
In the absence of adequate resting sympathetic
tone a number of secondary mechanisms, particu-
larly hormonal, attempt to maintain blood pressure.
An important component is the renin–angioten-
sin–aldosterone system, through the direct pressor
effects of angiotensin-II and the salt-retaining ef-
fects of aldosterone. Drugs that interfere with the
system, such as the angiotensin-converting enzyme
inhibitor, captopril, substantially lower supine
blood pressure in tetraplegics. Small doses of di-
uretics, that cause salt loss and lower intravascular
fluid volume, may cause a marked fall in supine
blood pressure. A low salt diet lowers blood pres-
sure, despite the ability of tetraplegics to reduce
salt excretion as do normal subjects (Sutters et al.,
1992). In tetraplegics, recumbency may induce a
diuresis but not a natriuresis; this differs from
subjects with primary autonomic failure, who also
have nocturnal polyuria (Mathias et al., 1986), and
in whom recumbency causes both diuresis and
natriuresis (Kooner et al., 1987). The difference
may relate to the ability of tetraplegics to mount
an adequate hormonal response to oppose natriur-
esis, unlike those in autonomic failure in whom
these responses often are muted. These observa-
tions have practical importance, as a period of re-
cumbency in people with high spinal lesions will
often result in accentuation of orthostatic hypo-
tension. This may be reduced or prevented by
head-up tilt.
Orthostatic hypotension
Subjects with high spinal cord lesions are prone to
orthostatic (postural) hypotension when changing
from the horizontal to upright position (Fig. 1).
Orthostatic hypotension is defined as a decrease in
systolic blood pressure of more than 20 mmHg, or
a fall in diastolic blood pressure of more than
10 mmHg, while upright or during head-up tilt to
60o, for at least 3 min (Schatz et al., 1996). The fall
in blood pressure is accompanied by a variety of
symptoms (Table 1); these can vary in nature and
intensity and are not necessarily related to the
degree of hypotension.
During head-up postural change there usually is
an immediate fall in blood pressure, often to ex-
tremely low levels. There usually is no loss of con-
sciousness, except in recently injured tetraplegics
or in chronic tetraplegics, following a period of
recumbency. This tolerance to a low cerebral per-
fusion pressure is similar to that observed in pa-
tients with primary autonomic failure, who have
the ability to autoregulate their cerebral circula-
tion at considerably lower perfusion pressure levels
than normal subjects (Goadsby, 2002).
Following the initial fall in blood pressure, the
subsequent responses vary. In some subjects,
blood pressure continues to fall as head-up tilt is
maintained. There is no rise in levels of plasma
noradrenaline following head-up postural change,

Fig. 1. Top panels: blood pressure (BP) and heart rate (HR) in
a tetraplegic before and after head-up tilt, in the early stages of
rehabilitation, when there were few muscle spasms and minimal
autonomic dysreflexia. Bottom panels: Blood pressure (BP) and
heart rate (HR) in a tetraplegic before, during, and after head-
up tilt to 45o. Blood pressure promptly falls but with partial
recovery, which in this case is linked to skeletal muscle spasms
(S) inducing spinal sympathetic activity. Some of the later os-
cillations may be due to a rise in plasma renin, which was
measured where there are interruptions in the blood pressure
records. In the later phases of tilt, skeletal muscle spasms occur
more frequently and further elevate the blood pressure. On re-
turn to the horizontal, blood pressure rises rapidly above the
previous level, and then slowly returns to a steady value. Heart
rate usually moves in the opposite direction to blood pressure,
except during muscle spasms, when there is an increase. From
Mathias and Frankel (1988). With Permission.
which is consistent with an inability to increase
sympathetic nervous activity reflexly in response to
postural change, as is expected. If tilt is prolonged,
blood pressure tends to partly recover. This
recovery may in part be related to activation of
the renin–angiotensin–aldosterone system. Release
of renin appears independent of sympathetic
stimulation and is probably secondary to renal
baroreceptor stimulation from the fall in renal per-
fusion pressure (Mathias et al., 1975). Renin results
in formation of the peptide angiotensin II, a pow-
erful direct-acting vasoconstrictor. Angiotensin II
also facilitates peripheral noradrenaline release and
activity, and stimulates the release of aldosterone
233
Table 1. Some of the symptoms resulting from orthostatic hy-
potension and impaired perfusion of various organs
Cerebral hypoperfusion
Dizziness
Visual disturbances
Blurred
Tunnel
Scotoma
Greying out
Blacking out
Colour defects
Loss of consciousness
Cognitive deficits
Muscle hypoperfusion
Paracervical and suboccipital (‘coat-hanger’) ache
Subclavian steal-like syndrome
Renal hypoperfusion
Oliguria
Non-specific
Weakness, lethargy, fatigue
Adapted from Mathias (2003a). With permission.
from the adrenal cortex. The salt and water
retaining effects of aldosterone increase intra-
vascular volume. These various actions of the
renin–angiotensin–aldosterone system help raise
blood pressure. During tilt there also may be
activation of spinal reflexes from stimulation
of skin, skeletal muscles or viscera, as part of
autonomic dysreflexia.
During head-up postural change the fall in
blood pressure is accompanied by a reduction in
central venous pressure, stroke volume, and car-
diac output, which is probably the result of venous
pooling, diminished venous return, and the inabil-
ity to increase sympathetic cardiac inotropic
activity. Venous pooling may cause cyanotic dis-
coloration of the legs and may account for ankle
oedema. Urine volume is usually reduced, often to
low levels and may raise the question of urinary
outflow tract obstruction. Oliguria is due to the
fall in blood pressure, which reduces renal plasma
flow and glomerular filtration rate. Increased levels
of the antidiuretic hormone, vasopressin, also
probably contribute. In subjects with high spinal
lesions there is an exaggerated rise in vasopressin
levels during head-up tilt when compared to nor-
mal subjects.
During head-up postural change there often is a
rapid rise in heart rate, which is inversely related to
234
the fall in blood pressure. This is due to with-
drawal of vagal tone presumably due to unloading
of baroreceptor afferents; it is markedly attenuat-
ed although not abolished by atropine. Prop-
ranolol also reduces the heart rate rise during tilt,
suggesting that beta-adrenoceptor stimulation par-
tially contributes. Despite marked orthostatic hy-
potension, the heart rate does not usually rise
above 100 beats/min. This is different from pa-
tients with an intact sympathetic nervous system,
in ‘shock’ with a similarly low level of blood pres-
sure, in whom heart rate usually rises considerably
above 100 beats/min.
The clinical problems resulting from orthostatic
hypotension in people with high spinal lesions are
not usually as severe and prolonged as the prob-
lems in subjects with primary autonomic failure.
There usually is no loss of consciousness, except in
recently injured tetraplegics, in the early stages of
rehabilitation, or in chronic tetraplegics following
a period of recumbency. This tolerance to a low
cerebral perfusion pressure is similar to that ob-
served in patients with primary autonomic failure,
who have greater autoregulation of their cerebral
circulation as described above (Goadsby, 2002).
Subjects may be affected in various ways, as out-
lined in Table 1. There may be transient dizziness
when changing to the upright position, impaired
concentration and attention (Critchley and Mathias,
2003), or a ‘coat-hanger’ neck ache (Cariga et al.,
2002). In some, arm movement that increases
upper limb blood flow may cause a subclavian
steal-like effect, and cause symptoms of cerebral
hypoperfusion by reducing blood flow to the brain
stem. Tiredness and lethargy may be related to a
low level of blood pressure, as has been noted in
primary autonomic failure (Mathias et al., 1999).
Management consists largely of non-pharmaco-
logical measures, based on knowledge of patho-
physiological processes (Table 2) Symptoms of
orthostatic hypotension often are diminished with
frequent postural change to the head-up position,
along with elevation of the head end of the bed at
night. This may activate the renin–angiotensin–al-
dosterone axis, causing vasoconstriction and plas-
ma volume expansion, which helps buffer the fall
in blood pressure during head-up postural change.
An improved ability to autoregulate cerebral
Table 2. Some of the non-pharmacological approaches used in
the management of orthostatic hypotension
To be avoided
Sudden head-up postural change (especially on waking)
Prolonged recumbency
Straining during micturition and defecation
High environmental temperature (including hot baths)
Drugs with vasodepressor properties
To be introduced
Head-up tilt during sleep
High salt intake
Adopting different body positions
To be considered
Elastic stockings
Thigh cuffs
Abdominal binders
Water ingestion
Adapted from Mathias (2003b). With permission.
blood flow at lower perfusion pressures probably
also contributes. Some of the newer treatments to
improve motor function, such as functional elec-
trical stimulation, may reduce orthostatic hypo-
tension (Sampson et al., 2000).
Awareness of the many factors that lower blood
pressure is important. These include simple meas-
ures such as avoiding rapid postural change, es-
pecially in the morning when getting out of bed.
The supine blood pressure often is lowest in the
morning, especially in those in whom nocturnal
polyuria reduces extracellular fluid volume. Pro-
longed bed rest and recumbency, especially post-
operatively, should be avoided. Straining during
micturition and bowel movement should be avoid-
ed. The blood pressure may fall to extremely low
levels during straining, similar to that observed
during the Valsalva manoeuvre even when
intrathoracic pressure is elevated only to 20 or
30 mmHg (van Lieshout et al., 1991). In hot
weather, body temperature may rise as thermo-
regulatory mechanisms such as sweating are im-
paired; this may further increase vasodilatation
and worsen orthostatic hypotension. In chronic
autonomic failure, ingestion of alcohol or large
meals, especially those containing a high carbohy-
drate content (Mathias et al., 1989), may cause
supine postprandial hypotension and aggravate
orthostatic hypotension post-meal (Mathias et al.,

1991; Chaudhuri et al., 1994). In tetraplegics, this
post-meal hypotension does not seem to occur to
the same extent; there is a modest fall in blood
pressure accompanied by an elevation in heart rate
(Baliga et al., 1997). Levels of forearm venous
plasma noradrenaline do not change, excluding a
generalized increase in sympathetic nerve activity.
The mechanisms responsible for this difference in
tetraplegics are unclear and could include activa-
tion of pressor reflexes from the gastrointestinal
tract (Mathias and Bannister, 2002). Arm exercise
may induce hypotension in subjects with high le-
sions (King et al., 1992; 1994).
In some cord-injured people, preventing venous
pooling when upright by using abdominal binders,
thigh cuffs, and lower limb elastic stockings may
be beneficial. Each has its limitations. Recent ob-
servations indicate that ingestion of 500 ml of wa-
ter raises supine blood pressure substantially and
improves orthostatic hypotension in primary au-
tonomic failure (Cariga et al., 2001; Mathias and
Young, 2004); it also appears to raise blood pres-
sure in people with high lesions (Tank et al., 2003).
Activation of spinal sympathetic reflexes, by in-
duction of muscle spasms or tapping of the ante-
rior abdominal wall suprapubically to contract the
urinary bladder, may cause autonomic dysreflexia
and thus elevate blood pressure.
A range of drugs is used to reduce orthostatic
hypotension in primary autonomic failure (Table 3);
some of these drugs have been used with success
in spinal injuries. However, subjects with high
lesions, unlike those with autonomic failure, are
prone to paroxysms of hypertension, the severity of
which may be exacerbated by such drugs. Drugs
with short half-lives are preferable and ideally
should be used only for limited periods, such as in
the early stages of rehabilitation. Ephedrine in a
dose of 15 mg, half an hour before postural change
often is of value. The alpha-adrenoceptor agonist
midodrine may have a role (Barber et al., 2000;
Mukand et al., 2001). In the majority of patients,
however, drugs are not needed. Although ortho-
static hypotension in people with high lesions is
primarily due to impaired sympathetic activation, in
clinical practice it is important to be aware that a
variety of non-neurogenic factors (Table 4) may
contribute.
235
Table 3. Outline of the major actions by which a variety of
drugs may reduce orthostatic hypotension
Reducing salt loss/plasma volume expansion
Mineralocorticoids (fludrocortisone)
Reducing nocturnal polyuria
V2-receptor agonists (desmopressin)
Vasoconstriction — sympathetic
On resistance vessels (ephedrine, midodrine, phenylephrine,
noradrenaline, clonidine, tyramine with monoamine oxidase
inhibitors, yohimbine, L-dihydroxyphenylserine)
On capacitance vessels (dihydroergotamine)
Vasoconstriction — non-sympathomimetic
V1 receptor agents — terlipressin
Ganglionic nicotinic-receptor stimulation
Cholinesterase inhibitors — pyridostigminc
Preventing vasodilatation
Prostaglandin synthetase inhibitors (indomethacin,
flurbiprofen)
Dopamine receptor blockade (metoclopramide,
domperidone)
Beta-adrenoceptor blockade (propranolol)
Preventing postprandial hypotension
Adenosine receptor blockade (caffeine)
Peptide release inhibitors (somatostatin analogue: octreotide)
Increasing cardiac output
Beta-blockers with intrinsic sympathomimetic activity
(pindolol, xamoterol)
Dopamine agonists (ibopamine)
Increasing red cell mass
Erythropoietin
Adapted from Mathias (2003b). With permission.
Paroxysmal hypertension
Subjects with high spinal lesions may have an ex-
aggerated rise in blood pressure in response to
stimuli that originate below the level of the lesion
(Fig. 2). These may arise from skin, abdominal and
pelvic viscera, or by contraction of skeletal mus-
cles. There usually is a fall in heart rate because of
increased vagal activity. In recently injured tetra-
plegics there usually is no change in blood pressure
or heart rate during such stimulation (Fig. 3), in
contrast to the chronic stage when spinal cord re-
flexes have recovered. The cardiovascular effects
are part of the syndrome of autonomic dysreflexia.
Other features include sweating in skin around and
above the level of the lesion, contraction of the
urinary bladder and rectum, penile erection, seminal
fluid emission and skeletal muscle spasms — com-
ponents of the ‘mass reflex’ described by Head and
236

Table 4. Examples of non-neurogenic causes of orthostatic hypotension

Low intravascular volume

Blood/plasma loss Haemorrhage, burns
Fluid/electrolyte deficiency
Diminished intake Vomiting
Loss from gut Diarrhea
Loss from kidney Salt losing nephropathy, diuretics
Endocrine deficiency Adrenal insufficiency (Addison’s disease)
Cardiac insufficiency
Myocardial Myocarditis
Impaired ventricular filling Atrial myxoma, constrictive pericarditis
Impaired output Aortic stenosis
Vasodilatation
Endogenous Hyperpyrexia
Exogenous Drugs such as glyceryl trinitrate (GTN)
Alcohol
Excessive heat

Fig. 2. Blood pressure (BP), heart rate (HR), intravesical pressure (IVP), plasma noradrenaline (NA) and adrenaline (A) levels in a
tetraplegic subject before, during and after bladder stimulation induced by suprapubic percussion of the anterior abdominal wall. The
rise in BP is accompanied by a fall in heart rate as a result of increased vagal activity in response to the rise in blood pressure. There is a
rise in levels of plasma NA (open histograms), but not A (filled histograms) suggesting an increase in sympathetic neural activity
independently of adrenomedullary activation. From Mathias and Frankel (1986). With permission.

Riddoch in 1917 (Table 5). The rise in blood pres-
sure, first reported by Guttmann and Whitteridge
in 1947, is secondary to vasoconstriction in resist-
ance vessels and capacitance vessels (Corbett et al.,
1971). Stroke volume and cardiac output increase
due to activation of spinal cardiac reflexes when the
lesion is rostral to the cardiac sympathetic
preganglionic neurons (Corbett et al., 1975). The
rapid rise in blood pressure after stimulation is
suggestive of reflex sympathetic activity, through
 237

Fig. 3. Average levels of mean blood pressure (MBP), heart rate (HR), plasma noradrenaline (NA, continuous line) and adrenaline
(A, interrupted line) in recently injured and chronic tetraplegics before, during and after bladder stimulation (BS). The bars indicate
7SEM. No changes occur in the recently injured tetraplegics, unlike the chronic tetraplegics in whom MBP and plasma NA levels rise
and HR falls. There are no changes in plasma A levels. From Mathias et al. (1979a). With permission.

Table 5. Clinical manifestations of autonomic dysreflexia
Paraesthesiae in neck, shoulders and arms
Fullness in head
Hot ears
Throbbing headache, especially in the occipital and frontal
regions
Tightness in chest and dyspnea
Hypertension and bradycardia
Occasionally cardiac dysrhythmias
Pupillary dilatation
Pallor initially, followed by flushing of face and neck and
sweating in areas above and around the lesion (above lesion)
Cold peripheries; piloerection (below lesion)
Contraction of urinary bladder and large bowelaa –4
Penile erection and seminal fluid emissiona
Adapted from Mathias and Frankel (2002).
a
May occur as part of the ‘mass’ reflex.
the isolated spinal cord. This interpretation is sup-
ported by the fact that plasma noradrenaline, but
not adrenaline, levels increase during dysreflexia
(Mathias et al., 1976a). Adrenomedullary secretion
does not contribute to this elevation in blood
pressure. This differs markedly from the extremely
high levels of plasma catecholamines often found
in pheochromocytoma where there also is blood
pressure lability. During autonomic dysreflexia, cir-
culating levels of other vasoconstrictor substances,
such as renin (and by inference angiotensin-II lev-
els), aldosterone, vasopressin, and atrial natriuretic
peptide remain unchanged or fall (Mathias et al.,
1981; Krum et al., 1992). Whether levels of other
vasoconstrictor peptides such as neuropeptide Y
(NPY) and endothelin rise in man is not known.
238
The rise in blood pressure and the widespread
involvement of the vasculature below the lesion,
despite a modest and often localized stimulus only
involving a few segments, suggest the spread of
neuronal impulses intraspinally and/or extraspin-
ally. In tetraplegics, microneurography indicates
only a moderate and transient rise in muscle sym-
pathetic nerve activity during autonomic dysre-
flexia (Stjernberg et al., 1986) with no correlation
between the cardiac cycle and muscle sympathetic
nerve discharge as occurs normally. Hyperactivity
of target organs innervated by the autonomic
nervous system has been demonstrated by re-
sponses of the dorsal foot vein of tetraplegics to
local intravenous noradrenaline (Arnold et al.,
1995). The exaggerated blood pressure response to
various stimuli, including to noradrenaline infused
intravenously (Mathias et al., 1976b), suggests su-
persensitivity of adrenoceptors, or that other
mechanisms are responsible for the enhanced vas-
cular response (see also McLachlan and Brock,
this volume). Overall, the results of physiological
and pharmacological studies, in conjunction with
the neurohormonal observations, indicate that au-
tonomic dysreflexia is a more appropriate term
than autonomic hyperreflexia. However, data ob-
tained from indirect techniques, such as total and
regional body noradrenaline spillover, indicate a
marked (332% total and 15-fold in the leg) in-
crease during bladder stimulation, and suggest
that in people with high lesions, greater quantities
of noradrenaline may be released per impulse dur-
ing dysreflexia (Karlsson et al., 1998; Gao et al.,
2002). This suggestion is based on the assumption
that the sparse firing of muscle sympathetic
fibres recorded by Stjernberg et al. (1986) during
dysreflexia can be generalized to reflect the entire
sympathetic outflow below the lesion. The spill-
over data of Gao et al. (2002) were interpreted as
possible evidence for a significant increase in sym-
pathetic nerve firing during dysreflexia, as has been
documented in the rat (Maiorov et al., 1997).
However, these findings could indicate greater
numbers of axons firing, higher firing rates, greater
release of transmitter per impulse, or reduced
noradrenaline reuptake by presynaptic terminals.
The mechanisms accounting for the substantial
rise in blood pressure during autonomic dysreflexia
are likely to include peripheral neural and vascular
changes, as well as central nervous system plastic-
ity (Weaver, 2002), and need further investigation,
in the light of clinical and recent experimental
data (see Weaver et al.; McLachlan and Brock;
Rabchevsky; Schramm, this volume).
The exaggerated pressor responses to stimuli
causing autonomic dysreflexia do not occur in le-
sions below the fifth thoracic segment (Fig. 4). This
indicates that the sympathetic neural outflow below
T5, which incorporates neural control of the large
splanchnic circulatory bed, is of major importance
in generating the blood pressure response during
dysreflexia and also in maintaining blood pressure
homeostasis. Mild episodes of autonomic dysreflex-
ia probably occur intermittently through the day
in response to various stimuli, but often are not
noticed and may be of little consequence. Severe
autonomic dysreflexia is of major clinical impor-
tance. Hypertension, especially linked to bladder
contractions and voiding, may not be accompa-
nied by symptoms (silent autonomic dysreflexia;
Linsenmeyer et al., 1996). When autonomic dysre-
flexia is prolonged there may be considerable
Fig. 4. Changes in mean blood pressure (DMBP) and heart rate
(DHR) in subjects with spinal cord lesions at different levels
(cervical and thoracic) after bladder stimulation induced by
suprapubic percussion of the anterior abdominal wall. In the
cervical and high thoracic lesions there is a marked elevation in
blood pressure and a fall in heart rate. In lesions below T5 there
are minimal cardiovascular changes. From Mathias and Frankel
(1986). With permission.
 239

morbidity, as a result of excessive sweating over Table 6. Causes of autonomic dysreflexia

the head and neck, and a throbbing headache. The Abdominal or pelvic visceral stimulation
latter is often, but not always, related to the level Ureter
of blood pressure and may be dependent on dis- Calculus
tension of pain-sensitive cranial blood vessels. Urinary bladder
With recurrent episodes of dysreflexia, headache Distension by blocked catheter or discoordinated bladder
Infection
may become a particularly severe symptom despite Irritation by calculus, catheter or bladder washout
modest elevations in blood pressure. Other com- Rectum and anus
plications of vasospasm and hypertension during Enemas
autonomic dysreflexia include myocardial failure Fecal retention
and neurological complications such as epileptic Anal fissure
Gastrointestinal organs
seizures, visual defects, and cerebral hemorrhage. Gastric dilatation
These may result in extensive and permanent Gastric ulceration
neurological deficits, or even death. Cholecystitis or cholelithiasis
The key factor in the management of autonomic Uterus
dysreflexia is to determine the provoking cause Contraction during pregnancy
Menstruation, occasionally
(Table 6), and rectify it. To lower blood pressure
rapidly, head-up tilt (which causes venous pooling) Cutaneous stimulation
initially may be used. A range of drugs, that act Pressure sores
Infected ingrowing toenails
through the proposed mechanisms responsible for Burns
autonomic dysreflexia, have been tried (Table 7).
Preventing afferent stimulation, for instance by the Skeletal muscle spasms
Especially in limbs with contractures
use of a local anaesthetic, such as lignocaine in the
urinary bladder, can be effective. Drugs that act Miscellaneous
partially (reserpine) or entirely (spinal anaesthet- Intrathecal neogstimine
Electroejaculatory procedures
ics) on the spinal cord are of particular use in se- Ejaculation
vere episodes of dysreflexia. Some may be given Vaginal dilatation
intrathecally (Middleton et al., 1996). The gangl- Urethra — insertion of catheter or abscess
ionic blocker, hexamethonium, was successfully Fracture of bones
used in the past, but, like other drugs that reduce
Adapted from Mathias and Frankel (2002). With permission.
sympathetic efferent activity, may cause profound
hypotension. The a2-adrenoceptor agonist, cloni-
dine, does not lower supine blood pressure in

Table 7. Some of the drugs used in the management of autonomic dysreflexia classified according
to their major site of action on the reflex arc and target organs

Afferent Topical lignocaine

Spinal cord Clonidinea
Reserpinea
Spinal anaesthetics
Efferent Sympathetic ganglia Hexamethonium
Sympathetic nerve terminals Guanethidine
a-adrenoceptors Phenoxybenzamine
Target Organs Blood vessels Glyceryl trinitrate
Nifedipine
Sweat glands Pro-Banthinea

Adapted from Mathias and Frankel (2002). With permission.

a
These drugs have multiple effects, some of which are peripheral
240

Fig. 5. Ambulatory blood pressure measurements demonstrating (A), a normal profile with a preserved physiological fall in blood
pressure at night between midnight and 04.00 h, (B) multiple episodes of autonomic dysreflexia in a complete tetraplegic and (C) the
same subject as in (B) after effective treatment of autonomic dysreflexia with the antihypertensive drug nifedipine. Despite improve-
ments in (C), the loss of the circadian regulation of blood pressure persists. In each panel, the top half shows systolic, mean, and
diastolic pressure (in mmHg, from top to bottom) and the bottom half shows heart rate in beats/min (bpm). From Curt et al. (1997).
 241

Fig. 6. Blood pressure (BP) and heart rate (HR) in a tetraplegic in the supine position before, during, and after bladder stimulation
(BS) by suprapubic percussion of the anterior abdominal wall, a procedure that induces hypertension. Sublingual glyceryl trinitrate
(GTN) (0.5 mg for 3.5 min) rapidly reverses the hypertension, elevates the heart rate and then causes substantial hypotension. Levels of
plasma renin activity (PRA) rise as a result of the fall in blood pressure. From Mathias and Frankel (1988).

people with high lesions (Reid et al., 1977), but
reduces hypertension during autonomic dysreflexia
(Mathias et al., 1979b). Alpha-adrenoceptor
blockers, such as phenoxybenzamine, and the se-
lective blockers prazosin and terazosin are useful
in autonomic dysreflexia due to bladder outflow
obstruction, as they relax the smooth muscle of the
urinary sphincter (Vaidyanathan et al., 1998).
Drugs acting directly on blood vessels, such as
glyceryl trinitrate and calcium channel blockers,
such as nifedipine are effective (Thyberg et al., 1994;
Curt et al., 1997) (Fig. 5), and have the advantage of
being given sublingually. They also have the poten-
tial to lower blood pressure substantially (Fig. 6).
In some subjects, autonomic dysreflexia may be a
major and recurring problem because of difficulty in
either defining or resolving the precipitating cause.
Unusual examples of the former are gastric ulcer-
ation or cholecystitis, which are difficult to detect
because of lack of pain. A common example, which
may be overlooked, is an anal fissure. Paroxysmal
hypertension can be a particular problem during
surgery, especially if either the urinary bladder or
the large bowel is involved. In these patients hyper-
tension is often controlled successfully by deepening
the level of spinal or general inhalation anesthesia,
along with an increase in positive pressure ventila-
tion (Welply et al., 1975). Short-acting ganglionic
blockers or direct-acting vasodilators have also been
successfully used during surgery. During labour,
severe hypertension may occur and should be
avoided, ideally by using a spinal anaesthetic.
Despite recognizing the cause, it may be ex-
tremely difficult to resolve problems that include
severe skeletal muscle spasms or recurrent urinary
bladder infection. Long-term drug therapy for au-
tonomic dysreflexia in such subjects is often only
partially successful and may result in undesirable
side effects. In severe cases, surgical procedures on
the spinal cord, such as rhizotomy and cordotomy,
or peripheral procedures, such as sacral and hypo-
gastric neurotomy, may need to be considered.
Non-surgical approaches, such as subarachnoid
block with alcohol or phenol, have also been uti-
lised. These procedures, however, usually abolish
spinal reflex activity and result in flaccidity of
skeletal muscles, and bladder and bowel atony,
with their attendant disadvantages.
References
Arnold, J.M., Feng, Q.P., Delaney, G.A. and Teasell, R.W.
(1995) Autonomic dysreflexia in tetraplegic patients: evidence
for alpha-adrenoceptor hyper-responsiveness. Clin. Auton.
Res., 5: 267–270.
Baliga, R.R., Catz, A.B., Watson, L.P., Short, D.J., Frankel,
H.L. and Mathias, C.J. (1997) Cardiovascular and hormonal
242
responses to food ingestion in humans with spinal cord tran-
section. Clin. Auton. Res., 7: 137–141.
Barber, D.B., Rogers, S.J., Fredrickson, M.D. and Able, A.C.
(2000) Midodrine hydrochloride and the treatment of ortho-
static hypotension in tetraplegia: two cases and a review of
the literature. Spinal Cord, 38: 109–111.
Cariga, P., Ahmed, S., Mathias, C.J. and Gardner, B.P. (2002)
The prevalence and association of neck (coat-hanger) pain
and orthostatic (postural) hypotension in human spinal cord
injury. Spinal Cord, 40: 77–82.
Cariga, P. and Mathias, C.J. (2001) Haemodynamics of the
pressor effect of oral water in human sympathetic denerva-
tion due to autonomic failure. Clin. Sci., 101: 313–319.
Chaudhuri, K.R., Maule, S., Thomaides, T., Pavitt, D. and
Mathias, C.J. (1994) Alcohol ingestion lowers supine blood
pressure, causes splanchnic vasodilatation and worsens post-
ural hypotension in primary autonomic failure. J. Neurol.,
241: 145–152.
Corbett, J.L., Debarge, O., Frankel, H.L. and Mathias, C.J.
(1975) Cardiovascular responses in tetraplegic man to muscle
spasm, bladder percussion and head-up tilt. Clin. Exp.
Pharmacol. Physiol., 2(Suppl): 189–193.
Corbett, J.L., Frankel, H.L. and Harris, P.J. (1971) Cardio-
vascular reflex responses to cutaneous and visceral stimuli in
spinal man. J. Physiol., 215(2): 395–409.
Critchley, H.D. and Mathias, C.J. (2003) Blood pressure, at-
tention and cognition: drivers and air traffic controllers. Clin.
Auton. Res., 13: 399–401.
Curt, A., Nitsche, B., Rodic, B., Schurch, B. and Dietz, V.
(1997) Assessment of autonomic dysreflexia in patients with
spinal cord injury. J. Neurol. Neurosurg. Psychiat., 62:
473–477.
Frankel, H.L., Mathias, C.J. and Spalding, J.M.K. (1975)
Mechanisms of reflex cardiac arrest in tetraplegic patients.
Lancet, ii: 1183–1185.
Frankel, H.L., Michaelis, L.S., Golding, D.R. and Beral, V.
(1972) The blood pressure in paraplegia-1. Paraplegia, 10:
193–198.
Gao, S.A., Ambring, A., Lambert, G. and Karlsson, A.-K.
(2002) Autonomic control of the heart and renal vascular bed
during autonomic dysreflexia in high spinal cord injury. Clin.
Auton. Res., 12: 457–464.
Goadsby, P.J. (2002) Autoregulation and autonomic control of
the cerebral circulation: implications and pathophysiology.
In: Mathias C.J. and Bannister R. (Eds.), Autonomic failure.
A Textbook of Clinical Disorders of the Autonomic Nervous
System (4th ed). Oxford University Press, Oxford, pp. 85–91.
Guttmann, L. and Whitteridge, D. (1947) Effects of bladder
distension on autonomic mechanisms after spinal cord injury.
Brain, 70: 361–404.
Head, H. and Riddoch, G. (1917) The autonomic bladder, ex-
cessive sweating and some other reflex conditions in gross
injuries of the spinal cord. Brain, 40: 188–263.
Karlsson, A.-K., Friberg, P., Lonnroth, P., Sullivan, L. and
Elam, M. (1998) Regional sympathetic function in high
spinal cord injury during mental stress and autonomic
dysreflexia. Brain, 121: 1711–1719.
King, M.L., Freeman, D.M., Pellicone, J.T., Wanstall, E.R.
and Bhansali, L.D. (1992) Exertional hypotension in thoracic
spinal cord injury: case report. Paraplegia, 30: 261–266.
King, M.L., Lichtman, S.W., Pellicone, J.T., Close, R.J. and
Lisanti, P. (1994) Exertional hypotension in spinal cord in-
jury. Chest, 106: 1166–1171.
Kooner, J.S., da Costa, D.F., Frankel, H.L., Bannister, R.,
Peart, W.S. and Mathias, C.J. (1987) Recumbency induces
hypertension, diuresis and natriuresis in autonomic failure,
but diuresis alone in tetraplegia. J. Hypertens., 5: 327–329.
Krum, H., Louis, W.J., Brown, D.J., Clarke, S.J., Fleming, J.A.
and Howes, L.G. (1992) Cardiovascular and vasoactive hor-
mone responses to bladder distension in spinal and normal
man. Paraplegia, 30: 348–354.
Linsenmeyer, T.A., Campagnolo, D.I. and Chou, I.H. (1996)
Silent autonomic dysreflexia during voiding in men with spi-
nal cord injuries. J. Urol., 155: 519–522.
Maiorov, D.N., Weaver, L.C. and Krassioukov, A.V. (1997)
Relationship between sympathetic activity and arterial pres-
sure in conscious spinal rats. Am. J. Physiol., 272: H625–H631.
Mathias, C.J. (1976) Bradycardia and cardiac arrest during
tracheal suction — mechanisms in tetraplegic patients. Eur. J.
Intens. Care Med., 2: 147–156.
Mathias, C.J. (2003a) Autonomic diseases — clinical features
and laboratory evaluation. J. Neurol. Neurosurg. Psychiat.,
74: 31–41.
Mathias, C.J. (2003b) Autonomic diseases — management.
J. Neurol. Neurosurg. Psychiat., 74: 42–47.
Mathias, C.J. and Bannister, R. (2002) Postprandial hypoten-
sion in autonomic disorders. In: Mathias C.J. and Bannister
R. (Eds.), Autonomic Failure: A Textbook of Clinical Dis-
orders of the Autonomic Nervous System (4th ed.). Oxford
University Press, Oxford, pp. 283–295.
Mathias, C.J., Christensen, N.J., Corbett, J.L., Frankel, H.L.,
Goodwin, T.J. and Peart, W.S. (1975) Plasma catechola-
mines, plasma renin activity and plasma aldosterone in tet-
raplegic man, horizontal and tilted. Clin. Sci. Mol. Med., 49:
291–299.
Mathias, C.J., Christensen, N.J., Corbett, J.L., Frankel, H.L.
and Spalding, J.M.K. (1976a) Plasma catecholamines during
paroxysmal neurogenic hypertension in quadriplegic man.
Circ. Res., 39: 204–208.
Mathias, C.J., Christensen, N.J., Frankel, H.L. and Spalding,
J.M.K. (1979a) Cardiovascular control in recently injured
tetraplegics in spinal shock. Q. J. Med. NS, 48: 273–287.
Mathias, C.J., da Costa, D.F., Fosbraey, P., Bannister, R.,
Wood, S.M., Bloom, S.R. and Christensen, N.J. (1989) Car-
diovascular, biochemical and hormonal changes during food
induced hypotension in chronic autonomic failure. J. Neurol.
Sci., 94: 255–269.
Mathias, C.J., Fosbraey, P., da Costa, D.F., Thornley, A. and
Bannister, R. (1986) The effect of desmopressin on nocturnal
polyuria, overnight weight loss and morning postural hypo-
tension in patients with autonomic failure. Br. Med. J., 293:
353–354.
Mathias, C.J. and Frankel, H. (2002) Autonomic disturbances
in spinal cord lesions. In: Mathias C.J. and Bannister R.

(Eds.), Autonomic Failure: A Textbook of Clinical Disorders
of the Autonomic Nervous System (4th Ed.). Oxford Uni-
versity Press, Oxford, pp. 494–513.
Mathias, C.J. and Frankel, H.L. (1986) The neurological and
hormonal control of blood vessels and heart in spinal man.
J. Auton. Nerv. System.(Suppl): 457–464.
Mathias, C.J. and Frankel, H.L. (1988) Cardiovascular control
in spinal man. Ann. Rev. Physiol., 50: 577–592.
Mathias, C.J., Frankel, H.L., Christensen, N.J. and Spalding,
J.M.K. (1976b) Enhanced pressor response to noradrenaline
in patients with cervical spinal cord transaction. Brain, 99:
757–770.
Mathias, C.J., Frankel, H.L., Davies, I.B., James, V.H.T. and
Peart, W.S. (1981) Renin and aldosterone release during sym-
pathetic stimulation in tetraplegia. Clin. Sci., 60: 399–604.
Mathias, C.J., Holly, E., Armstrong, E., Shareef, M. and Ban-
nister, R. (1991) The influence of food on postural hypoten-
sion in three groups with chronic autonomic failure: clinical
and therapeutic implications. J. Neurol. Neurosurg. Psych-
iat., 54: 726–730.
Mathias, C.J., Mallipeddi, R. and Bleasdale-Barr, K. (1999)
Symptoms associated with orthostatic hypotension in pure
autonomic failure and multiple system atrophy. J. Neurol.,
246: 893–898.
Mathias, C.J., Reid, J.L., Wing, L.M.H., Frankel, H.L. and
Christensen, N.J. (1979b) Antihypertensive effects of cloni-
dine in tetraplegic subjects devoid of central sympathetic
control. Clin. Sci., 57: 425s–428s.
Mathias, C.J. and Young, T.M. (2004) Water drinking in the
management of orthostatic intolerance due to orthostatic
hypotension, vasovagal syncope and the postural tachycardia
syndrome. Eur. J. Neurol., 11: 613–619.
Middleton, J.W., Siddall, P.J., Walker, S., Molloy, A.R. and
Rutkowski, S.B. (1996) Intrathecal clonidine and baclofen in
the management of spasticity and neuropathic pain following
spinal cord injury: a case study. Arch. Phys. Med. Rehabil.,
77: 824–826.
Mukand, J., Karlin, L., Barrs, K. and Lublin, P. (2001) Mid-
odrine for the management of orthostatic hypotension in
patients with spinal cord injury: A case report. Arch. Phys.
Med. Rehabil., 82: 694–696.
Nitsche, B., Perschak, H., Curt, A. and Dietz, V. (1996) Loss of
circadian blood pressure variability in complete tetraplegia.
J. Hum. Hypertens., 10: 311–317.
243
Reid, J.L., Wing, L.M.H., Mathias, C.J., Frankel, H.L. and
Neill, E. (1977) The central hypotensive effect of clonidine:
studies in tetraplegic subjects. Clin. Pharmacol. Ther., 21:
375–381.
Sampson, E.E., Burnham, R.S. and Andrews, B.J. (2000) Func-
tional electrical stimulation effect on orthostatic hypotension
after spinal cord injury. Arch. Phys. Med. Rehabil., 81:
139–143.
Schatz, I.J., Bannister, R., Freeman, R.L., Jankovic, J., Koller,
W.C., Low, P.A., Mathias, C.J., Polinsky, R.J., Quinn, N.P.,
Robertson, D. and Streeten, D.H.P. (1996) Consensus state-
ment on the definition of orthostatic hypotension, pure au-
tonomic failure and multiple system atrophy. Clin. Auton.
Res., 6: 125–126.
Stjernberg, L., Blumberg, H. and Wallin, B.G. (1986) Sympa-
thetic activity in man after spinal cord injury: outflow to
muscle below the lesion. Brain, 109: 695–715.
Sutters, M., Wakefield, C., O’Neil, K., Appleyard, M., Frankel,
H.L., Mathias, C.J. and Peart, W.S. (1992) The cardiovas-
cular, endocrine and renal response to tetraplegic and par-
aplegic subjects to dietary sodium restriction. J. Physiol.
(London), 457: 515–523.
Tank, J., Schroeder, C., Stoffels, M., Diedrich, A., Sharma,
A.M., Luft, F.C. and Jordan, J. (2003) Pressor effect of water
drinking in tetraplegic patients may be a spinal reflex. Hy-
pertension, 41: 1234–1239.
Thyberg, M., Ertzgaard, P., Gylling, M. and Granerus, G.
(1994) Effect of nifedipine on cystometry-induced elevation
of blood pressure in patients with a reflex urinary bladder
after a high level spinal cord injury. Paraplegia, 32: 308–313.
Vaidyanathan, S., Soni, B.M., Sett, P., Watt, J.W., Oo, T. and
Bingley, J. (1998) Pathophysiology of autonomic dysreflexia:
long-term treatment with terazosin in adult and paediatric
spinal cord injury patients manifesting recurrent dysreflexic
episodes. Spinal Cord, 36: 761–770.
van Lieshout, J.J., lmholz, B.P.M., Wesseling, K.H., Speelman,
J.D. and Wieling, W. (1991) Singing-induced hypotension: a
complication of high spinal cord lesion. Neth. J. Med., 38:
75–79.
Weaver, L.C. (2002) What causes autonomic dysreflexia after
spinal cord injury? Clin. Auton. Res., 12: 424–426.
Welply, N., Mathias, C.J. and Frankel, H.L. (1975) Circulatory
reflexes in tetraplegics during artificial ventilation and during
general anaesthesia. Paraplegia, 13: 172–182.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 16

Autonomic dysreflexia after spinal cord injury:

central mechanisms and strategies for prevention

Lynne C. Weaver1,, Daniel R. Marsh2, Denis Gris1, Arthur Brown1 and Gregory
A. Dekaban1

1
Spinal Cord Injury Team, BioTherapeutics Research Group, Robarts Research Institute, 100 Perth Drive, P.O. Box 5015,
London, ON N6A 5K8, Canada
2
Department of Anatomy and Cell Biology, Dalhousie University, Halifax, NS, Canada

Abstract: Spinal reflexes dominate cardiovascular control after spinal cord injury (SCI). These reflexes are
no longer restrained by descending control and they can be impacted by degenerative and plastic changes
within the injured cord. Autonomic dysreflexia is a condition of episodic hypertension that stems from
spinal reflexes initiated by sensory input entering the spinal cord caudal to the site of injury. This hy-
pertension greatly detracts from the quality of life for people with cord injury and can be life-threatening.
Changes in the spinal cord contribute substantially to the development of this condition. Rodent models
are ideal for investigating these changes. Within the spinal cord, injury-induced plasticity leads to nerve
growth factor (NGF)-dependent enlargement of the central arbor of a sub-population of sensory neurons.
This enlarged arbor can provide increased afferent input to the spinal reflex, intensifying autonomic
dysreflexia. Treatments such as antibodies against NGF can limit this afferent sprouting, and diminish the
magnitude of dysreflexia. To assess treatments, a compression model of SCI that leads to progressive
secondary damage, and also to some white matter sparing, is very useful. The types of spinal reflexes that
likely mediate autonomic dysreflexia are highly susceptible to inhibitory influences of bulbospinal pathways
traversing the white matter. Compression models of cord injury reveal that treatments that spare white
matter axons also markedly reduce autonomic dysreflexia. One such treatment is an antibody to the
integrin CD11d expressed by inflammatory leukocytes that enter the cord acutely after injury and cause
significant secondary damage. This antibody blocks integrin-mediated leukocyte entry, resulting in greatly
reduced white-matter damage and decreased autonomic dysreflexia after cord injury. Understanding the
mechanisms for autonomic dysreflexia will provide us with strategies for treatments that, if given early after
cord injury, can prevent this serious disorder from developing.

Introduction to autonomic dysreflexia entirely devoted to autonomic dysfunction after

In 2001, my co-authors and I published a chapter
reviewing the central mechanisms for autonomic
dysreflexia in a volume of Progress in Brain Re-
search that focused on spinal cord injury (SCI)
(Weaver et al., 2001a). The present volume,
Corresponding author. Tel.: +519 663 3776;
Fax: +519 663 3789; E-mail: lcweaver@robarts.ca
cord injury, contains many chapters written by
investigators quoted in that 2001 review. As the
work of others is presented elsewhere in this book,
this chapter will focus on the 10 years of investi-
gations of autonomic dysreflexia from our labo-
ratory that began in 1995.
After SCI, large increases in arterial pressure
(AP) can develop in response to sensory input en-
tering the spinal cord below the level of the lesion

DOI: 10.1016/S0079-6123(05)52016-8 245

246
(Mathias and Frankel, 1992a, b; Lee et al., 1994;
Krassioukov and Weaver, 1995; Maiorov et al.,
1998). This hypertension, part of a condition
termed autonomic dysreflexia, occurs in 50–90%
of people with tetraplegia or high paraplegia
(Corbett et al., 1975; Erickson, 1980; Lindan et
al., 1980; Mathias and Frankel, 1992a; Lee et al.,
1994; Giannantoni et al., 1998; Vaidyanathan
et al., 1998; Karlsson, 1999; Krassioukov et al.,
2003). The hypertension can result in debilitating
headaches, seizures, strokes and even death. As
blood pressure control normally depends upon
supraspinal regulation of sympathetic neurons
(Calaresu and Yardley, 1988), SCI leads to a con-
dition in which central nervous system (CNS)
regulation of AP is dominated by excitatory spinal
reflexes. This excitation is prominent, in part,
because cord injury blocks descending bulbospinal
inhibition of spinal sympathetic reflexes (Sato and
Schmidt, 1973; Dembowsky et al., 1978, 1980); the
unchecked activity of these reflexes leads to
autonomic dysreflexia.
Dysreflexia occurs after injury at or rostral to
the 6th thoracic spinal segment, because injury at
these levels leaves the sympathetic control of the
extensive abdominal circulation amenable to un-
restrained spinal reflexes (Mathias and Frankel,
1992a). These reflexes can be caused by innocuous
or noxious stimulation of the skin, pressure sores,
distension or inflammation of the urinary bladder
or gastrointestinal tract and also by muscle spasms
that often develop after SCI (Guttman and
Whitteridge, 1947; Corbett et al., 1975; Mathias
and Frankel, 1992a; Blackmer, 2003). They can be
initiated by routine daily procedures such as blad-
der catheterization and bowel evacuation. Dysre-
flexia is not always severe and may include only
sweating, flushed skin above the segment of injury,
piloerection and small increases in AP. However,
this condition can become uncontrolled, leading to
life-threatening hypertension (Shea et al., 1973;
Naftchi, 1990; Mathias and Frankel, 1992c; Lee
et al., 1994; Naftchi and Richardson, 1997;
Giannantoni et al., 1998; Blackmer, 2003). Mech-
anisms for autonomic dysreflexia may involve af-
ferent, interneuronal and efferent components of
the spinal sympathetic reflex and the vasculature,
as also discussed in other chapters of this book.
This chapter will review studies in rats that
demonstrated some of the changes within the
spinal cord that can contribute to the development
of autonomic dysreflexia, and experiments that
show the potential for treatments to prevent the
development of dysreflexia.
Rodent models of autonomic dysreflexia
Rats and mice readily develop autonomic dysre-
flexia after SCI (Osborn et al., 1990; Maiorov
et al., 1997a, b, 1998, 2001; Landrum et al., 1998;
Jacob et al., 2001, 2003). In response to distension
of the urinary bladder or colon, exaggerated in-
creases in AP are apparent within a day of a cord
transection at the upper thoracic (T) segments
(T2–4) and these responses augment with time af-
ter injury. This progression after cord transection
in rats was followed for up to 5 weeks in our
laboratory. A conscious, unrestrained rat with an
intact CNS has a transient, 5–15 mmHg increase in
mean arterial pressure (MAP) upon colon disten-
sion, accompanied by an increase in heart rate
(HR) (Krassioukov and Weaver, 1995; Maiorov
et al., 1997b). In this chapter, changes in blood
pressure will be expressed as changes in MAP,
unless otherwise indicated. The first day after
cord transection, the increase in AP, in response
to visceral stimulation, can be as great as
4172 mmHg, but within the next few days, these
responses become smaller, perhaps associated with
degenerative changes in the spinal sympathetic
preganglionic neurons caudal to the injury (see the
following sections) (Krassioukov and Weaver,
1995; Maiorov et al., 1997a, b). Then, in the en-
suing weeks, the increases in AP during dysreflexia
become larger, reaching values as high as
5274 mmHg by 5 weeks after SCI. The large re-
sponse on the first day likely reflects the full ca-
pacity of the spinal reflex, without descending
inhibitory restraint (Dembowsky et al., 1980), and
before the secondary effects of cord injury have
impacted greatly. After this time, the magnitude of
the hypertension appears to become a function of
many changes ongoing in the spinal cord and
vasculature, some degenerative and others, plastic
responses to the injury. These intraspinal changes,
 247

and their relationship to this time course of auto-

nomic dysreflexia will be discussed later in this
chapter.
Autonomic dysreflexia is readily apparent in
rats when SCI is caused by a method that repro-
duces the typical clinical injury. This method en-
tails brief (60 s) extradural compression of the cord
with a calibrated, modified aneurysm clip. The
clip-compression model mimics the key patho-
physiological features of human SCI. This model
produces mechanical injury (primary injury) and
secondary damage by a variety of well-character-
ized mechanisms including microvasculature dis-
ruption, hemorrhage, ischemia, increases in
intracellular calcium, calpain activation, progres-
sive axonal injury and glutamate toxicity (Wallace
and Tator, 1986; Wallace et al., 1986; Fehlings
et al., 1989; Koyanagi et al., 1993a, b; Agrawal
and Fehlings, 1996, 1997a, b; Agrawal et al., 1998;
Schumacher et al., 1999). Marked autonomic
dysreflexia is readily evoked after severe (50 g)
clip-compression cord injury at T2–4 in rats and
mice (Maiorov et al., 1998; Mayorov et al., 2001;
Weaver et al., 2001b; Jacob et al., 2003) (Fig. 1A).
Distension of the colon with a small balloon-
tipped catheter or cutaeous stimulation in con-
scious rats at two weeks after injury causes pressor
responses that continue to increase in magnitude
until 6 weeks after cord injury, when they appear
to plateau (Gris et al., 2004; Marsh and Weaver,
2004). Dysreflexia after clip-compression injury
can be evoked by visceral or cutaneous stimuli that
are noxious or non-noxious. Gently stroking the
skin of the back caudal to the segment of injury
causes pressor responses of 2375 mmHg, whereas
Fig. 1. The MAP, pulsatile AP and HR in two rats, 2 weeks
pinching the skin increases pressure by up to after clip-compression SCI: after intrathecal treatment with
3272 mmHg. Likewise, distension of the colon to control immunoglobulin G (IgG) (A) and after intrathecal
an extent that would be noxious in the intact an- treatment with trkA-IgG (B). Colon distension for 60 s (onset
imal increases AP by 4274 mmHg, whereas a and duration marked with a thick line) stimulated a large in-
crease in AP and MAP and a decrease in HR in the control
milder distension, that would not be noxious, still
IgG-treated rat. In contrast, colon distension caused only a
increases AP by 2473 mmHg. In a conscious rat modest decrease in AP and MAP after treatment with trkA-
with an intact CNS, this non-noxious cutaneous or IgG. Reprinted with permission from Marsh et al. (2002).
visceral stimulation would cause either no change
in AP or a transient (5–10 s) increase of abnormal, or dysreflexic responses, as they are
5–10 mmHg. These findings refute the idea that more than twice the magnitude of those in an
autonomic dysreflexia is initiated only by noxious intact rat.
stimuli. The responses to the non-noxious stimuli The large increase in noradrenaline spill-over re-
are smaller than those to noxious stimuli, but are ported during dysreflexia is consistent with intense
248
sympathetic firing relative to baseline values
(Karlsson et al., 1998). Direct measurements of
visceral vasomotor sympathetic firing can be made
in animals. The renal nerve has mostly vasomotor
sympathetic axons, and renal sympathetic firing
recorded in conscious rats, after cord transection
at T4, increases massively during episodes of au-
tonomic dysreflexia (Maiorov et al., 1998). The
increases are at least sixfold above baseline firing
and, during the first week after cord transection,
they parallel the increase in amplitude of the blood
pressure responses. Recordings at later times have
not been reported. These spinal reflexes are much
larger than they would be if the nervous system
were intact (Coote et al., 1969; Koizumi and
Brooks, 1972; Dembowsky et al., 1980) and reflect
loss of descending inhibition and, with time, plas-
tic changes within the spinal cord.
Basal firing of the renal sympathetic nerve is
greatly reduced by cord transection so that the
nerve targets are possibly approaching the state of
‘‘denervation’’ postulated by McLachlan and
Brock (see this volume and Yeoh et al., 2004).
The low firing rate certainly could prompt signif-
icant changes in the peripheral sympathetic system
and vascular targets. However, even the low basal
sympathetic firing in the first week after T4 tran-
section appears to make a contribution to vaso-
motor control, as ganglionic blockade decreases
resting AP by 4577 mmHg (Maiorov et al.,
1997b). This lowered sympathetic activity becomes
the norm after cord injury, and, if this activity is
greatly increased by a spinal reflex, coupled with a
more reactive vasculature and changes in periph-
eral neurotransmission, large hypertensive re-
sponses can occur.
Spinal cord transection or compression also has
significant initial effects on resting AP, decreasing
it by 2374 mmHg during the first 24 h after cord
injury (Maiorov et al., 1997b). This decrease is
transient and by 1 week after injury AP returns to
normal values (10173 mmHg). This range of AP
is stable for at least 6 weeks (Krassioukov and
Weaver, 1995; Maiorov et al., 1997a; Krenz et al.,
1999; Weaver et al., 2001b; Marsh et al., 2002; Gris
et al., 2004; Marsh and Weaver, 2004). In contrast,
HR initially is not altered by cord transection or
clip-compression at T4, but gradually increases
with time after injury, from
400 to 450 beats/min
in intact rats or cord-injured rats up to a week
after injury, to rates of
500–570 beats/min at 2–6
weeks after injury. In one study, we attributed this
tachycardia to increased afferent input to upper
thoracic preganglionic neurons, resulting from
sprouting of primary afferent fibers in the dorsal
horn (Krenz et al., 1999). This input, if tonically
active, could contribute to enhanced sympathetic
drive to the heart. Others have also reported in-
creases in HR after cord transection at T4
(Rodenbaugh et al., 2003), and have suggested
several mechanisms for this tachycardia (see
Collins et al., this volume).
Reorganization of the injured spinal cord
Caudal to the SCI, the loss of supraspinal input
may make the sympathetic preganglionic neuron
more sensitive to spinal afferent input, not only
because of the loss of tonically active inhibitory
supraspinal influences, but also because, in the
absence of descending excitatory synaptic input,
the remaining excitatory synapses within spinal
reflex circuits have a more dominant influence. As
reviewed by Llewellyn-Smith et al. (this volume),
glutamatergic inputs to preganglionic neurons are
largely lost within a few days of cord transection
(Llewellyn-Smith and Weaver, 2001), yet, at the
same time, the neurons respond vigorously to
reflex inputs, suggesting that a few dominant
synapses can activate the neurons. Reorganization
of the spinal pathways controlling sympathetic
preganglionic neurons after the loss of supraspinal
input is likely to change the makeup of inputs to
these neurons (Krassioukov and Weaver, 1996;
Cassam et al., 1997, 1999; Weaver et al., 2001b).
We searched for changes in glutamatergic spinal
reflex transmission via N-methyl-D-aspartate
(NMDA) and a-amino-3-hydroxy-5-methyl-4-
isoxazolpropionic acid (AMPA) receptors during
the first 2 weeks after spinal cord transection
(Maiorov et al., 1997a). Using selective receptor
antagonists, these two receptors appeared to con-
tribute about equally (
35% each) to the reflex
excitation causing autonomic dysreflexia, at 2 and
16 days after the injury, demonstrating no change

with time in the roles of these receptors during the
initial period after injury.
The enhanced transmission through the spinal
reflex circuits suggested by the recording of renal
nerve responses (Maiorov et al., 1997b) could be
mediated by changes in the afferent, interneuronal
or preganglionic neuronal components of the arc.
Earlier studies lead to the hypothesis that synaptic
input to preganglionic neurons undergoes plastic
reorganization after cord injury, resulting in
replacement of inputs from bulbospinal neurons
with those from interneurons (Krassioukov and
Weaver, 1996; Weaver et al., 1997). Thoracic
preganglionic neurons undergo a transient degen-
erative response, characterized by retraction of
dendrites and shrinkage of the soma, during the
first week after cord transection (the period of
synaptic loss), and then re-establish a normal
dendritic arbor and soma size within 2–4 weeks
after the injury (Krassioukov and Weaver, 1996;
Krenz and Weaver, 1998a; Llewellyn-Smith and
Weaver, 2001). This acute degenerative response
of the preganglionic neurons with later morpho-
logical recovery was also found in samples of in-
jured human spinal cord (Krassioukov et al.,
1999). These observations led us to propose a
model in which synapses from supraspinal neurons
were replaced with those from spinal circuits,
leading to exaggerated spinal reflex control of
preganglionic neurons. Immediately after cord
injury, the preganglionics are intact and descend-
ing inhibition is lost, leading to robust spinal
reflex-mediated pressor responses. Then, pregangl-
ionic neurons undergo a degenerative response
that is maximal at about 1 week, the time when
autonomic dysreflexia is decreased in magnitude.
Finally, as the preganglionic neurons recover from
the atrophy and have a normal soma size and
dendritic tree, with potentially new synaptic inputs
from interneurons, autonomic dysreflexia begins
to increase in magnitude. Although this was an
appealing model, an electron microscopic study
(Llewellyn-Smith and Weaver, 2001) demonstrated
that the loss of bulbospinal synaptic input to
thoracic preganglionic neurons reduces the num-
ber of synapses on these neurons by 50–70%, and
this loss is not replaced by intraspinal inputs.
Accordingly, the number of synapses from the
249
interneurons after cord injury does not appear to
increase and, therefore, an increased number of
synapses from interneurons to preganglionic neu-
rons does not mediate the time-dependent increas-
es in the spinal sympathetic reflexes and pressor
responses. The interneurons may still contribute if
they are more excited by an afferent input to the
spinal cord, and provide increased preganglionic
excitation through temporal rather than spatial
summation. Consistent with this view, an electro-
physiological study of responses of sympathetical-
ly correlated spinal interneurons to cutaneous
inputs, after spinal cord transection, showed that
the receptive fields of individual interneurons to
this input enlarge with time after spinal cord
transection, increasing their capacity to provide
excitatory input to the preganglionic neurons
(Krassioukov et al., 2002) (see also Schramm, this
volume). Moreover, the gray matter containing
thoracic preganglionic neurons, and interneurons
providing input to them, develops a network of
fibers immunoreactive to growth-associated
protein-43 that increases in density between 2
and 4 weeks after cord transection (Weaver et al.,
1997). This protein was also present in putative
growth cones and in cell bodies that likely were
interneurons. These data suggest that, although
the thoracic preganglionic neurons themselves did
not receive new synaptic inputs, growth within the
local neuropil likely generated new inputs to other
neurons antecedent to the preganglionics.
Changes in the primary afferent arbor contributing
to autonomic dysreflexia
In the dorsal horn, the arbors of small-diameter
primary afferent neurons can enlarge greatly in
rats and mice after SCI (Krenz and Weaver,
1998b; Krenz et al., 1999; Wong et al., 2000; Jacob
et al., 2001, 2003; Weaver et al., 2001b) (Fig. 2)
potentially leading to increased reflex excitation of
preganglionic neurons, via interneuronal path-
ways. These afferent neurons that are calcitonin
gene-related peptide (CGRP)-immunoreactive,
and the interneurons in laminae III–VII activated
by these afferent projections, make up the path-
way, mediating spinal reflex excitation of
250

Fig. 2. Digital photomicrographs of CGRP-immunoreactive (left, red) and substance P-immunoreactive (right, green) afferent fibers in
the dorsal horn of sham-injured rats (A and B) and of rats with clip-compression SCI at T4 (C-H). Examples are shown at 2 weeks after
sham or SCI. Tissue sections (30 mm) from T12 to L1 were processed for double immunofluorescence for CGRP (red) and substance P
(green) and visualized with a 10  objective (A–D, calibration bar ¼ 100 mm), a 20  objective (E–G, calibration bar ¼ 50 mm), or with
confocal microscopy (H, calibration bar ¼ 10 mm). Co-localization of CGRP and substance P in afferent fibers is indicated by yellow.
Reprinted with permission from Marsh and Weaver (2004).

preganglionic neurons (Sato and Schmidt, 1973;
Cabot et al., 1994; Joshi et al., 1995; Clarke et al.,
1998). After cord transection or compression in
rats and some strains of mice, the 2-week time
course for the increased arbor size correlates with
the gradual increase in the magnitude of auto-
nomic dysreflexia (described earlier) and may also
be a factor contributing to the increase. After
transection injury, the size of the increase in the
CGRP-immunoreactive afferent arbor correlates
with the magnitude of dysreflexia (Krenz et al.,
1999; Jacob et al., 2001, 2003). This afferent arbor
remains enlarged at 1 month after cord injury, a
time when autonomic dysreflexia is well developed
in rats (Krassioukov and Weaver, 1995). The
‘‘enlarged’’ CGRP-immunoreactive afferent arbor
might reflect only increased CGRP expression in
the fibers rather than a sprouting or growth
response. However, a confocal microscopic
analysis of changes in the afferent arbor after T4
cord transection, using retrograde transport of
wheat germ agglutinin, revealed a true increase in
the size of the arbor of small diameter fibers in the
lumbar spinal cord (Wong et al., 2000). Likewise,
transport of the b subunit of cholera toxin after
cord transection also revealed expansion of the
arbor of larger afferent fibers in the lumbar cord
(Krenz and Weaver, 1998b). Together, these find-
ings demonstrate that the increase in CGRP after
cord injury reflects a growth response, not simply an
increase in gene expression. The changes in the
larger diameter fibers probably underlie the re-
sponses to non-noxious cutaneous and visceral stim-
ulation described above (Marsh and Weaver, 2004).
The relationship between changes in the afferent
arbor and the magnitude of dysreflexia is less ro-
bust in the clip-compression cord-injury model
that spares some descending pathways. Enlarge-
ment of the CGRP-immunoreactive primary af-
ferent arbor is also characteristic of this model, but
changes are smaller than after cord transection,
whereas the magnitude of dysreflexia can also be
as great (Krenz and Weaver, 1998b; Krenz et al.,
1999; Weaver et al., 2001b). Injury to the cord
must be severe for enlargement of the primary af-
ferent arbor and autonomic dysreflexia to occur.
At 2 weeks after moderate/mild (20 g) clip-com-
pression cord injury at T4, rats have no autonomic
251
dysreflexia in response to colon distension and no
change in the CGRP-immunoreactive afferent
arbor. In contrast, after a more severe (50 g) inju-
ry, autonomic dysreflexia develops consistently
and the CGRP-immunoreactive arbor in the lower
thoracic and lumbar spinal cord enlarges. The dif-
ference between the transection and compression
injury model is the sparing of a limited number of
white-matter axons that pass through the lesion
site. If the compression is only moderate, and
many axons are spared, sufficient bulbospinal in-
put remains to prevent the development of
dysreflexia and afferent sprouting. However, after
severe compression injury, the small sparing of
descending fibers is not adequate to block dysre-
flexia and the sprouting response does develop,
albeit modestly.
The small diameter C-type sensory neurons in
the dorsal root ganglia contain a population that
express CGRP, and a subset of neurons within this
population also express substance P (Price, 1985;
Ju et al., 1987). In addition to the small-diameter
neurons, CGRP is also expressed in some of the
larger sensory neurons (McCarthy and Lawson,
1990; Lawson et al., 1993, 1996). The plasticity in
the CGRP-immunoreactive afferent arbor initiat-
ed by cord injury appears to target only the pop-
ulation that expresses CGRP without substance P
(Marsh and Weaver, 2004) (Fig. 2). The absence of
sprouting of substance P-containing afferent fibers
in the dorsal horn, in the presence of increased
densities of CGRP-containing fibers, shows the
selectivity of this response, suggesting that the
larger fibers play a greater role in dysreflexia.
Finally, as also discussed by Rabchevsky (this vol-
ume), increases or decreases in the afferent arbor
are only relevant to inputs entering the spinal cord
at the same segmental region. For example, be-
cause of the segmental arrangement of the primary
afferent projections to the spinal cord (Vizzard
et al., 2000), changes in the lumbosacral arbor will
impact on responses to colon or bladder distension
but not on those to cutaneous stimulation of the
mid-thoracic trunk. Likewise, changes in the mid-
thoracic region do not impact on responses to co-
lon distension (Cameron et al., 2003; Gris et al.,
2005). In our studies, the afferent arbor was par-
ticularly increased in the lower lumbar segments
252
after cord transection or compression (Krenz and
Weaver, 1998b; Krenz et al., 1999; Jacob et al.,
2001, 2003; Weaver et al., 2001b), making afferent
plasticity more of an issue to responses initiated
from the pelvic organs or hind limbs.
Enlargement of the afferent arbor depends upon
actions of nerve growth factor (NGF) in the in-
jured rat spinal cord. Intrathecal delivery of a
highly selective neutralizing antibody to NGF for
2 weeks after cord transection at T4 completely
blocked the sprouting of the small-diameter affer-
ent fibers in the dorsal horn (Krenz et al., 1999).
Delivery of the antibody subcutaneously to pe-
ripheral targets of the sensory neurons had no ef-
fect, demonstrating that the afferent sprouting was
not caused by a target-derived source of NGF.
Blockade of the sprouting response was due to the
neutralizing effects of the anti-NGF antibody on
the trkA-expressing central arbors of these sensory
neurons. Autonomic dysreflexia was measured in
the same rats and the increases in AP caused by
visceral stimulation were reduced by 43% in the
(intrathecal) antibody-treated rats. This reduced
the dysreflexia to the magnitude of the spinal reflex
that is present in the conscious rat within 48 h of
cord transection, before enlargement of the affer-
ent arbor could occur (Maiorov et al., 1997a).
Similarly, intrathecal delivery, for 2 weeks after
clip-compression injury at T4, of a trkA-IgG fu-
sion protein that sequesters NGF decreased
dysreflexia in rats by 30% (Marsh et al., 2002)
(Fig. 1B). These data led to two conclusions. First,
afferent sprouting in the spinal cord dorsal horn is
clearly associated with the time-dependent increase
in hypertensive responses to sensory stimulation,
characteristic of autonomic dysreflexia. Second,
the sprouting is caused by an intraspinal action of
NGF, presumably at the central arbors of the sen-
sory neurons containing CGRP.
These studies suggest that NGF is a culprit
in the development of autonomic dysreflexia.
Normally, the spinal cord contains very little
NGF but NGF levels within a few segments of a
cord injury site have been reported to increase to a
peak at 1 week post injury, remaining increased for
4 weeks (Bakhit et al., 1991). Using a two-site
enzyme-linked immunosorbant assay (ELISA) to
analyze spinal cord homogenates from rats with a
T4 SCI, we have confirmed that NGF levels in the
injured rat spinal cord are significantly greater in
and adjacent to the injury (T3–5) than rostral to
the injury (T1 and 2) or in the lumbar cord. NGF
at the injury site was approximately twofold great-
er than the content in this region in un-injured rats
(1.570.8 pg/mg). Immunocytochemistry on spinal
cord sections revealed increased NGF-immunore-
activity in cells and fibers in the dorsal root
entry zone, and in astrocytes, microglia and
leptomeningeal cells of cord-injured rats (Krenz
and Weaver, 2000). These were prevalent in
segments T3–8.
A second study of NGF in the injured spinal
cord extended these observations further by
searching for cells that produced NGF, i.e., con-
tained messenger RNA (mRNA) for NGF. In ad-
dition, double-labeling immunocytochemistry was
used to identify, with greater certainty, the cells
containing this protein (Brown et al., 2004). We
also compared the NGF profiles in spinal cords
with transection or compression injuries. In intact
spinal cords, NGF mRNA was in leptomeningeal
cells and neurons of the intermediate gray matter,
whereas NGF protein was only in the leptome-
ninges (Fig. 3). At 3–7 days after transection or
clip-compression injury NGF mRNA and protein
were expressed in the lesion and throughout the
intermediate gray matter and the white matter
rostral and caudal to the injury site. The NGF
mRNA and protein in spinal cord sections adja-
cent to each other were expressed in ramified mi-
croglia, astrocytes, intermediate gray neurons, pial
cells, leptomeningeal cells and Schwann cells in the
lateral white matter and lesion site (Fig. 4).
Rounded macrophages in the lesion contained
immunoreactivity for NGF, but the cells express-
ing NGF mRNA were not macrophages, and did
not appear to contain the protein. These findings
demonstrated that a variety of cells within the
injured spinal cord could produce NGF, but that
macrophages are not among them. Macrophages
may have contained NGF because they phagocy-
tosed it.
The studies described above show a strong
relationship between NGF and changes in the
afferent arbor that can contribute to dysreflexia.
The effectiveness of the anti-NGF antibody and
 253

Fig. 3. The NGF mRNA expression in the injured and un-injured (control) spinal cord. Hybridization signals visualized by alkaline
phosphatase are illustrated in longitudinal sections of the 5th thoracic spinal cord segment (T5). The lateral edge of the spinal cord is at
the top of each photomicrograph. In the un-injured control rats, NGF mRNA was intense in leptomeningeal cells at the edge of the un-
injured spinal cord and in gray-matter cells with morphology like neurons (A–C). Processing serial sections to detect either NGF
mRNA (C) or NeuN-immunoreactivity, confirmed that these gray-matter cells were neurons (D). The leptomeningeal cells at the edge
of the un-injured spinal cord had hybridization signal but none was present in white matter cells (B). NGF mRNA expression increased
after T4 transection-SCI (E–G). At one day, little NGF mRNA was present with the exception of patches of signal in the le-
ptomeningeal cells of the pia mater (E). At 3 days, expression of NGF mRNA was robust in white and gray matter and in the
leptomeninges, especially in segments close to the lesion (F). The NGF mRNA signals were still present but less abundant at 7 days
after SCI (G). Panels H–J show cells expressing NGF mRNA 7 days after SCI (at higher magnification). Within the lesion site, cells
were typically round and clustered (H), whereas those in the white matter were long and spindle shaped (I) and gray-matter cells
expressing NGF mRNA were oval and neuron-like in morphology or small and spindle-shaped (J); d, day; wm, white matter; gm, gray
matter; con, control. The scale bars on A and B are 50 mm; the bar on C also refers to D and is 50 mm; the bar for E–G, shown in G is
250 mm; the bar for H–J, shown in J is 50 mm. Reprinted from Brown et al. (2004) with permission from Elsevier.

the trkA-IgG fusion protein in limiting dysreflexia
shows that this strategy might be useful clinically
to prevent the development of dysreflexia. The
stimulus for the increased intraspinal NGF is now
in question. The inflammatory response to trau-
matic injury of the spinal cord is likely to promote
the production of NGF. The inflammatory
response to cord injury begins within minutes
and evolves for days, spreading throughout the
damaged cord and into adjacent, non-injured
regions (Blight, 1985; Tator and Fehlings, 1991;
Young, 1993; Popovich et al., 1997; Taoka and
Okajima, 1998). First, pro-inflammatory chemo-
kines and cytokines are released into the injured
254

Fig. 4. Identification of cells immunoreactive for the astrocytes marker glial fibrillary acidic protein (GFAP) and NGF in longitudinal
sections of the T5 spinal cord. Cells in the white matter are shown at 7 days after T4 transection-SCI (transect, A–C) or T4 clip-
compression SCI (clip, D–F). Immunoreactivity in the lesion is shown at 7 days after clip-SCI (G–I). Immunoreactivity for NGF (A, D
and G) and for GFAP (B, E and H) colocalized in the astrocytic glia limitans at the pial border of the white matter (C, arrow), and in
white-matter astrocytes (C, and F arrowheads). As few GFAP-immunoreactive astrocytes are in the lesion center, only small areas of
colocalization were present in the lesion (I, arrowhead). Scale bar in I refers to all panels and is 50 mm. Reprinted from Brown et al.
(2004) with permission from Elsevier.

area by astrocytes, microglia and endothelial cells.
Next, activation and proliferation of glia (gliosis)
occur, and neutrophils and monocyte/macrophages
enter the spinal cord from the circulation (Popovich
et al., 1997; Taoka et al., 1997; Leskovar et al.,
2000). Details of the inflammation in the injured
spinal cord will be presented below. Inflammation
and NGF expression and/or signaling are clearly
related. Interleukin-6 (IL-6) and IL-1b, two
cytokines found in the injured cord (Streit et al.,
1998; Hayashi et al., 2000), have well-documented
actions to increase the expression of NGF in the
CNS (Bandtlow et al., 1990; Spranger et al., 1990;
Saporito et al., 1993) and in Schwann cells of

peripheral nerves (Heumann et al., 1987a, b;
Lindholm et al., 1987). In response to cord
injury, microglial NGF mRNA transcription and
NGF protein release (Heese et al., 1998a, b) are
stimulated synergistically by IL-1 and tumor ne-
crosis factor (TNF)-a. The central processes of
sensory neurons have IL-6 receptors that would be
exposed to IL-6 in the injured cord, leading to
stimulation of NGF expression.
The NGF also has pro-inflammatory actions
(LaSala et al., 2000). The high-affinity trkA re-
ceptor is expressed on monocytes, B-lymphocytes
and T-lymphocytes. Although the signaling caused
by NGF on these receptors is not fully understood,
NGF promotes expression of Bcl-2 in monocytes
and mast cells, conferring resistance to cell death
(Bullock and Johnson, 1996; LaSala et al., 2000;
Saragovi and Gehring, 2000). The NGF also can
facilitate inflammation by promoting differentia-
tion of myeloid progenitor cells, inducing prolif-
eration and maturation of B-lymphocytes and
stimulating the release of inflammatory cytokines
such as TNF-a from basophils and mast cells.
In summary, inflammation can promote NGF
expression and NGF can promote the process of
inflammation. Clearly, the relationship between
inflammation and NGF could contribute to the
NGF-dependent increases in primary afferent
arbors in the spinal cord and the development of
autonomic dysreflexia.
Inflammation, secondary damage and autonomic
dysreflexia
Whereas blocking inflammation after SCI may
limit the production of NGF, it also has a host of
neuroprotective actions that make this issue a high
priority in devising strategies to prevent the devel-
opment of autonomic dysreflexia as well as the
other autonomic, sensory and motor dysfunctions
after cord injury. Secondary damage after an SCI
is caused, in part, by ischemia, cellular and tissue
edema, amino acid excitotoxicity and oxidative
damage. The extent of this damage directly corre-
lates with the magnitude of autonomic dysreflexia
(Weaver et al., 2001b). These changes often relate to
early inflammation and lead to myelin degradation
255
and intensified necrosis and apoptosis of neurons
and glia, events that contribute to increases in
lesion size (Taoka and Okajima, 1998; Bethea and
Dietrich, 2002). Neutrophils and hematogenous
macrophages invade the spinal cord, peaking at
12 h and 5–7 days after SCI, respectively
(Popovich et al., 1997; Taoka et al., 1997;
Leskovar et al., 2000), and releasing pro-inflam-
matory cytokines and reactive oxygen and nitro-
gen species (Taoka and Okajima, 1998; Popovich
et al., 1999). These cytotoxic substances can fur-
ther damage the white and gray matter and
contribute to scar formation (Jones et al., 2002).
The progression of secondary damage deterio-
rates neurological function, and renders the spinal
cord less amenable to repair strategies such as cell
replacement or therapeutic gene transfer. There-
fore, preventing this progression is of key impor-
tance. No practical way to control the early
inflammatory response has yet been devised,
despite our understanding of its destructive role
in the final outcome of an injury. We postulated
that a successful anti-inflammatory treatment
must selectively disrupt the early, destructive,
leukocyte-mediated actions in the injured cord,
while leaving an opportunity for later regenerative
interventions and wound-healing responses. After
SCI, intraspinal leukocyte infiltration first requires
leukocyte tethering, by selectins, to the surface of
endothelial cells (Bevilacqua, 1993). This is
followed by the interaction of endothelial cell-
adhesion molecules with integrins on the
leukocyte surface (Neish et al., 1995; Shanley
et al., 1998), facilitating leukocyte extravasation
through the blood-CNS barrier. To prevent this
interaction, we used a monoclonal antibody
(mAb) to the CD11d subunit of the CD11d/
CD18 integrin (Grayson et al., 1999; Van der
Vieren et al., 1999) (ICOS Corporation, Bothell,
WA). This mAb treatment substantially decreases
the numbers of neutrophils and macrophages at
the lesion at 3 days after SCI (Mabon et al., 2000;
Saville et al., 2004). It also markedly reduces oxi-
dative damage to the injured spinal cord (Bao
et al., 2004a, b). The anti-CD11d mAb was inject-
ed intravenously in three consecutive doses at 2, 24
and 48 h after clip-compression cord injury at T4,
and effects were assessed on autonomic dysreflexia
256

and on the spinal cord lesion at 2 and 6 weeks after
the injury (Gris et al., 2004) (Figs. 5 and 6).
At 2 weeks after cord injury, colon distension
increased AP and decreased HR for the duration
of stimulus (Fig. 5). The anti-CD11d mAb treat-
ment markedly reduced these increases in AP.
Mean increases during the 1 min stimulation peri-
od were reduced significantly from
30 mmHg in
the control rats to
20 mmHg in the treated rats.
As the normal reflex response to this stimulus in an
intact rat is an increase of
15 mmHg in AP
(Maiorov et al., 1997b), this treatment affected
much of the reflex increment attributable to cord
injury. The average decreases in HR during the
1 min colon distension in control rats (
65 beats/
min) tended to be greater than in treated rats (
45
beats/min), possibly because of the greater arterial
baroreceptor activation by the larger pressor re-
sponses. The treatment had little impact on resting
AP (
100 mmHg) or HR (
500 beats/min). At 6
weeks after compression cord injury, the increases
in AP during dysreflexia were greater than at 2
weeks and HR responses to colon distension in the
control rats more often included arrhythmias
(Fig. 5). AP responses of the treated rats remained
smaller than those of the control rats at this time.
Again, the mean changes in AP of the control rats
(
40 mmHg) were significantly greater than those
of the treated rats (
28 mmHg). At this time, as at
the earlier 2-week analysis, the anti-CD11d treat-
ment did not alter resting AP (
110 mmHg) or
HR (
500 beats/min). Additional experiments
were done to determine the effect of delaying the
onset of treatment to 6 h after compression injury.
The second and third injections of anti-CD11d
antibody were administered at 24 and 48 h, re-
spectively. This treatment also decreased dysre-
flexia by
50% at 6 weeks after cord injury,
demonstrating that a clinically relevant dosing
schedule still is highly effective.
The basis for the improved autonomic outcome
after this selective anti-inflammatory treatment
was sparing of the tissue surrounding the lesion ar-
ea. The spinal cords had increased areas of compact
myelin within 2–3 mm of the injury site at 2 and 6
weeks after injury (Fig. 6). Neurofilament in the

Fig. 5. Anti-CD11d mAb treatment reduces the magnitude of autonomic dysreflexia. Changes in pulsatile AP, MAP and HR are
shown during a 2 ml balloon distension of the colon in control (A and D) and mAb-treated rats (B and E) at 2 weeks (top row) and 6
weeks (bottom row) after SCI at T4. The large increase in pressure evoked at 6 weeks was often associated with arrhythmias (D).
Average change in MAP was assessed at 2 (C) and 6 (F) weeks after cord injury in control rats (& C, n ¼ 14; F, n ¼ 6) and mAb-
treated rats (’ C, n ¼ 10; F, n ¼ 7). The dotted lines on the bar graphs show the approximate change in pressure that would be elicited
by such stimulation in a rat with an intact spinal cord (Maiorov et al., 1997a). bpm, beats per minute; *Po0.05 compared to control
rats. Reprinted with permission from Gris et al. (2004), copyright 2004 by the Society for Neuroscience.
Fig. 6. Anti-CD11d mAb-treatment increases myelin in the injured cord after SCI. Solochrome cyanin-stained sections taken at 6 weeks after T4 SCI are shown at the
lesion epicenter and at 2.0 mm caudal to the epicenter in control and mAb-treated rats (a). Arrows on photomicrographs at the epicenter indicate patches of intact dark
blue compact myelin. Note the more intact neuropil at 2 mm caudal to the injury site in the rat treated with anti-CD11d mAb, reflected by more abundant compact myelin
and a smaller cavitation than in the control rat. Calibration bar equals 200 mm and applies to all photomicrographs. Treatment effects on normalized areas of compact
myelin are illustrated after SCI at T4 (b, 2 weeks; c, 6 weeks). Myelin was stained with luxol fast blue in b and solochrome cyanin in c. & control rats (b, n ¼ 10; c, n ¼ 5).
’ mAb-treated rats (b, n ¼ 8; c, n ¼ 7). *Po0.05 compared to control rats. +shortest distance where area is larger than area at epicenter, Po0.05. Reprinted with
permission from Gris et al. (2004) copyright 2004 by the Society for Neuroscience.

257
258

Fig. 7. Longitudinal section of rat lumbar (L2) spinal cord illustrating fibers immunoreactive for 5-HT (serotonin, green) streaming
from the spared white matter bundle at the edge of the spinal cord caudal to a clip-compression SCI at the 12th thoracic segment. This
section shows the abundance of these fibers at 4 weeks after cord injury in a rat that had been treated with the anti-CD11d mAb. Fibers
extend to the edge of the gray matter (bottom of illustration) and appear to surround sympathetic preganglionic neurons in the
intermediolateral cell column. The inset shows these lateral horn neurons stained for neurofilament immunoreactivity (red) with fibers
and possible terminals on their somata. This response suggests that axons spared by the anti-CD11d mAb treatment produced
collateral sprouting branches directed toward the preganglionic neurons and possibly innervating them. Calibration bar ¼ 50 mm (inset
10 mm).

gray and white matter was also significantly
increased, indicative of neuronal sparing. The
improvement in dysreflexia evoked by colon
distension did not relate to effects on the CGRP-
immunoreactive primary afferent arbor. No treat-
ment effects on this arbor were found in the dorsal
horn of the lumbar spinal cord, the segments
mediating the afferent input from colon distension
(Gris et al., 2005). Remarkably, the afferent arbor
at T9, caudal to the injury site, actually had a
greater area in antibody-treated than control rats.
This presumably reflected the tissue sparing in the
dorsal horn as the treated rats had almost intact
dorsal horns, whereas the controls had significant
secondary tissue damage that extended from T4 to
T9. This change would not have impacted on
dysreflexia initiated from the colon but might have
affected dysreflexia evoked from the mid-thoracic
region. This relationship was not assessed in our
study, as we did not anticipate the effect near the
lesion. However, other than effects due to tissue
sparing, the anti-integrin antibody treatment
did not alter the CGRP-immunoreactive arbor.
Instead effects appeared to be caused by retention
of some descending control of spinal neurons.
In a different study of effects of the anti-CD11d
antibody treatment (Oatway et al., 2005), we
found that it led to extensive sprouting of de-
scending serotonergic axons that passed through
the margins of the lesion. These axons were par-
ticularly attracted to the intermediolateral cell col-
umn, forming masses of projections to this area
(Fig. 7). The projections provided restored, even
enhanced, bulbospinal serotonergic inputs to the
preganglionic neurons caudal to the injury site. As
this study concerned only the injury and responses
of the lower thoracic and lumbar cord, it would
not have affected autonomic dysreflexia and we
can only speculate that the same phenomenon
would occur after a mid-thoracic injury, leading to
enhanced serotonergic input to all preganglionic
neurons caudal to the injury site. In summary,
blocking inflammation with a selective, transient
approach, which does not impair later wound
healing actions of leukocytes, markedly decreased
the development of autonomic dysreflexia, by
mechanisms that appear to be related to preserva-
tion or sprouting of spared descending pathways.
259
This approach has significant clinical promise as it
also has beneficial effects on motor function and
the development of neuropathic pain (Gris et al.,
2004; Oatway et al., 2005).
In conclusion, SCI leads to a host of changes
within the injured cord that promote pathological
secondary outcomes such as autonomic dysreflex-
ia. Although neurotrophic factors are useful and
probably necessary for the process of regeneration
of injured tracts and neurons, NGF does not ap-
pear to be a good candidate for this task as it
clearly contributes to secondary maladaptive dis-
orders. Likewise, inflammation is a necessary part
of healing but, in the CNS, inflammation can
cause great damage that is not readily reversed.
Therefore, the inflammatory response to cord
injury also is an ideal target for manipulation to
maintain its useful functions while limiting its
capacity to cause destructive or maladaptive
responses. Autonomic dysreflexia is one of many
secondary consequences of cord injury such as
bladder and sexual dysfunction, chronic pain,
and muscle spasticity. Understanding the mecha-
nisms of autonomic dysreflexia and development
of treatments to prevent its development may im-
pact on many of these disabling secondary
disorders.
Acknowledgments
This research was supported by grants from the
Canadian Institutes of Health Research, The
Heart and Stroke Foundation of Ontario
(T4053) and The Ontario Neurotrauma Founda-
tion. The authors wish to acknowledge that part of
the research reviewed in this chapter was conduct-
ed by previous postdoctoral fellows and graduate
students, Dr. Andrei Krassioukov, Dr. Dmitry
Mayorov, Dr. Natalie Krenz, Dr. Sharon Wong,
Mr. Aly Cassam, Ms. Mary-Jo Ricci and Mr.
Mark Oatway. We also are indebted to our past
and present excellent technical staff, Mrs. Barbara
Atkinson, Ms. Yuhua Chen, Ms. Eilis Hamilton,
Ms. Leyana Saville and Mrs. Carmen Simedrea
who have been essential to this research.
260
References
Agrawal, S.K. and Fehlings, M.G. (1996) Mechanisms of
secondary injury to spinal cord axons in vitro role Na+,
Na+–K+–ATPase, the Na+–H+ exchanger, and the
Na+–Ca+ exchanger. J. Neurosci., 16(2): 545–552.
Agrawal, S.K. and Fehlings, M.G. (1997a) Role of NMDA and
non-NMDA ionotropic glutamate receptors in traumatic
spinal cord axonal injury. J. Neurosci., 17(3): 1055–1063.
Agrawal, S.K. and Fehlings, M.G. (1997b) The effect of the
sodium channel blocker QX-314 on recovery after acute
spinal cord injury. J. Neurotrauma, 14: 81–88.
Agrawal, S.K., Theriault, E. and Fehlings, M.G. (1998) Role of
group I metabotropic glutamate receptors in traumatic spinal
cord white matter injury. J. Neurotrauma, 15: 929–941.
Bakhit, C., Armanini, M., Wong, W.L.T., Bennett, G.L. and
Wrathall, J.R. (1991) Increase in nerve growth factor-like
immunoreactivity and decrease in choline acetyltransferase
following contusive spinal cord injury. Brain Res., 554:
264–271.
Bandtlow, C.E., Meyer, M., Lindholm, D., Spranger, M.,
Heumann, R. and Thoenen, H. (1990) Regional and cellular
codistribution of interleukin 1 beta and nerve growth factor
mRNA in the adult rat brain: possible relationship to the
regulation of nerve growth factor synthesis. J. Cell Biol., 111:
1701–1711.
Bao, F., Chen, Y., Dekaban, G.A. and Weaver, L.C. (2004a)
An anti-CD11d integrin antibody reduces cyclooxygenase-2
expression and protein and DNA oxidation after spinal cord
injury in rats. J. Neurochem., 90: 1194–1204.
Bao, F., Chen, Y., Dekaban, G.A. and Weaver, L.C. (2004b)
Early anti-inflammatory treatment reduces lipid peroxidation
and protein nitration after spinal cord injury in rats.
J. Neurochem., 88: 1335–1344.
Bethea, J.R. and Dietrich, W.D. (2002) Targeting the host in-
flammatory response in traumatic spinal cord injury. Curr.
Opin. Neurol., 15(3): 355–360.
Bevilacqua, M.P. (1993) Endothelial-leukocyte adhesion mol-
ecules. Ann. Rev. Immunol., 11: 767–804.
Blackmer, J. (2003) Rehabilitation medicine: 1. Autonomic
dysreflexia. CMAJ, 169: 931–935.
Blight, A.R. (1985) Delayed demyelination and macrophage
invasion: a candidate for ‘‘secondary’’ cell damage in spinal
cord injury. Central Nervous System Trauma, 2: 299–315.
Brown, A., Ricci, M.-J. and Weaver, L.C. (2004) NGF message
and protein distribution in the injured rat spinal cord. Exp.
Neurol., 188: 115–127.
Bullock, E.D. and Johnson, E.M. (1996) Nerve growth factor
induces the expression of certain cytokine genes and bcl-2 in
mast cells. Potential role in survival promotion. J. Biol.
Chem., 271: 27500–27508.
Cabot, J., Alessi, V., Carroll, J. and Ligorio, M. (1994) Spinal
cord lamina V and lamina VII interneuronal projections to
sympathetic preganglionic neurons. J. Comp. Neurol., 347:
515–530.
Calaresu, F.R. and Yardley, C.P. (1988) Medullary basal sym-
pathetic tone. Ann. Rev. Physiol., 50: 511–524.
Cameron, A.A., Smith, G.M., Randall, D.C., Brown, D.R. and
Rabchevsky, A.G. (2003) Effects of nerve growth factor over-
expression on autonomic dysreflexia after spinal cord injury.
J. Neurotrauma, 20: 1086.
Cassam, A.K., Llewellyn-Smith, I.J. and Weaver, L.C. (1997)
Catecholamine enzymes and neuropeptides are expressed in
fibres and somata in the intermediate grey matter in chronic
spinal rats. Neuroscience, 78: 829–841.
Cassam, A.K., Rogers, K.A. and Weaver, L.C. (1999) Co-
localization of substance P and dopamine b-hydroxylase with
growth associated protein-43 is lost caudal to a spinal cord
transection. Neuroscience, 88: 1275–1288.
Clarke, H.A., Dekaban, G.A. and Weaver, L.C. (1998) Iden-
tification of lamina V and VII interneurons presynaptic to
adrenal sympathetic preganglionic neurons in rats using a
recombinant herpes simplex virus type 1. Neuroscience, 85:
863–872.
Coote, J.H., Downman, C.B.B. and Weber, W.V. (1969) Reflex
discharges into thoracic white rami elicited by somatic and
visceral afferent excitation. J. Physiol., 202: 147–159.
Corbett, J.L., Debarge, O., Frankel, H.L. and Mathias, C.
(1975) Cardiovascular responses in tetraplegic man to muscle
spasm, bladder percussion and head-up tilt. Clin. Exp.
Pharmacol. Physiol., Suppl., 2: 189–193.
Dembowsky, K., Czachurski, J., Amendt, K. and Seller, H.
(1978) Tonic, supraspinal, monoaminergic inhibition on spi-
nal somato-sympathetic reflexes. Pflugers Arch., 373: R76.
Dembowsky, K., Czachurski, J., Amendt, K. and Seller, H.
(1980) Tonic descending inhibition of the spinal somato-
sympathetic reflex from the lower brain stem. J. Auton. Nerv.
Syst., 2: 157–182.
Erickson, R.P. (1980) Autonomic hyperreflexia: pathophysiol-
ogy and medical management. Arch. Phys. Med. Rehabil.,
61: 431–440.
Fehlings, M.G., Tator, C.H. and Linden, R.D. (1989) The ef-
fect of nimodipine and dextran on axonal function and blood
flow following experimental spinal cord injury. J. Neurosurg.,
71: 403–416.
Giannantoni, A., Di, S.S., Scivoletto, G., Mollo, A., Silecchia,
A., Fuoco, U. and Vespasiani, G. (1998) Autonomic
dysreflexia during urodynamics. Spinal Cord, 36: 756–760.
Grayson, M.H., Van der Vieren, M., Sterbinsky, S.A., Gallantin,
W.M., Hoffman, P., Staunton, D. and Bochner, B.S. (1999)
alphadbeta2 integrin is a ligand for vascular cell adhesion
molecule-1. Int. Arch. Allergy Immunol., 118(2–4): 263–264.
Gris, D., Marsh, D.R., Oatway, M.A., Chen, Y., Hamilton,
E.F., Dekaban, G.A. and Weaver, L.C. (2004) Transient
blockade of the CD11d/CD18 integrin reduces secondary
damage after spinal cord injury, improving sensory, auto-
nomic, and motor function. J. Neurosci., 24: 4043–4051.
Gris, D., Marsh, D.R., Dekaban, G.A. and Weaver, L.C.
(2005). Comparison of methylprednisolone and anti-CD11d
antibody treatments on autonomic dysreflexia after spinal
cord injury. Exp. Neurol.
Guttman, F.L. and Whitteridge, D. (1947) Effects of bladder
distention on autonomic mechanisms after spinal cord inju-
ries. Brain, 70: 361–404.

Hayashi, M., Ueyama, T., Nemoto, K., Tamaki, T. and
Senba, E. (2000) Sequential mRNA expression for
immediate early genes, cytokines, and neurotrophins in
spinal cord injury. J. Neurotrauma, 17: 203–218.
Heese, K., Fiebich, B.L., Bauer, J. and Otten, U. (1998a)
NF-kappaB modulates lipopolysaccharide-induced micro-
glial nerve growth factor expression. Glia, 22: 401–407.
Heese, K., Hock, C. and Otten, U. (1998b) Inflammatory sig-
nals induce neurotrophin expression in human microglial
cells. J. Neurochem., 70: 699–707.
Heumann, R., Korsching, S., Bandtlow, C. and Thoenen, H.
(1987a) Changes of nerve growth factor synthesis in non-
neuronal cells in response to sciatic nerve transection. J. Cell
Biol., 104: 1623–1631.
Heumann, R., Lindholm, D., Bandtlow, C., Meyer, M.,
Radeke, M.J., Misko, T.P., Shooter, E. and Thoenen, H.
(1987b) Differential regulation of mRNA encoding nerve
growth factor and its receptor in rat sciatic nerve during
development, degeneration, and regeneration: role of
macrophages. Proc. Natl. Acad. Sci. USA, 84: 8735–8739.
Jacob, J.E., Gris, P., Fehlings, M.G., Weaver, L.C. and Brown,
A. (2003) Autonomic dysreflexia after spinal cord transection
and compression in 129Sv, C57BL and wallerian degenera-
tion slow mutant mice. Exp. Neurol., 183: 136–146.
Jacob, J.E., Pniak, A., Weaver, L.C. and Brown, A. (2001)
Autonomic dysreflexia in a mouse model of spinal cord
injury. Neuroscience, 108: 687–693.
Jones, L.L., Yamaguchi, Y., Stallcup, W.B. and Tuszynski,
M.H. (2002) NG2 is a major chondroitin sulfate proteogly-
can produced after spinal cord injury and is expressed by
macrophages and oligodendrocyte progenitors. J. Neurosci.,
22: 2792–2803.
Joshi, S., LeVatte, M.A., Dekaban, G.A. and Weaver, L.C. (1995)
Identification of spinal interneurons antecedent to adrenal
sympathetic preganglionic neurons using trans-synaptic trans-
port of herpes simplex virus type 1. Neuroscience, 65: 893–903.
Ju, G., Hokfelt, T., Brodin, E., Fahrenkrug, J., Fischer, J.A.,
Frey, P., Elde, R.P. and Brown, J.C. (1987) Primary sensory
neurons of the rat showing calcitonin gene-related peptide
immunoreactivity and their relation to substance P-,
somatostatin-, galanin-, vasoactive intestinal polypeptide-
and cholecystokinin-immunoreactive ganglion cells. Cell
Tissue Res., 247: 417–431.
Karlsson, A.-K. (1999) Autonomic dysreflexia. Spinal Cord, 37:
383–391.
Karlsson, A.-K., Friden, P.M., Lonnroth, P., Sullivan, L. and
Elam, M. (1998) Regional sympathetic function in high
spinal cord injury during mental stress and autonomic
dysreflexia. Brain, 121: 1711–1719.
Koizumi, K. and Brooks, C.M. (1972) The integration of
autonomic system reactions: a discussion of autonomic
reflexes, their control and their association with somatic
reflexes. In: Adrian R.H., Helmrich E., Holzer H., Jung R.,
Kramer K., Krayer O., Lynen F., Miescher P.A., Rasmussen
H., Renold A.E., Trendelenburg U., Ullrich K., Vogt W. and
Weber A. (Eds.), Reviews of Physiology: Biochemistry and
Experimental Pharmacology. Springer, Berlin, pp. 1–68.
261
Koyanagi, I., Tator, C.H. and Lea, P.J. (1993a) Three-dimen-
sional analysis of the vascular system in the rat spinal cord
with scanning electron microscopy of vascular corrosion
casts. Part 2: Acute spinal cord injury. Neurosurgery, 33:
285–291.
Koyanagi, I., Tator, C.H. and Theriault, E. (1993b) Silicone
rubber microangiography of acute spinal cord injury in the
rat. Neurosurgery, 32: 260–268.
Krassioukov, A.V., Bunge, M.B., Puckett, W.R. and Bygrave,
M.A. (1999) The changes in human spinal sympathetic
preganglionic neurons after spinal cord injury. Spinal Cord,
37: 6–13.
Krassioukov, A.V., Furlan, J.C. and Fehlings, M.G. (2003)
Autonomic dysreflexia in acute spinal cord injury: an under-
recognized clinical entity. J. Neurotrauma, 20: 707–716.
Krassioukov, A.V., Johns, D.G. and Schramm, L.P. (2002)
Sensitivity of sympathetically correlated spinal interneurons,
renal sympathetic nerve activity, and arterial pressure to so-
matic and visceral stimuli after chronic spinal injury.
J. Neurotrauma, 19: 1521–1529.
Krassioukov, A.V. and Weaver, L.C. (1995) Episodic
hypertension due to autonomic dysreflexia in acute and
chronic spinal cord-injured rats. Am. J. Physiol., 268:
H2077–H2083.
Krassioukov, A.V. and Weaver, L.C. (1996) Morphological
changes in sympathetic preganglionic neurons after spinal
cord injury in rats. Neuroscience, 70: 211–226.
Krenz, N.R., Meakin, S.O., Krassioukov, A.V. and Weaver,
L.C. (1999) Neutralizing intraspinal nerve growth factor
blocks autonomic dysreflexia caused by spinal cord injury.
J. Neurosci., 19(17): 7405–7414.
Krenz, N.R. and Weaver, L.C. (1998a) Changes in the mor-
phology of sympathetic preganglionic neurons parallel the
development of autonomic dysreflexia after spinal cord injury
in rats. Neurosci. Lett., 243: 61–64.
Krenz, N.R. and Weaver, L.C. (1998b) Sprouting of primary
afferent fibers after spinal cord transection in the rat.
Neuroscience, 85: 443–458.
Krenz, N.R. and Weaver, L.C. (2000) Nerve growth factor in
glia and inflammatory cells of the injured rat spinal cord.
J. Neurochem., 74: 730–739.
Landrum, L.M., Thompson, G.M. and Blair, R.W. (1998) Does
postsynaptic a1-adrenergic receptor supersensitivity contrib-
ute to autonomic dysreflexia? Am. J. Physiol., 274:
1090–1098.
LaSala, A., Corinti, S., Federici, M., Saragovi, H.U. and
Girolomoni, G. (2000) Ligand activation of nerve growth
factor receptor TrkA protects monocytes from apoptosis.
J. Leukoc. Biol., 68: 104–110.
Lawson, S.N., McCarthy, P.W. and Prabhakar, E. (1996)
Electrophysiological properties of neurones with CGRP-like
immunoreactivity in rat dorsal root ganglia. J. Comp. Ne-
urol., 365: 355–366.
Lawson, S.N., Perry, M.J., Prabhakar, E. and McCarthy, P.W.
(1993) Primary sensory neurones: neurofilament,
neuropeptides, and conduction velocity. Brain Res. Bull.,
30: 239–243.
262
Lee, B.Y., Karmakar, M.G., Herz, B.L. and Sturgill, R.A.
(1994) Autonomic dysreflexia revisited. J. Spinal Cord Med.,
18: 75–87.
Leskovar, A., Moriarty, L.J., Turek, J.J., Schoenlein, I.A. and
Borgens, R.B. (2000) The macrophage in acute neural injury:
changes in cell numbers over time and levels of cytokine
production in mammalian central and peripheral nervous
systems. J. Exp. Biol., 203(Part 12): 1783–1795.
Lindan, R., Joiner, E., Freehafer, A.A. and Hazel, C. (1980)
Incidence and clinical features of autonomic dysreflexia in
patients with spinal cord injury. Paraplegia, 18: 285–292.
Lindholm, D., Heumann, R., Meyer, M. and Thoenen, H. (1987)
Interleukin-1 regulates synthesis of nerve growth factor in
non-neuronal cells of rat sciatic nerve. Nature, 330: 658–659.
Llewellyn-Smith, I.J. and Weaver, L.C. (2001) Changes in
synaptic inputs to sympathetic preganglionic neurons after
spinal cord injury. J. Comp. Neurol., 435: 226–240.
Mabon, P.J., Weaver, L.C. and Dekaban, G.A. (2000) Inhibi-
tion of monocyte/macrophage migration to a spinal cord in-
jury site by an antibody to the integrin alphaD: a potential
new anti-inflammatory treatment. Exp. Neurol., 166: 52–64.
Maiorov, D.N., Fehlings, M.G. and Krassioukov, A.V. (1998)
Relationship between severity of spinal cord injury and ab-
normalities in neurogenic cardiovascular control in conscious
rats. J. Neurotrauma, 15(5): 365–374.
Maiorov, D.N., Krenz, N.R., Krassioukov, A.V. and Weaver,
L.C. (1997a) Role of spinal NMDA and AMPA receptors in
episodic hypertension in conscious spinal rats. Am. J.
Physiol., 273: H1266–H1274.
Maiorov, D.N., Weaver, L.C. and Krassioukov, A.V. (1997b)
Relationship between sympathetic activity and arterial pres-
sure in conscious spinal rats. Am. J. Physiol., 272:
H625–H631.
Marsh, D.R. and Weaver, L.C. (2004) Autonomic dysreflexia,
induced by noxious or innocuous stimulation, does not de-
pend on changes in dorsal horn substance P. J. Neurotrauma,
21: 817–828.
Marsh, D.R., Wong, S.T., Meakin, S.O., MacDonald, J.I.S.,
Hamilton, E.F. and Weaver, L.C. (2002) Neutralizing intra-
spinal NGF with a trkA-IgG fusion protein blocks develop-
ment of autonomic dysreflexia in a clip-compression model of
spinal cord injury. J. Neurotrauma, 19: 1531–1542.
Mathias, C.J. and Frankel, H.L. (1992a) Autonomic distur-
bances in spinal cord lesions. In: Mathias C.J. and Bannister
R. (Eds.) Autonomic Failure A Textbook of Clinical Disor-
ders of the Autonomic Nervous System, Vol. 4(51). Oxford
University Press, Oxford, pp. 494–513.
Mathias, C.J. and Frankel, H.L. (1992b) Autonomic distur-
bances in spinal cord lesions. In: Bannister R. and Mathias
C.J. (Eds.) Autonomic Failure, a Textbook of Clinical Dis-
orders of the Autonomic Nervous System, Vol. 3(43). Oxford
Medical Publications, Oxford, UK, pp. 839–881.
Mathias, C.J. and Frankel, H.L. (1992c) The cardiovascular
system in tetraplegia and paraplegia. In: Frankel H.L. (Ed.)
Handbook of Clinical Neurology, Vol. 17(25). Elsevier
Science Publishers, B.V., Amsterdam, The Netherlands,
pp. 435–456.
Mayorov, D.N., Adams, M.A. and Krassioukov, A.V. (2001)
Telemetric blood pressure monitoring in conscious rats be-
fore and after compression injury of spinal cord. J. Neuro-
trauma, 18: 727–736.
McCarthy, P.W. and Lawson, S.N. (1990) Cell type and con-
duction velocity of rat primary sensory neurons with ca-
lcitonin gene-related peptide-like immunoreactivity.
Neuroscience, 34: 623–632.
Naftchi, N.E. (1990) Mechanism of autonomic dysreflexia;
contributions of cathecholamine and peptide neurotransmit-
ters. Ann. NY Acad. Sci., 579: 133–148.
Naftchi, N.E. and Richardson, J.S. (1997) Autonomic dysreflexia:
pharmacological management of hypertensive crises in spinal
cord injured patients. J. Spinal. Cord. Med., 20: 355–360.
Neish, A.S., Read, M.A., Thanos, D., Pine, R., Maniatis, T.
and Collins, T. (1995) Endothelial interferon regulatory fac-
tor 1 cooperates with NF-kappa B as a transcriptional ac-
tivator of vascular cell adhesion molecule 1. Mol. Cell Biol.,
15(5): 2558–2569.
Oatway, M.A., Chen, Y., Bruce, J.C., Dekaban, G.A. and
Weaver, L.C. (2005) An anti-CD11d integrin antibody treat-
ment restores normal serotonergic projections to the dorsal,
intermediate and ventral horns of the injured spinal cord.
J. Neurosci., 25: 637–647.
Osborn, J.W., Taylor, R.F. and Schramm, L.P. (1990) Chronic
cervical spinal cord injury and autonomic hyperreflexia in
rats. Am. J. Physiol., 258: R169–R174.
Popovich, P.G., Guan, Z., Wei, P., Huitinga, I., van Rooijen,
N. and Stokes, B.T. (1999) Depletion of hematogenous ma-
crophages promotes partial hindlimb recovery and neuroan-
atomical repair after experimental spinal cord injury. Exp.
Neurol., 158: 351–365.
Popovich, P.G., Wei, P. and Stokes, B.T. (1997) Cellular in-
flammatory response after spinal cord injury in Sprague–Da-
wley and Lewis rats. J. Comp. Neurol., 377: 443–464.
Price, J. (1985) An immunohistochemical and quantitative ex-
amination of dorsal root ganglion neuronal subpopulations.
J. Neurosci., 5(8): 2051–2059.
Rodenbaugh, D.W., Collins, H.L. and DiCarlo, S.E. (2003)
Increased susceptibility to ventricular arrhythmias in hyper-
tensive paraplegic rats. Clin. Exp. Hypertens., 25: 349–358.
Saporito, M.S., Wilcox, H.M., Hartpence, K.C., Lewis, M.E.,
Vaught, J.L. and Carswell, S. (1993) Pharmacological induc-
tion of nerve growth factor mRNA in adult rat brain. Exp.
Neurol., 123: 295–302.
Saragovi, H.U. and Gehring, K. (2000) Development of phar-
macological agents for targeting neurotrophins and their re-
ceptors. Trends Pharmacol. Sci., 21: 93–98.
Sato, A. and Schmidt, R.F. (1973) Somatosympathetic reflexes:
afferent fibers, central pathways, discharge characteristics.
Physiol. Rev., 53(4): 916–947.
Saville, L.R., Pospisil, C.H., Mawhinney, L.A., Bao, F.,
Simedria, F.C., Peters, A.A., O’Connell, P.J., Weaver, L.C.
and Dekaban, G.A. (2004) A monoclonal antibody to
CD11d reduces the inflammatory infiltrate into the injured
spinal cord: a potential neuroprotective treatment. J. Ne-
uroimmunol., 156: 42–57.

Schumacher, P.A., Eubanks, J.H. and Fehlings, M.G. (1999)
Increased calpain I-mediated proteolysis and preferential loss
of dephosphorylated NF200, following traumatic spinal cord
injury. Neuroscience, 91: 733–744.
Shanley, T.P., Warner, R.L., Crouch, L.D., Dietsch, G.N.,
Clark, D.L., O’Brien, M.M., Gallatin, W.M. and Ward, P.A.
(1998) Requirements for ad in IgG immune complex-induced
rat lung injury. J. Immunol., 160: 1014–1020.
Shea, J.D., Gioffre, R., Carrion, H. and Small, M.P. (1973)
Autonomic hyperreflexia in spinal cord injury. Southern
Med. J., 66(8): 869–872.
Spranger, M., Lindholm, D., Bandtlow, C., Heumann, R.,
Gnahn, H., Naher-Noe, M. and Thoenen, H. (1990) Regu-
lation of nerve growth factor (NGF) synthesis in the rat
central nervous system: comparison between the effects of
interleukin-1 and various growth factors in astrocyte cultures
and in vivo. Eur. J. Neurosci., 2: 69–76.
Streit, W.J., Semple-Rowland, S.L., Hurley, S.D., Miller, R.C.,
Popovich, P.G. and Stokes, B.T. (1998) Cytokine mRNA
profiles in contused spinal cord and axotomized facial nu-
cleus suggest a beneficial role for inflammation and gliosis.
Exp. Neurol., 152: 74–87.
Taoka, Y. and Okajima, K. (1998) Spinal cord injury in the rat.
Prog. Neurobiol., 56: 341–358.
Taoka, Y., Okajima, K., Uchiba, M., Murakami, K.,
Kushimoto, S., Johno, M., Naruo, M., Okabe, H. and
Takatsuki, K. (1997) Role of neutrophils in spinal cord injury
in the rat. Neuroscience, 79(4): 1177–1182.
Tator, C.H. and Fehlings, M.G. (1991) Review of the secondary
injury theory of acute spinal cord trauma with emphasis on
vascular mechanisms. J. Neurosurg., 75: 15–26.
Vaidyanathan, S., Soni, B.M., Sett, P., Watt, J.W.H., Oo, T.
and Bingley, J. (1998) Pathophysiology of autonomic
dysreflexia: long-term treatment with terazosin in adult and
paediatric spinal cord injury patients manifesting recurrent
dysreflexic epidsodes. Spinal Cord, 36: 761–770.
Van der Vieren, M., Crowe, D.T., Hoekstra, D., Vazeux, R.,
Hoffman, P.A., Grayson, M.H., Bochner, B.S., Gallatin,
263
W.M. and Staunton, D.E. (1999) The leukocyte integrin
aDb2 binds VCAM-1: evidence for a binding interface be-
tween I domain and VCAM-1. J. Immunol., 163: 1984–1990.
Vizzard, M.A., Brisson, M. and De Groat, W.C. (2000)
Transneuronal labeling of neurons in the adult rat central
nervous system following inoculation of pseudorabies virus
into the colon. Cell Tissue Res., 299(1): 9–26.
Wallace, M.C. and Tator, C.H. (1986) Spinal cord blood flow
measured with microspheres following spinal cord injury in
the rat. Can. J. Neurol. Sci., 13: 91–96.
Wallace, M.C., Tator, C.H. and Frazee, P. (1986) Relationship
between posttraumatic ischemia and hemorrhage in the in-
jured rat spinal cord as shown by colloidal carbon angio-
graphy. Neurosurgery, 18: 433–439.
Weaver, L.C., Cassam, A.K., Krassioukov, A.V. and
Llewellyn-Smith, I.J. (1997) Changes in immunoreactivity
for growth associated protein-43 suggest reorganization of
synapses on spinal sympathetic neurons after cord transec-
tion. Neuroscience, 81: 535–551.
Weaver, L.C., Marsh, D.R., Gris, D., Meakin, S.O. and
Dekaban, G.A. (2001a) Central mechanisms for autonomic
dysreflexia after spinal cord injury. In: McKerracher L.,
Douchet D. and Rossignol S. (Eds.) Progress in Brain Re-
search, Vol. 7. Elsevier Science B.V., Amsterdam, pp. 83–95.
Weaver, L.C., Verghese, P., Bruce, J.C., Fehlings, M.G., Krenz,
N.R. and Marsh, D.R. (2001b) Autonomic dysreflexia and
primary afferent sprouting after clip-compression injury of
the rat spinal cord. J. Neurotrauma, 18: 1107–1119.
Wong, S.T., Atkinson, B.A. and Weaver, L.C. (2000) Confocal
microscopic analysis of sprouting small-diameter primary af-
ferent fibres after spinal cord injury. Neurosci. Lett., 296:
65–68.
Yeoh, M., McLachlan, E.M. and Brock, J.A. (2004) Chronic
decentralization potentiates neurovascular transmission in
the isolated rat tail artery, mimicking the effects of spinal
transection. J. Physiol., 561: 583–596.
Young, W. (1993) Secondary injury mechanisms in acute spinal
cord injury. J. Emerg. Med., 11: 13–22.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

Chapter 17

Segmental organization of spinal reflexes mediating

autonomic dysreflexia after spinal cord injury

Alexander G. Rabchevsky

University of Kentucky, Spinal Cord & Brain Injury Research Center and Department of Physiology, 741 South Limestone
Street, B371 BBSRB, Lexington, KY 40536-0509, USA

Abstract: Spinal cord injuries above mid-thoracic levels can lead to a potentially life-threatening hyper-
tensive condition termed autonomic dysreflexia that is often triggered by distension of pelvic viscera
(bladder or bowel). This syndrome is characterized by episodic hypertension due to sudden, massive
discharge of sympathetic preganglionic neurons in the thoracolumbar spinal cord. This hypertension is
usually accompanied by bradycardia, particularly if the injury is caudal to the 2nd to 4th thoracic spinal
segments. The development of autonomic dysreflexia is correlated with aberrant sprouting of peptidergic
afferent fibers into the spinal cord below the injury. In particular, sprouting of nerve growth factor-
responsive afferent fibers has been shown to have a major influence on dysreflexia, perhaps by amplifying
the activation of disinhibited sympathetic neurons. Using a model of noxious bowel distension after com-
plete thoracic spinal transection at the 4th thoracic segment in rats, we selectively altered C-fiber sprouting,
at specified spinal levels caudal to the injury, with microinjections of adenovirus encoding the growth-
promoting nerve growth factor or the growth-inhibitory semaphorin 3A. This was followed by assessment
of physiological responses to colorectal distension and subsequent histology. Additionally, anterograde
tract tracers were injected into the lumbosacral region to compare the extent of labeled propriospinal
rostral projections in uninjured cords to those in cords after complete 4th thoracic transection. In summary,
overexpression of chemorepulsive semaphorin 3A impeded C-fiber sprouting in lumbosacral segments and
mitigated hypertensive autonomic dysreflexia, whereas the opposite results were obtained with nerve
growth factor overexpression. Furthermore, compared to naı̈ve rats, there were significantly more labeled
lumbosacral propriospinal projections rostrally after thoracic injury. Collectively, our findings suggest that
distension of pelvic viscera increases the excitation of expanded afferent terminals in the disinhibited
lumbosacral spinal cord. This, in turn, triggers excitation and sprouting of local propriospinal neurons to
relay visceral sensory stimuli and amplify the activation of sympathetic preganglionic neurons in the
thoracolumbar cord, to enhance transmission in the spinal viscero-sympathetic reflex pathway. These
responses are manifested as autonomic dysreflexia.

Autonomic dysreflexia is a clinical syndrome that
develops following spinal cord injury above the
sixth thoracic (T) spinal segment (T6). It is present
after complete as well as incomplete injuries
(Karlsson, 1999) with an incidence of up to 70%
Corresponding author. Tel.: +(859) 323-0267;
Fax: +(859) 257-5737; E-mail: AGRab@uky.edu
after complete injuries (Snow et al., 1978). The
condition is commonly triggered by distension of
the pelvic viscera (bowel and bladder) and is man-
ifested by often debilitating hypertension and
sweating, dizziness, nausea, and often severe head-
aches (Snow et al., 1978; Lindan et al., 1980;
Harati, 1997; Karlsson, 1999). A reflex brad-
ycardia often accompanies these episodes of

DOI: 10.1016/S0079-6123(05)52017-X 265

266
hypertension, particularly if the injury is caudal to
the spinal segments providing sympathetic control
of the heart (T1–T4). During distension of the
bladder or bowel, the consequent afferent barrage
into the dorsal horn of the lumbosacral spinal cord
ultimately results in massive sympathetic reflex
discharge below the level of the injury. This results
in vasoconstriction of the muscular, splanchnic,
and cutaneous vascular beds (reviewed in Karlsson,
1999). The resultant paroxysmal hypertension pro-
duces a baroreceptor-mediated reflex bradycardia
accompanied by withdrawal of sympathetic activ-
ity above the lesion level with resultant vasodila-
tation that produces adverse symptoms such as
headaches and skin flushing.
The dysreflexic hypertension stems, in part,
from injury-induced loss of tonic control of sym-
pathetic preganglionic neurons in the intermedio-
lateral cell column of the thoracolumbar spinal
cord by medullo-spinal neurons in the rostral and
caudal ventrolateral medulla (Finestone and Teasell,
1993; Zagon and Smith, 1993). Anatomical and
physiological changes that occur in sympathetic
preganglionic neurons and sympathetically related
interneurons as descending inputs degenerate
following high thoracic spinal cord transection
have been the subject of considerable attention
(Weaver et al., 1997; Krenz and Weaver, 1998b;
Klimaschewski, 2001). Additionally, the influence
of injury-induced sprouting of primary afferent
fibers into the thoracolumbar spinal cord has been
the focus of investigations (Krenz and Weaver,
1998a; Weaver et al., 2001). Although autonomic
dyreflexia can be elicited by even non-noxious
stimuli below the injury level (Marsh and Weaver,
2004), a model for consistently inducing autonom-
ic dysreflexia employs colorectal distension in spi-
nal-injured rats. This model is designed to mimic
the common clinical manifestation of noxious fecal
impaction (Krassioukov and Weaver, 1995). We
have used this preparation to investigate the
undefined relationships among visceral sensory
afferents, lumbosacral relay neurons, and sympa-
thetic preganglionic neurons that trigger dysre-
flexic hypertension.
A contributing factor underlying autonomic
dysreflexia is the injury-induced elevation in spi-
nal levels of nerve growth factor (Brown et al.,
2004) and the subsequent sprouting of calcitonin
gene-related peptide (CGRP)+ afferent fibers in
the thoracolumbar spinal cord (Krenz and Weav-
er, 1998a; Krenz et al., 1999; Weaver et al., 2001;
Marsh et al., 2002). It is important to note that
CGRP immunoreactivity has been reported in Ab,
Ad, and C fiber afferent projections (McCarthy
and Lawson, 1990; Lawson et al., 1993; Lawson
et al., 1996). Because these CGRP+ fibers also can
be labeled with various neurotransmitter markers,
the identity of specific afferent fiber populations
that sprout after spinal cord injury remains
uncertain. For example, it is reported that sub-
populations of CGRP+ fibers sprouting distal to
spinal cord injury sites are nociceptive primary af-
ferents. This could be the anatomical substrate for
the development and maintenance of chronic pain
syndromes after spinal cord injury (Ondarza et al.,
2003). Conversely, there is evidence for a lack of
significant substance P+ fiber sprouting in parallel
with increased CGRP+ fiber sprouting in a rat
model of autonomic dysreflexia (Marsh and Weav-
er, 2004). Since substance P content can be used to
distinguish subpopulations of unmyelinated C
fibers from myelinated Ad fibers, the authors
suggest that sprouting of Ab and Ad fibers, and
not C fibers, contributed to increased sympathetic
outflow after injury. Accordingly, non-noxious
cutaneous and colonic stimulation below the
injury was also reported to elicit dysreflexic
hypertension, albeit to a lesser degree than with
noxious stimulation.
To investigate further the role of nerve growth
factor and CGRP+ afferent fiber sprouting in the
development of autonomic dysreflexia, we precise-
ly manipulated injury-induced CGRP+ C-fiber
sprouting in the dorsal horns with bilateral
microinjections of a well-characterized replica-
tion-defective, temperature-sensitive recombinant
adenovirus encoding growth-promoting nerve
growth factor (Romero et al., 2001). The princi-
pal goal was to target growth factor overexpres-
sion to selected regions of the dorsal gray matter of
the spinal cord, caudal to a complete T4 transec-
tion, to identify sites instrumental to augmenting
dysreflexic responses to colorectal distension 2
weeks after injury (Maiorov et al., 1998). Once the
critical spinal levels were identified, with subsequent
 267

immunohistochemical analyses for CGRP+ affer-
ent fiber sprouting (Krenz et al., 1999; Weaver
et al., 2001), we used recombinant adenovirus to
overexpress C-fiber growth-inhibitory semaphorin
3A (Tang et al., 2004) after injury in an attempt to
prevent sprouting and mitigate dysreflexic hyper-
tension. Regarding the aforementioned variability
in phenotypes of post-traumatic sprouting afferent
fibers, Tang et al. (2004) showed that overexpres-
sion of nerve growth factor within the spinal cord
directly increased sprouting of CGRP+ axons
without altering neuropeptide expression or
sprouting of other sensory axon populations.
These populations included myelinated and un-
myelinated, glial cell line-derived neurotrophic
factor-responsive subpopulations of nociceptive
axons. Moreover, overexpression of semaphorin
3A, a repulsive guidance molecule for nerve growth
factor-responsive C-fibers, reduced sprouting of
CGRP+ and substance P+ axons compared with
overexpression of control green fluorescent protein.
The sympathetic component of the intermedio-
lateral cell column extends from T1 to L2, and
primary afferent sprouting has been reported both
close to and distant from a T4 transection (Krenz
and Weaver, 1998a; Krenz et al., 1999). We first
tested whether increasing primary afferent sprout-
ing just caudal to at T4 transection would augment
hypertension evoked by colorectal distension
(Cameron et al., 2004). Two weeks after we over-
expressed nerve growth factor in the T5/T6 dorsal
horns (n ¼ 5), there was copious central sprouting
of CGRP+ primary afferent fibers into the
mid-thoracic spinal cord dorsal horns compared
to control injured rats injected with green fluores-
cent protein adenovirus at T5/T6 (n ¼ 3). Notably,
we found that sympathetic preganglionic neurons,
pre-labeled with FluoroGold (Anderson and
Edwards, 1994), were completely surrounded by
sprouting CGRP+ fibers (not shown). Physiolog-
ical recordings of mean arterial blood pressure and
heart rate in response to colorectal distension
showed that evoked hypertension was no greater
compared to injured controls (Fig. 1), even though
nerve growth factor induced robust afferent fiber
sprouting into the intermediolateral cell column.
Since this implied that neither the actions of nerve
growth factor nor sprouting CGRP+ fibers directly
influenced the sympathetic discharge in response to
colorectal distension, we chose to similarly manip-
ulate injury-induced sprouting at more caudal levels,
based on the evidence that sensory input from the
pelvic viscera enter the spinal cord caudal to T5/T6.
Electrophysiological (Al-Chaer et al., 1997) and
anatomical studies (Birder et al., 1991; Keast and
De Groat, 1992; Vizzard, 2000) report that the
primary afferents supplying the pelvic viscera
(bladder, distal colon, rectum) in rats run in the
pelvic nerve and distribute mainly to the L6 and S1
dorsal root ganglia and spinal cord segments. A
smaller percentage of afferents run in the hypogas-
tric nerve to the T13 and L1 dorsal root ganglia and
spinal cord segments, and neurons responsive to
colorectal distension are located in the superficial

Fig. 1. Overexpression of nerve growth factor (NGF) in lumbosacral spinal levels augments CGRP+ afferent fiber sprouting 2 weeks
post-injury. Photomicrographs showing CGRP+ staining in S1 spinal segments injected with adenovirus expressing control green
fluorescent protein (GFP Adts; left column) or NGF Adts (right column). Note the robust CGRP+ afferent fiber sprouting throughout
the S1 injection site in response to NGF overexpression compared to GFP controls. Scale bars ¼ 100 mm.
268
dorsal horns of the T13–L2 spinal segments (Ness
and Gebhart, 1988, 1989). Within the L6/S1 spinal
regions, however, lamina X and V–VIII contain
most of the neurons responding to visceral stim-
ulation (Al-Chaer et al., 1997). These projection
neurons respond physiologically to colorectal dis-
tension (Martinez et al., 1998; Landrum et al.,
2002), as well as to cutaneous stimulation.
Two groups of lumbosacral spinal interneurons
relay visceral information rostrally to supraspinal
targets, each through a separate pathway. One
group of neurons resides primarily in the dorsal
commissure of L6/S1 segments (Willis et al., 1999;
Vizzard et al., 2000) and projects axons in the me-
dial part of the dorsal columns (Hirshberg et al.,
1996; Wang et al., 1999). Most of these neurons
respond physiologically to colorectal distension
and their axons terminate in the medullary gracile
nucleus (Al-Chaer et al., 1996, 1999). The other
group of neurons lies in both the dorsal commis-
sure and lateral parasympathetic nucleus and their
axons project through the ventrolateral white mat-
ter to the thalamus (Ness and Gebhart, 1987).
These neurons are activated by colorectal disten-
sion as well as somatic (cutaneous) stimulation.
Propriospinal neurons connect lamina X/dorsal
commissure at multiple rostro-caudal levels of the
cord (Petkó and Antal, 2000) and injections of
anterograde tract tracers into lamina X/dorsal
commissure in the lumbosacral segments (L6–S2)
labels terminals in lamina X throughout the cord
up to cervical levels (Matsushita, 1998). These
studies show that long-ranging propriospinal neu-
rons exist in the dorsal commissure/deep dorsal
horn of the lumbosacral spinal cord. Such relays
extending the entire length of the cord could po-
tentially send off collaterals after spinal cord in-
jury and influence the activity of sympathetic
preganglionic neurons in the intermediolateral cell
column. Since the number of activated neurons
expressing c-Fos in response to colorectal disten-
sion is much greater after a chronic spinal cord
transection than after an acute injury (Landrum
et al., 2002), it suggests that larger numbers of
lumbosacral propriospinal neurons can be activat-
ed by visceral stimuli following injury. In view of
these findings and our preliminary results, we rea-
soned that autonomic dysreflexia may be mediated
primarily by visceral afferent activation of prop-
riospinal interneurons that project from the
lumbosacral to the thoracic spinal cord.
Therefore, we injected T4 spinal-transected rats
with adenovirus encoding nerve growth factor into
the T13/L1 (n ¼ 8) or L6/S1 (n ¼ 9) segments ver-
sus green fluorescent protein controls (n ¼ 7 per
level) to augment sprouting specifically and, po-
tentially, to augment dysreflexic hypertension
(Cameron et al., 2004). After 2 weeks of nerve
growth factor overexpression within either caudal
spinal level, hypertension induced by colorectal
distension was greater in magnitude (
35 mmHg)
than in green fluorescent protein-injected injured
controls (
20 mmHg) (Fig. 1). Conversely, L6/S1
overexpression of semaphorin 3A, a chemorepul-
sive factor for nerve growth factor-responsive pri-
mary afferent fibers, led to a marked reduction in
evoked hypertension compared to cord-injured
controls (
10 mmHg) (not shown).
Subsequent histological processing of the lum-
bosacral spinal cord confirmed that overexpression
of nerve growth factor in the S1 segment elicited
profuse sprouting of CGRP+ fibers 2 weeks after
T4 cord transection compared to modest sprouting
in response to control green fluorescent protein
overexpression (Fig. 2). In contrast, overexpres-
sion of semaphorin 3A in the S1 segment caused
diminished sprouting compared to controls (not
shown). Linear regression analysis of percent
CGRP+ fiber area covered in the dorsal horns
plotted against reflex-induced hypertensive chang-
es revealed a positive correlation between the ex-
tent of C-fiber sprouting into the lumbosacral
spinal level and the severity of autonomic dysre-
flexia among the three treatment groups (Cameron
et al., 2004). This suggests that, in this model of
autonomic dysreflexia induced by noxious colo-
rectal distension, nerve growth factor-induced
sprouting of primary afferents in the region where
colonic afferents enter the spinal cord causes ab-
normal activation of sympathetic preganglionic
neurons in the thoracolumbar spinal cord via pu-
tative propriospinal pathways (Matsushita, 1998;
Petkó and Antal, 2000).
These results are consistent with a nerve growth
factor-dependent increase in the number of pri-
mary afferent terminals. This may amplify the
 269

Fig. 2. Physiological responses to colorectal distension (CRD) 2 weeks after T4 spinal transection. Illustrative traces of pulsatile
arterial pressure (PAP), mean arterial pressure (MAP), and heart rate (HR) before, during, and after 1 min CRD with rectal balloon
catheter inflation (indicated by arrows) from injured rats with bilateral L6/S1 injections of (A) adenovirus expressing control green
fluorescent protein (GFP Adts) or (B) nerve growth factor (NGF Adts). Note that both injured groups show autonomic dysreflexia
with hypertension accompanied by bradycardia, but the severity of hypertension is almost twofold greater and more prolonged with
nerve growth factor overexpression.

synaptic action of colonic afferents, during colo-
rectal distension, on lumbosacral dorsal horn neu-
rons that project axons rostrally in the gray matter
and dorsal columns (Wang et al., 1999; Willis
et al., 1999) and may connect with, and influence
the activity of, preganglionic sympathetic neurons
in the intermediolateral cell column (Llewellyn-
Smith and Weaver, 2001). Intrathecal delivery of
nerve growth factor antibody following spinal
cord injury reduces the elevated nerve growth fac-
tor protein levels in L6 and S1 dorsal root ganglia
and spinal cord of cord-injured rats (Seki et al.,
2002). It seems likely that nerve growth factor-
mediated plasticity of CGRP+ afferents causes
hyperactivity of lumbosacral propriospinal neu-
rons and, as a result, excitation of sympathetically
correlated interneurons in the intermediolateral cell
column. In support of this, colorectal distension in
acute and chronic spinal-transected rats activates
sympathetically correlated neurons in the T10 seg-
ment, most likely via sacral–thoracic interneurons
(Chau et al., 2000; Krassioukov et al., 2002).
We investigated the potential influence of
intraspinal sprouting of the axons of lumbosacral
270
projection interneurons for activating sympathet-
ically correlated interneurons after complete high
thoracic spinal cord injury. To do this we injected
the anterograde tracer biotinylated dextran amine
(BDA) into the S1 dorsal horn of acute T4 spinal-
transected animals and of uninjured controls
(Cameron et al., 2004). Two weeks later, there
were conspicuously more labeled ipsilateral (and
contralateral) propriospinal projections in the ros-
tral thoracic gray matter in the injured spinal cords
(Fig. 3). In fact, many of these projections were
found in proximity to FluoroGold-labeled sympa-
thetic preganglionic neurons (Fig. 4). Additionally,
we found that transection of the L5 spinal cord
segment alone, or in addition to T4, markedly re-
duced colorectal distension-induced hypertension.
This suggests that relay neurons arising in the L6/
S1 segments are necessary for increased hyperten-
sion elicited by pelvic visceral distension. Moreo-
ver, transection of the T11 cord segment above the
Fig. 3. Injections of BDA tracer into L6/S1 spinal levels label
more ascending projections after T4 transection. Longitudinal
and horizontal sections at thoracolumbar spinal levels of (A)
non-transected or (B) T4-transected rats approximately 5 mm
rostral to 200 nl injections of BDA into the left L6/S1 dorsal
commissure. After 2 weeks to allow tracer transport and verify
physiologically that the injured rats had developed autonomic
dysreflexia, histological processing revealed significantly more
BDA-labeled projections (data not shown) within ipsilateral
gray matter extending to rostral levels of the thoracic spinal
cord. Scale bar ¼ 0.5 mm.
entry of the visceral afferents in the hypogastric
nerve, alone or in addition to T4 transection, pro-
duced a colorectal distension-induced rise in mean
arterial pressure with the same magnitude as T4-
and L5-transected animals (
10 mmHg above
baseline) (Cameron et al., 2004). This finding in-
dicates that the upper lumbar sympathetic neu-
rons, that are known to constrict hindlimb and
visceral blood vessels (Baron et al., 1985; Bahr
et al., 1986), are not sufficient to induce significant
dysreflexic hypertension. However, activation of
the entire sympathetic column after injury does
not appear necessary for eliciting modest hyper-
tension during distension of the pelvic viscera.
Notably, experimental autonomic dysreflexia was
not completely eliminated by impeding CGRP+
fiber sprouting with semaphorin 3A overexpression,
similar to previous findings in which endogenous
nerve growth factor was immunologically neutral-
ized after complete T4 spinal transection
Fig. 4. High magnification, dual immunofluorescent images at
the mid-thoracic spinal level 2 weeks after T4 transection dem-
onstrate close proximity of (A) FluoroGold-labeled sympathet-
ic preganglionic neurons (arrows) in the intermediolateral cell
column and (B) BDA-labeled fibers originating from lumbosa-
cral projection interneurons. Scale bar ¼ 50 mm.
 271

(Krenz et al., 1999; Marsh et al., 2002). Since the
high-affinity nerve growth factor trkA receptor is
located on cholinergic propriospinal neurons in
the deep dorsal horn of the rat spinal cord
(Michael et al., 1997), it is possible that, in addi-
tion to altering primary afferent plasticity, spinal
cord injury provokes both nerve growth factor-
dependent and -independent reorganization of
propriospinal pathways as well. Such changes
may provide a neural substrate for the amplifica-
tion of minor sensory signals entering the spinal
cord, resulting in the synchronous discharge of the
preganglionic sympathetic column.
In our attempts to identify neuronal substrates
for the autonomic hyperactivity after complete
spinal cord injury, we have proposed a model
(Cameron et al., 2004) which involves the sprout-
ing of CGRP+ visceral primary afferent C-fibers
and of the axons of lumbosacral projection
interneurons (Fig. 5). We hypothesize that in-
creased nerve growth factor-mediated sprouting of
sacral nociceptive C-fibers triggered by complete
thoracic spinal cord transection (de Groat et al.,
1990) drives larger numbers of propriospinal pro-
jection neurons located in the dorsal intercom-
missural nucleus and amplifies the information
relayed from colonic afferents during colorectal
distension to lumbosacral dorsal horn neurons.
These, in turn, convey the signal to rostral sym-
pathetically correlated neurons (Chau et al., 2000;
Tang et al., 2003) that activate the sympathetic
preganglionic neurons to elicit hypertension (Fig.
5). Since larger numbers of lumbosacral proprio-
spinal neurons are activated by visceral stimuli in
more chronic stages of spinal cord injury (Landrum
et al., 2002), this implies that acute functional re-
organization occurs, which results in signal ampli-
fication by spinal circuitry that has presumably

Fig. 5. Schematic illustration depicting the etiology of autonomic dysreflexia evoked by pelvic visceral distension and other stimuli.
Following complete spinal cord injury above the T6 level, sympathetic preganglionic neurons (blue) are released from descending
medullo-spinal control (dashed arrow) and autonomic spinal reflexes are rendered hyperactive. Consequently, pelvic visceral sensory
input (yellow) is relayed by propriospinal neurons (green) projecting from the dorsal gray commissure at the lumbosacral level to
sympathetic preganglionic neurons and/or sympathetically correlated interneurons located in the thoracolumbar cord (blue). Post-
traumatic C-fiber sprouting into the lumbosacral cord (yellow) further amplifies the central signals (green) to elicit hypertension,
ultimately causing profound peripheral vasoconstriction of splanchnic, muscle, and cutaneous vascular beds. Subsequent stimulation
of aortic depressor nerve (ADN) and carotid sinus nerve (CSN) baroreceptor afferents of the petrosal ganglion (red) is conveyed to the
nucleus tractus solitarius that elicits bradycardia via activation of the nucleus ambiguus. Note that flushing is likely the result of
inhibition of skin vasoconstrictor preganglionic neurons above the lesion by the baroreceptor reflex, and not due to the activation of
sympathetic preganglionic neurons below the lesion by the dysreflexia-inducing stimulus.
272
been modified as a result of the injury. As a result,
following acute thoracic spinal cord transection,
multiple neuronal networks in the thoracolumbar
cord generate sympathetic activity upon activation
from as yet unidentified neurons in the L6/S1 seg-
ments of the cord. The ascending terminals of
propriospinal neurons may also undergo morpho-
logic changes including the aberrant convergence
and formation of synapses with sympathetically
correlated neurons or sympathetic preganglionic
neurons in the intermediolateral cell column
(Weaver et al., 1997; Llewellyn-Smith and
Weaver, 2001). In various ways, the pelvic
visceral signal may be conveyed to the ultimate
vasoconstrictor post-ganglionic sympathetic
neurons to induce dysreflexia (Fig. 5).
Conclusion
Further studies with gene delivery, and perhaps
pharmacological agents that mimic plasticity-al-
tering agents, are needed to clarify the relative
contribution of changes in primary afferent versus
interneuronal/propriospinal systems to autonomic
dysreflexia. Such approaches may also address in-
traspinal plasticity in other forms of dysfunction
and/or recovery after spinal cord injury, such as
chronic pain and locomotion. These results are the
first demonstration that the remodeling of endog-
enous circuitry, which may underlie the develop-
ment of autonomic dysreflexia, can be altered by
genetic manipulation of axon guidance and inhib-
itory molecules to modulate plasticity-induced au-
tonomic pathophysiology after complete spinal
cord injury.
Acknowledgments
The author thanks Dr. Adrian A. Cameron for
contributions to the text and Drs. David C. Randall
and George M. Smith for critical review. Addi-
tionally, Kainath Durre, Johnna Shipp, Igor
Voskresensky, and Leslie Schwindel are appreci-
ated for expert technical assistance. AGR is sup-
ported by grants from the International Spinal
Research Trust, the Kentucky Spinal Cord and
Head Injury Research Trust, and the NIH/NINDS
(R01 NS-049901-01).
References
Al-Chaer, E.D., Feng, Y. and Willis, W.D. (1999) Comparative
study of viscerosomatic input onto postsynaptic dorsal
column and spinothalamic tract neurons in the primate. J.
Neurophysiol., 82: 1876–1882.
Al-Chaer, E.D., Lawand, N.B., Westlund, K.N. and Willis,
W.D. (1996) Pelvic visceral input into the nucleus gracilis is
largely mediated by the postsynaptic dorsal column pathway.
J. Neurophysiol., 76: 2675–2690.
Al-Chaer, E.D., Westlund, K.N. and Willis, W.D. (1997) Sen-
sitization of postsynaptic dorsal column neuronal responses
by colon inflammation. Neuroreport, 8: 3267–3273.
Anderson, C.R. and Edwards, S.L. (1994) Intraperitoneal in-
jections of Fluorogold reliably labels all sympathetic pre-
ganglionic neurons in the rat. J. Neurosci. Meth., 53:
137–141.
Bahr, R., Bartel, B., Blumberg, H. and Janig, W. (1986)
Functional characterization of preganglionic neurons
projecting in the lumbar splanchnic nerves: vasoconstrictor
neurons. J. Auton. Nerv. Syst., 15: 131–140.
Baron, R., Janig, W. and McLachlan, E.M. (1985) The afferent
and sympathetic components of the lumbar spinal outflow to
the colon and pelvic organs in the cat. III. The colonic nerves,
incorporating an analysis of all components of the lumbar
prevertebral outflow. J. Comp. Neurol., 238: 158–168.
Birder, L.A., Roppolo, J.R., Iadarola, M.J. and de Groat, W.C.
(1991) Electrical stimulation of visceral afferent pathways in
the pelvic nerve increases c-fos in the rat lumbosacral spinal
cord. Neurosci. Lett., 129: 193–196.
Brown, A., Ricci, M.J. and Weaver, L.C. (2004) NGF message
and protein distribution in the injured rat spinal cord. Exp.
Neurol., 188: 115–127.
Cameron, A.A., Smith, G.M., Randall, D.C., Brown, D.R. and
Rabchevsky, A.G. (2004) Genetic manipulation of afferent
fiber sprouting following spinal cord injury modulates the
severity of autonomic dysreflexia. J. Neurotrauma, 21: 1271.
Chau, D., Johns, D.G. and Schramm, L.P. (2000) Ongoing and
stimulus-evoked activity of sympathetically correlated neu-
rons in the intermediate zone and dorsal horn of acutely
spinalized rats. J. Neurophysiol., 83: 2699–2707.
de Groat, W.C., Kawatani, M., Hisamitsu, T., Cheng, C.L.,
Ma, C.P., Thor, K., Steers, W. and Roppolo, J.R. (1990)
Mechanisms underlying the recovery of urinary bladder
function following spinal cord injury. J. Auton. Nervous
System, 30(Suppl): S71–S77.
Finestone, H.M. and Teasell, R.W. (1993) Autonomic
dysreflexia after brainstem tumor resection. A case report.
Am. J. Phys. Med. Rehabil., 72: 395–397.
Harati, Y. (1997) Autonomic disorders associated with spinal
cord injury. In: Low P.A. (Ed.), Clinical Autonomic Disor-
ders (2nd ed). Lippincott-Raven, Philadelphia, pp. 455–461.
Hirshberg, R.M., Al-Chaer, E.D., Lawand, N.B., Westlund,
K.N. and Willis, W.D. (1996) Is there a pathway in the pos-
terior funiculus that signals visceral pain? Pain, 67: 291–305.
Karlsson, A.K. (1999) Autonomic dysreflexia. Spinal Cord, 37:
383–391.

Keast, J.R. and De Groat, W.C. (1992) Segmental distribution
and peptide content of primary afferent neurons innervating
the urogenital organs and colon of male rats. J. Comp.
Neurol., 319: 615–623.
Klimaschewski, L. (2001) Increased innervation of rat pre-
ganglionic sympathetic neurons by substance P containing
nerve fibers in response to spinal cord injury. Neurosci. Lett.,
307: 73–76.
Krassioukov, A.V., Johns, D.G. and Schramm, L.P. (2002)
Sensitivity of sympathetically correlated spinal interneurons,
renal sympathetic nerve activity, and arterial pressure to
somatic and visceral stimuli after chronic spinal injury.
J. Neurotrauma, 19: 1521–1529.
Krassioukov, A.V. and Weaver, L.C. (1995) Episodic hyper-
tension due to autonomic dysreflexia in acute and chronic
spinal cord-injured rats. Am. J. Physiol., 268: H2077–H2083.
Krenz, N.R., Meakin, S.O., Krassioukov, A.V. and Weaver,
L.C. (1999) Neutralizing intraspinal nerve growth factor
blocks autonomic dysreflexia caused by spinal cord injury. J.
Neurosci., 19: 7405–7414.
Krenz, N.R. and Weaver, L.C. (1998a) Sprouting of primary
afferent fibers after spinal cord transection in the rat.
Neuroscience, 85: 443–458.
Krenz, N.R. and Weaver, L.C. (1998b) Changes in the mor-
phology of sympathetic preganglionic neurons parallel the
development of autonomic dysreflexia after spinal cord injury
in rats. Neurosci. Lett., 243: 61–64.
Landrum, L.M., Jones, S.L. and Blair, R.W. (2002) The ex-
pression of Fos-labeled spinal neurons in response to colo-
rectal distension is enhanced after chronic spinal cord
transection in the rat. Neuroscience, 110: 569–578.
Lawson, S.N., McCarthy, P.W. and Prabhakar, E. (1996)
Electrophysiological properties of neurones with CGRP-
like immunoreactivity in rat dorsal root ganglia. J. Comp.
Neurol., 365: 355–366.
Lawson, S.N., Perry, M.J., Prabhakar, E. and McCarthy,
P.W. (1993) Primary sensory neurones: neurofilament,
neuropeptides, and conduction velocity. Brain Res. Bull.,
30: 239–243.
Lindan, R., Joiner, E., Freehafer, A.A. and Hazel, C. (1980)
Incidence and clinical features of autonomic dysreflexia in
patients with spinal cord injury. Paraplegia, 18: 285–292.
Llewellyn-Smith, I.J. and Weaver, L.C. (2001) Changes in
synaptic inputs to sympathetic preganglionic neurons after
spinal cord injury. J. Comp. Neurol., 435: 226–240.
Maiorov, D.N., Fehlings, M.G. and Krassioukov, A.V. (1998)
Relationship between severity of spinal cord injury and ab-
normalities in neurogenic cardiovascular control in conscious
rats. J. Neurotrauma, 15: 365–374.
Marsh, D.R. and Weaver, L.C. (2004) Autonomic dysreflexia,
induced by noxious or innocuous stimulation, does not de-
pend on changes in dorsal horn substance p. J. Neurotrauma,
21: 817–828.
Marsh, D.R., Wong, S.T., Meakin, S.O., MacDonald, J.I.,
Hamilton, E.F. and Weaver, L.C. (2002) Neutralizing
intraspinal nerve growth factor with a trkA-IgG fusion pro-
tein blocks the development of autonomic dysreflexia in a
273
clip-compression model of spinal cord injury. J. Neurotrau-
ma, 19: 1531–1541.
Martinez, V., Wang, L., Mayer, E. and Tache, Y. (1998)
Proximal colon distention increases Fos expression in the
lumbosacral spinal cord and activates sacral parasympathetic
NADPHd-positive neurons in rats. J. Comp. Neurol., 390:
311–321.
Matsushita, M. (1998) Ascending propriospinal afferents to
area X (substantia grisea centralis) of the spinal cord in the
rat. Exp. Brain Res., 119: 356–366.
McCarthy, P.W. and Lawson, S.N. (1990) Cell type and
conduction velocity of rat primary sensory neurons with
calcitonin gene-related peptide-like immunoreactivity.
Neuroscience, 34: 623–632.
Michael, G.J., Kaya, E., Averill, S., Rattray, M., Clary, D.O.
and Priestley, J.V. (1997) TrkA immunoreactive neurones in
the rat spinal cord. J. Comp. Neurol., 385: 441–455.
Ness, T.J. and Gebhart, G.F. (1987) Characterization of neu-
ronal responses to noxious visceral and somatic stimuli in the
medial lumbosacral spinal cord of the rat. J. Neurophysiol.,
57: 1867–1892.
Ness, T.J. and Gebhart, G.F. (1988) Characterization of neu-
rons responsive to noxious colorectal distension in the T13-
L2 spinal cord of the rat. J. Neurophysiol., 60: 1419–1438.
Ness, T.J. and Gebhart, G.F. (1989) Characterization of
superficial T13-L2 dorsal horn neurons encoding for
colorectal distension in the rat: comparison with neurons in
deep laminae. Brain Res., 46: 301–309.
Ondarza, A.B., Ye, Z. and Hulsebosch, C.E. (2003) Direct ev-
idence of primary afferent sprouting in distant segments fol-
lowing spinal cord injury in the rat: colocalization of GAP-43
and CGRP. Exp. Neurol., 184: 373–380.
Petkó, M. and Antal, M. (2000) Propriospinal afferent and
efferent connections of the lateral and medial areas of the
dorsal horn (laminae I–IV) in the rat lumbar spinal cord.
J. Comp. Neurol., 422: 312–325.
Romero, M.I., Rangappa, N., Garry, M.G. and Smith, G.M.
(2001) Functional regeneration of chronically injured sensory
afferents into adult spinal cord after neurotrophin gene ther-
apy. J. Neurosci., 21: 8408–8416.
Seki, S., Sasaki, K., Fraser, M.O., Igawa, Y., Nishizawa, O.,
Chancellor, M.B., de Groat, W.C. and Yoshimura, N. (2002)
Immunoneutralization of nerve growth factor in lumbosacral
spinal cord reduces bladder hyperreflexia in spinal cord
injured rats. J. Urol., 168: 2269–2274.
Snow, J.C., Sideropoulos, H.P., Kripke, B.J., Freed, M.M.,
Shah, N.K. and Schlesinger, R.M. (1978) Autonomic hype-
rreflexia during cystoscopy in patients with high spinal cord
injuries. Paraplegia, 15: 327–332.
Tang, X., Neckel, N.D. and Schramm, L.P. (2003) Locations
and morphologies of sympathetically correlated neurons
in the T(10) spinal segment of the rat. Brain Res., 976:
185–193.
Tang, X.Q., Tanelian, D.L. and Smith, G.M. (2004)
Semaphorin3A inhibits nerve growth factor-induced sprout-
ing of nociceptive afferents in adult rat spinal cord.
J. Neurosci., 24: 819–827.
274
Vizzard, M.A. (2000) Increased expression of spinal cord
Fos protein induced by bladder stimulation after spinal cord
injury. Am. J. Physiol. Regul. Integr. Comp. Physiol., 279:
R295–R305.
Vizzard, M.A., Brisson, M. and de Groat, W.C. (2000)
Transneuronal labeling of neurons in the adult rat central
nervous system following inoculation of pseudorabies virus
into the colon. Cell Tissue Res., 299: 9–26.
Wang, C.C., Willis, W.D. and Westlund, K.N. (1999) Ascend-
ing projections from the area around the spinal cord central
canal: a Phaseolus vulgaris leucoagglutinin study in rats. J.
Comp. Neurol., 415: 341–367.
Weaver, L.C., Cassam, A.K., Krassioukov, A.V. and
Llewellyn-Smith, I.J. (1997) Changes in immunoreactivity
for growth associated protein-43 suggest reorganization of
synapses on spinal sympathetic neurons after cord transec-
tion. Neuroscience, 81: 535–551.
Weaver, L.C., Verghese, P., Bruce, J.C., Fehlings, M.G., Krenz,
N.R. and Marsh, D.R. (2001) Autonomic dysreflexia and
primary afferent sprouting after clip-compression injury of
the rat spinal cord. J. Neurotrauma, 18: 1107–1119.
Willis, W.D., Al-Chaer, E.D., Quast, M.J. and Westlund, K.N.
(1999) A visceral pain pathway in the dorsal column of the
spinal cord. Proc. Natl. Acad. Sci. USA, 96: 7675–7679.
Zagon, A. and Smith, A.D. (1993) Monosynaptic projections
from the rostral ventrolateral medulla oblongata to identified
sympathetic preganglionic neurons. Neuroscience, 54:
729–743.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 18

Spinal cord injury alters cardiac electrophysiology

and increases the susceptibility to
ventricular arrhythmias

Heidi L. Collins1, David W. Rodenbaugh2 and Stephen E. DiCarlo1,

1
Department of Physiology, Wayne State University School of Medicine, Detroit, MI 48201, USA
2
Department of Molecular and Integrative Physiology, University of Michigan, Ann Arbor, MI 48109, USA

Abstract: The autonomic nervous system modulates cardiac electrophysiology and abnormalities of au-
tonomic function are known to increase the risk of ventricular arrhythmias. The abnormal and unstable
autonomic control of the cardiovascular system following spinal cord injury also is well known. For
example, individuals with mid-thoracic spinal cord injury have elevated resting heart rates, increased blood
pressure variability, episodic bouts of life-threatening hypertension as part of a condition termed auto-
nomic dysreflexia, and elevated sympathetic activity above the level of the lesion. Furthermore, cardio-
vascular morbidity and mortality are high in individuals with spinal cord injuries due to a relatively
sedentary lifestyle and higher prevalence of other cardiovascular risk factors, including obesity and di-
abetes. Therefore, spinal cord injury may alter cardiac electrophysiology and increase the risk for ven-
tricular arrhythmias. In this chapter, we discuss how the autonomic changes associated with cord injury can
influence cardiac electrophysiology and the susceptibility to ventricular arrhythmias.

Individuals with spinal cord injury in the cervical
and upper thoracic (T) segments (T1–T6) have
unstable arterial pressure and heart rate. In addi-
tion to other factors, the unstable arterial pressure
and heart rate after spinal cord injury enhances the
risk for blood pressure-related cardiovascular dis-
ease above levels seen in hypertensive subjects
(Rodenbaugh et al., 2003a). Thus, it is not sur-
prising that cardiovascular disease is a leading
cause of morbidity and mortality for individuals
with spinal cord injury. In this chapter, we will
discuss the impact of spinal cord injury on the
autonomic control of the circulation and its rela-
tionship to increased risk for cardiac arrhythmias.
An estimated 2 million individuals worldwide
live with spinal cord injuries. On average, 11,000
Corresponding author. Tel.: +(313) 577-1557;
Fax: +(313) 577-5494; E-mail: sdicarlo@med.wayne.edu
new injuries are reported each year in the United
States alone. Spinal cord injuries cost the United
States at least $9.7 billion per year for medical
care, equipment and disability support (DeVivo,
1997; Berkowitz et al., 1998). With the advances in
acute care and rehabilitation, the life expectancy of
individuals with spinal cord injury has increased to
near that for able-bodied individuals. However,
cardiovascular disease is now a leading cause of
death and morbidity for individuals with spinal
cord injury (Le and Price, 1982; Wicks et al.,
1983; Cardiovascular–Cardiopulmonary Second-
ary Disabilities, 1991; DeVivo et al., 1992, 1993;
Whiteneck et al., 1992; Frankel et al., 1998; Soden
et al., 2000). Cardiovascular morbidity and mor-
tality are high for individuals with spinal cord in-
juries, presumably due to a relatively sedentary
lifestyle and higher prevalence of other cardiovas-
cular risk factors, including obesity and diabetes

DOI: 10.1016/S0079-6123(05)52018-1 275

276

(Duckworth et al., 1980; Levi et al., 1995; Rodenbaugh et al., 2003a, b, c). This phenomenon
Karlsson, 1999). In fact, individuals with spinal is illustrated in Fig. 1 that shows the heart rate for
cord injury are at the lowest end of the human intact and cord-injured rats, recorded by radio-
fitness spectrum (Dearwater et al., 1986; Washburn telemetry, averaged over each week for 7 weeks
and Figoni, 1998). Similarly, cord-injured rats have (panel A) or for the entire 7 weeks of study (panel
lower levels of physical activity than healthy intact B). Animals with cord transection at T5 had
rats (Rodenbaugh et al., 2003a). Additionally, in- increased heart rates each week during the entire 7
dividuals with cord injury have blood lipid profiles weeks of study (panel A). Specifically, the average
characterized by elevated total cholesterol and heart rate was 19% higher in cord-injured rats
low-density lipoprotein cholesterol, and depressed during the entire 7 weeks of study (panel B).
high-density lipoprotein cholesterol, a lipid profile Numerous studies have documented that an
normally associated with, if not a direct result of, elevated heart rate is a strong risk factor for the
the sedentary lifestyle (Bauman et al., 1992, 1999; development of cardiovascular disease (Palatini
Cardus et al., 1992). There is a strong association and Julius, 1997, 1999; Julius et al., 1998). The
between low physical activity, increased sympa- elevated heart rates observed in individuals with
thetic nervous system activity, and cardiovascular mid-thoracic cord injury may be mediated by a
disease risk factors (Zimmet et al., 1991; Collins reflex compensatory mechanism to offset the de-
et al., 2000). Importantly, increased physical ac- creased stroke volume (Hjeltnes, 1977; Hopman
tivity lowers sympathetic nerve activity (DiCarlo et al., 1993a). The lower stroke volume has been
et al., 1997) and cardiovascular disease risk factors
(Jennings et al., 1984; Collins et al., 2000; DiCarlo
et al., 2002). Therefore, exercise with the arms is
often recommended for individuals with cord in-
jury, based on studies demonstrating improve-
ments in aerobic capacity and lipoprotein profiles
(DiCarlo, 1982, 1988; DiCarlo et al., 1983; Glaser,
1985; Hoffman, 1986; Hooker and Wells, 1989).
In fact, the Center for Disease Control has rec-
ommended further research to evaluate the effica-
cy of exercise to prevent the development of
cardiovascular disease in individuals with cord
injury (Cardiovascular–Cardiopulmonary Second-
ary Disabilities, 1991). It is interesting to speculate
that one mechanism contributing to the improved
cardiopulmonary status after exercise in indi-
viduals with spinal cord injury (DiCarlo, 1982;
DiCarlo et al., 1983) may be a reduced incidence
and/or severity of cardiac arrhythmias.
Cord-injured individuals have distinct hemody-
namic responses to a variety of activities of daily
living. For example, it is well established that Fig. 1. Average heart rate for intact and cord-injured rats, re-
individuals with mid-thoracic cord injury have corded by radio-telemetry, each week for 7 weeks (panel A) and
elevated heart rates and lower stroke volumes at the heart rates averaged during the entire 7 weeks of the study
rest and during activity than able-bodied indi- (panel B). Heart rate was higher in the cord-injured rats at every
time point during the 7 weeks of the study. Specifically, heart
viduals (Jacobs et al., 2002). Similarly, rats with
rate was 19% higher in cord-injured rats during the entire 7
cord injury at the 5th thoracic segment (T5) have weeks of the study (panel B). In this and subsequent figures,
elevated heart rates (Krassioukov and Weaver, variability is indicated by the standard error of the mean
1995; Maiorov et al., 1997; Mayorov et al., 2001;

Pp0:05:

attributed to reduced venous return from the in-
active regions below the level of the injury due to
loss of supraspinal control of sympathetically me-
diated veno- and vasoconstriction (Kinzer and
Convertino, 1989; Hopman et al., 1993b). The loss
of supraspinal control of sympathetically mediated
veno- and vasoconstriction impairs the ability to
effectively distribute blood throughout the vascu-
lar system and results in venous pooling. Preven-
tion of venous pooling with an external ‘‘muscle
pump’’ using transcutaneous electrical stimulation
markedly enhances stroke volume in cord-injured
individuals (Davis et al., 1990). Cardiac output,
the volume of blood pumped by the heart each
minute, is the product of heart rate and stroke
volume and is lower in cord-injured people (Jacobs
et al., 2002).
In able-bodied individuals the average heart rate
at rest is 60–80 beats per minute. As noted above,
heart rate is significantly higher in individuals with
mid-thoracic cord injury. During emotional ex-
citement or muscular activity, heart rate increases
reaching values as high as 220 beats per minute
minus the age of the individual. The maximum
heart rate is not altered in individuals with
mid-thoracic cord injury (Jacobs et al., 2002). The
increase in heart rate is mediated, mainly, by chang-
es in the autonomic nervous system. The process
of excitation of the heart originates in the sino-
atrial node, the intrinsic pacemaker of the heart,
located in the right atrium. These cells depolarize
spontaneously and generate action potentials at an
intrinsic rate of about 100 per minute (Jose, 1966;
Katona et al., 1982). The sinoatrial node is under
the direct influence of the sympathetic and para-
sympathetic divisions of the autonomic nervous
system. In general, the sympathetic division of the
autonomic nervous system increases heart rate by
increasing the rate at which the sinoatrial node
generates action potentials, whereas the parasym-
pathetic division of the autonomic nervous system
decreases heart rate by decreasing the rate of ac-
tion potential generation by the sinoatrial node.
Usually, changes in heart rate involve the recipro-
cal actions of the two divisions of the autonomic
nervous system. Specifically, an increased heart
rate is mediated by a reduction in parasympathetic
activity and a concomitant increase in sympathetic
277
activity whereas a decreased heart rate is mediated
by an increase in parasympathetic activity and a
concomitant decrease in sympathetic activity.
Overactivity, or dominant activity in the cardiac
parasympathetic nerves can produce or contribute
to a variety of brady-arrhythmias whereas over-
activity or dominant activity of the cardiac sym-
pathetic nerves can produce or contribute to a
variety of tachy-arrhythmias (Katz, 1992). In rest-
ing able-bodied individuals parasympathetic activ-
ity dominates. Thus, when both divisions of the
autonomic nervous system are blocked (by drugs
such as atropine for parasympathetic and prop-
ranolol for sympathetic) in resting able-bodied
individuals, the heart rate increases to approxi-
mately 100 beats per minute (Jose, 1966; Katona
et al., 1982). This increase in heart rate following
complete autonomic nervous system blockade doc-
uments the dominance of the parasympathetic sys-
tem on resting heart rate. The prevailing heart rate
after block of both sympathetic and parasympa-
thetic cardiac influences is called the intrinsic heart
rate. Importantly, because the cardiac parasym-
pathetic fibers never pass through the spinal cord,
spinal cord injury does not interrupt cardiac para-
sympathetic activity. Cardiac parasympathetic
fibers originate in the dorsal motor nucleus of
the vagus and the nucleus ambiguus in the medulla
oblongata (Loewy and Spyer, 1990) and travel in
the vagus nerve to the heart. Subsequently, these
fibers synapse with postganglionic cells on the
epicardial surface or within the walls of the heart
near the sinoatrial node and atrioventricular node.
In contrast, spinal cord injury above the 1st
thoracic segment (T1) disrupts central control of
preganglionic cardiac sympathetic activity. Specif-
ically, preganglionic cardiac sympathetic fibers
originate in the intermediolateral cell column of
the upper five or six thoracic segments of the spinal
cord (Bonica, 1968). These fibers exit the spinal
cord through the white communicating rami and
enter the paravertebral chains of ganglia. The
cardiac preganglionic and postganglionic neurons
synapse in the middle cervical and upper thoracic
paravertebral ganglia, especially the stellate
ganglion (Bonica, 1968). Cervical and upper tho-
racic spinal cord injury disrupts the interaction
of the cardiac sympathetic and parasympathetic
278
innervation. Disruption of this balance may have a
profound influence on cardiac rate, performance
and rhythm. For example, the higher heart rates
observed in individuals and animals with mid-
thoracic cord injury suggest higher cardiac sym-
pathetic activity or lower cardiac parasympathetic
activity. To test this hypothesis we measured heart
rate in intact rats and rats with cord injury
at T5 under normal resting conditions and
during ganglionic blockade achieved by intrave-
nous administration of hexamethonium chloride.
Hexamethonium chloride abolishes postganglionic
sympathetic and parasympathetic nervous system
activity and reduces arterial pressure and heart
rate slightly in intact rats and more substantially in
cord-injured rats. Resting heart rate was signifi-
cantly higher in injured rats and decreased signif-
icantly more in response to ganglionic blockade.
Figure 2 presents heart rate before and after
ganglionic blockade in intact and cord-injured
rats. Ganglionic blockade reduced heart rate more
in injured rats compared with intact rats (36% vs.
9%, respectively). These findings are consistent
with the idea that an increase in sympathetic ac-
tivity above the level of the lesion in cord-injured
rats contributes to a greater tonic sympathetic
support of heart rate.
Elevations in cardiac sympathetic activity or
reductions in cardiac parasympathetic activity
Fig. 2. Heart rate before (control) and after ganglionic block-
ade (Gag-X) in intact and cord-injured rats. Ganglionic block-
ade (equivalent to elimination of all neural input to the heart)
reduced heart rate more in cord-injured rats than in intact rats
(36% vs. 9%, respectively).
increase the risk for the development of ventricu-
lar arrhythmias and sudden cardiac death
(Schwartz and Stone, 1980; Schwartz et al.,
1993). Therefore, the elevated heart rates and in-
creased cardiac sympathetic activity in individuals
with mid-thoracic cord injury (Karlsson et al.,
1998; Rodenbaugh et al., 2003b) may result in an
increased mortality associated with changes in
cardiac electrophysiology and the incidence of
arrhythmias. In fact, it has recently been docu-
mented that rats with cord transection at the 5th
thoracic segment have a lower electrical stimula-
tion threshold to induce cardiac arrhythmias
(Rodenbaugh et al., 2003b, c).
Arterial baroreceptors, located in the aortic arch
and carotid sinuses, have a profound influence on
cardiac rate, performance, and rhythm by reflexly
altering cardiac sympathetic and parasympathetic
activity. The arterial baroreceptors respond to a
reduction in arterial pressure by decreasing activity
of the parasympathetic nerves and increasing
activity of the sympathetic nerves. This imbalance
of the autonomic nervous system in favor of sym-
pathetic dominance can lead to tachy-arrhythmias.
In contrast, the arterial baroreceptors respond to
an increase in arterial pressure by increasing
activity of the parasympathetic nerves and
decreasing activity of the sympathetic nerves. This
imbalance favoring parasympathetic dominance
can lead to brady-arrhythmias. Arterial pressure is
usually lower and unstable after cervical and upper
thoracic spinal cord injury. Hypotension occurs
immediately after the injury because of loss of
tonic supraspinal excitatory drive to spinal
sympathetic neurons (Calaresu and Yardley,
1988). Subsequently, resting arterial pressure
returns toward normal. However, episodic hyper-
tension often develops as part of the condition
termed autonomic dysreflexia (Naftchi, 1990;
Mathias and Frankel, 1992). In addition, activi-
ties of daily living produce large variations of
blood pressure in individuals with cord injury
(Stiens et al., 1995; Faghri et al., 2001). This is due,
in part, to the fact that arterial baroreflex control
of the sympathetic nervous system is lost below the
level of the lesion, resulting in reduced buffering of
changes in arterial pressure. Thus, individuals and
animals with spinal cord injury have increased
 279

blood pressure variability after mid-thoracic cord
injury (Rodenbaugh et al., 2003a). Increased blood
pressure variability and heart rate significantly in-
crease the risk of cardiovascular diseases (Rizzoni
et al., 1992; Stevenson et al., 1997; Julius and
Valentini, 1998). These risk factors are highly cor-
related with end organ damage, as well as vascular
structure changes and an increased incidence of
myocardial infarction, stroke, and cardiac ar-
rhythmias (Frattola et al., 1993; Stamler et al.,
1993; Palatini and Julius, 1997, 1999). The in-
creased blood pressure variability mediates chang-
es in arterial baroreceptor activity that reflexly
alters cardiac sympathetic and parasympathetic
activity and profoundly affects cardiac rate, per-
formance, and rhythm. Importantly, cord-injured
rats have increased blood pressure-related cardio-
vascular disease risk factors (Rodenbaugh et al.,
2003a). For example, Fig. 3 presents the average
systolic blood pressure standard deviation (SBP-
SD) for intact and cord-injured rats each week for
7 weeks (panel A) and SBP-SD averaged over the
entire duration of the study (panel C). Mid-
thoracic cord injury significantly increased SBP-
SD each week for the entire 7 weeks of the study.
Specifically, SBP-SD was 22% higher in the in-
jured versus intact rats over the entire 7 weeks
(panel C). Similarly, the average diastolic blood
pressure standard deviation (DBP-SD) for intact
and cord-injured rats was examined (panels B and
D) and cord injury increased DBP-SD during the
entire 7 weeks of the study. Specifically, DBP-SD
was 13% higher in the injured versus intact rats
during the entire 7 weeks (panel D). Taken to-
gether, these data suggest that the loss of arterial
baroreflex control of the autonomic nervous sys-
tem below the level of the lesion increases arterial
blood pressure variability.
As noted above, cardiac tachy-arrhythmias are
associated with elevations of cardiac sympathetic
efferent activity (Schwartz and Stone, 1980, 1982).
Cardiac a- and b-adrenergic receptors mediate the
response to cardiac sympathetic stimulation. b-
Adrenergic receptor blockade, which reduces the

Fig. 3. Systolic blood pressure standard deviation (SBP-SD) and diastolic blood pressure deviation (DBP-SD) for intact and cord-
injured rats, averaged each week for 7 weeks (panels A, B) and averaged over the entire duration of the study (panels C, D). Spinal cord
injury significantly increased SBP-SD and DBP-SD. Specifically, SBP-SD was 22% higher in the injured versus intact rats over the
entire 7 weeks (panel C) and DBP-SD was 13% higher in the injured vs. intact rats over the entire 7 weeks (panel D)
Pp0:05:
280
effect of cardiac sympathetic efferent activity, has
been shown to reduce cardiovascular mortality
(Frishman, 1992). The sympathetic nervous system
can affect the generation and conduction of action
potentials in the heart by activating mainly b-ad-
renergic receptors (Berne and Levy, 2001). b-Ad-
renergic receptor stimulation, which increases
intracellular cAMP, increases heart rate, atrioven-
tricular nodal conduction and contractile force,
and shortens atrial and ventricular refractoriness
(Murray and Roden, 1996). In addition, it en-
hances the plateau phase of the action potential by
increasing current through L-type Ca2+ channels
while repolarization is accelerated due to an in-
crease in both the delayed rectifier current and the
chloride current (Murray and Roden, 1996). Thus,
b-adrenergic receptor stimulation may shorten or
prolong action potential duration depending on
which of the currents predominate. b-Adrenergic
receptor stimulation also causes more rapid pace-
maker activity in the sinus node by its action on
diastolic currents (Murray and Roden, 1996). a-
Adrenergic receptor stimulation enhances cardiac
contractility due to Ca2+ influx (Berne and Levy,
2001). Furthermore, adrenergic stimulation en-
hances the development of after-depolarizations
that lead to electrocardiogram (ECG) complexes
termed triggered beats (Katz, 1992). In this situ-
ation, multiple ionic mechanisms are involved, and
elevated intracellular calcium is a common feature.
Adrenergic stimulation results in a reduction of the
electrical stimulus threshold to induce ventricular
fibrillation as well as an increase in the likelihood
of spontaneous ventricular arrhythmias. b-Ad-
renergic receptor blockade and enhanced para-
sympathetic tone inhibit these effects and are
known to be protective against ventricular tachy-
arrhythmias and sudden death (Engel, 1978; Zipes,
1991; Wharton et al., 1992; Schwartz et al., 1993).
Thus, interventions that reduce sympathetic activ-
ity or enhance parasympathetic activity may be
useful in preventing deadly tachy-arrhythmias.
Spinal cord injury, depending on the specific site
of the lesion, may affect sympathetic and para-
sympathetic activity in a manner that promotes
arrhythmias.
In addition to transmembrane signals that alter
ionic currents over the time course of seconds,
adrenergic signals alter cardiac electrophysiology
over the time course of hours to days by altering
transcription of genes encoding ion channels and
calcium-regulating proteins. For example, suscep-
tibility to cardiac arrhythmias may increase by al-
tering the expression of specific genes encoding
proteins critical to myocyte calcium homeostasis
[the sarco(endo)plasmic reticulum Ca2+ ATPase
(SERCA), the Na+/Ca2+ exchanger encoded by
the NCX1 gene, and the L-type calcium channel]
in a manner that will favor increasing intracellular
Ca2+ (Rodenbaugh et al., 2003c). The ability of
cardiac myocytes to maintain cytosolic Ca2+ with-
in a tightly controlled range is important to pre-
vent cardiac electrical disturbances. Disturbances
in Ca2+ homeostasis are closely related to cardiac
electrophysiological events. The Na+/Ca2+ ex-
changer and the L-type calcium channel provide
the primary calcium efflux and influx pathways,
respectively. SERCA is the predominant pathway
through which cytosolic calcium is sequestered.
Signaling through a- and b-adrenergic receptors
modulates calcium homeostasis by altering the ex-
pression of these calcium regulatory proteins. For
example, signaling through a- and b-adrenergic
receptors coordinately regulates the expression of
the a 1C-subunit of the L-type calcium channel
and the Na+/Ca2+ exchanger by activating pro-
tein kinase A and protein kinase C pathways
(Golden et al., 2000, 2002). Specifically, b-ad-
renergic receptor activation in vitro increases the
expression of the L-type Ca2+ channel and NCX1
genes, whereas a-adrenergic signaling reduces their
mRNA levels (Golden et al., 2000, 2001, 2002).
Several facts have been well documented. The
autonomic nervous system modulates cardiac
electrophysiology and abnormalities of autonom-
ic function can increase the risk of cardiac
arrhythmias. Furthermore, autonomic control of
the cardiovascular system is abnormal and unsta-
ble following spinal cord injury. For example, in-
dividuals with spinal cord injury have elevated
heart rates, increased blood pressure variability,
elevated sympathetic activity above the level of the
lesion, and episodic bouts of life-threatening hy-
pertension as part of a condition termed auto-
nomic dysreflexia. Autonomic dysreflexia occurs in
as many as 85% of individuals with spinal cord
 281

injuries above the 6th thoracic segment and is

characterized by severe hypertension (Lee et al.,
1995). If not prevented or treated promptly, the
hypertension may produce cerebral and sub-
arachnoid hemorrhage, seizures, renal failure, car-
diac arrhythmias, and may lead to death (McGuire
and Kumar, 1986). An example of autonomic
dysreflexia in rats is presented in Fig. 4, illustrating
an analog recording of arterial pressure, ECG, and
heart rate responses to colon distension in a quad-
riplegic rat (Collins and DiCarlo, 2002). Colon
distension produced an increase in arterial pres-
sure, bradycardia, and arrhythmias. This observa-
tion demonstrates that abnormal control of the
cardiovascular system after spinal cord injury can
increase the susceptibility to cardiac arrhythmias.
Autonomic dysreflexia is the second most com-
mon long-term secondary medical complication
associated with cord injury and thus is a major
health concern (McKinley et al., 1999). In fact,
autonomic dysreflexia is the most prominent life-
threatening situation for individuals with spinal
cord injury (Comarr and Eltorai, 1997). The long-
term consequence of repeated episodes of severe
hypertension has yet to be determined, however, it
is well documented that increased blood pressure
variability is a significant cardiovascular disease
risk factor (Frattola et al., 1993; Collins et al.,
2000). Thus, the incidence of cardiac arrhythmias
may be increased in individuals with spinal cord
injury due to the unstable arterial blood pressure.
In support of this concept, there are a large
number of case reports documenting cardiac
arrhythmias in individuals with spinal cord injury
(Guttmann and Whitteridge, 1947; Frankel et al.,
1975; Colachis and Clinchot, 1997). For example,
as early as 1947, Guttmann and Whitteridge re-
ported premature atrial and ventricular contrac-
tions as well as changes in the amplitude of the
T-waves in individuals with spinal cord injuries Fig. 4. Analog recording of arterial pressure, ECG, and heart
during urodynamic testing. Arrhythmias have also rate response to colon distension (50 mmHg for 1 min) in one
quadriplegic rat. Dashed line indicates the onset of colon dis-
been observed in a woman with a spinal lesion at
tension. Colon distension produced pressor and bradycardic
T3 during labor (Guttmann et al., 1965). Prema- responses. Of particular interest are the arrhythmias produced
ture ventricular contractions and atrio-ventricular by colon distension (insert). Reprinted with permission from
dissociation are common when individuals with spi- Collins and DiCarlo (2002).
nal cord injuries have distended bladders (Guttmann
and Whitteridge, 1947) and asystole as well as
other arrhythmias have been reported during
282
tracheal suctioning (Frankel et al., 1975) and
autonomic dysreflexia. Profound bradycardia is a
common complication in the early post-traumatic
period following cervical spinal cord damage. It is
thought to be due to temporary inactivity of the
sympathetic nervous system after separation from
supraspinal control, coupled with unopposed
parasympathetic dominance because of parasym-
pathetic, vagus nerve sparing (Mathias, 1976;
Piepmeier et al., 1985; Lehmann et al., 1987; Bravo
et al., 2004). Hypoxia, hypo-ventilation, and trache-
al suctioning appear to intensify the bradycardia. In
animals, acute spinal cord transection produces a
variety of arrhythmias (Greenhoot et al., 1972).
Several human studies have documented ECG
characteristics of individuals with spinal cord in-
jury. Blocker et al. (1983) analyzed the resting
ECG of 98 individuals with chronic cord injury
and compared their findings with results from two
studies of healthy able-bodied individuals. The in-
vestigators reported that ECG abnormalities were
more prevalent in the individuals with cord injury.
The mean age of the individuals with cord injury
was 47 years and more than half of the individuals
had sustained their injury before the age of 40. The
most common ECG abnormalities were ST seg-
ment depression and T-wave inversion. Axis devi-
ation, ventricular conduction delays, frequent
premature ventricular contractions, and low QRS
amplitude were also recorded. As might be ex-
pected, ECG abnormalities were most prevalent in
the 50–59 age group. The sample contained an
equal number of subjects with cervical and tho-
racic cord injury. ECG abnormalities were most
often recorded when the injury was cervical. The
incidence of ECG abnormalities did not relate to
the time interval after the cord injury. Individuals
with lumbar spinal cord injury had no ECG ab-
normalities. Lehmann et al. (1989) also analyzed
the ECG of individuals with spinal cord injury.
These investigators reported altered ventricular
depolarization in individuals with cervical spinal
cord injury. These results were confirmed recently
(Marcus et al., 2002). The ECG abnormalities
consisted of an upwardly concave ST-segment.
In contrast to the reports cited above, Prakash
et al. (2002) documented that the prevalence of
ECG abnormalities was the same for able-bodied
individuals and individuals with spinal cord
injury. These investigators examined the ECG of
26,734 able-bodied male veterans with a mean age
of 56 years and compared the data with the ECG
of 654 individuals with SCI with a mean age of 50
years. Similarly, Leaf et al. (1993) studied 47 in-
dividuals with chronic SCI (35–3605 days post-in-
jury). Twenty-five individuals were classified as
paraplegic (mean age 39 years). No differences
were recorded in the incidence of abnormalities
between the individuals with paraplegia and quad-
riplegia. Thus, the data regarding ECG abnormal-
ities associated with SCI are equivocal.
Two reported studies tested the effect of spinal
cord injury at T4 on cardiac electrophysiology and
the susceptibility to ventricular arrhythmias
(Rodenbaugh et al., 2003a, c). In the first study,
conscious female hypertensive cord-injured rats
had a significantly lower electrical stimulation
threshold to induce ventricular arrhythmias com-
pared to intact rats. The protocol used to induce
arrhythmias is presented in Fig. 5. Ventricular
arrhythmia was defined as sustained ventricular
tachycardia resulting in a reduction in arterial
pressure. The intensity of current required to
cause a ventricular arrhythmia was 62% lower in
cord-injured rats compared with intact rats. Cord-
injured rats also had a 35% lower effective refrac-
tory period compared to intact rats. Associated
with the increased susceptibility to ventricular
arrhythmias was a significantly higher resting
heart rate and cardiac sympathetic tone (as deter-
mined by the effect of propranolol on heart rate
after the administration of atropine (Chen and
DiCarlo, 1997). Triggered beats occur more fre-
quently in the presence of increased heart rate
(Rosen and Reder, 1981). The enhanced cardiac
sympathetic activity in the cord-injured rat may be
the cause of increased susceptibility to ventricular
arrhythmias. In this first study, female spontane-
ously hypertensive rats were studied because they
have elevated cardiac sympathetic activity
(Chandler and DiCarlo, 1998) and are susceptible
to cardiac arrhythmias. In the second study
(Rodenbaugh et al., 2003c), changes in cardiac
calcium regulatory proteins (Fig. 6), cardiac elect-
rophysiology parameters, and the susceptibility to
ventricular arrhythmias were examined in intact
 283

Fig. 5. Analog recording of arterial pressure and the ECG during step increases in current delivered to the heart in a conscious cord-
injured hypertensive rat. A MacLab programmable stimulator delivered 10 s trains of pulses (frequency 50 Hz and duration of 10 ms).
The current was increased every 10 s in 10 mA increments. Ventricular arrhythmia was identified by both the ECG as rapid, wide QRS
complexes and a decrease in arterial pressure to 40 mmHg (inset). In this animal, the threshold current required to induce the
arrhythmia was 100 mA. Normal sinus rhythm reappears upon termination of the stimulation (not shown). Reprinted with permission
from Rodenbaugh et al. (2003b).

and cord-injured rats (Fig. 7). Mid-thoracic cord
injury was associated with alterations in the
abundance of cardiac calcium regulatory proteins.
For example, cord injury increased the relative
protein expression of SERCA2 (45%) and the
Na+/Ca2+ exchanger (40%), whereas relative
protein expression of phospholamban was signif-
icantly decreased (28%, Fig. 6). It is well doc-
umented that reductions in phospholamban and/
or increases in SERCA protein abundance result
in an increased sarco(endo)plasmic reticulum
calcium load (Ji et al., 2000). The sarco(endo)-
plasmic reticulum calcium overload may produce
spontaneous calcium release, thereby leading to
ectopic activity. Importantly, these molecular
changes were associated with enhanced cardiac
electrophysiology parameters and a reduced elec-
trical stimulation threshold to induce ventricular
arrhythmias. The cord-injured rats had a reduced
electrical stimulation threshold to induce ventricu-
lar arrhythmias (48%, Fig. 7) as well as shorter
atrial-ventricular interval, sinus node recovery
time and Wenckebach cycle length. In this study,
mean arterial pressure was not significantly differ-
ent between the intact and cord-injured rats.
However, injured rats had significantly higher
284
Fig. 6. Western blots for sarco(endo)plasmic reticulum Ca2+
ATPase (SERCA; A), phospholamban (PLM; B), and sodium
calcium exchanger (NCX; C) using the hearts of two intact and
two cord-injured male Wistar rats. Quantified relative abun-
dance for each protein of interest in intact (n ¼ 8) and cord-
injured (n ¼ 8) rat hearts. Spinal cord injury increased the rel-
ative abundance of SERCA (by 45%) and Na/Ca exchanger (by
40%) with a concomitant decrease in phospholamban (28%);


Pp0:05; intact vs. injured rats. Reprinted with permission
from Rodenbaugh, et al. (2003c).
heart rates (Mayorov et al., 2001; Jacobs et al.,
2002; Rodenbaugh et al., 2003b, c).
These results were consistent with clinical
reports suggesting an increased susceptibility to
cardiac arrhythmias (Guttmann and Whitteridge,
1947; Frankel et al., 1975; Colachis and Clinchot,
1997) as well as alterations in the ECG of indi-
viduals with spinal cord injuries (Lehmann et al.,
1989; Marcus et al., 2002). The results were also
consistent with a cardiac sympatho-excitation in
rats with mid-thoracic cord injury (Maiorov et al.,
1997; Rodenbaugh et al., 2003b) and humans with
Fig. 7. Electrical stimulation threshold to induce ventricular
arrhythmias in intact (n ¼ 10) and cord-injured rats (n ¼ 6).
Spinal cord injury significantly reduced the stimulation thresh-
old by 48%. The stimulation threshold was not significantly
different in the rats that served as time controls.
Pp0:05; in-
tact vs. injured rats. Reprinted with permission from Rodenb-
augh et al. (2003c).
a similar injury (Davis and Shephard, 1988; Kinzer
and Convertino, 1989; Jacobs et al., 2002).
Intravenous epinephrine in humans mimicked the
effects of mid-thoracic spinal cord injury on heart
rate, effective refractory period and atrioventricu-
lar conduction (Morady et al., 1988). A b-ad-
renergic receptor antagonist blocked these effects
of epinephrine. Conversely, perturbations that
lower sympathetic activity and/or raise parasym-
pathetic activity slow the conductive properties
and intrinsic excitability of the heart (Stein et al.,
2002; Such et al., 2002). Taken together, these re-
sults suggest that the increased susceptibility to
ventricular arrhythmias in cord-injured rats may
be due, in part, to increased cardiac sympathetic
activity. The increased cardiac sympathetic activ-
ity, higher heart rates, and changes in calcium
regulatory proteins in the cord-injured rats favor
conditions of calcium overload that increases the
likelihood for ventricular arrhythmias.
Conclusion
Individuals with spinal cord injury have unstable
arterial pressure and heart rate due to profound
changes in the autonomic nervous system. Fur-
thermore, the autonomic nervous system modu-
lates cardiac electrophysiology and abnormalities
of autonomic function can increase the risk of

ventricular arrhythmias. Recent data suggest an
increased susceptibility to ventricular arrhythmias
with concomitant changes in cardiac electrophys-
iology parameters and the abundance of calcium
regulatory proteins in a conscious chronic model
of mid-thoracic spinal cord injury. These effects
may be mediated by an increased cardiac sympa-
thetic activity. It is interesting to speculate that
these cardiac changes contribute, in part, to the
fact that cardiovascular disease is a leading cause
of death for individuals with chronic spinal
cord injury.
Acknowledgments
We gratefully acknowledge the expert technical
assistance of Dustin G. Nowacek. This work was
supported, in part, by the National Heart Lung
and Blood Institute Grant HL-74122.
References
Bauman, W.A., Adkins, R.H., Spungen, A.M., Herbert, R.,
Schechter, C., Smith, D., Kemp, B.J., Gambino, R.,
Maloney, P. and Waters, R.L. (1999) Is immobilization as-
sociated with an abnormal lipoprotein profile? Observations
from a diverse cohort. Spinal Cord, 37: 485–493.
Bauman, W.A., Spungen, A.M., Zhong, Y.G., Rothstein, J.L.,
Petry, C. and Gordon, S.K. (1992) Depressed serum high
density lipoprotein cholesterol levels in veterans with spinal
cord injury. Paraplegia, 30: 697–703.
Berkowitz, M., O’Leary, P.K., Kruse, D.L. and Harvey, C.
(1998) Spinal Cord Injury: An Analysis of Medical and So-
cial Costs. Demos Medical Publishing, Inc., New York.
Berne, R. and Levy, M. (2001) Cardiovascular Physiology. 8th
Mosby-Year Book Inc., St. Loius, MO.
Blocker, W.P., Merrill, J.M., Krebs, M.A., Cardus, D.P. and
Ostermann, H.J. (1983) An electrocardiographic survey of
patients with chronic spinal cord injury. Am. Correct Ther.
J., 37: 101–104.
Bonica, J.J. (1968) Autonomic innervation of the viscera in re-
lation to nerve block. Anesthesiology, 29: 793–813.
Bravo, G., Guizar-Sahagun, G., Ibarra, A., Centurion, D. and
Villalon, C.M. (2004) Cardiovascular alterations after spinal
cord injury: an overview. Curr. Med. Chem. Cardiovasc.
Hematol. Agents, 2: 133–148.
Calaresu, F.R. and Yardley, C.P. (1988) Medullary basal sym-
pathetic tone. Ann. Rev. Physiol., 50: 511–524.
Cardiovascular–Cardiopulmonary Secondary Disabilities.
(1991) First colloquium on preventing secondary disabilities
among people with spinal cord injuries. Center for Disease
Control Atlanta, pp. 47–54.
285
Cardus, D., Ribas-Cardus, F. and McTaggart, W.G. (1992)
Lipid profiles in spinal cord injury. Paraplegia, 30: 775–782.
Chandler, M.P. and DiCarlo, S.E. (1998) Acute exercise and
gender alter cardiac autonomic tonus differently in hyper-
tensive and normotensive rats. Am. J. Physiol. Reg. Integ.
Comp. Physiol., 274: R510–R516.
Chen, C.-Y. and DiCarlo, S.E. (1997) Endurance exercise
training-induced resting bradycardia: a brief review. Sport.
Med. Train. Rehabil., 8: 37–77.
Colachis III, S.C. and Clinchot, D.M. (1997) Autonomic hype-
rreflexia associated with recurrent cardiac arrest: case report.
Spinal Cord, 35: 256–257.
Collins, H.L. and DiCarlo, S.E. (2002) TENS attenuates re-
sponse to colon distension in paraplegic and quadriplegic rats.
Am. J. Physiol. Heart Circ. Physiol., 283: H1734–H1739.
Collins, H.L., Rodenbaugh, D.W. and DiCarlo, S.E. (2000)
Daily exercise attenuates the development of arterial blood
pressure related cardiovascular risk factors in hypertensive
rats. Clin. Exp. Hypertens., 22: 193–202.
Comarr, A.E. and Eltorai, I. (1997) Autonomic dysreflexia/
hyperreflexia. J. Spinal Cord Med., 20: 345–354.
Davis, G.M., Servedio, F.J., Glaser, R.M., Gupta, S.C. and
Suryaprasad, A.G. (1990) Cardiovascular responses to arm
cranking and FNS-induced leg exercise in paraplegics.
J. Appl. Physiol., 69: 671–677.
Davis, G.M. and Shephard, R.J. (1988) Cardiorespiratory fit-
ness in highly active versus inactive paraplegics. Med. Sci.
Sports Exerc., 20: 463–468.
Dearwater, S.R., LaPorte, R.E., Robertson, R.J., Brenes, G.,
Adams, L.L. and Becker, D. (1986) Activity in the spinal
cord-injured patient: an epidemiologic analysis of metabolic
parameters. Med. Sci. Sports Exerc., 18: 541–544.
DeVivo, M.J. (1997) Causes and costs of spinal cord injury in
the United States. Spinal Cord, 35: 809–813.
DeVivo, M.J., Black, K.J. and Stover, S.L. (1993) Causes of
death during the first 12 years after spinal cord injury. Arch.
Phys. Med. Rehabil., 74: 248–254.
DeVivo, M.J., Shewchuk, R.M., Stover, S.L., Black, K.J. and
Go, B.K. (1992) A cross-sectional study of the relationship
between age and current health status for persons with spinal
cord injuries. Paraplegia, 30: 820–827.
DiCarlo, S.E. (1982) Improved cardiopulmonary status after a
two-month program of graded arm exercise in a patient with
C6 quadriplegia: a case report. Phys. Ther., 62: 456–459.
DiCarlo, S.E. (1988) Effect of arm ergometry training on
wheelchair propulsion endurance of individuals with quad-
riplegia. Phys. Ther., 68: 40–44.
DiCarlo, S.E., Collins, H.L., Rodenbaugh, D.W., Smitha,
M.R., Berger, R.D. and Yeragani, V.K. (2002) Daily exercise
reduces measures of heart rate and blood pressure variability
in hypertensive rats. Clin. Exp. Hypertens., 24: 221–234.
DiCarlo, S.E., Stahl, L.K. and Bishop, V.S. (1997) Daily ex-
ercise attenuates the sympathetic nerve response to exercise
by enhancing cardiac afferents. Am. J. Physiol., 273:
H1606–H1610.
DiCarlo, S.E., Supp, M.D. and Taylor, H.C. (1983) Effect
of arm ergometry training on physical work capacity of
286
individuals with spinal cord injuries. Phys. Ther., 63:
1104–1107.
Duckworth, W.C., Solomon, S.S., Jallepalli, P., Heckemeyer,
C., Finnern, J. and Powers, A. (1980) Glucose intolerance
due to insulin resistance in patients with spinal cord injuries.
Diabetes, 29: 906–910.
Engel, G.L. (1978) Psychologic stress, vasodepressor (vasova-
gal) syncope, and sudden death. Ann. Intern. Med., 89:
403–412.
Faghri, P.D., Yount, J.P., Pesce, W.J., Seetharama, S. and
Votto, J.J. (2001) Circulatory hypokinesis and functional
electric stimulation during standing in persons with spinal
cord injury. Arch. Phys. Med. Rehabil., 82: 1587–1595.
Frankel, H.L., Coll, J.R., Charlifue, S.W., Whiteneck, G.G.,
Gardner, B.P., Jamous, M.A., Krishnan, K.R., Nuseibeh, I.,
Savic, G. and Sett, P. (1998) Long-term survival in spinal
cord injury: a fifty year investigation. Spinal Cord, 36:
266–274.
Frankel, H.L., Mathias, C.J. and Spalding, J.M. (1975) Mech-
anisms of reflex cardiac arrest in tetraplegic patients. Lancet,
13: 1183–1185.
Frattola, A., Parati, G., Cuspidi, C., Albini, F. and Mancia, G.
(1993) Prognostic value of 24-hour blood pressure variability.
J. Hypertens., 11: 1133–1137.
Frishman, W.H. (1992) Beta-adrenegic blockers as cardiopro-
tective agents. Am. J. Cardiol., 70: 2I–6I.
Glaser, R.M. (1985) Exercise and locomotion for the spinal
cord injured. Exerc. Sport. Sci. Rev., 13: 263–303.
Golden, K.L., Fan, Q.I., Chen, B., Ren, J., O’Connor, J. and
Marsh, J.D. (2000) Adrenergic stimulation regulates Na(+)/
Ca(2+) exchanger expression in rat cardiac myocytes.
J. Mol. Cell. Cardiol., 32: 611–620.
Golden, K.L., Ren, J., Dean, A. and Marsh, J.D. (2002) Nor-
epinephrine regulates the in vivo expression of the L-type
calcium channel. Mol. Cell. Biochem., 236: 107–114.
Golden, K.L., Ren, J., O’Connor, J., Dean, A., DiCarlo, S.E.
and Marsh, J.D. (2001) In vivo regulation of Na/Ca ex-
changer expression by adrenergic effectors. Am. J. Physiol.
Heart Circ. Physiol., 280: H1376–H1382.
Greenhoot, J.H., Shiel, F.O. and Mauck, H.P.J. (1972) Exper-
imental spinal cord injury. Electrocardiographic abnormali-
ties and fuchsinophilic myocardial degeneration. Arch.
Neurol., 26: 524–529.
Guttmann, L., Frankel, H.L. and Paeslack, V. (1965) Cardiac
irregularities during labour in paraplegic women. Paraplegia,
3: 144–151.
Guttmann, L. and Whitteridge, D. (1947) Effects of bladder
distension on autonomic mechanisms after spinal cord inju-
ries. Brain, 70: 361–404.
Hjeltnes, N. (1977) oxygen uptake and cardiac output in graded
arm exercise in paraplegics with low level spinal lesions.
Scand. J. Rehabil. Med., 9: 107–113.
Hoffman, M.D. (1986) Cardiorespiratory fitness and training in
quadriplegics and paraplegics. Sports Med., 3: 312–330.
Hooker, S.P. and Wells, C.L. (1989) Effects of low- and mod-
erate-intensity training in spinal cord-injured persons. Med.
Sci. Sports Exerc., 21: 18–22.
Hopman, M.T., Pistorius, M., Kamerbeek, I.C. and Binkhorst,
R.A. (1993a) Cardiac output in paraplegic subjects at high
exercise intensities. Eur. J. Appl. Physiol. Occup. Physiol., 66:
531–535.
Hopman, M.T., Verheijen, P.H. and Binkhorst, R.A. (1993b)
Volume changes in the legs of paraplegic subjects during arm
exercise. J. Appl. Physiol., 75: 2079–2083.
Jacobs, P.L., Mahoney, E.T., Robbins, A. and Nash, M. (2002)
Hypokinetic circulation in persons with paraplegia. Med. Sci.
Sports Exerc., 34: 1401–1407.
Jennings, G.L., Nelson, L., Esler, M.D., Leonard, P. and
Korner, P.I. (1984) Effects of changes in physical activity on
blood pressure and sympathetic tone. J. Hypertens. Suppl., 2:
S139–S141.
Ji, Y., Lalli, M.J., Babu, G.J., Xu, Y., Kirkpatrick, D.L.,
Liu, L.H., Chiamvimonvat, N., Walsh, R.A., Shull, G.E. and
Periasamy, M. (2000) Disruption of a single copy of the SE-
RCA2 gene results in altered Ca2+ homeostasis and card-
iomyocyte function. J. Biol. Chem., 275: 38073–38080.
Jose, A.D. (1966) Effect of combined sympathetic and para-
sympathetic blockade on heart rate and cardiac function in
man. Am. J. Cardiol., 18: 476–478.
Julius, S., Palantini, P. and Nesbitt, S.D. (1998) Tachycardia:
an important determinant of coronary risk in hypertension.
J. Hypertens. Suppl., 16: S9–S15.
Julius, S. and Valentini, M. (1998) Consequences of the in-
creased autonomic nervous drive in hypertension, heart fail-
ure and diabetes. Blood Pressure., 3(Suppl): 5–13.
Karlsson, A.K. (1999) Insulin resistance and sympathetic func-
tion in high spinal cord injury. Spinal Cord, 37: 494–500.
Karlsson, A.K., Friberg, P., Lonnroth, P., Sullivan, L. and
Elam, M. (1998) Regional sympathetic function in high spi-
nal cord injury during mental stress and autonomic dysre-
flexia. Brain, 121: 1711–1719.
Katona, P.G., McLean, M., Dighton, D.H. and Guz, A. (1982)
Sympathetic and parasympathetic cardiac control in athletes
and nonathletes at rest. J. Appl. Physiol., 52: 1652–1657.
Katz, A.M. (1992) Physiology of the Heart. Raven Press,
New York.
Kinzer, S.M. and Convertino, V.A. (1989) Role of leg
vasculature in the cardiovascular response to arm work in
wheelchair-dependent populations. Clin. Physiol., 9:
525–533.
Krassioukov, A.V. and Weaver, L.C. (1995) Episodic hyper-
tension due to autonomic dysreflexia in acute and chronic
spinal cord-injured rats. Am. J. Physiol. Heart Circ. Physiol.,
268: H2077–H2083.
Le, C.T. and Price, M. (1982) Survival from spinal cord injury.
J. Chronic Dis., 35: 487–492.
Leaf, D.A., Bahl, R.A. and Adkins, R.H. (1993) Risk of cardiac
dysrhythmias in chronic spinal cord injury patients. Paraple-
gia, 31: 571–575.
Lee, B.Y., Karmakar, M.G., Herz, B.L. and Sturgill, R.A.
(1995) Autonomic dysreflexia revisited. J. Spinal Cord Med.,
18: 75–87.
Lehmann, K.G., Lane, J.G., Piepmeier, J.M. and Batsford,
W.P. (1987) Cardiovascular abnormalities accompanying

acute spinal cord injury in humans: incidence, time course
and severity. J. Am. Coll. Cardiol., 10: 46–52.
Lehmann, K.G., Shandling, A.H., Yusi, A.U. and Froelicher,
V.F. (1989) Altered ventricular repolarization in central sym-
pathetic dysfunction associated with spinal cord injury. Am.
J. Cardiol., 63: 1498–1504.
Levi, R., Hultling, C. and Seiger, A. (1995) The Stockholm
spinal cord injury study. 3. Health-related issues of the
Swedish annual level-of-living survey in SCI subjects and
controls. Paraplegia, 33: 726–730.
Loewy, A. and Spyer, K. (1990) Vagal preganglionic neurons.
Central Regulation of Autonomic Functions. Oxford Uni-
versity Press, New York, pp. 68–87.
Maiorov, D.N., Weaver, L.C. and Krassioukov, A.V. (1997)
Relationship between sympathetic activity and arterial pres-
sure in conscious spinal rats. Am. J. Physiol. Heart Circ.
Physiol., 272: H625–H631.
Marcus, R.R., Kalisetti, D., Raxwal, V., Kiratli, B.J., Myers, J.,
Perkash, I. and Froelicher, V.F. (2002) Early repolari-
zation in patients with spinal cord injury: prevalence and
clinical significance. J. Spinal Cord Med., 25: 33–38 discus-
sion 39.
Mathias, C.J. (1976) Bradycardia and cardiac arrest during
tracheal suction — mechanisms in tetraplegic patients. Eur. J.
Intensive Care Med., 2: 147–156.
Mathias, C.J. and Frankel, H.L. (1992) The cardiovascular
system in tetraplegia and paraplegia. In: Frankel H.L. (Ed.),
Handbook of Clinical Neurology. Elsevier Science Publish-
ers, Amsterdam, pp. 435–456.
Mayorov, D.N., Adams, M.A. and Krassioukov, A.V. (2001)
Telemetric blood pressure monitoring in conscious rats
before and after compression injury of spinal cord.
J. Neurotrauma, 18: 727–736.
McGuire, T.J. and Kumar, V.N. (1986) Autonomic dysreflexia
in the spinal cord injured: what the physician should know
about this medical emergency. Postgrad. Med., 80: 81–89.
McKinley, W.O., Jackson, A.B., Cardenas, D.D. and DeVivo,
M.J. (1999) Long-term medical complications after traumatic
spinal cord injury: a regional model systems analysis. Arch.
Phys. Med. Rehabil., 80: 1402–1410.
Morady, F., Nelson, S.D., Kou, W.H., Pratley, R., Schmaltz,
S., De Buitleir, M. and Halter, J.B. (1988) Electrophysiologic
effects of epinephrine in humans. J. Am. Coll. Cardiol., 11:
1235–1244.
Murray, K.T. and Roden, D.M. (1996) Cardiac arrhythmias
and sudden death. In: Robertson D., Low P.A. and Polinsky
R.J. (Eds.), Primer on the Autonomic Nervous System. Ac-
ademic Press, San Diego, CA, pp. 149–152.
Naftchi, N.E. (1990) Mechanism of autonomic dysreflexia:
contributions of catecholamine and peptide neurotransmit-
ters. Ann. NY Acad. Sci., 579: 133–148.
Palatini, P. and Julius, S. (1997) Association of tachycardia
with morbidity and mortality: pathophysiological consider-
ations. J. Hum. Hypertens., 11(Suppl 1): S19–S27.
Palatini, P. and Julius, S. (1999) Relevance of heart rate as a
risk factor in hypertension. Curr. Hypertens. Rep., 1:
219–224.
287
Piepmeier, J.M., Lehmann, K.B. and Lane, J.G. (1985) Cardi-
ovascular instability following acute cervical spinal cord
trauma. Cent. Nerv. Syst. Trauma, 2: 153–160.
Prakash, M., Raxwal, V., Froelicher, V.F., Kalisetti, D., Vieira,
A., O’Mara, G., Marcus, R., Myers, J., Kiratli, J. and
Perkash, I. (2002) Electrocardiographic findings in patients
with chronic spinal cord injury. Am. J. Phys. Med. Rehabil.,
81: 601–608.
Rizzoni, D., Muiesan, M.L., Montani, G., Zulli, R., Calebich,
S. and Agabiti-Rosei, E. (1992) Relationship between initial
cardiovascular structural changes and daytime and nighttime
blood pressure monitoring. Am. J. Hypertens., 5: 180–186.
Rodenbaugh, D.W., Collins, H.L. and DiCarlo, S.E. (2003a)
Paraplegia differentially increases arterial blood pressure re-
lated cardiovascular disease risk factors in normotensive and
hypertensive rats. Brain Res., 980: 242–248.
Rodenbaugh, D.W., Collins, H.L. and DiCarlo, S.E. (2003b)
Increased susceptibility to ventricular arrhythmias in hyper-
tensive paraplegic rats. Clin. Exp. Hypertens., 25: 349–358.
Rodenbaugh, D.W., Collins, H.L., Nowacek, D.G. and
DiCarlo, S.E. (2003c) Increased susceptibility to ventricular
arrhythmias is associated with changes in Ca2+ regulatory
proteins in paraplegic rats. Am. J. Physiol. Heart Circ.
Physiol., 285: H2605–H2613.
Rosen, M.R. and Reder, R.F. (1981) Does triggered activity
have a role in the genesis of cardiac arrhythmias? Ann. In-
tern. Med., 94: 794–801.
Schwartz, P.J., Priori, S.G. and Napolitano, C. (1993) Role of
the Autonomic Nervous System in Sudden Cardiac Death.
Sudden Cardiac Death. Blackwell Scientific Publications,
Cambridge, MA, pp. 16–37.
Schwartz, P.J. and Stone, H.L. (1980) Left stellectomy in the
prevention of ventricular fibrillation caused by acute myo-
cardial ischemia in conscious dogs with anterior myocardial
infarction. Circulation, 62: 1256–1265.
Schwartz, P.J. and Stone, H.L. (1982) The role of the auto-
nomic nervous system in sudden coronary death. Ann. NY
Acad. Sci., 382: 162–180.
Soden, R.J., Walsh, J., Middleton, J.W., Craven, M.L.,
Rutkowski, S.B. and Yeo, J.D. (2000) Causes of death after
spinal cord injury. Spinal Cord, 38: 604–610.
Stamler, J., Stamler, R. and Neaton, J.D. (1993) Blood pres-
sure, systolic and diastolic, and cardiovascular risks. Arch.
Intern. Med., 153: 598–615.
Stein, R., Medeiros, C.M., Rosito, G.A., Zimerman, L.I. and
Ribeiro, J.P. (2002) Intrinsic sinus and atrioventricular node
electrophysiologic adaptations in endurance athletes. J. Am.
Coll. Cardiol., 39: 1033–1038.
Stevenson, E.T., Davy, K.P., Jones, P.P., Desouza, C.A. and
Seals, D.R. (1997) Blood pressure risk factors in healthy
postmenopausal women: physical activity and hormone re-
placement. J. Appl. Physiol., 82: 652–660.
Stiens, S., Johnson, M. and Lyman, P. (1995) Cardiac rehabil-
itation in patients with spinal cord injury. Phys. Med. Re-
habil. Clin. N. Am., 6: 263–295.
Such, L., Rodriguez, A., Alberola, A., Lopez, L., Ruiz, R.,
Artal, L., Pons, I., Pons, M.L., Garcia, C. and Chorro, F.J.
288
(2002) Intrinsic changes on automatism, conduction, and re-
fractoriness by exercise in isolated rabbit heart. J. Appl.
Physiol., 92: 225–229.
Washburn, R.A. and Figoni, S.F. (1998) Physical activity and
chronic cardiovascular disease prevention in spinal cord in-
jury: a comprehensive literature review. Top. Spinal Cord
Injury Rehabil., 3: 16–32.
Wharton, J.M., Coleman, R.E. and Strauss, H.C. (1992) The
role of the autonomic nervous system in sudden cardiac
death. Trends Cardiovas. Med., 2: 65–71.
Whiteneck, G.G., Charlifue, S.W., Frankel, H.L., Fraser,
M.H., Gardner, B.P., Gerhart, K.A., Krishnan, K.R.,
Menter, R.R., Nuseibeh, I., Short, D.J. and Silver, J.R.
(1992) Mortality, morbidity, and psychosocial outcomes of
persons spinal cord injured more than 20 years ago. Para-
plegia, 30: 617–630.
Wicks, J.R., Oldridge, N.B., Cameron, N.B. and Jones, N.L.
(1983) Arm cranking and wheelchair ergometry in elite
spinal cord injured athletes. Med. Sci. Sports Exerc., 15:
224–231.
Zimmet, P.Z., Collins, V.R., Dowse, G.K., Alberti, K.G.,
Tuomilehto, J., Gareeboo, H. and Chitson, P. (1991) The
relation of physical activity to cardiovascular disease risk
factors in Mauritians. Mauritius Noncommunicable Disease
Study Group. Am. J. Epidemiol., 134: 862–875.
Zipes, D.P. (1991) The long QT interval syndrome. A Rosetta
stone for sympathetic related ventricular tachyarrhythmias
[comment]. Circulation, 84: 1414–1419.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 19

Adaptations of peripheral vasoconstrictor pathways

after spinal cord injury

Elspeth M. McLachlan and James A. Brock

Spinal Injuries Research Centre, Prince of Wales Medical Research Institute, Gate 1, Barker Street, Randwick,
NSW 2031, Australia

Abstract: The consequences of spinal cord injury on the function of sympathetic pathways in the periphery
have generally been ignored. We discuss two types of plasticity that follow disruption of sympathetic
pathways in rats. The first relates to the partial denervation of sympathetic ganglia that would follow the
loss of some preganglionic neurones. Sprouting of residual connections rapidly reinnervates many post-
ganglionic neurones, restoring functional transmission within a few weeks, but other neurones may be
permanently decentralized. Some of the new functional connections may generate inappropriate pathways
leading to abnormal reflexes. The second type of plasticity concerns the markedly enhanced and prolonged
contractile responses to nerve activity in arterial vessels to which ongoing sympathetic activity has been
reduced or silenced following spinal cord transection or ganglion decentralization. In a cutaneous artery
(the rat tail artery), the mechanisms underlying this arterial hyperreactivity differ from those in the
splanchnic arteries (the rat mesenteric artery). In the former, hyperreactivity is mainly postjunctional but
independent of changes in a1 -adrenoceptor sensitivity, whereas the increased responsiveness in the latter
vessels can be attributed to a greater responsiveness to a1 -adrenoceptor activation. There are enough data
from humans to suggest that both of these novel findings in experimental animals are likely to apply after
spinal cord injury and contribute to autonomic dysreflexia .

The disconnection of sympathetic vasoconstrictor
pathways from brainstem control that can follow
severe spinal cord injury leads initially to a pro-
found fall in resting arterial blood pressure that
gradually recovers. The long-term adjustment is
thought to occur via endocrine mechanisms, such
as the renin–angiotensin system and anti-diuretic
hormone. The fall in blood pressure presumably
reflects a marked decrease in sympathetic activity
in vasoconstrictor axons (Wallin and Stjernberg,
1984; Stjernberg et al., 1986). In the absence of
sympathetic vasoconstrictor tone, mean resting
arterial pressure is
70 mmHg when the subject
is upright (Lee et al., 1995). The loss of reflex
Corresponding author. Tel.: +61 2 9399 1031;
+61 2 9399 1034; E-mail: e.mclachlan@unsw.edu.au
baroreceptor control makes it difficult for people
with spinal injury to tolerate sitting up quickly
from a supine position, particularly if most of the
sympathetic preganglionic outflow is uncontrolled,
i.e. when the lesion is above the sixth thoracic
segment (T6). It has generally been found that the
involvement of the splanchnic vascular bed is crit-
ical, such that reflex recovery of blood pressure is
adequate provided the resistance of the splanchnic
component can be regulated through the bar-
oreceptors. The innervation of the splanchnic bed
largely arises below T6.
A secondary consequence of severe spinal inju-
ries above T6 is the development of autonomic
dysreflexia , an inappropriate and massive sympa-
Fax: thetic response to activation of nociceptive and
non-nociceptive afferents from viscera and skin.

DOI: 10.1016/S0079-6123(05)52019-3 289

290

When there are large bladder or colon distensions, consideration. Spinal cord injury not only dener-
undetected injuries or pressure sores, the arterial vates preganglionic neurones below the injury but,
pressure can rise excessively because reflex com- if the lesion is in the thoracic or upper lumbar seg-
pensation by vasodilation in the upper body and ments, is also likely to destroy preganglionic neu-
bradycardia is limited. Systolic blood pressure can rones that are located at the site of the injury or in
peak at over 300 mmHg (Lee et al., 1995). Reflex the surrounding area involved in secondary degen-
sweating may also occur that is not present in in- eration. While the loss of descending excitatory in-
tact individuals. Although removal of the stimulus put to preganglionic neurones below the level of the
will alleviate the dangerous hypertension , mean lesion silences the tonic activity of postganglionic
blood pressure remains above resting levels for
many minutes. This condition can be life-threaten-
ing and normally requires hospitalization; death by
stroke is not unusual when this cannot be achieved.
While it is widely believed that the basis for au-
tonomic dysreflexia is the loss of baroreflex com-
pensation, changes in the injured spinal cord are
now also thought to contribute. Peptidergic noc-
iceptor afferents (expressing calcitonin gene-relat-
ed peptide) have been reported to sprout within
the dorsal horn of thoracic segments below a se-
vere spinal injury at T4–T5 in rats that exhibit
autonomic dysreflexia (Krenz and Weaver, 1998;
Weaver et al., 2001). Both the sprouting and the
hypertensive response to colon distension are at-
tentuated by intrathecal administration of anti-
body to nerve growth factor (Krenz et al., 1999).
Rearrangement of synaptic connections below the
lesion, particularly an increase in peptidergic
synapses, might contribute to enhanced reflex
sympathetic excitation by noxious visceral disten-
sion. More recently, substantial reductions in the
abnormal hypertensive reflexes in rats were
achieved by administering antibodies to the
CD11d subunit of the CD11d/CD18 integrin (Gris
et al., 2004). This treatment blocks the infiltration
of neutrophils and haematogenous monocytes/ma-
crophages at the site of the injury during the first 2 Fig. 1. Diagrammatic representation of the connections in the
days and reduces free radical formation and the lower lumbar sympathetic paravertebral ganglia in the guinea
extent of secondary damage, resulting in improved pig. In the control situation, preganglionic axons arising in
several lumbar segments (grey) project to the ganglia associated
motor, sensory and autonomic function.
with segments below the lowest white ramus (at L4). In L5
paravertebral ganglion , each ganglion cell (black) receives in-
Changes in ganglionic transmission after loss of puts from several of these segments. When the paravertebral
preganglionic neurones chain was transected above L4 paravertebral ganglion (L4 on-
ly), only the preganglionic axons arising in L4 remained, leaving
many ganglion cells denervated (white). After 4–6 weeks, the L5
The possible role of changes in the peripheral
ganglion cells became reinnervated by L4 collaterals. As a re-
pathways from the spinal cord as a contribut- sult, signals from L4 preganglionic neurones would be trans-
ing factor in the generation of heightened and mitted to the targets of neurones previously not innervated by
abnormal autonomic responses has received little this spinal segment.
 291

vasoconstrictor neurones, the destruction of pre-

ganglionic neurones will partially denervate the pe-
ripheral ganglia in which they synapse.
The consequences of partial denervation can be
predicted from experiments in which the removal
of some preganglionic inputs led to sprouting and
reinnervation within sympathetic paravertebral
ganglia (Murray and Thompson, 1957; Maehlen
and Nja, 1984; Liestol et al., 1987). In a recent
electrophysiological study of synaptic transmission
in lower lumbar sympathetic ganglia of guinea
pigs, the changes in connectivity were analyzed
following transection of the lumbar trunk just
proximal to the last white ramus at the fourth
lumbar segment (L4) (Ireland, 1999). This lesion
removes all preganglionic axons (inputs) arising
from the cell bodies in T13 to L3 spinal segments
that project to the L5 ganglion , leaving only a Fig. 2. Frequency distributions of the number of preganglionic
small component (
10%) of the original pregangl- inputs received by ganglion cells in L5 paravertebral ganglia
ionic input (Fig. 1, middle panel). The responses of under the three conditions illustrated in Fig. 1. The number of
‘‘weak’’ or subthreshold (hatched columns) and ‘‘strong’’ or sup-
the postganglionic neurones in L5 ganglion to
rathreshold (black columns) inputs is shown, together with the
stimulation of all preganglionic axons (in the lum- frequency that these inputs were not detected (white columns). In
bar sympathetic trunk immediately rostral to the the control situation (upper panels), usually one strong and sev-
ganglion ) were recorded with intracellular micro- eral weak inputs are received by the ganglion cells, arising from
electrodes. These responses were recorded at 2–3 several spinal segments. Only the strong synapses transmit sig-
nals to the peripheral target. Inputs that arise from L4 only
days after cutting the chain, when the terminals of
(middle panels) are mainly weak and many ganglion cells ini-
the preganglionic axons originating proximally to tially lack any inputs after the paravertebral chain is transected.
the L4 segment would have degenerated, and at Surprisingly, when the ganglion cells became reinnervated by L4
4–5 weeks after cutting the chain when the re- collaterals after sprouting of the residual L4 preganglionic axons
maining L4 inputs would have sprouted within the (lower panels), most of the neurones received a strong input,
indicating that these were formed preferentially. Figure modified
ganglion. These synaptic responses were compared
from Fig. 3 in Ireland (1999).
to the responses in control ganglia.
Under control conditions (Fig. 1, upper panel),
each neurone in the L5 paravertebral ganglia re- 10% of L5 postganglionic neurones, one of the
ceives, on average,
4 preganglionic cholinergic inputs from the pool of preganglionic neurones in
inputs of which at least one is usually suprathresh- L4 was strong whereas 40% of the neurones re-
old or ‘‘strong’’ (Fig. 2, upper panels). The number ceived only weak inputs from L4, and there was no
of quanta of acetylcholine released from this L4 input at all to the remaining 50% of neurones
strong preganglionic axon is large enough to en- (Fig. 2, middle panels). The summed synaptic con-
sure that ganglionic transmission (i.e. initiation of ductance change produced by stimulating all L4
the postganglionic action potential) occurs with a inputs was only 26% of the total received from the
high safety factor, as at the skeletal neuromuscular entire preganglionic outflow.
junction. The other, subthreshold or ‘‘weak,’’ in-
puts generate post-synaptic potentials of only a Sprouting of residual preganglionic axons rapidly
few mV in amplitude that do not activate the cell. reinnervates many postganglionic neurones
That is, ganglionic transmission normally occurs
only by direct relay of the signal carried by the One month after transecting the paravertebral
strong input. Two to three days after section, in chain between L3 and L4 ganglia, the responses
292
recorded in the L5 postganglionic neurones had
substantially recovered, reaching
80% of the
original summed synaptic conductance. This was
achieved both by the formation of new synapses
and by the enlargement of existing ones. Indeed,
only 22% of neurones remained denervated where-
as
60% of neurones had a strong input of normal
magnitude (Fig. 2, lower panel). The probability
was 3.5 times higher that the collateral branch of
an L4 axon formed a strong, rather than a weak,
input. This implies that the postganglionic neu-
rones send a powerful signal to whichever pre-
ganglionic terminals make contact so that one of
them develops into a strong synapse. From these
data, it is evident that, when postganglionic neu-
rones lose some inputs after spinal injury, sprout-
ing of intact preganglionic inputs caudal to the
injury can quickly restore ganglionic transmission.
New connections in ganglia may be inappropriate
Because the preganglionic inputs that sprouted to
reinnervate the partially denervated ganglion arose
only from L4 and supplied neurones normally not
innervated from this segment, it is likely that some
of the strong (i.e. effective) synapses that devel-
oped in the first few weeks after the injury were
formed on functionally incorrect postganglionic
neurones. Preganglionic axons from segments con-
taining the original source of innervation will dis-
place the erroneous sprouts if they are able to
reenter the ganglion (Murray and Thompson,
1957; Liestol et al., 1987) although, even after dis-
placement of the sprouted connections, the degree
of reinnervation is poor compared with that fol-
lowing reinnervation after complete denervation
(Maehlen and Nja, 1984). However, if the original
preganglionic neurones have been destroyed, the
potential for correction of inappropriate novel
connections is lost.
It is not clear whether some postganglionic neu-
rones may have remained permanently decentral-
ized if the survival period in the above experiments
had been longer, but clearly, in humans, this could
follow spinal injuries that destroyed part of the
thoracolumbar cord. The likelihood that all the
denervated postganglionic neurones would be-
come reinnervated would depend on the extent to
which individual preganglionic neurones are re-
stricted in the number of additional synapses they
can support, i.e. whether the size of the sympa-
thetic ‘‘neural unit’’ (Purves and Wigston, 1983) is
limited. The neural unit is the equivalent of a mo-
tor unit, being the number of postganglionic neu-
rones with which a single preganglionic neurone
synapses. Thus, the remaining L4 preganglionic
neurones might not have the capacity to take over
control of all denervated postganglionic neurones.
To summarize, the evidence in animal experi-
ments indicates that rapid sprouting of remaining
sympathetic preganglionic terminals in ganglia re-
stores transmission to many postganglionic neu-
rones that have lost their strong inputs when
preganglionic neurones are damaged. Thus, post-
ganglionic neurones denervated by spinal cord dam-
age may soon be reconnected with the cord below
the lesion and can participate at least in spinal re-
flexes. The potential for functionally inappropriate
reconnections may explain some of the abnormal
reflex responses observed in cord-injured people.
Changes in neurovascular transmission after spinal
transection
Another contributing factor to autonomic dysre-
flexia might reside in the properties of the arterial
resistance vessels after spinal cord injury. Subjects
with cervical spinal lesions have increased pressor
responses to intravenously infused noradrenaline
(Mathias et al., 1976; Krum et al., 1992). While
this change is likely to reflect, in part, loss of bar-
oreceptor reflexes (Mathias et al., 1976), it has
been suggested that increased reactivity of the vas-
cular smooth muscle to a-adrenoceptor activation
contributes to the augmented pressor responses to
noradrenaline (Krum et al., 1992). Consistent with
this idea, the foot veins of quadriplegic patients
with recurrent bouts of autonomic dysreflexia have
a markedly increased sensitivity to locally infused
noradrenaline (Arnold et al., 1995). This interven-
tion produced marked local venoconstriction in
the absence of baroreflex involvement.
Because the responses to exogenous agonists
are often unlike the responses to nerve-released
 293

transmitter (Hirst et al., 1992), it seems more rel-
evant to identify whether responses to nerve ac-
tivity are enhanced. The responses of cutaneous
vessels to brief bursts of vasoconstrictor nerve ac-
tivity elicited by increasing bladder pressure, or by
electrical stimulation of the skin of the contralat-
eral leg, led to vasoconstrictions lasting over 30 s in
subjects with cervical or high thoracic spinal inju-
ries. In intact individuals responses to similar
bursts lasted only
10 s (Wallin and Stjernberg,
1984). These observations suggest that spinal dam-
age modifies the response of cutaneous vessels to
sympathetic activity in such a way that neurogenic
vasoconstriction is prolonged.
Transmitter release may be increased in tail arteries
after spinal cord transection
By examining the electrophysiological responses in
the smooth muscle of the isolated rat -tail artery to
stimulation of the perivascular nerves , we have
obtained direct evidence that sympathetic neuro-
vascular transmission is enhanced after spinal cord
injury. The spinal cord of female Wistar rat was
transected at T7–T8, well above the preganglionic
outflow to the tail artery (T13–L2) (Rathner and
McAllen, 1998). Segments of tail artery about 1 cm
long were dissected from the animals and studied
in vitro using intracellular microelectrodes inserted
in the vascular smooth muscle.
In control tail arteries, stimulation of the peri-
vascular nerves evokes an excitatory junction po-
tential in the vascular smooth muscle cells (Fig. 3a,
upper record). Following a train of five such ex-
citatory junction potentials evoked at 1 Hz, a slow
depolarization develops, which peaks at 15 s and
lasts for 1 min (Fig. 3a, lower record). The exci-
tatory junction potential is blocked by purinergic
receptor (P2X) antagonists such as suramin and so
is likely to be due to the release of adenosine 50 -
triphosphate. The slow depolarization is blocked
by a2 -adrenoceptor antagonists such as idazoxan,
and reflects the release of noradrenaline .
When these signals were recorded in arteries
from animals at
2 months after spinal cord tran-
section, the amplitude of both the excitatory junc-
tion potentials and the slow depolarizations
evoked by supramaximal electrical stimuli were
about double those recorded in vessels from age-
matched control animals (Fig. 3b). In addition,
the frequency of spontaneous excitatory junction

Fig. 3. Spinal transection at T7–T8 increases the amplitude of intracellularly recorded excitatory junction potentials (EJP) and the
slow noradrenaline -induced depolarization (NAD) in a proximal segment of the rat -tail artery. Tissues were studied 2 months post-
operatively. (a) Traces showing the averaged response to 5 pulses at 1 Hz. The upper trace is the first part of the lower record shown on
an expanded time base. (b) EJP and NAD amplitude for arteries isolated from sham-operated rats (white columns, n ¼ 11) and rats
after spinal cord transection (hatched columns, n ¼ 13) rats , *Po0:01;**Po0:001; unpaired t-test.
294

potentials (thought to reflect the release of indi-
vidual quanta of adenosine triphosphate) was in-
creased in recordings from many preparations
after spinal cord transection. The simplest expla-
nation of these findings is that transmitter release
from the perivascular terminals is increased.
Nerve-evoked contractions are enhanced in tail
arteries from rats with spinal cord transection
The more dramatic change observed in the tail ar-
tery of rats after spinal cord transection was the
increase in the contractile responses evoked by
nerve stimulation. Contractile responses to brief
trains of perivascular stimuli were recorded from
isolated ring segments of tail artery mounted in a
myograph (Yeoh et al., 2004a). Two weeks and 2
months following spinal cord transection, brief
trains of supramaximal electrical stimuli at
0.1–10 Hz applied transmurally led to contractions
that were up to 25-fold larger after spinal cord
transection than in control arteries (Fig. 4a). The
degree of enhancement was more marked during
trains at lower stimulation frequencies (0.1 Hz).
Further, consistent with Wallin and Stjernberg’s
observations in the cutaneous vascular beds of
cord-injured subjects, the contractile responses to
nerve stimulation were prolonged in time course
relative to control (Fig. 4a). Taken together, these
data show that the rat -tail artery quite rapidly
becomes more reactive to nerve activity after spi-
nal transection and that this hyperreactivity is
maintained.
The degree of block of nerve-evoked responses
produced by supramaximal concentrations of a-
antagonists was less (
88%) than in controls
(95%) (Yeoh et al., 2004a). This suggests a greater
role for co-transmitters, such as adenosine tripho-
sphate, after spinal transection. However, in com-
parison with control arteries, the effectiveness of
partially blocking concentrations of the competi-
tive a1 - and a2 -adrenoceptors antagonists, prazo-
sin and idazoxan, was markedly lower in arteries
from cord-injured animals. This finding may be
explained by an increase in the amount of nor-
adrenaline released per impulse from the perivas-
cular axons following spinal transection.
Much of the enhancement of the responses to
nerve stimulation occurred postjunctionally, be-
cause the amplitude of contractions to depolari-
zation evoked by applying saline containing 60
mM K+ almost doubled (Yeoh et al., 2004a). Like
the responses to nerve stimulation, the duration of

Fig. 4. Spinal transection markedly increases nerve-evoked contractions of both (a) rat -tail artery and (b) rat mesenteric artery. The
spinal cord was transected at T7–T8 for the tail artery experiments and at T4 for the mesenteric artery experiments. Tissues were
studied 2 months post-operatively. Both panels show traces for arteries isolated from a sham-operated rat (upper) and a cord-
transected rat (lower). In (a), the tail artery was stimulated with 25 pulses at 0.1, 0.3 and 0.5 Hz and 100 pulses at 1 Hz. In (b), the
mesenteric artery was stimulated with 100 pulses at 1, 2, 3 and 5 Hz.

the contraction to increased [K+] was also pro-
longed. The arteries also showed increased sensi-
tivity to the a2 -adrenoceptor agonist, clonidine,
and a transient increase in sensitivity to the a1 -
adrenoceptor agonist, phenylephrine. After 2
months, however, the marked increase in the
nerve-evoked response was maintained but was
not associated with enhanced a1 -adrenoceptor sen-
sitivity. Interestingly, as contractions to both high
[K+] and clonidine are dependent on extracellular
Ca2+ (Abe et al., 1987; Chen and Rembold, 1995),
the augmentation of contraction to both these
agents may indicate that the contractile mecha-
nism is selectively sensitized to Ca2+ entering the
cell or that there in an increase in stimulus-induced
Ca2+ entry.
Decentralization mimics spinal cord transection in
enhancing vascular reactivity
The underlying mechanism for the hyperreactivity
after spinal transection might be related to the re-
duced activity of sympathetic postganglionic ax-
ons. We tested this idea by silencing the
sympathetic outflow to the tail by cutting the lum-
bar sympathetic chain below L3, i.e. decentralizing
the postganglionic neurones supplying the tail ar-
tery that are located in the sacral sympathetic
chain (Sittiracha et al., 1987). In the tail arteries
from these animals, the responses to stimulation of
the perivascular nerves were enhanced and pro-
longed, mimicking the changes after spinal tran-
section (Yeoh et al., 2004b). In this case, enhanced
responses to nerve stimulation were observed 2
days following decentralization, they had in-
creased after 2 weeks and were maintained for at
least 7 weeks.
The data indicate that the vascular changes after
decentralization and spinal cord transection are
not identical (Yeoh et al., 2004a, b). At two weeks,
the augmentation of nerve-evoked responses was
consistently larger after decentralization than it
was after cord transection. This difference may
reflect residual vasoconstrictor nerve activity to
arteries from cord-injured rats , evoked reflexly
from afferent input below the lesion. In addition,
there was no evidence for an increased role of
295
co-transmitters in the decentralized arteries, as the
blockade of the nerve-evoked contraction by sup-
ramaximal concentrations of a-adrenoceptor an-
tagonists was similar to control (i.e.
95%
blockade).
Nerve-evoked responses in mesenteric arteries are
also potentiated after spinal transection
More recent experiments have confirmed that re-
sponses of second-order mesenteric arteries from
rats with spinal cord transection at T4 are en-
hanced in a manner similar to the responses of tail
arteries from rats with spinal cord transection
(Fig. 4b). Mesenteric vessels are normally control-
led by preganglionic neurones that project from
T4 to T13 (Anderson et al., 1989; Taylor and
Weaver, 1992). In the rat mesenteric artery, the
contractile responses to supramaximal electrical
stimulation of the sympathetic axons are very
small so that the lowest frequency of stimulation
that can be evaluated in vitro is 1 Hz.The potent-
iation of mesenteric artery contraction at 1 Hz was
relatively greater (8-fold) than that of the tail ar-
tery (1.5-fold) (Fig. 4). Unlike for the tail artery,
the change in the nerve-evoked responses of the
mesenteric arteries was closely correlated with an
increased sensitivity to a1 -adrenoceptor agonists.
Further, the responses of the mesenteric artery to
depolarization with increased extracellular [K+]
were not larger. Thus, the enhancement of neuro-
vascular transmission in the mesenteric artery ap-
pears to be due primarily to an increased
responsiveness to a1 -adrenoceptor activation.
These findings indicate that the mechanisms
underlying the augmented responses differ bet-
ween the cutaneous and splanchnic vascular beds.
These differences probably reflect the equally
distinctive mechanisms by which sympathetic
vasoconstrictor nerves normally activate smooth
muscle of these two vascular beds. In the tail
artery, both a1 - and a2 -adrenoceptors contribute
to the nerve-evoked contraction in vitro and
co-transmitters play only a minor role (Bao
et al., 1993; Brock et al., 1997; Yeoh et al.,
2004b). In contrast, in the mesenteric vascular bed,
nerve-released noradrenaline produces contraction
296
through activation of a1 -adrenoceptors , while
both adenosine triphosphate and neuropeptide Y
appear to contribute a significant component (Do-
noso et al., 1997; Han et al., 1998). In other ar-
terial vessels, the mechanisms underlying neurally
evoked contraction differ from those in either the
tail or the mesenteric artery. For example, con-
tractions of guinea-pig submucosal arterioles are
entirely due to neurally released adenosine tripho-
sphate (Evans and Surprenant, 1992). Further-
more, in arteries like the mesenteric artery that are
innervated by a perivascular plexus of afferent ax-
ons, it is not clear whether ongoing activity in
these peptidergic terminals has an effect on the
behaviour of the vascular smooth muscle. There-
fore, the effects of silencing ongoing sympathetic
nerve activity on hyperreactivity are likely to differ
between vascular beds.
Conclusions
The result of our experiments clearly show signif-
icant changes in the peripheral components of the
sympathetic pathways involved in autonomic
dysreflexia .
(i) The experiments in guinea pigs imply that
any loss of preganglionic neurones that occurs as a
direct or indirect result of spinal damage will den-
ervate postganglionic neurones and lead to cessa-
tion of their activity. This state will not persist for
long, as sprouting of residual intact terminals
within the ganglia will soon restore strong (effec-
tive) synaptic transmission to many of the dener-
vated neurones. However, this might lead to
erroneous connections that could account for in-
appropriate functional reflexes such as sweating in
response to noxious stimuli.
(ii) The experiments in rats show that the re-
sponsiveness of both the tail artery (which supplies
a major thermoregulatory cutaneous bed) and the
small mesenteric arteries is dramatically enhanced
after spinal cord injury.
These changes appear to result from the reduc-
tion in sympathetic traffic as similar hyperreactiv-
ity follows surgical decentralization.
Irrespective of whether or not spinal shock re-
moves spinal reflex -evoked sympathetic nerve ac-
tivity, which is debated (Silver, 2000), the changes
in the vascular muscle produced by abolishing
normal ongoing nerve activity by spinal cord tran-
section or decentralization are evident within a few
days. Damage to preganglionic neurones that par-
tially denervates vasoconstrictor neurones might
silence them and also rapidly result in hyperreac-
tivity of the vasculature. Whether or not this
would be reversed when the postganglionic neu-
rones become reinnervated in the ensuing weeks is
unclear, but it is likely that some vasoconstrictor
innervation remains chronically inactive.
There is no doubt that changed responsiveness
of the vasculature is likely to contribute to the
prolonged vasoconstriction that occurs when sym-
pathetic activity is evoked below a spinal injury
(Wallin and Stjernberg, 1984; Teasell et al., 2000).
Similar increases in vascular reactivity may well
occur in most types of blood vessel below a spinal
injury, although the underlying mechanisms may
differ. What is less clear is whether vascular hyper-
reactivity contributes to other problems in patients
such as inadequate temperature regulation and the
poor healing of pressure sores. Our data imply that
sympathetic nerve activity modulates the process
of neurovascular transmission and the behaviour
of the vascular smooth muscle itself. Until more is
known about these processes, both in control sit-
uations and in conditions when nerve activity is
modified, it will not be easy to predict which ther-
apeutic approaches are the most appropriate to
ameliorate the functional disorders that appear
after injury. Another question is whether non-vas-
cular target organs in the periphery, whose pat-
terns of autonomic nerve activity are modified
after spinal cord injury have as yet undiscovered
changes in neuroeffector function.
Acknowledgments
This work was supported by grants from the Na-
tional Health & Medical Research Council of
Australia (970852, 209632) and the Christopher
Reeve Paralysis Foundation (BAC1-0101-1,
BAC1-0101-2). We thank David Ireland and Me-
lanie Yeoh for their major contributions to the
conduct of the experiments.

References
Abe, K., Matsuki, N. and Kasuya, Y. (1987) Pharmacological
and electrophysiological discrimination of contractile re-
sponses to selective a1- and a2-adrenoceptor agonists in rat
tail artery. Jpn. J. Pharmacol., 45: 249–261.
Anderson, C.R., McLachlan, E.M. and Srb-Christie, O. (1989)
Distribution of sympathetic preganglionic neurons and mon-
oaminergic nerve terminals in the spinal cord of the rat. J.
Comp. Neurol., 283: 269–284.
Arnold, J.M., Feng, Q.P., Delaney, G.A. and Teasell, R.W.
(1995) Autonomic dysreflexia in tetraplegic patients: evidence
for a-adrenoceptor hyper-responsiveness. Clin. Auton. Res.,
5: 267–270.
Bao, J.X., Gonon, F. and Stjarne, L. (1993) Frequency- and
train length-dependent variation in the roles of postjunctio-
nal a1- and a2-adrenoceptors for the field stimulation-in-
duced neurogenic contraction of rat tail artery. Naunyn
Schmiedebergs Arch. Pharmacol., 347: 601–616.
Brock, J.A., McLachlan, E.M. and Rayner, S.E. (1997) Con-
tribution of a-adrenoceptors to depolarization and contrac-
tion evoked by continuous asynchronous sympathetic nerve
activity in rat tail artery. Br. J. Pharmacol., 120: 1513–1521.
Chen, X.L. and Rembold, C.M. (1995) Phenylephrine contracts
rat tail artery by one electromechanical and three pharmaco-
mechanical mechanisms. Am. J. Physiol., 268: H74–H81.
Donoso, M.V., Steiner, M. and Huidobro-Toro, J.P. (1997)
BIBP 3226, suramin and prazosin identify neuropeptide Y,
adenosine 50 -triphosphate and noradrenaline as sympathetic
cotransmitters in the rat arterial mesenteric bed. J. Pharma-
col. Exp. Ther., 282: 691–698.
Evans, R.J. and Surprenant, A. (1992) Vasoconstriction of
guinea-pig submucosal arterioles following sympathetic nerve
stimulation is mediated by the release of ATP. Br. J. Pharma-
col., 106: 242–249.
Gris, D., Marsh, D.R., Oatway, M.A., Chen, Y., Hamilton,
E.F., Dekaban, G.A. and Weaver, L.C. (2004) Transient
blockade of the CD11d/CD18 integrin reduces secondary
damage after spinal cord injury, improving sensory, auto-
nomic, and motor function. J. Neurosci., 24: 4043–4051.
Han, S., Yang, C.L., Chen, X., Naes, L., Cox, B.F. and West-
fall, T. (1998) Direct evidence for the role of neuropeptide Y
in sympathetic nerve stimulation-induced vasoconstriction.
Am. J. Physiol., 274: H290–H294.
Hirst, G.D., Bramich, N.J., Edwards, F.R. and Klemm, M.
(1992) Transmission at autonomic neuroeffector junctions.
Trends Neurosci., 15: 40–46.
Ireland, D.R. (1999) Preferential formation of strong synapses
during re-innervation of guinea-pig sympathetic ganglia. J.
Physiol., 520(Part 3): 827–837.
Krenz, N.R., Meakin, S.O., Krassioukov, A.V. and Weaver,
L.C. (1999) Neutralizing intraspinal nerve growth factor
blocks autonomic dysreflexia caused by spinal cord injury. J.
Neurosci., 19: 7405–7414.
Krenz, N.R. and Weaver, L.C. (1998) Sprouting of primary
afferent fibers after spinal cord transection in the rat. Ne-
uroscience, 85: 443–458.
297
Krum, H., Louis, W.J., Brown, D.J. and Howes, L.G. (1992)
Pressor dose responses and baroreflex sensitivity in quadri-
plegic spinal cord injury patients. J. Hypertens., 10:
245–250.
Lee, B.Y., Karmakar, M.G., Herz, B.L. and Sturgill, R.A.
(1995) Autonomic dysreflexia revisited. J. Spinal Cord Med.,
18: 75–87.
Liestol, K., Maehlen, J. and Nja, A. (1987) Two types of
synaptic selectivity and their interrelation during sprouting in
the guinea-pig superior cervical ganglion. J. Physiol., 384:
233–245.
Maehlen, J. and Nja, A. (1984) Rearrangement of synapses on
guinea-pig sympathetic ganglion cells after partial interrup-
tion of the preganglionic nerve. J. Physiol., 348: 43–56.
Mathias, C.J., Frankel, H.L., Christensen, N.J. and Spalding,
J.M. (1976) Enhanced pressor response to noradrenaline in
patients with cervical spinal cord transection. Brain, 99:
757–770.
Murray, J.G. and Thompson, J.W. (1957) The occurrence and
function of collateral sprouting in the sympathetic nervous
system of the cat. J. Physiol., 135: 133–162.
Purves, D. and Wigston, D.J. (1983) Neural units in the supe-
rior cervical ganglion of the guinea-pig. J. Physiol., 334:
169–178.
Rathner, J.A. and McAllen, R.M. (1998) The lumbar pre-
ganglionic sympathetic supply to rat tail and hindpaw. J.
Auton. Nervous System, 69: 127–131.
Silver, J.R. (2000) Early autonomic dysreflexia. Spinal Cord,
38: 229–233.
Sittiracha, T., McLachlan, E.M. and Bell, C. (1987) The in-
nervation of the caudal artery of the rat. Neuroscience, 21:
647–659.
Stjernberg, L., Blumberg, H. and Wallin, B.G. (1986) Sympa-
thetic activity in man after spinal cord injury. Outflow to
muscle below the lesion. Brain, 109: 695–715.
Taylor, R.B. and Weaver, L.C. (1992) Spinal stimulation to
locate preganglionic neurons controlling the kidney, spleen,
or intestine. Am. J. Physiol., 263: H1026–H1033.
Teasell, R.W., Arnold, J.M., Krassioukov, A. and Delaney,
G.A. (2000) Cardiovascular consequences of loss of supra-
spinal control of the sympathetic nervous system after spinal
cord injury. Arch. Phys. Med. Rehabil., 81: 506–516.
Wallin, B.G. and Stjernberg, L. (1984) Sympathetic activity in
man after spinal cord injury. Outflow to skin below the le-
sion. Brain, 107: 183–198.
Weaver, L.C., Verghese, P., Bruce, J.C., Fehlings, M.G., Krenz,
N.R. and Marsh, D.R. (2001) Autonomic dysreflexia and
primary afferent sprouting after clip-compression injury of
the rat spinal cord. J. Neurotrauma, 18: 1107–1109.
Yeoh, M., McLachlan, E.M. and Brock, J.A. (2004a) Tail ar-
teries from chronically spinalized rats have potentiated re-
sponses to nerve stimulation in vitro. J. Physiol., 556:
545–555.
Yeoh, M., McLachlan, E.M. and Brock, J.A. (2004b) Chronic
decentralization potentiates neurovascular transmission in
the isolated rat tail artery, mimicking the effects of spinal
transection. J. Physiol., 561: 583–596.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 20

Genetic approaches to autonomic dysreflexia

A. Brown and J.E. Jacob

Biotherapeutics Research Group, The Spinal Cord Injury Team, Robarts Research Institute and The Graduate Program in
Neuroscience, The University of Western Ontario, P.O. Box 5015, 100 Perth Drive, London, ON N6A 5K8, Canada

Abstract: Autonomic dysreflexia is a potentially life-threatening condition in which episodic hypertension

occurs after injuries above the mid-thoracic segments of the spinal cord. Despite the seriousness of this
condition, little is known of the molecular mechanisms that lead to its development. The completed
sequencing of the mouse genome, its dense genetic map, and the large repository of engineered and
spontaneous mouse mutants, make the mouse an ideal model organism in which to study the molecular
mechanisms underlying autonomic dysreflexia. We subjected two wild-type strains of mice, 129Sv and
C57BL/6, and one spontaneous mouse mutant, Wallerian degeneration slow (Wlds), to spinal cord tran-
section and clip-compression injury. We found that the incidence of autonomic dysreflexia is greatly
reduced, compared to spinal cord-transected wild-type mice, in Wlds mice after both injury paradigms and
in 129Sv and C57BL/6 that have undergone the clip-compression injury. We also found that the amplitude
of the dysreflexic response was greater in cord-compressed 129Sv than in C57BL/6 mice. These results
implicate axonal degeneration as an important source of signals that trigger the development of autonomic
dysreflexia and are discussed in the context of mouse genetics, interstrain differences and possible molecular
mechanisms underlying autonomic dysreflexia after spinal cord injury.

Introduction Yardley, 1988). However, after spinal cord injury,

Autonomic dysreflexia is a condition that often
develops after midthoracic or higher spinal cord
injuries and is characterized by episodes of hyper-
tension and other signs of sympathetic hyperac-
tivity that are triggered by sensory input entering
the spinal cord below the level of the lesion. Au-
tonomic dysreflexia develops in 50–90% of people
with tetraplegia or high paraplegia (Corbett et al.,
1975; Lindan et al., 1980; Mathias and Frankel,
1993; Lee et al., 1995; Giannantoni et al., 1998). In
uninjured animals and people, blood pressure con-
trol depends upon supraspinal regulation of sym-
pathetic preganglionic neurons (Calaresu and
Corresponding author. Tel.: +519 663 5777 ext. 34308;
Fax: +519 663 3789; E-mail: abrown@robarts.ca
spinal reflexes that increase sympathetic outflow
from sympathetic preganglionic neurons dominate
the regulation of arterial pressure. The activity of
the spinal reflexes, unchecked by supraspinal inhi-
bition, leads to autonomic dysreflexia. These ex-
aggerated reflexes begin within weeks of cord
injury and can be caused by stimulation of the
skin, pressure sores, distension or inflammation of
the urinary bladder or gastrointestinal tract, and
also by muscle spasms that often develop after
spinal cord injury (Corbett et al., 1975; Mathias
and Frankel, 1993). Autonomic dysreflexia may be
mild, characterized only by sweating, piloerection,
and small increases in arterial pressure or it may be
severe and lead to debilitating headaches, seizures,
strokes, and death. Even in rehabilitated tetra-
plegics and paraplegics, this condition can become

DOI: 10.1016/S0079-6123(05)52020-X 299

300
uncontrolled, leading to life-threatening hyperten-
sion (Naftchi, 1990; Mathias and Frankel, 1993;
Lee et al., 1995; Giannantoni et al., 1998).
As is the case for any biological phenomenon,
autonomic dysreflexia can be studied using a va-
riety of techniques. Unfortunately, few if any lab-
oratories are undertaking a genetic approach to
this disorder. We will describe our own experience
in using a spontaneous mouse mutant to uncover
some of the molecular underpinnings of autonom-
ic dysreflexia. We will also describe some of the
strain differences between C57BL/6 and 129Sv
mice that we found with respect to their expression
of autonomic dysreflexia and their development of
afferent arbor plasticity after spinal cord injury.
Finally, we will briefly discuss and evaluate genetic
and genomic approaches to the study of autonom-
ic dysreflexia.
A mouse model of autonomic dysreflexia
We have been interested in developing a mouse
model to study autonomic dysreflexia because of
the many advantages of the mouse over other
model systems. First, with the exception of man,
we know much more about the genetics of mice
than of any other mammal (Copeland and Jenkins,
1991; Rinchik, 1991; Copeland et al., 1993;
Dietrich et al., 1994, 1998; Collins et al., 1998;
Marra et al., 1999; Nusbaum et al., 1999). Second,
there are a large number of spontaneous mouse
mutants that can be studied with respect to spinal
cord injury (Green and Witham, 1991; Roths
et al., 1999). Third, large genetic screens have been
undertaken in mice (Schimenti and Bucan, 1998).
Fourth, embryonic stem cell technology provides
the ability to design mice with mutations in any
nonlethal gene of interest to test its potential role
in autonomic dysreflexia (Roths et al., 1999).
Although autonomic dysreflexia had been well
documented in humans and rats (Osborn et al.,
1990; Krassioukov and Weaver, 1995; Maiorov
et al., 1997a, b; Bravo et al., 2004), there were no
previous reports characterizing autonomic dysre-
flexia in the spinal cord-injured mouse. Using 129Sv
mice and a simple transection injury at the second
thoracic spinal segment (T2) we were able to
demonstrate autonomic dysreflexia in mice. Specif-
ically, we found that we could elicit a mean arterial
blood pressure elevation of approximately 35 mmHg
in response to colon distension and cutaneous
stimulation below the level of the lesion in spinal
cord-transected mice (Fig. 1) (Jacob et al., 2001).
A definition of autonomic dysreflexia
Once we began to use spinal cord-injured mice to
address mechanisms in the development of auto-
nomic dysreflexia, it became obvious that this
could not be done without a proper definition of
autonomic dysreflexia. The importance of this def-
inition was evident by the observation that even
uninjured mice could be shown to display eleva-
tions in mean arterial blood pressure in response
to colon distension (Fig. 1). To formulate a def-
inition of autonomic dysreflexia, we compared the
pressor responses of uninjured and spinal cord-
transected mice to colon distension. We considered
cord-transected animals to be obligately dysre-
flexic since their sympathetic reflexes cannot be
affected by supraspinal inputs. We considered un-
injured mice to be obligately eureflexic because of
their intact neuraxis. Uninjured mice demonstrat-
ed an erratic blood pressure response to colon dis-
tention with readings falling above and below
baseline values during the stimulus. Upon the
withdrawal of the stimulus the blood pressures of
uninjured mice returned, within a 10 s interval, to
baseline values (the longest time taken for any
unoperated animal’s blood pressure to return to
baseline was 10.2 s). The blood pressure response
of cord-transected mice to colon distension usually
increased above baseline for the duration of the
stimulus and remained increased well after the
stimulus ceased. Thus we consider a mouse to be
dysreflexic if (1) its blood pressure is consistently
increased above baseline in response to colon dis-
tension and, (2) its blood pressure returns to base-
line with a delay greater than 10 s after colon
distension is terminated. It is interesting to note
that spinal cord-injured mice show a bimodal dis-
tribution with respect to the amount of time it
takes blood pressure to return to baseline after
colo-rectal distension. Their blood pressures either
 301

Fig. 1. Recordings of arterial blood pressure changes in response to colon distension in uninjured and spinal cord-transected mice. An
uninjured C57BL/6 mouse (A) and a C57BL/6 spinal cord-transected mouse (B) were subjected to a 0.3 ml colon distension using a
balloon-tipped catheter. On the blood pressure tracings the time points marked ‘‘start’’ indicate the point at which the balloons were
fully inflated and the time points marked ‘‘stop’’ indicate the time at which the balloons were fully deflated and removed. The dashed
lines indicate the baseline blood pressure before balloon inflation. Note that while this stimulus produces a pressor response in
uninjured mice, the blood pressure elevation is erratic and the blood pressure returns to baseline or below baseline before the stimulus
ceased. In the spinal cord-transected mice there is a rapid increase in blood pressure that remains stable above baseline for the duration
of the stimulus and for a prolonged period afterward. Modified from Jacob et al. (2003). With permission from Elsevier Science B.V.

returned to baseline within 10 s of terminating the
stimulus (a non-dysreflexic response) or their
blood pressures remained elevated for minutes af-
ter the stimulus was terminated. Once a mouse is
judged dysreflexic, its degree of dysreflexia can be
determined by measuring the increase in its blood
pressure above baseline in response to a particular
stimulus.
Proposed mechanisms for the development of
autonomic dysreflexia
Both peripheral and central mechanisms have been
proposed to explain the development of autonom-
ic dysreflexia. Peripheral mechanisms for the de-
velopment of autonomic dysreflexia include
increased expression and/or responsiveness of vas-
cular catecholamine receptors and increased neu-
ral release of catecholamines (Naftchi, 1990; Lee
et al., 1995; Karlsson et al., 1998; Karlsson, 1999;
Teasell et al., 2000). An attempt to quantify the
contribution of peripheral mechanisms to the de-
velopment of autonomic dysreflexia in the rat sug-
gests that about half of the dysreflexic response
may be attributed to changes in vascular respon-
siveness after spinal cord injury (Collins and
Dicarlo, 2002). Central mechanisms include
loss of the baroreceptor reflex and loss of tonic
bulbospinal inhibitory input to spinal neurons
(Mathias and Frankel, 1993). Indeed, the modu-
lation of spinal reflex excitation of sympathetic
preganglionic neurons by baroreceptors and other
supraspinal inhibitory systems may either be lost
following complete spinal cord transection or re-
duced following partial injuries. This lack of inhi-
bition likely has a role in determining the
magnitude and duration of the hypertensive epi-
sodes. However, the loss of inhibition develops al-
most instantaneously after the injury, whereas
autonomic dysreflexia in humans and experimental
animals take weeks or months to develop (Mathias
and Frankel, 1992; Krassioukov and Weaver,
1995; Maiorov et al., 1997a). Thus the loss of
supraspinal inputs alone cannot account for the
development of autonomic dysreflexia.
We and others have suggested that the time-
dependent, progressive development of autonomic
dysreflexia implies that in addition to spinal dys-
inhibition, synaptic plasticity leading to the grad-
ual remodeling of the spinal reflexes that control
sympathetic preganglionic neuron output may
underlie the development of this disorder (Mathias
and Frankel, 1992; Krassioukov and Weaver,
1996; Cassam et al., 1997; Weaver et al., 1997;
Jacob et al., 2001). Synaptic plasticity may be due
to the unmasking of ‘‘silent’’ synapses (Guth,
1976; Goshgarian et al., 1989), adjustments to
synaptic strength (Kang and Schuman, 1995;
302
Scanziani et al., 1996), or to axonal sprouting that
culminates in reinnervation of denervated neurons
(Steward, 1989). Many of these synaptic altera-
tions are triggered by axonal degeneration (Grob-
stein and Chow, 1987). We therefore set out to
evaluate the importance of axonal degeneration to
the development of autonomic dysreflexia by: (1)
assessing autonomic dysreflexia in a strain of mice,
designated Wallerian degeneration slow (Wlds)
mice, that undergo delayed Wallerian degenera-
tion, and (2) comparing autonomic dysreflexia in
cord-transected mice and -compressed mice in
which maximal axonal degeneration would be ex-
pected to be delayed for sometime after the initial
injury (Jacob et al., 2003).
Wlds mice, a genetic model to study the effects of
delayed axonal degeneration
The Wlds mutation arose in a substrain of C57BL/
6 mice being bred in England. The distinguishing
feature of Wlds mutants is that while the distal
segment of an injured nerve normally degenerates
within 48 h, their axons survive for a long time
after being severed and separated from their cell
bodies (Perry et al., 1990a, b, 1991). In fact, the
distal segments of severed Wlds axons not only
appear intact but may also conduct electrical im-
pulses up to 2 weeks after injury. While the genetic
lesion accounting for the Wlds mutation has been
identified as a fusion of the N-terminal portion of
the ubiquitination factor E4B (Ube4b) to the ni-
cotinamide mononucleotide adenylyltransferase
(Nmnat) gene, the mechanism through which this
mutation results in delayed Wallerian degenera-
tion is unclear (Mack et al., 2001). Wlds mice have
previously been utilized to demonstrate the im-
portance of Wallerian degeneration as a trigger for
the plasticity involved in locomotor recovery after
spinal cord injury in mice (Zhang et al., 1998).
The clip-compression injury, a nongenetic model to
study the effects of delayed axonal degeneration
The clip-compression injury model in the mouse
has been previously described (Joshi and Fehlings,
2002a, b). Clip-compression injury is carried out
using a modified aneurysm clip calibrated to a 24 g
weight, which produces a severe injury in the
mouse (Joshi and Fehlings, 2002a). This model of
spinal cord injury closely replicates the key patho-
physiological features of human injury by produc-
ing a prolonged, rapidly applied, extradural
compression. This model produces mechanical in-
jury (primary injury) and secondary damage by a
variety of well-characterized mechanisms includ-
ing microvasculature disruption, hemorrhage, is-
chemia, increases in intracellular calcium, calpain
activation, progressive axonal loss, and glutamate
toxicity (Agrawal and Fehlings, 1996, 1997a, b;
Agrawal et al., 1998; Schumacher et al., 1999;
Weaver et al., 2001, 2002). As an alternative to the
use of Wlds mice to evaluate the role of axonal
degeneration in the development of autonomic
dysreflexia, we compared mice after spinal
cord transection when all axons should undergo
Wallerian degeneration starting at the time of in-
jury to mice after clip-compression injury when one
would expect a great deal of Wallerian degeneration
to begin days after the initial insult due to the pro-
gressive nature of the secondary injury (Dumont
et al., 2001). The progressive loss of neurons due to
secondary events triggered by optic nerve crush has
been well demonstrated (Yoles and Schwartz, 1998).
Histological assessments of spinal cord transection
and clip-compression injury in wild-type and
Wlds mice
We began our analysis by evaluating the lesion
sites after spinal cord transection and clip-
compression injury in two strains of wild-type
mice, 129Sv and C57BL/6, and in Wlds mice. The
three strains of mice demonstrated very similar
histopathological changes after each injury para-
digm as demonstrated by staining with hem-
atoxylin and eosin. Two weeks after spinal cord
transection and clip-compression injury, lesion ep-
icenters were largely filled by fibrous material, ma-
crophages, and fibroblasts. In all cases small
cavities were present around the lesion, however,
large central cavities such as are found in lesioned
rat spinal cords were not found (Fig. 2). These
findings corroborate other reports that C57BL/6
Fig. 2. Histology of the lesion epicenter in the mouse and rat after spinal cord injury. Two weeks after spinal cord transection (SCT, A), and clip-compression injury
(CCI, B and C) the spinal cords of C57BL/6 mice and rats were sectioned longitudinally on a cryostat and stained with hematoxylin and eosin. The lesion epicenters in
spinal cord-transected mice (A) were characterized by well-demarcated scars that were fibrous compact, whereas the lesion epicenters in mice with clip-compression injury
(B) were less fibrous and more diffuse and cellular. Unlike spinal cord-injured rats, that demonstrate large central cavities after spinal cord injury (C), mouse spinal cords
after either injury paradigm demonstrate multiple small cavities confined to within 1–2 mm of the lesion epicenter. Cav, cavity; FS, fibrous scar. Scale bars ¼ 0.5 mm. Rat
spinal cord illustration is a photomontage. Modified from Jacob et al. (2003). With permission from Elsevier Science B.V.

303
304
and Wlds spinal cords undergo very little cavita-
tion after spinal cord injury compared to the pro-
gressive necrosis and cavitation seen in rats (Kuhn
and Wrathall, 1998; Steward et al., 1999; Jakeman
et al., 2000; Ma et al., 2001; Joshi and Fehlings,
2002a, b). Other groups have performed more de-
tailed analysis comparing the cellular reactions of
C57BL/6 and Wlds mice to spinal cord injury
(Fujiki et al., 1996; Zhang et al., 1996). These
studies have shown that in Wlds mice there is a
delay in wound healing that becomes apparent af-
ter 2 weeks post-injury and a delay in macrophage
and astrocyte activation that is obvious at 1 week
post-injury. While we were unable to demonstrate
any histological differences between the strains for
either injury paradigm, we noted an obvious differ-
ence between the mice that had undergone spinal
cord transection versus those that had undergone
clip-compression injury. The lesion epicenters in
mice that had undergone spinal cord transection
were thinner, more fibrous, and better circumscribed
than the lesions after clip-compression injury.
Incidence of autonomic dysreflexia in Wlds mice
Since injured axons in Wlds mice undergo delayed
Wallerian degeneration, we used this mutant to
assess the role of axon degeneration in the devel-
opment of autonomic dysreflexia. We reasoned
that if autonomic dysreflexia develops as a result
of events triggered by axonal degeneration, such as
changes in synaptic strength, axonal sprouting, or
reactive synaptogenesis, then the development of
autonomic dysreflexia should be delayed in Wlds
mice. In our studies the incidence of autonomic
dysreflexia at 2 weeks post-spinal cord transection
was reduced by half (4/8) in Wlds mice compared
to its wild-type parental strain, C57BL/6 (7/7).
This supports the hypothesis that the development
of autonomic dysreflexia depends in part, on
synaptic plasticity that is triggered by signals elab-
orated by degenerating axons.
Incidence of autonomic dysreflexia after spinal cord
transection and clip-compression injury
A nongenetic way to test the hypothesis that the
development of autonomic dysreflexia depends in
part, on signals released by degenerating axons is
to evaluate autonomic dysreflexia in an injury
paradigm in which axonal degeneration is delayed
compared to transection. We have argued above
that clip-compression injury is an injury paradigm
in which axonal degeneration increases over time
and is therefore somewhat delayed compared to
spinal cord transection. Blood pressure responses
after spinal cord transection and clip-compression
injury were measured at 2 weeks post-injury in re-
sponse to colon distension and cutaneous pinch
caudal to the injury. While all spinal cord-
transected C57BL/6 mice developed autonomic
dysreflexia (7/7), only approximately 50% did so
after clip-compression injury (5/9). Another wild-
type strain being studied in our laboratory, 129Sv,
also consistently demonstrated half the incidence
of autonomic dysreflexia after clip-compression
injury (5/9) compared to after spinal cord transec-
tion (6/6). Thus we suggest that the incidence of
autonomic dysreflexia after clip-compression inju-
ry is lower than the incidence after spinal cord
transection because some of the Wallerian degen-
eration that is immediate after spinal cord tran-
section is delayed after clip-compression injury.
We speculate that at sometime point greater than 2
weeks post-injury, when Wallerian degeneration
has peaked in cord-compressed mice and in cord-
transected Wlds mice, all these mice would develop
autonomic dysreflexia (assuming that they eventu-
ally show the same amount of Wallerian degener-
ation as in spinal cord transection wild-type mice).
This work using genetic models and injury para-
digms that delay axonal degeneration suggests that
the development of autonomic dysreflexia does
depend on signals from degenerating axons. This
finding has profound importance to the treatment
of spinal cord injury, suggesting that neuroprotec-
tive measures that delay Wallerian degeneration
and support axonal survival may be able to pre-
vent the synaptic plasticity associated with auto-
nomic dysreflexia.
Strain differences in the amplitude of
autonomic dysreflexia
During the analysis of autonomic dysreflexia in
C57BL/6 and Wlds mice, identical experiments

were ongoing in our laboratory to evaluate the
same responses to spinal cord transection and clip-
compression injury in another wild-type strain,
129Sv. A comparison of our results revealed in-
teresting similarities and potentially important in-
ter-strain differences in response to spinal cord
injury. Both 129Sv and C57BL/6 mice had a sim-
ilar reduction in the incidence of autonomic
dysreflexia after clip-compression injury compared
to after spinal cord transection (see above). Fur-
thermore, both strains showed approximately the
same magnitude of the blood pressure responses to
colon distension and cutaneous stimulation after
spinal cord transection. However among the mice
that underwent clip-compression injury and were
judged to be dysreflexic (5/9 for both 129Sv and
C57BL/6) the magnitude of the blood pressure re-
sponses during episodes of dysreflexia were signif-
icantly greater in 129Sv than C57BL/6 mice.
Inspection of the data shows that 129Sv and
C57BL/6 mice experienced a mean arterial blood
pressure increase of approximately 35 mmHg after
spinal cord transection in response to colon dis-
tension or cutaneous pinch. 129Sv mice demon-
strated the same degree of dysreflexia after clip-
compression injury, whereas C57BL/6 mice only
showed a mean arterial blood pressure increase of
approximately 20 mmHg after clip-compression
injury (Fig. 3).
Increases in the size of the small diameter primary
afferent arbor as a mechanism for the development
of autonomic dysreflexia
It has been proposed that sprouting of the small
diameter primary afferent arbor may lead to au-
tonomic dysreflexia by increasing sensory input
onto interneurons in the dorsal horn that control
sympathetic preganglionic neuron output. In sup-
port of this hypothesis, increases in the arbor of
small diameter afferent fibers in laminae III–V of
the dorsal horn have been demonstrated to corre-
late with the development of autonomic dysreflex-
ia (Krenz and Weaver, 1998; Krenz et al., 1999).
Pain, another complication arising from spinal
cord injury, has also been shown to be associated
with sprouting of small diameter afferent fibers
305
(Christensen and Hulsebosch, 1997a). Indeed ther-
apeutic strategies designed to stop the sprouting of
primary afferent fibers have been used to decrease
autonomic dysreflexia and hyperalgesia in animal
models following spinal cord injury (Christensen
and Hulsebosch, 1997b; Krenz et al., 1999). In-
creases in size of the small diameter primary af-
ferent arbor after spinal cord injury may be
detected by an increased area of Calcitonin gene-
related peptide-immunoreactivity (CGRP-Ir) in
laminae III–V of the dorsal horn. In our studies,
129Sv, C57BL/6, and Wlds mice sham and spinal
cord-injured animals were perfused with fixative 2
weeks post-injury and their spinal cords were sec-
tioned and processed immunohistochemically for
CGRP-Ir. Sections were analyzed and an area val-
ue (mm2) for CGRP-immunoreactive fibers was
determined within laminae III–V (Fig. 4). Where-
as, the area of CGRP-Ir was increased at all spinal
segments in 129Sv mice after both spinal cord
transection and clip-compression injury, in both
C57BL/6 and Wlds mice, CGRP-Ir was not sig-
nificantly increased by spinal cord injury despite
the fact that all mice analyzed for CGRP-Ir were
dysreflexic. We therefore concluded that sprouting
of small diameter primary afferent fibers after spi-
nal cord injury is not mandatory for the develop-
ment of autonomic dysreflexia, at least in C57BL/6
and Wlds mice. However, from these experiments,
we cannot determine the potential importance of
an increased small diameter primary afferent arbor
to the development of autonomic dysreflexia in
animals that demonstrate this phenomenon.
Strain differences in CGRP-Ir in the dorsal horn
Interesting strain differences were also observed in
the size of the small diameter primary afferent ar-
bor measured in control and injured spinal cords.
First, we found that the size of the small diameter
primary afferent arbor, as measured by CGRP-Ir,
was significantly greater in control C57BL/6 and
Wlds mice than in control 129Sv mice. Second, we
also found that, regardless of which injury was
used, spinal cord transection or clip-compression
injury, an increase in the size of the primary af-
ferent arbor depended solely on the strain of
306

Fig. 3. Blood pressure changes in dysreflexic 129Sv (gray bars) and C57BL/6 (black bars) mice 2 weeks after clip-compression injury.
The change in blood pressure was measured by subtracting the baseline blood pressure from the maximal blood pressure reached
during stimulation. In each group, the blood pressure changes were significantly different from baseline, *po0:05: In addition, the
blood pressure changes were significantly greater in 129Sv than C57BL/6 dysreflexic mice; +po0:05: From Jacob et al. (2003) and
partially reproduced with permission from Elsevier Science B.V.

mouse under consideration. Thus, whether as-
sessed after spinal cord transection or clip-com-
pression injury, 129Sv mice inevitably
demonstrated an increased afferent arbor while
C57BL/6 mice (and their mutant derivative Wlds)
did not.
Other groups also have documented strain dif-
ferences between C57BL/6 and 129Sv mice. For
example, it has been shown that kainic acid pro-
duces excitotoxic cell death in 129/SvEMS mice, as
described in the rat, but that C57BL/6 are highly
resistant to this neurotoxin (Schauwecker and
Steward, 1997). Interestingly, both strains devel-
op seizures after the same kainic acid regime. The
same group also showed that after spinal cord
crush injury, 129/SvEMS mice developed a signif-
icantly greater lesion size than C57BL/6 mice at
late (21 and 56 days post-injury) but not at early
(7 and 14 days post-injury) time points (Inman
et al., 2002). Other previously reported strain
differences point to a possible explanation for the
129Sv/C57BL/6 strain differences observed after
spinal cord injury. For example, the recruitment
of neutrophils and macrophages is defective in
129X1/SvJ compared to C57BL/6 mice in response
to chemical inflammation (White et al., 2002).
Similarly, it has also been shown that the cellular
response after spinal cord injury is different in

Fig. 4. Area of CGRP-Ir fibers (mm2) in the dorsal horn of
129Sv, C57Bl, and Wlds mice 2 weeks after a sham operation,
spinal cord transection or clip-compression injury. The area of
CGRP-Ir was measured within an area of interest in laminae
III–V of the dorsal horn as described in the text. The areas of
CGRP-Ir were increased at all spinal segments in the 129Sv
cord-transected and clip-compressed mice (A) but not in C57BL
(B) or Wlds (C) mice after either injury. *Significantly different
from sham-operated group of the same segment, po0:05: From
Jacob et al. (2003) and reproduced with permission from
Elsevier Science B.V.
307
129X1/SvJ compared to C57BL/6 mice (Ma et al.,
2004). In particular this group has shown that the
lesion site in C57BL/6 mice is occupied by more
macrophages than the lesion sites in 129X1/SvJ
mice and that the macrophages are distributed
uniformly throughout the lesion in C57BL/6,
whereas the macrophages in 129X1/SvJ mice are
distributed in patches separated by tissue matrix.
They also demonstrated more astrocytic processes
and axon profiles in the lesions of 129X1/SvJ
compared to C57BL/6 mice. Thus this study sug-
gests that the inflammatory response generated by
129X1/SvJ mice in response to spinal cord injury
results in a lesion that encourages more plasticity,
as evidenced by axonal sprouting into the lesion,
than that seen in C57BL/6 mice.
The strain differences summarized above sug-
gest that, after an equivalent injury, 129X1/SvJ
mice may elicit greater plasticity at the lesion ep-
icenter than C57BL/6. Can this help us to under-
stand strain differences in autonomic dysreflexia
that are more likely to be affected by plasticity
caudal to the injury where sympathetic outflow
controlling the major splanchnic vascular beds
originates? We have shown that 129Sv mice have a
greater degree of plasticity below the level of the
lesion, as measured by greater changes in CGRP-
Ir than C57BL/6. Assuming that the slightly dif-
ferent substrain of 129Sv mice, 129S3Svimj, used
in our studies, demonstrate the same muted in-
flammatory response to spinal cord injury as
shown for 129X1/SvJ mice, it is reasonable to
speculate that the mechanisms generating in-
creased plasticity at the lesion epicenter and at
more caudal segments may be the same. In this
regard there are two possibilities. First, since the
inflammatory response after spinal cord injury
adds to the destruction of spinal cord tissue sur-
rounding the lesion epicenter (Bethea, 2000), we
propose that a less robust inflammatory reaction
allows for greater sparing in the injured cord that
may provide more substrate for the development
of synaptic plasticity even at segmental levels cau-
dal to the primary injury. Thus, the less efficient
recruitment of inflammatory cells to the injury site
in 129Sv mice after clip-compression injury, might
lead to greater plasticity and therefore greater
blood pressure responses to sensory stimulation
308
than cord-compressed C57BL/6 dysreflexic mice.
Second, there may be a greater proclivity for plas-
ticity in the injured spinal cord in 129Sv mice
compared to C57BL/6 mice for genetic reasons
unrelated to inflammatory response. Genetic map-
ping studies that might allow these important
genes to be identified will be discussed below. We
speculate that we do not see differences in the de-
gree of dysreflexia between these strains after spi-
nal cord transection because the total disruption of
descending inputs to sympathetic preganglionic
neurons promotes maximal dysreflexia regardless
of other modulating effects.
Conclusions
The study of C57BL/6 and Wlds mice was a useful
attempt to use mouse genetics to address the mo-
lecular mechanisms responsible for autonomic
dysreflexia and supports the importance of axon-
al degeneration as a trigger to the cascade of
events that leads to autonomic dysfunction. How-
ever, because of the nature of the mutation (a fu-
sion of the N-terminus of Ube4b to the Nmnat
gene) it is difficult to speculate on the molecular
signals that might be affected. Others have used
engineered mouse mutants (knockouts and trans-
genics) to ask the roles of particular genes in var-
ious aspects of recovery from spinal cord injury
(Kim et al., 2003; Simonen et al., 2003; Wells et al.,
2003; Zheng et al., 2003; Kerr and Patterson, 2004;
Song et al., 2004). Unfortunately, none of these
studies have evaluated autonomic dysreflexia in
these mouse mutants and, while one might assume
that better locomotor recovery will translate into
less autonomic dysreflexia, caution is warranted.
For example, whereas analysis of spinal cord in-
jury in a mouse mutant that demonstrates im-
proved neuroprotection and tissue sparing might
be safely expected to have a concomitant reduction
in autonomic dysreflexia, it is also possible that
mutants that show enhanced axonal growth,
sprouting or regeneration and therefore improved
locomotor outcomes might actually have more se-
vere autonomic dysreflexia. This is because neuro-
protection and tissue sparing might allow for
enough descending inhibitory inputs into thoracic
sympathetic preganglionic neurons to suppress au-
tonomic dysreflexia, whereas axonal growth might
reflect a spinal cord environment that encourages
enhanced plasticity and promotes the development
of autonomic dysreflexia. Thus we stress the im-
portance of adding autonomic dysreflexia to the
outcome measures used by those studying spinal
cord injury in mouse mutants as it is a sensitive
measure of both axonal sparing and of deleterious
plasticity (whereas regenerative growth allowing
for improved locomotor function might be con-
sidered ameliorative plasticity). These types of
studies may then have the potential to delineate
the genetic pathways that lead to productive as
opposed to harmful plasticity. Our study and the
work of Zhang et al. (1996, 1998), suggest that the
trigger to deleterious and beneficial plasticity may
be the same, namely axonal degeneration. The
identification of molecules that are unique to the
development of one form of plasticity over the
other will be critical to therapeutic advances.
In contrast to the study of mutant mice in which
the genetic lesion has been identified, studying
phenotypic differences between strains of mice is
somewhat more resistant to genetic dissection. The
different responses to spinal cord injury observed
between C57BL/6 and 129Sv must be due to ge-
netic differences between these two strains of mice.
Inter strain genetic differences can be identified
with proper analysis and have yielded dividends in
a number of fields including obesity, atherosclero-
sis, and cancer research (Devereux and Kaplan,
1998; Diament et al., 2003; Smith, 2003). For ex-
ample, we have found that after clip-compression
injury, 129Sv mice have nearly a two-fold greater
blood pressure response to tail pinch or colon dis-
tention than C57BL/6 mice. These two mouse
strains could be crossed and the resulting F1 gen-
eration sibmated or backcrossed to one of the pa-
rental strains. By phenotyping (autonomic testing)
the F2 or backcross generation and then genotyp-
ing them using molecular markers (microsatellite
sequences, restriction fragment length polymorph-
isms, or single nucleotide polymorphisms) an as-
sociation between one or more genetic loci and an
autonomic dysreflexia phenotype could be
achieved. These types of genetic mapping studies
can narrow down the genomic region of interest to

about 1 cM (Diament et al., 2003). Through the
judicious use of congenic and recombinant inbred
strains, the genomic area of interest may be further
narrowed and allow one to evaluate candidate
genes from that locus (Mullerova and Hozak,
2004). This genetic approach may require patience
to identify genes responsible for autonomic dysre-
flexia but it offers the following two very powerful
advantages. First, because the analysis begins with
an interesting autonomic phenotype (a mouse
more or less vulnerable to develop autonomic
dysreflexia after spinal cord injury) one can be
fairly certain that the genes being sought do have a
role to play in autonomic physiology. This stands
in stark contrast to a candidate gene approach that
begins with a gene that may or may not be in-
volved in the phenomenon under study. Second,
this approach offers the potential to identify genes
with an important role to play in autonomic
dysreflexia that may not have been proposed based
on the literature. This unbiased approach to iden-
tify genes involved in autonomic dysreflexia is not
limited by our imagination to conjecture a priori
what that gene may be. Thus this type of research
may possibly open up new avenues of research into
causes of autonomic dysreflexia.
Another unbiased strategy to identify genes in-
volved in autonomic dysreflexia would be to em-
ploy genomic technologies. In contrast to the
analysis of mouse mutants or gene mapping stud-
ies that allow one to consider the role of one or a
few genes in a particular process, genomics allows
one to analyze tens-of-thousands of genes at once.
As genomic analysis rests on gene expression pro-
files that correlate gene expression with a physio-
logical process, a gene or group of genes may be
implicated with a particular process but their in-
volvement in that process must be proven by func-
tional studies. The advantage of this approach lies
in: (1) the ability to interrogate the entire tran-
scriptome for correlations to autonomic dysreflex-
ia, (2) the unbiased nature of this approach, and
(3) the speed with which one can generate candi-
date genes responsible for autonomic dysreflexia
compared to mapping studies. The biggest disad-
vantage to this approach is the problem of han-
dling and mining the volumes of data. A genomics
approach to autonomic dysreflexia requires gene
309
profiling of multiple conditions that can be corre-
lated to a degree of autonomic dysfunction. While
several groups have conducted gene expression
studies at various time points after spinal cord in-
jury in the rat using microarrays (Carmel et al.,
2001; Song et al., 2001; Nesic et al., 2002; Tachi-
bana et al., 2002; Pan et al., 2004; Resnick et al.,
2004; Zhang et al., 2004), none have tried to cor-
relate these expression patterns to autonomic
dysreflexia. However, based on the premise that
sparing of descending inputs may protect against
autonomic dysreflexia, a related approach may be
to profile gene expression after experimental treat-
ments that promote neuroprotection and tissue
sparing. For example one group has profiled gene
expression using microarrays after treating spinal
cord explants with a variety of anti-inflammatory
compounds (Pan et al., 2004). This analysis re-
vealed a neuroprotective gene cluster that was in-
duced by a COX-2 inhibitor. In a similar fashion,
we are analyzing changes in gene expression asso-
ciated with the administration of an early anti-in-
flammatory strategy after spinal cord injury in the
rat. This anti-integrin strategy has been shown to
be neuroprotective and to reduce greatly auto-
nomic dysreflexia in spinal cord-injured rats (Gris
et al., 2004). The identification of a gene or gene
cluster induced or suppressed by this treatment
may be a step toward identification of genes in-
volved in autonomic dysreflexia.
Autonomic dysreflexia is a complex disorder
that is likely the product of trauma and many ge-
netic interactions. Strategies to identify genes im-
portant to its development include the analysis of
mouse mutants and the use of genetics and gen-
omics. Our own genetic studies have helped to
uncover the role of axonal degeneration in trig-
gering the development of autonomic dysreflexia.
Our studies also reveal interesting strain differenc-
es between 129Sv and C57BL/6 mice in their
dysreflexic responses and in their expression of
plasticity after spinal cord injury as expressed by
changes in their primary afferent arbor. We
strongly urge those carrying out genetic and gen-
omic studies of spinal cord injury in rodents to use
autonomic function as an outcome measure as it is
as important to identify the genes involved in
harmful plasticity, that may lead to autonomic
310
dysfunction, as it is to identify the genes involved
in regenerative plasticity that may lead to im-
proved locomotor function. Analysis of strain dif-
ferences, new studies employing genetically
modified mice, and genomic analysis of cord-in-
jured rodents under conditions that lead to varying
degrees of dysreflexia will undoubtedly lead to the
identification of genes responsible for autonomic
dysreflexia and point the way to new innovative
approaches to its treatment.
References
Agrawal, S.K. and Fehlings, M.G. (1996) Mechanisms of sec-
ondary injury to spinal cord axons in vitro: role of Na+
Na(+)-K(+)-ATPase the Na(+)-H(+) exchanger and the
Na(+)-Ca2+ exchanger. J. Neurosci., 16: 545–552.
Agrawal, S.K. and Fehlings, M.G. (1997a) The effect of the
sodium channel blocker QX-314 on recovery after acute spi-
nal cord injury. J. Neurotrauma, 14: 81–88.
Agrawal, S.K. and Fehlings, M.G. (1997b) Role of NMDA and
non-NMDA ionotropic glutamate receptors in traumatic
spinal cord axonal injury. J. Neurosci., 17: 1055–1063.
Agrawal, S.K., Theriault, E. and Fehlings, M.G. (1998) Role of
group I metabotropic glutamate receptors in traumatic spinal
cord white matter injury. J. Neurotrauma, 15: 929–941.
Bethea, J.R. (2000) Spinal cord injury-induced inflammation: a
dual-edged sword. Prog. Brain Res., 128: 33–42.
Bravo, G., Guizar-Sahagun, G., Ibarra, A., Centurion, D. and
Villalon, C.M. (2004) Cardiovascular alterations after spinal
cord injury: an overview. Curr. Med. Chem. Cardiovasc.
Hematol. Agents, 2: 133–148.
Calaresu, F.R. and Yardley, C.P. (1988) Medullary basal sym-
pathetic tone. Annu. Rev. Physiol., 50: 511–524.
Carmel, J.B., Galante, A., Soteropoulos, P., Tolias, P., Recce,
M., Young, W. and Hart, R.P. (2001) Gene expression pro-
filing of acute spinal cord injury reveals spreading inflamma-
tory signals and neuron loss. Physiol. Genomics, 7: 201–213.
Cassam, A., Llewellyn-Smith, I. and Weaver, L. (1997) Cat-
echolamine enzymes and neuropeptides are expressed in fi-
bres and somata in the intermediate gray matter in chronic
spinal rats. Neuroscience, 78: 829–841.
Christensen, M. and Hulsebosch, C. (1997a) Chronic central
pain after spinal cord injury. J. Neurotrauma, 14: 517–537.
Christensen, M.D. and Hulsebosch, C.E. (1997b) Spinal cord
injury and anti-NGF treatment results in changes in CGRP
density and distribution in the dorsal horn in the rat. Exp.
Neurol., 147: 463–475.
Collins, F.S., Patrinos, A., Jordan, E., Chakravarti, A., Gest-
eland, R. and Walters, L. (1998) New goals for the U.S.
Human Genome Project: 1998–2003. Science, 282: 682–689.
Collins, H.L. and Dicarlo, S.E. (2002) Acute exercise reduces
the response to colon distension in T(5) spinal rats. Am. J.
Physiol. Heart Circ. Physiol., 282: H1566–H1570.
Copeland, N.G. and Jenkins, N.A. (1991) Development and
applications of a molecular genetic linkage map of the mouse
genome. Trends Genet., 7: 113–118.
Copeland, N.G., Jenkins, N.A., Gilbert, D.J., Eppig, J.T.,
Maltais, L.J., Miller, J.C., Dietrich, W.F., Weaver, A.,
Lincoln, S.E., Steen, R.G., Stein, L.D., Nadeau, J.H. and
Lander, E.S. (1993) A genetic linkage map of the mouse:
current applications and future prospects. Sci. New Series,
No. 5130, Genome Issue, 262: 57–66.
Corbett, J.L., Debarge, O., Frankel, H.L. and Mathias, C.
(1975) Cardiovascular responses in tetraplegic man to muscle
spasm bladder percussion and head-up tilt Clin. Exp.
Pharmacol. Physiol., 2: 189–193.
Devereux, T.R. and Kaplan, N.L. (1998) Use of quantitative
trait loci to map murine lung tumor susceptibility genes. Exp.
Lung Res., 24: 407–417.
Diament, A.L., Fisler, J.S. and Warden, C.H. (2003) Studies of
natural allele effects in mice can be used to identify genes
causing common human obesity. Obes. Rev., 4: 249–255.
Dietrich, W., Katz, H., Lincoln, S., Shin, H., Friedman, J.,
Dracopoli, N. and Lander, E. (1998) A genetic map of the
mouse suitable for typing intraspecific crosses. Genetics, 131:
423–447.
Dietrich, W., Miller, J., Steen, R., Merchant, M., Damron, D.,
NAhf, R., Gross, A., Joyce, D., Wessel, M., Dredge, R.,
Marquis, A., Stein, L., Goodman, N., Page, D. and Lander,
E. (1994) A genetic map of the mouse with 4,006 simple
sequence length polymorphisms. Nat. Genet., 7: 220–245.
Dumont, R.J., Okonkwo, D.O., Verma, S., Hurlbert, R.J.,
Boulos, P.T., Ellegala, D.B. and Dumont, A.S. (2001) Acute
spinal cord injury, part I: pathophysiologic mechanisms.
Clin. Neuropharmacol., 24: 254–264.
Fujiki, M., Zhang, Z., Guth, L. and Steward, O. (1996) Genetic
influences on cellular reactions to spinal cord injury: activa-
tion of macrophages/microglia and astrocytes is delayed in
mice carrying a mutation (WldS) that causes delayed Wall-
erian degeneration. J .Comp. Neurol., 371: 469–484.
Giannantoni, A., Di Stasi, S.M., Scivoletto, G., Mollo, A.,
Silecchia, A., Fuoco, U. and Vespasiani, G. (1998) Auto-
nomic dysreflexia during urodynamics [see comments]. Spinal
Cord, 36: 756–760.
Goshgarian, H.G., Yu, X.J. and Rafols, J.A. (1989) Neuronal
and glial changes in the rat phrenic nucleus occurring within
hours after spinal cord injury. J. Comp. Neurol., 284: 519–533.
Green, M.C. and Witham, B.A. (1991) Handbook on Genet-
ically Standardized Jax Mice. Jackson Laboratory, Bar Har-
bor, Maine.
Gris, D., Marsh, D.R., Oatway, M.A., Chen, Y., Hamilton,
E.F., Dekaban, G.A. and Weaver, L.C. (2004) Transient
blockade of the CD11d/CD18 integrin reduces secondary
damage after spinal cord injury, improving sensory auto-
nomic, and motor function. J. Neurosci., 24: 4043–4051.
Grobstein, P. and Chow, K.L. (1987) Visual system develop-
ment plasticity. In: Adelman G. (Ed.), Encyclopedia of
Neuroscience. Birkhauser, Boston, pp. 1294–1297.
Guth, L. (1976) Functional plasticity in the respiratory pathway
of the mammalian spinal cord. Exp. Neurol., 51: 414–420.

Inman, D., Guth, L. and Steward, O. (2002) Genetic influences
on secondary degeneration and wound healing following
spinal cord injury in various strains of mice. J. Comp.
Neurol., 451: 225–235.
Jacob, J.E., Gris, P., Fehlings, M.G., Weaver, L.C. and Brown,
A. (2003) Autonomic dysreflexia after spinal cord transection
or compression in 129Sv, C57BL, and Wallerian degenera-
tion slow mutant mice. Exp. Neurol., 183: 136–146.
Jacob, J.E., Pniak, A., Weaver, L.C. and Brown, A. (2001)
Autonomic dysreflexia in a mouse model of spinal cord in-
jury. Neuroscience, 108: 687–693.
Jakeman, L.B., Guan, Z., Wei, P., Ponnappan, R., Dzwonczyk,
R., Popovich, P.G. and Stokes, B.T. (2000) Traumatic spinal
cord injury produced by controlled contusion in mouse.
J. Neurotrauma, 17: 299–319.
Joshi, M. and Fehlings, M.G. (2002a) Development and char-
acterization of a novel, graded model of clip compressive
spinal cord injury in the mouse: part 1. Clip design, behavi-
oral outcomes, and histopathology. J. Neurotrauma, 19:
175–190.
Joshi, M. and Fehlings, M.G. (2002b) Development and char-
acterization of a novel, graded model of clip compressive
spinal cord injury in the mouse: part 2. Quantitative neuro-
anatomical assessment and analysis of the relationships be-
tween axonal tracts, residual tissue, and locomotor recovery.
J. Neurotrauma, 19: 191–203.
Kang, H. and Schuman, E.M. (1995) Long-lasting neurotro-
phin-induced enhancement of synaptic transmission in the
adult hippocampus. Science, 267: 1658–1662.
Karlsson, A.K. (1999) Autonomic dysreflexia [see comments].
Spinal Cord, 37: 383–391.
Karlsson, A.K., Friberg, P., Lonnroth, P., Sullivan, L. and
Elam, M. (1998) Regional sympathetic function in high spi-
nal cord injury during mental stress and autonomic dysre-
flexia. Brain, 121: 1711–1719.
Kerr, B.J. and Patterson, P.H. (2004) Potent pro-inflammatory
actions of leukemia inhibitory factor in the spinal cord of the
adult mouse. Exp. Neurol., 188: 391–407.
Kim, J.E., Li, S., GrandPre, T., Qiu, D. and Strittmatter, S.M.
(2003) Axon regeneration in young adult mice lacking Nogo-
A/B. Neuron, 38: 187–199.
Krassioukov, A.V. and Weaver, L.C. (1995) Episodic hyper-
tension due to autonomic dysreflexia in acute and chronic
spinal cord-injured rats. Am. J. Physiol., 268: H2077–H2083.
Krassioukov, A.V. and Weaver, L.C. (1996) Morphological
changes in sympathetic preganglionic neurons after spinal
cord injury in rats. Neuroscience, 70: 211–225.
Krenz, N. and Weaver, L. (1998) Changes in the morphology of
sympathetic preganglionic neurons parallel the development
of autonomic dysreflexia after spinal cord injury in rats.
Neurosci. Lett., 243: 61–64.
Krenz, N.R., Meakin, S.O., Krassioukov, A.V. and Weaver,
L.C. (1999) Neutralizing intraspinal nerve growth factor
blocks autonomic dysreflexia caused by spinal cord injury.
J. Neurosci., 19: 7405–7414.
Kuhn, P. and Wrathall, J. (1998) A mouse model of graded
contusive spinal cord injury. J. Neurotrauma, 15: 125–140.
311
Lee, B.Y., Karmakar, M.G., Herz, B.L. and Sturgill, R.A.
(1995) Autonomic dysreflexia revisited. J. Spinal Cord Med.,
18: 75–87.
Lindan, R., Joiner, E., Freehafer, A.A. and Hazel, C. (1980)
Incidence and clinical features of autonomic dysreflexia in
patients with spinal cord injury. Paraplegia, 18: 285–292.
Ma, M., Basso, D.M., Walters, P., Stokes, B.T. and Jakeman,
L.B. (2001) Behavioral and histological outcomes following
graded spinal cord contusion injury in the C57Bl/6 mouse.
Exp. Neurol., 169: 239–254.
Ma, M., Wei, P., Wei, T., Ransohoff, R.M. and Jakeman, L.B.
(2004) Enhanced axonal growth into a spinal cord contusion
injury site in a strain of mouse (129X1/SvJ) with a diminished
inflammatory response. J. Comp. Neurol., 474: 469–486.
Mack, T.G., Reiner, M., Beirowski, B., Mi, W., Emanuelli, M.,
Wagner, D., Thomson, D., Gillingwater, T., Court, F.,
Conforti, L., Fernando, F.S., Tarlton, A., Andressen, C.,
Addicks, K., Magni, G., Ribchester, R.R., Perry, V.H. and
Coleman, M.P. (2001) Wallerian degeneration of injured ax-
ons and synapses is delayed by a Ube4b/Nmnat chimeric
gene. Nat. Neurosci., 4: 1199–1206.
Maiorov, D.N., Krenz, N.R., Krassioukov, A.V. and Weaver,
L.C. (1997a) Role of spinal NMDA and AMPA receptors in
episodic hypertension in conscious spinal rats. Am. J.
Physiol., 273: H1266–H1274.
Maiorov, D.N., Weaver, L.C. and Krassioukov, A.V. (1997b)
Relationship between sympathetic activity and arterial pres-
sure in conscious spinal rats. Am. J. Physiol., 272:
H625–H631.
Marra, M., Hillier, L.D., Kucaba, T., Allen, M., Barstead, R.,
Beck, C., Blistain, A., Bonaldo, M., Bowers, Y., Bowles, L.,
Cardenas, M., Chamberlain, A., Chappell, J., Clifton, S.,
Favello, A., Geisel, S., Gibbons, M., Harvey, N., Hill, F.,
Jackson, Y., Kohn, S., Lennon, G., Mardis, E., Martin, J.,
Mila, L.A., McCann, R., Morales, R., Pape, D., Person, B.,
Prange, C., Ritter, E., Soares, M., Schurk, R., Shin, T.,
Steptoe, M., Swaller, T., Theising, B., Underwood, K.,
Wylie, T., Yount, T., Wilson, R. and Waterston, R. (1999)
An encyclopedia of mouse genes. Nat. Genet., 21: 191–194.
Mathias, C.J. and Frankel, H.L. (1992) The cardiovascular
system in tetraplegia and paraplegia. In: Frankel H.L. (Ed.),
Handbook of Clinical Neurology. Elsevier Science Publish-
ers, B.V., pp. 435–456.
Mathias, C.J. and Frankel, H.L. (1993) Autonomic disturbanc-
es in spinal cord lesions. In: Bannister R. and Mathias C.J.
(Eds.), Autonomic Failure. A Textbook of Clinical Disorders
of the Autonomic Nervous System. Oxford Medical Publi-
cations, Oxford.
Mullerova, J. and Hozak, P. (2004) Use of recombinant con-
genic strains in mapping disease-modifying genes. News.
Physiol. Sci., 1: 105–109.
Naftchi, N.E. (1990) Mechanism of autonomic dysreflexia.
Contributions of catecholamine and peptide neurotransmit-
ters. Ann. NY. Acad. Sci., 579: 133–148.
Nesic, O., Svrakic, N.M., Xu, G.Y., McAdoo, D., Westlund,
K.N., Hulsebosch, C.E., Ye, Z., Galante, A., Soteropoulos,
P., Tolias, P., Young, W., Hart, R.P. and Perez-Polo, J.R.
312
(2002) DNA microarray analysis of the contused spinal cord:
effect of NMDA receptor inhibition. J. Neurosci. Res., 68:
406–423.
Nusbaum, C., Slonim, D.K., Harris, K.L., Birren, B.W., Steen,
R.G., Stein, L.D., Miller, J., Dietrich, W.F., Nahf, R., Wang,
V., Merport, O., Castle, A.B., Husain, Z., Farino, G., Gray,
D., Anderson, M.O., Devine, R., Horton Jr., L.T., Ye, W.,
Wu, X., Vardouhie, K., Zemsteva, I.S., Wu, Y., Collymore,
A.J., Courtney, D.F, Tam, J., Cadman, M., Haynes, A.R.,
Heuston, C., Marsland, T., Southwell, A., Trickett, P., Striv-
ens, M.A., Ross, M.T., Makalowski, W., Xu, Y., Boguski,
M.S., Carter, N.P., Denny, P., Brown, S.D.M., Hudson, T.J.
and Lander, E.S. (1999) A YAC-based physical map of the
mouse genome. Nat. Genet., 22: 388–393.
Osborn, J.W., Taylor, R.F. and Schramm, L.P. (1990) Chronic
cervical spinal cord injury and autonomic hyperreflexia in
rats. Am. J. Physiol., 258: R169–R174.
Pan, J.Z., Jornsten, R. and Hart, R.P. (2004) Screening anti-
inflammatory compounds in injured spinal cord with mic-
roarrays: a comparison of bioinformatics analysis approach-
es. Physiol. Genomics, 17: 201–214.
Perry, V., Brown, M. and Lunn, E. (1991) Very slow retrograde
and Wallerian degeneration in the CNS of C57BL6/Ola mice.
Eur. J. Neurosci., 3: 102–105.
Perry, V., Brown, M., Lunn, E., Tree, P. and Gordon, S.
(1990a) Evidence that very slow Wallerian degeneration in
C57BL/Ola mice is an intrinsic property of the peripheral
nerve. Eur. J. Neurosci., 2: 802–808.
Perry, V., Lunn, E., Brown, M., Cahusac, S. and Gordon, S.
(1990b) Evidence that the rate of Wallerian degeneration is
controlled by a single autosomal dominant gene. Eur. J.
Neurosci., 2: 408–413.
Resnick, D.K., Schmitt, C., Miranpuri, G.S., Dhodda, V.K.,
Isaacson, J. and Vemuganti, R. (2004) Molecular evidence of
repair and plasticity following spinal cord injury. Neurore-
port, 15: 837–839.
Rinchik, E.M. (1991) Chemical mutagenesis and fine-structure
functional analysis of the mouse genome. Trends Genet., 7:
15–21.
Roths, J.B., Foxworth, W.B., McArthur, M.J., Montgomery,
C.A. and Kier, A.B. (1999) Spontaneous and engineered
mutant mice as models for experimental and comparative
pathology: history comparison and developmental technolo-
gy. Lab. Anim. Sci., 49: 12–34.
Scanziani, M., Malenka, R.C. and Nicoll, R.A. (1996) Role of
intercellular interactions in heterosynaptic long-term depres-
sion. Nature, 380: 446–450.
Schauwecker, P.E. and Steward, O. (1997) Genetic determi-
nants of susceptibility to excitotoxic cell death: implications
for gene targeting approaches. Proc. Natl. Acad. Sci. USA.,
94: 4103–4108.
Schimenti, J. and Bucan, M. (1998) Functional genomics in the
mouse: phenotype-based mutagenesis screens. Genome Res.,
8: 698–710.
Schumacher, P.A., Eubanks, J.H. and Fehlings, M.G. (1999)
Increased calpain I-mediated proteolysis, and preferential
loss of dephosphorylated NF200, following traumatic spinal
cord injury. Neuroscience, 91: 733–744.
Simonen, M., Pedersen, V., Weinmann, O., Schnell, L., Buss,
A., Ledermann, B., Christ, F., Sansig, G., van der Putten, H.
and Schwab, M.E. (2003) Systemic deletion of the myelin-
associated outgrowth inhibitor Nogo-A improves regenera-
tive and plastic responses after spinal cord injury. Neuron,
38: 201–211.
Smith, J. (2003) Quantitative trait locus mapping for athero-
sclerosis susceptibility. Curr. Opin. Lipidol., 14: 499–504.
Song, G., Cechvala, C., Resnick, D.K., Dempsey, R.J. and
Rao, V.L. (2001) GeneChip analysis after acute spinal cord
injury in rat. J. Neurochem., 79: 804–815.
Song, X.Y., Zhong, J.H., Wang, X. and Zhou, X.F. (2004)
Suppression of p75NTR does not promote regeneration of
injured spinal cord in mice. J. Neurosci., 24: 542–546.
Steward, O. (1989) Reorganization of neuronal connections
following CNS trauma: principles and experimental para-
digms. J. Neurotrauma, 6: 99–152.
Steward, O., Schauwecker, P.E., Guth, L., Zhang, Z., Fujiki,
M., Inman, D., Wrathall, J., Kempermann, G., Gage, F.H.,
Saatman, K.E., Raghupathi, R. and McIntosh, T. (1999)
Genetic approaches to neurotrauma research: opportunities
and potential pitfalls of murine models. Exp. Neurol., 157:
19–42.
Tachibana, T., Noguchi, K. and Ruda, M.A. (2002) Analysis of
gene expression following spinal cord injury in rat using
complementary DNA microarray. Neurosci. Lett., 327:
133–137.
Teasell, R.W., Arnold, J.M., Krassioukov, A. and Delaney,
G.A. (2000) Cardiovascular consequences of loss of supra-
spinal control of the sympathetic nervous system after spinal
cord injury. Arch. Phys. Med. Rehabil., 81: 506–516.
Weaver, L., Cassam, A., Krassioukov, A. and Llewellyn-Smith,
I. (1997) Changes in immunoreactivity for growth associated
protein-43 suggest reorganization of synapses on spinal sym-
pathetic neurons after cord transection. Neuroscience, 81:
535–551.
Weaver, L., Verghese, P., Bruce, J., Fehlings, M., Krenz, N.R.
and Marsh, D. (2001) Autonomic dysreflexia and primary
afferent sprouting after clip compression injury of the rat
spinal cord. J. Neurotrauma, 18: 1107–1119.
Weaver, L.C., Marsh, D.R., Gris, D., Meakin, S.O. and
Dekaban, G.A. (2002) Central mechanisms for autonomic
dysreflexia after spinal cord injury. Prog. Brain Res., 137:
83–95.
Wells, J.E., Rice, T.K., Nuttall, R.K., Edwards, D.R., Zekki,
H., Rivest, S. and Yong, V.W. (2003) An adverse role for
matrix metalloproteinase 12 after spinal cord injury in mice.
J. Neurosci., 23: 10107–10115.
White, P., Liebhaber, S.A. and Cooke, N.E. (2002) 129X1/SvJ
Mouse strain has a novel defect in inflammatory cell recruit-
ment. J. Immunol., 168: 869–874.
Yoles, E. and Schwartz, M. (1998) Degeneration of spared ax-
ons following partial white matter lesion: implications for
optic nerve neuropathies. Exp. Neurol., 153: 1–7.

Zhang, K.H., Xiao, H.S., Lu, P.H., Shi, J., Li, G.D., Wang,
Y.T., Han, S., Zhang, F.X., Lu, Y.J., Zhang, X. and Xu,
X.M. (2004) Differential gene expression after complete spi-
nal cord transection in adult rats: an analysis focused on a
subchronic post-injury stage. Neuroscience, 128: 375–388.
Zhang, Z., Fujiki, M., Guth, L. and Steward, O. (1996) Genetic
influences on cellular reactions to spinal cord injury: a
wound-healing response present in normal mice is impaired
in mice carrying a mutation (WldS) that causes delayed
Wallerian degeneration. J. Comp. Neurol., 371: 485–495.
313
Zhang, Z., Guth, L. and Steward, O. (1998) Mechanisms of
motor recovery after subtotal spinal cord injury insights
from the study of mice carrying a mutation (Wlds)
that delays cellular responses to injury. Exp. Neurol., 149:
221–229.
Zheng, B., Ho, C., Li, S., Keirstead, H., Steward, O. and
Tessier-Lavigne, M. (2003) Lack of enhanced spinal regen-
eration in Nogo-deficient mice. Neuron, 38: 213–224.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 21

Gastrointestinal symptoms related to autonomic

dysfunction following spinal cord injury

Eric A.L. Chung and Anton V. Emmanuel

St Mark’s Hospital, Northwick Park, Watford Road, Harrow, Middlesex, HA1 3UJ, UK

Abstract: The impact of spinal cord injury on an individual’s gastrointestinal tract function is often poorly
understood by the general public and also by those involved with persons with spinal cord injury. This
chapter reviews the anatomy, physiology and function of the gastrointestinal tract, with particular em-
phasis on neurological control mechanisms. In turn, it relates the effect that spinal cord injury has on the
neurological control of the gastrointestinal tract. The symptoms that are encountered by patients in the
acute phase following injury, and by individuals in the months/years after injury, with particular reference
to the effect of altered autonomic nervous system control of the gastrointestinal tract, are discussed.
Together with a following summary of current bowel management regimens and techniques, this chapter
aims to provide an overall view of the effect that autonomic dysfunction due to spinal cord injury has on
gastrointestinal function.

Introduction published work has been dedicated to the urolog-

Among the general population, the perceived im-
pact that spinal cord injury has on people is often
limited to the noticeable effects of impaired mo-
bility. Less well appreciated by the general popu-
lation is the impact that spinal cord injury has on
pelvic function, resulting in bladder, bowel and
sexual dysfunction.
Control of the gastrointestinal system involves
complex interactions between autonomic and so-
matic innervation acting ultimately at the level of
the intrinsic enteric nervous system. Following
spinal cord injury, this fine control mechanism is
interrupted to varying degrees, dependent upon
the level and extent of the spinal cord injury. The
result is a spectrum of possible gastrointestinal
symptoms. To date, the effects of spinal cord in-
jury on bowel function and management have
been poorly understood. A much larger volume of
Corresponding author. Tel.: +020-8235-4084;
Fax: +020-8235-4162; E-mail: a.emmanuel@imperial.ac.uk
ical sequelae of spinal cord injury, compared to the
effects on the gastrointestinal system. However, in
recent years this has begun to be researched. This
has been fuelled by work that has shown that a
considerable proportion of cord-injured people
rate bowel dysfunction as a greater source of
distress than bladder and sexual problems (Stone
et al., 1990a; Glickman and Kamm, 1996; De Looze
et al., 1998; Han et al., 1998).
In both the acute and long-term phases of spinal
cord injury, patients report high levels of gastroin-
testinal morbidity (Cosman et al., 1991; Krogh
et al., 1997; De Looze et al., 1998; Miller et al.,
2001). Gastrointestinal problems are a large cause of
rehospitalization among people with spinal cord in-
jury, accounting for 11% of readmissions in a recent
Australian study (Middleton et al., 2004). This mor-
bidity has significant cost implications to healthcare
systems in the acute hospital and community setting
(Harvey et al., 1992; Johnson et al., 1996).
In the acute spinal cord injury setting, symp-
toms can affect any region of the gastrointestinal

DOI: 10.1016/S0079-6123(05)52021-1 317

318
tract. At the upper end of the gut, problems in-
clude gastric dilatation, ileus, superior mesenteric
artery syndrome, peptic ulceration and pan-
creatitis. Chronic gastrointestinal symptoms en-
countered by people with spinal cord injury
include poorly localized abdominal pain, bloating,
upper gastrointestinal symptoms such as nausea
and vomiting, incontinence and constipation. This
list of symptoms which is partially attributable to
autonomic dysfunction, exists alongside the po-
tential for any other acute abdominal pathology,
the diagnosis and treatment of which may often be
complicated by the reduction of visceral sensitivity
(Bar-On and Ohry, 1995; Miller et al., 2001). The
symptoms of lower gastrointestinal tract dysfunc-
tion following spinal cord injury are more appar-
ent to clinicians, typically presenting later with
constipation and fecal incontinence.
This chapter aims to give an overview of neuro-
logical control within the gastrointestinal tract and
also to review the current understanding of the
symptoms and pathology behind the effect that spi-
nal cord injury has on the gastrointestinal system.
Background
Bowel anatomy and innervation
The gastrointestinal tract from the oesophagus to
the rectum follows a similar structural pattern of a
tube whose lumen is formed by concentric layers
of mucosa, submucosa, circular and longitudinal
muscle layers, and an outer serosal covering layer
(Fig. 1). Between the muscular layers and beneath
the mucosa are collections of nerve cells that form
plexuses (the submucosal Meissner’s and muscular
Auerbach’s plexuses) that participate in the con-
trol of gut peristalsis and secretion. Dependent
upon the position in the gastrointestinal tract these
layers vary in thickness and complexity. This pat-
tern is consistent between the gastrointestinal
tracts of most vertebrates.
The term ‘neurogenic bowel’ relates to colonic
dysfunction (constipation, fecal incontinence and
disordered defecation) following disruption of
normal control, and is the largest contributor to
gastrointestinal symptoms following spinal cord
injury. The adult human large intestine consists of
a compliant tubular sac, approximately 1.5 m in
length (Sinnatamby, 1999), which can be divided
anatomically into five parts; appendix, cecum,
colon, rectum and anus (Fig. 2). Embryologically,
the large bowel develops from two separate sourc-
es: the proximal colon up to the transverse colon
arising from the mid-gut, and the colon distal to
the mid-transverse colon, arising from the embry-
onic hindgut. Proximally it commences at the
ileocecal valve and distally it ends with the anal
sphincter. The former is of little functional signif-
icance, whereas the latter has obvious major phys-
iological importance. The colon follows the
general structure of the gastrointestinal tract, with
an inner circular smooth muscle layer and a thin
outer longitudinal muscle layer that is gathered up
into thickened cords forming the taenia coli. At
the distal rectum is the anal canal formed from
anal mucosa overlying two layers of muscle, the
internal and external anal sphincters. The internal
anal sphincter is formed from a condensation of
the inner circular smooth muscle and hence is not
under voluntary control. The external anal sphinc-
ter is made up of a circumferential ‘voluntary’
striated muscle band, which is continuous with the
pelvic floor. These sphincters work in conjunction
with the puborectalis muscle, which forms a sling
around the distal rectum and is tethered to the
pubic symphysis to maintain the puborectal angle,
a minor contributor to the maintenance of fecal
continence. Tonic contraction of the internal anal
sphincter provides 80% of resting anal pressure
(Schweiger, 1979). When the urge to defecate
occurs with rectal distension and puborectalis
stretch, contraction of the external anal sphincter
and puborectalis helps to maintain continence un-
til there is a suitable moment to void. In addition
to the voluntary control of external anal sphincter
function, there is a reflex component, which can be
experimentally triggered by a cough or Valsalva
manoeuvre, and which serves physiologically to
maintain continence during episodes of raised
intra-abdominal pressure.
The enteric nervous system
The enteric nervous system controls the gastroin-
testinal tract, via a network of sensory neurones
 319

Fig. 1. The layers of the gastrointestinal tract. From Feldman’s GastroAtlas Online, with permission.

Fig. 2. Colon anatomy. From Feldman’s GastroAtlas Online, with permission.

320
relaying information from the gut which in turn
communicates with a network of interneurones
and effector neurones to produce an effect on gut
secretion, blood flow and motor function. It has
been estimated that the enteric nervous system
contains 80–100 million neurones (Furness and
Costa, 1987) a similar number to that in the spinal
cord itself (Goyal and Hirano, 1996). The enteric
nervous system can function independently of the
central nervous system but the central nervous
system plays a large role in coordinating gut func-
tion. This has led to the concept of the brain gut
axis, with the ‘larger brain’ in the cranium and the
‘mini brain’ in the abdomen. The central nervous
system exerts its effect on the bowel via afferent
and efferent sympathetic, parasympathetic and so-
matic innervation (Sarna, 1991), which interacts
with the intrinsic nervous system. An analogy
would be of the enteric nervous system being like
constant traffic running through a town, while the
central nervous system is the system of traffic
lights and roundabouts that controls the smooth
flow of that traffic.
Nerve cell bodies in the enteric nervous system
are grouped into small ganglia that are connected
to each other by nerve processes, producing two
main plexuses that constitute this intrinsic nervous
system. The myenteric (Auerbach’s) plexus is well
developed, made up of unmyelinated fibres and
postganglionic parasympathetic cell bodies and
lies between the longitudinal and circular muscles
of the gut and coordinates peristalsis. It supplies
the mucosa with secretomotor innervation and has
connections with the sympathetic ganglia (Fig. 3).
The submucosal (Meissner’s) plexus lies on the
luminal side of the circular muscle in the submu-
cosa together with connective tissue, glands and
small vessels. It conveys local sensory and motor
responses to Auerbach’s plexus and to the central
nervous system (Stiens et al., 1997). It also has a
role in the control of secretions, endocrine cells
and the submucosal vasculature. The exact signal-
ling controls between gut neurones have yet to be
fully elucidated. Neurones have been shown to
contain the classical adrenergic and cholinergic
neurotransmitters, together with putative trans-
mitters such as peptides (e.g., substance P), amino
acids [e.g., glutamate, g amino-butyric acid
(GABA)] and smaller molecules (e.g., nitric oxide)
(Olsson and Holmgren, 2001). The role of each
transmitter varies, in each region of the gut, de-
pendent upon the interaction with other local
transmitters and receptor density on target cells
(Schemann and Neunlist, 2004). A functional prin-
ciple states that neuropeptides often act as
neuromodulators, as opposed to direct neuro-
transmitters, in the gut. Enteric nervous system
neurones can be broadly classed into intrinsic af-
ferents, interneurones and motor neurones (Lynch
et al., 2001). The intrinsic afferents form the sen-
sory limb of motor and secretory reflexes, project-
ing into the interneurones of both plexuses.
Excitatory motor and secretory neurones project
to circular muscle locally or rostrally, whereas in-
hibitory neurones project caudally. This pattern of
proximal relaxation with local and distal contrac-
tion helps coordinate churning and peristaltic gut
contractions. The afferent and motor neurones are
linked by interneurones, forming multi-synaptic
pathways that fine-tune gut secretion and gut mo-
tility. While the enteric nervous system coordinates
segmental motility and some peristaltic movement,
global colonic movements are triggered by spinal
cord-mediated reflexes, acting via pelvic nerves.
Parasympathetic innervation
Parasympathetic autonomic innervation from the
vagus (10th cranial nerve), originating from the
brainstem, supplies the gastrointestinal tract from
the esophagus up to the colonic splenic flexure
(Devroede and Lamarche, 1974) (Fig. 4). Para-
sympathetic innervation to the splenic flexure, de-
scending colon and rectum arises from the sacral
(S) spinal roots S2–S4 that form the pelvic plexus
and give rise to the nervi erigentes. In man, the
precise point at which the vagal innervation to the
bowel stops and pelvic innervation starts is a
source of controversy with some authors describ-
ing vagal innervation down to the rectum, and
others reporting pelvic nerve branches travelling
proximally to innervate the entire colon (Stiens
et al., 1997). The parasympathetic supply to the
internal anal sphincter is derived from the sacral
spinal cord and joins the pelvic nerves. It relaxes

Fig. 3. Layers of the submucosal and myenteric plexuses. From Feldman’s GastroAtlas Online, with permission.
the sphincter by the effect of pre-ganglionic
cholinergic neurones exciting nicotinic and mu-
scarinic receptors.
Sympathetic innervation
Preganglionic axons from thoracic (T) nerve roots
T6–T12 pass via rami communicantes to the sym-
pathetic chain and travel via thoracic splanchnic
nerves, synapsing at the celiac and superior mes-
enteric plexus, supplying small bowel and the as-
cending colon. Sympathetic innervation distal to
the splenic flexure to the upper rectum arises from
lumbar (L) nerve roots L1–L3. The nerves travel
to the sympathetic chain and via the lumbar
splanchnic nerves, synapse at the inferior mesen-
teric ganglia. The supply then follows the arterial
blood supply to the left colon. The lower rectum
and anal canal sympathetic supply is derived from
the aortic and lumbar splanchnics which unite to
form the hypogastric plexus, giving off the pre-
sacral branches to form the sacral plexus, whose
321
postganglionic fibres innervate the rectum and
anus. The internal anal sphincter has sympathetic
supply from the inferior mesenteric ganglion via
the hypogastric nerves. Sympathetic tone is exci-
tatory to the internal sphincter musculature and
helps to maintain continence.
Colonic function, reflexes and control
The colon serves several functions: stool storage;
stool propulsion when socially appropriate; the
provision of an environment for symbiotic bacte-
rial growth; and even absorption of amino acids,
short chain fatty acids and fluid. Following the
disruption of supraspinal control systems, the
dominant autonomic tone is inhibitory to colonic
propulsion, which contributes towards constipa-
tion. The understanding of these neural mecha-
nisms and their aberrant behaviour after spinal
cord injury may provide a basis to treat symptoms.
Normal patterns of colonic contraction can be
classified into three groups: (i) individual phasic
322
Fig. 4. Autonomic innervation of the colon. From Feldman’s GastroAtlas Online, with permission.
contractions (of long or short duration), which
have the effect of kneading and mixing stool; (ii)
organized groups (migratory and non-migratory
motor complexes), which are propulsive in small
regions of colon; (iii) giant migratory contractions,
which produce movements of content and expel
stool during defecation (Christensen, 1991; Sarna,
1993). In addition, distension of the wall of the
colon causes proximal muscle contraction and
distal relaxation, resulting in caudal propagation.
The colon has intrinsic rhythmic slow wave
activity that is thought to be important in encour-
aging fluid reabsorption from the colonic mucosa.
The origin of this activity varies and the
mechanisms controlling it are inadequately under-
stood. Enteric reflexes, with serotonin as the
neurotransmitter, stimulate peristalsis (Hansen,
2003), as demonstrated by its continuation after
the gut is removed from the body. These enteric
reflexes also contribute to the colonic slow wave
activity (Olsson and Holmgren, 2001). Recent
studies have suggested that interstitial cells of
Cajal, found in the submucosa, intra- and inter-
muscle layers of the gut, especially in the right co-
lon, generate spontaneously active pacemaker
currents (Horowitz et al., 1999; Takaki, 2003).
These ‘pacemaker’ cells in the colon are likely to
have profound effects on colonic smooth muscles.

One of the specialized aspects of gastrointestinal
motility is the gastrocolic reflex. This reflex causes
increased small bowel and colonic propulsive mo-
tility, and is mediated by neural and endocrine
mechanisms. The neural substrates comprise cho-
linergic motor neurones which are activated by the
ingestion of a meal (Connell et al., 1963). The
stomach is not the source of the stimulus as the
response is triggered if food bypasses the stomach
(by a feeding tube) and enters the duodenum di-
rectly (Snape, Jr. et al., 1979; Christensen, 1991)
and even by the psychological anticipation or
smell of food. The proposed mechanisms for the
reflex include a role for central vagal mediation,
possibly long enteric reflexes via the enteric nerv-
ous system and humoral components related to
release of cholecystokinin, gastrin and motilin
(Christensen, 1991; Saltzstein et al., 1995). Studies
in spinal cord-injured subjects have demonstrated
differing recorded responses of the reflex, either
showing it to be intact or absent (Glick et al.,
1984). The reasons for this discrepancy may simply
be methodological, and in the clinical setting at
least the reflex is frequently used as a management
tool for treating constipation in cord-injured peo-
ple. By ingesting food or a calorific drink approx-
imately 30 min before bowel management is
planned, reflex colonic contractions can aid stool
emptying (Longo et al., 1989). Fatty foods tend to
have larger and longer action on the reflex com-
pared to protein or carbohydrate-dominant foods
(Spiller, 2000).
Pelvic sacral reflexes are excitatory, with the re-
flex arc conveyed from the sacral spinal cord seg-
ments in the conus, to and from the colon via pelvic
nerves. Parasympathetic stimulation of splanchnic
nerves leads to a significant propulsive colonic re-
sponse. From the colon, enteric nerves trigger this
reflex in response to stretch or dilation, reinforcing
inherent colonic enteric-mediated peristalsis. The
rectocolic reflex is another pelvic reflex that is trig-
gered by mechanical or chemical stimulation in the
rectum or anus. It also produces colonic peristalsis,
which brings stool down to the rectum. Stool
entering the rectum can then trigger the recto-anal
inhibitory reflex, a reflex relaxation of the internal
anal sphincter in response to rectal distension
allowing expulsion of stool from the rectum. The
323
properties of these two reflexes are exploited in
bowel management, to aid defecation, as long as
the spinal cord lesion is above the level of the conus.
Spinal cord injury results in disruption of the
interaction that normally occurs between the in-
trinsic and extrinsic nervous system. Studies look-
ing at the effect of spinal cord and peripheral nerve
lesions on the enteric nervous system have shown
ganglion cell loss and secondary Schwann
cell proliferation in the colon (Devroede and
Lamarche, 1974; Devroede et al., 1979). Once
established, recovery of this disruption is restrict-
ed. However, recent work with enteric glia cells has
shown that they have the potential to aid axonal
growth (Jiang et al., 2003) and may be a source of
future regenerative therapies.
Gastrointestinal dysfunction with acute spinal
cord injury
In the acute phase of spinal cord injury, tonic ex-
citatory input to ganglionic and enteric neurones is
lost and the neurones are less excitable, resulting in
overall lack of neural input to the gut. Commonly
encountered complications in this acute period of
‘spinal shock’ include ileus, gastric dilatation, pep-
tic ulcer disease, pancreatitis and superior mesen-
teric artery syndrome (Tibbs et al., 1979; Gore
et al., 1981; Berlly and Wilmot, 1984). The latter
complication, which is compression of the third
part of the duodenum by the superior mesenteric
artery, tends to occur due to alterations of vascular
tone to the viscera. This spectrum of ‘spinal shock’
abnormalities tends to settle over a varying period
of days to weeks after the initial injury (Ditunno
et al., 2004). Peptic ulceration in the acute setting
is thought to occur as a result of unopposed para-
sympathetic activity from the vagus and transient
loss of sympathetic innervation, resulting in raised
gastrin levels and a reduced pH (Pollock and
Finkelman, 1954; Bowen et al., 1974; Tanaka
et al., 1979). Analgesic and corticosteroid admin-
istration following spinal injury may exacerbate
the condition. These drugs have also been impli-
cated in the prevalence of pancreatitis in acute spi-
nal cord injury. Other causes of pancreatitis which
have been described include autonomic imbalance
324
causing over stimulation of the sphincter of Oddi,
hypercalcaemia due to immobilization and thick-
ened pancreatic secretions (Hyman et al., 1972;
Carey et al., 1977; Maynard and Imai, 1977).
With regard to lower gut function, animal
studies have shown decreased colonic motility
immediately after thoracic cord transection
(Meshkinpour et al., 1985). Inhibitory reflexes
below the lesion are lost and there is a loss of
facilitation from above. This loss of supra-lesional
input to the bowel in large part explains the
reduced transit through the bowel that is found.
Ileus occurs almost immediately in patients with
thoracolumbar cord injury but can be delayed in
high thoracic and cervical nerve injuries. However,
it is most commonly seen in patients with higher
lesions when cord injury occurs at or above the
level of visceral innervation, namely T5.
Gastrointestinal dysfunction with chronic spinal
cord injury
Upper gastrointestinal symptoms
Little work has been published on the extent and
mechanisms of upper gastrointestinal dysfunction
affecting cord-injured people, and much of it is
contradictory. Mild upper gastrointestinal symp-
toms have been reported to affect a third of cord-
injured people (Lu et al., 1998). Heartburn and
dysphagia have been reported in 61 and 30%,
respectively, of injured individuals, which is of
greater prevalence than in matched controls. This
symptom-burden is associated with high levels
of endoscopic and histological evidence of es-
ophagitis (Stinneford et al., 1993). Oesophageal
motility studies also show abnormal slow wave
peristaltic propagation, the equivalent of the
slowed motor abnormalities seen further down
the gastrointestinal tract. The cause and relevance
of these findings remains unknown.
A high prevalence of hiatus hernia is found after
spinal cord injury, which appears to be related to a
reduction of diaphragmatic motion, muscle atro-
phy and weakening of fibrous tissue at the gastro-
oesophageal junction due to chronically raised
intra-abdominal pressures. Treatment of hiatus
hernia and other reflux type symptoms is along the
same lines as in able-bodied patients. The focus is
on antacids (alginates), acid suppressants (hista-
mine (H)2-receptor antagonists and proton pump
inhibitors) and motility stimulating agents (dom-
peridone, metoclopramide).
Nausea and vomiting related to gastric dilata-
tion and ileus are common symptoms in the acute
spinal cord injury setting. Both tend to improve as
spinal shock resolves. Nausea, however, can also
be a persistent and troublesome symptom in the
longer term (Stone et al., 1990a; Glickman and
Kamm, 1996). There are a number of possible
causes for these symptoms: gastric stasis (second-
ary to denervation), gallstone disease (which is
more prevalent in spinal cord injury patients) and
constipation (Camilleri, 1990; Pfeifer et al., 1996;
Tola et al., 2000; Cubeddu, 2003).
Gastric emptying is delayed after spinal cord
injury (Lu et al., 1998; Kao et al., 1999). Vagal
parasympathetic innervation to the upper gastro-
intestinal tract originates from the brainstem and
its control tends to be preserved in spinal cord in-
jury. However, the sympathetic outflow arises
from the thoracic lumbar cord (T5–T12) and its
loss with a lesion about this level results in exces-
sive splanchnic sympathetic activity from the tho-
racic cord, and hence gastroparesis (decreased
gastric emptying). After lower level injuries, there
is no loss of supra-spinal influence and the auto-
nomic hyper-reflexia and delayed gastric emptying
is not seen (Fealey et al., 1984; Nino-Murcia and
Friedland, 1992). There is some evidence to sug-
gest that gastric emptying tends to return towards
normal over time as a degree of autonomic nerv-
ous system-mediated homeostasis and regulation
returns in long-term spinal cord-injured individu-
als (Segal et al., 1987).
Orocecal transit times are delayed after spinal
cord injury (Chen et al., 2004). Using a non-
invasive hydrogen breath test method, these au-
thors have shown that cord-injured subjects have
overall mean orocecal transit times of 180 min com-
pared to 98 min in controls. The net result of this
prolonged small bowel transit time, is to predispose
to disturbance of digestion and bacterial over-
growth which can exacerbate nausea. Nausea has
also been attributed, in part, to the higher incidence
of gallstone disease found in cord-injured patients

(Ketover et al., 1996). Studies looking at the gall-
bladder following spinal cord injury, have shown
reduced contractility and this has been postulated
as possible reason for the increase in incidence
(Fong et al., 2003). Gallstones have been implicated
in causing non-specific symptoms such as nausea
and bloating in cord-injured people (Moonka et al.,
1999). However, given the prevalence of gallstones
after spinal cord injury, the presence of vague ab-
dominal symptoms that occur should not be put
down to the presence of gallstones alone (Moonka
et al., 2000). Gastric or colonic pathology should be
considered in the diagnostic workup.
Pain
Abdominal pain experienced by people after spinal
cord injury needs to be carefully investigated to
exclude common abdominal pathology (such as
neoplasm, peptic ulceration and ischemia), as di-
agnosis in this population can be fraught with dif-
ficulty (Ingersoll, 1985; Bar-On and Ohry, 1995).
Chronic neurological visceral pain does affect
cord-injured people but the extent of this problem
is poorly documented and understood partly as a
result of poor classification (Beric, 2003). Studies
quote the prevalence of chronic visceral pain as
between 3 and 10% (Cardenas et al., 2002). Al-
though not as common as musculoskeletal or ne-
uropathic pain, visceral pain is perceived as being
of higher intensity (severe/excruciating) compared
to musculoskeletal pain. Visceral pain tends to de-
velop months or years after injury, compared with
other pain types that more likely have an early
onset. This probably results following the devel-
opment of visceral organ problems associated with
spinal cord injury such as constipation, bladder
infection and renal calculi. Visceral pain may be
due to normal afferent sensation via the sympa-
thetic and vagus nerves in paraplegics, or vagal
innervation alone in tetraplegics (Richards, 1992;
Siddall and Loeser, 2001).
Lower gastrointestinal symptoms and pathology
Colorectal dysfunction following spinal cord
injury is the major source of gastrointestinal
325
symptoms in these patients. Constipation, fecal
incontinence and incoordinated defecation are the
most frequently reported symptoms. The incidence
of constipation reported in the literature ranges
from 20 to 58%. This discrepancy in the reported
figures can be attributed to a disparity between
definitions of constipation and bowel management
practices used between spinal injury units. Incon-
tinence to feces and flatus is reported to affect up
to 75% of the spinal cord injury population al-
though the percentage of cord-injured people in
whom this occurs on a regular basis (more than
monthly) is only approximately 15% (Krogh et al.,
1997). However, the threat of episodes of fecal in-
continence causes psychological stresses to cord-
injured people and their carers and can result in
social isolation.
Colonic diverticulae are found more frequently
and at younger ages in the spinal cord injury pop-
ulation compared with controls (Gore et al., 1981).
This may be due to the contribution of the high
pressures that uncoordinated segmental peristalsis
can produce and of chronic intraluminal disten-
sion (Gore et al., 1981). Hemorrhoids are also
common with up to three-quarters of cord-
injured people having the problem (Stone et al.,
1990a).
Upper versus lower motor neurone lesions: effect on
the bowel
When describing symptoms attributable to bowel
dysfunction after spinal cord injury, an under-
standing of the effect that the level of injury has on
the bowel is necessary, as this determines the pat-
tern of colonic motility. Upper motor neurone
spinal cord injury lesions occur above the level of
the conus medullaris, which in adults lies at the
lumbar (L)1,2 level. The colon in these cases is
described as ‘spastic’ with increased colonic wall
and striated external anal sphincter muscle tone.
Baseline colonic activity is higher in this group
compared to controls (Aaronson et al., 1985).
Rectal tone is high (Krogh et al., 2002), resulting
in a reduced capacity to hold stool and therefore
increasing the risk of fecal incontinence episodes.
This gives rise to poorly coordinated peristalsis,
with excessive segmental and reduced propulsive
326
peristalsis. The resulting slow whole gut transit
results in constipation that is often exacerbated by
changes in puborectalis muscle function. Evacua-
tion of feces is achieved by triggering reflex def-
ecation either by mechanical (digitation) or
chemical means (suppositories or enemas).
Lower motor neurone lesions occur with injuries
at the level of the conus, cauda equina or pelvic
nerves resulting in the disruption of parasympa-
thetic innervation to the bowel. Loss of parasym-
pathetic control results in a flaccid bowel and low
internal anal sphincter tone. There is also an ab-
sence of spinal cord-mediated reflex peristalsis, and
hence stool propulsion occurs with intrinsic my-
enteric plexus-triggered segmental peristalsis. With
the loss of external anal sphincter control and the
absence of internal anal sphincter parasympathetic
supply, the anal sphincter complex resting tone is
low. This low pressure makes cord-injured people
susceptible to passive fecal leakage. This tendency
is exacerbated by the associated loss of rectal tone,
resulting in a capacious rectum full of stool. Man-
ual removal of stool, aided by increases in intra-
abdominal pressure (such as with a Valsalva) is the
mainstays of management of such people.
Incoordinate anal sphincter function
There are a number of reports in the literature of
abnormal anorectal physiology in spinal cord-
injured people. Resting sphincter tone (mainly a
reflection of internal anal sphincter function) is
reduced in cord-injured people compared to con-
trols, and is maintained mainly by internal anal
sphincter activity possibly due to tonic excitatory
sympathetic discharge (Lynch et al., 2000). The
Valsalva manoeuvre causes a rise in intra-abdom-
inal pressure that is thought to stimulate pelvic
floor tension receptors into triggering reflex exter-
nal anal sphincter contraction (MacDonagh et al.,
1992). In able-bodied individuals, a cortically me-
diated pathway relaxes the external anal sphincter
during straining to defecate. With upper motor
neurone lesions where the reflex pathway is intact
but the supra-lesional input is absent, Valsalva
manoeuvres for bowel emptying may actually
worsen attempts to evacuate stool, as the external
anal sphincter tone increases on straining. The
recto-anal inhibitory reflex described above is
present after spinal cord injury but differs from
controls in that it can be triggered with lower vol-
umes compared to controls. It has been hypoth-
esized that the cause of this could be decreased
rectal compliance (a less distensible rectum) re-
sulting in lower threshold for stimulating the reflex
(Meshkinpour et al., 1983; Glick et al., 1984). This
combination of a less distensible rectum and reflex
anal relaxation contributes towards triggering ep-
isodes of fecal incontinence in spinal cord injury.
Constipation and incontinence
Constipation is common after spinal cord injury
(Glickman and Kamm, 1996). The frequency of
constipation is affected by the level of injury, with
up to three-quarters of quadriplegics being affect-
ed, falling to a third in paraplegics with lesions
between the T10 and L2 cord segments (De Looze
et al., 1998). This is due to delayed colonic transit,
disordered evacuation and changes in visceral sen-
sitivity. Investigating and researching constipation
is difficult since imaging and measuring bowel
motility is not straightforward. However there are
techniques available: radionucleotide and radio-
opaque marker studies (van der Sijp et al., 1993);
solid state pressure catheters (Fajardo et al., 2003);
balloon distension and Barostat recorders
(Bruninga and Camilleri, 1997). Using these tech-
niques, it is possible to understand the patient’s
bowel motility pattern and allow therapeutic in-
terventions to be directed appropriately. Most
studies show overall colonic transit times to be
prolonged following spinal cord injury (Menardo
et al., 1987). Some studies have suggested that this
delay in colonic transit is segmental, most mark-
edly in the distal colon and rectum (Menardo
et al., 1987; Beuret-Blanquart et al., 1990). By
contrast, others show a pan-colonic increase in
transit times (Keshavarzian et al., 1995). In the
clinical setting, groups of cord-injured patients
benefit from oral laxatives in addition to rectal
medications and reflex stimulation, indicating a
reduction in whole colon motility. Poor coordina-
tion of the anal sphincter complex described pre-
viously, leading to outlet obstruction, can also
contribute towards constipation. The importance

of a good bowel care programme, to prevent con-
stipation, is underlined by some studies that show
high levels (73%) of megacolon (dilated colon sec-
ondary to constipation) in people with spinal cord
injury (Harari and Minaker, 2000).
Incontinence is a threat to cord-injured people
due to a combination of factors — lack of aware-
ness of rectal fullness, overflow following poor
evacuation management, and weak sphincter func-
tion as described above (particularly in lower mo-
tor neurone lesion patients). Poor control of flatus
and fecal leakage can lead to physical, psycholog-
ical, sexual and social problems (DeLisa and
Kirshblum, 1997). Effective bowel management
strategies to prevent these symptoms are impor-
tant for the well being of cord-injured people.
Regular evacuation and preventing loose stool
formation by ensuring an adequate fibre intake,
preventing gastrointestinal infections and regulat-
ing diet are core techniques that will be discussed
further later in this chapter.
Therapies that exacerbate symptoms
Spinal cord injuries cause dysfunction to many
organ systems and the treatment of these can result
in adverse effects on the bowel. Side effects from
prescribed medications are very common. Anti-
cholinergics used in the treatment of bladder
dyssynergia, opiates and anti-spasmodics slow
bowel transit and dry the stool thereby exacerbat-
ing the constipation. Broad-spectrum antibiotics
can cause diarrhea by altering the balance of com-
mensal enteric flora in the gut. Additionally, ther-
apies used in bowel management can exacerbate
symptoms. Anal digitation, evacuation and rectal
medication administration can cause local trauma
potentially irritating hemorrhoids and predispos-
ing towards anal fissure and solitary rectal ulcer
formation. All treatments and therapies should
therefore be evaluated for side effects, and vigi-
lance observed for their onset.
Bowel management
Early implementation of a regulated-controlled
bowel management program is held to be the best
practice for patients after their injuries. During the
327
phase of spinal shock when peristalsis is reduced,
digital or manual evacuation of stool is required
(Halm, 1990). When bowel function stabilizes, a
regular bowel care program can be initiated. Cur-
rent programs vary between institutions, where
management is often empirical, given the lack of
well-designed controlled trials (Wiesel et al., 2001).
There have been published proposed bowel pro-
grams (Correa and Rotter, 2000) but none have
been universally adopted and for many cord-in-
jured people, bowel management regimens are far
from ideal. One series reported 41% of cord-in-
jured individuals spending more than 1 h on bowel
evacuation (Harari et al., 1997) and some people
report having to spend 3 h or more a day on their
bowel care. It has been quoted that the ideal bowel
management regime should be self controlled and
spontaneous, with or without oral medication,
performed at least once every 2 days, and com-
pleted within 30 min to result in effective evacua-
tion without complication and this was achieved
by only 32% of subjects in one study (Han et al.,
1998). In the clinical setting, it is accepted that
there have been improvements in bowel care over
the last two decades although hard evidence to
back this up is scarce.
The foundation of good bowel management
program involves the implementation of a regular
routine, which addresses the specific issues such as
constipation, incontinence and functional mobility,
using the appropriate interventions. A bowel care
routine should be timed to coincide with colonic
giant migratory contractions, to take advantage of
any stool propulsion. Giant migratory contrac-
tions tend to occur after meals and in the morning,
on waking. This regime should take into account
the person’s social, sexual, cultural and vocational
beliefs. Also, the question of functional mobility
needs to be addressed, ensuring carers and appro-
priate equipment such as commode chairs are
available. Logistical issues such as access to toilet
facilities, are fundamental but often overlooked.
Diet
Simple dietary measures can benefit bowel man-
agement. Adequate fluid intake aids gut transit by
softening stool. In people with spinal cord injury,
328

fibre does not increase colonic transit time but acts fecal incontinence. Senna may be used as an oc-
to absorb excess water and to keep the stool soft casional night-time dose to ‘prime’ the bowel for a
and formed, thereby reducing the problems of in- morning evacuation. Combinations of the differ-
continence (Banwell et al., 1993). ent classes of laxatives often deliver the desired
results. For example, a regular bulking agent with
Laxatives a stimulant laxative can lead to the regular evac-
uation of soft, formed stools.
There are several groups of medications that can
be taken orally to aid bowel movements or to re-
Suppositories and enemas
lieve constipation. Lubricants and stool softeners
(e.g., docusate sodium, liquid paraffin and mineral
Glycerine suppositories are used to stimulate rectal
oils) ‘grease’ the stool and make passage through
contraction due to its irritant and hyperosmotic
the bowel easier. Bulking agents (e.g., isogel gran-
action and result in bowel movements in
ules, ispaghula husk, methylcellulose, psyllium)
15–30 min. Bisacodyl can be administered in sup-
are indicated if dietary fibre cannot be adequately
pository form and it acts on sensory afferent
ingested. They act by absorbing water in the gut
nerves of the mucosa producing a parasympathet-
thereby softening and bulking stool. Fluid intake
ically mediated reflex peristaltic contraction of the
must be adequate when on these treatments, al-
entire colon (Stiens et al., 1998), which aids bowel
though this can be difficult to achieve in people
emptying and reduces time spent on bowel man-
with bladder management difficulties. Bloating
agement (Frisbie, 1997). Saline, water or docusate
and flatulence can be problematic but tend to set-
sodium enemas can be used. These work by trig-
tle if people can persevere with the treatment.
gering reflex colonic peristalsis, lubricating and in
Indigestible carbohydrates (lactitol, lactulose,
the case of docusate softening the stool. Auto-
polyethylene glycols) and salts [e.g., magnesium
nomic dysreflexia is the condition of abrupt onset,
(Epsom) salts] act osmotically to draw fluid into
potentially lethal hypertension in people with spi-
the colon. Stimulant laxatives (bisacodyl, da-
nal cord injury above the level of T6 (see chapters
nthron, senna) induce and augment peristaltic
addressing cardiovascular dysfunction, this vol-
movement of the bowel, thus aiding stool progres-
ume). It is caused by uncontrolled sympathetic
sion and reducing the time allowed for water and
discharge triggered by any noxious stimulus and
electrolyte resorption. Senna is broken down in
many innocuous stimuli below the level of the le-
absorbable anthraquinones which directly stimu-
sion. Common triggers of this condition relate to
late the myenteric plexus. Oral stimulant laxatives
bladder and bowel distension or irritation (Adsit
can all cause the side effects of cramps, diarrhea
and Bishop, 1995). The maneuvers involved with
and dehydration. Their chronic use can lead to
bowel management such as digitation, use of en-
colonic mucosal staining due to macrophage
emas and evacuation can trigger autonomic
phagocytosis of pigments derived from laxatives
dysreflexia particularly if there is local pathology
(melanosis coli) (Menter et al., 1997). There is no
such as anal fissure or rectal ulceration. Use of
evidence in cord-injured people that their already
local anesthetic agents, such as lidocaine gel can
very slow transit is further compromised by reg-
reduce the incidence of attacks of autonomic
ular use of these agents. The delay of onset for
dysreflexia during bowel management.
these laxatives is 1–2 days, except for magnesium
salts that have a faster onset of action of about 4 h
(Frisbie, 1997; Amir et al., 1998). The attraction of Prokinetic agents
osmotic agents is their speed of action and their
ability to be titrated, done according to stool con- Metoclopramide and domperidone are dopamine
sistency. Care must be taken however to avoid too antagonists that increase the rate of gastric emp-
loose a stool, especially in people with sphincter tying and of small gut transit. They have no effect
compromise, as they may lead frequent episodes of on colonic peristalsis and are most commonly used

in the acute setting of spinal cord injury when try-
ing to overcome the gastric dilatation and ileus that
accompanies spinal shock (Miller and Fenzl, 1981;
Segal et al., 1987). Erythromycin is a macrolide
antibiotic with prokinetic effects which is also used
in the acute setting to enhance transit through the
upper gut (Clanton and Bender, 1999).
The parasympathomimetic drugs neostigmine,
bethanechol, distigmine and pyridostigmine, all
enhance parasympathetic effects on the gut to in-
crease motility but are rarely used in the clinical
setting for this purpose, due to their side effects.
They may have a role in treating the rare situation
of acute pseudo-obstruction of the gut (Ogilvie’s
Syndrome) which is seen in some cases after acute
injury, related to sudden loss of autonomic tone to
the viscera (Delgado-Aros and Camilleri, 2003).
Cisapride was used for its prokinetic properties on
the upper and lower gut and increased colonic
transit speed (Binnie et al., 1988; Geders et al.,
1995; Longo et al., 1995) but has been withdrawn
from clinical use because of an association with
fatal cardiac arrhythmias (Prescrire Int, 2000;
Cubeddu, 2003).
Mechanical devices and surgical interventions
Anal plug devices can be utilized to prevent leak-
age of flatus and feces (Kim et al., 2001) for cord-
injured people with lower motor neurone lesions,
who often have an atonic anal sphincter. They are
best tolerated by people who have no preservation
of anal sensation, but tend to be inefficient if large
volumes of stool are being lost. Pulsed irrigation
enemas have been used, in which a catheter with
an inflatable retention cuff is passed into the rec-
tum followed by a program of tap water pulses
(Puet et al., 1997). This loosens and suspends stool
that is removed via a conduit drain running
through the centre of the catheter. Antegrade con-
tinence enemas require appendicocecostomies to
be surgically fashioned to deliver washout fluid
into the proximal colon to allow controlled daily
emptying. Water or saline is infused into the
cecum and passes through to produce bowel evac-
uation minutes later (Malone, 2004). The tech-
nique has been modified to allow radiological
329
(Chait et al., 1997) or endoscopic placement (De
Peppo et al., 1999). These irrigation methods have
been most studied in pediatric practice, especially
in children with myelomeningocoele. While often
efficient in the short term, infective and mechanical
complications around the tube entry site can be
problematic. Furthermore, there is evidence that
with time, the irrigation method may become less
efficient, and indeed the antegrade continence en-
ema openings in the abdominal wall frequently
stenose (McAndrew and Malone, 2002).
Severe refractory constipation, prolonged bowel
care time, fecal incontinence and chronic peri-anal
ulcers are reasons for cord-injured people to con-
sider stoma formation for their bowel care
(Deshmukh et al., 1996; Pfeifer et al., 1996). How-
ever, the decision to opt for surgery should not be
taken lightly as there are issues of assessment of
the current bowel care program, body image, life-
style, and required nursing assistance to be taken
into account together with the high risks of surgery
in this group of patients. That said, the formation
of a stoma (ileostomy or colostomy) has been
shown in several studies to improve the quality of
life for people who opt for this treatment option as
it can simplify bowel management, reduce incon-
tinence and bloating and increase independence
(Stone et al., 1990b; Randell et al., 2001; Branagan
et al., 2003). Choice of type and position of stoma
should be assessed according to the person, de-
pendent upon their colonic transit characteristics
and their mobility (Safadi et al., 2003).
Sacral anterior root stimulators have been
implanted since 1977 (Brindley et al., 1986), orig-
inally being used for functional electrical stimula-
tion to control bladder emptying (Brindley and
Rushton, 1990; Binnie et al., 1991; Brindley, 1994).
The implant consists of a subcutaneous radio-re-
ceiver connected to S2, S3 and S4 nerve roots via
tunnelled wires. Trains of high-frequency stimula-
tion trigger complex, high pressure, phasic colonic
and rectal contractions resembling peristaltic
movements, which act to bring stool down distal-
ly. Following stimulation, defecation may occur
(Varma et al., 1986; MacDonagh et al., 1990;
Varma, 1992) and indeed this procedure was found
to improve bowel management in addition to
bladder care in some patients.
330
Conclusion
Gastrointestinal symptoms and the required pro-
cedures for management of the bowel in people
with spinal cord injury are very problematic and
distressing. Unlike the advances in limb and blad-
der dysfunction, the understanding of the effects of
spinal cord injury on the bowel is still very poor.
Improvement in the quality of life for this group of
people requires on-going basic clinical research in
this field. Greater understanding of the influence
of neural disconnection on the residual function of
the gut (via the enteric nervous system) is needed.
Specifically, the understanding of the role of pelvic
reflexes in controlling evacuation and continence is
required to offer the prospect of possible future
neuromodulation of these reflexes. Finally, under-
standing of the quality of life implications of bowel
dysfunction and developments of simple means to
remedy the socially isolating problems is required.
The above necessitates a combination of labora-
tory and clinical research, in addition to advances
in nursing care. All of this requires a greater pri-
ority to be placed on the research and the clinical
agenda for spinally injured people.
References
Aaronson, M.J., Freed, M.M. and Burakoff, R. (1985) Colonic
myoelectric activity in persons with spinal cord injury. Dig.
Dis. Sci., 30: 295–300.
Adsit, P.A. and Bishop, C. (1995) Autonomic dysreflexia —
don’t let it be a surprise. Orthop. Nurs., 14: 17–20.
Amir, I., Sharma, R., Bauman, W.A. and Korsten, M.A. (1998)
Bowel care for individuals with spinal cord injury: compar-
ison of four approaches. J. Spinal Cord Med., 21: 21–24.
Banwell, J.G., Creasey, G.H., Aggarwal, A.M. and Mortimer,
J.T. (1993) Management of the neurogenic bowel in patients
with spinal cord injury. Urol. Clin. North Am., 20: 517–526.
Bar-On, Z. and Ohry, A. (1995) The acute abdomen in spinal
cord injury individuals. Paraplegia, 33: 704–706.
Beric, A. (2003) Spinal cord injury pain. Eur. J. Pain, 7:
335–338.
Berlly, M.H. and Wilmot, C.B. (1984) Acute abdominal emer-
gencies during the first four weeks after spinal cord injury.
Arch. Phys. Med. Rehabil., 65: 687–690.
Beuret-Blanquart, F., Weber, J., Gouverneur, J.P.,
Demangeon, S. and Denis, P. (1990) Colonic transit time
and anorectal manometric anomalies in 19 patients with
complete transection of the spinal cord. J. Auton. Nerv.
Syst., 30: 199–207.
Binnie, N.R., Creasey, G.H., Edmond, P. and Smith, A.N.
(1988) The action of cisapride on the chronic constipation of
paraplegia. Paraplegia, 26: 151–158.
Binnie, N.R., Smith, A.N., Creasey, G.H. and Edmond, P.
(1991) Constipation associated with chronic spinal cord in-
jury: the effect of pelvic parasympathetic stimulation by the
Brindley stimulator. Paraplegia, 29: 463–469.
Bowen, J.C., Fleming, W.H. and Thompson, J.C. (1974) In-
creased gastrin release following penetrating central nervous
system injury. Surgery, 75: 720–724.
Branagan, G., Tromans, A. and Finnis, D. (2003) Effect of
stoma formation on bowel care and quality of life in patients
with spinal cord injury. Spinal Cord, 41: 680–683.
Brindley, G.S. (1994) The first 500 patients with sacral anterior
root stimulator implants: general description. Paraplegia, 32:
795–805.
Brindley, G.S., Polkey, C.E., Rushton, D.N. and Cardozo, L.
(1986) Sacral anterior root stimulators for bladder control in
paraplegia: the first 50 cases. J. Neurol. Neurosurg. Psychiat.,
49: 1104–1114.
Brindley, G.S. and Rushton, D.N. (1990) Long-term follow-up
of patients with sacral anterior root stimulator implants.
Paraplegia, 28: 469–475.
Bruninga, K. and Camilleri, M. (1997) Colonic motility and
tone after spinal cord and cauda equina injury. Am. J.
Gastroenterol., 92: 891–894.
Camilleri, M. (1990) Disorders of gastrointestinal motility in
neurologic diseases. Mayo Clin. Proc., 65: 825–846.
Cardenas, D.D., Turner, J.A., Warms, C.A. and Marshall, H.M.
(2002) Classification of chronic pain associated with spinal
cord injuries. Arch. Phys. Med. Rehabil., 83: 1708–1714.
Carey, M.E., Nance, F.C., Kirgis, H.D., Young, H.F., Megison
Jr., L.C. and Kline, D.G. (1977) Pancreatitis following spinal
cord injury. J. Neurosurg., 47: 917–922.
Chait, P.G., Shandling, B., Richards, H.M. and Connolly, B.L.
(1997) Fecal incontinence in children: treatment with percu-
taneous cecostomy tube placement — a prospective study.
Radiology, 203: 621–624.
Chen, C.Y., Chuang, T.Y., Tsai, Y.A., Tai, H.C., Lu, C.L., Kang,
L.J., Lu, R.H., Chang, F.Y. and Lee, S.D. (2004) Loss of sym-
pathetic coordination appears to delay gastrointestinal transit
in patients with spinal cord injury. Dig. Dis. Sci., 49: 738–743.
Christensen, J. (1991) The motor function of the colon. In:
Yamada T. (Ed.), Textbook of Gastroenterology. Lippincott,
Philadelphia, pp. 180–196.
Clanton Jr., L.J. and Bender, J. (1999) Refractory spinal cord
injury induced gastroparesis: resolution with erythromycin
lactobionate, a case report. J. Spinal Cord Med., 22: 236–238.
Connell, A.M., Frankel, H. and Guttman, L. (1963) The
motility of the pelvic colon following complete lesions of the
spinal cord. Paraplegia, 104: 98–115.
Correa, G.I. and Rotter, K.P. (2000) Clinical evaluation and
management of neurogenic bowel after spinal cord injury.
Spinal Cord, 38: 301–308.
Cosman, B.C., Stone, J.M. and Perkash, I. (1991) Gastrointes-
tinal complications of chronic spinal cord injury. J. Am.
Paraplegia Soc., 14: 175–181.

Cubeddu, L.X. (2003) QT prolongation and fatal arrhythmias:
a review of clinical implications and effects of drugs. Am. J.
Ther., 10: 452–457.
De Looze, D., Van Laere, M., De Muynck, M., Beke, R. and
Elewaut, A. (1998) Constipation and other chronic gastro-
intestinal problems in spinal cord injury patients. Spinal
Cord, 36: 63–66.
De Peppo, F., Iacobelli, B.D., De Gennaro, M., Colajacomo,
M. and Rivosecchi, M. (1999) Percutaneous endoscopic
cecostomy for antegrade colonic irrigation in fecally incon-
tinent children. Endoscopy, 31: 501–503.
Delgado-Aros, S. and Camilleri, M. (2003) Clinical manage-
ment of acute colonic pseudo-obstruction in patients: a sys-
tematic review of the literature. Gastroenterol. Hepatol., 26:
646–655.
DeLisa, J.A. and Kirshblum, S. (1997) A review: frustrations
and needs in clinical care of spinal cord injury patients.
J. Spinal. Cord Med., 20: 384–390.
Deshmukh, G.R., Barkel, D.C., Sevo, D. and Hergenroeder, P.
(1996) Use or misuse of colostomy to heal pressure ulcers.
Dis. Colon Rectum, 39: 737–738.
Devroede, G., Arhan, P., Duguay, C., Tetreault, L., Akoury, H.
and Perey, B. (1979) Traumatic constipation. Gastroenterol-
ogy, 77: 1258–1267.
Devroede, G. and Lamarche, J. (1974) Functional importance
of extrinsic parasympathetic innervation to the distal colon
and rectum in man. Gastroenterology, 66: 273–280.
Ditunno, J.F., Little, J.W., Tessler, A. and Burns, A.S. (2004)
Spinal shock revisited: a four-phase model. Spinal Cord, 42:
383–395.
Fajardo, N.R., Pasiliao, R.V., Modeste-Duncan, R., Creasey,
G., Bauman, W.A. and Korsten, M.A. (2003) Decreased co-
lonic motility in persons with chronic spinal cord injury. Am.
J. Gastroenterol., 98: 128–134.
Fealey, R.D., Szurszewski, J.H., Merritt, J.L. and DiMagno,
E.P. (1984) Effect of traumatic spinal cord transection on
human upper gastrointestinal motility and gastric emptying.
Gastroenterology, 87: 69–75.
Fong, Y.C., Hsu, H.C., Sun, S.S., Kao, A., Lin, C.C. and Lee,
C.C. (2003) Impaired gallbladder function in spinal cord in-
jury on quantitative Tc-99m DISIDA cholescintigraphy. Ab-
dom. Imaging, 28: 87–91.
Frisbie, J.H. (1997) Improved bowel care with a polyethylene
glycol based bisacadyl suppository. J. Spinal Cord Med., 20:
227–229.
Furness, J.B. and Costa, M. (1987) The Enteric Nervous Sys-
tem. Churchill Livingstone, Edinburgh.
Geders, J.M., Gaing, A., Bauman, W.A. and Korsten, M.A.
(1995) The effect of cisapride on segmental colonic transit
time in patients with spinal cord injury. Am. J. Gas-
troenterol., 90: 285–289.
Glick, M.E., Meshkinpour, H., Haldeman, S., Hoehler, F.,
Downey, N. and Bradley, W.E. (1984) Colonic dysfunction in
patients with thoracic spinal cord injury. Gastroenterology,
86: 287–294.
Glickman, S. and Kamm, M.A. (1996) Bowel dysfunction in
spinal-cord-injury patients. Lancet, 347: 1651–1653.
331
Gore, R.M., Mintzer, R.A. and Calenoff, L. (1981) Gastroin-
testinal complications of spinal cord injury. Spine, 6:
538–544.
Goyal, R.K. and Hirano, I. (1996) The enteric nervous system.
N. Engl. J. Med., 334: 1106–1115.
Halm, M.A. (1990) Elimination concerns with acute spinal cord
trauma. Assessment and nursing interventions. Crit. Care
Nurs. Clin. North Am., 2: 385–398.
Han, T.R., Kim, J.H. and Kwon, B.S. (1998) Chronic gastro-
intestinal problems and bowel dysfunction in patients with
spinal cord injury. Spinal Cord, 36: 485–490.
Hansen, M.B. (2003) Neurohumoral control of gastrointestinal
motility. Physiol. Res., 52: 1–30.
Harari, D. and Minaker, K.L. (2000) Megacolon in patients
with chronic spinal cord injury. Spinal Cord, 38: 331–339.
Harari, D., Sarkarati, M., Gurwitz, J.H., McGlinchey-Berroth,
G. and Minaker, K.L. (1997) Constipation-related symptoms
and bowel program concerning individuals with spinal cord
injury. Spinal Cord, 35: 394–401.
Harvey, C., Wilson, S.E., Greene, C.G., Berkowitz, M. and
Stripling, T.E. (1992) New estimates of the direct costs of
traumatic spinal cord injuries: results of a nationwide survey.
Paraplegia, 30: 834–850.
Horowitz, B., Ward, S.M. and Sanders, K.M. (1999) Cellular
and molecular basis for electrical rhythmicity in gastrointes-
tinal muscles. Annu. Rev. Physiol., 61: 19–43.
Hyman, L.R., Boner, G., Thomas, J.C. and Segar, W.E. (1972)
Immobilization hypercalcemia. Am. J. Dis. Child, 124: 723–727.
Ingersoll, G.L. (1985) Abdominal pathology in spinal cord in-
jured persons. J. Neurosurg. Nurs., 17: 343–348.
Jiang, S., Wang, J., Khan, M.I., Middlemiss, P.J., Salgado-
Ceballos, H., Werstiuk, E.S., Wickson, R. and Rathbone,
M.P. (2003) Enteric glia promote regeneration of transected
dorsal root axons into spinal cord of adult rats. Exp. Neurol.,
181: 79–83.
Johnson, R.L., Brooks, C.A. and Whiteneck, G.G. (1996) Cost
of traumatic spinal cord injury in a population-based regis-
try. Spinal Cord, 34: 470–480.
Kao, C.H., Ho, Y.J., Changlai, S.P. and Ding, H.J. (1999)
Gastric emptying in spinal cord injury patients. Dig. Dis. Sci.,
44: 1512–1515.
Keshavarzian, A., Barnes, W.E., Bruninga, K., Nemchausky,
B., Mermall, H. and Bushnell, D. (1995) Delayed colonic
transit in spinal cord-injured patients measured by indium-
111 Amberlite scintigraphy. Am. J. Gastroenterol., 90:
1295–1300.
Ketover, S.R., Ansel, H.J., Goldish, G., Roche, B. and Geb-
hard, R.L. (1996) Gallstones in chronic spinal cord injury: is
impaired gallbladder emptying a risk factor? Arch. Phys.
Med. Rehabil., 77: 1136–1138.
Kim, J., Shim, M.C., Choi, B.Y., Ahn, S.H., Jang, S.H. and
Shin, H.J. (2001) Clinical application of continent anal plug
in bedridden patients with intractable diarrhea. Dis. Colon
Rectum, 44: 1162–1167.
Krogh, K., Mosdal, C., Gregersen, H. and Laurberg, S. (2002)
Rectal wall properties in patients with acute and chronic
spinal cord lesions. Dis. Colon Rectum, 45: 641–649.
332
Krogh, K., Nielsen, J., Djurhuus, J.C., Mosdal, C., Sabroe, S.
and Laurberg, S. (1997) Colorectal function in patients with
spinal cord lesions. Dis. Colon Rectum, 40: 1233–1239.
Longo, W.E., Ballantyne, G.H. and Modlin, I.M. (1989) The
colon, anorectum and spinal cord patient. A review of the
functional alterations of the denervated hindgut. Dis. Colon
Rectum, 32: 261–267.
Longo, W.E., Woolsey, R.M., Vernava, A.M., Virgo, K.S.,
McKirgan, L. and Johnson, F.E. (1995) Cisapride for con-
stipation in spinal cord injured patients: a preliminary report.
J. Spinal Cord Med., 18: 240–244.
Lu, C.L., Montgomery, P., Zou, X., Orr, W.C. and Chen, J.D.
(1998) Gastric myoelectrical activity in patients with cervical
spinal cord injury. Am. J. Gastroenterol., 93: 2391–2396.
Lynch, A.C., Anthony, A., Dobbs, B.R. and Frizelle, F.A.
(2000) Anorectal physiology following spinal cord injury.
Spinal Cord, 38: 573–580.
Lynch, A.C., Antony, A., Dobbs, B.R. and Frizelle, F.A. (2001)
Bowel dysfunction following spinal cord injury. Spinal Cord,
39: 193–203.
MacDonagh, R., Sun, W.M., Thomas, D.G., Smallwood, R.
and Read, N.W. (1992) Anorectal function in patients with
complete supraconal spinal cord lesions. Gut, 33: 1532–1538.
MacDonagh, R.P., Sun, W.M., Smallwood, R., Forster, D. and
Read, N.W. (1990) Control of defecation in patients with
spinal injuries by stimulation of sacral anterior nerve roots.
BMJ, 300: 1494–1497.
Malone, P.S. (2004) The antegrade continence enema proce-
dure. BJU Int., 93: 248–249.
Maynard, F.M. and Imai, K. (1977) Immobilization hype-
rcalcemia in spinal cord injury. Arch. Phys. Med. Rehabil.,
58: 16–24.
McAndrew, H.F. and Malone, P.S. (2002) Continent catheter-
izable conduits: which stoma, which conduit and which res-
ervoir? BJU Int., 89: 86–89.
Menardo, G., Bausano, G., Corazziari, E., Fazio, A.,
Marangi, A., Genta, V. and Marenco, G. (1987) Large-bowel
transit in paraplegic patients. Dis. Colon Rectum, 30: 924–928.
Menter, R., Weitzenkamp, D., Cooper, D., Bingley, J.,
Charlifue, S. and Whiteneck, G. (1997) Bowel management
outcomes in individuals with long-term spinal cord injuries.
Spinal Cord, 35: 608–612.
Meshkinpour, H., Harmon, D., Thompson, R. and Yu, J.
(1985) Effects of thoracic spinal cord transection on colonic
motor activity in rats. Paraplegia, 23: 272–276.
Meshkinpour, H., Nowroozi, F. and Glick, M.E. (1983)
Colonic compliance in patients with spinal cord injury. Arch.
Phys. Med. Rehabil., 64: 111–112.
Middleton, J.W., Lim, K., Taylor, L., Soden, R. and Rutkowski,
S. (2004) Patterns of morbidity and rehospitalisation follow-
ing spinal cord injury. Spinal Cord, 42: 359–367.
Miller, B.J., Geraghty, T.J., Wong, C.H., Hall, D.F. and
Cohen, J.R. (2001) Outcome of the acute abdomen in patients
with previous spinal cord injury. ANZ J. Surg., 71: 407–411.
Miller, F. and Fenzl, T.C. (1981) Prolonged ileus with acute
spinal cord injury responding to metaclopramide. Paraplegia,
19: 43–45.
Moonka, R., Stiens, S.A., Eubank, W.B. and Stelzner, M.
(1999) The presentation of gallstones and results of biliary
surgery in a spinal cord injured population. Am. J. Surg.,
178: 246–250.
Moonka, R., Stiens, S.A. and Stelzner, M. (2000) Atypical
gastrointestinal symptoms are not associated with gallstones
in patients with spinal cord injury. Arch. Phys. Med. Re-
habil., 81: 1085–1089.
Nino-Murcia, M. and Friedland, G.W. (1992) Functional ab-
normalities of the gastrointestinal tract in patients with spinal
cord injuries: evaluation with imaging procedures. Am. J.
Roentgenol., 158: 279–281.
Olsson, C. and Holmgren, S. (2001) The control of gut motility.
Comp. Biochem. Physiol. A Mol. Integr. Physiol., 128: 481–503.
Pfeifer, J., Agachan, F. and Wexner, S.D. (1996) Surgery for
constipation: a review. Dis. Colon Rectum, 39: 444–460.
Pollock, L.J. and Finkelman, I. (1954) The digestive apparatus
in injuries to the spinal cord and cauda equina. Surg. Clin.
North Am., 98: 259–268.
Prescrire, Int. (2000) Severe cardiac arrythmia on cisapride.
Prescrire Int., 9: 144–145.
Puet, T.A., Jackson, H. and Amy, S. (1997) Use of pulsed ir-
rigation evacuation in the management of the neuropathic
bowel. Spinal Cord, 35: 694–699.
Randell, N., Lynch, A.C., Anthony, A., Dobbs, B.R., Roake,
J.A. and Frizelle, F.A. (2001) Does a colostomy alter quality
of life in patients with spinal cord injury? A controlled study.
Spinal Cord, 39: 279–282.
Richards, J.S. (1992) Chronic pain and spinal cord injury: re-
view and comment. Clin. J. Pain, 8: 119–122.
Safadi, B.Y., Rosito, O., Nino-Murcia, M., Wolfe, V.A. and
Perkash, I. (2003) Which stoma works better for colonic
dysmotility in the spinal cord injured patient? Am. J. Surg.,
186: 437–442.
Saltzstein, R.J., Mustin, E. and Koch, T.R. (1995) Gut hor-
mone release in patients after spinal cord injury. Am. J. Phys.
Med. Rehabil., 74: 339–344.
Sarna, S.K. (1991) Physiology and pathophysiology of colonic
motor activity (2). Dig. Dis. Sci., 36: 998–1018.
Sarna, S.K. (1993) Colonic motor activity. Surg. Clin. North
Am., 73: 1201–1223.
Schemann, M. and Neunlist, M. (2004) The human enteric nerv-
ous system. Neurogastroenterol. Motil., 16(Suppl. 1): 55–59.
Schweiger, M. (1979) Method for determining individual con-
tributions of voluntary and involuntary anal sphincters to
resting tone. Dis. Colon Rectum, 22: 415–416.
Segal, J.L., Milne, N., Brunnemann, S.R. and Lyons, K.P.
(1987) Metoclopramide-induced normalization of impaired
gastric emptying in spinal cord injury. Am. J. Gastroenterol.,
82: 1143–1148.
Siddall, P.J. and Loeser, J.D. (2001) Pain following spinal cord
injury. Spinal Cord, 39: 63–73.
Sinnatamby, C.S. (1999) Last’s Anatomy, Regional and Ap-
plied (10th ed.). Churchill Livingstone, Edinburgh.
Snape Jr., W.J., Wright, S.H., Battle, W.M. and Cohen, S.
(1979) The gastrocolic response: evidence for a neural mech-
anism. Gastroenterology, 77: 1235–1240.

Spiller, R.C. (2000) Haute cuisine and the colon. Gut, 46:
150–151.
Stiens, S.A., Bergman, S.B. and Goetz, L.L. (1997) Neurogenic
bowel dysfunction after spinal cord injury: clinical evaluation
and rehabilitative management. Arch. Phys. Med. Rehabil.,
S86–S102.
Stiens, S.A., Luttrel, W. and Binard, J.E. (1998) Polyethylene
glycol versus vegetable oil based bisacodyl suppositories to
initiate side-lying bowel care: a clinical trial in persons with
spinal cord injury. Spinal Cord, 36: 777–781.
Stinneford, J.G., Keshavarzian, A., Nemchausky, B.A., Doria,
M.I. and Durkin, M. (1993) Esophagitis and esophageal
motor abnormalities in patients with chronic spinal cord in-
juries. Paraplegia, 31: 384–392.
Stone, J.M., Nino-Murcia, M., Wolfe, V.A. and Perkash, I.
(1990a) Chronic gastrointestinal problems in spinal cord in-
jury patients: a prospective analysis. Am. J. Gastroenterol.,
85: 1114–1119.
Stone, J.M., Wolfe, V.A., Nino-Murcia, M. and Perkash, I.
(1990b) Colostomy as treatment for complications of spinal
cord injury. Arch. Phys. Med. Rehabil., 71: 514–518.
Takaki, M. (2003) Gut pacemaker cells: the interstitial cells of
Cajal (ICC). J. Smooth Muscle Res., 39: 137–161.
333
Tanaka, M., Uchiyama, M. and Kitano, M. (1979) Gastrodu-
odenal disease in chronic spinal cord injuries. An endoscopic
study. Arch. Surg., 114: 185–187.
Tibbs, P.A., Bivins, B.A. and Young, A.B. (1979) The problem
of acute abdominal disease during spinal shock. Am. Surg.,
45: 366–368.
Tola, V.B., Chamberlain, S., Kostyk, S.K. and Soybel, D.I.
(2000) Symptomatic gallstones in patients with spinal cord
injury. J. Gastrointest. Surg., 4: 642–647.
van der Sijp, J.R., Kamm, M.A., Nightingale, J.M., Britton,
K.E., Mather, S.J., Morris, G.P., Akkermans, L.M. and
Lennard-Jones, J.E. (1993) Radioisotope determination of
regional colonic transit in severe constipation: comparison
with radio opaque markers. Gut, 34: 402–408.
Varma, J.S. (1992) Autonomic influences on colorectal motility
and pelvic surgery. World J. Surg., 16: 811–819.
Varma, J.S., Binnie, N., Smith, A.N., Creasey, G.H. and
Edmond, P. (1986) Differential effects of sacral anterior root
stimulation on anal sphincter and colorectal motility in spin-
ally injured man. Br. J. Surg., 73: 478–482.
Wiesel, P.H., Norton, C. and Brazzelli, M. (2001) Management
of faecal incontinence and constipation in adults with central
neurological diseases. Cochrane Database. Syst. Rev., 4:
CD002115.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 22

Colorectal motility and defecation after spinal cord

injury in humans

A.C. Lynch and F.A. Frizelle

Colorectal Unit, Department of Surgery, Christchurch Hospital and Burwood Spinal Unit, Christchurch, New Zealand

Abstract: Following spinal cord injury, colorectal problems are a significant cause of morbidity, and
chronic gastrointestinal problems remain common with increasing time after injury. Although many cord-
injured patients achieve an adequate bowel frequency with drugs and manual stimulation, the risk and
occurrence of fecal incontinence, difficulties with evacuation, and need for assistance remain significant
problems. The underlying physiology of colorectal motility and defecation is reviewed, and consequences of
spinal cord injury on defecation are reported. A discussion of present management techniques is under-
taken and new directions in management and research are suggested. There is need for more intervention in
regard to bowel function that could improve quality of life, but there is also a need for more research in this
area.

Introduction with increasing time after injury (Stone et al.,

Bowel dysfunction is a problem following spinal
cord injury
Bowel dysfunction following spinal cord injury is
increasingly recognized as an area of major phys-
ical and psychological difficulty. Surveys of spinal
cord-injured people show that bowel function is as
much of a problem as loss of mobility or sexual
function. The problem is twofold as, not only does
spinal cord injury result in changes to bowel mo-
tility and sphincter control, but also the concur-
rent loss of mobility and gross motor dexterity
makes bowel management a major life-limiting
problem.
Immediately after spinal cord injury, colorectal
problems are a significant cause of morbidity, and
chronic gastrointestinal problems remain common
Corresponding author. Tel.: +64-3-3640-640;
Fax: +64-3-3640-352; E-mail: frank.frizelle@chmeds.ac.nz
1990a). The inability to defecate normally means
that bowel care often occupies a significant part of
the day, and, although many cord-injured people
achieve an adequate bowel frequency with drugs
and manual stimulation, the risk and occurrence
of fecal incontinence, difficulties with evacuation,
and need for assistance remain significant life-lim-
iting problems (Stone et al., 1990a, b; Levi et al.,
1995; Glickman and Kamm, 1996; Han et al.,
1998; Lynch et al., 2001).
Long-term gastrointestinal complications can
develop in cord-injured people. Fecal impaction is
common. Diverticular disease and volvulus are
more frequent and are perhaps related to higher
intracolonic pressures in those with upper motor
neuron lesions. The occurrence of hemorrhoids
and mucosal prolapse was also identified by Lynch
et al. (2000c) as occurring more frequently after
spinal cord injury by an incidence of hemorrhoi-
dectomy of 9% compared with a control group
incidence of 1.5% (po0.001). This may be

DOI: 10.1016/S0079-6123(05)52022-3 335

336
multifactorial due to altered anorectal tone or
trauma with manual evacuation, and is a frequent
source of bleeding or autonomic dysreflexia.
Impact on lifestyle
Spinal cord-injured people rate difficulties with
bowel management as similar to problems associ-
ated with loss of mobility and sexual function.
Hanson and Franklin (1976) reported that 80% of
male paraplegics and 46% of male tetraplegics
would rank bladder and bowel as their greatest
functional loss after loss of mobility. It was inter-
esting that when they asked the same question to
spinal unit staff, only 39% ranked bladder and
bowel problems as high.
Toileting
Survey data shows that 61% of cord-injured peo-
ple would spend more than 15 min per day toile-
ting, compared with only 9% of controls (Lynch
et al., 2001). Those doing manual evacuations
spend the longest time. Half of all cord-injured
people need assistance with toileting. The need for
assistance with toileting relates very closely with
level of injury and has implications for provision
of carers and dependence on family members. It is
recognised that having family members perform
such intimate tasks can be emotionally charged
and negatively affect family interrelationships.
This was significantly associated with the report-
ed perception that bowel function was a source of
distress.
Colonic function following spinal cord injury
Questionnaires exploring bowel function in spinal
injured people have found that over half of those
with an injury above the second lumbar segment
(L2) suffer from constipation (DeLooze et al.,
1998). People with higher injuries defecate less
frequently compared to those with lower injuries
and the general population. It is apparent that
even with medications and other methods, spinal
cord-injured people do not achieve a bowel motion
frequency similar to a control population. Chang-
es to the extrinsic autonomic innervation of the
bowel are presumed to decrease the normal post-
prandial increase in motility and to decrease
colonic compliance. Laxative use among the
cord-injured population is common as a means
of regulating bowel habit. Two-thirds of those
with high injuries report using laxatives either
occasionally or regularly, compared to the 4% in a
control group drawn from the general population
(po0.0001, Fisher’s exact test, Lynch et al.,
2000c). However, the diarrhea produced by laxa-
tives and the resultant risk of incontinence may
limit their use for some.
Patterns of gut dysmotility have been described
for different levels and degrees of spinal cord in-
jury with the level of the spinal cord lesion deter-
mining the effect on colonic motility. Marker
transit studies show that lesions above the first
thoracic segment (T1) result in delayed mouth-to-
caecum time, but lesions below this level result in
normal transit times to the caecum. Beyond the
ileocaecal valve, transit times are markedly de-
layed (Menardo et al., 1987). For people with an
upper motor neuron lesion, transit studies and
scintigraphy have demonstrated variable changes
in colonic transit. If the spinal cord lesion is above
the lumbar region, transit is slowed throughout the
whole colon. The velocity of the median position
of bowel contents throughout the colon was sig-
nificantly slower in cord-injured people (0.637
0.33 cm/h in cord-injured; 2.5871.2 cm/h in con-
trols, po0.001). One study by Nino-Murcia et al.
(1990), involving 28 cord-injured subjects, also
demonstrated distal small bowel dilatation in 10
people, all of whom had abdominal symptoms and
9 of whom had a spinal cord lesion above T5. A
lower motor neuron injury from a lesion affecting
the conus, cauda equina or pelvic nerves results in
interruption of the parasympathetic supply to the
colon and reduced spinal cord-mediated reflex
peristalsis. Stool propulsion is by segmental co-
lonic peristalsis only.
The mechanism for colonic dysmotility follow-
ing a spinal cord injury may be a loss of descend-
ing inhibitory modulation from the sympathetic
nervous system. This theory is supported by
studies in the cat by Gillis et al. (1987) in which

a2-adrenergic receptor activation resulted in pro-
found inhibition of colonic motility, and section-
ing of splanchnic nerves (containing preganglionic
sympathetic innervation of the intestine) produced
an increase in colonic contraction.
In cord-injured people, colonic transit delays are
more profound in higher injuries, where the sym-
pathetic injury should be more pronounced. The
delay may in part be due to loss of colonic com-
pliance. With a spinal cord lesion above L1, the
left colon has an abnormal response to increasing
volume. Distension with water produces a pres-
sure–volume curve (colometrogram) showing a
steep increase in intracolonic pressure with in-
creasing volume. This is similar to the hype-
rreflexic response described by Meshkinpour
et al. (1983) during bladder cystometry for inju-
ries at a similar level. For injuries above T5, the
right colon is also affected. The lack of compliance
leads to functional obstruction, increased transit
times, abdominal distension, bloating and discom-
fort. It suggests that the central nervous system
(CNS) is necessary to modulate colonic motility.
Colonic myoelectric activity has been recorded in a
group of spinal injury subjects with injuries at
varying levels and controls. This demonstrated a
significantly higher level of basal colonic activity in
cord-injured subjects (12.6 vs. 3.3 spikes per
10 min), and no demonstrable gastrocolic reflex.
This would support the assumption that the CNS
exerts a tonic inhibitory influence on basal colonic
activity and is consistent with the hypertonicity
seen on colometrograms.
Colonic neurotransmitters following spinal
cord injury
The intramural distribution of regulatory neuro-
peptides within the bowel wall is distinct. Sub-
stance P is exclusively localized in nerves (Ferri
et al., 1983). Large numbers of vasoactive intestinal
polypeptide- and substance P-containing enteric
nerves supply the ganglionated plexuses and are
especially numerous in the circular muscle layer.
They have a role in colon motility while those
supplying the mucosa are involved with electrolyte
and fluid transport. Substance P has been shown
337
to accelerate the transit of a charcoal meal in rats.
It increases intraluminal pressure mainly by circu-
lar muscle contraction by direct action on the
muscle as well as by simultaneous activation of
excitatory cholinergic pathways and of inhibitory
vasoactive intestinal polypeptide-independent, ni-
tric oxide-regulated pathways.
Substance P is reduced in the colonic mucosa of
patients with chronic constipation, and mucosal
substance P levels correlate significantly with dis-
ease state (Goldin et al., 1989). A similar scenario
exists with diabetic constipation where substance P
in the rectal mucosa of diabetics with constipation
is significantly lower than in diabetics with normal
bowel function (Lysy et al., 1993). The fact that
mucosal substance P levels are associated with two
disorders of colonic transit suggests a role in the
pathogenesis of intestinal transit disorders.
Whereas mucosal substance P may be decreased
with chronic constipation, concentrations in the
muscle layers may be increased. Sjolund et al.
(1997) examined tissue from the colon of 18 sub-
jects with slow-transit constipation. Tissue con-
centrations of vasoactive intestinal polypeptide
and substance P were measured by radioimmuno-
assay. Significantly increased concentrations of
both peptides were found in the ascending colon,
and in the descending colon, substance P was in-
creased in the myenteric plexus.
Recovery of bladder and bowel function follow-
ing traumatic spinal cord injury is dependent on
reorganisation of reflex pathways in the periphery
and CNS. Part of this reorganisation may be in-
fluenced by spinal cord–target organ interactions
mediated by neurotrophic factors released by the
peripheral organs. Interrupting the descending
modulation from the CNS may lead to changes
in the autonomic and somatic outflow reaching
target organs from the spinal cord caudal to the
lesion and alter target organ function. In rats, spi-
nal cord injury leads to hypertrophy of the bladder
as well as electrophysiological and morphological
changes in bladder afferent neurons (de Groat
et al., 1993; Yoshimura et al., 1993). In the colon,
however, the intrinsic enteric nervous system ap-
pears intact. Nerve fibres containing the intrinsic
neurotransmitters substance P and vasoactive
intestinal polypeptide appear to be present in
338
approximately similar amounts in specimens from
cord-injured and control subjects (Lynch et al.,
2000b). The colon may therefore continue to
function independently of CNS modulation after
spinal cord injury.
Anorectal function
Continence
The incidence of fecal incontinence in people with
spinal cord injury is more common than in the
general population. When compared to matched
controls by using standardized scoring systems,
the mean fecal incontinence score was higher for
cord-injured people than controls (po0.0001), and
for complete spinal cord injury compared with in-
complete injury ðp ¼ 0:0023Þ: Having fecal incon-
tinence also impacts on the quality of life of those
with a spinal cord injury more frequently than of
neurologically intact persons [62% of cord-injured
people report that fecal incontinence impacts upon
their everyday life, compared to 8% of control
subjects, po0.0001 (Lynch et al., 2000c)].
Fecal continence requires the ability to maintain
internal anal sphincter resting tone and to contract
the external anal sphincter in response to increased
intra-abdominal pressure, rectal distension and
rectal contraction. These are spinal reflexes that
are intact following spinal cord injury, but no
longer modulated by cortical input. Basal sphinc-
ter tone is mainly an activity of the internal anal
sphincter, the maintenance of which seems to be
due to a tonic excitatory sympathetic discharge.
Frenckner and Ihre (1976) observed that anorectal
manometry performed on cord injury subjects
shows a persistent anal tone that is reduced com-
pared to control subjects. They described changes
in anal tone in eight healthy subjects following
spinal anesthesia. High spinal anesthesia resulted
in a significantly lower resting anal pressure than
either low spinal or pudendal block. Of note, peo-
ple with lumbosacral injuries, who have external
anal sphincter paralysis but persistent internal anal
sphincter activity, still appear to maintain a degree
of anorectal tone, higher than rectal pressure, but
lower than normal. The external anal sphincter
also continues to show tonic activity, but again
generates a lower than normal pressure. Mano-
metric studies on cord-injured subjects show a
maximal mean basal sphincter pressure (which
probably reflects external anal sphincter pressure)
significantly lower than control group pressures.
Spinal cord-injured subjects can produce a small
rise in sphincter pressure with voluntary squeeze
(p40.05). This is, however, much less than the in-
crease in pressure generated by control subjects
performing a similar maneuver who can generate a
four-fold increase in external anal sphincter pres-
sure. People with incomplete injuries can produce
a greater increase in sphincter pressure with a
Valsalva maneuver than those with complete inju-
ries. The Valsalva maneuver is expiration against a
closed glottis. This may reflect the greater increase
in intra-abdominal pressure, as measured by rectal
pressure, that people with incomplete or low inju-
ries are able to generate due to incomplete paral-
ysis of their abdominal musculature. For those
with a complete injury, attempts at squeezing re-
sult in a straining response rather than a true
squeeze. People with lesions above T5 will be un-
able to use abdominal muscles and rely on inter-
costal and diaphragmatic muscle contraction to
increase intra-abdominal pressure. Those with cer-
vical injuries can only use the diaphragm. These
observations fit the concept that external anal
sphincter contraction is mediated by a spinal re-
flex, triggered by tension receptors in the pelvic
floor that respond to an increase in intra-abdom-
inal pressure. This is supported by another study
by MacDonagh et al. (1992) that found the rise in
sphincter pressure with the Valsalva maneuver to
be directly proportional to the rise in intra-
abdominal pressure.
Rectal sensation
All normal subjects experiencing rectal distension
as part of anorectal manometry studies report a
range of sensation starting at a rectal volume of
about 10 ml, and ranging from sensations of ‘wind’
to pain. This is compared to 78% of cord-injured
subjects with complete injuries, and 43% of those
with incomplete injuries, who report no sensation

on rectal distension (Lynch et al., 2000a). Those
that did report sensation described non-specific
abdominal sensation that did not prevent further
rectal distension. A previous study by MacDonagh
et al. (1992), examining similar sensations, pro-
posed that sympathetic nerves entering the tho-
racic spinal cord above the level of the injury
conveyed this dull pelvic sensation. However, such
sensations have also been identified in five people
with complete cervical injuries, making the origin
of this sensation unclear.
Rectal compliance
A normally compliant rectum accommodates an
increase in volume with little change in pressure.
As rectal volume increases, a normal sphincter re-
sponse is the relaxation of the internal anal sphinc-
ter with continence being maintained by continued
external anal sphincter contraction. The ability of
the rectum to distend to store bowel volume is an
important component of normal bowel function,
as it means defecation can be delayed until an ap-
propriate time.
People with complete cervical injuries can have
increased rectal tone with low compliance, i.e., a
sharp rise in rectal pressure occurs with rectal dis-
tension, as the rectum does not expand to accom-
modate the increase in volume. Sphincter tone can
also increase with rectal distension, because inter-
nal anal sphincter contraction is an enteric reflex,
normally suppressed by descending inhibitory
pathways. The loss of inhibitive sympathetic tone
has also been proposed as a mechanism for the
absent rectal relaxation and linear pressure/vol-
ume relationship during rectal distension (Mac-
Donagh et al., 1992). Most people with low
lumbosacral injuries have an areflexic rectum with
an attenuated sphincter response to rectal disten-
sion (Shafik, 1995). The rectum is flaccid and ca-
pacious producing no rise in rectal pressure with
increasing volume.
Urgency
Fecal urgency, or an inability to delay defecation,
is more often a problem following spinal cord
339
injury. It becomes even more significant as a qual-
ity of life issue when reduced mobility and poor
hand dexterity are compounded by the difficulties
associated with finding a wheelchair-able toilet.
Fecal urgency can be assessed by asking respond-
ents how long defecation can be delayed. Overall
81% of controls can delay defecation, compared
with only 41% of cord-injured people. There is
also an approximately ten-fold increase in the
proportion of cord-injured people, compared to
controls, who have to defecate immediately. Of
note, many people with complete injuries have no
sensation, and thus never sense the need to defe-
cate. The incidence of fecal incontinence is often
higher for cord-injured people who are unable to
delay defecation.
Defecation
Many defecatory problems seen after spinal cord
injury are the result of altered anorectal function.
Defecation requires the complex integration of re-
flex and voluntary muscular control. Cortical in-
hibition of the external anal sphincter occurs in a
coordinated fashion on straining as the rectal
smooth muscle contracts. This modulation of the
intrinsic nervous system by the extrinsic system is
disrupted following a complete supra-conal spinal
cord injury. This means that straining by increas-
ing intra-abdominal pressure does not improve
evacuation by the usual external anal sphincter
relaxation in response to the rectal and intra-ab-
dominal pressure increase. Rather, external anal
sphincter tone increases. Thus, coordinated reflex
defecation is difficult for cord-injured people. It is
often inefficient and incomplete, resulting in in-
continence and/or constipation. Therefore, defe-
cation is planned on regular basis to avoid
constipation or an increased chance of fecal in-
continence.
High spinal cord injuries
The difficulties with defecation following high
spinal cord injury result from discoordinate anal
sphincter function. The normal synergistic activity
of colonic smooth muscle and pelvic striated muscle
340
is lost. There is a loss of conscious sphincter con-
trol and, due to abdominal muscle paralysis, an
inability to significantly increase intra-abdominal
pressure. Loss of rectal sensation and a spastic
external anal sphincter require defecation to be
anticipated.
The conus-mediated increase in external anal
sphincter tone with increasing intra-abdominal
pressure acts against straining to defecate. How-
ever, reflex relaxation of the internal and external
anal sphincters by mucosal stimulation, either dig-
itally or with a suppository, can be exploited in
order to defecate. Insertion of a gloved finger into
the rectum with gentle sustained pressure towards
the sacrum relaxes the spastic external anal sphinc-
ter and pelvic muscles. Rapid or excessive stretch-
ing can precipitate sphincter spasm. Rotation of
the finger continues the stimulation until a reflex
peristaltic wave is generated in the rectum, flatus is
passed and stool comes down. The recto-anal in-
hibitory reflex is initiated, the internal anal sphinc-
ter relaxes, and the recto-colic reflex stimulates
pelvic nerve-mediated peristalsis. If there is no re-
flex relaxation of the external anal sphincter com-
plex, evacuation will not occur or be incomplete.
Manual evacuation or enemata are often required
in this situation.
This reflex relaxation can also lead to fecal in-
continence by two means. The anal sphincter may
relax at relatively low rectal volume in response to
a small increase in intra-abdominal pressure, or as
people with high-level injuries often have no sen-
sation of rectal fullness, reflex defecation in re-
sponse to a full rectum can result in fecal
incontinence that is unpredictable and episodic.
Low spinal cord injuries
Complete or partial injuries to the cauda equina
result in a lower motor neuron pattern of injury. A
person with a lower motor neuron lesion following
spinal cord injury will have absent external anal
sphincter tone, flaccid pelvic muscles and decreased
reflex peristalsis. Rectal compliance is increased in
response to rectal distension. The loss of parasym-
pathetic control of the internal anal sphincter means
that resting anal tone is low and unresponsive
to changes in intra-abdominal pressure. Thus, a
Valsalva maneuver can result in fecal leakage, and
the rectum has to be kept empty to avoid fecal
incontinence. Reflex-mediated defecation does not
occur with spinal cord lesions below the conus, so
stool has to be removed digitally, assisted by a
Valsalva maneuver and abdominal massage.
Change in bowel function with time from injury
Studies do not demonstrate a change in fecal in-
continence with either duration of injury or in-
creasing age for cord-injured people. One survey
of bowel dysfunction in cord-injured people by
Stone et al. (1990a) found that chronic gastroin-
testinal problems were rare in the first 5 years fol-
lowing injury, but problems with defecation
became more common with increasing time. The
incidence of abdominal pain and distension was
increased in long-term cord-injured people (more
than 18 years since injury) whose bowel regimen
was less frequent than once a day (Stone et al.,
1990a). Another study of chronic gastrointestinal
problems in cord-injured subjects by Han et al.
(1998) concluded that bowel habit appeared to
settle by about 6 months after injury, and that
subsequent bowel dysfunction was not related to
age, duration or level of injury.
Current management strategies
The approach to bowel management after spinal
cord injury should address specific issues such as
fecal incontinence, constipation and functional mo-
bility. This must be within the context of the patient
as a whole person and consider his/her cultural,
social, sexual and vocational roles. A bowel care
regimen needs to be generated that fits the person’s
long-term routine. The aim should be effective
colonic evacuation without fecal incontinence or
other complications. Regularity of evacuation pre-
vents excessive build-up of feces and impaction.
Appropriate equipment, such as commode chairs
and wheelchair-able toilets, needs to be supplied for
an adequate long-term bowel program.
Dietary manipulation is important. Adequate
water intake promotes transit by keeping the stool

soft. Fibre is promoted to give the stool bulk and
plasticity. This is thought to assist colonic transit
in neurologically intact patients, probably by pro-
moting propulsive activity secondary to increased
colonic wall distension. The effects of increased
fibre on colonic function after spinal cord injury
are not yet fully understood, and may be counter-
productive. A study carried out by Cameron et al.
(1996) at the Spinal Injury Unit, Austin Hospital,
Heidelberg, Australia demonstrated that signifi-
cantly increasing dietary fibre in a group of cord-
injured subjects with a range of injuries resulted in
an increase in mean colonic transit time from 28.2
to 42.2 h (po0.05), and in mean recto-sigmoid
transit time from 7.9 to 23.3 h (po0.02). However,
an increase in stool bulk may mean more time
spent with bowel care.
Stool softeners other than fibre, such as docu-
sate sodium increase the amount of water in stool
without increasing volume and have no effect on
bowel motility. They can also affect the intestinal
absorption of other drugs, resulting in higher plas-
ma levels. The stool is more likely to be liquid, so
continence will not be improved. They are most
useful when fecal incontinence is not a risk and
straining is to be avoided, such as for patients with
hemorrhoids or autonomic dysreflexia.
Stimulant laxatives act by increasing intestinal
motility, resulting in less time for water reabsorp-
tion. Senna has a direct stimulant effect on the
myenteric plexus and also increases intraluminal
fluid. Bisacodyl has a similar mode of action and is
often used as a suppository to initiate bowel evac-
uation. Dose-dependent side effects can occur.
These include abdominal cramping, diarrhea and
electrolyte imbalance. Chronic use of stimulant
laxatives, especially senna, can result in a progres-
sive unresponsiveness. Osmotic laxatives such as
lactulose draw fluid into the colon. They can result
in more liquid stool and cause cramping. Proki-
netic agents such as cisapride have been employed
to reduce constipation in cord-injured people.
Transit times are improved, but cardiac arrhyth-
mias have been noted with long-term use. Enemata
are often employed when suppositories or digital
stimulation fail. Long-term use can result in enema
dependences and side effects such as rectal trauma
and autonomic dysreflexia can occur.
341
Spinal cord-injured people with upper motor
neuron lesions can exploit the recto-colic reflex to
effect defecation. Digital stimulation can result in
a reflex wave of conus-mediated rectal peristalsis.
The recto-anal inhibitory reflex is intact, so this
causes internal anal sphincter relaxation and def-
ecation. Rectal sensation is reduced, however, so
defecation has to be anticipated on a regular basis.
These people require a bowel management pro-
gram that keeps the rectum empty to reduce the
incidence of incontinence. If people with upper
motor neuron lesions are unable to defecate using
the recto-colic reflex, then a management plan
needs to minimize anorectal trauma but still allow
adequate rectal evacuation to avoid constipation.
Cord-injured people with lower motor neuron
lesions have a rectum that is areflexic, reduced
sphincter tone and an attenuated sphincter re-
sponse to rectal distension or a Valsalva mane-
uver. These dysfunctions lead to an increased risk
of incontinence, especially with liquid stool. The
aim, therefore, is to keep stool consistency firm.
Local anorectal reflexes are often insufficient to
result in defecation, and a compliant rectum acts
as a large reservoir, so stool is digitally removed.
Arnold et al. (1986) noted that a Brindley sacral
anterior nerve root stimulator (S2–S4) can be used
for electromicturition to achieve regular, complete
bladder emptying. Often deafferentiation of the
sacral posterior nerve roots is performed before
the stimulator is implanted to produce detrusor
areflexia and interim urinary continence. The deaf-
ferentiation also results in loss of the sacral reflexes
necessary for defecation. The stimulator can then
be used to initiate defecation. This does not occur
during stimulation due to the simultaneous rectal
and sphincter contraction, but when stimulation
stops the external anal sphincter relaxes instanta-
neously and the rectum relaxes slowly, resulting in
spontaneous defecation. This method has been
shown to result in quicker, more controllable def-
ecation than the reflex method.
Does a colostomy improve bowel function?
Colostomy has been reported to result in improved
quality of life for people after spinal cord injury
342
(Frisbie et al., 1986; Saltzstein and Romano, 1990;
Stone et al., 1990b; Randell et al., 2001). It can not
only reduce fecal incontinence, but also simplify
bowel care in those for whom bowel evacuation is
difficult, reducing the amount of time spent on
bowel care from 99 to 18 min per day (Stone et al.,
1990b). The only long-term management problem
reported has been occasional appliance leakage
and mucoid discharge per rectum. All subjects in-
volved in a study by Craven and Etchells (1998)
found the stoma had impacted significantly on
their lifestyle with increased feelings of independ-
ence, freedom and raised self-esteem.
Sacral pressure ulcers are a common reason for
stoma formation in people who had no bowel
function problems previously. Deshmukh et al.
(1996) examined the use of colostomy as an adjunct
measure in the healing of pressure sores in cord-
injured people. Their findings similarly showed that
a colostomy improved bowel management and was
well accepted. However, they noted that the pri-
mary goal of pressure sore healing was accom-
plished in only 6 out of 27 cord-injured people, and
stoma formation resulted in significant morbidity
with stomal prolapse and difficulties with wound
healing. Of the 19 subjects reviewed in this study,
none wanted their stoma reversed, and all felt that
their quality of life was improved.
A colostomy can improve bowel management
for some people after cord injury. Dependence on
assistance with toileting is decreased as even tet-
raplegics with limited dexterity are able to manage
suitable stoma appliances. Enthusiasm for bowel
diversion procedures should be tempered with an
acknowledgment that the risks associated with any
surgical procedure may not make it acceptable to
all cord-injured people.
Future objectives for the investigation and
management of bowel dysfunction
Bowel dysfunction has a major impact on the
quality of life for many people with spinal cord
injury. This difficulty has been shown to improve
with appropriate early identification and manage-
ment of their problems. Interview and clinical ex-
amination can generate an impression of their
general bowel function and identify problems such
as constipation, fecal impaction, anal fissures and
hemorrhoids. Simple tests of anorectal function
are available that can be performed on all cord-
injured people in the same manner that bladder
dysfunction is investigated. Anorectal manometry
will identify those with dyssynergic sphincter func-
tion. For those with abdominal bloating and con-
stipation, abdominal X-ray and colonic motility
studies can be helpful.
Although improving colonic motility and appro-
priate bowel management may help, some cord-
injured people will have ongoing bowel problems.
Colostomy formation has been used to provide the
patient with relief from constipation and anorectal
dysfunction and with an independent means of
managing their own bowel function. Further re-
search needs to be done to examine the differences
in quality of life and bowel function following co-
lostomy formation in cord-injured people.
References
Arnold, E.P., Gowland, S.P., MacFarlane, M.R., Bean, A.R.
and Utley, W.L.F. (1986) Sacral anterior root stimulation of
the bladder in paraplegics. Aust. N.Z. J. Surg., 56: 319–324.
Cameron, K.J., Nyulasi, I.B., Collier, G.R. and Brown, D.J.
(1996) Assessment of the effect of increased dietary fibre in-
take on bowel function in patients with spinal cord injury.
Spinal Cord, 34(5): 277–283.
Craven, M.L. and Etchells, J. (1998) A review of the outcome of
stoma surgery on spinal cord injured patients. J. Adv. Nurs.,
27: 922–926.
de Groat, W.C. (1993) Anatomy and physiology of the lower
urinary tract. Urol. Clin. North Am., 20(3): 383–401.
DeLooze, D., Van Laere, M., De Muynck, M., Beke, R. and
Elewault, A. (1998) Constipation and other chronic gastro-
intestinal problems in spinal cord injury patients. Spinal
Cord, 36: 63–66.
Deshmukh, G.R., Barkel, D.C., Sevo, D. and Hergenroeder, P.
(1996) Use or misuse of colostomy to heal pressure ulcers.
Dis. Colon Rectum, 39: 737–738.
Ferri, G.-L., Adrian, T.E., Ghatei, M.A., O’Shaughnessy, D.J.,
Probert, L., Lee, Y.C., Buchan, A.M., Polak, J.M. and
Bloom, S.R. (1983) Tissue localisation and relative distribu-
tion of regulatory peptides in separated layers from the hu-
man bowel. Gastroenterology, 84: 777–786.
Frenckner, B. and Ihre, T. (1976) Influence of autonomic nerves
on the internal anal sphincter in man. Gut, 17: 306–312.
Frisbie, J.H., Tun, C.G. and Nguyen, C.H. (1986) Effect of
enterostomy on quality of life in spinal cord injury patients.
J. Am. Paraplegia Soc., 9(1–2): 3–5.

Gillis, R.A., Dias Souza, J., Hicks, K.A., Mangel, A.W.,
Pagani, F.D., Hamilton, B.L., Garvey III, T.Q., Pace,
D.G., Browne, R.K. and Norman, W.P. (1987) Inhibitory
control of proximal colonic motility by the sympathetic
nervous system. Am. J. Physiol., 253(4, Pt 1): G531–G539.
Glickman, S. and Kamm, M.A. (1996) Bowel dysfunction in
spinal-cord-injury patients. Lancet, 347: 1651–1653.
Goldin, E., Karmeli, F., Selinger, Z. and Rachmilewitz, D.
(1989) Colonic substance P levels are increased in ulcerative
colitis and decreased in chronic severe constipation. Dig. Dis.
Sci., 34(5): 754–757.
Han, T.R., Kim, J.H. and Kwon, B.S. (1998) Chronic gastro-
intestinal problems and bowel dysfunction in patients with
spinal cord injury. Spinal Cord, 36: 485–490.
Hanson, R.W. and Franklin, M.R. (1976) Sexual loss in rela-
tion to other functional losses for spinal cord injured males.
Arch. Phys. Med. Rehabil., 57: 291–293.
Levi, R., Hultling, C., Nash, M.S. and Seiger, Å. (1995) The
Stockholm spinal cord injury study: 1. medical problems in a
regional SCI population. Paraplegia, 33: 308–315.
Lynch, A., Anthony, A., Dobbs, B. and Frizelle, F. (2000a)
Anorectal physiology following spinal cord injury. Spinal
Cord, 38: 73–80.
Lynch, A., Anthony, A., Dobbs, B. and Frizelle, F. (2000b)
Colonic neurotransmitters following spinal cord injury. Tech.
Coloproct., 4: 93–97.
Lynch, A., Anthony, A., Dobbs, B. and Frizelle, F. (2001)
Bowel dysfunction following spinal cord injury; a review.
Spinal Cord, 39: 193–203.
Lynch, A., Wong, C., Anthony, A., Dobbs, B. and Frizelle, F.
(2000c) Bowel dysfunction following spinal cord injury:
a description of bowel function in a spinal cord-injured pop-
ulation and comparison with age and gender matched con-
trols. Spinal Cord, 38: 717–723.
Lysy, J., Karmeli, F. and Goldin, E. (1993) Substance P levels in
the rectal mucosa of diabetic patients with normal bowel func-
tion and constipation. Scand. J. Gastroenterology, 28: 49–52.
343
MacDonagh, R., Sun, W.M., Thomas, D.G., Smallwood, R.
and Read, N.W. (1992) Anorectal function in patients
with complete supraconal spinal cord lesions. Gut, 33:
1532–1538.
Menardo, G., Bausano, G., Corazziari, E., Fazio, A., Marangi,
A., Genta, V. and Marenco, G.L. (1987) Large bowel transit
in paraplegic patients. Dis. Colon Rectum, 30(12): 924–928.
Meshkinpour, H., Nowroozi, F. and Glick, M.E. (1983) Co-
lonic compliance in patients with spinal cord injury. Arch.
Phys. Med. Rehabil., 64: 111–112.
Nino-Murcia, M., Stone, J.M., Chang, P.J. and Perkash, I.
(1990) Colonic transit in spinal cord-injured patients. Invest.
Radiol., 25(2): 109–112.
Randell, N., Lynch, A., Anthony, A., Dobbs, B., Roake, J. and
Frizelle, F. (2001) Does a colostomy alter quality of life in
patients with spinal cord injury? A controlled study. Spinal
Cord, 39(5): 279–282.
Saltzstein, R.J. and Romano, J. (1990) The efficacy of colos-
tomy as a bowel management alternative is selected spinal
cord injury patients. J. Am. Paraplegia Soc., 13(2): 9–13.
Shafik, A. (1995) Electrorectogram study of the neuropathic
rectum. Paraplegia, 33: 346–349.
Sjolund, K., Fasth, S., Ekman, R., Hulten, L., Jiborn, H.,
Nordgren, S. and Sundler, F. (1997) Neuropeptides in
idiopathic chronic constipation (slow transit constipation).
Neurogastroenterol. Motil., 9(3): 143–150.
Stone, J.M., Nino-Murcia, M., Wolfe, V.A. and Perkash, I.
(1990a) Chronic gastrointestinal problems in spinal cord in-
jury patients: a prospective analysis. Am. J. Gastroenterol.,
85(9): 1114–1119.
Stone, J.M., Wolfe, V.A., Nino-Murcia, M. and Perkash, I.
(1990b) Colostomy as treatment for complications of spinal
cord injury. Arch. Phys. Med. Rehabil., 71: 514–518.
Yoshimura, N. and de Groat, W.C. (1993) Changes in elect-
rophysiological and pharmacological properties of rat blad-
der afferent neurons following spinal cord injury. J. Urol.,
149: 340A.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 23

Mechanisms controlling normal defecation and the

potential effects of spinal cord injury

A.F. Brading1, and T. Ramalingam2

1
Oxford Continence Group, University Department of Pharmacology, Mansfield Road, Oxford OX1 3QT, UK
2
Specialist Registrar in Colorectal Surgery, Headquarters Army Medical Directorate, Former Army Staff College,
Camberley, Surrey GU16 4NP, UK

Abstract: Spinal cord injury frequently leads to bowel dysfunction with the result that emptying the bowel
can occupy a significant part of the day and reduce the quality of life. This chapter contains an overview of
the function and morphology of the normal distal gut in the human, and of gut behaviour in normal
defecation. In humans, this can be monitored and is described, but knowledge of the mechanisms con-
trolling it is limited. Work on animals has shown that the intrinsic activity of the smooth muscles and their
interactions with the enteric nervous system can program the activity that is necessary to expel waste
material, but the external anal sphincter is controlled through somatic nerves. The gut however also
receives input from the central nervous system through autonomic nerves, and a spinal reflex centre exists.
Voluntary effort to induce defecation can influence all the control mechanisms, but the precise importance
of each is not understood.
The behaviour and properties of the individual muscles in the normal human rectum and anal canal are
described, including their responses to intrinsic nerve stimulation and adrenergic and cholinergic agonists.
The effects of established spinal cord injury are then considered. For convenience, supraconal and conal/
cauda equina lesions are considered as two categories. Prolongation of transit times and disordered def-
ecation are common problems. Supraconal lesions result in reduced resting anal pressures and increased
risk of fecal incontinence. The acute effects of spinal cord injury are described, with injury causing ileus
(prolonged total gastrointestinal transit times), constipation (prolonged colonic transit times) and fecal
incontinence (passive leakage).

Introduction cause of morbidity and remain a problem long

Problems related to the control of defecation are a
source of physical, psychological and social dis-
tress to people with spinal cord injuries (SCI).
Approximately 30% of these people consider bow-
el dysfunction to be a greater concern than either
bladder or sexual dysfunction (Krogh et al., 1997).
Chronic gastrointestinal problems are a significant
Corresponding author. Tel.: +44 1865 271875;
Fax: +44 1865 281120;
E-mail: alison.brading@pharm.ox.ac.uk
after the time of injury. Constipation and dis-
turbed defecation means that this routine bodily
function can occupy a significant part of the day in
those with spinal cord injuries. Successful bowel
management often requires considerable control of
dietary intake, strict adherence to a regime, man-
ual stimulation and the use of a concoction of
drugs to aid the process. Those with higher lesions
and poor arm and hand control may have to rely
on others for assistance with the accompanying
loss of dignity, making bowel management a ma-
jor life-limiting problem. Some also experience

DOI: 10.1016/S0079-6123(05)52023-5 345

346
symptoms of fecal incontinence, with its inevitable
consequences on their quality of life (Glickman
and Kamm, 1996).
Our current understanding of the control of
normal defecation is still somewhat sketchy, and
thus the precise mechanisms through which spinal
cord injury affects the process are also not entirely
clear. In this chapter, the mechanisms controlling
normal defecation will be discussed. We will in-
clude colonic morphology in relation to motility
and transit, and the behaviour and role of the
anorectum during evacuation, since effective def-
ecation requires both these components to func-
tion properly. We will go on to describe the
current appreciation of the potential effects of SCI
in each of these component areas.
Normal gut
Outline of colorectal function and morphology
The large intestine extends from the ileocaecal
valve of the terminal ileum to the anus. It com-
prises of the caecum, ascending colon, transverse
colon, descending colon, sigmoid colon, rectum
and the anus (Bannister, 1995). Its main functions
are the absorption of water and salts, the degra-
dation of short-chain fatty acids and the transport
and storage of luminal contents to the anorectum
until it is deemed convenient for defecation.
The colon contains two layers of smooth mus-
cle, an inner, circular layer and an outer longitu-
dinal layer. The anatomy of the longitudinal layer
is not consistent throughout the large intestine.
From the caecum to the sigmoid colon, it is con-
centrated in three bands that are symmetrically
placed: the taenia coli.
Bundles of smooth muscle cells within these lay-
ers are connected by gap junctions at intermittent
points, enabling them to function as a syncitium.
The anorectum forms the distal end of the
colon. At the rectum, the taenia coli coalesce to
form the longitudinal smooth muscle layer. The
anal sphincter consists of an inner ring of smooth
muscle, the internal anal sphincter, surrounded by a
thin outer longitudinal smooth muscle, the conjoint
longitudinal coat, and a thicker outer ring of
striated muscle, the external anal sphincter. The
internal anal sphincter is considered to be the con-
tinuation of the circular muscle while the longitu-
dinal muscle layer of the rectum splits to envelope
the internal sphincter ultimately inserting into the
subcutaneous tissue (see Fig. 1).
Overview of defecation
In the gut, waste material, usually either solid or
semi-solid feces or flatus (gas), can be stored in the
rectum and distal colon. The anal sphincter pro-
vides continence and is equivalent to the external
urethral sphincter. Functional differences between
the distal gut and the urinary tract are due to the
fact that fecal material can be returned to the co-
lon from the rectum, and that a detection mech-
anism is present in the anal canal which allows
assessment of the rectal content, so that flatus can
be voided at times when voiding feces would not
be appropriate.
The sequence of events that occurs in storage
and defecation is as follows. The rectum remains
empty for most of the time, but will fill as fecal
matter accumulates in the descending and sigmoid
colon, and is pushed forward through the occa-
sional peristaltic waves (mass movements) that
occur. As fecal matter enters the rectum, the walls
relax, and filling can occur with little increase in
rectal pressure. Distension of the rectum triggers a
recto-anal inhibitory reflex that lowers the pressure
in the anal canal, allowing the rectal contents to
enter and contact the mucosa. Periodic relaxations
Fig. 1. Diagram of the human rectum and anal canal.
 347

of the internal anal sphincter allow anorectal sam-

pling of the contents (Miller et al., 1988a, b) using
the rich sensory innervation of the anal canal
which allows us to discriminate between gas, liquid
and solid. During these phases, the external stri-
ated sphincter is contracted to maintain conti-
nence. As filling continues, sensory information
ascending to the brain leads to the sensation of
rectal fullness. If defecation is deemed appropriate,
voluntary relaxation of the external sphincter oc-
curs and peristalsis in the colon and rectum is in-
itiated, usually by abdominal straining resulting in
relaxation of the internal sphincter and expulsion
of the rectal contents. If defecation is not appro-
priate, the rectal contents may return to the colon.

Overview of neural control

Unlike the situation in the urinary tract, however,

the gut wall contains all of the machinery (intrinsic
pacemakers and neural networks) to programme
the activity of the smooth muscle that is necessary
to expel the waste material (relaxation of the in-
ternal anal sphincter, initiation and co-ordination
of peristalsis). The intrinsic or enteric nervous sys-
tem consists of two major plexuses of intercon-
necting ganglia, namely the submucous
Fig. 2. Innervation of the distal gut. CP, celiac plexus; HGN,
(Meissner’s) plexus and the myenteric (Auer-
hypogastric nerve; PEL, pelvic nerve; PP, pelvic plexus; PUD,
bach’s) plexus (Furness and Costa, 1987) whose pudendal nerve; S2–4, sacral roots; SC, sympathetic chain; SHP,
complex circuitry is the focus of considerable ex- superior hypogastric plexus; SN, sacral nerve.
perimental interest, and which can choreograph
most of the activity of the gut. It is only the ex- filling, and to help switch on the peristaltic activity
ternal striated sphincter that absolutely requires necessary to produce defecation, as well as to in-
extrinsic innervation to contract, through activa- itiate the correct pattern of activity in the somatic
tion of the somatic motor neurons whose axons nerves to the extrinsic sphincter. The voluntary
run in the pudendal nerves. The distal gut does, input to defecation is through control of the stri-
however, receive extrinsic innervation through the ated muscles of the anal sphincter and the abdom-
autonomic nervous system, with parasympathetic inal muscles.
input to the proximal colon through the vagus,
and to the whole colon and anorectum via the
sacral roots (S2–S4) through the pelvic nerves, and Underlying smooth muscle properties
with sympathetic input from the lumbar cord run-
ning with the superior and inferior mesenteric Basic rhythmic activity
blood vessels and passing through the mesenteric Smooth muscles of the non-sphincteric elements of
and pelvic plexuses (Figs. 2 and 3). A spinal reflex the gut show basic rhythmic changes in electrical
centre exists which uses this autonomic extrinsic activity, called slow waves. These slow waves,
pathway to maintain rectal compliance during if large enough, trigger action potentials through
348
Fig. 3. Diagram of the autonomic innervation of the gut wall. CMS, circular smooth muscle; LSM, longitudinal smooth muscle; Epi,
epithelium. Green: preganglionic parasympathetic nerves. Blue: pre- and post-ganglionic sympathetic nerves, brown: sensory nerves,
orange and yellow: interneurones. Note that this diagram does not include interstitial cells of Cajal.
activation of L-type Ca channels, and resulting
smooth muscle contraction (Tomita, 1981).
Rhythmic contractions of the longitudinal mus-
cles may generate pendular activity in the gut wall,
and similar contractions in the circular muscle
generate segmentation. These types of activity may
occur in the absence of neural input. The rhythms
and sizes of the underlying slow waves can be
modulated by neuronal input and by local gut
hormones.
Pacemaking
This spontaneous, or more accurately, non-
neurogenic contractile activity in gastrointestinal
smooth muscles is now known not to be generated
by the smooth muscles themselves, but by
another cell type, the interstitial cells of Cajal.
These cells are named after Santiago Ramon
y Cajal, a Spanish histologist who described them
at the end of the 19th century (Cajal, 1893). Recent
studies of these cells have been facilitated by the
discovery that they express on their surface
membrane a receptor tyrosine kinase that is the
gene product of c-kit, a proto-oncogene (for
reviews, see Sanders et al., 1999, 2002; Ward and
Sanders, 2001). These cells can now be identified
using immunohistochemical labelling with anti-
bodies to the Kit receptor. Interstitial cells are ar-
ranged in distinct ways in different parts of the
gut. In the longitudinal elements there is an ex-
tensive plexus running with the neurons in the
myenteric plexus, and often another plexus at the
submucosal surface of the circular muscle. In
these, the interstitial cells are linked to each other,
and to the adjacent smooth muscle cells through
gap-junctions. Interstitial cells also run along the
outside of smooth muscle bundles within the mus-
cle layers. The interstitial cells associated with the
myenteric plexus and submucosal plexus are
thought to be pacemakers in the human colon
and rectum. The interstitial cells along the muscle
bundles may be involved in helping conduct the
activity within the muscle layers. Isolated intersti-
tial cells undergo rhythmic large depolarizations at
frequencies similar to the phasic contractions seen
in intact smooth muscles, and when held under
voltage clamp conditions, will generate depolariz-
ing currents with the same rhythm. The underlying
mechanism has not been completely resolved, and
may vary in different parts of the gut. Integrated
Ca2+ handling by the endoplasmic reticulum and
the mitochondria seem to be involved (Ward et al.,
2000), and it is thought that changes in [Ca2+]i

determined by phasic release of Ca2+ from endo-
plasmic reticulum through an IP3-dependent
mechanism and also uptake of Ca2+ by mi-
tochondria, lead to activation of Ca2+-dependent
conductances (possibly Cl channels or non-selec-
tive cation channels) that can generate depolariz-
ing (inward) currents. When linked together in
networks and activated synchronously, the inter-
stitial cells can inject sufficient current into adja-
cent smooth muscle cells to produce the
characteristic slow waves of depolarization seen
in gastrointestinal smooth muscles.
Underlying mechanisms in defecation
Experimental approaches
Our understanding of the basic mechanisms comes
from work on animals. Initially much work was
carried out on the overall organisation (central
and peripheral) controlling defecation using anaes-
thetised cats or dogs (Martner, 1975; de Groat and
Krier, 1978; Mackel, 1979; de Groat et al., 1981,
1982; Fukuda et al., 1981; Fukuda and Fukai,
1986; Takaki et al., 1987), but the current climate
with respect to animal research has limited the use
of these species. Studies of the properties of iso-
lated segments of the whole gut have been useful
for studying reflexes intrinsic to the enteric nerv-
ous system, but to keep these preparations alive
and functioning in vitro, it is necessary to use in-
testine from small animals to enable proper oxy-
genation of the tissue and survival of the intrinsic
nerves; thus more recent experimental work has
concentrated on rodents. Many of the detailed in-
vestigations into the properties of the enteric nerv-
ous system have been carried out on guinea-pig gut
(e.g., see Furness and Costa, 1987), including el-
egant demonstrations of the morphological,
immunohistochemical and functional properties of
the individual neurons in the plexuses. The motor
control of the smooth muscles of the lower gut can
be examined in the organ bath by dissecting strips
of smooth muscle and recording their spontaneous
contractile behaviour and their evoked responses to
electrical field stimulation of the intrinsic nerves.
Many of the experimental techniques used on
animals cannot be applied to normal humans, and
349
our knowledge of defecation in humans comes
from less direct methodologies such as imaging the
gut (e.g., X-rays and ultrasound) behavioural
studies, recordings of pressure in the gastrointes-
tinal tract, and of electrical activity using needle or
surface electrodes. Information about the control
of the individual muscles can be obtained using
human tissue resected during surgical procedures.
In the next section, examination of the gross and
more detailed properties of the various compo-
nents will be made. Much of the work on the hu-
man anorectum and its intrinsic innervation
described in this chapter is from research using
tissue retrieved from patients at surgery by clini-
cians working for higher degrees in the Oxford
Continence Group, using a superfusion organ bath
set-up (Brading and Sibley, 1982).
Colorectal motility and transport
In health, the proximal colon (ascending and
transverse segments) acts as a reservoir while the
descending colon serves as a conduit (Proano
et al., 1990). The movement of colonic contents is
a discontinuous process, such that residue may be
retained for prolonged periods in the ascending
colon and mass movements can deliver its contents
to the sigmoid colon in seconds. Colonic contrac-
tions can either be phasic or tonic. The former are
the result of spike potentials and have definite be-
ginnings and endings, causing elevations in intra-
luminal pressure (Garcia et al., 1991; Gregersen
and Ehrlein, 1996). Ambulatory studies have
shown there to be several patterns of phasic co-
lonic contractions: single non-propagating con-
tractions, antegrade pressure waves, retrograde
pressure waves and periodic colonic motor activ-
ity. Of note, high-amplitude propagating contrac-
tions or giant migrating/mass contractions, the
result of antegrade pressure waves, occur as iso-
lated events or in bursts (Sarna, 1991). They may
or may not result in propagation of colonic con-
tents. Mass contractions are mostly generated in
the daytime, especially after awakening or after
meals (known as the gastro-colonic response).
In relation to the gastro-colonic response, some
investigators report the response to be most
intense in the sigmoid and descending colon
350
segments (Quigley, 2002). Localized phasic
contractions function to mix and move colonic
contents over short distances (Cook et al., 2000;
Rao et al., 2001). Tonic colonic contractions are
less well defined than phasic contractions. They
are long lasting and may or may not be associated
with increased intraluminal pressure (Gregersen
and Ehrlein, 1996).
Three types of phasic rectal contractions have
been described: isolated ones, short clusters of low
amplitude contractions (frequency 5–6/min) and
powerful contractions termed rectal phasic motor
activity, every 60–120 min (Akervall et al., 1989).
The phasic activity is most common during the
night and is usually associated with simultaneous
colonic contractions and contractions of the anal
sphincter, thus preventing defecation.
Rectal tone appears to be influenced by the rate
of volume change. Rapid volume waves lasting less
than 2 min have been frequently associated with
increase in luminal pressure, while slow volume
waves of greater than 2 min duration have not.
Although the significance of these volume waves is
uncertain, changes in the rectal tone may act to
alter its function between a capacious reservoir
and a conduit (Akervall et al., 1989).
The transit time for colonic contents is the cul-
mination of anatomical and motility effects and is
relatively easily quantifiable (Abrahamsson et al.,
1988). Normally, the average mouth to caecum
transit time for ingested food is about 6 h and re-
gional transit times, measured using radio-opaque
marker method, are about 2 h each in the ascend-
ing, descending and sigmoid colon segments.
Although the mean colonic transit time is 36 h,
the range is wide since many factors can affect this
such as diet, drugs, etc. (Metcalf et al., 1987).
Properties of the rectal smooth muscles
The rectum is surrounded by a complete layer of
outer longitudinal and inner circular smooth mus-
cle. The properties of the two layers are probably
rather similar to those in other longitudinal ele-
ments of the gut. In humans, both the circular and
longitudinal smooth muscle layers generate phasic
contractions at a frequency of about 3–4/min
(Stebbing, 1998). In the circular smooth muscle,
there is little if any intrinsic tone, and these
contractions thus arise from a zero baseline. In
contrast, the longitudinal smooth muscle may de-
velop a basal tension of some 0.2 g/mg tissue, and
the phasic contractions rise from this. These spon-
taneous contractions are unaffected by tetrodo-
toxin, and are probably evoked by slow waves
generated in the interstitial cells of Cajal (see
above). Both muscle layers respond to activation
of their muscarinic receptors by contracting.
Catecholamines abolish the spontaneous contrac-
tions in both muscle layers and reduce the baseline
tension in the longitudinal smooth muscle (Fig. 4).
Activation of either a- or b-adrenoceptors can
induce relaxation, exemplified by the fact that
phenylephrine and isoproterenol produce the
same relaxant responses as norepinephrine. a-
adrenoceptors are known to induce relaxation of
the longitudinal elements of the gut in many mam-
malian species through opening of Ca-activated K
channels (Kuriyama et al., 1998).
The enteric nervous system is well developed
in the rectum (O’Kelly et al., 1994), as shown in
Fig. 5. Extrapolation from detailed studies in other
species (e.g., Gabella, 2001) suggests that the den-
sity of innervation directly to the smooth muscle
cells is fairly sparse, and that the interstitial cells of
Cajal may be the main targets of the intrinsic
nerves (Sanders, 1996). Transmural stimulation of
Fig. 4. Contractile responses of small strips of rectal smooth
muscle to agonist application.

Fig. 5. Preparations of the human myenteric plexus stained
with NADPH diaphorase to indicate nerves containing nitric
oxide synthase. (A) Myenteric plexus from the main rectum. (B)
Enlarged view of ganglion cells in rectal myenteric plexus
showing pronounced staining. (C) Myenteric plexus from the
lower rectum. (D) Myenteric plexus from the anal canal. Scale
bars in (A), (B) and (D): 1 mm; in (C): 100 mm. From O’Kelly
et al. (1994), with permission.
strips of rectal smooth muscle produce responses
that are normally quite small and varied — where
there is initial tone, the normal response is a tran-
sient small relaxation followed by a small contrac-
tion. If tone is initiated by application of an
agonist such as histamine, field stimulation initiates
a robust relaxant response that is attenuated by
nitric oxide synthase inhibition (Stebbing, 1998).
Properties of the internal anal sphincter
The internal anal sphincter is in a continuous state
of contraction as a result of its intrinsic myogenic
properties (O’Kelly et al., 1993a, b) and extrinsic
innervation by the autonomic nerves (Gutierrez
and Shah, 1975; Frenckner and Ihre, 1976). It
contributes up to 85% of the overall resting anal
pressure. Using in vivo experiments, investigators
have shown that reduction in sphincter tone can be
achieved following sympathetic blockade by either
high spinal (T6–T12) anaesthesia (Frenckner and
Ihre, 1976) or by infusion of phentolamine, an
351
a-adrenoceptor antagonist, suggesting a tonic ex-
citatory sympathetic discharge. There does not
appear to be a tonic parasympathetic input to the
internal sphincter in vivo.
In addition, superimposed on the tonic state of
the internal sphincter are two intermittent wave-
form activities, namely, slow and ultraslow, ac-
cording to their respective frequencies (Sun et al.,
1990). The relevance of these is not yet clear but it
has been suggested that they may act either to op-
pose rectal pressure waves (Sorensen et al., 1989)
or to keep the anal canal closed to prevent desen-
sitization of the anoderm (Zbar et al., 2000).
The intrinsic innervation of the anal canal
has been examined in sections, (Holmes, 1961;
Aldridge and Campbell, 1968; Weinberg, 1970;
Baumgarten et al., 1971) and also in whole mounts
where it can be shown that the myenteric plexus
penetrates as far as the middle third of the anal
canal, although the number of ganglia seems to be
less than in the rectum. Nitric oxide-producing
nerves divide and ramify into the internal anal
sphincter, as shown in Fig. 5 (O’Kelly et al., 1994).
Functionally, the internal anal sphincter has
a robust inhibitory innervation. Burleigh and
colleagues showed that electric field stimulation
caused a relaxation, which is blocked by tetrodo-
toxin but not by atropine or guanethidine (Burleigh
et al., 1979). This suggested that the relaxation
is nerve mediated, but not by cholinergic or ad-
renergic pathways, paving the way for a possible
non-adrenergic non-cholinergic mediator. The prop-
erties of the human sphincter including the inhibitory
innervation were reviewed by Penninckx et al. (1992).
O’Kelly and colleagues extended this work on
the human sphincter, and showed that strips of the
sphincter relaxed to transmural stimulation of
their intrinsic nerves, and also relaxed in response
to nitric oxide donors. Nerve-mediated relaxation
was inhibited in a concentration-dependent man-
ner by the nitric oxide synthase blocker, L-N-9-
nitroarginine, which competes with the normal
substrate for nitric oxide synthesis, L-arginine. The
antagonistic effect was countered by the addition
of excess L-arginine (O’Kelly et al., 1993a),
providing good evidence for a nitrergic inhibitory
innervation (see Fig. 6). O’Kelly also showed that
the sphincter relaxes in response to muscarinic
352
Fig. 6. Responses of a strip of human internal anal sphincter to
transmural stimulation (10 V, 0.5 ms, 8 Hz, 1 s train) at the dots.
The responses were abolished by application of the nitric oxide
synthase blocker L-NOARG (10 5 M), were not affected by
additional application of the inactive D isomer of arginine, but
restored by application of the natural substrate L-arginine
which competes with L-NOARG for the enzyme.
Fig. 7. Responses of strips of human internal anal sphincter to
10 s applications of 5  10 5 M carbachol (arrows) and trans-
mural field stimulation (10 V, 0.5 ms, 8 Hz, 1 sec train) at the
dots. The relaxant responses to both stimuli were abolished by
L-NOARG, and restored by additional application of L-argi-
nine.
receptor stimulation, contracts in response to
activation of a-adrenoceptors, and relaxes in re-
sponse to stimulation of b-adrenoceptors (O’Kelly
et al., 1993b). He postulated that muscarinic
receptors were present on nitrergic neurons since
inhibitors of nitric oxide synthase attenuated the
relaxant response of the sphincter to carbachol, an
acetylcholine analogue (Fig. 7). In contrast to the
internal anal sphincter, strips from the conjoined
longitudinal coat only generate a small amount of
basic tone but show little spontaneous phasic
activity (O’Kelly et al., 1993b). They respond to
activation of both a-adrenoceptors and muscarinic
receptors by contracting (Fig. 8).
Table 1 compares the basic properties of the
four smooth muscles of the anal canal and rectum,
emphasising their heterogeneity. The responsive-
ness of these smooth muscles to stimulation of
their adrenergic receptors reflects their overall
response to circulating epinephrine and activation
of sympathetic nerves, which will ensure that in
‘flight and fight’ conditions, activity will be re-
duced in the longitudinal parts of the gut, while the
smooth muscle sphincters will be closed, ensuring
continence whilst reducing overall energy expend-
iture. However, under extreme conditions of fear,
if the circulating levels of epinephrine become too
high, the relaxant responses to b-adrenoceptor
stimulation on the sphincteric smooth muscle may
underlie the anecdotal accounts of involuntary loss
of feces and urine.
Properties of the external anal sphincter
Since the inherent automaticity of the internal anal
sphincter confers the sphincter complex with a
resting tone, which in turn helps maintain our
subconscious control of continence, there has been
a greater emphasis on research looking into the
physiology and pharmacology of this compared to
the external sphincter. The striated muscle of the
external sphincter has a somatic nerve supply with
acetylcholine as the excitatory transmitter at the
neuromuscular junction working via nicotinic re-
ceptors. The cell bodies of the motor neurons lie in
the ventral horn of the second and third sacral
segments of the spinal cord (Schroder, 1981).
The external sphincter can contribute up to 50%
of the anal pressure during periods of rectal dis-
tension (Frenckner and Euler, 1975). When
required, the voluntary component can be called
upon to momentarily raise the anal pressure in
order to defer defecation.
Main areas of ignorance
The above account perhaps illustrates that we
can describe overall events, and have increasing
knowledge of the properties of the individual
 353

Fig. 8. A comparison of the effects of carbachol and norepinephrine (5  10 5 M, 10 sec) on the strips of smooth muscle dissected
from the conjoined longitudinal coat and the internal anal sphincter. Hexamethoium (10 6 M) was present to prevent any stimulation
of nicotinic acetylcholine receptors.

Table 1. Properties of smooth muscle from human ano-rectum

Tone Response to muscarinic Response to

agonist noradrenaline

Rectal circular m k
Rectal longitudinal + m k
Internal anal sphincter +++ k m
Conjoined longitudinal + m m

m contraction; k relaxation

muscles involved in defecation. However, we are
extremely ignorant about exactly how the process
of defecation is controlled. Although evidence
strongly suggests that there is a spinal defecation
centre, the relative roles of the enteric nervous
system, the spinal centre and voluntary control is
hard to determine. Voluntary control via increas-
ing intra-abdominal pressure can trigger increased
activity in the enteric nervous system through ac-
tivation of pressure sensitive nerves, and these may
enhance activity at any of the several sites — lo-
cally in the enteric nervous system, through influ-
encing autonomic ganglia en route to the CNS, at
the level of the spinal reflex centre and at higher
centres. Sensory nerve activity may enter the CNS
along with the parasympathetic, sympathetic, or
somatic nerves, and the CNS can alter activity in
the enteric nervous system through the parasym-
pathetic and sympathetic nerves. The relative
importance of these different systems will be of
considerable significance in patients with spinal
injury, since injuries at different levels of the cord,
and the likelihood that there may also be addi-
tional damage to autonomic nerves means that the
spectra of disruption to defecation must be ex-
tremely varied. Exact knowledge of the extent of
the damage would be required to predict the dis-
ruption to defecation that is likely to occur for a
particular patient, as well as the relative impor-
tance of the potential levels of control. More re-
search, in particular on whole animal models, may
help throw light onto the underlying processes.
Spinal cord injury
Introduction
Traditionally, lesions of motor neurons are classi-
fied into supranuclear (supraconal or upper motor
neuron) or infranuclear (conal/cauda equina or
354
lower motor neuron) lesions. In striated muscles
and the smooth muscle of the urinary bladder,
upper motor neuron lesions result in hyperactive
or spastic paresis while lower motor neuron lesions
lead to hyporeactive or flaccid paresis.
However, the effects of spinal cord injury on the
colon, rectum and sphincter complex are less well
known. This is in part due to our relatively poor
understanding of the complex regulation of bowel
function and, in particular, defecation. This is
compounded by the variable clinical effects ob-
served in spinal cord injury with 60% of those af-
flicted having incomplete lesions and 59% of these
having significant recovery of function (Buckle
et al., 1999).
We will therefore describe the potential effects
of spinal cord injury in relation to the clinical
manifestation of disturbed colorectal transport/
motility, and impaired defecation. As a matter of
convenience, we have two categories: people with
supraconal or conal/cauda equina lesions. We will
also briefly describe colorectal dysfunction after
acute injury.
Effects on colorectal motility and transport
In a study of the effects of spinal cord injuries on
the total gastrointestinal transit time and colonic
transit times, Krogh and colleagues showed that
these parameters were significantly increased
chronically (Krogh et al., 2000). The regions in
the colon and rectum responsible for the delays are
unclear, however. Devroede and colleagues found
prolonged transit times of the entire colon and
rectum in four subjects with incomplete lumbosa-
cral lesions (Devroede et al., 1979). In two other
studies, colorectal transit times were also shown to
be prolonged after cervical and thoracic lesions.
Nino-Murcia, who looked at transit times after
cervical or thoracic lesions, found the total and left
(descending) colon transit times to be prolonged
but not the right (ascending) colon transit times
(Nino-Murcia et al., 1990). In contrast, Beuret-
Blanquat and colleagues, in a study of 19 people
with complete thoracic, lumbar or sacral lesions,
found neither the mean right (ascending) colon nor
left (descending) colon transit times to be pro-
longed (Beuret-Blanquart et al., 1990).
In their relatively large study that examined the
changes in segmental colonic transit times among
32 people with spinal cord injuries, Krogh and
colleagues observed that chronic supraconal injury
significantly prolonged transit times at the trans-
verse and descending colon segments but not at the
rectosigmoid (Krogh et al., 1997). Chronic conal
or cauda equina lesions were shown to prolong
transit times significantly, not only at the trans-
verse and descending colonic segments, but also at
the rectosigmoid region. They also showed that
there was no statistically significant difference in
the total gastrointestinal transit times or segmental
colonic transit times between people with lesions
affecting the sympathetic outflow and those with
lesions beyond this level, reinforcing the percep-
tion that the parasympathetic input is more im-
portant in colorectal transport.
A subsequent study by the same group of in-
vestigators, albeit using a new scintigraphic tech-
nique for quantitative assessment of segmental
colorectal transport, showed impaired emptying of
the rectosigmoid segment in people with sacral le-
sions when compared with a healthy volunteer
group (Krogh et al., 2003). The gastro-colonic
response in people with supraconal lesions is also
generally reduced or absent (Bruninga and
Camilleri, 1997).
Disordered defecation
In a questionnaire survey of 424 members of the
Danish Paraplegia Association, constipation and
disturbed defecation were the most common colo-
rectal concern affecting 81% of those with spinal
cord injuries (Krogh et al., 1997). Symptoms as-
sociated with autonomic dysreflexia, a response
usually generated by distended bladder or bowel,
were noted in 25% before and during defecation.
Studies have suggested rectal tone to be in-
creased after supraconal lesions (Sun et al., 1995)
and decreased after conal or cauda equina lesions
(Krogh et al., 2002), similar to the effects seen in
the urinary bladder (Koldewijn et al., 1994). There
also appears to be a relative increase in giant rectal

contractions in people with supraconal injuries
that may be important in their ability to induce
defecation by means of mechanical (digital) stim-
ulation of the rectum. It is postulated that this
action may trigger both giant rectal contractions
and the recto-anal inhibitory reflex simultaneously
(Krogh et al., 2003).
The effects of spinal cord injury on the recto-
anal inhibitory reflex are unclear. The reflex has
been shown to be both increased (Sun et al.,
1990b) and not (Frenckner, 1975) in those with
chronic supraconal injury. It has also been shown
to be increased (Devroede et al., 1979) or normal
(Sun et al., 1990) in people with conal or cauda
equina lesions.
The anal sphincter resting pressure has been
shown to be lowered in subjects with supraconal
lesions but not in those with conal or caudal le-
sions (Krogh et al., 2002). Fecal incontinence can
therefore be a common problem and is strongly
associated with complete lesions (Krogh et al.,
1997). Inevitably a component of fecal inconti-
nence can be the result of using laxatives to over-
come constipation. Finally, although we are
increasingly aware of the dynamic role of the pel-
vic floor musculature in defecation, the extent of
its likely dysfunction in people with spinal cord
injury is unclear.
Colorectal transport and defecation in acute spinal
cord injury
Acutely (within 3 weeks of injury) colorectal tran-
sit times (and total gastrointestinal transit times) in
those with both supraconal and conal injuries have
been found to be at least twice as long as those of
a healthy control group (Krogh et al., 2002).
Rectosigmoid transit times were also significantly
prolonged after conal/caudal lesions. No doubt
these prolonged transit times may also, however,
be the consequence of other factors encountered
by the acutely ill spinal cord injury patients such
as immobilisation, usage of opioid analgesics,
reduced enteral diet intake and the unpredictable
effects of spinal shock and trauma generally.
With regard to the anorectum, Denny-Brown
and Robertson showed that the rectal wall lacked
355
any contractile activity in a subject who had sus-
tained supraconal cord injury days earlier (Denny-
Brown and Robertson, 1935). Reflex defecation is
however, thought to return between 4 and 8 weeks
of the injury (Kuhn, 1947; Krogh et al., 2002).
Finally, the anal resting pressures were not shown
to be significantly lowered in an acute setting with
either supraconal or conal lesions (Krogh et al.,
2002) but the ability to voluntarily contract the
external sphincter was reduced or absent (Frenck-
ner, 1975; Sun et al., 1990a).
In the acute setting, therefore, spinal cord injury
causes ileus (prolonged total gastrointestinal tran-
sit times), constipation (prolonged colonic transit
times) and fecal incontinence (passive leakage).
Conclusion
The number of people with traumatic spinal cord
injury in most countries can only be estimated,
although the overall incidence is thought to be in
the region of 12–40 per million annually (Illis,
2004). With improved medical awareness and care,
the life expectancy for paraplegics surviving the
acute phase is now approaching the life expectancy
of the general population (Hartkopp et al., 1997).
Consequently, activities that affect quality of life
such as effective bowel management are deemed
by patients to be important, more so even than
bladder or sexual functions.
The inability to defecate normally means that
this ‘normal’ bodily function not only occupies a
significant part of daily living but also acts as a
drain on resources e.g., the use of laxatives, toi-
letries, dependence on physical assistance (Glick-
man and Kamm, 1996; Krogh et al., 1997). There
are also potential long-term effects of poor colo-
rectal motility and function in these people who
may go on to develop diverticular disease, volvulus
and haemorrhoids (Steins et al., 1997). As a result,
some have advocated early colostomies (Branagan
et al., 2003).
The mechanisms and control of defecation both
in healthy and spinal cord injured individuals are
still poorly understood. The recent development
of a pig model of defecation (Ramalingam,
personal communication), should enable research
356
into defecation and its control in a species with
similar ano-rectal function to the human. Areas of
research using imaging techniques, such as dy-
namic magnetic resonance imaging, therapeutic
techniques, such as sacral nerve stimulation, and
genetic manipulation, such as stem cell therapy,
will also no doubt help further unfold the
physiology of defecation.
References
Abrahamsson, H., Antov, S. and Bosaeus, I. (1988) Gastroin-
testinal and colonic segmental transit time evaluated by a
single abdominal X-ray in healthy subjects and constipated
patients. Scand. J. Gastroenterol,, 152(Suppl.): 72–80.
Akervall, S., Fasth, S., Nordgren, S., Oresland, T. and Hulten,
L. (1989) Rectal reservoir and sensory function studied by
graded isobaric distension in normal man. Gut, 30: 496–502.
Aldridge, R.T. and Campbell, P.E. (1968) Ganglion cell distri-
bution in the normal rectum and anal canal. A basis for the
diagnosis of Hirschsprung’s disease by anorectal biopsy.
J. Pediatr. Surg., 3: 475–490.
Bannister, L.H. (1995) Alimentary system. In: Williams P.L.
(Ed.), Gray’s Anatomy (38th ed.). Churchill-Livingstone,
London, pp. 1774–1786.
Baumgarten, H.G., Holstein, A.F. and Stelzner, F. (1971) Dif-
ferences in the innervation of the large intestine and the in-
ternal sphincter of the anus in mammals and humans. Verh.
Anat. Ges., 66: 43–47.
Beuret-Blanquart, F., Weber, J., Gouverneur, J.P., De-
mangeon, S. and Denis, P. (1990) Colonic transit time and
anorectal manometric anomalies in 19 patients with complete
transection of the spinal cord. J. Auton. Nerv. Syst., 30:
199–207.
Brading, A.F. and Sibley, G.N. (1982) A superfusion apparatus
to study field stimulation of smooth muscle from mammalian
urinary bladder. J. Physiol., 334: 11–12.
Branagan, G., Tromans, A. and Finnis, D. (2003) Effect of
stoma formation on bowel care and quality of life in patients
with spinal cord injury. Spinal Cord, 41: 680–683.
Bruninga, K. and Camilleri, M. (1997) Colonic motility and
tone after spinal cord and cauda equina injury. Am. J. Gas-
troenterol., 92: 891–894.
Buckle, D.A., McKenna, J. and Guanci, M. (1999) Spinal cord
trauma. Nurs. Clin. North Am., 34: 661–687.
Burleigh, D.E., D’Mello, A. and Parks, A.G. (1979) Responses
of isolated human internal anal sphincter to drugs and elec-
trical field stimulation. Gastroenterology, 77: 484–490.
Cajal, S.R. (1893) Sur les ganglions et plexus nerveux de l’in-
testin. C.R. Soc. Biol., 45: 217–223.
Cook, I.J., Furukawa, Y., Panagopoulos, V., Collins, P.J. and
Dent, J. (2000) Relationships between spatial patterns of co-
lonic pressure and individual movements of content. Am. J.
Physiol. Gastrointest. Liver Physiol., 278: G329–G341.
de Groat, W.C., Booth, A.M., Milne, R.J. and Roppolo, J.R.
(1982) Parasympathetic preganglionic neurons in the sacral
spinal cord. J. Auton. Nerv. Syst., 5: 23–43.
de Groat, W.C. and Krier, J. (1978) The sacral parasympathetic
reflex pathway regulating colonic motility and defaecation in
the cat. J. Physiol., 276: 481–500.
de Groat, W.C., Nadelhaft, I., Milne, R.J., Booth, A.M.,
Morgan, C. and Thor, K. (1981) Organization of the sacral
parasympathetic reflex pathways to the urinary bladder and
large intestine. J. Auton. Nerv. Syst., 3: 135–160.
Denny-Brown, D. and Robertson, E.G. (1935) An investigation
of the nervous control of defecation. Brain, 58: 256–310.
Devroede, G., Arhan, P., Duguay, C., Tetreault, L., Akoury, H.
and Perey, B. (1979) Traumatic constipation. Gastroenterol-
ogy, 77: 1258–1267.
Frenckner, B. (1975) Function of the anal sphincters in spinal
man. Gut, 16: 638–644.
Frenckner, B. and Euler, C.V. (1975) Influence of pudendal
block on the function of the anal sphincters. Gut, 16: 482–489.
Frenckner, B. and Ihre, T. (1976) Influence of autonomic nerves
on the internal and sphincter in man. Gut, 17: 306–312.
Fukuda, H. and Fukai, K. (1986) Postural change and straining
induced by distension of the rectum, vagina and urinary
bladder of decerebrate dogs. Brain Res., 380: 276–286.
Fukuda, H., Fukai, K., Yamane, M. and Okada, H. (1981)
Pontine reticular unit responses to pelvic nerve and colonic
mechanical stimulation in the dog. Brain Res., 207: 59–71.
Furness, J.B. and Costa, M. (1987) The Enteric Nervous Sys-
tem. Churchill-Livingstone, New York.
Gabella, G. (2001) Development and ageing of intestinal mus-
culature and nerves: the guinea-pig taenia coli. J. Ne-
urocytol., 30: 733–766.
Garcia, D., Hita, G., Mompean, B., Hernandez, A., Pellicer, E.,
Morales, G. and Parrilla, P. (1991) Colonic motility: electric
and manometric description of mass movement. Dis. Colon
Rectum,, 34: 577–584.
Glickman, S. and Kamm, M.A. (1996) Bowel dysfunction in
spinal-cord-injury patients. Lancet, 347: 1651–1653.
Gregersen, H. and Ehrlein, H.-J. (1996) Motility studies in lab-
oratory animals. In: Jensen S.L., Grgersen H., Amiri S. and
Moody F. (Eds.), Essentials of Experimental Surgery. Gastro-
enterology. Harwood Academic Publishers, New York,
pp. 17/11–17/39.
Gutierrez, J.G. and Shah, A.N. (1975) Autonomic control of
the internal anal sphincter in man. In: Trappen V. (Ed.),
Fifth International Symposium on Gastrointestinal Motility.
Typoff Press, Leuven, pp. 363–373.
Hartkopp, A., Bronnum-Hansen, H., Seidenschnur, A.M. and
Biering-Sorensen, F. (1997) Survival and cause of death after
traumatic spinal cord injury. A long-term epidemiological
survey from Denmark. Spinal Cord, 35: 76–85.
Holmes, A.M. (1961) Observations on the intrinsic innervation
of the rectum and anal canal. J. Anat., 95: 416–422.
Illis, L.S. (2004) The case for specialist units. Spinal Cord,
42: 443–446.
Koldewijn, E.L., Van Kerrebroeck, P.E., Rosier, P.F.,
Wijkstra, H. and Debruyne, F.M. (1994) Bladder compliance

after posterior sacral root rhizotomies and anterior sacral
root stimulation. J. Urol., 151: 955–960.
Krogh, K., Mosdal, C., Gregersen, H. and Laurberg, S. (2002)
Rectal wall properties in patients with acute and chronic
spinal cord lesions. Dis. Colon Rectum, 45: 641–649.
Krogh, K., Mosdal, C. and Laurberg, S. (2000) Gastrointestinal
and segmental colonic transit times in patients with acute and
chronic spinal cord lesions. Spinal Cord, 38: 615–621.
Krogh, K., Nielsen, J., Djurhuus, J.C., Mosdal, C., Sabroe, S.
and Laurberg, S. (1997) Colorectal function in patients with
spinal cord lesions. Dis. Colon Rectum,, 40: 1233–1239.
Krogh, K., Olsen, N., Christensen, P., Madsen, J.L. and
Laurberg, S. (2003) Colorectal transport during defecation in
patients with lesions of the sacral spinal cord. Ne-
urogastroenterol. Motil., 15: 25–31.
Kuhn, W.G. (1947) The care and rehabilitation of patients with
injuries of the spinal cord and cauda equina. J. Neurosurg., 4:
40–68.
Kuriyama, H., Kitamura, K., Itoh, T. and Inoue, R. (1998)
Physiological features of visceral smooth muscle cells, with
special reference to receptors and ion channels. Physiol. Rev.,
78: 811–920.
Mackel, R. (1979) Segmental and descending control of the
external urethral and anal sphincters in the cat. J. Physiol.,
294: 105–122.
Martner, J. (1975) Influences on the defecation and micturition
reflexes by the cerebellar fastigial nucleus. Acta Physiol.
Scand., 94: 95–104.
Metcalf, A.M., Phillips, S.F., Zinsmeister, A.R., MacCarty, R.L.,
Beart, R.W. and Wolff, B.G. (1987) Simplified assessment of
segmental colonic transit. Gastroenterology, 92: 40–47.
Miller, R., Bartolo, D.C., Cervero, F. and Mortensen, N.J.
(1988a) Anorectal sampling: a comparison of normal and
incontinent patients. Br. J. Surg., 75: 44–47.
Miller, R., Lewis, G.T., Bartolo, D.C., Cervero, F. and Mor-
tensen, N.J. (1988b) Sensory discrimination and dynamic
activity in the anorectum: evidence using a new ambulatory
technique. Br. J. Surg., 75: 1003–1007.
Nino-Murcia, M., Stone, J.M., Chang, P.J. and Perkash, I.
(1990) Colonic transit in spinal cord-injured patients. Invest.
Radiol., 25: 109–112.
O’Kelly, T., Brading, A. and Mortensen, N. (1993a) Nerve
mediated relaxation of the human internal anal sphincter: the
role of nitric oxide. Gut, 34: 689–693.
O’Kelly, T.J., Brading, A. and Mortensen, N.J. (1993b) In vitro
response of the human anal canal longitudinal muscle layer
to cholinergic and adrenergic stimulation: evidence of sphinc-
ter specialization. Br. J. Surg., 80: 1337–1341.
O’Kelly, T.J., Davies, J.R., Brading, A.F. and Mortensen, N.J.
(1994) Distribution of nitric oxide synthase containing neu-
rons in the rectal myenteric plexus and anal canal. Morpho-
logic evidence that nitric oxide mediates the rectoanal
inhibitory reflex. Dis. Colon Rectum,, 37: 350–357.
Penninckx, F., Lestar, B. and Kerremans, R. (1992) The inter-
nal anal sphincter: mechanisms of control and its role in
maintaining anal continence. Baillieres Clin. Gastroenterol.,
6: 193–214.
357
Proano, M., Camilleri, M., Phillips, S.F., Brown, M.L. and
Thomforde, G.M. (1990) Transit of solids through the hu-
man colon: regional quantification in the unprepared bowel.
Am. J. Physiol., 258: G856–G862.
Quigley, E.M. (2002) Colonic motility and colonic function. In:
Pemberton J., Swash M. and Henry M. (Eds.), The Pelvic
Floor: Its Function and Disorders. W.B Saunders, London,
pp. 84–93.
Rao, S.S., Sadeghi, P., Beaty, J., Kavlock, R. and Ackerson, K.
(2001) Ambulatory 24-h colonic manometry in healthy hu-
mans. Am. J. Physiol. Gastrointest. Liver Physiol., 280:
G629–G639.
Sanders, K.M. (1996) A case for interstitial cells of Cajal as
pacemakers and mediators of neurotransmission in the gas-
trointestinal tract. Gastroenterology, 111: 492–515.
Sanders, K.M., Ordog, T., Koh, S.D., Torihashi, S. and Ward,
S.M. (1999) Development and plasticity of interstitial cells of
Cajal. Neurogastroenterol. Motil., 11: 311–338.
Sanders, K.M., Ordog, T. and Ward, S.M. (2002) Physiology
and pathophysiology of the interstitial cells of Cajal: from
bench to bedside. IV. Genetic and animal models of GI mo-
tility disorders caused by loss of interstitial cells of Cajal. Am.
J. Physiol. Gastrointest. Liver Physiol., 282: G747–G756.
Sarna, S.K. (1991) Physiology and pathophysiology of colonic
motor activity (1). Dig. Dis. Sci., 36: 827–862.
Schroder, H.D. (1981) Onuf’s nucleus X: a morphological study
of a human spinal nucleus. Anat. Embryol. (Berl.), 162:
443–453.
Sorensen, S.M., Gregersen, H., Sorensen, S. and Djurhuus, J.C.
(1989) Spontaneous anorectal pressure activity. Evidence of
internal anal sphincter contractions in response to rectal
pressure waves. Scand. J. Gastroenterol., 24: 115–200.
Stebbing, J.F. (1998) Nitric oxide synthase neurones and ne-
uromuscular behaviour of the anorectum. Ann. R. Coll.
Surg. Engl., 80: 137–145.
Steins, S.A. and Bergman. S.B. and Goetz, L.L. (1997) Ne-
urogenic bowel dysfunction after spinal cord injury: clinical
evaluation and rehabilitative management. Arch. Phys. Med.
Rehabil., 78: S86–S102.
Sun, W.M., MacDonagh, R., Forster, D., Thomas, D.G.,
Smallwood, R. and Read, N.W. (1995) Anorectal function in
patients with complete spinal transection before and after
sacral posterior rhizotomy. Gastroenterology, 108: 990–998.
Sun, W.M., Read, N.W. and Donnelly, T.C. (1990a) Anorectal
function in incontinent patients with cerebrospinal disease.
Gastroenterology, 99: 1372–1379.
Sun, W.M., Read, N.W., Miner, P.B., Kerrigan, D.D. and
Donnelly, T.C. (1990b) The role of transient internal sphinc-
ter relaxation in faecal incontinence? Int. J. Colorectal Dis.,
5: 31–36.
Takaki, M., Neya, T. and Nakayama, S. (1987) Functional role
of lumbar sympathetic nerves and supraspinal mechanism in
the defecation reflex of the cat. Acta. Med. Okayama,, 41:
249–257.
Tomita, T. (1981) Electrical activity (spikes and slow waves) in
gastrointestinal smooth muscle. In: Bulbring E., Brading
358
A.F., Jones A.W. and Tomita T. (Eds.), Smooth Muscle: An
assessment of Current Knowledge. E. Arnold, London.
Ward, S.M., Ordog, T., Koh, S.D., Baker, S.A., Jun, J.Y.,
Amberg, G., Monaghan, K. and Sanders, K.M. (2000) Pace-
making in interstitial cells of Cajal depends upon calcium
handling by endoplasmic reticulum and mitochondria. J.
Physiol., 525(Part 2): 355–361.
Ward, S.M. and Sanders, K.M. (2001) Physiology and patho-
physiology of the interstitial cell of Cajal: from bench to
bedside. I. Functional development and plasticity of intersti-
tial cells of Cajal networks. Am. J. Physiol. Gastrointest.
Liver Physiol., 281: G602–G611.
Weinberg, A.G. (1970) The anorectal myenteric plexus: its re-
lation to hypoganglionosis of the colon. Am. J. Clin. Pathol.,
54: 637–642.
Zbar, A.P., Jayne, D.G., Mathur, D., Ambrose, N.S. and
Guillou, P.J. (2000) The importance of the internal anal
sphincter (IAS) in maintaining continence: anatomical, phys-
iological and pharmacological considerations. Colorectal
Dis., 2: 193–202.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 24

Alterations in eliminative and sexual reflexes after

spinal cord injury: defecatory function and
development of spasticity in pelvic floor musculature

Yvette S. Nout1, Gail M. Leedy2, Michael S. Beattie1 and Jacqueline C. Bresnahan1,

1
Department of Neuroscience, Laboratory of CNS Repair and Spinal Trauma and Repair Laboratories, The Ohio State
University College of Medicine and Public Health, Columbus, OH, USA
2
Division of Social Work, University of Wyoming, Laramie, WY, USA

Abstract: Spinal cord injury often results in loss of normal eliminative and sexual functions. This chapter is
focused on defecatory function, although aspects of micturition and erectile function will be covered as well
due to the overlap in anatomical organization and response to injury. These systems have both autonomic
and somatic components, and are organized in the thoracolumbar (sympathetic), lumbosacral (somatic),
and sacral (parasympathetic) spinal cord. Loss of supraspinal descending control and plasticity-mediated
alterations at the level of the spinal cord, result in loss of voluntary control and in abnormal functioning of
these systems including the development of dyssynergies and spasticity. There are several useful models of
spinal cord injury in rodents that exhibit many of the autonomic dysfunctions observed after spinal cord
injury in humans. Numerous studies involving these animal models have demonstrated development of
abnormalities in bladder, external anal sphincter, and erectile function, such as detrusor–sphinc-
ter–dyssynergia and external anal sphincter hyperreflexia. Here we review many of these studies and
show some of the anatomical alterations that develop within the spinal cord during the development of
these hyperreflexias. Furthermore, we show that spasticity develops in other pelvic floor musculature as
well, such as the bulbospongiosus muscle, which results in increased duration and magnitude of pressures
developed during erectile events and increased duration of micturition. Advances and continued improve-
ment in the use of current animal models of spinal cord injury should encourage and increase the laboratory
work devoted to this relatively neglected area of experimental spinal cord injury.

Introduction and somatic motor pools in the caudal lumbar/

Both eliminative and sexual functions are com-
promised following spinal cord injury (Comarr,
1970; Frenckner, 1975; Pedersen, 1983). These sys-
tems have both autonomic and somatic compo-
nents, and are organized by sympathetic neuronal
pools in the caudal thoracic/upper lumbar cord,
parasympathetic neuronal pools in the sacral cord,
Corresponding author. Tel.: +614-292-2206;
E-mail: bresnahan.1@osu.edu
upper sacral cord. Although the basic afferent and
efferent limbs of the reflex components are seg-
mentally organized, the loss of brainstem and
forebrain descending input results in loss of vol-
untary control and in abnormal functioning of
these systems including the development of dyssy-
nergies and spasticity. Urinary function has been
studied most in animal models, and the complete
loss of descending input initially causes an absence
of bladder emptying which is gradually replaced
by spinal reflex circuit activity that is initiated by a

DOI: 10.1016/S0079-6123(05)52024-7 359

360
different population of afferents (e.g., c-fiber
afferents) than in the normal condition (De Groat
et al., 1990, 1998). While some recovery of this
function occurs, residual deficits remain in the co-
ordination of the autonomic efferents to the
detrusor muscle and the somatic efferents to the
external urethral sphincter muscle (so-called
bladder-sphincter dyssynergia) producing inade-
quate bladder emptying. Defecatory function has
been less well studied but exhibits similar respons-
es to the loss of descending control causing inad-
equate voiding, fecal impaction, and chronic
constipation in man (Frenckner 1975; Cosman
et al., 1991) due, perhaps, to dyssynergia and
external anal sphincter spasticity (Holmes et al.,
1998, 2005). Sexual reflexes are similarly disrupted;
in humans, loss of descending input results in
facilitated spinally mediated erectile activity but
disruption of ejaculatory function and seminal flu-
id composition (Higgins, 1979; Basu et al., 2004).
The focus of the present chapter will be on def-
ecation as other eliminative and sexual functions
will be covered in other chapters (for urinary blad-
der and sexual function, see other chapters in this
volume). However, we will discuss these other sys-
tems to some extent here as well, as their anatom-
ical organization is co-extensive, their functions
overlap, and response to injury is similar. We will
first describe the anatomy of the distal gastrointes-
tinal tract, its innervation including the central and
peripheral components; this will be followed by a
discussion of normal eliminative function and then
of dysfunctions resulting from spinal cord injury.
Neuroanatomy
Peripheral components
The most distal portion of the gastrointestinal
tract is formed by the rectum and anus. Innerva-
tion of the rectum is similar to that of the colon,
but innervation of the anus is more complex (for a
comprehensive review, see Jänig and McLachlan,
1987). The anus controls passage of fecal matter
by the degree of constriction of the internal and
external anal sphincters (Schuster, 1968; Gonella
et al., 1987). The internal anal sphincter is formed
by thickening of circular smooth muscle that lies
immediately inside the anus. Similar to the rest of
the gastrointestinal smooth muscle, the internal
anal sphincter contains a myenteric plexus within
the tunica muscularis. Furthermore, the internal
sphincter muscle receives autonomic innervation
via the pelvic plexus. The external anal sphincter is
composed of striated muscle that surrounds the
internal anal sphincter and extends distal to it. The
external sphincter is controlled by nerve fibers in
the pudendal nerve, which in the rat has motor,
sensory, and autonomic components (Rexed, 1954;
Katagiri et al., 1986; Holstege and Tan, 1987;
Paxinos, 1995). While the internal anal sphincter is
the main mechanism for continence, accounting for
the majority of the resting pressure, subconsciously
the external sphincter is usually kept constricted
unless conscious signals inhibit constriction
(Gonella et al., 1987). Motor units of the external
anal sphincter have been shown to be tonically
active (Krier, 1985; Gonella et al., 1987), whereas
the other pudendal motor units are effectively qui-
escent (Dubrovsky et al., 1985; Holmes et al., 1994).
The rectum and anal canal are supported by the
pelvic diaphragmatic musculature, which partici-
pates in the functions of fecal continence and def-
ecation. Paired rectococcygeal muscles originate
on each side of the rectum and attach dorsally to
the base of the tail. These smooth muscles are in-
nervated by autonomic fibers from the pelvic plex-
us and shorten the rectum during defecation to
assist evacuation of feces. Laterally the illeocauda-
lis and pubocaudalis and the coccygeus muscles
surround the rectum. These striated muscles are
innervated by the 6th lumbar (L6) to 1st sacral
(S1) nerves in the rat (Bremer et al., 2003) and help
to compress the rectum during defecation.
Pelvic plexus
The pelvic plexus is a crossroad of autonomic
nerves and ganglia supplying the rectum, the lower
urinary tract, and the genital tract (Paxinos, 1995).
In the male rat, the main components of this plex-
us are the right and left pelvic ganglia, also referred
to as the hypogastric ganglia, or the major pelvic
ganglia. These ganglia are positioned on either side

of the prostate, ventral to the rectum, and caudal
to the ureter and vas deferens, and are unusual
among autonomic ganglia, since they contain both
parasympathetic and sympathetic post-ganglionic
neurons. The main afferents to the pelvic ganglia
are the hypogastric nerve and the pelvic nerve
(Paxinos, 1995).
The majority of fibers comprising the hypogastric
nerve originate in the intermediolateral cell column
of the thoracolumbar cord [thoracic (T)11–L2]
(Baron and Jänig, 1991). These preganglionic sym-
pathetic neurons send their axons to sympathetic
ganglion cells of the bladder in the sympathetic
chain (T12–L6), pelvic ganglia, and to the inferior
mesenteric ganglia (Vera and Nadelhaft, 1992). The
caudal continuation of the inferior mesenteric gan-
glion forms the hypogastric nerve, which carries the
majority of the sympathetic input to the pelvic
plexus. A small number of hypogastric fibers have
their cells of origin located in either the sympathetic
chain or the inferior mesenteric ganglion.
The pelvic nerve carries the parasympathetic in-
put and originates from the last lumbar (L6) and
first sacral (S1) spinal nerves in the rat (Purinton
et al., 1973). The pelvic nerve also carries a small
portion of sympathetic fibers to the pelvic viscera.
Numerous small efferents arise from the pelvic
ganglion and supply the rectum, the ureter, the
vas deferens, the seminal vesicles, the prostate, the
bladder, and the urethra. In the female rat
the pelvic ganglion, which is also referred to as
the paracervical or Frankenhauser ganglion, is
smaller than that in males and lies against the lat-
eral wall of the uterine cervix. Afferent neurons
with fibers in the pelvic nerve are located in the
L6–S1 dorsal root ganglia in the rat and carry
sensory information from the descending colon,
bladder, urethra, and sex organs (Nadelhaft and
Booth, 1984). These fibers enter the tract of
Lissauer, encircle the dorsal horn with a medial
component terminating in the dorsal commissural
gray and the lateral component terminating in the
sacral parasympathetic nucleus.
Pudendal nerve
The pudendal nerve arises from L5–L6 in the rat
and carries efferent fibers to the coccygeus, internal
361
obturator, ventral and dorsal bulbospongiosus
(also termed bulbocavernosus), ischiocavernosus,
external urethral and anal sphincter muscles, and
afferent fibers from the penis, prepuce, scrotum,
and ventral-proximal tail (McKenna and Nadelhaft,
1986). The sacral plexus is the complex formed by
the bridge-like structure connecting the pudendal
nerve with the lumbosacral trunk, and two nerve
branches emerging from it, one innervating the
proximal half of the scrotal skin, and the other,
known as the motor branch, innervating the mus-
cles at the base of the penis (Pacheco et al., 1997).
These branches are only considered as a part of the
sacral plexus because they integrate axons from
both the lumbosacral trunk and pudendal nerve.
In rats, retrograde tracer studies of the pudendal
nerve have demonstrated the presence of two mo-
tor nuclei in the ventral horn gray matter of the
L5–L6 spinal cord segments, the dorsomedial, and
dorsolateral cell columns (Fig. 1; Breedlove and
Arnold, 1980; McKenna and Nadelhaft, 1986;
Vera and Nadelhaft, 1992). These nuclei are the
homologs to the nucleus of Onuf, which is found
in other species, and contains motor neurons in-
nervating the anal and urethral sphincters, bulb-
ospongiosus, and ischiocavernosus muscles, all
located in one cell group. McKenna and Nadelhaft
(1986) also demonstrated that in the female rat,
urethral sphincter motor neurons accounted for
almost all the dorsolateral nucleus motor neurons,
and anal sphincter motor neurons accounted for
almost all the dorsomedial motor neurons; the is-
chiocavernosus and bulbospongiosus muscles are
vestigial in the female rat. In the male, neurons
innervating the external anal sphincter and bulb-
ospongiosus muscles are found intermingled in the
dorsomedial nucleus. In contrast, in the dorsolat-
eral nucleus, the urethral sphincter neurons are
located in the lateral portion of the nucleus and the
ischiocavernosus neurons are located in the medial
portion (McKenna and Nadelhaft, 1986; Collins
et al., 1992; Hermann et al., 1998). There are sig-
nificantly more neurons in both nuclei in the male
than in the female and the neurons are larger in the
male. However, perineal muscle tracer injections
have established that there is no difference be-
tween males and females in the number of motor
neurons innervating the external anal or urethral
362

Fig. 1. The caudal lumbar and rostral sacral spinal cord contains the neuronal circuitry that organizes the parasympathetic com-
ponents of bladder, bowel, and sexual function. DCG: dorsal commissural gray; SPN: sacral parasympathetic nucleus; DL: dorso-
lateral nucleus; DM: dorsomedial nucleus; DF: dorsal funiculus; DH: dorsal horn; CC: central canal; VM: ventromedial nucleus; RDL:
retrodorsolateral nucleus. (A) The cord at this level contains the SPN and the associated somatic motor columns innervating the pelvic
floor musculature, the DM, and DL nuclei. (B, C) The L6 segment contains the DM and DL nuclei that are characterized by a lack of
myelinated fibers; this is most easily seen in the DL. The boxed area in (B) is shown at higher magnification in (C). In this particular
section, bulbospongiosus motor neuron cell bodies are retrogradely labeled with horseradish peroxidase (HRP) and can be easily
identified in the higher power image shown in (C). This is a plastic embedded section containing the labeled bulbospongiosus motor
neurons cell bodies; diaminobenzidine was used to visualize the horseradish peroxidase in the neurons. After processing the tissue for
HRP, it was stained en bloc with osmium and uranyl acetate and embedded in plastic for subsequent sectioning for electron mi-
croscopy. Myelinated fibers appear black (note the darker staining of the ventral white commissure just above the DM nuclei and in the
white matter of the dorsal lateral and ventral funiculi). The retrogradely labeled neurons also appear dark in this section (arrow in C).
(D) This micrograph shows a 1.0 mm thick, toluidine blue stained section of the DM area enclosed by the circle in (C). Two motor
neuron cell bodies are shown in cross section; the cell on the left has a clear nucleus, clumps of rough endoplasmic reticulum in the
cytoplasm, and a large primary dendrite extending dorsally (toward the letter D). The other cell body has clumps of rough endoplasmic
reticulum and short primary dendrites that extend mostly out of the plane of section. There is a high density of cross-cut dendrites that
appear as small white circles; these are rostrocaudally oriented bundles of dendrites. Black myelinated fibers can be seen at the top of
the micrograph and a couple of fibers can be seen to traverse the nucleus between the two cell bodies. The magnification is too low to
identify the HRP-stained elements in the cytoplasm. (E, F). These low power dark field micrographs of adjacent sections show the
extensive dendritic arborizations of bulbospongiosus motor neurons retrogradely labeled with horseradish peroxidase from muscle
injections (tetramethyl benzidine processing). The bundles extend into the DCG region that receives pudendal and pelvic nerve afferent
input, into the DL nucleus that contains motor neurons of the ischiocavernosus and external urethral sphincter muscles, and over to
the contralateral DM nucleus.

sphincters (McKenna and Nadelhaft, 1986). These
nuclei have a number of special anatomic features
(see Fig. 1) similar to the phrenic nucleus, a motor
nucleus in the cervical spinal cord that innervates
the diaphragm. These nuclei lack myelinated fibers
making them appear translucent, similar to the
substantia gelatinosa. The motor neurons exhibit
extensive dendritic bundling (Roney et al., 1979;
Schroder, 1980; McKenna and Nadelhaft, 1986;
Sasaki, 1994); the rostrocaudally oriented dendri-
tic bundles can be seen in Fig. 1D. Dense bundles
of dendrites from bulbospongiosus motor neurons
can be seen in Figs. 1E and F (arrows), intercon-
necting the dorsomedial nuclei on the two sides as
well as the dorsomedial and dorsolateral nuclei on
the same side. A dorsally extending group projects
into the dorsal commissural gray (Figs. 1E and F).
Peshori et al. (1995) demonstrated that motor
neurons in the contralateral dorsomedial nucleus
could be transneuronally labeled with wheatgerm-
agglutinin horseradish peroxidase after a unilater-
al muscle injection and dye-coupling has been
demonstrated between bulbocavernosus motor
neurons by Matsumoto et al. (1988). At the ultra-
structural level, the dendrites are connected with
tight junctions or puncta adherentia (Ramirez-
Leon and Ulfhake, 1993; see Figs. 2A and B) as
well as through presynaptic dendrites (PSD; see
Figs. 2A–C, and D). The synaptic arrangements in
this region are also somewhat unique, in that axon
terminals are commonly contacting more than one
dendrite (see, e.g., AT1 and AT2 in Fig. 2;
Ramirez-Leon and Ulfhake, 1993). The dense in-
nervation of this region by g-amino butyric acid
(GABA)-containing elements is also of note
(Ramirez-Leon and Ulfhake, 1993; Li et al., 1995).
It is likely that these anatomical arrangements can
provide for a more synchronized activation of
motor neurons controlling pelvic floor musculature
(e.g., the sphincters and erectile musculature).
The pudendal motor neuron dendrites occupy
most of the ventral horn and reach dorsally be-
yond the central canal (Sasaki, 1994; Peshori et al.,
1995; Hermann et al., 1998). In the cat, the sphinc-
ter motor neuron dendritic arbors have been
shown to extend into the regions where function-
ally associated preganglionic neurons in the sacral
parasympathetic nucleus are located (Beattie et al.,
363
1990). For example, the external anal sphincter
motor neurons project dendrites into the dorsal
band of the sacral parasympathetic nucleus where
the colonic preganglionics are located, and the ex-
ternal urethral sphincter motor neurons into its
lateral band where the bladder preganglionics re-
side. Similarly, the bulbospongiosus motor neurons
in the rat extend profuse dendrites into the dorsal
commissural region (Fig. 1E) where interneurons
important in sexual reflex function are located
(Peshori et al., 1995; Nadelhaft and Vera, 2001).
Pudendal nerve afferent neurons are located in
the L6 and S1 dorsal root ganglia (Nadelhaft and
Booth, 1984). In both sexes, pudendal afferent
fibers in the spinal cord are located in the dorsal
columns, the medial half of Lissauer’s tract, the
extreme medial edge of the dorsal horn, both ipsi-
laterally and contralaterally, and in a large termi-
nal field in the dorsal gray commissure. No
afferents have been detected in the intermediate
or ventral gray (McKenna and Nadelhaft, 1986).
Defecation
Defecation reflexes
In normal animals and humans, defecation is in-
itiated by defecation reflexes (Gonella, 1987;
Guyton and Hall, 2000). One of these reflexes is
an intrinsic reflex mediated by the local enteric
nervous system in the rectal wall. Distention of the
rectal wall initiates afferent signals that spread
through the myenteric plexus to initiate peristaltic
waves in the descending colon, sigmoid, and rec-
tum, forcing feces toward the anus. As the per-
istaltic wave approaches the anus, the internal anal
sphincter is relaxed by inhibitory signals from the
myenteric plexus. A rectoanal inhibition reflex oc-
curs in which as the rectum fills and the internal
anal sphincter relaxes, the external anal sphincter
contracts involuntarily and defecation only occurs
following further rectal distension inducing con-
comitant external anal sphincter relaxation
(Gonella et al., 1987). This local intrinsic defeca-
tion reflex, however, is by itself relatively weak and
usually must be fortified by another type of def-
ecation reflex, a parasympathetic defecation reflex
364

Fig. 2. Electron micrographs showing a variety of interactions between dendritic elements in Onuf’s nucleus in the cat, the homolog of
the rat dorsomedial (DM), and dorsolateral (DL) nuclei. Two dendrites (D1 and D2) of retrogradely identified pudendal motoneurons
receive synaptic input from a variety of axon terminals, some of which contact both dendrites (AT1 and AT2). The dendrites also are
closely apposed and have regions of direct appositions with puncta adherentia (small arrows in B). An adjacent dendrite (PSD) gives
rise to an element that appears to be presynaptic to D1 (synaptic specializations on D1 are indicated by the small arrows on the bottom
of the higher power figure shown in D). This dendritic excrescence is also post-synaptic to two adjacent axon terminals, one on the right
and one on the left, both of which are presynaptic to D1. In the higher power figure shown in (D), synaptic vesicles in the right terminal
are clustered at the membrane and a clear post-synaptic density is visible between the small arrows. The dendrite (D1) also has a region
containing vesicles (VES) and may be presynaptic to a small dendrite (shown in C at higher magnification). There is a synaptic
specialization between the small arrows at the interface with D1 as well as with a synaptic terminal surrounding this dendritic element
(small arrows on the left).

that involves the sacral segments of the spinal
cord. When nerve endings in the rectum are stim-
ulated by distension, signals are transmitted first
into the spinal cord and then reflexively back to
the descending colon, rectum, and anus by way of
parasympathetic nerve fibers in the pelvic nerves.
These parasympathetic signals greatly intensify the
peristaltic waves and relax the internal anal
sphincter, and thus convert the intrinsic myenter-
ic defecation reflex from a weak effort into a pow-
erful process of defecation.
The afferent defecation signals entering the spi-
nal cord initiate other effects such as closure of the
glottis and contracture of the abdominal wall
muscles to force fecal contents of the colon down-
ward and at the same time cause the pelvic floor to

relax downward and pull outward on the anal ring
to evaginate feces.
Supraspinal control mechanisms
Although in the rat reflex defecation is mainly or-
ganized at the spinal level, there is evidence for
supraspinal centers that may modify the functional
response (Maggi et al., 1988). These supraspinal
inputs strongly appear to influence the magnitude
and coordination of these reflexes, and generally
this descending supraspinal input is inhibitory to
the lumbosacral reflexes (described above as the
parasympathetic defecation reflex). Retrograde
tracer injections into the ventromedial gray of
the lumbar spinal cord labeled supraspinal input
primarily from the vestibular nuclei, the giganto-
cellular reticular nuclei, the medullary raphe nu-
clei, and the hypothalamic paraventricular nuclei
(Marson and McKenna, 1990; Shen et al., 1990;
Monaghan and Breedlove, 1991; Marson et al.,
1992). The ventrolateral gray of the lumbar spinal
cord has supraspinal input from the dorsolateral
pontine tegmental region, also referred to as Bar-
rington’s nucleus or the pontine micturition center
(Ding et al., 1995). Furthermore, transneuronal
tracing studies using pseudorabies virus injections
directly into the muscles involved in pelvic floor
reflexes have consistently confirmed these same
regions to be the source of supraspinal input
(Marson and McKenna, 1996; Tang et al., 1999;
Vizzard et al., 2000). Some studies have focused on
supraspinal input originating from the nucleus
raphe obscurus in the brainstem and the region in
the ventrolateral medulla referred to as the gig-
antocellular–lateral paragigantocellular complex.
Projections from this latter area are widespread
and include fibers in close apposition to external
anal sphincter and bulbospongiosus motor neu-
rons, and although targets from the nucleus raphe
obscurus are much more restricted, they also in-
clude contacts between nucleus raphe obscurus
terminals and bulbospongiosus and external anal
sphincter motor neurons (Hermann et al., 1998,
2003). More specifically, fibers from the nucleus
raphe obscurus ramify in close apposition to both
bulbospongiosus and external anal sphincter
365
motor neurons and terminals of nucleus raphe
obscurus projections in the lumbosacral spinal
cord have been shown to co-localize with se-
rotonin, thyrotropin-releasing hormone, and sub-
stance P (Ramirez-Leon et al., 1994).
In fact, the nucleus raphe obscurus is thought to
be an overall regulator of autonomic functions
controlled by both cranial and spinal autonomic
efferents. Rostral portions of the nucleus raphe
obscurus project to the dorsal vagal complex,
which controls proximal digestive functions
(McCann et al., 1989), whereas caudal portions
of the nucleus raphe obscurus project to the caudal
intermediolateral cell column, sacral parasympa-
thetic nucleus, and related somatic motor neurons,
which control distal digestive and other pelvic au-
tonomic functions. Also, the nucleus raphe ob-
scurus maintains direct projections to both
autonomic and somatic regions and to regions of
the intermediate gray, which contain putative in-
terneurons that contribute to the organization of
pelvic floor reflexes. The nucleus raphe obscurus is
thus in a position to modulate autonomic pre-
ganglionic and functionally related skeletal motor
neuron activity (Hermann et al., 1998).
Electrical stimulation of the nucleus raphe
obscurus results in reduction of spontaneous ano-
rectal activity, providing evidence for a direct
brainstem inhibitory circuit (Holmes et al., 1997a).
Also, nucleus raphe obscurus lesions elicit tran-
sient increases in anorectal reflex activity (Beattie
et al., 1996; Holmes et al., 2002) and disrupt male
copulatory behavior in rats (Yamanouchi and
Kakeyama, 1992). However, nucleus raphe ob-
scurus lesions do not have an effect on measures of
ex copula penile reflexes (Holmes et al., 2002). In
contrast, lesions of the gigantocellular–lateral
paragigantocellular complex did not affect meas-
ures of external anal sphincter activity but altered
penile reflexes significantly (Holmes et al., 2002). It
has been shown that the bulbospongiosus muscle
displays electromyographic activity during the
passage of urine (Sachs and Leipheimer, 1988;
Schmidt et al., 1995), and it is suggested that the
nucleus raphe obscurus projections may be specific
to pudendal eliminative reflexes, while other
descending brainstem projections are specific to
sexual reflexes (Holmes et al., 2002).
366
Projections from the gigantocellular–lateral
paragigantocellular complex are diffuse, descend-
ing to all levels of the spinal cord affecting sensory,
motor, and autonomic control circuits beyond
those for pudendal reflexes. The gigantocellu-
lar–lateral paragigantocellular complex may glo-
bally modulate multiple behaviors with somatic
and autonomic components such as reproductive
reflexes, acoustic startle, and nociception as pro-
posed by Hermann et al. (2003).
Neurotransmitters implicated in the descending
supraspinal control of pelvic and pudendal motor
neurons include serotonin, thyrotropin-releasing
hormone, and GABA (Arvidsson et al., 1990).
Medullary thyrotropin-releasing hormone neurons
diffusely project to the ventral horn of the spinal
cord and sympathetic neurons within the inter-
mediolateral cell column (Appel et al., 1987;
Hirsch and Helke, 1988). When applied intrathec-
ally at the level of the L5–L6 spinal cord, thy-
rotropin-releasing hormone resulted in contraction
of the internal anal sphincter, predominantly
through activation of the pelvic nerve. Although
sectioning of the hypogastric nerve had no effect
on thyrotropin-releasing hormone-induced inter-
nal anal sphincter activity, a role for thyrotropin-
releasing hormone activation of sympathetic
preganglionic neurons should not be excluded
since the pelvic nerve has been shown to contain
both sympathetic and parasympathetic fibers
(Hulsebosch and Coggeshall, 1982). It is suggest-
ed that the relative roles of the pelvic and
hypogastric nerves in internal anal sphincter
contractility differ with changes in activity of rec-
tal afferents (Holmes et al., 1995). Also, pudendal
motor neurons are under spinal descending
inhibitory control. Similar to the effect thyrotro-
pin-releasing hormone has on the internal anal
sphincter, high doses of thyrotropin-releasing hor-
mone applied to the lumbosacral spinal cord re-
sulted in an increase in firing rate of the external
anal sphincter (Holmes et al., 1997b). Thus, thy-
rotropin-releasing hormone appears to be involved
in maintaining fecal continence. Other studies have
shown that thyrotropin-releasing hormone plays
a critical role in modulation of gastrointestinal
function (McCann et al., 1989). Thyrotropin-
releasing hormone has been shown to increase
gastric secretion and motility, intestinal transport,
and is thought to play a role in stress-induced in-
creases in defecation (Miyata et al., 1992).
Although thyrotropin-releasing hormone appears
to have an excitatory effect on external anal sphinc-
ter motor neurons, intrathecally applied thyrotro-
pin-releasing hormone has an inhibitory effect on
penile erections, suggesting that although external
anal sphincter and bulbospongiosus motor neurons
are co-mingled within the same spinal nucleus, they
are discretely and differentially regulated by sepa-
rate neural circuits (Holmes et al., 1997b).
Eliminative functions following spinal cord injury
Following acute injury to the spinal cord an initial
period of ‘‘spinal shock’’ occurs, which is charac-
terized by areflexia and generally lasts for approx-
imately 24 h in humans (Ditunno et al., 2004).
Disruption of spinal cord tracts proximal to the
lumbosacral cord leads to loss of supraspinal con-
trol over normal eliminative and reproductive be-
haviors. These functions, which reflect coordinated
activity of both somatic and autonomic compo-
nents of the nervous system, are important ther-
apeutic targets as they represent critical problems
for the spinal cord injured population, and have
been less well studied than locomotion in models
of spinal cord injury (Anderson, 2004). After the
initial post-injury period of spinal shock, the basic
spinal reflexes control these functions independent
from supraspinal input. During the recovery phase
from spinal shock (1 day to 12 months), a gradual
return of reflexes may take place as well as the
development of hyperreflexia and spasticity
(Ditunno et al., 2004). Interestingly in the rat,
the initial loss and subsequent recovery of func-
tions that have autonomic components show a
similar pattern of recovery as locomotor function
(Holmes et al., 2005), suggesting commonality of
underlying recovery mechanisms.
In human patients with spinal cord injury,
coordination of autonomic and somatic defeca-
tion reflexes is lost secondary to disruption of
supraspinal control pathways. Compressive de-
struction of the conus medullaris of the spinal cord
can destroy the distal segments of the cord where,

in humans, the cord defecation reflex is integrated,
and this almost paralyzes defecation. More fre-
quently the spinal cord is injured more proximal,
between the conus medullaris and the brain, in
which case the voluntary portion of the defecation
act is blocked while the basic spinal cord reflex for
defecation is still intact.
In rat models, in addition to loss of control of
eliminative and sexual function, the lumbosacral
reflexes develop hyperreactivity, including in-
creased external anal sphincter contractions after
distention and hyperreflexia of erections in re-
sponse to slight tactile stimuli (Holmes et al., 1998,
2005). This is consistent with the finding in
humans that removal of supraspinal control leads
to spasticity and dyssynergia of urethral and anal
sphincters. In addition to hyperreflexia of the
external anal sphincter, it is well recognized that
hyperreflexia of the detrusor muscle occurs after
spinal cord injury, as well as detrusor–sphinc-
ter–dyssynergia. Detrusor-sphincter-dyssynergia is
thought to be the main reason for the observed
increased voiding pressures and decreased voiding
efficiency, and has been documented in rats (Kruse
et al., 1993; Pikov et al., 1998; Yoshiyama et al.,
2000; Cheng and de Groat, 2004). Although ex-
ternal urethral sphincter bursting activity can be
mediated by spinal reflex mechanisms, the bursting
activity that is seen after spinal cord injury is ab-
normal, leading to shorter urethral opening times
and this presumably contributes to the inefficient
voiding and increased voiding pressures (Cheng
and de Groat, 2004). Also, the amplitude from
external urethral sphincter electromyographic re-
cordings has been shown to increase following
spinal cord injury (Pikov and Wrathall, 2001).
Suppression of external urethral sphincter activity
results in improvement of all voiding parameters
(Kruse et al., 1993; Yoshiyama et al., 2000). In
addition to dyssynergia, detrusor hyperreflexia has
been well characterized in rats following spinal
cord injury (Mitsui et al., 2003) and is thought
to be due to a lack of supraspinal inhibition
with or without an increase of afferent signaling.
Moreover, enlargement of the bladder may result
in plasticity of afferents, further contributing to
this hyperreflexia (De Groat et al., 1998). The
significance of this syndrome in the recovery
367
of micturition, however, is unknown. Recovery
of micturition function has been observed to oc-
cur through amelioration of detrusor–sphinc-
ter–dyssynergia without noticeable change of de-
trusor hyperreflexia (Mitsui et al., 2003).
We suggest that in addition to these muscles,
other pelvic floor muscles, and specifically the
bulbospongiosus musculature develops hype-
rreflexia and spasticity following spinal cord inju-
ry. This change in reflex activity is reminiscent
of hyperreflexia observed in other segmentally
mediated reflexes (Bose et al., 2002). Interestingly,
McKenna et al. (1991) observed simultaneous
contraction of all pelvic floor musculature (bulb-
ospongiosus, ischiocavernosus, external anal
sphincter, and external urethral sphincter) during
the urethrogenital reflex after spinal cord transec-
tion. This reflex is induced by stimulation of the
urethra and these authors suggest that the motor
program is controlled by a spinal pattern genera-
tor in the absence of descending input and that the
spinal circuitry is responsible for the rhythmic,
coordinated bursting activity of all these muscles.
Certainly, the extensive co-mingling of the dend-
rites of these motor neurons in the spinal cord
(as well as their electrotonic and synaptic cou-
pling) could be an anatomical substrate for this
and other coordinated activities (Coolen et al.,
2004; Giuliano and Rampin, 2004).
A closer examination of the bulbospongiosus
muscle activity after contusion injury of the spinal
cord (Nout et al., 2005) in awake behaving rats,
shows that the loss of descending input produces
an exaggerated contraction pattern in this muscle
similar to that seen in the external anal sphincter.
This activity was observed during erectile events as
well as during micturition. Using pressure record-
ings from the corpus spongiosum penis it is pos-
sible to reliably assess both micturition and erectile
events (Schmidt et al., 2004, 2005). Following
spinal cord damage, increased corpus spongiosum
penis mean pressures and increased duration
of micturition as well as of erectile events occur
(Figs. 3 and 4). Increased corpus spongiosum penis
pressures indicate bulbospongiosus muscle spasti-
city, since pressure within the bulb of the corpus
spongiosum penis is directly dependent on the
activity in the surrounding bulbospongiosus
368

Fig. 3. Pressures recorded from the corpus spongiosum penis in

conscious awake rats during micturition. A moderate spinal
cord contusion [12.5 g cm weight drop Multicenter Animal
Spinal Cord Injury Study (MASCIS) injury] results in a signif-
icant trend toward the development of slightly higher pressures
Fig. 4. Pressures recorded from the corpus spongiosum penis in
and prolonged duration of these events. This is likely due to
conscious awake rats during erectile events. A moderate spinal
development of bulbospongiosus muscle spasticity (time scale: a
cord contusion (12.5 g cm MASCIS injury) results in the de-
25 s interval is shown). (Adapted from Nout et al., 2005.)
velopment of slightly higher pressures and prolonged duration
of the peaks occurring during these events. This is likely due to
development of bulbospongiosus muscle spasticity (time scale: a
musculature (Schmidt et al., 1995). In addition to
25 sc interval is shown). (Adapted from Nout et al., 2005.)
changes in corpus spongiosum penis pressures, the
total number of erectile events per 24-h period was
decreased after injury. We hypothesize that the erections. In addition, during ex copula reflex test-
number of partial erectile events decreased due to ing in these same preparations, there was a short-
disinhibition of supraspinal input allowing almost ened latency to induction of erections consistent
all initiated erectile events to develop into full with previous reports (Hart, 1968; Schmidt et al.,

1999; Hubscher and Johnson, 2000; Holmes et al.,
2001, 2005).
One of the possible explanations of our finding
of a reduced number of erectile events following
spinal cord injury is the assumption that erections
occur secondary to genital grooming, which is re-
duced markedly following spinal cord injury. This
could then account for a reduction of the number
of erectile events following spinal cord injury. Also,
ascending sensory pathways from the male genita-
lia, that are bilaterally located within the dorsal
quadrant at the midthoracic level of the spinal
cord, are damaged in this model of spinal cord in-
jury (Hubscher and Johnson, 1999). Another pos-
sible explanation for a reduction of the number of
erectile events following spinal cord injury, would
be through disruption of descending excitatory
mechanisms in addition to disruption of inhibitory
tracts. Although most studies have concluded that
the descending tracts modulating sexual reflexes
are inhibitory in nature, recent studies have dem-
onstrated the significance of descending excitatory
tracts particularly from the paraventricular nucleus
of the hypothalamus (Giuliano and Rampin, 2000).
In addition to loss of descending inhibitory con-
trol of spinal reflex pathways, plasticity within the
spinal cord is thought to contribute to recovery
but also to development of exaggerated spinal re-
flex responses. For example, Weaver and her col-
leagues have shown that sprouting of dorsal root
afferents containing calcitonin gene-related pep-
tide may contribute to the development of auto-
nomic dysreflexia after severe spinal cord injury
(Weaver et al., 2001). Similarly, calcitonin gene-
related peptide fibers contributing to the parasym-
pathetic components of bladder, bowel, and sexual
reflex function have also been demonstrated to
sprout after spinal cord injury in cats (Beattie
et al., 2000). Spinal cord injury, in this species,
produces a chronic (measured at 6 weeks) dener-
vation of identified parasympathetic preganglionic
neurons (Beattie et al., 1993); the proportion of
the somatic and proximal dendritic membrane of
these efferent neurons contacted by synaptic ter-
minals is significantly reduced, the size of the ter-
minals is smaller, and glial coverage is reciprocally
increased. In additional studies, an increase in the
proportion of the membrane area contacted by
369
calcitonin gene-related peptide immunoreactive
terminals appears to increase (Beattie et al.,
2000), suggesting that the effect of such input
might be even more functionally significant after
spinal cord injury. Interestingly, the synaptic in-
puts to Onuf’s nucleus in the same studies showed
similar synaptic rearrangements, i.e., reduced size
of terminals apposed to Onuf’s motor neurons af-
ter spinal cord injury (Beattie et al., 1993), and an
increase in their terminal coverage by GABA-
immunoreactive terminals (Beattie et al., 2000).
The plasticity that is seen in this area of the cord
forms an important target for potential therapeu-
tic interference. For example, it has been shown
that interference in the development of the prima-
ry afferent plasticity affected the development of
autonomic dysreflexia (Weaver et al., 2001) and
detrusor–sphincter–dyssynergia (Seki et al., 2004).
Recently, Cameron et al. (2004) presented a strat-
egy for modulation of post-traumatic spinal plas-
ticity in both sacral afferents and propriospinal
projection neurons via targeted gene therapy in a
model of autonomic dysreflexia.
Conclusions
Defecation, urination, and male sexual function are
compromised after spinal cord injury. The resulting
dysfunction represents a combination of recovery
of reflexes and plasticity that produce hyperactive
reflexes. Treatments aimed at ameliorating second-
ary injury and enhancing sprouting or regeneration
may have positive effects on lumbosacral auto-
nomic function, but more information regarding
the normal and pathophysiological eliminative and
sexual functions is needed to plan treatment strat-
egies. The strategies should include pharmacolog-
ical therapies guided by experimental studies. The
availability of several useful models of spinal cord
injury in rodents that mimic many of the auto-
nomic functions of human spinal cord injury
should encourage an increase in the laboratory
work devoted to this relatively neglected area of
experimental spinal cord injury. This should be
especially true since spinal cord-injured people
consider recovery of these functions to be of high-
est priority (Anderson, 2004).
370
Acknowledgments
We would like to thank Dr. Gregory M. Holmes
for his helpful comments on this manuscript, and
Mr. John Komon for preparing the figures. Sup-
ported by NIH grant NS-31193.
References
Anderson, K.D. (2004) Targeting recovery: priorities of the
spinal cord injured population. J. Neurotrauma, 21:
1371–1383.
Appel, N.M., Wessendorf, W. and Elde, R.P. (1987) Thryrotro-
pin-releasing hormone in spinal cord: coexistence with se-
rotonin and with substance P in fibers and terminals
apposing identified preganglionic sympathetic neurons. Brain
Res., 415: 137–143.
Arvidsson, U., Cullheim, S., Ulfhake, B., Bennett, G.W., Fone,
K.C., Cuello, A.C., Verhofstad, A.A.J., Visser, T.J. and
Hokfelt, T. (1990) 5-Hydroxytryptamine, substance P, and
thyrotropin releasing hormone in the adult cat spinal cord
segment L7: immunohistochemical and chemical studies.
Synapse, 6: 237–270.
Baron, R. and Janig, W. (1991) Afferent and sympathetic neu-
rons projecting into lumbar visceral nerves of the male rat. J.
Comp. Neurol., 314: 429–436.
Basu, S., Aballa, T.C., Ferrell, S.M., Lynne, C.M. and Brack-
ett, N.L. (2004) Inflammatory cytokine concentrations are
elevated in seminal plasma of men with spinal cord injuries. J.
Androl., 25: 250–254.
Beattie, M.S., Holmes, G.M., Hermann, G.E., Rogers, R.C. and
Bresnahan, J.C. (1996) Pudendal motoneuron function in in-
tact and CNS injured rats: alterations following brainstem le-
sions and spinal transection. Soc. Neurosci. Abstr., 22: 722.4.
Beattie, M.S., Leedy, M.G. and Bresnahan, J.C. (1993) Evi-
dence for alterations of synaptic inputs to sacral spinal reflex
circuits after spinal cord transection in the rat. Exp. Neurol.,
123: 35–50.
Beattie, M.S., Li, Q. and Bresnahan, J.C. (2000) Cell death and
plasticity after experimental spinal cord injury. In: Seil F.J.
(Ed.) Progress in Brain Research, Vol. 128. Elsevier, Am-
sterdam, pp. 9–21.
Beattie, M.S., Li, Q., Leedy, M.G. and Bresnahan, J.C. (1990)
Motoneurons innervating the external anal and urethral
sphincters of the female cat have different patterns of den-
dritic arborization. Neurosci. Lett., 111: 69–74.
Bose, P., Parmer, R. and Thompson, F.J. (2002) Velocity-
dependent ankle torque in rats after contusion injury of the
midthoracic spinal cord: time course. J. Neurotrauma, 19:
1231–1249.
Breedlove, S.M. and Arnold, A.P. (1980) Hormone accumula-
tion in a sexually dimorphic motor nucleus of the rat spinal
cord. Science, 210: 564–566.
Bremer, R.E., Barber, M.D., Coates, K.W., Dolber, P.C. and
Thor, K.B. (2003) Innervation of the levator ani and co-
ccygeus muscles of the female rat. Anat. Rec., 275: 1031–1041.
Cameron, A.A., Smith, G.M., Randall, D.C. and Rabchevsky,
A.G. (2004) Genetic manipulation of afferent fiber sprouting
following spinal cord injury modulates the severity of auto-
nomic dysreflexia. Soc. Neurosci. Abstr., Program No. 459.15.
Cheng, C.L. and de Groat, W.C. (2004) The role of capsaicin-
sensitive afferent fibers in the lower urinary tract dysfunction
induced by chronic spinal cord injury in rats. Exp. Neurol.,
187: 445–454.
Collins, W.F.I., Alspaugh, W.H. and Erichsen, J.T. (1992)
Morphology of external urethral sphincter motoneurons and
localization of putative sphincter interneurons in the rat spi-
nal cord. Soc. Neurosci. Abstr., 18: 1050.
Comarr, A.E. (1970) Sexual function among patients with spi-
nal cord injury. Urol. Int., 25: 134–168.
Coolen, L.M., Allard, J., Truitt, W.A. and McKenna, K.E. (2004)
Central regulation of ejaculation. Physiol. Beh., 83: 203–215.
Cosman, B.C., Stone, J.M. and Perkash, I. (1991) Gastrointes-
tinal complications of chronic spinal cord injury. J. Am.
Paraplegia Soc., 14: 175–181.
De Groat, W.C., Araki, I., Vizzard, M.A., Yoshiyama, M.,
Yoshimura, M., Sugaya, K., Tai, C. and Roppolo, J.R.
(1998) Developmental and injury induced plasticity in the
micturition reflex pathway. Behav. Brain Res., 92: 127–140.
De Groat, W.C., Kawatani, M., Hisamitsu, T., Cheng, C.L.,
Ma, C.P., Thor, K., Steers, W. and Roppolo, J.R. (1990)
Mechanisms underlying the recovery of urinary bladder
function following spinal cord injury. J. Autonom. Nerv.
Sys., 30(Suppl): S71–S77.
Ding, Y.Q., Takada, M., Tokuno, H. and Mizuno, N. (1995)
Direct projections from the dorsolateral pontine tegmentum
to pudendal motoneurons innervating the external urethral
sphincter muscle in the rat. J. Comp. Neurol., 357: 318–330.
Ditunno, J.F., Little, J.W., Tessler, A. and Burns, A.S. (2004)
Spinal shock revisited: a four-phase model. Spinal Cord, 42:
383–395.
Dubrovsky, B., Martinez-Gomez, M. and Pacheco, P. (1985) Spi-
nal control of pelvic floor muscles. Exp. Neurol., 88: 277–287.
Frenckner, B. (1975) Function of the anal sphincters in spinal
man. Gut, 16: 638–644.
Giuliano, F. and Rampin, O. (2000) Central neural regulation
of penile erection. Neurosci. Biobehav. Rev., 24: 517–533.
Giuliano, F. and Rampin, O. (2004) Neural control of erection.
Physiol. Behav., 83: 189–201.
Gonella, J., Bouvier, M. and Blanquet, F. (1987) Extrinsic
nervous control of motility of small and large intestined and
related sphicters. Physiol. Rev., 67: 902–961.
Guyton, A.C. and Hall, J.E. (2000) Textbook of Medical Phys-
iology (10th ed.). W.B. Saunders Co., Philadelphia, PA.
Hart, B.L. (1968) Sexual reflexes and mating behavior in the
male rat. J. Comp. Physiol. Psychol., 65: 453–460.
Hermann, G.E., Bresnahan, J.C., Holmes, G.M., Rogers, R.C.
and Beattie, M.S. (1998) Descending projections from the
nucleus raphe obscurus to pudendal motoneurons in the male
rat. J. Comp. Neurol., 397: 458–474.

Hermann, G.E., Holmes, G.M., Rogers, R.C., Beattie, M.S.
and Bresnahan, J.C. (2003) Descending spinal projections
from the rostral gigantocellular reticular nuclei complex. J.
Comp. Neurol., 455: 210–221.
Higgins Jr, G.E. (1979) Sexual responses in spinal cord injured
adults: a review of the literature. Arch. Sex. Behav., 8: 173–196.
Hirsch, M.D. and Helke, C.J. (1988) Bulbospinal thryrotrophin
releasing hormone projections to the intermediolateral cell
column: a double fluorescence immunohistochemical retro-
grade tracing study in the rat. Neuroscience, 25: 625–637.
Holmes, G.M., Bresnahan, J.C., Rogers, R.C., Hermann, G.E.
and Beattie, M.S. (1994) Differential characteristics of rat
dorsomedial motoneurons. Soc. Neurosci. Abstr., 20: 791.
Holmes, G.M., Hermann, G.E., Rogers, R.C., Bresnahan, J.C.
and Beattie, M.S. (2002) Dissociation of the effects of nucleus
raphe obscurus or rostral ventrolateral medulla lesions
on eliminatory and sexual reflexes. Physiol. Behav., 75: 49–55.
Holmes, G.M., Martau, J.M., Hermann, G.E., Rogers, R.C.,
Bresnahan, J.C. and Beattie, M.S. (1997a) Nucleus raphe
obscurus (nRO) regulation of anorectal motility in rats. Brain
Res., 759: 197–204.
Holmes, G.M., Rogers, R.C., Bresnahan, J.C. and Beattie,
M.S. (1995) Thyrotropin-releasing hormone (TRH) and CNS
regulation of anorectal motility in the rat. J. Auton. Nerv.
Syst., 56: 8–14.
Holmes, G.M., Rogers, R.C., Bresnahan, J.C. and Beattie,
M.S. (1997b) Differential effects of intrathecal thyrotropin-
releasing hormone (TRH) on perineal reflexes in male rats.
Physiol. Behav., 61: 57–63.
Holmes, G.M., Rogers, R.C., Bresnahan, J.C. and Beattie,
M.S. (1998) External anal sphincter hyperreflexia following
spinal transection in the rat. J. Neurotrauma, 15: 451–457.
Holmes, G.M., Stephens, R.L., Bresnahan, J.C. and Beattie, M.S.
(2001) Intrathecal methiothepin and thyrotropin-releasing hor-
mone effects on penile erection. Physiol. Behav., 73: 59–64.
Holmes, G.M., Van Meter, M.T., Beattie, M.S. and Bresnahan,
J.C. (2005) Serotonergic fiber sprouting to external anal
sphincter motoneurons after spinal cord contusion. Exp. Ne-
urol., 193: 29–42.
Holstege, G. and Tan, J. (1987) Supraspinal control of moto-
neurons innervating the striated muscles of the pelvic floor
including urethral and anal sphincters in the cat. Brain,
110(Part 5): 1323–1344.
Hubscher, C.H. and Johnson, R.D. (1999) Effects of acute and
chronic midthoracic spinal cord injury on neural circuits for
male sexual function. I. Ascending pathways. J. Neuro-
physiol., 82: 1381–1389.
Hubscher, C.H. and Johnson, R.D. (2000) Effects of acute and
chronic midthoracic spinal cord injury on neural circuits for
male sexual function. II. Descending pathways. J. Neuro-
physiol., 83: 2508–2518.
Hulsebosch, C.E. and Coggeshall, R.E. (1982) An analysis of
the axon populations in the nerves to the pelvic viscera in the
rat. J. Comp. Neurol., 211: 1–10.
Jänig, W. and McLachlan, E.M. (1987) Organization of lumbar
spinal outflow to distal colon and pelvic organs. Physiol.
Rev., 67: 1332–1404.
371
Katagiri, T., Gibson, S.J., Su, H.C. and Polak, J.M. (1986) Com-
position and central projections of the pudendal nerve in the rat
investigated by combined peptide immunocytochemistry and
retrograde fluorescent labelling. Brain Res., 372: 313–322.
Krier, J. (1985) Discharge patterns of pudendal efferent fibers
innervating the external anal sphincter of the cat. J. Physiol.,
368: 471–480.
Kruse, M.N., Belton, A.L. and De Groat, W.C. (1993) Changes
in bladder and external urethral sphincter function after spi-
nal cord injury in the rat. Am. J. Physiol., 264: R1157–R1163.
Li, Q., Beattie, M.S. and Bresnahan, J.C. (1995) Onuf’s nucleus
(ON) motoneurons (MNs) have more GABA-ergic synapses
than other somatic MNs: a quantitative immunocytochem-
ical study in the cat. Soc. Neurosci. Abstr., 21: 1201.
Maggi, C.A., Giuliani, S., Santicioli, P., Patacchini, R. and Meli,
A. (1988) Neural pathways and pharmacological modulation
of defecation reflex in rats. Gen. Pharmacol., 19: 517–523.
Marson, L., List, M.S. and McKenna, K.E. (1992) Lesions of
the nucleus paragigantocellularis alter ex copula penile re-
flexes. Brain Res., 592: 187–192.
Marson, L. and McKenna, K.E. (1990) The identification of a
brainstem site controlling spinal sexual reflexes in male rats.
Brain Res., 515: 303–308.
Marson, L. and McKenna, K.E. (1996) CNS cell groups in-
volved in the control of the ischiocavernosus and bulb-
ospongiosus muscles: a transneuronal tracing study using
pseudorabies virus. J. Comp. Neurol., 374: 161–179.
Matsumoto, A., Arnold, A.P., Zampighi, G.A. and Micevych,
P.E. (1988) Androgenic regulation of gap junctions between
motoneurons in the male rat. J. Neurosci., 8: 4177–4183.
McCann, M.J., Hermann, G.E. and Rogers, R.C. (1989) Nu-
cleus raphe obscurus (nRO) influences vagal control of gas-
tric motility in rats. Brain Res., 486: 181–184.
McKenna, K.E. and Nadelhaft, I. (1986) The organization of
the pudendal nerve in the male and female rat. J. Comp.
Neurol., 248: 532–549.
McKenna, K.E., Sookja, K.C. and McVary, K.T. (1991) A
model for the study of sexual function in anesthetized male
and female rats. Am. J. Physiol., 261(Regul. Integr. Comp.
Physiol. 30): R1276–R1285.
Mitsui, T., Kakizaki, H., Tanaka, H., Shibata, T., Matsuoka, I.
and Koyanagi, T. (2003) Immortalized neural stem cells
transplanted into the injured spinal cord promote recovery of
voiding function in the rat. J. Urol., 170: 1421–1425.
Miyata, K., Kamato, T., Nishida, A., Ito, H., Yuki, H.,
Yamano, M., Tsutsumi, R., Katsuyama, Y. and Honda, K.
(1992) Role of the serotonin3 receptor in stress-induced def-
ecation. J. Pharmacol. Exp. Ther., 261: 297–303.
Monaghan, E.P. and Breedlove, S.M. (1991) Brain sites pro-
jecting to the spinal nucleus of the bulbocavernosus. J.
Comp. Neurol., 307: 370–374.
Nadelhaft, I. and Booth, A.M. (1984) The location and mor-
phology of preganglionic neurons and the distribution of
visceral afferents from the rat pelvic nerve: a horseradish
peroxidase study. J. Comp. Neurol., 226: 238–245.
Nadelhaft, I. and Vera, P.L. (2001) Separate urinary bladder
and external urethral sphincter neurons in the central nervous
372
system of the rat: simultaneous labeling with wo immunohisto-
chemically distinguishable pseudorabies viruses. Brain Res.,
903: 33–44.
Nout, Y.S., Schmidt, M.H., Tovar, C.A., Culp, E., Beattie,
M.S. and Bresnahan, J.C. (2005) Telemetric monitoring of
corpus spongiosum penis pressures in conscious rats for as-
sessment of micturition and sexual function following spinal
cord contusion injury. J. Neurotrauma, 22(4): 429–444.
Pacheco, P., Camacho, M.A., Garcia, L.I., Hernandez, M.E.,
Carrillo, P. and Manzo, J. (1997) Electrophysiological evi-
dence for the nomenclature of the pudendal nerve and sacral
plexus in the male rat. Brain Res., 763: 202–208.
Paxinos, G. (1995) The Rat Nervous System (2nd ed.). Aca-
demic Press Inc., San Diego, CA.
Pedersen, E. (1983) Regulation of bladder and colon-rectum in
patients with spinal lesions. J. Auton. Nerv. Syst., 7: 329–338.
Peshori, K.R., Erichsen, J.T. and Collins 3rd, W.F. (1995) Dif-
ferences in the connectivity of rat pudendal motor nuclei as
revealed by retrograde transneuronal transport of wheat
germ agglutinin. J. Comp. Neurol., 353: 119–128.
Pikov, V., Gillis, R.A., Jasmin, L. and Wrathall, J.R. (1998)
Assessment of lower urinary tract functional deficit in rats
with contusive spinal cord injury. J. Neurotrauma, 15:
375–386.
Pikov, V. and Wrathall, J.R. (2001) Coordination of the blad-
der detrusor and the external urethral sphincter in a rat
model of spinal cord injury: effect of injury severity. J. Ne-
urosci., 21: 559–569.
Purinton, P.T., Fletcher, T.F. and Bradley, W.E. (1973) Gross
and light microscopic features of the pelvic plexus in the rat.
Anat. Rec., 175: 697–705.
Ramirez-Leon, V. and Ulfake, B. (1993) GABA-like immuno-
reactive innervation and dendro-dendritic eontacts in the
ventrolateral dendritic bundle in the cat S1 spinal segment:
and electron microscopic study. Exp. Brain Res., 97: 1–12.
Ramirez-Leon, V., Ulfake, B., Arvidsson, U., Verhofstad,
A.A., Visser, T.J. and Hokfelt, T. (1994) Serotonergic,
peptidergic, and GABAergic innervation of the ventrolater-
al and dorsolateral motor nuclei in the cat S1/S2 segments:
an immunofluorescence study. J. Chem. Neuroanat., 7:
87–103.
Rexed, B. (1954) A cytoarchitectonic atlas of the spinal cord in
the cat. J. Comp. Neurol., 100: 297–379.
Roney, K.J., Scheibel, A.B. and Shaw, G.L. (1979) Dendritic
bundles: survey of anatomical experiments and physiological
theories. Brain Res. Rev., 1: 225–271.
Sachs, B.D. and Leipheimer, R.E. (1988) Rapid effect of testos-
terone on striated muscle activity in rats. Neuroendocrino-
logy, 48: 453–458.
Sasaki, M. (1994) Morphological analysis of external urethral
and external anal sphincter motoneurones of cat. J. Comp.
Neurol., 349: 269–287.
Schmidt, M.H., Nout, Y.S., Culp, E., Tovar, C.A., Beattie,
M.S. and Bresnahan, J.C. (2004) A new technique for the
simultaneous recording of micturition and penile erections in
freely behaving rats: telemetric monitoring of corpus
spongiosum penis (CSP) pressure. Society for Neuroscience,
Online., Program No. 922.16. Abstract Viewer/Itinerary
Planner. Washington, DC.
Schmidt, M.H., Nout, Y.S., Culp, E., Tovar, C.A., Beattie,
M.S. and Bresnahan, J.C. (2005) A new technique for the
simultaneous recording of micturition and penile erections in
freely behaving rats: telemetric monitoring of corpus
spongiosum penis (CSP) pressure, in preparation.
Schmidt, M.H., Sakai, K., Valatx, J.L. and Jouvet, M. (1999)
The effects of spinal or mesencephalic transections on sleep-
related erections and ex-copula penile reflexes in the rat.
Sleep, 22: 409–418.
Schmidt, M.H., Valatx, J.L., Sakai, K., Debilly, G. and Jouvet,
M. (1995) Corpus spongiosum penis pressure and perineal
muscle activity during reflexive erections in the rat. Am. J.
Physiol., 269: R904–R913.
Schroder, H.D. (1980) Organization of the motor neurons in-
nervating the pelvic muscles of the male rat. J. Comp. Ne-
urol., 192: 567–587.
Schuster, M.M. (1968) Motor action of rectum and anal
sphincters in continence and defecation. In: Code C.F. (Ed.)
Handbook of Physiology, Alimentary Canal Motility, Sec-
tion 6, Vol. IV. Chapter 103. American Physiological Society,
Washington, DC, pp. 2121–2146.
Seki, S., Sasaki, K., Igawa, Y., Nishizawa, O., Chancellor,
M.B., DeGroat, W.C. and Yoshimura, N. (2004) Suppres-
sion of detrusor-sphincter dyssynergia by immunoneutrali-
zation of nerve growth factor in lumbosacral spinal cord in
spinal cord injured rats. J. Urol., 171: 478–482.
Shen, P., Arnold, A.P. and Micevych, P.E. (1990) Supraspinal
projections to the ventromedial lumbar spinal cord in adult
male rats. J. Comp. Neurol., 300: 263–272.
Tang, Y., Rampin, O., Giuliano, F. and Ugolini, G. (1999)
Spinal and brain circuits to motoneurons of the bulb-
ospongiosus muscle: retrograde transneuronal tracing with
rabies virus. J. Comp. Neurol., 414: 167–192.
Vera, P.L. and Nadelhaft, I. (1992) Afferent and sympathetic
innervation of the dome and the base of the urinary bladder
of the female rat. Brain Res. Bull., 29: 651–658.
Vizzard, M.A., Brisson, M. and De Groat, W.C. (2000)
Transneuronal labeling of neurons in the adult rat central
nervous system following inoculation of pseudorabies virus
into the colon. Cell Tissue Res., 299: 9–26.
Weaver, L.C., Verghese, P., Bruce, J.C., Fehlings, M.G., Krenz,
N.R. and Marsh, C.R. (2001) Autonomic dysreflexia and
primary afferent sprouting after clip-compression injury of
the rats spinal cord. J. Neurotrauma, 18: 1107–1119.
Yamanouchi, K. and Kakeyama, M. (1992) Effect of medullary
raphe lesions on sexual behavior in male rats with or without
treatments of p-chlorophenylalanine. Physiol. Behav., 51:
575–579.
Yoshiyama, M., De Groat, W.C. and Fraser, M.O. (2000) In-
fluences of external urethral sphincter relaxation induced by
alpha-bungarotoxin, a neuromuscular junction blocking
agent, on voiding dysfunction in the rat with spinal cord
injury. Urology, 55: 956–960.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 25

Upper and lower gastrointestinal motor and sensory

dysfunction after human spinal cord injury

Paul Enck1,, Irmgard Greving2, Sibylle Klosterhalfen3 and Beate Wietek4

1
Department of Psychosomatic Medicine, University Hospitals Tuingen, Schaffhausenstr 113, 72072 Tubingen, Germany
2
Department of Internal Medicine, Elisabeth Hospital, Gelsenkirchen, Germany
3
Institute of Medical Psychology, University Hospitals Dusseldorf, Germany
4
Department of Radiology, University Hospitals Tubingen, Germany

Abstract: This chapter describes the results of investigations of the upper and lower gastrointestinal tract in
subjects with complete and incomplete spinal cord injury. In one study, gastric emptying was investigated
and found delayed. The delay was tentatively attributed to a colo-gastric inhibitory reflex triggered by
inappropriate colonic emptying. In another study, anorectal motor and sensory functions were measured.
Decreased tone of the internal anal sphincter, exaggerated recto-anal reflexes following rectal distension
and spontaneous high-amplitude rectal contractions at low distension volumes were among the findings of
the study. Some of the subjects, classified as having a complete injury according to usual clinical criteria
(American Spinal Injury Association, ASIA), reported sensation of distension of the rectum. This raises the
issue of the need for better methods for the clinical assessment of sensory transmission in the spinal cord.
Promising results obtained with functional magnetic resonance imaging of the brain during rectal stim-
ulation in a small group of paraplegics, with complete injuries by ASIA criteria, showed evidence of
activation of several brain regions.

Introduction innervation is unknown. The purpose of this

People with traumatic spinal cord lesions often
suffer from persistent upper and lower gastroin-
testinal complaints such as dyspepsia, nausea,
slowed defecation, reflex and uncontrolled rectal
evacuation, and chronic abdominal pain (Glickman
and Kamm, 1996; Kirk et al., 1997; Menter et al.,
1997; De Looze et al., 1998a, b; Han et al., 1998).
Symptoms tend to increase with time and peak at
around 5 years post-trauma (Stone et al., 1990),
but the reasons for these long-term changes are
unclear. The effect of disconnection of the spinal
cord from supraspinal centers on autonomic (para-
sympathetic/sympathetic) and intrinsic (enteric)
Corresponding author. Tel.: +49 7071 9387374;
Fax.: +49 7071 9387379; E-mail: paul.enck@uni-tuebingen.de
chapter is to review data from functional investi-
gations of the upper and lower gastrointestinal
tract in spinal cord-injured people with complete
or incomplete injury. Both motor and sensory
dysfunctions are addressed.
Gastric emptying
People with incomplete or complete spinal cord
injury often complain of upper gastrointestinal
symptoms such as heartburn, bloating, and post-
prandial fullness (De Vault et al., 1996; Kao et al.,
1999; Singh and Triadafilopoulos, 2000; Chen
et al., 2004), but these symptoms have rarely been
investigated in a systematic fashion. Clinically,
they are often regarded as secondary to immobility

DOI: 10.1016/S0079-6123(05)52025-9 373

374
and bed rest, to inadequate eating habits, or as a
consequence of inappropriate colonic emptying
and constipation, i.e. secondary to a colo-gastric
inhibitory reflex (Youle and Read, 1984). While
this may be the case with lower spinal cord inju-
ries, a higher cord lesion may also directly affect
visceral motor functions such as gastric emptying,
via reflex spinal pathways.
The following study of gastric emptying was done
using 15 subjects with complete injury of the spinal
cord (according to ASIA criteria) (Ditunno et al.,
1994), 5 of whom had lesions in the cervical (C)
segments C1 to C7 and 10 who had lesions in the
thoracic (TH) and lumbar (L) regions (TH1 to L5).
The ratio of males to females investigated was 14:1.
Studies were undertaken between 3 months and 28
years following trauma. All subjects completed a
questionnaire related to gastrointestinal symptoms.
Gastric emptying was measured by means of the 13C
(carbon 13) breath test: in short, a standardized
mixed test meal labeled with [13C]octanoic acid for
measurement of emptying of solids (Zahn et al.,
2003), and a standard test drink, labeled with
[13C]sodium acetate for emptying of liquids (Braden
et al., 1995), were provided on two separate days in
randomized sequence. Breath samples were collect-
ed at baseline, and every 10 min over 2 h for liquid
emptying and every 15 min for 4 h for emptying of
solids, and analyzed by infrared spectrometry (IRIS,
Wagner Analysentechnik, Bremen, Germany) for
the 12C:13C ratio. The time series of each data set
permits the relevant parameters of gastric emptying
to be calculated, i.e. half-time emptying (T/2) and
the lag-phase (t-lag) that measures the time between
the end of the meal and the first bolus leaving the
stomach (Chew et al., 2003).
All subjects reported relevant upper gastrointes-
tinal symptoms such as heartburn, post-prandial
fullness, nausea, and bloating. Compared to
the normal values of the test (T/2o75 min,
t-lago50 min for solid emptying; T/2o60 min,
t-lago20 min for liquid emptying), subjects with
high spinal cord lesions had significantly delayed
gastric emptying for solids (T/2 ¼ 313 min,
t-lag ¼ 167 min) but not for liquids, whereas sub-
jects with lower spinal cord lesions had significant
delay of both solid (T/2 ¼ 153, t-lag ¼ 109) and
liquid emptying (T/2 ¼ 58, t-lag ¼ 36).
Delayed gastric emptying seems to be quite
common in people with complete spinal cord in-
jury and may explain the occurrence of upper gas-
trointestinal symptoms in these people in everyday
life (Fealey et al., 1984). However, the data are
conflicting, since some authors found no delay in
gastric emptying in people with spinal cord injury
or concluded that, if a delay occurs early after spi-
nal trauma, it diminishes over time (Mollen
et al., 1999). It may be argued that the delays are
secondary to immobility and bed rest, or to inad-
equate eating habits; more likely, however, they are
a consequence of inappropriate colonic emptying
and constipation, i.e. are secondary to a colo-gas-
tric inhibitory reflex. Delayed gastric emptying has
been shown to occur following experimental rectal
balloon distension in healthy volunteers (Youle
and Read, 1984; Mollen et al., 1999), and also in
patients with chronic constipation (Coremans
et al., 2004). Final proof of direct or indirect
effects of spinal cord injury on upper gastrointes-
tinal transit remains to be obtained in the future.
Anorectal motor and sensory function
People with complete or incomplete spinal cord
lesions due to trauma often also suffer from def-
ecation disorders (Longo et al., 1989; Enck et al.,
1991; Bruninga and Camilleri, 1997; De Looze
et al., 1998a). As early as 1948 (Gaston, 1948), this
problem was attributed to loss of motor and/or
sensory control of the anorectum and/or pelvic
floor. In studies of anorectal motor function after
traumatic complete (MacDonagh et al., 1992) or
incomplete (Sun et al., 1990) spinal injury, various
continence mechanisms have been shown to be
disturbed. Patients with spinal tumors (May, 1991)
and other causes of denervation (Wheatley et al.,
1977) also have defecation disorders, resulting in
either constipation or incontinence or both. This
problem may relate to motor and sensory deficits
in the somatic and visceral compartments of the
nervous system. Colonic transit time is usually de-
layed in people with spinal cord injury (Devroede
et al., 1979; Glick et al., 1984; Beuret-Blanquart
et al., 1990; Keshavarzian et al., 1995), but since
transit in upper gastrointestinal compartments of-
ten is also abnormal (Fealey et al., 1984; see also

the preceding section), constipation may be either
of the slow transit-type or of the ‘‘outlet obstruc-
tion’’-type. Incontinence, if present, is mostly at-
tributable to the loss of visceral sensation on the
one hand, and to the loss of voluntary control over
the external anal sphincter on the other (Sun et al.,
1995; De Looze et al., 1998b). In the case of a low
spinal cord lesion, the combination of intact sen-
sation and voluntary control of the external anal
sphincter suggests a lesion of the conus medullaris
or cauda equina (MacDonagh et al., 1992). In
people with high spinal cord lesions, loss of vol-
untary contraction of the external anal sphincter,
blunted or absent rectal sensation and exaggerated
reflex activity of the sphincter, without inconti-
nence, are often seen. In these people, the residual
pressure, i.e. the remaining pressure during sphinc-
ter relaxation induced by rectal distension, is sig-
nificantly higher than in people with a low spinal
cord injury (Read and Sun, 1992). No data are
available on the short-term recovery of anorectal
functions following trauma, except after spinal
surgery (Sun et al., 1995). Furthermore, the rela-
tionship between clinical data and anorectal func-
tions has not been determined.
We conducted studies to establish a set of criteria
that would best predict short-term recovery of
anorectal functions, permitting selection of patients
for therapy programs such as biofeedback training.
Thirty-two subjects with spinal cord injury were
recruited from the Spinal Trauma Unit of the
Department of Surgery at the University Hospital
‘‘Bergmannsheil’’, Bochum, Germany. A mini-
mum of 4 days elapsed between trauma and the
clinical functional investigations to allow recovery
from traumatic inflammation (Table 1). Subjects
underwent assessment of completeness of spinal
injury by clinical signs (ASIA criteria; Ditunno
et al., 1994), of clinical symptoms of incontinence
(urinary, fecal), of constipation or diarrhea, and of
the necessity for medication, compared to pre-
trauma history. Fifteen of the subjects had com-
plete injury and the remaining 17 had incomplete
injury. The subjects had anorectal manometry that
included visceral (rectal) sensitivity testing (see
below) and this testing was repeated after an
average of 64 days (n ¼ 29) and again after 140
days (n ¼ 20). Anorectal manometry (Read and
375
Sun, 1992; Rao et al., 2002) was routinely per-
formed in left-lateral position after a conventional
bowel preparation (saline or sorbid enema). The
following parameters were assessed (Tables 2 and
3) — anal resting pressure, representing the func-
tion of the smooth muscle of the internal anal
sphincter; squeeze pressure, the increment above
resting pressure due to voluntary contraction of
the external anal sphincter; reflex relaxation of the
internal anal sphincter (%) during balloon disten-
sion of the rectum 10 cm above the anal canal
(called the recto-anal inhibitory reflex, RAIR);
reflex contraction of the external anal sphincter
during a sudden abdominal pressure increase
produced by coughing, as an indicator of stress
continence (both anal and abdominal pressures are
presented); and paradoxical contraction of pelvic
floor muscles, especially of the external anal
sphincter, with attempts to strain for defecation
(% occurrence). If spontaneous expulsion (% oc-
currence) of the distending balloon was observed,
the respective balloon volume (ml) was noted. If
the subject reported sensation of balloon disten-
sion, the pressure at which the sensation appeared
was noted (sensation threshold) and the procedure
was repeated at least once to ensure reproducibil-
ity of the sensation. If, as the rectal balloon was
inflated, the subject experienced pain, the pressure
at which this happened was noted (pain threshold).
Data from healthy subjects served as controls for
the manometry measurements.
Most anorectal parameters analyzed in the cord-
injured subjects differed significantly from those of
the control subjects (Table 2). In addition, spon-
taneous defecation of the rectal balloon, never ob-
served in the healthy subjects, was initially present
in nine subjects with complete, and one subject
with incomplete lesions, usually preceded by giant
rectal contractions. These responses were inde-
pendent of the spinal level of the lesions. Signif-
icant changes in anorectal function, as assessed by
manometry, were never observed in the cord-in-
jured subjects investigated two or three times dur-
ing the course of disease (Table 3), and no patient
reported a significant clinical change of symptoms,
either with incontinence or with constipation.
People with traumatic spinal cord lesions
usually have disturbances of voluntary anorectal
376

Table 1. Patient characteristics and clinical investigations performed

No Init Age Sex C/IC Level Invest Interv Neuro UI FI

1 RV 20 1 IC T5 2 5 Y N N
2 MW 27 1 IC L3 3 33 Y Y Y
3 RS 23 2 IC T12 3 6 N Y Y
4 HDS 41 2 C T11 3 14 N Y Y
5 JS 38 2 IC C6 3 6 N Y Y
6 PP 25 2 IC L2 1 17 N Y Y
7 NM 17 1 C T7 3 8 Y Y Y
8 HJL 48 2 C C4 3 46 N Y Y
9 JH 32 1 IC T12 2 17 Y Y Y
10 EH 65 1 IC C6 2 22 Y Y Y
11 WE 37 2 IC L1 2 67 N Y Y
12 MBO 21 2 C C6 3 25 N Y Y
13 OB 18 2 C C4 2 32 N Y Y
14 RC 31 2 IC T6 1 42 N Y Y
15 RG 53 1 IC L1 3 9 Y Y Y
16 BP 16 2 IC C5 3 23 N Y Y
17 CT 24 2 C T8 3 12 N Y Y
18 WK 58 2 IC T4 3 74 N N N
19 MH 20 2 C C7 3 29 N Y Y
20 MB 33 2 C C4 3 51 N Y Y
21 HI 49 2 IC T11 3 51 N Y Y
22 HJS 58 2 IC C4 2 20 Y Y Y
23 DH 19 2 IC L1 2 22 Y N N
24 AL 40 2 C L3 2 301 Y N N
25 FT 18 2 IC T10 3 38 Y Y Y
26 CW 21 2 C T12 2 4 Y N N
27 MN 19 2 C C6 3 11 Y Y Y
28 JB 21 2 C C4 3 17 Y Y Y
29 IW 47 2 C C6 3 34 N Y Y
30 HE 29 2 C T6 3 26 Y Y Y
31 MM 24 1 IC T7 3 10 Y Y N
32 HN 17 2 C C6 1 13 Y Y Y

Note: C/IC, complete, incomplete; Level, spinal level of lesion; Invest, investigated 1, 2, or 3 times; Interv, interval (days) between trauma and 1st
investigation; Neuro, neurological investigation performed (Yes, No); UI, urinary incontinence (Yes, No); FI, fecal incontinence (Yes, No); Sex,
1 ¼ male, 2 ¼ female.

motor function (MacDonagh et al., 1992), and
these problems originate from various kinds of
dysfunctions (Sun et al., 1990, 1995; MacDonagh
et al., 1992). Our data are in accordance with these
previous reports. People with complete lesions ex-
hibit lower resting anal pressures, i.e. decreased
tone of the internal anal sphincter muscle, some-
times exaggerated recto-anal reflexes following
rectal distension, and some abnormalities not
found in normal controls or in patients with in-
continence of other origins. For example, some
patients have spontaneous high-amplitude rectal
contractions, at rather low distension volumes that
uncontrollably expel the distending balloon from
the rectum. It is of interest and clinical relevance
that anorectal functions assessed early post-trau-
ma, within the first few weeks, do not change
significantly during the subsequent period, irre-
spective of completeness or incompleteness of cord
injury, despite the fact that this time usually in-
cludes intense rehabilitation of the other motor
functions. However, since these programs do not
include pelvic floor training for regaining conti-
nence control, one cannot dismiss the possibility
that patients, at least those with incomplete le-
sions, would benefit from programs such as bio-
feedback training. This hypothesis should be
tested in the future.
 377

Table 2. Manometric study of cord-injured subjects and controls (1st investigation)

Patients Controls Statistics

N 32 15
Age 32714 2674 po0.05
Sex (male/female) 25:7 11:4 n.s.
Resting pressureb 65725 78711 po0.02
Squeeze pressureb 66740 206740 po0.01
RAIRa (%) 75715 73713 n.s.
Sensation thresholdb 53743 (n ¼ 20) 45732 n.s.
Urge to defecate 95774 (n ¼ 19) 94745 n.s.
Pain threshold 154767 (n ¼ 14) 167770 n.s.
Cough reflex (anal)b 60732 (n ¼ 20) 176752 po0.01
Cough reflex (abdom)b 40742 (n ¼ 20) 104732 po0.01
Paradoxical contractiond 18 0 po0.01
Spontaneous defecationd 10 0 po0.01
Volumec 154765 (n ¼ 10)
a
Recto-anal inhibitory reflex. b All pressures are reported in mmHg. c Balloon volume (ml) initiating spontaneous defecation. d Number
of subjects with response.

Table 3. Manometric study of the course of disease

Investigation number

First Second Third

N 32 29 20
Age 32714 33714 31714
Sex (male/female) 25:7 22:7 16:4
Resting pressureb 65725 65721 74727
Squeeze pressureb 66740 76765 67738
RAIRa (%) 75715 (n ¼ 32) 74715 (n ¼ 29) 68711 (n ¼ 20)
Sensation thresholdb 53743 (n ¼ 20) 66742 (n ¼ 20) 57738 (n ¼ 11)
Urge to defecateb 95774 (n ¼ 19) 62732 (n ¼ 16) 82740 (n ¼ 11)
Pain thresholdb 154767 (n ¼ 14) 147778 (n ¼ 13) 190788 (n ¼ 7)
Cough reflex (anal)b 60732 (n ¼ 20) 65740 (n ¼ 18) 59735 (n ¼ 15)
Cough reflex (abdom)b 40742 (n ¼ 20) 43730 (n ¼ 15) 45722 (n ¼ 8)
Paradoxical contractiond 18 13 10
Spontaneous defecationd 10 11 8
Volumec 154765 (n ¼ 10) 156.3786 (n ¼ 11) 180778 (n ¼ 8)
a
Recto-anal inhibitory reflex. b All pressures are reported in mmHg. c Balloon volume (ml) initiating spontaneous defecation. d Number
of subjects with response.

Completeness of spinal injury is assessed clini-
cally by the absence of sensory perception and the
lack of motor control of the lower limbs. Trans-
cranial recordings of somatosensory and motor
cortical potentials, evoked by stimulation of the
pelvic floor or lower limbs, and recordings of the
electrical activity, evoked by transcranial stimula-
tion of the motor cortex, in muscles of the pelvic
floor or lower limb often are used to confirm the
diagnosis (Neill and Swash, 1980; Herdmann et al.,
1991; Enck et al., 1992). Occasionally, people with
complete spinal cord injury report diffuse sensa-
tions from the viscera, e.g., the feeling of rectal
fullness preceding involuntary colonic mass move-
ments and defecation. These reports challenge the
reliability of the ASIA criteria (Ditunno et al.,
378

1994) for clinical assessment of cases of complete

spinal cord injury. However, these reports have
usually been attributed to re-interpretation of
bowel sounds arising from the abdominal cavity.
In the above study of anorectal motor function,
we also assessed anorectal sensory functions by
testing the subject’s sensitivity to rectal balloon
distension. During anorectal manometry, as de-
scribed above, the thresholds for minimal sensa-
tion, urge to defecate and pain following stepwise
rectal balloon distension (ml) were determined. If a
sensation of the distension (which was blinded for
the patients) was noted, it was repeated at least
once to assure reproducibility.
When subjects were grouped according to com-
plete or incomplete spinal cord injury, by clinical
assessment, 5 of 15 cases of complete injury had
some sensory function and 3 of 15 even reported
pain during rectal distension. In one subject, pain
thresholds were identical when retested and in two
subjects, minimal perception thresholds to disten-
sion were almost identical on repetition. Three
subjects (No. 17, 20, and 27, Table 1) who were
tested three times consistently reported perception
of the distension on all three occasions. One of
them had been evaluated neurologically (No. 27).
Conversely, in one subject judged as incomplete,
no remaining motor or sensory function could be
detected by anorectal manometry (No. 3, Table 1).
Patients with remaining visceral sensation had re-
producible perception thresholds and experienced
pain with rectal distension volumes up to 200 ml.
Persistence of visceral sensation, despite the clas- Fig. 1. Level of injury in subjects with complete (left) and in-
sification of ‘‘complete’’ injury, occurred with complete (right) spinal injuries. Numbers refer to the individ-
lesions at all spinal levels (Fig. 1). uals in Table 1. Bold (outer) numbers refer to those subjects
investigated neurophysiologically. See Table 1 for clinical data
As described above, 5 of 15 cases diagnosed as
and Table 3 for cases with residual sensation. Reproduced by
complete spinal cord injury by clinical assessment permission of Blackwell from Greving et al., 1998.
had some anorectal sensory function. Because of
this discrepancy between the anorectal manometry
data and the clinical assessment, electrophysiolog- area of the cortex. In the same two groups, con-
ical techniques were used to provide another indi- centric needle recording electrodes were placed
cation of completeness of injury. In 15 subjects in the external anal sphincter (Herdmann et al.,
diagnosed as complete injury and in 17 others di- 1991) while transcranial stimulation of the motor
agnosed as having incomplete injury by ASIA cri- cortex was performed with a magneto-electric coil.
teria (Ditunno et al., 1994), electrical stimulation The subjects of the incomplete group showed
of the anal mucosa (Enck et al., 1992) was per- sensory- and motor-evoked potentials with late-
formed while recording electrodes were placed bi- ncies of 53.8723.5 and 27.875.1 ms, respectively.
laterally on the scalp overlying the sensory-motor No evoked potentials were recorded in the

complete group. The fact that the 5 cases with
rectal sensation (out of 15 cases), diagnosed as
complete by ASIA criteria, showed neither senso-
ry- nor motor-evoked potentials suggests that
electrophysiological testing is not more ‘‘sensitive’’
than the routine neurological examination. On the
other hand, anorectal manometry provides infor-
mation on sensory function that seems more useful
than either electrophysiology or routine neurolog-
ical examination for the diagnosis of completeness
of spinal cord injury.
These findings about visceral sensation in sub-
jects with spinal cord injury also raise the question
of the pathways by which these sensations are
processed. The pelvic nerve innervates the distal
parts of the colon, rectum, and the urogenital
organs. Although afferent nerves from different
intra-abdominal nerve trunks enter multiple seg-
ments of the cord, each nerve shows peak projec-
tions to one or two adjacent segments. Thus,
innervation of different visceral organs has con-
siderable segmental overlap, which probably
explains the poor viscerotopic localization of
sensation in the gastrointestinal tract. Additional
convergence of somatic afferent projections onto
the same spinal segments is thought to be the basis
for the referral of visceral sensation to somatic
structures (Aziz and Thompson, 1998).
Visceral afferent information is transmitted
along the spinal cord via a number of tracts, of
which the spinothalamic tracts and the dorsal col-
umns are the most important. The lateral and me-
dial subdivisions of the spinothalamic tract project
to the ventral, ventral posterior lateral, medial, and
intralaminar nuclei of the thalamus, respectively.
The lateral spinothalamic neurons mediate the
sensory-discriminative aspects of pain whereas the
medial spinothalamic neurons mediate the motiva-
tional-affective aspects of pain. In contrast to con-
ventional wisdom that the dorsal columns do not
mediate visceral afferent information, recent evi-
dence from human studies now suggests that they
do, since posterior midline myelotomy that inter-
rupts the dorsal columns alleviates pelvic visceral
pain in patients with colon cancer (Gildenberg and
Hirshberg, 1984). Recent evidence from animal
studies also confirms that the dorsal column is
an important pathway for transmitting visceral
379
nociceptive information to the thalamus (Al-Chaer
et al., 1996a, b). However, since evoked potential
recordings are thought to represent predominantly
dorsal column transmission of information, other
pathways must process the remaining sensation in
our subjects with ‘‘complete’’ lesions.
Visceral afferent information is also carried in
the spinoreticular, spinomesencephalic and spino-
solitary tracts that project to the thalamus via re-
lays in the brainstem (e.g. the nucleus of the
tractus solitarius) and in the midbrain (see Aziz
and Thompson, 1998). These pathways mediate
the integration of somatic and visceral input from
wide areas of the body, and also allow afferent
information encoded within vagal afferent projec-
tions to modulate afferent information encoded
within spinal afferent nerves (Randich and Gebhart,
1992; Mayer and Gebhart, 1994). Whereas spinal
afferent nerves are usually only considered as the
pathway for transmission of nociceptive informa-
tion to the CNS, the majority of afferent projections
have stimulus response functions that cover both
physiological and nociceptive ranges of stimulation
(Aziz and Thompson, 1998). This idea is supported
by the fact that, in our study, five cord-injured sub-
jects reported non-painful and three subjects re-
ported painful sensation following rectal distension.
Due to the complex innervation of the viscera,
pathways that remain partially intact following
trauma may still be able to transmit sensory in-
formation from the rectum, and may even be able
to take over processing of visceral information to
and from the brain. Whereas plasticity has been
shown to be a feature of reorganization of the in-
jured spinal cord (Schnell et al., 1994; Schwab and
Bartholdi, 1996), the application of this to visceral
functions remains to be shown in the future.
Cortical representation of sensory functions from
the anorectum
Anecdotal evidence suggests that residual anorec-
tal sensation is present in some cord-injured people
with clinically classified complete injury of the
spinal cord. This has also been shown with inves-
tigations of anorectal functions in these cord-
injured subjects (Greving et al., 1998, and previous
section). We examined this phenomenon with
380
brain activation imaging. Cortical reorganization
following traumatic spinal cord injury is a com-
mon phenomenon, but is investigated predomi-
nantly for motor functions of the upper and lower
extremities. In people with complete spinal cord
injury, cortical areas adjacent to the denervated
one may eventually take over its functions, as
shown by cortical functional representations, and
this phenomenon is accompanied by clinical im-
provements (Lotze et al., 1999).
Ten paraplegic subjects with complete traumatic
spinal cord injury (according to ASIA criteria)
(Ditunno et al., 1994) at different levels (TH3–L3)
were investigated during non-painful stimulation
of the distal rectum and anal canal by means of
pneumatic dilatation with a balloon probe. Func-
tional magnetic resonance imaging data were ac-
quired across the whole brain with a commercial
1.5 Tesla tomograph (Siemens, Vision) using Echo
Planar Imaging (Lotze et al., 2001).
Although complete impairment was clinically
diagnosed in all 10 cord-injured subjects, four of
the subjects experienced reproducible sensations
during anal and/or rectal stimulation. In the re-
maining six subjects who had no sensations, inju-
ries of the dura or an empty spinal canal had been
reported from a surgical procedure. For the four
patients with sensation, data analysis for each was
undertaken using SPM99 (Welcome Department
of Cognitive Neurology, London) to identify the
cortical areas that show significant activation in an
event-related stimulus design. Figure 2 shows im-
ages from one individual during rectal stimulation
in sagittal, coronal, and transverse projection
planes. Individual data analysis in these four pa-
tients revealed predominant activation in the right
secondary somatosensory cortex, the posterior
cingulate gyrus, and the left posterior cerebellar
lobe. In addition, left orbitofrontal cortex activa-
tion was observed in one of the patients.
Significant activation was found with individual
data analysis but not for the whole group, most
likely due to differences between subjects in injury
patterns and in the levels of the spinal cord lesions.
With a less strict statistical criterion, group anal-
ysis was possible. For group comparison it is con-
ventional to compute a contrast between two
conditions (condition A minus condition B, e.g.
healthy controls minus patients or post-treatment
minus pre-treatment). Parameter estimates of in-
terest of each subject were computed, and mean
images across all four were used to contrast the
responses of the cord-injured subject and those of
a healthy volunteer group (Lotze et al., 2001). In
comparison with a healthy group, contrast anal-
yses revealed only the activation pattern of the
healthy subjects (Fig. 3). This implies that subjects
with residual visceral perception exhibit cortical
activation in areas similar to that found in healthy
volunteers, but this activation is less extensive and
less well coordinated. We recently were able to re-
investigate one of these four patients with residual
subjective perception and cortical activation fol-
lowing anorectal stimulation. In addition to the
stimulation, putative activation was recorded dur-
ing anticipation of stimuli that were not delivered
(sham stimulation). In the initial investigation in
this patient, relevant activation was revealed in
prefrontal (rectal stimulation) and the second so-
matosensory cortex (anal stimulation) only. After
one year, significant activation was detected bilat-
erally in the cerebellum for both stimulation sites,
as well as in the second somatosensory cortex for
rectal stimulation and in the first somatosensory
cortex for anal stimulation. This was accompanied
by further improvement of subjective awareness of
anorectal sensations. Sham stimulation did not re-
veal any cortical activation at all.
These data indicate that the diagnosis of com-
plete spinal cord injury by ASIA criteria (Ditunno
et al., 1994) alone may be insufficient, as has been
speculated previously after interoception testing in
these patients (Greving et al., 1998). Cord-injured
people with residual visceral sensation exhibit cor-
tical activation in areas similar to those found in
healthy volunteers but this activation is less ex-
tensive and less well coordinated. It exhibits, how-
ever, significant plasticity in the course of the
disease and may improve over time.
Several questions remain that cannot be an-
swered by results from the small number of pa-
tients that we have investigated. Do sensations
arising from other gastrointestinal segments acti-
vate similar brain responses? What is the time
course of the cortical activations? Is the cortical
network activated sufficient to explain clinical
Fig. 2. Functional magnetic resonance imaging of the brain of a cord-injured subject during non-painful stimulation of the distal rectum or of the anal canal with a
balloon. The injury was complete in the basis of ASIA criteria but the subject reported anal/rectal sensation. The brain is shown in three projections (sagittal, coronal, and
transverse). Significant cortical activation is present (po0.05 corrected) in different regions during rectal (top) and anal (bottom) stimulation. Top: Second somatosensory
cortex bilaterally, right posterior cingulate cortex, and left cerebellum. Bottom: right first and bilateral second somatosensory cortex, posterior cingulate cortex, and left
cerebellum.

381
382

Fig. 3. Group analysis (contrasts: healthy minus paraplegic subjects) following rectal stimulation revealed significant activation in the
right second somatosensory cortex and left cerebellar cortex during rectal stimulation (see text for details).

symptoms? Does it respond to therapeutic inter-
ventions with plasticity and change? Can thera-
peutic approaches and patient management be
built upon this phenomenon, e.g. biofeedback
therapy? This warrants future investigations.
Summary and conclusions
The research discussed above demonstrates that
both sensory and motor functions of the gastro-
intestinal system can be significantly impaired in
people with complete or incomplete spinal cord
injury. Multiple mechanisms may underlie this im-
pairment, ranging from disruption of sympathetic
and parasympathetic (extrinsic) pathways that af-
fect autonomic motor and sensory functions to
changes secondary to altered individual mobility
and behaviors. Further research is needed to in-
vestigate these possibilities, and to develop man-
agement strategies for the bowel problems in those
with spinal cord injury.
Acknowledgment
Supported by grants from Deutsche Forschungs-
gemeinschaft, DFG En 50/18 and 50/21.

References
Al-Chaer, E.D., Lawand, N.B., Westlund, K.N. and Willis,
W.D. (1996a) Visceral nociceptive input into the ventral
posterolateral nucleus of the thalamus. A new function for
the dorsal column pathway. J. Neurophysiol., 76: 2661–2674.
Al-Chaer, E.D., Lawand, N.B., Westlund, K.N. and Willis,
W.D. (1996b) Pelvic visceral input into the nucleus gracilis is
largely mediated by the postsynaptic dorsal column pathway.
J. Neurophysiol., 76: 2675–2690.
Aziz, Q. and Thompson, D.G. (1998) Brain-gut axis in health
and disease. Gastroenterology, 114: 559–578.
Beuret-Blanquart, F., Weber, J., Gouverneurm, J.P., Demangeon,
S. and Denis, P. (1990) Colonic transit time and anorectal
manometric anomalies in 19 patients with complete transection
of the spinal cord. J. Auton. Nerv. Syst., 30: 199–208.
Braden, B., Adams, S., Duan, L.P., Orthm, K.H., Maul, F.D.,
Lembcke, B., Hor, G. and Caspary, W.F. (1995) The [13C]ac-
etate breath test accurately reflects gastric emptying of liquids
in both liquid and semisolid test meals. Gastroenterology,
108: 1048–1055.
Bruninga, K. and Camilleri, M. (1997) Colonic motility and tone
after spinal cord injury. Am. J. Gastroenterol., 92: 891–894.
Chen, C.Y., Chuang, T.Y., Tsai, Y.A., Tai, H.C., Lu, C.L.,
Kang, L.J., Lu, R.H., Chang, F.Y. and Lee, S.D. (2004) Loss
of sympathetic coordination appears to delay gastrointestinal
transit in patients with spinal cord injury. Dig. Dis. Sci., 49:
738–743.
Chew, C.G., Bartholomeusz, F.D., Bellon, M. and Chatterton,
B.E. (2003) Simultaneous 13C/14C dual isotope breath test
measurement of gastric emptying of solid and liquid in nor-
mal subjects and patients: comparison with scintigraphy.
Nucl. Med. Rev. Cent. East Eur., 6: 29–33.
Coremans, G., Geypens, B., Vos, R., Tack, J., Margaritis, V.,
Ghoos, Y. and Janssens, J. (2004) Influence of continuous
isobaric rectal distension on gastric emptying and small bow-
el transit in young healthy women. Neurogastroenterol.
Motil., 16: 107–111.
De Looze, D.A., De Muynck, M.C., Van Laere, M., De Vo,
M.M. and Elewaut, A.G. (1998b) Pelvic floor function in
patients with clinically complete spinal cord injury and its
relation to constipation. Dis. Colon Rectum,, 41: 778–786.
De Looze, D.A., Van Laere, M., De Muynck, M.C., Beke, R.
and Elewaut, A.G. (1998a) Constipation and other chronic
gastrointestinal problems in spinal cord injury patients. Spi-
nal Cord, 36: 63–66.
DeVault, K.R., Beacham, S., Castell, D.O., Streletz, L.J. and
Ditunno, J.F. (1996) Esophageal sensation in spinal cord-
injured patients: balloon distension and cerebral evoked po-
tential recording. Am. J. Physiol., 271: G937–941.
Devroede, G., Arhan, P., Duguay, C., Tétreault, L., Akoury, H.
and Perey, B. (1979) Traumatic constipation. Gastroenterol-
ogy, 77: 1258–1267.
Ditunno, J.F., Young, W., Donovan, W.H. and Creasey, G.
(1994) The international standard booklet for neurological
and functional classification of spinal cord injury. Paraplegia,
32: 70–80.
383
Enck, P., Bielefeldt, K., Rathmann, W., Purrmann, J., Tschöpe,
D. and Erckenbrecht, J.F. (1991) Epidemiology of faecal in-
continence in selected patient groups. Int. J. Colorect. Dis., 6:
143–146.
Enck, P., Herdmann, J., Börgermann, K., Theisen, U., Zacchi,
P. and Lübke, H.J. (1992) Up and down the spinal cord:
afferent and efferent innervation of the human external anal
sphincter. J. Gastrointest. Motil., 4: 271–277.
Fealey, R.D., Szurszewski, J.H., Merrit, J.L. and DiMagno,
E.P. (1984) Effect of traumatic spinal cord transection on
human upper gastrointestinal motility and gastric emptying.
Gastroenterology, 87: 69–75.
Gaston, E.A. (1948) The physiology of fecal continence. Surg.
Gynec. Obstr., 87: 280–290.
Gildenberg, P.L. and Hirshberg, R.M. (1984) Limited myelo-
tomy for the treatment of intractable cancer pain. J. Neurol.
Neurosurg. Psychiatry,, 47: 94–96.
Glick, M.E., Meshkinpour, H., Haldeman, S., Hoehler, F.,
Downey, N. and Bradley, W.E. (1984) Colonic dysfunction in
patients with thoracic spinal cord injury. Gastroenterology,
86: 287–294.
Glickman, S. and Kamm, M.A. (1996) Bowel dysfunction in
spinal cord injury patients. Lancet, 347: 1651–1653.
Greving, I., Nedjat, S., Orth, G., Tegenthoff, M., Bötel, U.,
Micklefieldt, G., May, B. and Enck, P. (1998) Anorectal
functions in patients with spinal cord injury. Ne-
urogastroenterol. Motil., 10: 509–515.
Han, T.R., Kim, J.H. and Kwon, B.S. (1998) Chronic gastro-
intestinal problems and bowel dysfunction in patients with
spinal cord injury. Spinal Cord, 36: 485–490.
Herdmann, J., Bielefeldt, K. and Enck, P. (1991) Quantification
of motor pathways to the pelvic floor in humans. Am. J.
Physiol., 260: G720–723.
Kao, C.H., Ho, Y.J., Changlai, S.P. and Ding, H.J. (1999)
Gastric emptying in spinal cord injury patients. Dig. Dis. Sci.,
44: 1512–1515.
Keshavarzian, A., Barnes, W.E., Bruninga, K., Nemchausky, B.,
Mermall, H. and Bushnell, D. (1995) Delayed colonic transit
in spinal cord-injured patients measured by Indium-111
amberlite scintigraphy. Am. J. Gastroenterol., 90: 1295–1300.
Kirk, P.M., King, R.B., Temple, R., Bourjaily, J. and Thomas,
P. (1997) Long-term follow-up of bowel management after
spinal cord injury. SCI Nurs., 14: 56–63.
Longo, W.E., Ballantyne, G.H. and Modlin, I.M. (1989) The
colon, anorectum, and spinal cord patient. A review of the
functional alterations of the denervated hindgut. Dis. Colon
Rectum, 32: 261–267.
Lotze, M., Laubis-Herrmann, U., Erb, M., Topka, H. and
Grodd, W. (1999) Reorganization in the primary motor cor-
tex after spinal cord injury — a functional magnetic reso-
nance study. Restor. Neurol.Neurosci., 14: 183–187.
Lotze, M., Wietek, B., Birbaumer, N., Ehrhardt, J., Grodd, W.
and Enck, P. (2001) Cerebral activation during anal and rec-
tal stimulation. NeuroImage, 14: 1027–1034.
MacDonagh, R., Sun, W.M., Thomas, D.G., Smallwood, R.
and Read, N.W. (1992) Anorectal function in patients with
complete supraconal spinal cord lesions. Gut, 33: 1532–1538.
384
May, R.J. (1991) Fecal incontinence in a 39-year-old man.
Gastroenterology, 101: 1120–1126.
Mayer, E.A. and Gebhart, G.F. (1994) Basic and clinical aspects
of visceral hyperalgesia. Gastroenterology, 107: 271–293.
Menter, R., Weitzenkamp, D., Cooper, D., Bingley, J., Charlifue,
S. and Whiteneck, G. (1997) Bowel management outcomes in
individuals with long-term spinal cord injuries. Spinal Cord,
35: 608–612.
Mollen, R.M., Hopman, W.P., Kuijpers, H.H. and Jansen, J.B.
(1999) Abnormalities of upper gut motility in patients with
slow-transit constipation. Eur. J. Gastroenterol. Hepatol., 11:
701–708.
Neill, M.E. and Swash, M. (1980) Increased motor unit fibre
density in the external anal sphincter muscle in ano-rectal
incontinence: a single fibre EMG study. J. Neurol. Ne-
urosurg. Pychiatry,, 43: 343–347.
Randich, A. and Gebhart, G.F. (1992) Vagal afferent modu-
lation of nociception. Brain Res. Rev., 17: 77–99.
Rao, S.S.R., Azpiroz, F., Diamant, N., Enck, P., Tougas, G.
and Wald, A. (2002) Minimum standards of anorectal man-
ometry. Neurogastroenterol. Motil., 14: 553–559.
Read, N.W. and Sun, W.M. (1992) Anorectal manometry. In:
Henry M.M. and Swash M. (Eds.), Coloproctology and the
Pelvic Floor (2nd ed). Butterworth, London, pp. 119–145.
Schnell, L., Schneider, R., Kolbeck, R., Barde, Y.A. and
Schwab, M.E. (1994) Neurotrophin-3 enhances sprouting of
corticospinal tract during development and after adult spinal
cord lesions. Nature, 367: 170–173.
Schwab, M.E. and Bartholdi, D. (1996) Degeneration and
regeneration of axons in the lesioned spinal cord. Physiol.
Rev., 76: 319–370.
Singh, G. and Triadafilopoulos, G. (2000) Gastroesophageal
reflux disease in patients with spinal cord injury. J. Spinal
Cord Med., 23: 23–27.
Stone, J.M., Nino-Murcia, M. and Wolfe, V.A. (1990) Gastro-
intestinal problems in spinal cord injury patients: a prospec-
tive analysis. Am. J. Gastroenterol., 85: 1114–1119.
Sun, W.M., MacDonagh, R., Forster, D., Thomas, D.G.,
Smallwood, R. and Read, N.W. (1995) Anorectal function
in patients with complete spinal transection before and
after sacral posterior rhizotomy. Gastroenterology, 109:
990–998.
Sun, W.M., Read, N.W. and Donnelly, T.C. (1990) Anorectal
function in incontinent patients with cerebrospinal disease.
Gastroenterology, 99: 1372–1379.
Wheatley, I.C., Hardy, K.J. and Dent, J. (1977) Anal pressure
studies in spinal patients. Gut, 18: 488–490.
Youle, M.S. and Read, N.W. (1984) Effect of painless rectal
distension on gastrointestinal transit of solid meal. Dig. Dis.
Sci., 29: 902–906.
Zahn, A., Langhans, C.D., Hoffner, S., Haberkorn, U., Rating,
D., Haass, M., Enck, P., Stremmel, W. and Ruhl, A. (2003)
Measurement of gastric emptying by 13C-octanoic acid
breath test versus scintigraphy in diabetics. Z. Gastroenterol.,
41: 383–390.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 26

Problems of sexual function after spinal cord injury

Stacy L. Elliott1,2,3,4,

1
Departments of Psychiatry and Urology, University of British Columbia, BC, Canada
2
British Columbia Center for Sexual Medicine, Echelon-5, 855 West 12th Avenue, Vancouver, BC, V5Z 1M9, Canada
3
Vancouver Sperm Retrieval Clinic, Vancouver Hospital, Vancouver, BC, Canada
4
G.F. Strong Rehabilitation Centre, Vancouver, BC, Canada

Abstract: Sex is a legitimate and fundamental need in humans. Substantial changes to both the autonomic
and somatic nervous system occur after spinal cord injury, and result in altered sexual function and fertility
potential. This chapter provides a clinical overview of the main sexual and reproductive concerns and
priorities men and women face after spinal cord injury. Besides genital functioning, other autonomic
functions affect sexuality, such as bladder and bowel function, cardiovascular control and temperature
regulation. These interlinked autonomic functions are presented in their impact on sexuality. The mind-
body interaction and spinal feedback loops are discussed. It is proposed that human sexuality after spinal
cord injury can be a model for investigating integrated autonomic function. Recent research on the meas-
urement of cardiovascular parameters during vibrostimulation and ejaculation demonstrates the discord-
ance between objective and subjective signs of autonomic dysreflexia. It is hoped that health care
professionals and researchers will become motivated to attend to the unmet sexual health care needs of this
population.

What is sex? Sex is a highly individualized experience. The

Sex is a legitimate and fundamental need in human
beings. Sexual functioning is recognized by the
health care profession as an area of joy for many
people, but it can also be an area of great mental
and physical suffering. Medicine is mandated to
relieve suffering. After spinal cord injury, in gen-
eral, sexual satisfaction decreases (Berkman et al.,
1983; Alexander et al., 1993; Reitz et al., 2004).
However, sexuality after spinal cord injury remains
a central motivating factor in life (Reitz et al.,
2004). Both health care professionals treating
patients and clients and researchers in the area of
somatic and autonomic function must address this
significant area.
Corresponding author. Tel.: +604-875-8273;
Fax: +604-875-8249; E-mail: stacy.elliott@vch.ca
drive to be sexual is a combination of biological
drive, the degree of success in presenting (and ac-
cepting) oneself as a sexual person to the world,
positive sexual experiences and harmonic integra-
tion with cultural and spiritual perspectives. Ulti-
mately, the priority any one individual places on
sex depends on the ‘‘pay-off’’ that comes with be-
ing sexual: does it result in a pleasurable physical
experience one wishes to repeat, a more intimate
relationship with a partner, an improved mood
and attitude about life, or better societal accept-
ance in some way?
Physiologically, for sex to unfold in a natural
and rewarding way, people have to trust their
bodies. The sexual ‘‘pay-off’’ has to be worth it, or
there will be less motivation to repeat the be-
havior. Thus, an erection should occur with pre-
dictability and not be a source of embarrassment
or disappointment, and vaginal lubrication from

DOI: 10.1016/S0079-6123(05)52026-0 387

388
adequate sexual arousal should assist with com-
fortable sexual intercourse. Although orgasm is
not a prerequisite to a rewarding sexual experi-
ence, reaching orgasm with reliability after spinal
cord injury is a bonus.
When a person sustains a spinal cord injury,
there is a new body to learn, to adapt to and to
trust. Not only do the sensory and motor changes
affect the ability to get on with the tasks of daily
living, they severely affect sexual capacities as well.
Without, for example, sensation in one’s genitalia,
how can one interpret genital touch by a partner as
sexual? With interruption of sexual reflexes, not
only are automaticity and reliability of sexual re-
sponses diminished or gone, they are sometimes
cross-wired with other pelvic automatic responses
such as perspiration, bladder and bowel function.
What a daunting experience to relearn how to
make one’s sexual body parts be functional again
and to use therapies to accomplish this! Even more
complicated is the integration of this discovered
function with the learned subtleties of post-injury
sexual arousal and gratification that come with
experimentation, experience (positive and nega-
tive) and patience. Some men and women with
cord injury claim their injury has spurred them
onto a level of sexual depth and understanding
they would not have otherwise known.
Readiness to address sexuality will vary from
queries during the acute management (‘‘Will I ever
be able to have sex again? Can I still have kids?’’)
to queries several years after injury (‘‘How do I
have intercourse with my new boyfriend with this
indwelling catheter?’’). It takes time and shaping
experiences to become open to sexual experimen-
tation, to learn a ‘‘map’’ of sexual potential on
sensate areas not traditionally thought to be sex-
ually rewarding and to appreciate the sexual ca-
pacity that high mental arousal alone can provide.
Clinicians in the field of sexual medicine are ac-
quainted with the tenacity and positiveness that
many men and women with spinal cord injury
bring to the table when sexuality is finally a focus
in their lives.
The majority of studies of sex following spinal
cord injury focus on physical changes, with far
fewer studies on sexual satisfaction and prediction
of positive sexual outcomes. Tepper et al. (2001)
examined the psychological, emotional and rela-
tionship aspects of sexuality in the lives of 15
women who sustained a complete spinal cord in-
jury between the levels of the thoracic (T)6 and
lumbar (L)2 segments. These researchers were in-
terested in the ‘‘real-life experience’’. The theme of
readiness was addressed; even though the study
subjects had regained their sense of identity as
people in the world, their sense of sexual identity
lagged years behind their overall self-esteem re-
covery. With time, experience with a partner and
changes in self and relationships, these women
eventually set sexual exploration as a higher pri-
ority. A set of continuous themes was identified:
cognitive genital dissociation (purposefully ‘‘shut-
ting out’’ sexuality) followed by sexual disenfran-
chisement (feeling ‘‘shut out’’ and having poor
sexual self-esteem) and eventually sexual rediscov-
ery. The latter was often triggered by a turning
point (e.g., new partner, orgasmic experience, etc.).
The subjects identified a relationship with a part-
ner to be the most significant event affecting their
sexual recovery. Similarly, in another study (Ek-
land and Lawrie, 2004), women with spinal cord
injury underwent three distinct phases while redis-
covering sexuality after injury: avoidance, increas-
ing comfort and exploration. Our clinical
experience with men following spinal cord injury
suggests that men move faster in their sexual re-
habilitation than women. Regaining sexual func-
tion appears to be an earlier priority for them.
However, this has not been substantiated by
‘‘lived-experience’’ studies similar to those done
in women.
These findings emphasize that sexual recovery is
part of post-spinal cord injury rehabilitation, and
that sex is important in the lives of men and wom-
en after injury. Past researchers have not always
felt that way. Cole et al. (1973) fought against the
dogma that sexual rehabilitation rated low against
other rehabilitation variables, by noting that in the
lay literature many persons with cord injury
mourned the loss of sexual function more than
other deficits such as loss of walking. For the next
30 years many well-meaning rehabilitation profes-
sionals did not recognize that sex was a priority for
their patients: their focus was survival, then inde-
pendence. Sexual rehabilitation was a luxury, and

an uncomfortable one at that to discuss or treat. In
the 1970s and early 1980s, not many attractive
medical therapies were available; sexual counseling
and adoption of children rather than procreation
were the mainstays.
In her 2004 survey of priorities for over 600
persons with spinal cord injury, Anderson con-
firmed what health care professionals dealing with
sexuality have always known: sexuality is very im-
portant to the majority of persons after spinal cord
injury (Anderson, 2004). The survey found regain-
ing sexual function to be either the first or second
highest priority of 28.3% of quadriplegics and
45.5% of paraplegics, compared with other prior-
ities such as recovering motor function and blad-
der and bowel issues. Sex is no longer to be
disregarded as relatively unimportant for the ma-
jority of persons with spinal cord injury.
In this era of early scientific progress in sexuality
following cord injury, it is critical to provide use-
ful, practical assistance to persons with spinal cord
injury as soon as they are ready. What a difference
proper sexual rehabilitation can make: addressing
the mind-body connection together versus focus-
ing only on the genitalia, can make all the differ-
ence to a successful outcome. Misinformation or
unrealistic expectations can lead to months or
years of non-productive sexual experience accom-
panied by negative psychological overlay. For ex-
ample, even though the reflex component of
orgasm for women may need to be relearned and
reinforced with practice, starting too early after
injury to experiment with recovering reflexes may
undermine a woman’s sexual confidence and make
her give up hope of a future sex life. Comarr and
Vigue (1978) suggested that it might take at least 6
months to know the level of erectile recovery, es-
pecially among men with incomplete lesions. For
patients with lower motor neuron or cauda equina
injury, it may take a year since there is more pos-
sibility of nerve regeneration as compared with
injury located within the spinal cord. Providing a
pharmacological rigid erection for a young injured
man may provide tremendous relief and opportu-
nity to regain that part of his life soon after injury.
But when he is unable to ejaculate or experience
orgasm for years he may feel discouraged. What if
his only source of pleasure with a partner is
389
vicarious? He may need assistance to maximize the
great sexual potential that remains and to pursue
higher arousal and even orgasmic release through
more non-conventional avenues (e.g., Tantric sex
experimentation discussed by Tepper: www.sexu-
alhealth.com). Therefore the issues of past sexual
experiences, willingness to experiment and readi-
ness are of paramount importance. These are
espoused in a set of sexual rehabilitation principles
(Elliott, 2003). These principles include the max-
imization of the underlying physiology, followed
by adaptation to the remaining limitations with
the use of medical enhancement and, finally, being
open to potentially new therapies and attitudes
previously not part of pre-injury sexual practice.
Impact on sexual function of the autonomic,
motor and sensory changes associated with spinal
cord injury
Despite the recent advances in the knowledge of
male sexual function and in therapies for male
sexual dysfunction, sexual neurophysiology, as a
science, is a complicated subject with many un-
knowns. The highly complex role of central auto-
nomic and somatic control, the interconnection
between spinal reflex pathways, and the continual
local and generalized moment-to-moment cerebral
and intraspinal feedback that happens during sex-
ual activity are very hard to systematize accurate-
ly, especially in human basic science research.
Human spinal cord injury has been a paramount
model in our understanding of sexual neurophys-
iology. The role of the autonomic nervous system
is crucial in sexual functioning. Human sexuality
after spinal cord injury is a model for investigating
integrated autonomic function. To learn, or rather
to relearn, a sexual sense of self in a new body
requires the brain’s interpretation, not just of sen-
sory and motor, but also of autonomic, compo-
nents. In the simplest of terms, the physiology
associated with sexual arousal is primarily para-
sympathetic. As sexual arousal increases, a critical
neurological threshold (often termed the ‘‘orgas-
mic’’ or ‘‘ejaculatory’’ threshold) is reached, and
the sympathetic nervous system predominates.
When the autonomic nervous system is altered
390
below the level of spinal cord injury, sexual arous-
al can generate altered cardiovascular responses
and disordered sexual, bladder and bowel proc-
esses. More stimuli may be required to trigger
what remains of sexual reflexes, and extragenital
stimuli (i.e., around the hypersensitive level of in-
jury, or the face, neck and ears) may take on
a new, amplified and critical importance in the
genesis of sexual arousal.
The use of animal models to research sexual
functioning, while critical to our current knowl-
edge base, is limited, as subjective data about sex-
ual experiences is not available. If we listen and
observe carefully, the men and women with spinal
cord injury are the best teachers of sexuality after
injury. They dictate what needs to be researched at
a human trial level. What have we, as health care
professionals and researchers, learned in terms of
autonomic responses?
Sexual feelings and responses in the spinal cord
injured population (and in the able-bodied for that
matter) constitute a head-to-toe approach. After
spinal cord injury, the changes instigated by the
altered autonomic system affecting sexual function
and manifestations of sexuality are as equally, if
not more, important as those from motor and
sensory alterations. This suggested head-to-toe
‘‘autonomic’’ framework of sexual changes fol-
lowing spinal cord injury is meant to elucidate the
interconnection between autonomic functions that
either interfere with sexuality, or are a conse-
quence of sexual activity. With real-life spinal cord
injury experiences, altered genital functioning may
or may not be viewed, by the injured person, as the
factor most affecting his or her sexuality. This is
important for researchers and clinicians to appre-
ciate. Research and clinical efforts need to be di-
rected at the most important sexual rehabilitation
priorities identified by men and women with spinal
cord injury. To date, such preference studies are
few.
An altered nervous system can affect sexuality
from head to toe.
1. Cerebral functioning: The mind is the ruler of
sexual desire, sexual self-image and mood al-
terations following spinal cord injury. De-
pression from overwhelming life changes,
sadness and anger at being robbed of sexual
sensations, frustration at altered motor abil-
ity and autonomic coordination and all the
influences mentioned below make the mind
the ultimate arbitrator of sexual pursuit.
2. Genital functioning: The spinal sexual reflexes
are under tonic inhibitory control by supra-
spinal neurons (Chuang and Steers, 1999). If
this control is removed by spinal cord injury,
the spinal reflexes are free to be triggered
without conscious control (i.e., reflex erec-
tions). However, the intactness of the genital
reflexes required for sexual functioning
(arousal, orgasm, etc.) is not necessarily re-
lated to the intactness of sensory or motor
function. Remaining reflexes can be thera-
peutically enhanced: alternate learning is re-
quired for sexual rehabilitation.
3. Bladder functioning: Bladder functioning de-
pends on remaining autonomic pathways.
Factors such as bladder continence dictate
freedom from leakage or accidents during
sexual activity. Neurogenic bladders increase
the risk for urinary tract and kidney infec-
tions. Urine odors and various external urine
collection apparatuses are usually associated
negatively with sexuality.
4. Bowel function: Bowel-emptying practices are
often time-consuming. Bowel evacuation pat-
terns can affect general wellness and energy.
Flatulence or stool incontinence can be one
of the most socially embarrassing and anti-
sexual events that a person with a spinal cord
injury can experience.
5. Cardiovascular and respiratory function: Al-
tered cardiovascular responses during sexual
arousal and orgasm/ejaculation occur after
cord injury. Altered heart rates and rhythms,
symptomatic hypotension in quadriplegics or
the hypertension of autonomic dysreflexia
can alter the safety and willingness to be sex-
ual. The same effect can result from respira-
tory problems like loss of vital capacity due
to paralysis of the intercostal muscles, in-
crease in airway resistance due to the decrease
in bronchodilator sympathetic neuronal ac-
tivity and loss or attenuation of the cough
reflex. In most cases, autonomic dysreflexia is

a deterrent to safe or enjoyable sexual prac-
tices and can interfere with sperm retrieval
for fertility purposes in men with spinal cord
injury.
6. Altered thermal regulation and skin integrity:
Unnatural appearance or texture of the skin
can affect sexual self-esteem. Skin integrity is
altered and extended pressure leads to skin
breakdown, affecting sexual positioning.
Lack of sweating and vasodilatory responses
below the level of lesion or unattractive ab-
normal flushing and sweating above the level
of lesion can impede the motivation to be
sexual. Inability to tolerate nakedness due to
chilling can be restrictive for both partners.
Cold extremities are not uncommon after
spinal cord injury, and are often avoided by a
sexual partner.
7. Sensory deficits: These severely affect the
neural pathways reinforced from previous
sexual learning. The highly motivated person
with spinal cord injury can reprogram senso-
ry afferents from previously ‘‘non-sexual’’ in-
puts on the body to have sexual flavoring.
This may be brain plasticity or other relearn-
ing phenomenon (Bach-y-Rita, 1999), a vir-
tually unexplored area.
8. Motor deficits: Loss of abdominal tone (i.e.,
‘‘quad belly’’) affects sexual self-view. De-
creased muscle tone and bulk affects sexual
self-esteem and pride, especially if this was a
source of sexual confidence before the injury.
Loss of motor coordination or strength can
preclude even holding or caressing a partner.
Inability to balance or support the upper
trunk affects sexual positioning options and
ability to thrust with the pelvis. Muscle
spasms and clonus can affect sexual enjoy-
ment and may preclude sexual positioning
options.
Medications and other pre-existing medical con-
ditions can also influence sexual function and re-
ceptivity.
There are many clinical examples of interplay of
various autonomic functions during sexual activ-
ity. For example, a woman with spinal cord injury
may be at risk for bowel incontinence during
391
sexual intercourse if her bowel reflexes are trig-
gered by penile penetration (Szasz, 1983). Bladder
or bowel fullness can impede or facilitate the trig-
gering of ejaculation by vibrostimulation for
sperm retrieval. Autonomic dysreflexia can result
in avoidance of sexual activity by both sexes. Al-
though these examples demonstrate the negative
consequences of sexual arousal and activity, there
are positive consequences also. Obviously, the
pursuit of sexual activity after injury is a result of
weighing the pros and cons, and fortunately for
most, the pros prevail. For example, some men
and women with spinal cord injury who experience
mild autonomic dysreflexia learn, with time, to in-
terpret those altered cardiovascular processes as
sexually enhancing.
With spinal cord injury, disturbances in the sen-
sory input to the brain, descending autonomic
control of reflex centers in the spinal cord and al-
tered efferent reflex abilities result in various com-
plete, incomplete or even aborted sexual responses.
Along with the changes seen in physiological sex-
ual responses following spinal cord injury, the
person with this injury must learn and adapt to his
or her sexual new body. Psychological feedback
feeds into recovery progression. This mind-body
interaction leads to, or deters, sexual curiosity, sex-
seeking behaviors and motivation to express inti-
macy in a sexual manner.
Sexual activity in people with spinal cord injury
may have positive consequences, such as relief
from spasms following ejaculation (Courtois et al.,
2004). This anti-spastic effect has been reported
following ejaculation either by vibrostimulation
(Szasz and Carpenter, 1989; Elliott, 2003; Laessoe
et al., 2004) or electroejaculation (Halstead et al.,
1993) and can last for several hours.
Changes in sexual responses after spinal cord injury
Masters and Johnston (1966) outlined the complex
genital and cardiovascular changes associated with
arousal, orgasm and ejaculation through the de-
velopment of a sexual response cycle. They defined
four phases of a ‘‘sexual mountain’’, including ex-
citement, plateau, orgasm and resolution. Not
much has been written about how the sexual
392
response cycle changes after cord injury. For ex-
ample, Cole (1975) noted that with sexual arousal,
the labia in women with spinal cord injury only
swell, whereas they swell and open in non-injured
women. In men with spinal cord injury, ejaculation
is almost always absent. When erection occurs, it
may not be as reliable as experienced before injury
and may detumesce with pressure or the penis may
have an altered filling capacity (i.e., the glans may
not fill with high arousal). Although typically with
sexual arousal and orgasm, muscle tone increases,
and heart rate, respiratory rate and blood pressure
increase (Masters and Johnson, 1966), cardiovas-
cular responses with arousal in spinal cord injured
people may be overexaggerated (i.e., autonomic
dysreflexia). Resolution of cardiovascular changes
may take longer than pre-injury, especially if there
is atypical autonomic dysreflexia (Elliott and
Krassioukov, 2005).
Clinical issues of female sexual function after spinal
cord injury
The literature on sexuality after spinal cord injury
is dominated by men. Men are approximately 75%
of the spinal cord injury population and, therefore,
too few women are available to warrant statisti-
cally significantly sized samples in many studies
(Siosteen et al., 1990). Like children, women have
traditionally been excluded from many medical
studies. Surveys of women with spinal cord injury
show definite loss of sexual functioning and sexual
satisfaction post injury. While rates of sexual in-
tercourse and orgasmic success both drop after
injury, they both increase in frequency over time,
although not to pre-injury levels (Jackson and
Wadley, 1999). In one large survey by the Kinsey
Institute (Donohue and Gebhard, 1995), about
60% of the women with spinal cord injury mas-
turbated after injury, with half experiencing vag-
inal lubrication and only a third experiencing
orgasm. In another survey (White et al., 1993), the
researchers noted less intercourse but relatively
good sexual satisfaction among women with spinal
cord injury. They also found that injury before the
age of 18 years implied a greater risk of not having
an active sex life.
Approximately 50% of women with spinal cord
injury (including women with complete injury)
who participated in a study by Sipski et al. (1995)
were able to achieve orgasm (see also Sipski and
Arenas, this volume). In the Kinsey survey (Dono-
hue and Gebhard, 1995), women with incomplete
lesions attained orgasm more frequently than
women with complete lesions, regardless of the
level of injury. For women with spinal cord injury,
the incidence of orgasm generated by self-stimula-
tion decreased, and the time required to attain or-
gasm increased (Donohue and Gebhard, 1995;
Sipski et al., 1995; Ferrerio-Velasco et al., 2005).
Breast stimulation and mechanical genital stimu-
lation were often employed. Many compensated
for the loss of genital sensitivity by stimulating a
sensate part of their body. Greater sexual knowl-
edge and higher sex drive were two variables in
women that could predict a higher chance of
reaching orgasm after spinal cord injury (Donohue
and Gebhard, 1995).
After spinal cord injury, women have many sex-
ual concerns other than orgasmic difficulties. They
include altered arousal capacity (i.e., vaginal lu-
brication and accommodation), impaired motor
ability, decreased sexual satisfaction and concerns
about attractiveness and partnerships. In their
survey of women with spinal cord injury, Jackson
and Wadley (1999) noted an increased concern
about sexually transmitted diseases. They also
noted that, whereas fertility is not affected, preg-
nancy is associated with risks and complications
(increased urinary tract infections, changes in
bladder management, increased risk for skin
breakdown, difficult transfers, increased risk for
deep vein thrombosis and delayed bowel empty-
ing). Labor and delivery must be monitored since,
depending on the level of lesion, labor may not be
felt. Women with spinal cord injury also give birth
to lower birth weight infants (Jackson and Wad-
ley, 1999), the etiology of which is not clearly un-
derstood.
As noted previously, many surveys have shown
that the greatest sexual concerns for women with
spinal cord injury are problems associated with
urinary and bowel accidents and with autonomic
dysreflexia (Jackson and Wadley, 1999). While
dysreflexia usually presents in women with injury

higher than neurological level T6, this survey also
noted reports of autonomic dysreflexia in a small
number of women with lumbar/sacral injury.
Menopausal changes mirror able-bodied coun-
terparts (Dannels and Charlifue, 2004). The peri-
menopause presents a unique challenge for women
with spinal cord injury since the symptoms may
mimic or mask conditions associated with spinal
cord injury, such as autonomic dysreflexia, infec-
tions, impaired temperature regulation and spinal
cord cyst development. Autonomic dysreflexia
during pregnancy and labor must be distinguished
from pre-eclampsia, another hypertensive condi-
tion associated with pregnancy that requires dif-
ferent treatment (Pope et al., 2001; Yarkony and
Chen, 1995). There are reports of cerebral intra-
ventricular hemorrhage (McGregor and Meeuw-
sen, 1985) with resultant neurological deficits and
death (Abouleish, 1980) associated with unrecog-
nized autonomic dysreflexia during labor or deliv-
ery. Methods to prevent autonomic dysreflexia in
these circumstances include the use of epidural
anesthesia to block the reflex arc or prompt deliv-
ery by caesarian section (Yarkony and Chen, 1995;
Pereira, 2003).
Clinical issues of male sexual function after spinal
cord injury
Central to male sexual functioning after spinal
cord injury is the presence of an erection and the
option of sexual intercourse. While it is true that
non-coital activities can be highly satisfying sexu-
ally, most men with spinal cord injury need to
know whether an erection adequate for sexual in-
tercourse is possible, either naturally or assisted
(Dr. Claus Hultling, personal communication).
Erection requires participation of sacral parasym-
pathetic (pelvic), thoracolumbar sympathetic
(hypogastric and lumbar sympathetic chain) and
somatic (pudendal) nerves (Chuang and Steers,
1999). Vasodilator preganglionic neurons (prima-
rily parasympathetic) become activated, the activ-
ity of the vasoconstrictor preganglionic neurons
(mainly sympathetic) become suppressed and pe-
nile tumescence occurs. Activation of the somatic
motoneurons causes contraction of the pelvic floor
393
muscles (McKenna, 1999). The result is an increase
in penile intracavernosal pressure, leading to pe-
nile rigidity (Schmidt and Schmidt, 1993).
As will be discussed by Brown, Hill and Baker
(this volume), basically two distinct control mech-
anisms induce penile erection: reflexogenic and
psychogenic. To understand why this distinction is
clinically relevant on a daily basis for the man with
spinal cord injury, a brief description of the auto-
nomic involvement is presented. The tactile-de-
pendent reflexogenic erection is mediated by a
reflex arc that is complete at the sacral spinal level.
The afferent limb is composed of the dorsal nerve
of the penis/pudendal nerve, and the efferent limb
consists of preganglionic axons traveling in the
pelvic nerve to the pelvic plexus, where ganglion
cells send axons to the penis via the cavernous
nerve (Chuang and Steers, 1999). Spinal cord in-
jury above the sacral level will not only preserve
the reflex but can even enhance it, especially if the
lesion is complete (Chuang and Steers, 1999), due
to loss of tonic inhibitory control. This loss results
in a decrease in the sensory threshold and latency
of erection (McKenna, 1999). Clinically, men with
cervical injuries come to expect reflex erections to
non-sexual touch stimulation associated with sit-
uations such as catheterization, chafing from
clothing, or bumping on rough road in their
wheelchair. Such reflex erections are labeled
‘‘spontaneous’’; they are reflex in origin and are
produced by the same touch mechanism that pro-
duces sexual erections.
Alternately, the ability to have reflex erections is
lost if the sacral spinal cord is injured or if the
pudendal nerve or pelvic nerve is destroyed
(Chuang and Steers, 1999). The bulbocavernosus
reflex, a polysynaptic response elicited by low
threshold pudendal sensory fibers, activates pu-
dendal motorneurons to contract the striated peri-
neal muscles (McKenna, 1999). Since tactile
stimulation, especially to the glans penis, activates
the bulbocavernosus reflex, Szasz (1986) suggests
that the clinical test of the bulbocavernosus reflex
can determine the potential for reflex erections by
demonstrating an intact sacral reflex.
Psychogenic erections occur in men with an in-
tact nervous system in response to various stimuli
(Chuang and Steers, 1999). The sensory inputs
394
(or afferent limb) are processed in the higher centers
(imaginative in the limbic system, olfactory in the
rhinencephalon, visual in the occipital regions and
tactile in the thalamus inputs) and are integrated in
the medial pre-optic and anterior hypothalamic
regions and paraventricular nucleus. The brain also
receives sensory input from the penis via ascending
spinal pathways. The efferent limb of this reflex
pathway from the brain is through the lumbar
sympathetic and sacral parasympathetic outflow to
the penis. Men with injuries to their sacral cord
or lower are often dependent on these intact
psychogenic pathways to elicit an erection; they
are also dependent on maintaining their mental
sexual arousal, unlike the men with reflex erections,
in order to obtain and maintain their erection.
Oral phosphodiesterase-5 inhibitor (PDE5 in-
hibitors) medications such as Viagras (sildenafil),
Levitras (vardenafil), or Cialiss (tadalafil) are
used with efficacy in the spinal cord injured pop-
ulation. PDE5 inhibitors may make spontaneous
or reflex erections more frequent when the drug is
still active in the body. Clinically, men with sacral,
cauda equina or conus medullaris lesions do not
seem to respond as well to the PDE5 inhibitors.
Although not studied, this may be due to less ne-
urogenic nitric oxide release (a neurotransmitter
essential for penile smooth muscle relaxation) with
poorly mediated psychogenic erections in men with
flaccid paralysis as compared with the more robust
reflexogenic erections in men with spastic paralysis.
Granting the oversimplification, while the para-
sympathetic nervous system is primarily responsi-
ble for vasodilatation of the penile vasculature and
erection, and the sympathetic nervous system is
responsible for detumescence (loss of erection), it
appears that the sympathetic nervous system also
maintains erections after injury to parasympathet-
ic pathways (Chuang and Steers, 1999). This com-
pensatory mechanism brings with it another
potential clinical problem for the man dependent
on psychogenic erections; while he struggles to
maintain his erection through activation of
sexually arousing mental thought, he is activating
the fibers of the sympathetic chain, and such ad-
renergic stimulation will bring about the first stage
of ejaculation, seminal emission. This flow of
seminal emission results in detumescence. This
results in frustration and feeling of worthlessness
(Szasz, 1983).
Reflexogenic and psychogenic mechanisms
probably act synergistically to determine the erec-
tile response via a final common pathway involv-
ing a sacral parasympathetic route (Chuang and
Steers, 1999). Courtois et al. (1993) state that neu-
rologically, a spinal cord lesion between the two
erection centers should maintain both the reflex
and psychogenic erection potential. However, in
my clinical experience and that of others (Szasz,
1983), this specific lesion can result in the impor-
tant loss of cord communication between the two
centers, resulting in a poorer erection than expec-
ted. McKenna (1999) suggests that a strict division
between psychogenic and reflexogenic erections
may not be possible, since even erections generated
by higher neural activity may be facilitated by
sensory stimuli elicited by peripheral sexual arous-
al, and that loss of this positive feedback system
following spinal cord injury may explain why
erections are often not sustained and why ejacu-
lation often requires very strong stimuli.
Ejaculation and orgasm pose an even bigger
problem. In a large-scale study done at the Kinsey
Institute, only 12–15% of men with all levels of
spinal cord injury could ejaculate (Donohue and
Gebhard, 1995). The ejaculatory reflex, consisting
of seminal emission (sympathetic) and ejaculation
(parasympathetic and somatic) (Elliott, 2003) can
be disrupted by spinal cord injury, but the most
common problem is that of absence of ejaculation
or of triggering the reflex. Unlike reflex erections,
the triggering of ejaculation in a man with com-
plete upper motor neuron injury requires a specific
stimulus, and usually a sexual stimulus alone is not
adequate. Penile vibrostimulation increases the
chance of ejaculation if the thoracolumbar reflexes
are intact, but again, the vibrator needs to be spe-
cifically calibrated for the response to occur
(Sonksen et al., 1994). Fertility in men is there-
fore compromised due to erection and ejaculatory
problems. Fertility is also affected due to changes
in semen quality after spinal cord injury (Lin-
senmeyer and Perkash, 1991; Brackett et al., 1997;
Elliott, 2003).
Ejaculation is not necessarily associated with
erection, and erection itself may be variable

throughout the ejaculatory procedure in men with
spinal cord injury (Szasz and Carpenter, 1989;
Sonksen et al., 1994; Elliott, 2003). In sperm re-
trieval procedures, for example, the loss of erection
is not an indication to stop vibrostimulation, nor is
the presence of a strong erection a reliable predic-
tor of ejaculation (Szasz and Carpenter, 1989;
Elliott, 2003). This dissociation of erection and
ejaculation is also seen in other neurogenic condi-
tions, such as post-radical prostatectomy and mul-
tiple sclerosis. Despite difficulties with ejaculation,
up to 42% of men with various levels and com-
pleteness of spinal cord injury have reported or-
gasm (Alexander et al., 1993). Other researchers
(Talbot, 1949; Cole and Cole, 1981; Szasz, 1983)
have stated that their subjects were able to expe-
rience what they considered to be highly pleasur-
able orgasms by focusing on a sensate part of their
body and intensifying the sensation to be sexual,
or reassigning the sensation to their genitals. In the
Kinsey study (Donohue and Gebhard, 1995), al-
though almost all males prior to spinal cord injury
could attain orgasm with masturbation, just under
50% could attain orgasm after injury, and not
with reliability. It appeared that those with para-
plegia and those with incomplete injuries had a
better chance of experiencing orgasm than those
with quadriplegia and those with complete inju-
ries. Orgasm attainment also required more time
(Donohue and Gebhard, 1995). Tepper (Tepper
dissertation, website) discussed lived experiences
that could either impede or facilitate orgasm for
both men and women with spinal cord injury. The
role of the PDE5 inhibitor Levitras (vardenafil),
used successfully for erection enhancement in men
with spinal cord injury, was recently noted to
double the ejaculation rate in men with spinal cord
injury (19% with vardenafil vs. 10% for placebo)
(Giuliano, 2004). This was not noted in the orig-
inal studies with Viagras in men with cord injury
(Derry et al., 1998; Maytom et al., 1999).
From an autonomic point of view, the neuro-
physiology of orgasm is unclear. In men, although
both orgasm and ejaculation occur almost simul-
taneously, they are two distinct entities (Elliott,
2002) that can be separated by medical conditions.
Orgasm is traditionally described in urology texts
as the cerebral processing of afferent stimuli via
395
the pudendal nerve activated during phases of
emission and ejaculation (deGroat and Booth,
1980). It is associated with reversal of the physi-
ological changes occurring with the buildup of
sexual excitement, or the release of this sexual
tension (Klein, 1988). However, clinical examples
have shown that orgasm appears to be far more
complicated. For example, orgasm can occur in
men with spinal cord injury who are not capable of
ejaculation, and altered capacities for orgasm seem
possible in some spinal cord injured men and
women through stimulation of non-genital or oth-
er sensory inputs (Elliott, 2003). Orgasm could be
a genitally based, learned reflex as Sipski suggests
from her research (see Sipski and Arenas, this vol-
ume), an ‘‘efferent’’ cerebral interpretation of ei-
ther centrally generated activity or of sensory
input arriving from a different neural system, such
as the vagus nerve (Komisaruk and Whipple,
1991) or other typical genital or non-genital sourc-
es (Bach-y-Rita, 1999). Orgasm is likely a combi-
nation of a genitally, reflex-based component (that
improves with reinforcement) and non-genital or
non-pudendal nerve-dependent components that
may play a larger, compensatory role when the
reflex component of orgasm is impaired.
Therefore, although an erection can be provided
through safe and effective erection enhancement
methods – a big breakthrough – not much can be
done to provide feelings to the penis in order for
the man to fully experience touch, intercourse and
genital orgasm. Nor can his urine loss with genital
stimulation and ejaculation be eliminated without
drastic measures, or his risk of autonomic dysre-
flexia with ejaculation be removed. Fertility op-
tions for men with spinal cord injury have had the
biggest improvement over the last ten years with
effective sperm retrieval methods and the refine-
ment of in vitro fertilization and intracytoplasmic
sperm injection. However, we cannot reverse his
ejaculatory dysfunction.
Sexual consequences: the example of autonomic
dysreflexia
Autonomic dysreflexia is a response to a noxious
or non-noxious stimulus to the body below the
level of the injury, resulting in symptoms such as
396
headache, nausea, sweating above the level of in-
jury and severe hypertension that could lead to
sequelae such as stroke and death (Teasell et al.,
2000). By definition, autonomic dysreflexia is only
recognized in the neurologically disabled popula-
tion, such as those after spinal cord injury or with
multiple sclerosis (Bateman and Goldish, 2002).
Sexual arousal, a phenomenon of vasoconges-
tion and neuromuscular tension (Masters and
Johnson, 1966) resulting in penile erection in
males or vaginal lubrication and accommodation
in females, has a cardiovascular component. Blood
pressure increases with arousal and tends to peak
with the advent of genital orgasm. In one study of
able-bodied healthy males (Nemec et al., 1976),
mean blood pressure at orgasm was in the range of
161/77 mmHg–163/81 mmHg depending on the
sexual position. The risk of cardiac complications
during sex is minimal in the able-bodied popula-
tion (Drory, 2002), due to the briefness of the car-
diovascular stress and the intact compensatory
mechanisms for blood pressure control. Whereas
autonomic dysreflexia during sexual activity or
sperm retrieval procedures is usually short-lived
due to the causative stimuli being removed, occa-
sionally a worsening and or prolongation of the
dysreflexia induced by penile vibrostimulation has
been observed after this stimulation is removed.
For example, at the Vancouver Sperm Retrieval
Clinic, one incomplete quadriplegic (ASIA C) who
is often symptomatically hypotensive (blood pres-
sure approximately 60/40 mmHg) states that, at
ejaculation induced by vibrostimulation, he feels
an intense adrenaline rush leading to an out-
of-body experience, almost like ‘‘death’’. His
blood pressure during ejaculation on several oc-
casions has been consistently recorded at greater
than 200 mmHg systolic with a diastolic pressure
of over 100 mmHg. Over the next 10 min after
ejaculation, his blood pressure fluctuates around
three times his resting value, until it settles at
140–160 mmHg systolic within 20 min post-ejacu-
lation. Understandably, this atypically prolonged
autonomic dysreflexia makes him highly anxious
about repeating the process. On the other hand,
silent dysreflexia, a condition in which systolic
blood pressure increases by at least 20 mmHg in
the absence of any subjective symptoms, can also
occur during such ejaculatory procedures (Elliott
et al., 2005).
Clinically, various adjustments to the severity
and/or personal interpretation of autonomic dys-
reflexia symptoms can occur as time progresses.
Some persons have ceased sexual activity because
of it, others have adapted to the discomfort, and
some have incorporated it as a positive aspect of
sexual activity. Others have noted a drop in sub-
jective awareness of dysreflexia with continued
sexual activity (primarily with repeated ejacula-
tion), even to the point of dysreflexia becoming
symptomatically ‘‘silent’’. Silent autonomic dysre-
flexia is a recognized phenomenon in voiding and
bowel procedures (Linsenmeyer and Perkash,
1991; Kirshbium et al., 2002) and during sperm
retrieval procedures. Sexual activity has also pro-
voked atypical dysreflexia termed ‘‘malignant’’
(Elliott and Krassioukov, 2005) lasting for days
beyond the sexual experience.
Descriptions of autonomic dysreflexia and sex-
ual activity from patients with cord injury have
included both positive and negative consequences.
For some, the rising feeling of headache, facial
warmth and discomfort or nausea is anti-sexual. If
there is no accompanying orgasm or other sexual
payoff, a pounding headache from dysreflexia
leads to avoidance of high arousal or ejaculation,
except when medically required (i.e., fertility pro-
cedures). For example, one woman who had a very
active and rewarding sexual life pre-injury stated
she felt a tremendous loss at not being able to at-
tain high arousal without autonomic dysreflexia
making her sweaty and nauseated, and turning her
and her partner off. Despite this, she was persistent
in being sexual, determined to regain some sem-
blance of her past; sex remained important to her.
Other patients have stated that they have learned
to use their particular experience of autonomic
dysreflexia to enhance or extend the sexual feelings
they were already experiencing. What was formerly
distressing was consciously enveloped into the over-
all sexual arousal, and eventually interpreted as pos-
itive and sexually heightening. These anecdotal
remarks have come primarily from men with in-
complete spinal cord injury who can ejaculate. The
severity of subjective symptoms of autonomic
dysreflexia (headache, facial warmth and discomfort

or nausea) has also been noted to decrease with re-
peated ejaculations in some men at our clinic. These
men who are long-term patients are now able to
ejaculate with very little symptoms of dysreflexia.
For example, symptoms of headache are either ab-
sent or are substantially reduced in duration.
Our current research in ejaculation-induced au-
tonomic dysreflexia demonstrates that subjective
symptoms do not predict or reflect objective blood
pressure readings, and that cardiac arrhythmias
can occur at this time (Claydon et al., 2005; Sheel
et al., 2005). We have confirmed the established
knowledge that men with cervical injuries are more
likely to experience autonomic dysreflexia during
vibrostimulation than are men with thoracic inju-
ries. However, an important clinical point is that
symptoms are not predictive of severity of blood
pressure rise, and so cannot be used to estimate
safety of sexual practices or clinical procedures.
Conclusions
In conclusion, just as human spinal cord injury is
the model for the neurophysiology of sexual spinal
reflexes, human sexuality after spinal cord injury
can be the model for integrated autonomic func-
tion. Within the framework of a head-to-toe au-
tonomic approach to sexuality, multiple
autonomic systems could be investigated. Ejacu-
lation also provides a good predictive model for
observable autonomic dysreflexia and how it
changes over time with repeated challenges. Over-
all, the uniqueness of sexual rehabilitation lies in
its potential to grow even when the motor and
sensory recovery has met its limits.
It is unfortunate that the time devoted to re-
search or clinical rehabilitation in the sexual and
fertility area is not reflective of the need, since sex
is clearly a priority to men and women following
spinal cord injury. Researchers and clinicians in-
terested in sexual and fertility rehabilitation can
find an area in autonomic dysfunction where they
feel they could make a difference. To paraphrase
Anderson (2004), basic science researchers trained
in physiological function need to appreciate that
any improvement their work leads to in quality of
life for persons with spinal cord injury, results in
397
major psychological, social and economic reper-
cussions. Knowing that autonomic function lies at
the basis of sexuality and sexual ability should
hopefully spur research in this area and add to the
very important downstream effects that are fun-
damental to human happiness.
Acknowledgments
International Collaboration of Repair Discoveries
(ICORD), and its Director, John Steeves, PhD, for
his research support and encouragement, The Rick
Hansen Institute for their funding of research in
sexuality over the last ten years, collaborator And-
rei Krassioukov, MD, PhD, research colleagues
(in particular Marci Ekland RN, BSN, CRRN and
Kate McBride RN, BSN, CRRN), clinical co-di-
rector Mark Nigro MD, FRCP(C) and mentor,
Professor Emeritus George Szasz, MD.
References
Abouleish, E. (1980) Hypertension in a paraplegic parturient.
Anesthesiology, 53: 348–349.
Alexander, C.J., Sipski, M.L. and Findley, T.W. (1993) Sexual
activities, desire and satisfaction in males pre- and post-spinal
cord injury. Arch. Sex. Behav., 22(3): 217–228.
Anderson, K.D. (2004) Targeting recovery: Priorities of the spinal
cord-injured population. J. Neurotrauma, 21(10): 1371–1383.
Bach-y-Rita, P. (1999) Theoretical aspects of sensory substitu-
tion and of neurotransmission-related reorganization in spi-
nal cord injury. Spinal Cord, 37: 465–474.
Bateman, A.M. and Goldish, G.D. (2002) Autonomic dysre-
flexia in multiple sclerosis. J. Spinal Cord Med., 25(1): 40–42.
Berkman, A.H., Weissman, R. and Friedlich, M.H. (1983) Sex-
ual adjustment of spinal cord injured veterans living in the
community. Arch. Phys. Med. Rehabil., 64: 47–52.
Brackett, N.L., Padron, O.F. and Lynne, C.M. (1997) Semen
quality of spinal cord injured men is better when obtained by
vibratory stimulation versus electroejaculation. J. Urol., 157:
151.
Chuang, A.T. and Steers, W.D. (1999) Neurophysiology of pe-
nile erection. In: Carson C., Kirby R. and Goldstein I. (Eds.),
Textbook of Erection Dysfunction. ISIS Medical Media Inc.,
Oxford, UK, pp. 59–72.
Claydon, V., Elliott, S., Sheel, A.W. and Krassioukov, A.V.
(2005) Incidence of cardiac dysrhythmias during sperm re-
trieval in men with SCI. Proceedings of American Spine In-
jury Association. The ASIA abstracts were published in
Journal of Spinal Cord Medicine, American Paraplegic
Association (United States).
398
Cole, T. and Cole, S. (1981) Sexual attitude reassessment
programs for spinal cord injured adults. Their partners and
health care professionals. In: Sha’ked A. (Ed.), Human
Sexual and Rehabilitative Medicine. Sexual functioning after
Spinal Cord Injury. Williams and Wilkins, Baltimore, pp.
80–90.
Cole, T.M. (1975) Sexuality and the spinal cord injured. In:
Green R. (Ed.), Human Sexuality: A Health Practitioner’s
Text. Williams and Wilkins, Baltimore, pp. 243–263.
Cole, T.M., Chilgrem, R. and Rosenberg, P. (1973) A new
programme of sex education and counseling for spinal cord
injured adults and health care professionals. Paraplegia, 11:
111–124.
Comarr, A.E. and Vigue, M. (1978) Sexual counseling among
male and female patients with spinal cord injury and/or
cauda equina injury. Am. J. Phys. Med., 57(Part 1): 107–122.
Courtois, F.J., Charvier, K.F., Leriche, A. and Raymond, D.P.
(1993) Sexual function in spinal injury men I. Assessing sex-
ual capacity. Paraplegia, 31: 771–784.
Courtois, F.J., Geoffrion, R., Landry, E. and Belanger, M.H.
(2004) Reflex and physiologic measures of ejaculation in men
with spinal cord injury. Arch. Phys. Med. Rehab., 85:
910–918.
Dannels, A. and Charlifue, S. (2004) The perimenopause expe-
rience for women with spinal cord injury. SCI Nurs., 21(1):
9–13.
deGroat, W.C. and Booth, A.M. (1980) Physiology of male
sexual function. Ann. Intern. Med., 92(Pt 2): 329–331.
Derry, F.A., Dinsmore, W.W., Fraser, M., Gardner, B.P.,
Glass, G.A., Maytom, M.C. and Smith, M.D. (1998) Efficacy
and safety of oral sildenafil (viagra) in men with erectile dys-
function caused by spinal cord injury. Neurology, 51:
129–163.
Donohue, J. and Gebhard, P. (1995) The Kinsey Institute/In-
diana University report on sexual and spinal cord injury. Sex
Disabil., 13(1): 7–84.
Drory, Y. (2002) Sexual activity and cardiovascular risk. Eur.
Heart J., 4(Suppl. H): 413–418.
Ekland, M. and Lawrie, B. (2004) How a woman’s sexual ad-
justment after sustaining a spinal cord injury impacts sexual
health interventions. SCI Nurs., 21(1): 14–19.
Elliott, S. (2002) Ejaculation and orgasm: Sexuality in men with
SCI. Top Spinal Cord Inj. Rehabil., 8(1): 1–15.
Elliott, S. (2003) Sexual dysfunction and infertility in men with
spinal cord disorders. In: Lin V. (Ed.), Spinal Cord Medicine:
Principles and Practice. Demos Medical Publishing, New
York, pp. 350–365.
Elliott, S.L. and Krassioukov, A. (2005) Malignant autonomic
dysreflexia following ejaculation in spinal cord injured men.
Spinal Cord, Stockholm Press, England.
Elliott, S.L., McBride, K., Ekland, M. and Krassioukov, A.
(2005) Characteristics of autonomic dysreflexia in men with
SCI undergoing 296 procedures of vibrostimulation for
sperm retrieval. Proceedings of American Spine Injury As-
sociation. The ASIA abstracts were published in Journal of
Spinal Cord Medicine, American Paraplegic Association
(United States).
Ferrerio-Velasco, M.E., Barca-Buyo, A., Salvador DeLaBarr-
era, S., Monto, M., Arques, A., Miguens Vazquez, X. and
Rodriguez-Sotillo, A. (2005) Sexual issues in a sample of
women with spinal cord injury. Spinal Cord, Advance on
Line Publication. Spinal Cord, 43(1): 51–55.
Giuliano, F. (2004) Vardenafil in challenging – to - treat pa-
tients: The RESPITE study Presented symposium 11th
World Congress of the International Society for Sexual and
Impotence Research. Buenos Aires, Argentina.
Halstead, L.S., Seager, S.W., Houston, J.M., Whitesell, K.,
Dennis, M. and Nance, P.W. (1993) Relief of spasticity in
men and woman using rectal probe stimulation. Paraplegia,
31: 715–721.
Jackson, A.B. and Wadley, V. (1999) A multicenter study of
women’s self-reported reproductive health after spinal cord
injury. Arch. Phys. Med. Rehabil., 80: 1420–1428.
Kirshbium, S.C., House, J.G. and O’Connor, K.C. (2002) Silent
autonomic dysreflexia during a routine bowel program in
persons with traumatic spinal cord injury: A preliminary
study. Arch. Phys. Med. Rehabil., 83: 1774–1776.
Klein, E.A. (1988) The anatomy and physiology of normal male
sexual function. In: Montague D. (Ed.), Disorders of Male
Sexual Function. Year Book Medical Publishers, Inc.,
Chicago, USA, pp. 2–19.
Komisaruk, B.R. and Whipple, B. (1991) Physiological and
perceptual correlates of orgasm produced by genital or non-
genital stimulation. In: Kothari P. (Ed.), The Proceedings of
the First International Conference on Orgasm, India. VRP
Publishers, Bombay, pp. 69–73.
Laessoe, L., Nielsen, J.B., Biering-Sorensen, F. and Sonksen, J.
(2004) Antispastic effect of penile vibration in men with spi-
nal cord injury. Arch. Phys. Med. Rehabil., 85: 919–924.
Linsenmeyer, T.A. and Perkash, I. (1991) Infertility in men
with spinal cord injury. Arch. Phys. Med. Rehab., 72:
747–754.
Masters, W.H. and Johnson, V.E. (1966) Human Sexual
Response. Little Brown and Co, Boston, pp. 174–176.
Maytom, M.C., Derry, F.A., Dinsmore, W.W., Glass, C.A.,
Smith, M.D., Orr, M. and Osterloh, I.H. (1999) A two-part
pilot study of sildenafil (Viagra) in men with erectile dys-
function caused by spinal cord injury. Spinal Cord, 37:
110–116.
McGregor, J.A. and Meeuwsen, J. (1985) Autonomic hype-
rreflexia: A mortal danger for spinal cord-damaged women in
labor. Am. J. Obstet. Gynecol., 151(3): 330–333.
McKenna, K. (1999) The brain is the master organ in sexual
function: Central nervous system control of male and female
sexual function. Int. J. Impot. Res., 11(Suppl. 1): S48–S55.
Nemec, E.D., Mansfield, L. and Kennedy, J.W. (1976) Heart
rate and blood pressure responses during sexual activity in
normal males. Am. Heart J., 92(3): 274–277.
Pereira, L. (2003) Obstetric management of the patient with
spinal cord injury. Obstet. Gynecol. Surv., 58(10): 678–687.
Pope, C.S., Markenson, G.R., Bayer-Zwirello, L.A. and
Maissel, G.S. (2001) Pregnancy complicated by chronic
spinal cord injury and history of autonomic hyperreflexia.
Obstet. Gynecol., 97(5): 802–803.

Reitz, A., Tobe, V., Knapp, P.A. and Schurch, B. (2004)
Impact of spinal cord injury on sexual health and quality of
life. Int. J. Impotence Res., 16: 167–174.
Schmidt, M.H. and Schmidt, H.S. (1993) The ischiocavernosus
and bulbospongiosus muscles in mammalian penile rigidity.
Sleep, 16: 171–183.
Sheel, A.W., Krassioukov, A.V., Inglis, J.T. and Elliott, S.L.
(2005) Autonomic dysreflexia during sperm retrieval in spinal
cord Injury: Influence of lesion level and sildenafil citrate. J.
Appl. Physiol., 99(July): 53–58.
Siosteen, A., Lundqvist, C., Blomstrand, C., Sullivan, L. and
Sullivan, M. (1990) Sexual ability, activity, attitudes and
satisfaction as art of adjustment in spinal cord injured
subjects. Paraplegia, 28: 285–295.
Sipski, M.L., Alexander, C.J. and Rosen, R.C. (1995) Orgasm
in women with spinal cord injuries: A laboratory based
assessment. Arch. Phys. Med. Rehab., 76: 811–818.
Sonksen, J., Biering-Sorensen, F. and Kristensen, J.K. (1994)
Ejaculation by penile vibratory stimulation in men with spi-
nal cord injured. Paraplegia, 32: 651–660.
Szasz, G. (1983) Sexual health care. In: Zejdlik C. (Ed.),
Management of the Spinal Cord Injured, Wadsworth Health
Sciences Division. Monterey, California, pp. 125–152.
399
Szasz, G. (1986) Sexual function in the spinal cord injured. In:
Bloch R.F. and Basbaum M. (Eds.), Management of Spinal
Cord Injuries. Williams and Wilkins, Baltimore.
Szasz, G. and Carpenter, C. (1989) Clinical observations in vi-
bratory stimulation of the penis in men with spinal cord in-
jury. Arch. Sex. Behav., 18: 461–474.
Talbot, H. (1949) A report on sexual function in paraplegics.
J. Urol., 61: 265–270.
Teasell, R.W., Arnold, J.M. and Krassioukov, A. (2000)
Cardiovascular consequences of loss of supraspinal control
of the sympathetic nervous system after spinal cord injury.
Arch. Phys. Med. Rehabil., 81(4): 506–516.
Tepper, M.S., Whipple, B., Richard, E. and Komisaruk, B.R.
(2001) Women with complete spinal cord injury: A phen-
omenological study of sexual experiences. J. Sex Marital
Ther., 27: 615–623.
White, M.J., Rintala, D.H., Hart, K.A. and Fuhrer, M.J. (1993)
Sexual activities, concerns and interests of women with spinal
cord injury living in the Community. Am. J. Phys. Med.
Rehabil., 72(6): 372–378.
Yarkony, G.M. and Chen, D. (1995) Sexuality in patients with
spinal cord injury. Phys. Med. Rehab., 9(2): 325–343.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Published by Elsevier B.V.

CHAPTER 27

Ascending spinal pathways from sexual organs:

effects of chronic spinal lesions

Charles H. Hubscher

Department of Anatomical Sciences and Neurobiology, University of Louisville School of Medicine, Louisville,
KY 40292, USA

Abstract: A recent survey of paraplegics indicates that regaining sexual function is of the highest priority
for both males and females (Anderson, K.D. (2004) Targeting recovery: priorities of the spinal cord-injured
population J. Newrotrauma, 21: 1371–1383). Our understanding of the neural pathways and mechanisms
underlying sexual behavior and function is limited at the present time. More studies are obviously needed to
direct experiments geared toward developing effective therapeutic interventions. In this chapter, a review of
studies on the processing of sensory inputs from the male and female reproductive organs is presented with
a review of what is known about the location of ascending spinal pathways conveying this information. The
effect of spinal cord injury on sexual function and the problems that ensue are discussed.

Afferent innervation of the male and female
reproductive organs
Compared to the voluminous literature on the
neural pathways involved in the sensation and
motor control of the limbs, there have been
relatively few studies on the male and female
reproductive organs. This may partly be due to the
complexity of the neural pathways involved in
sexual behavior and functions, which include
somatic, sympathetic, and parasympathetic nerves
integrated with two interconnected portions of the
spinal cord, which in turn make connections with
the brainstem and cerebral cortex (de Groat and
Booth, 1980, 1993; de Groat et al., 1981; deGroat
and Steers, 1988; Berkley and Hubscher, 1995b;
Giuliano et al., 1995; McKenna, 2000). The main
focus of this review is on experiments that use
rodents, since the majority of studies on sexual
function use rats to model humans.
Corresponding author. Tel.: +(502)852-3058;
Fax: +(502)852-6228; E-mail: chhubs01@louisville.edu
Peripheral structures of primary importance for
reproduction include pelvic visceral organs, the
external genitalia and somatic elements of the
hindquarters. The internal pelvic visceral organs
involved in reproduction include the ovaries,
oviduct, uterus, cervix and vagina in females and
the testis, epididymis, vas deferens, seminal vesicle,
ejaculatory duct and prostate in males. These
organs are innervated by visceral afferent fibers,
which consist primarily of unmyelinated and some
thinly myelinated fibers that convey information
centrally mainly via the hypogastric and pelvic
nerves. Organ-specific and organ-characteristic
information is conveyed in a rostro-caudal topo-
graphic array to the caudal spinal cord (Berkley
and Hubscher, 1995b). The external genitalia
(clitoris/penis), skin and muscles of the perineum
are innervated by somatic myelinated and un-
myelinated afferent fibers of all sizes. These fibers
convey information to the lower lumbar and upper
sacral segments of the spinal cord mainly via the
pudendal nerve, but also via branches of the pelvic,
genitofemoral, ilioinguinal and anococcygeal
nerves (Peters et al., 1987; Bonica, 1990).

DOI: 10.1016/S0079-6123(05)52027-2 401

402
Hypogastric nerve
The hypogastric nerve contains both afferent and
efferent fibers. Removal of one axonal component
(e.g. dorsal root ganglionectomy from the 12th
thoracic (T12) to 2nd sacral (S2) ganglia to remove
sensory fibers) demonstrated that most of the
axonal population is efferent (92%) (Hulsebosch
and Coggeshall, 1982). Cell bodies of hypogastric
nerve afferents have been identified (using the
transganglionic transport of horseradish per-
oxidase) from T10 to the 4th lumbar (L4)
ganglion (Neuhuber, 1982). In females, a variety
of tracing techniques and electrophysiological
recordings have been used to demonstrate that
the T13–L3 (predominantly) dorsal root ganglion,
via the hypogastric nerve (Nance et al., 1988),
provide afferent innervation to the cervix, uterus
and broad ligament (Peters et al., 1987; Berkley
et al., 1988, 1993a; Nance et al., 1988). Sensory
axons in the hypogastric nerve of male and female
adult rats (with cell bodies concentrated in the L1
and L2 dorsal root ganglia) are similar in number,
110 and 148, respectively. This is not the case
for efferent hypogastric fibers. In male rats, an
average of 415 sympathetic preganglionic neurons
were found, while in females, only 110 were found
(Nadelhaft and McKenna, 1987).
The reproductive structures innervated by the
hypogastric nerve are involved in contraction and
movement of substances. In the male, stimulation
of the hypogastric nerve produces contractions of
the distal vas deferens and seminal vesicle but
not erection (Quinlan et al., 1989). Thus, the
hypogastric nerve is believed to play a role in
sperm motility (ejaculation). Bilateral sectioning of
the hypogastric nerve has no effect on mating be-
havior (Larsson and Swedin, 1971). These findings
are consistent with those of Bacq (1930) who
showed that sympathetic denervation affects sem-
inal discharge but not mounts and intromissions.
In the female, stimulation of the hypogastric nerve
produces contractions of the uterus (Sato et al.,
1989). Hypogastric afferents in the female rat
convey information about intense stimulation
(mechanical and chemical) to the central nervous
system (Berkley et al., 1993c). Intense mechanical
or chemical stimulation of the uterus or cervix is
necessary to evoke a response in the hypogastric
nerve (Berkley et al., 1987, 1993c). Behavioral
studies also show that rats will engage in an escape
behavior only at noxious levels of uterine disten-
sion, and that this behavior is eliminated after
bilateral hypogastric neurectomy (Berkley et al.,
1995b; Temple et al., 1999). The noxious level
is equivalent to the distension volume at which
ischemia of the uterine vessels is produced. Blood
vessels in the uterus of the non-pregnant rat have
been shown to be highly innervated (Garfield,
1986; Haase et al., 1997). Uterine distension late in
pregnancy leads to hormonal changes, which ac-
count for the initiation of delivery and onset of
maternal behavior (Graber and Kristal, 1977).
Whether or not this nerve serves a similar role in
males (re-nociception) is not known.
Pelvic nerve
The pelvic nerve in the male rat contains both
afferent (34%) and efferent (66%) fibers with the
majority of the myelinated axons and slightly less
than half of the unmyelinated axons being sensory
(Hulsebosch and Coggeshall, 1982). Cell bodies of
pelvic nerve afferents (95%) are located in the
L6 and S1 dorsal root ganglia (Nadelhaft and
Booth, 1984). In the female rat, the pelvic nerve
was shown to bifurcate into a viscero-cutaneous
branch and a muscular branch shortly after
separating from the lumbosacral trunk (Pacheco
et al., 1989), with both branches conveying sensory
and motor information. In females, afferent
innervation of the cervical portion of the uterus,
vagina, base of the bladder and rectum via the
pelvic nerve was demonstrated with HRP injec-
tions of the L6 and S1 dorsal root ganglia (Nance
et al., 1988). Electrophysiological studies showed
that pelvic nerve fibers have receptive fields
extending from the uterine cervix to the distal
end of the vagina (Berkley et al., 1990). Sensory
input from the midline perineal skin and viscera
travel through the viscerocutaneous branch and
from the iliococcygeus and pubococcygeus muscles
through the muscular branch (Pacheco et al.,
1989).
Stimulation of the pelvic nerve produces penile
erection and contractions of the distal vas deferens

and seminal vesicle in the male (Quinlan et al.,
1989) and produces displacement of the vaginal
wall (Pacheco et al., 1989) and contractions of the
uterus in the female (Sato et al., 1989). Although
there is a substantial amount of information
available regarding the function of the pelvic nerve
in the female rat, less experimentation has been
done in the male. In females, pelvic afferents serve
important informative, affective and behavioral
functions during mating, conception and parturi-
tion. The pelvic nerve, which has both internal and
external sensory fields, conveys information that is
considerably different from the hypogastric nerve.
The external sensory field is involved in triggering
lordosis and may play a role in positioning for
intromission, whereas the internal fields are
sensitive to vaginal stimulation, especially at the
cervical end of the vagina (Peters et al., 1987).
Responses to noxious pinch stimulation can
be attenuated by probing the vaginal cervix
(Komisaruk and Wallman, 1977). It is hypothe-
sized that this attenuation may be especially sig-
nificant during parturition, because it may reduce
stress, which would otherwise interfere with ma-
ternal behavior. Bilateral sectioning of the pelvic
nerve renders the entire intravaginal mucosa com-
pletely insensitive to gentle and intense mechanical
stimulation, and prevents pseudopregnancy (arrest
of estrus) upon intense mechanical stimulation of
the cervix (Kollar, 1953; Carlson and De Feo,
1965; Ross et al., 1979). Prolongation of the
process of parturition is seen in bilaterally pelvic
neurectomized rats due to the elimination of the
fetus-expulsion reflex (contraction of abdominal
muscles and diaphragm) in response to mechanical
stimulation of the upper vagina and cervix
(Higuchi et al., 1987). Only the sensory branch
of the pelvic nerve is required for normal vaginal
delivery in the rat (Burden et al., 1990). Pelvic
neurectomy also eliminates the lordosis response
(Carlson and De Feo, 1965). Komisaruk and
Wallman (1977) hypothesize that immobilization
during lordosis could facilitate sperm transport
and consequently pregnancy. The lordosis reflex
has been studied in the male, but not in the context
of neural control [for example, effects of cholecys-
tokinin (Bloch et al., 1988); modulation by the main
olfactory system (Chateau and Aron, 1990)]. The
403
pelvic nerve has also been shown to be important
for the display of paced mating behaviors (Erskine,
1992). Mating pattern (position, contact) is affected
in cervically denervated females (Diakow, 1970).
Pudendal nerve
The pudendal nerve in the male and female rat,
which shares its origin with the pelvic nerve, runs
from the L6–S1 trunk to the sacral plexus and
contains fibers that arise from the L6–S1 trunk as
well as the lumbosacral trunk (Reiner et al., 1981;
McKenna and Nadelhaft, 1986; Pacheco et al.,
1997). Revised nomenclature of the pudendal
nerve in the male rat includes three components:
a sensory branch (dorsal nerve of the penis
(Calaresu, 1970; Nunez et al., 1986)) that supplies
the penis and prepuce, a motor branch that supplies
the perineal muscles (McKenna and Nadelhaft,
1986) and cutaneous branches that supply the
scrotum (Pacheco et al., 1997). The glans penis has
an abundance of free nerve endings, and the pri-
mary afferent population in the dorsal nerve of the
penis contains both slowly and rapidly adapting
fibers (Kitchell et al., 1982; Johnson and Halata,
1991). In the female, afferents arising from the
clitoris and perigenital skin send information to
the L6 and S1 dorsal root ganglia through the pu-
dendal nerve sensory branch (McKenna and
Nadelhaft, 1986). A few of the dorsal root gangli-
on cells at L6 respond to electrical stimulation of
myelinated fibers in both the pudendal and sciatic
nerves (Pierau et al., 1982; Taylor et al., 1982),
demonstrating prespinal convergence of sensory
nerve fibers. The pudendal nerve in male and
female rats varies greatly, however, in both size
and number of afferent and efferent neurons. For
example, there are twice as many dorsal root
ganglion neurons from the pudendal nerve sensory
branch in the male rat than in the female
(McKenna and Nadelhaft, 1986) and in cross sec-
tion, the sensory branch is three times larger in males
(Moore and White, 1996). One well-known differ-
ence between the male and female rat, for example,
involves the spinal nucleus of the bulbocavernosus
muscle (Breedlove and Arnold, 1980). Adult
female rats lack both this spinal nucleus and its
target muscles (perineal bulbocavernosus). The
404
presence of the nucleus is dependent on the action
of androgens at an early stage of life (Breedlove
and Arnold, 1980; Breedlove, 1985).
In males, bilateral transection of the pudendal
nerves, affecting the dorsal nerve of the penis,
prevents erection, intromission, ejaculation, sexual
motivation and some (mounts and intromissions),
but not all (time-outs and mount-bout periods),
mating behaviors (Larsson and Sodersten, 1973;
Lodder and Zeilmaker, 1976). Stimulation of the
pudendal nerve does not, however, produce penile
erection (Quinlan et al., 1989). Desensitization of
the penis by local anesthesia also causes an
impairment of reflexogenic erection, intromission
and ejaculation (Hart and Leedy, 1985). In fe-
males, application of the local anesthetic lidocaine
to the clitoris (sensory innervation by pudendal
nerve) prevents the occurrence of intromission
patterns (Baum et al., 1974). When the pudendal
nerve is sectioned, there is a significant decrease in
lordosis, although this decrease can be overridden
by elevated estrogen levels (Kow and Pfaff, 1973;
Kow, 1976).
Ovarian/testicular nerves
Sensory fibers in the ovarian plexus nerves and the
superior ovarian nerves innervate the ovaries, ovi-
duct and rostral pole of the uterine horns (Burden
and Lawrence, 1978; Baljet and Drukker, 1979,
1980; Marchetti et al., 1987; Klein et al., 1989;
Berkley and Hubscher, 1995b; Serghini et al.,
1997). Dorsal root ganglion neurons from T10 to
L2 are labeled unilaterally following injection of
HRP into the ipsilateral ovary (Burden et al., 1983;
Nance et al., 1988). In males, the testis is supplied
by the superior spermatic nerve (Kumazawa,
1986). Most of the testicular afferents are poly-
modal, responding to mechanical, chemical and
thermal (noxious heat) stimulation (Kumazawa
et al., 1995). In dogs, HRP injection labels dorsal
root ganglia between T10 and L4, with most cells
labeled in L1 and L2 (Kumazawa et al., 1995).
Vagus nerve
In female rats, the innervation of the reproductive
tract by abdominal branches of the vagus includes
the ovaries, oviduct, the entire uterine horns and
cervix (Burden et al., 1983; Ortega-Villalobos
et al., 1990; Collins et al., 1999). Abdominal vago-
tomy disrupts the rat’s estrous cycle (Burden et al.,
1981). Whether or not the vagus innervates
portions of the male urogenital tract is unknown.
Central processing of inputs from the male and
female reproductive organs
Sensory information conveyed centrally from the
internal reproductive organs has important conse-
quences for both reproduction and for sensation
(from pleasure to pain). In females for example,
cervix stimulation in both rats and humans pro-
duces effects that are of considerable importance
for reproductive behaviors such as during mating
and parturition (Komisaruk and Whipple, 1988),
and for pain, such as is evident in women during
invasive gynecological procedures or for many,
with deep penile penetration during vaginal inter-
course (dyspareunia — affecting 10–15% of sex-
ually active women) (Meana et al., 1997).
Spinal processing
In males, the lumbosacral segments (L5–S1 in the
rat) contain a center for erection and the expulsive
part of ejaculation mediated by preganglionic
parasympathetic motor axons in the pelvic nerve
and somatic motor axons in the pudendal nerve
supplying the striated perineal muscles of the
pelvic floor. Stimulation of the dorsal nerve of
the penis induces Fos labeling in the dorsal horn,
dorsal gray commissure and sacral parasympa-
thetic nucleus (Rampin et al., 1997). Electrophys-
iological techniques have been used to investigate
single spinal cord interneurons in the dorsal horn
and intermediate zone of primarily L6–S1 that re-
ceive input from dorsal nerve of the penis afferents
(Johnson, 1989). All of the penile interneurons
exhibit receptive fields on the penis that are sig-
nificantly larger than the receptive fields for single
primary afferent neurons, thereby demonstrating
a central convergence of penile sensory input.
Almost all of the penile interneurons have recep-
tive fields on both sides of the body and their

response characteristics strongly suggest a mo-
nosynaptic input from both ipsilateral and cont-
ralateral dorsal nerve of the penis fibers. There is
also an extensive representation from the distal
glans (cup) region in these spinal cord interneu-
rons. Afferent fibers in the dorsal nerve of the
penis produce bilateral (crossed and uncrossed)
reflex facilitation of pudendal motoneurons locat-
ed in L5–L6. A lumbar (L3–L4) reflex region for
ejaculation has also been proposed (Truitt and
Coolen, 2002) that depends on input from afferent
systems releasing Substance P.
In females, information from the reproductive
organs is conveyed in a rostro-caudal topographic
array to the caudal spinal cord (Berkley et al.,
1993c; Berkley and Hubscher, 1995b). Electro-
physiological and anatomical studies indicate that
there is an extensive system of neurons in thoracic,
lumbar and sacral spinal cord that receive repro-
ductive organ input (Berkley et al., 1993b; Lee and
Erskine, 1996, 2000). The cervix, which is inner-
vated by both the hypogastric and pelvic nerves,
has inputs to dorsal horn neurons at all three lev-
els, although the neurons are concentrated vent-
rally in the dorsal horn at T13–L1 and throughout
the dorsal horn at L4–L5 and L6–S1 segments
(Berkley et al., 1993b). Cervix-responsive dorsal
horn neurons in both regions receive convergent
inputs from other pelvic organs as well as cutane-
ous regions, although the receptive fields tend to
be larger at T13–L1 and more confined to the
perineum for the neurons at L6-S2 that are located
in the dorsal part of the dorsal horn (Berkley et al.,
1993b). Many cervix-responsive neurons at L6–S2
respond to uterine distension by being inhibited.
This uterine input originates from distant roots
(uterus innervated by the hypogastric nerve), as
shown in experiments, where T13–L2 roots were
sectioned bilaterally (Wall et al., 1993). How these
interactions sculpt the actions of these neurons for
various aspects of reproduction is still unclear.
Supraspinal processing
Many regions throughout the brain receive input
from the reproductive organs. These regions in-
clude (but are not limited to) the nucleus tractus
solitarius, nucleus gracilis, nucleus reticularis
405
gigantocellularis, nucleus reticularis gigantocellularis
pars alpha, lateral paragigantocellular reticular
nucleus, raphe pallidus, raphe magnus, A5 nor-
adrenergic cell groups, parapyramidal area, lateral
vestibular nucleus, Barrington’s nucleus, nucleus
locus coeruleus, nucleus subcoeruleus, caudal pon-
tine reticular nucleus, periaqueductal gray, intra-
laminar thalamic nuclei, ventroposterolateral
thalamic nucleus, ventrolateral thalamic nucleus,
ventromedial thalamic nucleus, anterior thalamic
nuclei, nucleus submedius, subparafascicular nu-
cleus, medial preoptic area, lateral septum, bed
nucleus of the stria terminalis, ventromedial hypo-
thalamus, hypothalamic paraventricular nucleus,
medial amygdala, globus pallidus, mesencephalic
central gray and cortex (Allen et al., 1981;
Haldeman et al., 1982; Haskins and Moss, 1983;
Akaishi et al., 1988; Shen et al., 1990; Baum and
Everitt, 1992; Berkley et al., 1993a, 1995a; Marson
et al., 1993; Tetel et al., 1993; Hubscher and
Berkley, 1994; Berkley and Hubscher, 1995b;
Hubscher and Johnson, 1996, 2003; Pfaus et al.,
1996; Bradley et al., 1998; Papka et al., 1998;
Van der Horst and Holstege, 1998; Ding et al.,
1999; Lee and Erskine, 2000). Areas receiving
reproductive organ input such as those targeted by
Hubscher and colleagues, for example, which in-
clude the nucleus reticularis gigantocellularis and
surrounding nuclei within the medullary reticular
formation, the nucleus gracilis and solitarius, and
various subregions of the thalamus have also been
shown to process and relay a vast array of con-
vergent somatic and pelvic visceral sensory inputs
(Hubscher and Berkley, 1994; Berkley and
Hubscher, 1995a; Hubscher and Johnson, 1996,
2001, 2003). Our electrophysiological data from
extracellular single unit recordings in the rostral
ventromedial medulla in male rats has demon-
strated a significant degree of convergence from
the reproductive organs (dorsal nerve of penis;
pelvic nerve) and rostral skin areas (including the
ears and forepaws) (Hubscher and Johnson, 1996).
Examples from recordings obtained in a male and
female rat are provided in Fig. 1. The majority of
these neuronal responses were to noxious levels of
stimulation. Some neurons required windup (re-
petitive stimuli) to respond to bilateral dorsal nerve
of the penis (or clitoris) stimulation. In males,
406

it is hypothesized that these neurons may be firing
near the ejaculatory threshold (Hubscher and
Johnson, 1996), which depends on a gradual
build-up of activity produced by multiple intr-
omissions (Sachs and Meisel, 1988). Large lesions
that include the gigantocellular (ventral and pars
alpha) and lateral paragigantocellular nuclei have
been shown to affect ejaculatory bursts in perineal
muscles (Marson and McKenna, 1990).
In female rats, neurons in the nucleus reticularis
gigantocellularis and surrounding regions respond-
ing to probing of the vaginal canal (Hornby and
Rose, 1976; Hubscher and Johnson, 2001) have
been shown to be involved in female circuitry
responsible for lordosis behavior (Modianos and
Pfaff, 1979; Schwartz-Giblin et al., 1996; Daniels
et al., 1999). Although many of these neurons are
likely involved in mating, some likely play a role in
nociceptive processing as well (Bowsher, 1976;
Peschanski and Besson, 1984; Zhuo and Gebhart,
1991; Berkley et al., 1993a; Al-Chaer et al., 1996b;
Yang et al., 1998; Mason, 2001).
Ascending pathways
Limited information is available on the location of
ascending spinal pathways that convey information
originating from the internal reproductive organs
and external genitalia. The results of several stud-
ies, when taken together, indicate that there are
likely many spinal pathways that convey this input
to the brain, and these projections are likely all
bilateral. These pathways include most, if not
all, of the following: dorsal column, post-synaptic

Fig. 1. Example on the left shows excitatory responses in a male rat of a single neuron located in right n. reticularis gigantocellularis
pars alpha to uni- and bi-lateral stimulation of the pelvic nerve and dorsal nerve of the penis as well as stimulation of the abdominal
branch of the vagus, distention of the colon and mechanical stimulation of the penis. This neuron had excitatory responses to pinching
of the entire body, including the face, dorsal trunk, forepaw and ears (responses shown are only to pinch of the glans penis). Note that
the only low threshold responses obtained were from stroking of the penis (rest of body responded to noxious pinch). Example on the
right shows excitatory responses in a female rat of a single neuron located in left nucleus reticularis gigantocellularis to bilateral (b)
stimulation of the dorsal nerve of the clitoris (DNC) and pelvic nerve (PN), pressure on the cervix, distention of the left uterine horn
and distal colon. The same neuron also had excitatory responses to distention of the right uterine horn and bladder, and to bilateral
stimulation (pinching) of the perineum, anus, trunk, ears, toes of the hindpaw and forepaw (not shown). Note that this particular
neuron did not respond to distention of the vaginal canal (not shown); the vagus was not tested. For both neurons shown, note the
stimulus artifacts at the onset of the electrical nerve stimulation train (see arrows, which indicate the onset of the electrical stimulation
train). The horizontal bar indicates the duration of a natural stimulus. The two large dots with smaller dots in between indicate the
onset and end, respectively, of maximal balloon stimulus intensity (by vol.). Adapted from Hubscher et al. (2004b).

dorsal column, spinoreticular, spinothalamic, spino-
solitary, spinoparabrachial, spinohypothalamic,
spinoamygdalar, spinomesencephalic and spinocer-
ebellar pathways (Menetrey and de Pommery, 1991;
Berkley and Hubscher, 1995b).
In males, the most extensive data exist for
central projections of input from the penis. The
location of projections originating from the dorsal
nerve of the penis to the medullary reticular
formation neurons has been examined (Hubscher
and Johnson, 1999, 2004). The results from these
studies indicate that there are at least two central
projections originating from the male genitalia
located in the dorsal quadrant at the mid-thoracic
spinal level (conveyed rostrally from lumbosacral
dorsal horn cells to the medullary reticular forma-
tion). All responses to bilateral electrical stimula-
tion of the dorsal nerve of the penis were lost
following a complete dorsal hemisection. Specifi-
cally, responses to gentle stimulation of the penis
were lost following either an acute or chronic
mid-thoracic dorsal column lesion (bilateral).
Responses to pinching of the penis were not lost
following a complete dorsal column lesion or
following a subsequent unilateral lesion of the
dorsolateral quadrant, but were lost after both
dorsolateral quadrant’s had been lesioned. It
is important to note that, although a chronic
mid-thoracic spinal cord dorsal hemisection
eliminates medullary reticular formation neuronal
responses to bilateral dorsal nerve of the penis
stimulation (Hubscher and Johnson, 1999), this
electrical search stimulus activates dorsal nerve
of the penis afferents in the A-b and A-d range
only. Therefore, additional pathways likely exist
that convey, to the medullary reticular formation,
information originating from the large population
of C-fibers that are contained within the
dorsal nerve of the penis (Johnson and Halata,
1991). Candidates include spinothalamic or spin-
oreticulothalamic pathways in the ventrolateral
quadrant (Giesler et al., 1981; Peschanski and
Besson, 1984). Specific thalamic subregions relay
extensive inputs from the dorsal nerve of the
penis (Truitt and Coolen, 2002; Hubscher and
Johnson, 2003). Other thalamic subregions may be
associated with ejaculation (Truitt and Coolen,
2002).
407
At the present time, it is unclear whether these
projections reach the medullary reticular forma-
tion directly or indirectly via one or more synaptic
contacts in other regions of the brain. There is
little evidence in the literature that would strongly
support or refute either of the two possibilities.
For example, the location of direct spinoreticular
projections ascending in the white matter of the rat
spinal cord is unknown, with the exception of one
anatomical study that used large medullary injec-
tions of horseradish peroxidase to demonstrate the
loss of retrogradely labeled L4–L6 cells around the
central canal following a ventrolateral quadrant
lesion at T12 (Nahin et al., 1986). The evidence
from that study relative to Hubscher and Johnson
(1999) is inconclusive, since in Nahin et al. (i) the
lesions appeared to encroach upon the ventral
portion of the dorsolateral quadrant, (ii) the
lesions were at T12 (T7/T8 in Hubscher and
Johnson) and there is evidence for a dorsal shift
of axons as they ascend rostrally (Willis and
Coggeshall, 1991), and (iii) the study focused on
cells adjacent to the central canal, which are some
of many dorsal horn neurons to (a) project directly
upon neurons in the medullary reticular formation
(Menetrey et al., 1980; Chaouch et al., 1983) and
(b) respond to bilateral dorsal nerve of the penis
stimulation (Johnson, 1989). There is also evidence
for the existence of several different ascending
pathways within the dorsolateral quadrant of the
rat, any of which could provide indirect projec-
tions to medullary reticular formation and thus,
may convey pelvic and visceral information. Such
pathways include a spinomesencephalic pathway
and a spinohypothalamic pathway (Zemlan et al.,
1978; Menetrey et al., 1980; McMahon and Wall,
1983; Cliffer et al., 1991; Burstein et al., 1996;
Kostarczyk et al., 1997).
In addition, the medullary reticular formation
has been shown in the rat to be interconnected
with both the gracilis nucleus and the solitary nu-
cleus in the caudal brainstem (Tomasulo and
Emmers, 1972; Odutola, 1977; Jean, 1991; Mtui
et al., 1995). In females, neurons in the gracilis nu-
cleus have been shown to receive input (albeit in-
direct) from female reproductive organs (Hubscher,
1994; Berkley and Hubscher, 1995a). The dorsal
column-medial lemniscal pathway has therefore
408
been implicated as a source of input about
innocuous and noxious events from pelvic viscera
(uterus, cervix, vagina and colon) and skin to
lateral thalamus, suggesting that both the dorsal
column-medial lemniscal and spinothalamic
pathways are involved in pain (Hubscher, 1994;
Berkley and Hubscher, 1995a). The dorsal
columns have also been shown to convey visceral
nociceptive inputs from the colon (Al-Chaer et al.,
1996a, b; Willis et al., 1999) and pancreas
(Houghton et al., 1997). Ascending projections
from neurons in the area around the central canal,
an area known to receive primary afferent input
from somatic and pelvic visceral structures in the
cat (Honda, 1985), has been implicated in the
transmission of second order afferent information
for visceral nociception to the gracilis nucleus
(Wang et al., 1999).
The solitary nucleus has also been shown to
relay input (potentially noxious) from pelvic
visceral sources centrally (from the cervix, vagina
and uterus), via both a spino-solitary and
vagal-solitary pathway (Hubscher and Berkley,
1994, 1995; Komisaruk et al., 1996). Different
pathways were identified for different portions of
the reproductive tract, which involved electrophys-
iological recording of responses to pelvic organ
stimulation in the solitary nucleus, pre- and post-
combinations of acute spinal transection and bi-
lateral vagotomy (Hubscher and Berkley, 1995).
The results demonstrating a vagal-solitary projec-
tion from the uterus are consistent with anatom-
ical tracing and other types of experiments
(Ortega-Villalobos et al., 1990; Collins et al.,
1999; Guevara-Guzman et al., 2001). In addition,
although bilateral vagotomy had an effect on the
responsiveness of neurons in the solitary nucleus to
cervix/vaginal stimulation (suggestive of an ana-
tomical connection), the responses were only elim-
inated after a subsequent acute spinal transection
(Hubscher, 1994; Hubscher and Berkley, 1995).
Clinical implications of damage to these spinal
pathways: males versus females
In man, the degree of sexual dysfunction caused
by spinal cord injury depends on the level of the
lesion. In male patients with clinically complete
spinal cord injuries cranial to T9–10, for example,
the spinal reflex arcs for erection and ejaculation
are intact, although supraspinal input has been
removed (see Sachs and Bitran, 1990). In females,
having a spinal cord injury has important conse-
quences for pregnancy (Cross et al., 1992), such as
frequent urinary tract infections, and the location
of the spinal lesion has important consequences
for childbirth (Sauer and Harvey, 1993), since
damage above T5 can result in the life-threatening
complication of autonomic dysreflexia (Crosby
et al., 1992).
Sensation
Most male spinal cord-injured patients with intact
reflex arcs demonstrate reflexogenic erections of
varying degrees in response to very slight mechan-
ical stimulation of the penis (Sarkarati et al., 1987;
Ver Voort, 1987). Although these erections are
easily initiated, they are not easily sustained or
firm enough for vaginal penetration (Bodner et al.,
1987), which has been proposed to be a result of
altered penile sensitivity (Goldstein, 1988). Studies
show that central pathway impairment and not
peripheral pathways may play a role in erectile
dysfunction in diabetic patients (Sartucci et al.,
1999). In cases where the spinal reflex arc is dam-
aged (such as for conus medullaris or cauda equina
injury), sensations are completely absent or signif-
icantly diminished (Pavlakis et al., 1983).
Among women with spinal cord injuries, only
some cannot achieve orgasm, although most of
those with lower motor neuron dysfunction cannot
achieve orgasm (Sipski et al., 2001). For those
spinal cord-injured women who can achieve or-
gasm, time to orgasm significantly increases versus
able-bodied controls (Sipski et al., 2001). A survey
shows that following injury, the majority of wom-
en are satisfied with their sexual experiences
(Charlifue et al., 1992). Women with spinal cord
injuries can also perceive menstrual cramps (Axel,
1982). There is growing evidence of a nociceptive
vagal-solitary pathway from the vaginocervix
region in humans with clinically complete spinal
cord injuries (Komisaruk et al., 1997; Komisaruk

and Sansone, 2003; Komisaruk et al., 2004). Such
a pathway is consistent with some of the basic
scientific studies that have been done using rats
(Ortega-Villalobos et al., 1990; Hubscher and
Berkley, 1994, 1995; Komisaruk et al., 1996; Collins
et al., 1999). However, in pregnancy, perception of
fetal movement and uterine contractions does not
occur in women with injuries above T8 (Sauer and
Harvey, 1993).
Fertility
Infertility in males relates to impairments in ejac-
ulation and poor quality of semen once ejaculation
is achieved (Beretta et al., 1989; Linsenmeyer et al.,
1994; Sedor and Hirsch, 1995). In 95% of male
spinal patients with lesions cranial to T10, ejacu-
lation is severely impaired or impossible (Ver
Voort, 1987; Seftel et al., 1991). Despite the intact
spinal reflex arc, the segmental ejaculatory reflex
circuitry may be dependent on (1) supraspinal
facilitation or disinhibition and/or (2) the integrity
of a spino–bulbo–spinal loop for the coordination
of perineal muscle activities. Thus, any decrease in
penile sensation resulting from damage to ascend-
ing pathways would affect this ejaculatory circuit-
ry (mechanisms discussed in Chapter 28 by
Johnson in this book). There are a number of
techniques used to assist spinal cord-injured males
with fertility (Beretta et al., 1989; Chung et al.,
1995; Dahlberg et al., 1995; Nehra et al., 1996).
Many spinal cord-injured patients who do not re-
spond to normal tactile stimulation of the penis
may ejaculate to very intense vibratory stimulation
of the ventral penile midline (Sonksen and Ohl,
2002) suggesting that massive recruitment of all
low and high threshold penile mechanoreceptive
afferent neurons can provide enough input to the
spinal ejaculatory circuit. Quality of semen has
been shown to vary with different assisted ejacu-
lation procedures, with vibratory stimulation
being better than electroejaculation (Brackett
et al., 1997).
In females, although menses is initially altered
(on average 5 months in 60% of patients) follow-
ing spinal cord injury, cycling resumes to pre-
injury status (Comarr, 1966; Axel, 1982; Charlifue
409
et al., 1992; Westgren and Levi, 1994). A delay in
the reproductive cycle has also been found exper-
imentally in adult rats (44%) following contusion
injury at the mid-thoracic spinal level (Hubscher
et al., 2004a). Only a minority of spinal cord-in-
jured women report heavier, longer and irregular
periods post injury, although many report changes
in frequency and amount of vaginal discharge
(Charlifue et al., 1992). Thus, fertility is less of an
issue for females than males with spinal cord in-
jury. Although there are significant risks associat-
ed with pregnancy following spinal cord injury,
most women have full-term healthy babies (Cross
et al., 1991, 1992; Burns and Jackson, 2001) and
succeed in parenting despite their physical limita-
tions (Westgren and Levi, 1994; Alexander et al.,
2002).
References
Akaishi, T., Robbins, A., Sakuma, Y. and Sato, Y. (1988)
Neural inputs from the uterus to the paraventricular magno-
cellular neurons in the rat. Neurosci. Lett., 84: 57–62.
Al-Chaer, E.D., Lawand, N.B., Westlund, K.N. and Willis,
W.D. (1996a) Pelvic visceral input into the nucleus gracilis is
largely mediated by the postsynaptic dorsal column pathway.
J. Neurophysiol., 76: 2675–2690.
Al-Chaer, E.D., Lawand, N.B., Westlund, K.N. and Willis,
W.D. (1996b) Visceral nociceptive input into the ventral
posterolateral nucleus of the thalamus: a new function for the
dorsal column pathway. J. Neurophysiol., 76: 2661–2674.
Alexander, C.J., Hwang, K. and Sipski, M.L. (2002) Mothers
with spinal cord injuries: impact on marital, family, and
children’s adjustment. Arch. Phys. Med. Rehabil., 83: 24–30.
Allen, T.O., Adler, N.T., Greenberg, J.H. and Reivich, M.
(1981) Vaginocervical stimulation selectively increases meta-
bolic activity in the rat brain. Science, 211: 1070–1072.
Axel, S.J. (1982) Spinal cord injured women’s concerns: men-
struation and pregnancy. Rehabil. Nurs., 7: 10–15.
Bacq, Z.M. (1930) Impotence of the male rodent after sympa-
thetic denervation of the genital organs. In: From the Lab-
oratories of Physiology in the Harvard Medical School,
pp. 321–330.
Baljet, B. and Drukker, J. (1979) The extrinsic innervation of
the abdominal organs in the female rat. Acta Anat., 104:
243–267.
Baljet, B. and Drukker, J. (1980) The extrinsic innervation of
the pelvic organs in the female rat. Acta Anat., 107: 241–267.
Baum, M.J. and Everitt, B.J. (1992) Increased expression of c-
fos in the medial preoptic area after mating in male rats: role
of afferent inputs from the medial amygdala and midbrain
central tegmental field. J. Neurosci., 50: 627–646.
410
Baum, M.J., Sodersten, P. and Vreeburg, J.T. (1974) Mounting
and receptive behavior in the ovariectomized female rat: in-
fluence of estradiol, dihydrotestosterone, and genital an-
esthetization. Horm. Behav., 5: 175–190.
Beretta, G., Chelo, E. and Zanollo, A. (1989) Reproductive
aspects in spinal cord injured males. Paraplegia, 27: 113–118.
Berkley, K.J., Benoist, J.M., Gautron, M. and Guilbaud, G.
(1995a) Responses of neurons in the caudal intralaminar
thalamic complex of the rat to stimulation of the uterus, va-
gina, cervix, colon and skin. Brain Res., 695: 92–95.
Berkley, K.J., Guilbaud, G., Benoist, J.M. and Gautron, M.
(1993a) Responses of neurons in and near the thalamic vent-
robasal complex of the rat to stimulation of uterus, cervix,
vagina, colon and skin. J. Neurophysiol., 69: 557–568.
Berkley, K.J., Hotta, H., Robbins, A. and Sato, Y. (1990)
Functional properties of afferent fibers supplying reproduc-
tive and other pelvic organs in pelvic nerve of female rat. J.
Neurophysiol., 63: 256–272.
Berkley, K.J. and Hubscher, C.H. (1995a) Are there separate
central nervous system pathways for touch and pain? Nat.
Med., 1: 766–773.
Berkley, K.J. and Hubscher, C.H. (1995b) Visceral and somatic
sensory tracks through the neuraxis and their relation to pain:
lessons from the rat female reproductive system. In: Gebhart
G.F. (Ed.), Visceral Pain. IASP Press, Seattle, pp. 195–216.
Berkley, K.J., Hubscher, C.H. and Wall, P.D. (1993b) Neu-
ronal responses to stimulation of the cervix, uterus, colon and
skin in the rat spinal cord. J. Neurophysiol., 69: 545–556.
Berkley, K.J., Robbins, A. and Sato, Y. (1987) Uterine afferent
fibers in the rat. In: Schmidt R.F., Schaible H.-G. and Vahle-
Hinz C. (Eds.), Fine Afferent Nerve Fibers and Pain. VCH,
Weinheim, pp. 129–136.
Berkley, K.J., Robbins, A. and Sato, Y. (1988) Afferent fibers
supplying the uterus in the rat. J. Neurophysiol., 59: 142–163.
Berkley, K.J., Robbins, A. and Sato, Y. (1993c) Functional
differences between afferent fibers in the hypogastric and
pelvic nerves innervating female reproductive organs in the
rat. J. Neurophysiol., 69: 533–544.
Berkley, K.J., Wood, E., Scofield, S.L. and Little, M. (1995b)
Behavioral responses to uterine or vaginal distension in the
rat. Pain, 61: 121–131.
Bloch, G.J., Babcock, A.M., Gorski, R.A. and Micevych, P.E.
(1988) Effects of cholecystokinin on male copulatory behavior
and lordosis behavior in male rats. Physiol. Behav., 43: 351–357.
Bodner, D.R., Lindan, R., Leffler, E., Kursh, E.D. and
Resnick, M.I. (1987) The application of intracavernous
injection of vasoactive medications for erection in men with
spinal cord injury. J. Urol., 138: 310–311.
Bonica, J.J. (1990) General considerations of pain in the pelvis
and perineum. In: Bonica J.J. (Ed.), The Management of Pain
(2nd ed). Lea & Febiger, Philadelphia, PA, pp. 1283–1312.
Bowsher, D. (1976) Role of the reticular formation in responses
to noxious stimulation. Pain, 2: 361–378.
Brackett, N.L., Padron, O.F. and Lynne, C.M. (1997) Semen
quality of spinal cord injured men is better when obtained by
vibratory stimulation versus electroejaculation. J. Urol., 157:
151–157.
Bradley, W.E., Farrell, D.F. and Ojemann, G.A. (1998) Human
cerebrocortical potentials evoked by stimulation of the dorsal
nerve of the penis. Somatosens Mot. Res., 15: 118–127.
Breedlove, S.M. (1985) Hormonal control of the anatomical
specificity of motoneuron-to-muscle innervation in rats. Sci-
ence, 227: 1357–1359.
Breedlove, S.M. and Arnold, A.P. (1980) Hormone accumula-
tion in a sexually dimorphic motor nucleus of the rat spinal
cord. Science, 210: 564–566.
Burden, H.W. and Lawrence Jr., I.E. (1978) Experimental
studies on the acetylcholinesterase-positive nerves in the ova-
ry of the rat. Anat. Rec., 190: 233–241.
Burden, H.W., Lawrence Jr., I.E., Louis, T.M. and Hodson,
C.A. (1981) Effect of abdominal vagotomy on the estrous
cycle of the rat and the induction of pseudopregnancy. Ne-
uroendocrinology, 33: 218–222.
Burden, H.W., Leonard, M., Smith, C.P. and Lawrence Jr., I.E.
(1983) The sensory innervation of the ovary: a horseradish
peroxidase study in the rat. Anat. Rec., 207: 623–627.
Burden, H.W., Price, G.T., Renegar, R.H. and Hodson, C.A.
(1990) Effects of peripheral nerve lesions during pregnancy
on parturition in rats. Anat. Embryol., 182: 499–501.
Burns, A.S. and Jackson, A.B. (2001) Gynecologic and repro-
ductive issues in women with spinal cord injury. Phys. Med.
Rehabil. Clin. N. Am., 12: 183–199.
Burstein, R., Falkowsky, O., Borsook, D. and Strassman, A.
(1996) Distinct lateral and medial projections of the spin-
ohypothalamic tract of the rat. J. Comp. Neurol., 373:
549–574.
Calaresu, F.R. (1970) Experimental studies on the dorsal nerve
of the penis of the rat. Am. J. Anat., 127: 415–422.
Carlson, R.R. and De Feo, V.J. (1965) Role of the pelvic nerve
vs. the abdominal sympathetic nerves in the reproductive
function of the female rat. Endocrinology, 77: 1014–1022.
Chaouch, A., Menetrey, D., Binder, D. and Besson, J.M. (1983)
Neurons at the origin of the medial component of the bulb-
opontine spinoreticular tract in the rat: an anatomical study
using horseradish peroxidase retrograde transport. J. Comp.
Neurol., 214: 309–320.
Charlifue, S.W., Gerhart, K.A., Menter, R.R., Whiteneck,
G.G. and Manley, M.S. (1992) Sexual issues of women with
spinal cord injuries. Paraplegia, 30: 192–199.
Chateau, D. and Aron, C. (1990) Peripheral anosmia and
display of lordosis behaviour in the male rat. Behav. Proc., 22:
33–40.
Chung, P.H., Yeko, T.R., Mayer, J.C., Sanford, E.J. and
Maroulis, G.B. (1995) Assisted fertility using electroejaculation
in men with spinal cord injury—a review of literature. Fertil.
Steril., 64: 1–9.
Cliffer, K.D., Burstein, R. and Giesler Jr., G.J. (1991)
Distributions of spinothalamic, spinohypothalamic, and
spinotelencephalic fibers revealed by anterograde transport
of PHA-L in rats. J. Neurosci., 11: 852–868.
Collins, J.J., Lin, C.E., Berthoud, H.R. and Papka, R.E.
(1999) Vagal afferents from the uterus and cervix provide
direct connections to the brainstem. Cell Tissue Res., 295:
43–54.

Comarr, A.E. (1966) Observations on menstruation and preg-
nancy among female spinal cord injury patients. Paraplegia,
3: 263–272.
Crosby, E., St-Jean, B., Reid, D. and Elliott, R.D. (1992) Ob-
stetrical anaesthesia and analgesia in chronic spinal cord-in-
jured women. Can. J. Anaesth., 39: 487–494.
Cross, L.L., Meythaler, J.M., Tuel, S.M. and Cross, A.L.
(1991) Pregnancy following spinal cord injury. West J. Med.,
154: 607–611.
Cross, L.L., Meythaler, J.M., Tuel, S.M. and Cross, A.L.
(1992) Pregnancy, labor and delivery post spinal cord injury.
Paraplegia, 30: 890–902.
Dahlberg, A., Ruutu, M. and Hovatta, O. (1995) Pregnancy
results from a vibrator application, electroejaculation, and a
vas aspiration programme in spinal-cord injured men. Hum.
Reprod., 10: 2305–2307.
Daniels, D., Miselis, R.R. and Flanagan-Cato, L.M. (1999)
Central neuronal circuit innervating the lordosis-producing
muscles defined by transneuronal transport of pseudorabies
virus. J. Neurosci., 19: 2823–2833.
de Groat, W.C. and Booth, A.M. (1980) Physiology of male
sexual function. Ann. Intern. Med., 92: 329–331.
de Groat, W.C. and Booth, A.M. (1993) Neural control of pe-
nile erection. In: Maggi C.A. (Ed.), The Autonomic Nervous
System. Harwood Academic, London, pp. 465–516.
de Groat, W.C., Nadelhaft, I., Milne, R.J., Booth, A.M., Mor-
gan, C. and Thor, K. (1981) Organization of the sacral para-
sympathetic reflex pathways to the urinary bladder and large
intestine. J. Auton. Nerv. Syst., 3: 135–160.
deGroat, W.C. and Steers, W.D. (1988) Neuroanatomy and ne-
urophysiology of penile erection. In: Tanagho E.A., Lue T.F.
and McClure R.D. (Eds.), Contemporary Management Of
Impotence and Infertility. Williams and Wilkins, Baltimore,
pp. 3–27.
Diakow, C. (1970) Effects of genital desensitization on the
mating pattern of female rats as determined by motion pic-
ture analysis. Am. Zool., 10: 486.
Ding, Y.Q., Shi, J., Wang, D.S., Xu, J.Q., Li, J.L. and Ju, G.
(1999) Primary afferent fibers of the pelvic nerve terminate in
the gracile nucleus of the rat. Neurosci. Lett., 272: 211–214.
Erskine, M.S. (1992) Pelvic and pudendal nerves influence the
display of paced mating behavior in response to estrogen and
progesterone in the female rat. Behav. Neurosci., 106: 690–697.
Garfield, R.E. (1986) Structural studies of innervation on non-
pregnant rat uterus. Am. J. Physiol., 251: C41–C54.
Giesler Jr., G.J., Spiel, H.R. and Willis, W.D. (1981) Organ-
ization of spinothalamic tract axons within the rat spinal
cord. J. Comp. Neurol., 195: 243–252.
Giuliano, F.A., Rampin, O., Benoit, G. and Jardin, A. (1995)
Neural control of penile erection. Urol. Clin. North Am., 22:
747–766.
Graber, G.C. and Kristal, M.B. (1977) Uterine distention fa-
cilitates the onset of maternal behavior in pseudopregnant
but not in cycling rats. Physiol. Behav., 19: 133–137.
Guevara-Guzman, R., Buzo, E., Larrazolo, A., de la Riva, C.,
Da Costa, A.P. and Kendrick, K.M. (2001) Vaginocervical
stimulation-induced release of classical neurotransmitters
411
and nitric oxide in the nucleus of the solitary tract varies as
a function of the oestrus cycle. Brain Res., 898: 303–313.
Haase, E.B., Buchman, J., Tietz, A.E. and Schramm, L.P.
(1997) Pregnancy-induced uterine neuronal degeneration in
the rat. Cell Tissue Res., 288: 293–306.
Haldeman, S., Bradley, W.E., Bhatia, N.N. and Johnson, B.K.
(1982) Pudendal evoked responses. Arch. Neurol., 39:
280–283.
Haskins, J.T. and Moss, R.L. (1983) Action of estrogen and
mechanical vaginocervical stimulation on the membrane
excitability of hypothalamic and midbrain neurons. Brain
Res. Bull., 10: 489–496.
Higuchi, T., Uchide, K., Honda, K. and Negoro, H. (1987)
Pelvic neurectomy abolishes the fetus-expulsion reflex and
induces dystocia in the rat. Exp. Neurol., 96: 443–455.
Honda, C.N. (1985) Visceral and somatic afferent convergence
onto neurons near the central canal in the sacral spinal cord
of the cat. J. Neurophysiol., 53: 1059–1078.
Hornby, J.B. and Rose, J.D. (1976) Responses of caudal brain
stem neurons to vaginal and somatosensory stimulation in
the rat and evidence of genital-nociceptive interactions. Exp.
Neurol., 51: 363–376.
Houghton, A.K., Kadura, S. and Westlund, K.N. (1997)
Dorsal column lesions reverse the reduction of homecage
activity in rats with pancreatitis. Neuroreport, 8: 3795–3800.
Hubscher, C.H. (1994) Sensory input from pelvic reproductive
organs to the gracile and solitary nuclei in the female rat.
Department of Psychology: Florida State University,
Tallahassee, p. 80.
Hubscher, C.H. and Berkley, K.J. (1994) Responses of neurons
in caudal solitary nucleus of female rats to stimulation of
vagina, cervix, uterine horn and colon. Brain Res., 664: 1–8.
Hubscher, C.H. and Berkley, K.J. (1995) Spinal and vagal in-
fluences on the responses of rat solitary nucleus neurons to
stimulation of uterus, cervix and vagina. Brain Res., 702:
251–254.
Hubscher, C.H., Brooks, D.L. and Johnson, J.R. (2004a)
Effects of spinal cord injury on the rat estrous cycle. Program
No. 458.15. In: 2004 Abstract Viewer/Itinerary Planner.
Washington, DC: Society for Neuroscience.
Hubscher, C.H. and Johnson, R.D. (1996) Responses of med-
ullary reticular formation neurons to input from the male
genitalia. J. Neurophysiol., 76: 2474–2482.
Hubscher, C.H. and Johnson, R.D. (1999) Effects of acute and
chronic midthoracic spinal cord injury on neural circuits
for male sexual function. I. Ascending pathways. J.
Neurophysiol., 82: 1381–1389.
Hubscher, C.H. and Johnson, R.D. (2001) Supraspinal
processing of inputs from the cervix: effects of hormonal
status. Soc. Neurosci. Abstr., 27: Program No. 840–841.
Hubscher, C.H. and Johnson, R.D. (2003) Responses of
thalamic neurons to input from the male genitalia. J. Ne-
urophysiol., 89: 2–11.
Hubscher, C.H. and Johnson, R.D. (2004) Effects of chronic
dorsal column lesions on pelvic viscerosomatic convergent
medullary reticular formation neurons. J. Neurophysiol., 92:
3596–3600.
412
Hubscher, C.H., Kaddumi, E.G. and Johnson, R.D. (2004b)
Brain stem convergence of pelvic viscerosomatic inputs via
spinal and vagal afferents. Neuroreport, 15: 1299–1302.
Hulsebosch, C.E. and Coggeshall, R.E. (1982) An analysis of
the axon populations in the nerves to the pelvic viscera in the
rat. J. Comp. Neurol., 211: 1–10.
Jean, A. (1991) The nucleus tractus solitarius: neuroanatomic,
neurochemical and functional aspects. Arch. Int. Physiol.
Biochem. Biophys., 99: A3–A52.
Johnson, R.D. (1989) Physiology of single spinal cord sensory
neurons responding to penile stimulation in the rat. Soc.
Neurosci. Abstr., 15: 756.
Johnson, R.D. and Halata, Z. (1991) Topography and ultra-
structure of sensory nerve endings in the glans penis of the
rat. J. Comp. Neurol., 312: 299–310.
Kitchell, R.L., Gilanpour, H. and Johnson, R.D. (1982) Elect-
rophysiologic studies of penile mechanoreceptors in the rats.
Exp. Neurol., 75: 229–244.
Klein, C.M., Ray, R.H. and Burden, H.W. (1989) Direct elec-
trical stimulation of the superior ovarian nerve in rats causes
an increase in neuronal activity in the ipsilateral ovarian
plexus nerve. Brain Res., 479: 194–200.
Kollar, E.J. (1953) Reproduction in the female rat after pelvic
nerve neurectomy. Anat. Rec., 115: 641–658.
Komisaruk, B.R., Bianca, R., Sansone, G., Gomez, L.E.,
Cueva-Rolon, R., Beyer, C. and Whipple, B. (1996) Brain-
mediated responses to vaginocervical stimulation in spinal
cord-transected rats: role of the vagus nerves. Brain Res.,
708: 128–134.
Komisaruk, B.R., Gerdes, C.A. and Whipple, B. (1997) ‘Com-
plete’ spinal cord injury does not block perceptual responses
to genital self-stimulation in women. Arch. Neurol., 54:
1513–1520.
Komisaruk, B.R. and Sansone, G. (2003) Neural pathways
mediating vaginal function: the vagus nerves and spinal cord
oxytocin. Scand. J. Psychol., 44: 241–250.
Komisaruk, B.R. and Wallman, J. (1977) Antinociceptive ef-
fects of vaginal stimulation in rats: neurophysiological and
behavioral studies. Brain Res., 137: 85–107.
Komisaruk, B.R. and Whipple, B. (1988) The role of vaginal
stimulation-produced analgesia in reproductive processes. In:
Genazzani A.R., Nappi G., Facchinetti F. and Martignoni E.
(Eds.), Pain and Reproduction. Parthenon Publishing
Group, New Jersey, pp. 125–140.
Komisaruk, B.R., Whipple, B., Crawford, A., Liu, W.C.,
Kalnin, A. and Mosier, K. (2004) Brain activation during
vaginocervical self-stimulation and orgasm in women with
complete spinal cord injury: fMRI evidence of mediation by
the vagus nerves. Brain Res., 1024: 77–88.
Kostarczyk, E., Zhang, X. and Giesler Jr., G.J. (1997) Spin-
ohypothalamic tract neurons in the cervical enlargement of
rats: locations of antidromically identified ascending axons
and their collateral branches in the contralateral brain. J.
Neurophysiol., 77: 435–451.
Kow, L.M. (1976) Sensory requirements for the lordosis reflex
in female rats. Brain Res., 101: 47–66.
Kow, L.M. and Pfaff, D.W. (1973) Effects of estrogen treat-
ment on the size of receptive field and response threshold of
pudendal nerve in the female rat. Neuroendocrinology, 13:
299–313.
Kumazawa, T. (1986) Sensory innervation of reproductive or-
gans. Prog. Brain Res., 67: 115–131.
Kumazawa, T., Mizumura, K., Koda, H., Tamura, R. and
Sato, J. (1995) Mechanisms of chemical modulation of tes-
ticular afferents. In: Gebhart G.F. (Ed.), Visceral Pain
Progress in Pain Research and Management. IASP Press,
Seattle.
Larsson, K. and Sodersten, P. (1973) Mating in male rats after
section of the dorsal penile nerve. Physiol. Behav., 10:
567–571.
Larsson, K. and Swedin, G. (1971) The sexual behavior of male
rats after bilateral section of the hypogastric nerve and re-
moval of the accessory genital glands. Physiol. Behav., 6:
251–253.
Lee, J.W. and Erskine, M.S. (1996) Vaginocervical stimulation
suppresses the expression of c-fos induced by mating in tho-
racic, lumbar and sacral segments of the female rat. Neuro-
science, 74: 237–249.
Lee, J.W. and Erskine, M.S. (2000) Pseudorabies virus tracing
of neural pathways between the uterine cervix and CNS: ef-
fects of survival time, estrogen treatment, rhizotomy, and
pelvic nerve transection. J. Comp. Neurol., 418: 484–503.
Linsenmeyer, T.A., Pogach, L.M., Ottenweller, J.E. and
Huang, H.F. (1994) Spermatogenesis and the pituitary-tes-
ticular hormone axis in rats during the acute phase of spinal
cord injury. J. Urol., 152: 1302–1307.
Lodder, J. and Zeilmaker, G.H. (1976) Effects of pelvic nerve
and pudendal nerve transection on mating behavior in the
male rat. Physiol. Behav., 16: 745–751.
Marchetti, B., Cioni, M., Badr, M., Follea, N. and Pelletier, G.
(1987) Ovarian adrenergic nerves directly participate in the
control of luteinizing hormone-releasing hormone and beta-
adrenergic receptors during puberty: a biochemical and au-
toradiographic study. Endocrinology, 121: 219–226.
Marson, L. and McKenna, K.E. (1990) The identification of a
brainstem site controlling spinal sexual reflexes in male rats.
Brain Res., 515: 303–308.
Marson, L., Platt, K.B. and McKenna, K.E. (1993) Central
nervous system innervation of the penis as revealed by the
transneuronal transport of pseudorabies virus. Neuroscience,
55: 263–280.
Mason, P. (2001) Contributions of the medullary raphe and
ventromedial reticular region to pain modulation and other
homeostatic functions. Ann. Rev. Neurosci., 24: 737–777.
McKenna, K.E. (2000) The neural control of female sexual
function. NeuroRehabilitation, 15: 133–143.
McKenna, K.E. and Nadelhaft, I. (1986) The organization of
the pudendal nerve in the male and female rat. J. Comp.
Neurol., 248: 532–549.
McMahon, S.B. and Wall, P.D. (1983) A system of rat spinal
cord lamina 1 cells projecting through the contralateral
dorsolateral funiculus. J. Comp. Neurol., 214: 217–223.

Meana, M., Binik, Y.M., Khalife, S., Bergeron, S., Pagidas, K.
and Berkley, K.J. (1997) Dyspareunia: more than bad sex.
Pain, 71: 211–212.
Menetrey, D., Chaouch, A. and Besson, J.M. (1980) Location
and properties of dorsal horn neurons at origin of spin-
oreticular tract in lumbar enlargement of the rat. J. Neuro-
physiol., 44: 862–877.
Menetrey, D. and de Pommery, J. (1991) Origins of spinal as-
cending pathways that reach central areas involved in vis-
ceroception and visceronociception in the rat. Eur. J.
Neurosci., 3: 249–259.
Modianos, D. and Pfaff, D. (1979) Medullary reticular forma-
tion lesions and lordosis reflex in female rats. Brain Res., 171:
334–338.
Moore, C.L. and White, R.H. (1996) Sex differences in sensory
and motor branches of the pudendal nerve of the rat. Horm.
Behav., 30: 590–599.
Mtui, E.P., Anwar, M., Reis, D.J. and Ruggiero, D.A. (1995)
Medullary visceral reflex circuits: local afferents to nucleus
tractus solitarii synthesize catecholamines and project to
thoracic spinal cord. J. Comp. Neurol., 351: 5–26.
Nadelhaft, I. and Booth, A.M. (1984) The location and mor-
phology of preganglionic neurons and the distribution of
visceral afferents from the rat pelvic nerve: a horseradish
peroxidase study. J. Comp. Neurol., 226: 238–245.
Nadelhaft, I. and McKenna, K.E. (1987) Sexual dimorphism in
sympathetic preganglionic neurons of the rat hypogastric
nerve. J. Comp. Neurol., 256: 308–315.
Nahin, R.L., Madsen, A.M. and Giesler Jr., G.J. (1986) Funic-
ular location of the ascending axons of neurons adjacent to the
spinal cord central canal in the rat. Brain Res., 384: 367–372.
Nance, D.M., Burns, J., Klein, C.M. and Burden, H.W. (1988)
Afferent fibers in the reproductive system and pelvic viscera
of female rats: anterograde tracing and immunocytochemical
studies. Brain Res. Bull., 21: 701–709.
Nehra, A., Werner, M.A., Bastuba, M., Title, C. and Oates,
R.D. (1996) Vibratory stimulation and rectal probe electro-
ejaculation as therapy for patients with spinal cord injury:
semen parameters and pregnancy rates. J. Urol., 155:
554–559.
Neuhuber, W. (1982) The central projections of visceral pri-
mary afferent neurons of the inferior mesenteric plexus and
hypogastric nerve and the location of the related sensory and
preganglionic sympathetic cell bodies in the rat. Anat. Em-
bryol., 164: 413–425.
Nunez, R., Gross, G.H. and Sachs, B.D. (1986) Origin and
central projections of rat dorsal penile nerve: possible direct
projection to autonomic and somatic neurons by primary
afferents of nonmuscle origin. J. Comp. Neurol., 247:
417–429.
Odutola, A.B. (1977) On the location of reticular neurons pro-
jecting to the cuneogracile nuclei in the rat. Exp. Neurol., 54:
54–59.
Ortega-Villalobos, M., Garcia-Bazan, M., Solano-Flores, L.P.,
Ninomiya-Alarcon, J.G., Guevara-Guzman, R. and Wayner,
M.J. (1990) Vagus nerve afferent and efferent innervation of
413
the rat uterus: an electrophysiological and HRP study. Brain
Res. Bull., 25: 365–371.
Pacheco, P., Camacho, M.A., Garcia, L.I., Hernandez, M.E.,
Carrillo, P. and Manzo, J. (1997) Electrophysiological evi-
dence for the nomenclature of the pudendal nerve and sacral
plexus in the male rat. Brain Res., 763: 202–208.
Pacheco, P., Martinez-Gomez, M., Whipple, B., Beyer, C. and
Komisaruk, B.R. (1989) Somato-motor components of the
pelvic and pudendal nerves of the female rat. Brain Res., 490:
85–94.
Papka, R.E., Williams, S., Miller, K.E., Copelin, T. and Puri, P.
(1998) CNS location of uterine-related neurons revealed by
trans-synaptic tracing with pseudorabies virus and their re-
lation to estrogen receptor-immunoreactive neurons. Neuro-
science, 84: 935–952.
Pavlakis, A.J., Siroky, M.B., Goldstein, I. and Krane, R.J.
(1983) Neurourologic findings in conus medullaris and cauda
equina injury. Arch. Neurol., 40: 570–573.
Peschanski, M. and Besson, J.M. (1984) A spino-reticulo-
thalamic pathway in the rat: an anatomical study with ref-
erence to pain transmission. Neuroscience, 12: 165–178.
Peters, L.C., Kristal, M.B. and Komisaruk, B.R. (1987) Sen-
sory innervation of the external and internal genitalia of the
female rat. Brain Res., 408: 199–204.
Pfaus, J.G., Marcangione, C., Smith, W.J., Manitt, C. and
Abillamaa, H. (1996) Differential induction of Fos in the
female rat brain following different amounts of vaginocervi-
cal stimulation: modulation by steroid hormones. Brain Res.,
741: 314–330.
Pierau, F.K., Taylor, D.C., Abel, W. and Friedrich, B. (1982)
Dichotomizing peripheral fibres revealed by intracellular re-
cording from rat sensory neurones. Neurosci. Lett., 31:
123–128.
Quinlan, D.M., Nelson, R.J., Partin, A.W., Mostwin, J.L. and
Walsh, P.C. (1989) The rat as a model for the study of penile
erection. J. Urol., 141: 656–661.
Rampin, O., Gougis, S., Giuliano, F. and Rousseau, J.P. (1997)
Spinal Fos labeling and penile erection elicited by stimulation
of dorsal nerve of the rat penis. Am. J. Physiol., 272:
R1425–R1431.
Reiner, P., Woolsey, J., Adler, N. and Morrison, A. (1981) A
gross anatomical study of the peripheral nerves associated
with reproductive function in the female albino rat. In: Adler
N.T. (Ed.), Neuroendocrinology of Reproduction. Plenum
Press, New York, pp. 545–549.
Ross, E.L., Komisaruk, B.R. and O’Donnell, D. (1979) Evi-
dence that probing the vaginal cervix is analgesic in rats,
using an operant paradigm. J. Comp. Physiol. Psychol., 93:
330–336.
Sachs, B.D. and Meisel, R.L. (1988) The physiology of male
sexual behavior. In: Knobil E. and Neill J.D. (Eds.), The
Physiology of Reproduction. Raven, New York,
pp. 1393–1485.
Sarkarati, M., Rossier, A.B. and Fam, B.A. (1987) Experience
in vibratory and electro-ejaculation techniques in spinal cord
injury patients: a preliminary report. J. Urol., 138: 59–62.
414
Sartucci, F., Piaggesi, A., Logi, F., Bonfiglio, L., Bongioanni,
P., Pellegrinetti, A., Baccetti, F., Navalesi, R. and Murri, L.
(1999) Impaired ascendant central pathways conduction in
impotent diabetic subjects. Acta Neurol. Scand., 99: 381–386.
Sato, S., Hayashi, R.H. and Garfield, R.E. (1989) Mechanical
responses of the rat uterus, cervix, and bladder to stimulation
of hypogastric and pelvic nerves in vivo. Biol. Reprod., 40:
209–219.
Sauer, P.M. and Harvey, C.J. (1993) Spinal cord injury and
pregnancy. J. Perinat. Neonatal. Nurs., 7: 22–34.
Schwartz-Giblin, S., McCarthy, M.M. and Robbins, A. (1996)
The medullary reticular formation is a site of muscle relaxant
action of diazepam on deep back and neck muscles in the
female rat. Brain Res., 710: 178–188.
Sedor, J.F. and Hirsch, I.H. (1995) Evaluation of sperm mor-
phology of electroejaculates of spinal cord-injured men by
strict criteria. Fertil. Steril., 63: 1125–1127.
Seftel, A.D., Oates, R.D. and Krane, R.J. (1991) Disturbed
sexual function in patients with spinal cord disease. Neurol.
Clin., 9: 757–778.
Serghini, R., Prud’homme, M.J., Vaudry, H. and Rousseau,
J.P. (1997) Involvement of the pelvic plexus and the supra-
renal ganglia in the neuropeptide Y (NPY) innervation of the
cervix and the uterus of the rat. J. Auton. Nerv. Syst., 67:
38–50.
Shen, P., Arnold, A.P. and Micevych, P.E. (1990) Supraspinal
projections to the ventromedial lumbar spinal cord in adult
male rats. J. Comp. Neurol., 300: 263–272.
Sipski, M.L., Alexander, C.J. and Rosen, R. (2001) Sexual
arousal and orgasm in women: effects of spinal cord injury.
Ann. Neurol., 49: 35–44.
Sonksen, J. and Ohl, D.A. (2002) Penile vibratory stimulation
and electroejaculation in the treatment of ejaculatory dys-
function. Int. J. Androl., 25: 324–332.
Taylor, D.C., Korf, H.W. and Pierau, F.K. (1982) Distribution
of sensory neurons of the pudendal nerve in the dorsal root
ganglia and their projection to the spinal cord. Horseradish-
peroxidase studies in the rat. Cell Tissue Res., 226: 555–564.
Temple, J.L., Bradshaw, H.B., Wood, E. and Berkley, K.J.
(1999) Effects of hypogastric neurectomy on escape responses
to uterine distention in the rat. Pain, 6(Suppl.): S13–S20.
Tetel, M.J., Getzinger, M.J. and Blaustein, J.D. (1993) Fos ex-
pression in the rat brain following vaginal-cervical stimula-
tion by mating and manual probing. J. Neuroendocrinol., 5:
397–404.
Tomasulo, K.C. and Emmers, R. (1972) Activation of neurons
in the gracile nucleus by two afferent pathways in the rat.
Exp. Neurol., 36: 197–206.
Truitt, W.A. and Coolen, L.M. (2002) Identification of a po-
tential ejaculation generator in the spinal cord. Science, 297:
1566–1569.
Van der Horst, V.G. and Holstege, G. (1998) Sensory and mo-
tor components of reproductive behavior: pathways and
plasticity. Behav. Brain Res., 92: 157–167.
Ver Voort, S. (1987) Infertility in spinal cord injured male.
Urology, XXIX(29): 157–165.
Wall, P.D., Hubscher, C.H. and Berkley, K.J. (1993) Intraspinal
modulation of neuronal responses to uterine and cervix stim-
ulation in rat L1 and L6 dorsal horn. Brain Res., 622: 71–78.
Wang, C.C., Willis, W.D. and Westlund, K.N. (1999) Ascend-
ing projections from the area around the spinal cord central
canal: a Phaseolus vulgaris leucoagglutinin study in rats. J.
Comp. Neurol., 415: 341–367.
Westgren, N. and Levi, R. (1994) Motherhood after traumatic
spinal cord injury. Paraplegia, 32: 517–523.
Willis, W.D., Al-Chaer, E.D., Quast, M.J. and Westlund, K.N.
(1999) A visceral pain pathway in the dorsal column of the
spinal cord. Proc. Natl. Acad. Sci. USA, 96: 7675–7679.
Willis, W.D. and Coggeshall, R.E. (1991) Sensory Mechanisms
of the Spinal Cord (2nd ed). Plenum Press, New York.
Yang, S.W., Follett, K.A., Piper, J.G. and Ness, T.J. (1998) The
effect of morphine on responses of mediodorsal thalamic
nuclei and nucleus submedius neurons to colorectal disten-
sion in the rat. Brain Res., 779: 41–52.
Zemlan, F.P., Leonard, C.M., Kow, L.M. and Pfaff, D.W.
(1978) Ascending tracts of the lateral columns of the rat spi-
nal cord: a study using the silver impregnation and horse-
radish peroxidase techniques. Exp. Neurol., 62: 298–334.
Zhuo, M. and Gebhart, G.F. (1991) Spinal serotonin receptors
mediate descending facilitation of a nociceptive reflex from
the nuclei reticularis gigantocellularis and gigantocellularis
pars alpha in the rat. Brain Res., 550: 35–48.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Published by Elsevier B.V.

CHAPTER 28

Descending pathways modulating the spinal circuitry

for ejaculation: effects of chronic spinal cord injury

Richard D. Johnson

Department of Physiological Sciences, College of Veterinary Medicine and the McKnight Brain Institute, University of
Florida, Gainesville, FL 32610-0144, USA

Abstract: Sexual dysfunction is a common complication in men with chronic spinal cord injury. In par-
ticular, ejaculation is severely compromised or absent and the resulting infertility issues are important to
this group of predominantly young men. To investigate the neural circuits and descending spinal pathways
involved in ejaculation, animal models have been developed in normal and spinal cord-injured prepara-
tions. Primarily through studies in rats, spinal ejaculatory circuits have been described including (i) au-
tonomic circuits at the thoracolumbar and lumbosacral levels mediating the emission phase of ejaculation,
(ii) somatic circuits at the lumbosacral level controlling the expulsion phase of ejaculation through se-
quential and rhythmic contraction of perineal striated muscles (e.g. bulbospongiosus), and (iii) a proposed
ejaculatory pattern generator in the lumbar cord. Midthoracic incomplete chronic spinal cord injury has
revealed the dependency of spinal ejaculatory circuits on bilateral spinal pathways from the brainstem via
modulation of pudendal motor neuron reflexes and pudendal nerve autonomic fibers. Accordingly, sensory
input from the dorsal nerve of the penis, required to trigger the ejaculatory response in animals and
humans, is no longer inhibited from the lateral paragigantocellularis nucleus in the ventrolateral medulla.
This inhibitory effect, likely presynaptic through a serotonergic pathway, is thought to be necessary to
provide the rhythmic, bursting, and sequential contractions of the perineal muscles during ejaculation.
Chronic lateral hemisection injury, which severs half of the descending lateral funiculus-located pathways,
results in new functional connections of the pudendal reflex inhibitory and pudendal sympathetic activation
pathways across the midline, above and below the lesion, respectively. Clinical correlations in spinal cord-
injured men have demonstrated the validity of the rodent animal for the study of ejaculatory dysfunction
after chronic injury.

Introduction recent survey of men with spinal cord injury by

Male sexual dysfunction is a common complica-
tion resulting from chronic spinal cord injury. The
human spinal cord-injured population is predom-
inantly male and of this group, 85% are men
between the ages of 18 and 45 (Ohl et al., 1989). A
Corresponding author. Tel.: +352-392-4700 ext 3834;
Fax: +352-392-5145; E-mail: johnson@ufbi.ufl.edu
Anderson (2004) has demonstrated that among
paraplegics, regaining normal sexual function is
the most important goal, even surpassing locomo-
tion. Moreover, the ability to ejaculate is severely
compromised making infertility a critical issue.
Clinical and animal studies over the last 25 years
have investigated the integrity and degree of
recovery of male sexual function following spinal
injury in order to (i) document the neural pathways

DOI: 10.1016/S0079-6123(05)52028-4 415

416
affected, (ii) evaluate the effects of different de-
grees of chronic injury, and (iii) devise treatment
strategies to improve functional outcome.
Of paramount importance has been the devel-
opment of animal models for the study of male
sexual function following acute and chronic spinal
cord injury. Although the investigation of normal
sexual neural mechanisms have included impor-
tant studies in the cat and dog, the rat has emerged
as the most appropriate animal model particularly
with regard to studying the chronic injury. Several
advantages to using the rat model are (i) the ex-
istence of a substantial behavioral database on
normal sexual function, (ii) a variety of spinal le-
sions can be performed with few complications,
(iii) the accessibility of single neurons or nerve/
muscle ensembles for recording in anesthetized
preparations, and (iv) combined electrophysiolog-
ical, neuroanatomical, pharmacological, and be-
havioral studies can be carried out in the same
animal.
In this chapter, a consideration of the studies
using animal models will concentrate on the chronic
spinal cord injury-induced alteration of ejaculato-
ry circuitry and provide a structural template on
which future studies can be designed. When ap-
propriate, data from human clinical studies will be
correlated with data from animal models.
Spinal reflex circuitry for ejaculation
Much of the neural circuitry for erection and ejac-
ulation is contained within specific regions of the
spinal cord (reviewed by McKenna, 2000; Steers,
2000; Coolen et al., 2004; Giuliano and Rampin,
2004). Spinal cord-injured humans and animals,
however, lose portions of these sexual responses
when the lesion severs ascending and descending
pathways to the brainstem and higher centers,
suggesting the segmental circuits require a long-
loop connection to maintain optimum function.
The spinal circuits are very complex primarily be-
cause they involve the integration of both divisions
of the autonomic motor system (parasympathetic
and sympathetic), the somatic motor system, and
all three divisions of general sensory afferent fibers
(visceral, somatic, mucocutaneous) in a network of
intraspinal connections that span several spinal
segments [in the rat; thoracic (T)10–sacral (S)1].
Studies in animals and humans, detailed below,
have shown that descending control from supra-
spinal sites are required for coordination of the
visceral and somatic motor events mediating the
forceful, rhythmic expulsion of semen in ejacula-
tion and moving the blood into the glans penis in
rigid erection. These two propulsive-like sexual
events are dependent on a coordinated contraction
sequence of smooth and striated muscles similar to
other eliminative viscerosomatic responses like
defecation and micturition.
The ejaculation response consists of two phases.
The initial emission phase involves accessory gland
secretions and smooth muscle-mediated movement
of sperm and of the resultant mixture of sperm
with glandular secretions (semen) through the in-
ternal segments of the reproductive tract into the
pelvic urethra. The emission phase is mediated by
the sympathetic and parasympathetic divisions of
the autonomic nervous system. Sympathetic pre-
ganglionic emission neurons are located in the in-
termediolateral cell column and central gray of the
T12 to lumbar (L)2 spinal cord and send axons
into the hypogastric nerve and sympathetic trunk
(Nadelhaft and McKenna, 1987). Parasympathetic
preganglionic emission neurons reside in the sacral
portion of the intermediolateral cell column (also
known as the sacral parasympathetic nucleus) and
project peripherally through the pelvic nerve. Fol-
lowing emission, the expulsion phase of ejacula-
tion consists of a rapid, forceful, and rhythmic
propulsion of semen through the penile urethra
and out through the external urethral orifice. Ex-
pulsion of semen is produced by the coordinated
and rhythmic contraction of the striated perineal
muscles; bulbospongiosus, ischiocavernosus, and
external urethral sphincter. In the rat, the motor
neurons innervating these muscles are located in
the dorsomedial (bulbospongiosus) and dorsolat-
eral (ischiocavernosus, external urethral sphincter)
motor nuclei in the ventral horn of L5–L6
(McKenna and Nadelhaft, 1986; Collins et al.,
1991). In humans and certain other species (primate,
cat, dog), these motoneuronal pools are combined
in Onuf’s nucleus (Schroder, 1985; Beattie et al.,
1993) located in the sacral segments. The perineal

motor neurons send their axons peripherally via
the pudendal nerve (via the motor branch in rats).
The perineal muscle that is primarily involved in
ejaculatory expulsion and in the firm erection of
the glans penis (Johnson, 1988; Holmes et al.,
1991; Schmidt and Schmidt, 1993; Yang and
Bradley, 1999) is the bulbospongiosus (also known
as the bulbocavernosus), a striated muscle that
surrounds the bulb of the penis and its internal
corpus spongiosum.
The segmental reflex circuit for perineal muscle
contraction resides in the L5–S1 spinal cord of the
rat. The bulbospongiosus and ischiocavernosus
motor neurons have many properties that are dif-
ferent from typical somatic motor neurons inner-
vating skeletal muscles and have many similarities
to autonomic motor neurons. For example, they
are characterized by lack of a monosynaptic reflex
arc from the homonymous muscle spindle recep-
tors (Bowens et al., 1984; McKenna and Nadelhaft,
1989; Collins et al., 1991), resistance to paradox-
ical sleep-triggered inhibition (Mann et al., 2003)
and to damage from amyotrophic lateral sclerosis
(Carvalho et al., 1995) but exhibit susceptibility to
autonomic motor neuron disorders (Sung et al.,
1979; Dubrovsky and Filipini, 1990). In addition,
these neurons are trophically dependent on the
presence of testosterone (Breedlove and Arnold,
1980), are each bilaterally organized for simulta-
neous contraction, and can have dendritic arbors
that cross the midline forming gap junction-medi-
ated electrical communication with their contra-
lateral counterpart (Collins et al., 1991; Coleman
and Sengelaub, 2002). The polysynaptic segmental
reflex activation of these pudendal motoneurons
requires the participation of many spinal interneu-
rons that have been identified, via transsynaptic
chemical and viral retrograde tracers (Collins
et al., 1991; Marson and McKenna, 1996), bilat-
erally in the central and intermediate gray of the
lumbosacral cord. Spinal interneurons receiving
bilateral synaptic input from dorsal nerve of the
penis afferent terminations (Nunez et al., 1986)
have been recorded from throughout the T13–S1
spinal cord (R.D. Johnson, unpublished obser-
vations). In addition to connecting pudendal
motoneurons synaptically with segmental primary
afferent terminals (see below), the interneurons
417
may connect with a proposed spinal pattern
generator for ejaculation, located in the L3–L4
central gray with connections to the autonomic
and somatic spinal circuits for sexual function
(Truitt and Coolen, 2002; Carro-Juarez et al.,
2003; Coolen et al., 2004).
The primary sensory input to the pudendal seg-
mental reflex circuit and probably to the proposed
ejaculatory spinal pattern generator, is the dorsal
nerve of the penis, the most distal portion of the
pudendal nerve innervating sensory endings in the
penis and prepuce (Johnson, 1988; Johnson and
Halata, 1991). Penile mechanoreceptors have been
shown to (i) respond to vibratory and tangential
surface tactile stimulation (Johnson and Murray,
1992), (ii) spontaneously respond to blood flow
engorging the erectile tissue in the absence of skin
stimulation, an internal response to stretch (Johnson,
1988), and (iii) increase their tactile sensitivity
during erection possibly through autonomic fiber
modulation of the sensory ending (Johnson, 1988;
Johnson and Halata, 1991). Behavioral and func-
tional studies in animals and humans have dem-
onstrated that the integrity of the penile sensory
axons is essential for triggering ejaculation (Hart
and Leedy, 1985; Meisel and Sachs, 1994; Wieder
et al., 2000). Visceral afferents innervating the
urethral mucosa (McKenna and Nadelhaft, 1989)
and erectile tissue in the corpus cavernosum/
spongiosum (Johnson, 1988; Johnson and Halata,
1991) along with the mucocutaneous afferents in-
nervating the glans penis and external urethral or-
ifice (Johnson and Halata, 1991) are all present
within the dorsal nerve of the penis and pudendal
nerve. Electrical, tactile, or chemical activation of
the pudendal primary afferents triggers a polysy-
naptic reflex discharge of pudendal motoneurons
(McKenna and Nadelhaft, 1989; Johnson, 1995;
Johnson and Hubscher, 1998), produces ejacula-
tion-like motor patterns (McKenna et al., 1991;
Carro-Juarez et al., 2003) and drives the bulbo-
cavernosus reflex in men (Yang and Bradley, 1999).
To investigate the electrophysiological charac-
teristics of the pudendal reflex circuit in male rats
with and without chronic spinal cord injury, our
laboratory at the University of Florida developed
an in vivo animal model. By unilaterally and
bilaterally recording and stimulating the pudendal
418
reflex elements simultaneously with activation of
descending modulatory pathways from the medul-
la (Fig. 1), the effects of different spinal lesions and
postlesion intervals could be determined. In unin-
jured anesthetized animals, stimulation of penile
afferents produced polysynaptic bilateral reflex
discharges that were subject to rate depression and
presynaptic inhibition (Johnson, 1995; Johnson
and Hubscher, 1998). In contrast, stimulation of
pelvic nerve afferents elicited bilateral reflex dis-
charges that were resistant to rate depression and
presynaptic inhibition. These results suggested that
the spinal interneurons activated by these two dif-
ferent inputs are separate and may be modulated by
different descending pathways from the brainstem.
Descending brainstem control of ejaculation
Several supraspinal regions have been shown to
exert various degrees of facilitatory or inhibitory
control on the spinal centers for ejaculation (re-
viewed by Coolen et al., 2004). Excitatory influ-
ences have been proposed from several regions of
the hypothalamus, the medial preoptic area (Pehek
et al., 1989; Markowski et al., 1994), the paraven-
tricular nucleus (Marson and McKenna, 1994),
and the lateral hypothalamus (Kippin et al., 2004).
Strong inhibitory influences descend to the spinal
cord from the lateral portion of the nucleus par-
agigantocellularis in the medullary reticular for-
mation (Marson and McKenna, 1992). Numerous
neuroanatomical studies have shown robust axon-
al projections to the perineal motoneuronal pools
from the lateral paragigantocellularis (Marson and
McKenna, 1996; Hermann et al., 2003), the adja-
cent nucleus raphe obscurus (Hermann et al.,
1998; Holmes et al., 2002), and the paraventricular
nucleus in the thalamus (Wagner and Clemens,
1991). Although the medial preoptic area does not
project axons to the spinal cord, it likely has an
influence through connections to the lateral par-
agigantocellularis via the periaqueductal gray
(Murphy and Hoffman, 2001). Other supraspinal
regions above the brainstem that may be active
prior to, during, or after ejaculation have been
identified in neuroanatomical studies in rats (Coolen
et al., 2004) and positron emission tomography
(PET)-scan studies in men (Holstege and Georgiadis,
2004). However, as these areas do not project to
the spinal cord, any modulatory effect they might
have would likely be relayed through the medul-
lary spinal pathways.
The descending spinal pathways from the vent-
rolateral medulla and medullary reticular forma-
tion modulate many viscerosomatic functions in
the spinal cord including nociceptive reflexes, how-
ever, they have been proposed to organize and
temporally sequence complex sensorimotor activ-
ities which require both somatic and visceral
(autonomic) motor neurons such as gagging, vom-
iting, defecation, and ejaculation (Holstege, 1991;
Mason, 2001). Electrical stimulation of the lateral
paragigantocellularis nucleus has been shown to
produce field potentials in the lumbosacral spinal
cord near the pudendal motor nuclei (Tanaka and
Arnold, 1993). Large lesions that included the
reticularis gigantocellularis nuclear complex
eliminated ejaculatory bursts in perineal muscles
triggered by urethral stimulation (urethrogenital
reflex; Marson and McKenna, 1990). Smaller
lesions of the nucleus raphe obscurus or lateral
paragigantocellularis nucleus had significant effects
on external anal sphincter and perineal muscle
reflexes, respectively (Holmes et al., 2002) and
reduced ejaculatory behavior (Yells et al., 1992).
Using bilateral electrical stimulation of the
dorsal nerve of the penis as the search stimulus,
we found penile-responsive neurons in the gig-
antocellularis complex in anesthetized, uninjured
male rats (Hubscher and Johnson, 1996; see
Hubscher, this volume). The majority of these
neurons exhibited an excitatory response to me-
chanical stimulation of receptive fields on either
side of the glans penis and many exhibited a wind-
up of firing during repeated stimulation as would
occur during copulation. This suggested that a
spino–bulbo–spinal loop for coordination and
control of the lumbosacral ejaculatory reflex cir-
cuitry involved the gigantocellularis nuclear com-
plex. Unilateral microstimulation of neurons in and
adjacent to the lateral paragigantocellularis nucleus
(Fig. 1; left panel) produced a profound bilateral
inhibition (decrease in amplitude and increase in
latency) of the short latency pudendal motoneuron
reflex discharges elicited by stimulation of the
Fig. 1. Schematic representation of two descending pathways that modulate activity of somatic and autonomic motor neurons in the
pudendal nerve, a nerve innervating the perineal muscles involved in the expulsive phase of ejaculation. Data from Johnson and
Hubscher (1998, 2000) and Hubscher and Johnson (1999, 2000) were obtained from anesthetized spinal cord-injured and -uninjured
male rats in which electrodes were placed bilaterally around the dorsal nerve of the penis and pelvic nerve for the stimulation (S) of
segmental sensory afferents and the motor branch of the pudendal nerve for the recording (R) of elicited polysynaptic pudendal motor
neuron reflex discharges and the firing of postganglionic sympathetic fibers. The most robust medullary site (gray regions) for
inhibitory modulation of pudendal motor neuron reflexes (left panel) and activation of pudendal sympathetic fibers (right panel) were
obtained by electrical microstimulation of the lateral paragigantocellularis (LPGi) nuclear region located ventrolateral to the gig-
antocellularis (Gi, GiA) nuclear complex, dorsolateral to the pyramidal (py) tracts and lateral to the raphe magnus (RMg). The
location of each descending pathway in the lateral funiculus of T8 was determined by documenting the effect of microstimulating the
ipsilateral or contralateral medulla after a variety of acute and chronic midthoracic spinal lesions. Bilateral chronic lesions of the dorsal
3/5 of T8 were necessary to eliminate the effects of both pathways. Chronic (30 day), but not acute (several hours) lateral hemisection
lesions revealed evidence of reorganization and/or plasticity. The pudendal reflex inhibitory pathway, normally ipsilateral and un-
crossed above T8 but exerting a bilateral inhibition on pudendal segmental circuits, exhibited crossed inhibition after chronic lateral
hemisection lesions suggesting the development of novel crossed connections above the lesion (dashed lines). In addition, the pudendal
sympathetic activation pathway, normally unilateral and uncrossed below T8, exhibited bilateral excitation of sympathetic post-
ganglionics suggesting the development of novel crossed connections below the lesion. Pudendal motor neurons are located in the
dorsomedial (DM) and dorsolateral (DL) nuclei of the L5–L6 ventral horn. Pudendal sympathetic neurons are located in the in-
termediate zone of the T13–L2 spinal cord. Templates of the medullary and spinal sections modified from Paxinos and Watson (1998).
420
dorsal nerve of the penis (Johnson and Hubscher,
1998). In marked contrast, no inhibitory effects
could be seen on pudendal reflex discharges elic-
ited by pelvic nerve afferents. Bilateral microstim-
ulation was always more effective in depression of
penile afferent-elicited reflexes than unilateral mi-
crostimulation for motoneuron pools on either
side of the cord. Microstimulation (conditioning
stimulus alone) never produced direct firing of pu-
dendal motoneurons and the latency of effect
strongly suggests the lumbosacral site of action
involves presynaptic inhibition of dorsal nerve of
the penis afferents. The function of the descending
inhibitory pathway may be to coordinate, through
the periodic presynaptic inhibition of excitatory
penile afferent input, the proper sequence of pro-
pulsive contraction bursts interfaced with closure
of the bladder neck and emission of semen. The
50–75 ms period of penile afferent inhibition elic-
ited from lateral paragigantocellularis microstim-
ulation corresponds to the period of quiescence
between sequential bursts of contractions of the
bulbospongiosus during ejaculation in the behav-
ing rat (Meisel and Sachs, 1994). Neurons in the
lateral paragigantocellularis nucleus contain se-
rotonin (5-HT; Marson and McKenna, 1992), a
neurotransmitter known to produce presynaptic
inhibition on primary afferent terminals in the
spinal cord (Peng et al., 2001; Schwartz et al.,
2005), and have been shown to project axons
through the spinal cord to terminate in and adja-
cent to the pudendal motor nuclei (Marson and
McKenna, 1992; Hermann et al., 2003). The lack
of inhibitory effect on pelvic nerve-elicited puden-
dal reflexes argues against a direct postsynaptic
inhibition of pudendal motoneurons from these
microstimulation sites. Presynaptic inhibition of
pudendal and urethral afferents of cats has been
shown in the spinal micturition circuit (reviewed
by Shefchyk, 2002) and recent data has shown
5-HT receptors on primary afferent terminals
(Maxwell et al., 2003).
In addition to the descending inhibition of pe-
nile afferent-elicited pudendal motoneuron reflex-
es, sympathetic postganglionic neurons in the
motor branch of the pudendal nerve are strongly
activated by ipsilateral or contralateral microstim-
ulation of a slightly more caudolateral region of
the lateral paragigantocellularis nucleus (Fig. 1,
right panel; Johnson and Hubscher, 1998). As was
true for the pudendal reflex depression sites, si-
multaneous microstimulation of both left and
right sympathetic premotor axons, as opposed to
unilateral stimulation, produced the maximum
level of the long latency sympathetic postgangl-
ionic axon firing, again demonstrating a decussat-
ion across the midline. Terminations of lateral
paragigantocellularis axons have been found in the
intermediate zone of the thoracic cord using an-
terograde (Hermann et al., 2003) and viral tracers
(Stornetta et al., 2004). The unmyelinated sympa-
thetic postganglionic axons in the motor branch of
the pudendal nerve are synaptically connected to
T13–L1 preganglionic neurons in the intermedio-
lateral cell column as shown by pseudorabies viral
tracing from the perineal muscles (Marson and
McKenna, 1996; Marson and Carson, 1999) and
travel through the caudal sympathetic trunk, the
L5 spinal nerve, but not the pelvic nerve (Johnson
and Hubscher, 1998). Although the peripheral tar-
get of the activated sympathetic axons is still un-
known, they likely innervate blood vessels, glands
in the penile bulb, and/or structures in the prox-
imal portions of the corpus cavernosum erectile
tissue (Galindo et al., 1997).
Effects of chronic spinal cord injury on
sexual function
Development of an animal model for the study of
sexual function following chronic spinal cord in-
jury must carefully take into account the level of
injury. Since the spinal cord segments containing
erectile and ejaculatory circuits span the T10–S1
levels in the rat, studies designed to investigate the
effects of removing long ascending or descending
pathways need to avoid injuries within this span
that would disrupt intrinsic spinal segmental cir-
cuits and their putative interconnections as well as
produce the equivalent of lower motor neuron
signs. Historically, studies of the effects of chronic
(at least 30 day) spinal cord transection in rats
have used injuries to the midthoracic (T6–T9) spi-
nal cord, rostral to the autonomic centers for
erection and emission, and caudal to the T1–T5

levels within which severe injuries could produce
complications from autonomic dysreflexia (re-
viewed by Weaver, this volume). It is also impor-
tant to use postlesion intervals of at least 30 days
(chronic lesion), since studies on acute injuries
(hours to a few days) are subject to lingering ef-
fects of spinal shock and active plasticity, show
vastly different results when compared to the
chronic lesion, and do not provide the degree of
clinical correlation to chronic injuries in humans.
Early behavioral studies on rats with spinal cord
transections at the midthoracic level for approxi-
mately 30 days (Sachs and Garinello, 1979; Hart
and Odell, 1981; Mas et al., 1987) documented
enhanced erectile and depressed ejaculatory reflex-
es due to loss of supraspinal influences. Dorsal
nerve of the penis afferents that are required for
ejaculation and reflexogenic erection likely play a
role in the reorganization of spinal circuits follow-
ing chronic injury. In normal unoperated animals
and in those with an acute 4–6 h T8 transection,
the magnitude of the pudendal reflex evoked by
supramaximal stimulus strength of the penile
afferent nerves was significantly greater than
that evoked by pelvic nerve afferent stimulation
(Johnson, 1995). This ratio changed over the
course of 30–60 days after transection such that
penile afferent stimulation became increasingly less
effective than pelvic afferent stimulation in eliciting
a reflex discharge. The decrease in synaptic efficacy
of penile afferents was particularly pronounced on
the contralateral (crossed) activation of pudendal
motoneurons. Although the pelvic nerve afferent
synaptic efficacy did not decrease and may have
increased slightly, the contralateral changes were
not seen with this input circuit.
At 30 and 60 days after spinal cord transection,
a reduction in synaptic efficacy of dorsal nerve of
the penis afferents was coupled with the relative
strengthening of the pelvic nerve afferent inputs
suggesting that some descending pathway(s), likely
from the brainstem, normally exerts some type of
selective facilitation of pudendal afferent input
onto pudendal motoneuron circuitry. Elimination
of that facilitatory influence after transection
injury may have left those circuits approachable
by other segmental systems, as demonstrated by
Beattie et al. (1993) in the cat, including possibly
421
the visceral afferent inputs originating in the pelvic
nerve. This reorganization of inputs on the pu-
dendal motor neuron circuits likely takes time as
shown by the lack of immediate effect of an acute
transection (4–6 h). Alternatively, injury discharg-
es and spinal shock may have contributed to some
of the acute–chronic injury differences.
Utilizing the electrophysiological model for in-
vestigating the spino–bulbo–spinal loop (Fig. 1),
various degrees of chronic (30 day) incomplete le-
sions of the midthoracic cord were used to deter-
mine the bilateral nature of the descending
pudendal inhibitory and sympathetic facilitatory
pathways, the midthoracic white matter pathway
location, and the behavioral effect on sexual re-
flexes (Hubscher and Johnson, 2000). As expected,
lateral midthoracic hemisections did not eliminate
the effects of microstimulating either descending
pathway from the medulla, corroborating the bi-
lateral terminal architecture in electrophysiological
(Johnson and Hubscher, 1998, 2000) and neuro-
anatomical (Hermann et al., 2003) studies, and
had a negligible effect on sexual or bladder reflex-
es. However, chronic lateral hemisections revealed
a reorganization of both descending pathways that
may have contributed to the functional recovery.
The pudendal reflex inhibitory pathway, normally
ipsilateral and uncrossed above T8 but exerting a
bilateral inhibition on pudendal segmental circuits,
exhibited crossed inhibition after chronic lateral
hemisection lesions suggesting the development of
novel crossed connections above the lesion (Fig. 1;
broken lines in left panel). In addition, the puden-
dal sympathetic activation pathway, normally uni-
lateral and uncrossed below T8, exhibited bilateral
excitation of sympathetic postganglionics suggest-
ing the development of novel crossed connections
below the lesion (Fig. 1; broken lines in right pan-
el). These novel synaptic connections were not seen
in acute (4–6 h) lateral hemisections (Hubscher
and Johnson, 2000). The new connections revealed
across the thoracic spinal cord below the level of a
chronic lesion may be due a reorganization of seg-
mental autonomic circuits as described by Weaver
et al. (1997, 2001).
In contrast to the spared and novel bilateral
connections following chronic lateral hemisections,
severe contusion or dorsal hemisection injuries
422
(dorsal 3/5 of the cord) eliminated the effects of
both descending pathways and produced a behavi-
oral status similar to spinal transection; reduction
of erectile reflex initiation latency, and the devel-
opment of bladder-sphincter dyssynergia (de
Groat, 1995). It is important to note that func-
tional behavioral measures of mating-triggered
ejaculation cannot be performed because of the
inability of the severely injured rat to physically
mount a receptive female. By reconstructing and
overlapping the lesion extent of the variable acute
and chronic injuries (Hubscher and Johnson,
2000), the approximate lateral white matter loca-
tion of these two pathways was determined (Fig. 1)
in general agreement with neuroanatomical tracing
studies on normal animals (Martin et al., 1985;
Hermann et al., 2003). In support of this finding,
pseudorabies virus injections into the perineal mus-
cles failed to infect neurons in the gigantocellularis
complex 30 days after lesions of the dorsal 3/5 of
the T8 spinal cord (Chadha et al., 2004).
Chronic spinal cord injury compromises normal
sexual function and the disruption of the descend-
ing pathways modulating the pudendal moto-
neuron reflex circuit likely contributes to the
incoordination of perineal muscle contractions.
The inappropriate tonic contractions of these mus-
cles following severe chronic injury likely interfere
with the timing of phasic events mediating elim-
inative functions such as ejaculation, defecation,
and micturition. The differential descending inhi-
bition of penile nerve but not pelvic nerve inputs to
segmental reflex circuits may correlate with the
finding that the segmental inputs from the penile
nerve but not the pelvic nerve become progres-
sively ineffective following long-term transection
injury, possibly resulting from a reorganization of
input-specific interneurons.
Clinical correlations in men with spinal injury
Studies of sexual function in men with spinal cord
injury (reviewed by Elliot, this volume) have
documented complications, signs, and treatment
strategies that correlate with data obtained in ani-
mal studies. In some instances, in which the verbal
descriptions of sensations (e.g. orgasm) or emotional
feelings (e.g. arousal, libido) cannot be obtained in
animal models, the clinical picture can greatly
expand the interpretation of neuronal mechanisms.
The great majority of men with severe lesions
cranial to the T10 spinal segment (i) can obtain
reflexogenic (parasympathetically mediated) erec-
tions but usually lack the desired penile rigidity
and/or duration even under high states of arousal
(Szasz and Carpenter, 1989), (ii) cannot ejaculate
or sense genital-based orgasm (Donohue and
Gebhard, 1995), and (iii) often have difficulty in
controlling urethral and anal sphincters during
sexual activity (Elliott, 2003). All these scenarios
point to inappropriate, weak, or uncoordinated
perineal muscle contractions likely due to the loss
of the spinal–bulbo–spinal or segmental coordina-
tion pathways described above.
By activating the afferent fibers in the dorsal
nerve of the penis with high-intensity penile vibra-
tory stimulation (PVS), an ejaculation sufficient
for sperm collection can often be elicited (Sønksen
and Ohl, 2002; Brackett et al., 1997), provided the
T11–S4 spinal cord, the dorsal nerve of the penis,
and the bulbocavernosus reflex are intact (Szasz
and Carpenter, 1989; Wieder et al., 2000; Bird
et al., 2001). Thus it would seem that the penile
afferent input to the pudendal reflex circuitry and
the spinal ejaculation centers inclusive to the T11
segment, are crucial for the application of this
therapy. The high magnitude and long duration of
PVS needed for optimum effects (2.5 mm displace-
ment; Sønksen et al., 1994) are beyond the stim-
ulation intensity needed for reflexogenic erection
and normal sensation in able-bodied men, sug-
gesting that the central synaptic efficacy of penile
afferents gradually diminishes after injury, similar
to the data obtained in rat studies.
Another therapeutic strategy, electroejaculation,
involves the stimulation of the pelvic nerve affer-
ents through the wall of the rectum (Halstead
et al., 1987). When successful, this procedure trig-
gers emission but not the expulsive phase of ejac-
ulation and often produces retrograde emission
into the bladder (Biering-Sorensen and Sønksen,
2001). Therefore, as in the spinal cord-injured rat,
an increased synaptic efficacy of pelvic nerve af-
ferents on segmental ejaculatory circuits may oc-
cur in chronic injury. However, because of

unpleasant sensations, electroejaculation can only
be applied under anesthesia if any residual as-
cending sensory pathways are retained (Sønksen
and Ohl, 2002), presumably in the ventral half of
the spinal cord. PVS and electroejaculation utilize
the dorsal nerve of the penis and pelvic nerve sen-
sory afferents, respectively, to activate autonomic
circuits in the thoracolumbar spinal cord including
the proposed spinal ejaculation center (at L3–L4 in
the rat; Truitt and Coolen, 2002). If the lesion is
cranial to T7, however, autonomic dysreflexia may
be produced by either procedure (Frankel and
Mathias, 1980; Szasz and Carpenter, 1989) or by
stimulation of the dorsal nerve of the penis (Reitz
et al., 2003). A recent study has demonstrated
positive sympathetic skin (sudomotor) responses
in the hand following stimulation of the urethra in
men with incomplete spinal cord injuries caudal to
the upper thoracic levels (Schmid et al., 2004).
Recording a sympathetic skin response in the hand
following stimulation of the penile or pelvic nerves
may become useful in determining whether upper
thoracic sympathetic centers can be activated by
ascending intraspinal pathways through the lesion.
Pharmacological considerations
The neurotransmitters, neuromodulators, and re-
ceptor subtypes identified in the neurons subserv-
ing erection and ejaculation are numerous (see
Giuliano and Rampin, 2004). Some of these are
likely important for the mechanisms described
above. Several animal studies have demonstrated
that serotonin (5-HT), released by several descend-
ing brainstem pathways, is involved through acti-
vation of certain receptor subtypes primarily as an
inhibitory neurotransmitter (see McKenna, 2000;
Marson and Gravitt, 2004). The 5-HT1A receptor,
however, facilitates ejaculation in rats (Carro-
Juarez et al., 2003; Truitt et al., 2003) and has been
proposed to be hypersensitive in men with prema-
ture ejaculation (Waldinger, 2004). The 5-HT2C
receptor has been found on perineal motor
neurons (Bancila et al., 1999) along with the 5-
HT5A receptor (Doly et al., 2004) and may inhibit
ejaculation (Waldinger, 2004). Serotonin release
from brainstem pathways has been shown to pro-
duce presynaptic inhibition of spinal primary
423
afferent terminals (Peng et al., 2001; Schwartz
et al., 2005) possibly through the 5-HT3 receptor
found on primary afferent terminals (Maxwell
et al., 2003). In addition to serotonin, studies by
Holmes et al. in normal and spinal-injured rats
(2001) describe a medullary-spinal pathway from
the nucleus raphe obscurus which releases thyrotro-
phin releasing hormone on pudendal motor neurons
and acts synergistically with serotonin to specifically
inhibit sexual and anal sphincter reflexes.
The inhibitory neurotransmitter gamma amino-
butyric acid (GABA), found in some of the
spinally projecting neurons of the lateral par-
agigantocellularis nucleus (Jones et al., 1991) as
well as in spinal interneurons, likely plays a role in
ejaculatory dysfunction after chronic spinal cord
injury. The GABAB receptor is found on primary
afferent terminals, mediates presynaptic inhibi-
tion, and is the site of action of baclofen
(Li et al., 2004). In rats and humans, baclofen
administration inhibits ejaculation and erection of
the glans penis (Leipheimer and Sachs, 1988;
Vaidyanathan et al., 2004) possibly through the
tonic presynaptic inhibition of penile afferents.
Summary
The ejaculatory dysfunction experienced by the
vast majority of spinal cord-injured men is likely
caused by the removal of supraspinal modulatory
pathways that normally act to coordinate the ac-
tivity in the ejaculatory spinal circuits and spinal
pattern generator. Continued development of an-
imal models for the investigation of electrophys-
iological, neuroanatomical, pharmacological, and
behavioral characteristics of male sexual responses
following chronic spinal cord injury will provide
valuable information about a complex visceroso-
matic system dependent on the rhythmic and co-
ordinated contraction of smooth and striated
muscles. Injury-induced alteration of segmental
and descending control of the spinal circuits sub-
serving somatic and autonomic elements of the
perineal muscles, including the development of nov-
el connections, involves spinal pathways from the
lateral paragigantocellularis nucleus in the medulla
acting on reflex inputs from the dorsal nerve of the
penis and associated autonomic interneurons.
424
Acknowledgments
Supported by grants from the National Institutes
of Health, the Christopher Reeve Paralysis Asso-
ciation, the Paralyzed Veterans of America, and
the Brain and Spinal Cord Injury Trust Fund of
Florida.
References
Anderson, K.D. (2004) Targeting recovery: priorities of the
spinal cord-injured population. J. Neurotrauma, 21:
1371–1383.
Bancila, M., Verge, D., Rampin, O., Backstrom, J.R.,
Sanders-Bush, E., McKenna, K.E., Marson, L., Calas, A.
and Giuliano, F. (1999) 5-Hydroxytryptamine2C receptors
on spinal neurons controlling penile erection in the rat.
Neuroscience, 92: 1523–1537.
Beattie, M.S., Leedy, M.G. and Bresnahan, J.C. (1993)
Evidence for alterations of synaptic inputs to sacral spinal
reflex circuits after spinal cord transection in the cat. Exp.
Neurol., 123: 35–50.
Biering-Sorensen, F. and Sønksen, J. (2001) Sexual function in
spinal cord lesioned men. Spinal Cord, 39: 455–470.
Bird, V.G., Brackett, N.L., Lynne, C.M., Aballa, T.C. and
Ferrell, S.M. (2001) Reflexes and somatic responses as
predictors of ejaculation by penile vibratory stimulation in
men with spinal cord injury. Spinal Cord, 39: 514–519.
Bowens, J.M., Hughes, B.J. and Bradley, W.E. (1984)
Anatomic, physiologic, and pharmacologic properties of the
feline bulbospongiosus muscle. Am. J. Vet. Res., 45: 308–313.
Brackett, N.L., Padron, O.F. and Lynne, C.M. (1997) Semen
quality of spinal cord injured men is better when obtained by
vibratory stimulation versus electroejaculation. J. Urol., 157:
151.
Breedlove, S.M. and Arnold, A.P. (1980) Hormone accumula-
tion in a sexually dimorphic motor nucleus of the rat spinal
cord. Science, 210: 564–566.
Carro-Juarez, M., Cruz, S.L. and Rodriguez-Manzo, G. (2003)
Evidence for the involvement of a spinal pattern generator in
the control of the genital motor pattern of ejaculation. Brain
Res., 975: 222–228.
Carvalho, M., Schwartz, M.S. and Swash, M. (1995) Involve-
ment of the external anal sphincter in amyotrophic lateral
sclerosis. Muscle Nerve, 18: 848–853.
Chadha, H.K., Johnson, R.D. and Hubscher, C.H. (2004)
Effects of chronic partial mid-thoracic spinal cord injury on
the transneuronal tracing of pseudorabies virus from the
perineal musculature to the brainstem. Soc. Neurosci. Abstr.,
30: 458.16.
Coleman, A.M. and Sengelaub, D.R. (2002) Patterns of dye
coupling in lumbar motor nuclei of the rat. J. Comp. Neurol.,
454: 34–41.
Collins III, W.F., Erichsen, J.T. and Rose, R.D. (1991) Puden-
dal motor and premotor neurons in the male rat: A WGA
transneuronal study. J. Comp. Neurol., 308: 28–41.
Coolen, L.M., Allard, J., Truitt, W.A. and McKenna, K.E. (2004)
Central regulation of ejaculation. Physiol. Behav., 83: 203–215.
de Groat, W.C. (1995) Mechanisms underlying the recovery of
lower urinary tract function following spinal cord injury.
Paraplegia, 33: 493–505.
Doly, S., Fischer, J., Brisorgueil, M.J., Verge, D. and Conrath,
M. (2004) 5-HT5A receptor localization in the rat spinal cord
suggests a role in nociception and control of pelvic floor
musculature. J. Comp. Neurol., 476: 316–329.
Donohue, J. and Gebhard, P. (1995) The Kinsey Institute/In-
diana University report on sexual and spinal cord injury. Sex.
Disabil., 13(1): 7–84.
Dubrovsky, B. and Filipini, D. (1990) Neurobiological aspects
of the pelvic floor muscles involved in defecation. Neurosci.
Biobehav. Rev., 14: 157–168.
Elliott, S. (2003) Sexual dysfunction and infertility in men with
spinal cord disorders. In: Lin V. (Ed.), Spinal Cord Medicine:
Principles and Practice (Chapter 26). Demos Medical Pub-
lishing, New York, pp. 350–365.
Frankel, H.L. and Mathias, C.J. (1980) Severe hypertension in
patients with high spinal cord lesions undergoing electro-
ejaculation–management with prostaglandin E2. Paraplegia,
18: 293–299.
Galindo, R., Barba, V. and Dail, W.G. (1997) The sensory
branch of the pudendal nerve is the major route for adrenergic
innervation of the penis in the rat. Anat. Rec., 247: 479–485.
Giuliano, F. and Rampin, O. (2004) Neural control of erection.
Physiol. Behav., 83(2): 189–201.
Halstead, L.S., VerVoort, S. and Seager, S.W. (1987) Rectal
probe electrostimulation in the treatment of anejaculatory
spinal cord injured men. Paraplegia, 25: 120–129.
Hart, B.L. and Leedy, M.G. (1985) Neurological bases of male
sexual behavior: a comparative analysis. In: Adler N., Pfaff
D. and Goy R.W. (Eds.) Handbook of Behavioral Neuro-
biology, Vol. 7. Reproduction. Plenum Press, New York,
pp. 373–422.
Hart, B.L. and Odell, V. (1981) Elicitation of ejaculation and
penile reflexes in spinal male rats by peripheral electric shock.
Physiol. Behav., 26: 623–626.
Hermann, G.E., Bresnahan, J.C., Holmes, G.M., Rogers, R.C.
and Beattie, M.S. (1998) Descending projections from the
nucleus raphe obscurus to pudendal motoneurons in the male
rat. J. Comp. Neurol., 397: 458–474.
Hermann, G.E., Holmes, G.M., Rogers, R.C., Beattie, M.S.
and Bresnahan, J.C. (2003) Descending spinal projections
from the rostral gigantocellular reticular nuclei complex.
J. Comp. Neurol., 455: 210–221.
Holmes, G.M., Bresnahan, J.C. and Beattie, M.S. (2001) Inhi-
bition of pudendal reflexes in spinal rats. Reassessing the role
of serotonin. Physiol. Behav., 74: 57–64.
Holmes, G.M., Chapple, W.D., Leipheimer, R.E. and Sachs,
B.D. (1991) Electromyographic analysis of male rat perineal
muscles during copulation and reflexive erections. Physiol.
Behav., 49: 1235–1246.

Holmes, G.M., Hermann, G.E., Rogers, R.C., Bresnahan, J.C.
and Beattie, M.S. (2002) Dissociation of the effects of
nucleus raphe obscurus or rostral ventrolateral medulla
lesions on eliminatory and sexual reflexes. Physiol. Behav.,
75: 49–55.
Holstege, G. (1991) Descending motor pathways and the spinal
motor system: limbic and non-limbic components. Prog.
Brain Res., 87: 307–421.
Holstege, G. and Georgiadis, J.R. (2004) The emotional brain:
neural correlates of cat sexual behavior and human male
ejaculation. Prog. Brain Res., 143: 39–45.
Hubscher, C.H. and Johnson, R.D. (1996) Responses of med-
ullary reticular formation neurons to input from the male
genitalia. J. Neurophysiol., 76: 2474–2482.
Hubscher, C.H. and Johnson, R.D. (1999) Effects of acute and
chronic midthoracic spinal cord injury on neural circuits for
male sexual function. I. Ascending pathways. J. Neuro-
physiol., 82: 1381–1389.
Hubscher, C.H. and Johnson, R.D. (2000) Effects of acute and
chronic midthoracic spinal cord injury on neural circuits for
male sexual function. II. Descending pathways. J. Neuro-
physiol., 83: 2508–2518.
Johnson, R.D. (1988) Efferent modulation of penile me-
chanoreceptor activity. In: Iggo A. and Hamann W. (Eds.)
Progress in Brain Research, Vol. 73. Transduction and Cel-
lular Mechanisms in Sensory Receptors. Elsevier, Amster-
dam, pp. 319–324.
Johnson, R.D. (1995) Effects of chronic spinal cord injury on
segmental circuits controlling eliminative and sexual func-
tions. J. Neurotrauma, 12: 359.
Johnson, R.D. and Halata, Z. (1991) Topography and ultra-
structure of sensory nerve endings in the glans penis of the
rat. J. Comp. Neurol., 312: 299–310.
Johnson, R.D. and Hubscher, C.H. (1998) Brainstem micro-
stimulation differentially inhibits pudendal motoneuron
reflex inputs. Neuroreport, 9: 341–345.
Johnson, R.D. and Hubscher, C.H. (2000) Brainstem micro-
stimulation activates sympathetic fibers in pudendal nerve
motor branch. Neuroreport, 11: 1–4.
Johnson, R.D. and Murray, F.T. (1992) Reduced sensitivity of
penile mechanoreceptors in aging rats with sexual dysfunc-
tion. Brain Res. Bull., 28: 61–64.
Jones, B.E., Holmes, C.J., Rodriguez-Viega, E. and Mainville,
L. (1991) GABA-synthesizing neurons in the medulla: their
relationship to serotonin-containing and spinally projecting
neurons in the rat. J. Comp. Neurol., 312: 1–19.
Kippin, T.E., Sotiropoulos, V., Badih, J. and Pfaus, J.G. (2004)
Opposing roles of the nucleus accumbens and anterior lateral
hypothalamic area in the control of sexual behaviour in the
male rat. Eur. J. Neurosci., 19: 698–704.
Leipheimer, R.E. and Sachs, B.D. (1988) GABAergic regula-
tion of penile reflexes and copulation in rats. Physiol. Behav.,
42: 351–357.
Li, Y., Li, X., Harvey, P.J. and Bennett, D.J. (2004) Effects of
baclofen on spinal reflexes and persistent inward currents in
motoneurons of chronic spinal rats with spasticity. J. Neuro-
physiol., 92: 2694–2703.
425
Mann, K., Pankok, J., Connemann, B. and Roschke, J. (2003)
Temporal relationship between nocturnal erections and rapid
eye movement episodes in healthy men. Neuropsychobiol.,
47: 109–114.
Markowski, V.P., Eaton, R.C., Lumley, L.A., Moses, J. and
Hull, E.M. (1994) A D1 agonist in the MPOA facilitates
copulation in male rats. Pharmacol. Biochem. Behav., 47:
483–486.
Marson, L. and Carson III, C.C. (1999) Central nervous system
innervation of the penis, prostate, and perineal muscles:
a transneuronal tracing study. Mol. Urol., 3: 43–50.
Marson, L. and Gravitt, K. (2004) Spinal neurons activated
with the urethrogenital reflex in the male rat. Brain Res.,
1026: 108–115.
Marson, L. and McKenna, K.E. (1990) The identification of a
brainstem site controlling spinal sexual reflexes in male rats.
Brain Res., 515: 303–308.
Marson, L. and McKenna, K.E. (1992) A role for 5-hydroxy-
tryptamine in descending inhibition of spinal sexual reflexes.
Exp. Brain Res., 88: 313–320.
Marson, L. and McKenna, K.E. (1994) Stimulation of the
hypothalamus initiates the urethrogenital reflex in male rats.
Brain Res., 638: 103–108.
Marson, L. and McKenna, K.E. (1996) CNS cell groups
involved in the control of ischiocavernosus and bulb-
ospongiosus muscles: a transneuronal tracing study using
pseudorabies virus. J. Comp. Neurol., 374: 161–179.
Martin, G.F., Vertes, R.P. and Waltzer, R. (1985) Spinal pro-
jections of the gigantocellular reticular formation in the rat.
Evidence for projections from different areas to laminae
I and II and lamina IX. Exp. Brain Res., 58: 154–162.
Mas, M., del Castillo, A.R., Guerra, M., Davidson, J.M. and
Battaner, E. (1987) Neurochemical correlates of male sexual
behavior. Physiol. Behav., 41: 341–345.
Mason, P. (2001) Contributions of the medullary raphe and
ventromedial reticular region to pain modulation and other
homeostatic functions. Annu. Rev. Neurosci., 24: 737–777.
Maxwell, D.J., Kerr, R., Rashid, S. and Anderson, E. (2003)
Characterisation of axon terminals in the rat dorsal horn that
are immunoreactive for serotonin 5-HT3A receptor subunits.
Exp. Brain Res., 149: 114–124.
McKenna, K.E. (2000) Some proposals regarding the organi-
zation of the central nervous system control of penile erec-
tion. Neurosci. Biobehav. Rev., 24: 535–540.
McKenna, K.E., Chung, S.K. and McVary, K.T. (1991) A
model for the study of sexual reflexes in anesthetized male
and female rats. Am. J. Physiol., 261: R1276–R1285.
McKenna, K.E. and Nadelhaft, I. (1986) The organization of
the pudendal nerve in the male and female rat. J. Comp.
Neurol., 248: 532–549.
McKenna, K.E. and Nadelhaft, I. (1989) The pudendo-puden-
dal reflex in male and female rats. J. Autonom. Nerv. Syst.,
27: 67–77.
Meisel, R.L. and Sachs, B.D. (1994) The physiology of male
sexual behavior. In: Knobil E. and Neill J.D.D. (Eds.) The
Physiology of Reproduction, Vol. 2. Raven Press, New York,
pp. 3–105.
426
Murphy, A.Z. and Hoffman, G.E. (2001) Distribution of go-
nadal steroid receptor-containing neurons in the preoptic-
periaqueductal gray-brainstem pathway: a potential circuit
for the initiation of male sexual behavior. J. Comp. Neurol.,
438: 191–212.
Nadelhaft, I. and McKenna, K.E. (1987) Sexual dimorphism in
sympathetic preganglionic neurons of the rat hypogastric
nerve. J. Comp. Neurol., 256: 308–315.
Nunez, R., Gross, G.H. and Sachs, B.D. (1986) Origin and
central projections of rat dorsal penile nerve: possible direct
projection to autonomic and somatic neurons by primary af-
ferents of nonmuscle origin. J. Comp. Neurol., 247: 417–429.
Ohl, D.A., Bennett, C.J., McCabe, M., Menge, A.C. and
McGuire, E.J. (1989) Predictors of success in electroejacula-
tion of spinal cord injured men. J. Urol., 142: 1483–1486.
Paxinos, G. and Watson, C. (1998) The Rat Brain In Stereo-
taxic Coordinates (4th ed.). Academic Press, San Diego.
Pehek, E.A., Thompson, J.T. and Hull, E.M. (1989) The effects
of intracranial administration of the dopamine agonist apo-
morphine on penile reflexes and seminal emission in the rat.
Brain Res., 500: 325–332.
Peng, Y.B., Wu, J., Willis, W.D. and Kenshalo, D.R. (2001)
GABA(A) and 5-HT(3) receptors are involved in dorsal root
reflexes: possible role in periaqueductal gray descending in-
hibition. J. Neurophysiol., 86: 49–58.
Reitz, A., Schmid, D.M., Curt, A., Knapp, P.A. and Schurch,
B. (2003) Autonomic dysreflexia in response to pudendal
nerve stimulation. Spinal Cord, 41: 539–542.
Sachs, B.D. and Garinello, L.D. (1979) Spinal pacemaker con-
trolling sexual reflexes in male rats. Brain Res., 171: 152–156.
Schmid, D.M., Reitz, A., Curt, A., Hauri, D. and Schurch, B.
(2004) Urethral evoked sympathetic skin responses and vis-
cerosensory evoked potentials as diagnostic tools to evaluate
urogenital autonomic afferent innervation in spinal cord in-
jured patients. J. Urol., 171: 1156–1160.
Schmidt, M.H. and Schmidt, H.S. (1993) The ischiocavernosus
and bulbospongiosus muscles in mammalian penile rigidity.
Sleep, 16: 171–183.
Schroder, H.D. (1985) Anatomical and pathoanatomical studies
on the spinal efferent systems innervating pelvic structures.
1. Organization of spinal nuclei in animals. 2. The nucleus
X-pelvic motor system in man. J. Auton. Nerv. Syst., 14: 23–48.
Schwartz, E.J., Gerachshenko, T. and Alford, S. (2005) 5-HT
prolongs ventral root bursting via presynaptic inhibition
of synaptic activity during fictive locomotion in lamprey.
J. Neurophysiol., 93: 980–988.
Shefchyk, S.J. (2002) Spinal cord neural organization control-
ling the urinary bladder and striated sphincter. Prog. Brain
Res., 137: 71–82.
Sønksen, J., Biering-Sorensen, F. and Kristensen, J.K. (1994)
Ejaculation by penile vibratory stimulation in men with
spinal cord injured. Paraplegia, 32: 651–660.
Sønksen, J. and Ohl, D.A. (2002) Penile vibratory stimulation
and electroejaculation in the treatment of ejaculatory dys-
function. Int. J. Androl., 25: 324–332.
Steers, W.D. (2000) Neural pathways and central sites involved
in penile erection: neuroanatomy and clinical implications.
Neurosci. Biobehav. Rev., 24: 507–516.
Stornetta, R.L., McQuiston, T.J. and Guyenet, P.G. (2004)
GABAergic and glycinergic presympathetic neurons of rat
medulla oblongata identified by retrograde transport of
pseudorabies virus and in situ hybridization. J. Comp.
Neurol., 479: 257–270.
Sung, J.H., Mastri, A.R. and Segal, E. (1979) Pathology of
Shy-Drager syndrome. J. Neuropathol. Exp. Neurol., 38:
353–368.
Szasz, G. and Carpenter, C. (1989) Clinical observations in vi-
bratory stimulation of the penis in men with spinal cord in-
jury. Arch. Sexual Behav., 18: 461–474.
Tanaka, J. and Arnold, A.P. (1993) An electrophysiological
study of descending projections to the lumbar spinal cord in
adult male rats. Exp. Brain Res., 96: 117–124.
Truitt, W.A. and Coolen, L.M. (2002) Identification of a
potential ejaculation generator in the spinal cord. Science,
297: 1566–1569.
Truitt, W.A., Shipley, M.T., Veening, J.G. and Coolen, L.M.
(2003) Activation of a subset of lumbar spinothalamic neu-
rons after copulatory behavior in male but not female rats.
J. neurosci., 23: 325–331.
Vaidyanathan, S., Watt, J.W., Singh, G., Hughes, P.L.,
Selmi, F., Oo, T., Soni, B.M. and Sett, P. (2004) Man-
agement of recurrent priapism in a cervical spinal cord
injury patient with oral baclofen therapy. Spinal Cord,
42: 134–135.
Wagner, C.K. and Clemens, L.G. (1991) Projections of the
paraventricular nucleus of the hypothalamus to the sexually
dimorphic lumbosacral region of the spinal cord. Brain Res.,
539: 254–262.
Waldinger, M.D. (2004) Lifelong premature ejaculation: from
authority-based to evidence-based medicine. BJU Int., 93:
201–207.
Weaver, L.C., Cassam, A.K., Krassioukov, A.V. and
Llewellyn-Smith, I.J. (1997) Changes in immunoreactivity
for growth associated protein-43 suggest reorganization of
synapses on spinal sympathetic neurons after cord transec-
tion. Neuroscience, 81: 535–551.
Weaver, L.C., Verghese, P., Bruce, J.C., Fehlings, M.G., Krenz,
N.R. and Marsh, D.R. (2001) Autonomic dysreflexia and
primary afferent sprouting after clip-compression injury of
the rat spinal cord. J. Neurotrauma, 18: 1107–1119.
Wieder, J.A., Brackett, N.L., Lynne, C.M., Green, J.T. and
Aballa, T.C. (2000) Anesthetic block of the dorsal penile
nerve inhibits vibratory-induced ejaculation in men with spi-
nal cord injuries. Urology, 55: 915–917.
Yang, C.C. and Bradley, W.E. (1999) Somatic innervation of
the human bulbocavernosus muscle. Clin. Neurophysiol.,
110: 412–418.
Yells, D.P., Hendricks, S.E. and Prendergast, M.A. (1992)
Lesions of the nucleus paragigantocellularis: effects on
mating behavior in male rats. Brain Res., 596: 73–79.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 29

Male fertility and sexual function after spinal

cord injury

D.J. Brown1,, S.T. Hill1,2 and H.W.G. Baker2,3

1
Victorian Spinal Cord Service, Austin Health, Heidelberg, Vic., Australia
2
Melbourne IVF Reproductive Services, Royal Women’s Hospital, Melbourne, Vic., Australia
3
University of Melbourne Department of Obstetrics and Gynaecology, Carlton, Vic. 3058, Australia

Abstract: Spinal cord injury has an enormous impact upon the sexual relationship of a man and his
partner. Erection may be partial or absent, orgasm altered or impossible, and fertility severely impaired.
New understanding of the physiology of sexual function and improved treatment can enable most cord-
injured men to achieve erections suitable for sexual satisfaction. Modern methods of sperm collection and
fertility treatment mean that many can also be fathers. The best results are obtained by a team approach
involving rehabilitation and reproductive medicine clinicians, nurses, spinal cord injury specialists and
counselors with the cord-injured man and his partner. Erections can be achieved by drugs, such as si-
ldenafil, that block phosphodiesterase 5, prolonging the action of nitric oxide with resultant smooth muscle
relaxation. Intracavernosal prostaglandin E1 and mechanical systems, such as vacuum pumps and con-
striction rings, are also effective. Sexual gratification can be promoted in the context of an understanding
relationship in which the cord-injured person can gain pleasure from pleasing his partner and also from his
partner’s exploration of erotogenic areas not affected by the spinal cord injury. An emphasis on the broader
view of sexuality in relationships allows for a continuance and strengthening of bonds between the couple.
Vibration ejaculation or electroejaculation can be used to collect semen. For a limited period in the acute
phase, usually for about 6–12 days after injury, normal semen can be obtained by electroejaculation from
some cord-injured men. With chronic spinal cord injury the semen is of variable quality. Some patients have
necrospermia, which may be improved by regular ejaculation. Others have poor quality semen or sperm-
atogenic disorders and, in this situation, in vitro fertilization techniques must be used to achieve parent-
hood. Trials of assisted ejaculation help individualize cost-effective management of the infertility.

Introduction significant impact upon the man and his partner.

‘‘Sexual disturbances, which inevitably follow a se-
vere injury of the spinal cord, regardless of whether
the lesion is complete or incomplete, constitute a
complex problem in the rehabilitation of paraplegics
and even more in tetraplegics’’ (Guttmann, 1973).
The profound changes in sexual function and fer-
tility in men after spinal cord injury have a very
Corresponding author. Tel.: +94963030; Fax: +9496 3626;
E-mail: DouglasJ.Brown@austin.org.au
In younger people in particular, the bigger picture
of sexuality and personal relationship is lost as
they focus on sexual performance and hopes of
parenthood. To the cord-injured man and his
partner, the closely integrated facets of erection,
ejaculation and orgasm become separated and,
with this, the prospects of sexual pleasure and
parenthood seem like lost dreams. The loss of
erections suitable for intercourse is a major blow
to the cord-injured man’s self-esteem and self-con-
fidence and can lead to a belief that no woman

DOI: 10.1016/S0079-6123(05)52029-6 427

428
would want him. The prospect of infertility in
a partner also affects the woman’s hopes and
expectations. The more important coitus and hav-
ing children are to a cord-injured person and his
partner, the more profound the impact on their
relationship.
And so it was. But now the scene has changed
almost beyond recognition due to improved un-
derstanding of the physiology and pathophysiolo-
gy of sexual function. The pioneering work of
people like Giles Brindley in the treatment of im-
potence and the development of in vitro fertiliza-
tion programs, in particular intracytoplasmic
sperm injection, have given hope of an improved
sexual life and of parenthood (Brindley, 1981a, b,
1982, 1983a, 1983b, 1984, 1988; Brindley et al.,
1982; Schatte et al., 2000). The introduction of
sexual counseling about these developments dur-
ing rehabilitation has given new hope to cord-in-
jured individuals and their partners.
Modern management of sexual dysfunction fol-
lowing spinal cord injury is based upon better un-
derstanding of the physiology of erection, the
development of sophisticated techniques for re-
trieval of sperm and the achievement of fertiliza-
tion in spite of highly adverse circumstances.
However, our understanding of the mechanisms
whereby disruption of the nervous system causes
such profound sexual disability remains poor. We
still do not know clearly how erection, ejaculation
and orgasm are normally integrated, nor do we
know the cause of male infertility after spinal cord
injury. Treatments are therefore incomplete.
Male fertility and sexual function are dramati-
cally impaired by spinal cord injury with less than
5% of men able to procreate without medical in-
tervention (Talbot, 1955; Thomas, 1983). Modern
vibroejaculation and electroejaculation equipment
allows sperm to be collected from the majority of
cord-injured men (Lim et al., 1994; Brackett,
1999). However, with chronic spinal cord injury,
the semen quality is variable and generally poor;
particularly, there is necrospermia (low sperm mo-
tility and viability, Mallidis et al., 2000). The fact
that collected sperm is normal in the acute phase
of spinal cord injury indicates that this defect takes
some time to develop (Mallidis et al., 1994).
Research aimed at discovering the mechanism of
the acquired necrospermia suggests that genital
tract stasis is the main cause but a number of
unresolved questions remain (Mallidis et al., 2000).
This work has also aided understanding the devel-
opment of sperm defects, particularly necrospermia,
in able-bodied men. In this chapter we review the
pathophysiology of autonomic dysfunction of the
male reproductive tract after spinal cord injury,
previous studies of the associated testicular and
sperm defects, studies on necrospermia after chron-
ic spinal cord injury, a possible mechanism of the
development of the common sperm defects present
after cord injury and clinical approaches for the
management of the male infertility and sexual dys-
function after spinal cord injury.
Pathophysiology of autonomic dysfunction affecting
male fertility and sexual function
The names of Bors and Comarr are associated
with the early studies quantifying sexual dysfunc-
tion following spinal cord injury. They found, in
their classic work, that the higher the spinal cord
injury, the more likely the man was to have reflex
but no psychogenic erections, whereas lesions in
the sacral region were likely to be associated with
psychogenic, rather than reflex, erections (Bors
and Comarr, 1960). Ejaculation in men with com-
plete upper motor neuron lesions was rare. Men
with lower motor neuron lesions could rarely have
erections or ejaculate. The less complete the lesion,
the more likely was the man to have reflex and
psychogenic erections and ejaculate. However,
some men suffer pain and spasm with sexual ac-
tivity (Slot et al., 1989).
Erection
Penile erection occurs in 95% of men with com-
plete spinal cord injury and upper motor neuron
lesions, although the erection is often of poor
quality (insufficient rigidity) or poorly sustained.
Why is this so, particularly when the reflex path-
ways appear to be intact? Contact stimulation of-
ten does not lead to erections as good as those
which occurred before the injury. Before exploring
the causes for this dysfunction, the innervation

and neurotransmission key to the erectile response
must be reviewed (see also Burns et al., 2001). The
male sex organs are innervated by three sets of
nerves: the pelvic nerves (nervi erigentes, para-
sympathetic), the hypogastric nerves (sympathetic)
and the pudendal nerves (somatic). The pelvic
nerves contain the axons of sacral parasympathetic
preganglionic neurons situated in the intermedio-
lateral column of sacral (S) spinal cord segments
S2–S4. These axons leave the spinal cord in the
ventral roots S2–S4. Electrical stimulation of these
ventral roots causes erection (Brindley et al.,
1982). The hypogastric nerves contain the axons
of the sympathetic preganglionic neurons situated
in the intermediolateral column of the 11th and
12th thoracic (T) and 1st and 2nd lumbar (L) spi-
nal segments. The pudendal nerves contain the
axons of somatic motoneurons situated in the
ventral horn of sacral cord segments S2–S4. Sym-
pathetic and parasympathetic preganglionic axons
in the hypogastric and pelvic nerve, respectively,
synapse in the pelvic plexus with ganglionic neu-
rons. The postganglionic axons form the cavern-
ous nerves that innervate the smooth muscle of
blood vessels, erectile tissue and nonerectile tissue
(e.g., vas deferens) and glandular tissue. Some of
these axons are cholinergic, some are adrenergic
and the rest are noncholinergic, nonadrenergic.
Somatic motor axons in the pudendal nerves con-
trol the bulbo-cavernous and ischio-cavernous
muscles that are striated muscles situated at the
base of the penis. Sensory axons from the male sex
organs run mainly in the pudendal nerves. Erec-
tion is produced by dilatation of the arterioles that
supply the erectile tissue of the corpora cavernosa.
As the erectile tissue fills with blood, the veins that
drain the erectile tissue are compressed against the
stiff fibrous envelope of the corpora cavernosa and
venous outflow decreases. Erection is likely to in-
volve inhibition of the activity of noradrenergic
axons innervating the corpora cavernosa, since
noradrenaline causes constriction of the arterioles
that supply the erectile tissue by acting on alpha-
adrenergic receptors. Erection is also likely to
involve increased activity of cholinergic axons.
Acetylcholine is thought to act by inhibiting the
release of noradrenaline by noradrenergic axons
and to facilitate the release of vasoactive intestinal
429
polypeptide by the same axons. This peptide is a
powerful smooth muscle relaxant. In addition,
acetylcholine triggers the release of nitric oxide,
another powerful smooth muscle relaxant, by end-
othelial cells and other cells of the erectile tissue.
The nitric oxide binds to smooth muscle receptors
of the corpus cavernosum to increase cyclic gua-
nosine monophosphate (cGMP) levels and thus
relax smooth muscles. Reversal of this process by
phosphodiesterase 5, converting cGMP into GMP,
can be inhibited to promote and prolong erection.
This is the basis of the use of sildenafil and other
phosphodiesterase 5 inhibitors, such as vardenafil
and taladafil, in the treatment of impotence (Derry
et al., 2002).
While acetylcholinesterase positive axons have
been described in human cavernous tissue, and
acetylcholine synthesis and release has been dem-
onstrated in the same tissue (Blanco et al., 1988),
Shirai et al. (1973) could find no evidence of ace-
tylcholinesterase in cavernous tissue from able-
bodied men or from men with a variety of diseases
causing impotence. The mechanism of action of
the parasympathetic nerves in producing erections
must still be considered open to question (Lopez
and Koller, 2000).
The sympathetic nerves of the pelvic plexus can
also cause erection when stimulated (Brindley,
1988). However, it is known that infusion of
alpha2-adrenergic antagonists cause erections,
showing that noradrenalin is, in fact, the main
anti-erectile neurotransmitter (Levin and Wein,
1980). In a study of corpora cavernosa noradren-
aline and acetylcholine content, Melman et al.
(1980) found normal levels of noradrenalin in men
with spinal cord injury. Thus the sympathetic sys-
tem seems to have both erectile and anti-erectile
actions, particularly the latter. How these oppos-
ing actions are regulated and integrated with para-
sympathetic activity is not known.
Intracavernosal injection therapy aims at in-
hibiting sympathetic tone or at relaxing smooth
muscle by a direct action, thereby causing an
erection. Brindley (1983b) showed that the alpha-
adrenergic receptor blocker, phenoxybenzamine,
was effective. The drugs most commonly used to-
day are papaverine, phentolamine and prostaglan-
din E1. Papaverine increases cyclic adenosine
430
monophosphate, promotes calcium efflux from
cells and thereby induces smooth muscle relaxa-
tion. Phentolamine blocks alpha1- and alpha2-ad-
renergic receptors and has a direct relaxant effect
on smooth muscles. Given alone it does not pro-
duce an erection. It is commonly used in conjunc-
tion with papaverine. Prostaglandin E1, the most
widely used drug for intracavernosal injection,
causes vasodilatation, smooth muscle relaxation
and therefore penile erection. Men with spinal
cord injury are much more sensitive to prostaglan-
din E1 than men with other forms of impotence
and may respond to as little as 1–2 mg.
A vacuum pump and ring can be used to cause an
erection suitable for intercourse (Denil et al., 1996).
Penile prostheses are used rarely now that less
destructive treatments for impotence are available.
Objective measurement of sexual function in the
clinical setting can be difficult as it largely relies on
self-reporting. The International Index of Erectile
Function (IIEF) has been developed to measure, in
a standardized way, five areas of sexual function –
erectile function, intercourse satisfaction, orgasm
function, sexual desire and overall satisfaction.
This tool has been used for measuring the efficacy
of phosphodiesterase inhibitors and comparing the
relative benefits of newer agents such as tadalafil
and vardenafil to the more established agent
sildenafil (Del Popolo, 2004).
Ejaculation
Ejaculation is a complex process involving coor-
dination of erection, propulsion of semen and
prevention of retrograde seminal flow. A review by
Thomas (1983) provides a clear, concise account of
these events. Ejaculation may be defined as the
expulsion of semen (sperm plus secretions of pros-
tate, seminal vesicles and other glands) from the
distal urethra. It is produced by rhythmic contrac-
tions of the urethral smooth muscle (innervated by
sympathetic axons) and of two striated muscles
at the base of the penis, the ischio- and bulbo-
cavernosus muscle (innervated by somatic motor
axons). The external bladder sphincter (striated
muscle innervated by somatic motor axons) is
closed. Emission may be defined as the delivery of
semen to the proximal urethra. The delivery is due
to contraction of the smooth muscle of the
epididymis, vas deferens, prostate and seminal
vesicles (innervated by sympathetic axons). Spinal
cord injury causes loss of coordination of these
reflexes and loss of emission and ejaculation.
Psychogenic and genital stimulation usually are
unsuccessful in producing ejaculation in complete
injuries above T10. Injuries below that level may
permit psychogenic ejaculation (Thomas, 1983).
Testicular disorders associated with spinal
cord injury
Many factors could affect testicular function in
men with spinal cord injury. Hypogonadism and
gynecomastia were described to be common in
paraplegic men from the time of the Second World
War. It is possible that the testicular failure was
related to a severe catabolic state following the
spinal cord injury and, with recovery in the sur-
vivors, there was a period of relative estrogen
excess causing the development of gynecomastia
[re-feeding gynecomastia (Baker, 2001)]. This
pattern is not seen today because of better
nutrition and rehabilitation techniques. Other
coincidental injuries could cause gonadotropin
deficiency, direct testicular damage or disruption
of the male genital tract. Drugs used in the
management of spasticity and other complications
of chronic spinal cord injury could also cause
hyperprolactinemia or gonadotropin suppression.
While hormonal abnormalities are still reported in
some men with chronic spinal cord injury (Naderi
and Safarinejad, 2003), most investigators do not
consistently find abnormalities (Baker, 2001).
Autonomic dysfunction has been shown in
animals with experimental spinal cord damage to
affect spermatogenesis (Ohl et al., 2001; Huang
et al., 2004). Specific patterns of disrupted sperm
production have been claimed to occur with spinal
cord lesions at particular levels in men; however,
this has not been confirmed (Bors and Comarr,
1963; Chapelle et al., 1993; Elliott et al., 2000).
Nonspecific defects of spermatogenesis such as
hypospermatogenesis and germ cell arrest are
found in some men with chronic spinal cord

injury but they may be coincidental (Elliott et al.,
2000; Mallidis et al., 2000). Urinary infections
spreading to the vasa deferentia, epididymides and
testes may also temporarily or permanently impair
sperm production or genital tract patency.
Urinary bladder catheterization may cause genital
tract obstruction. Obstruction may result in the
formation of sperm antibodies (Hirsch et al.,
1992). Fever from sepsis or disturbed temperature
regulation could also impair spermatogenesis or
epididymal function (see below). However, these
conditions are not generally present in all men
with spinal cord injury and the most frequent
dysfunction is defective ejaculation that occurs in
about 95% of cord-injured men (Talbot, 1955).
Alterations of the semen associated with spinal
cord injury
Methods of sperm collection
Guttmann (1973) developed assisted ejaculation
with intrathecal neostigmine for humans with spi-
nal cord injury in the 1940s. Brindley (1981b)
pioneered the technique of electroejaculation,
using small electrode-containing probes embedded
in a plastic sheath that fit over the index finger.
After insertion into the rectum the electrodes were
placed in different positions until an emission was
obtained upon stimulation. Probes with electrodes
with larger area have become standard as they
seem more effective and safer than the original
probes (Halstead et al., 1987). The large area elec-
trode equipment has been further improved. The
probe is placed in the rectum with the electrodes
anterior and the electrical stimulation is presumed
to stimulate directly the accessory sex organs and
nerves responsible for emission. By performing
repeated electroejaculation in patients in the acute
phase of spinal cord injury, we found that during
spinal shock semen with few or no sperm was
obtained. Thereafter, good quality semen was
often produced for a few days until about 2 weeks
after the injury when sperm numbers or motility
became low (Mallidis et al., 1994). We believe this is
consistent with a direct effect of electroejaculation
on the autonomic nerve supply of the prostate,
431
seminal vesicles and distal vas deferens accounting
for the azoospermic samples obtained during
spinal shock. There is also a requirement for func-
tional perivasal nerves for stimulation of contrac-
tion of the cauda epididymis to deliver the stored
sperm, which is possible once the spinal shock
abates. Failure to stimulate the pudendal nerves is
probably the reason that emission, rather than
ejaculation, occurs.
For electroejaculation we use a rechargeable,
battery-powered, electrical stimulator, the ‘‘CGS
Electrojector’’ (Ratek Industries P/L 60 Wadhurst
Drive, Boronia, Vic., 3155, Australia, Lim et al.,
1994; Mallidis et al., 1994). It provides a sine wave
current at 20 Hz of progressively increasing
amplitude to a maximum of 500 mA. The stimula-
tor is fitted with a blunt-end Delrin probe which
houses three anteriorly placed, stainless steel,
longitudinal, bar electrodes. A thermocouple in
the middle electrode monitors temperature at the
electrode-mucosal interface. The temperature is
read on a screen built into the stimulator casing.
In 9–41% men with complete spinal cord injury
undergoing electroejaculation, there is partial or
complete retrograde ejaculation, that is, retrograde
flow of semen into the bladder (Brindley, 1984;
Siosteen et al., 1990; Lim et al., 1994). This is rec-
ognized by failure of antegrade flow of semen and
the presence of sperm in the urine collected from
the bladder after the procedure. To prevent retro-
grade ejaculation, we use a nontoxic, all-silicone
Foley catheter inserted into the bladder. Lubricat-
ing gels are avoided, as they are known to be toxic
to sperm (Lim et al., 1994). Once the catheter is in
situ, the balloon is inflated with 10 ml of saline.
Urine is drained into a collection bag. An assistant
applies constant, gentle traction to the catheter to
tamponade the bladder neck (Lim et al., 1994).
After proctoscopy to detect any pre-existing
lesions, the rectal probe is inserted and the
stimulator output progressively increased toward
the maximum output or until emission occurs. The
rectal temperature is kept below 401C. Antegrade
emission occurs through the urethra around the
catheter. The semen is collected in a warm, sterile,
plastic jar. The proctoscope is reintroduced to
check the condition of the rectal mucosa and the
catheter removed.
432
The technique of vibration ejaculation (vibro-
ejaculation) has also improved recently (Brindley,
1981a; Brackett, 1999). The specifications of the
machinery vary. The key is adequate vibration fre-
quency (80 Hz) and amplitude (2.5 mm; Brindley,
1981a; Brackett, 1999). The technique relies on an
intact spinal reflex arc, including sacral segments
and thoracolumbar segments up to T10, and
works by an augmentation of the normal ejacula-
tory reflex (Wieder et al., 2000). The vibrator is
applied to the ventral surface of the glans penis to
initiate the ejaculatory reflex. Brackett (1999)
reported a 50% success rate of vibroejaculation
in 211 cord-injured men using a 2.5 mm amplitude
vibrator. The success rate was higher than when a
standard 1.6 mm vibrator was used. The success
rate with the 2.5 mm vibrator was related to level
of spinal cord injury. Lesions at C3–7, T1–5,
T6–10, and T11–L3 were associated with success
rates of 66%, 54%, 41% and 36%, respectively. It
has been noted that, when successful, vibroejacu-
lation produces sperm of better quality than does
electroejaculation (Brackett et al., 1997, 1998).
Assisted ejaculation may induce autonomic
dysreflexia in dysreflexia-prone men, i.e., those
with spinal cord injury above T6 (Guttmann, 1973;
Brackett, 1999; Scheutzow and Bockenek, 2000).
Careful monitoring during trials of vibroejaculation
or electroejaculation and the prophylactic use
of the calcium channel blocker nifedipine are
essential (Brackett, 1999). Electroejaculation
requires general anesthesia in men with incom-
plete lesions as pain is induced in those with
sensory awareness in their sacral dermatomes.
Rectal trauma may be induced by the procedure
for electroejaculation. Other methods of stimulat-
ing emission of semen with oral or intrathecal
drugs (e.g., the catecholamine uptake blocker,
imipramine, or the cholinesterase inhibitor,
neostigmine) are rarely effective and are potential-
ly dangerous (Guttmann, 1973).
Sperm may also be obtained from the genital
tract by prostatic massage or surgically by hemi-
section of the vas deferens, epididymal aspiration
or needle aspiration or open testicular biopsy
(Guttmann, 1973; Hovatta and von Smitten, 1993;
Buch, 1994; Watkins et al., 1996; Marina et al.,
1999; Brackett et al., 2000). While these approaches
may be used if assisted ejaculation is not available
or fails, intracytoplasmic sperm injection is
required as usually insufficient mature sperm are
obtained for artificial insemination.
Semen abnormalities
y it seems that non-drainage is one factor in the
poor quality of paraplegic semen. Almost certainly
chronic infection in the genital tract is another, yA
third potentially remediable factor is raised scrotal
temperaturey (Brindley, 1983a)
Analysis of semen collected by assisted ejacula-
tion gives variable results from person to person
after cord injury. In some subjects there are no
sperm; occasionally the semen is normal; usually
there is a normal to high sperm concentration with
extremely low sperm motility and viability. The
latter semen pattern is called necrospermia and it
also occurs, rarely, in able-bodied infertile men
who can ejaculate (see below; Wilton et al., 1988;
Mallidis et al., 2000). A number of abnormalities
of the semen collected from men with chronic
spinal cord injury have been described, including:
increased leukocyte numbers (from very low num-
bers, o1 million/ml), increased levels of inflam-
matory cytokines (normally found in semen),
increased presence of reactive oxygen species (from
very low values normally) and changes in prostate-
specific antigen (De Lamirande et al., 1995; Aird
et al., 1999; Lynne et al., 1999; Trabulsi et al.,
2002; Basu et al., 2004). The increased numbers of
leukocyte and levels of cytokines may occur
because of inflammation and they further impair
sperm quality but probably are not the only, or
main, cause of semen abnormality. Some have
reported that the seminal plasma will impair the
motility of normal test sperm (Brackett et al.,
1996; Monga et al., 2001). While some investiga-
tors believe abnormalities of the secretions of the
accessory sex organs are responsible for the defec-
tive sperm motility, at this stage it remains unclear
whether these changes cause the sperm defect or
are merely associations.
It is important to consider the possibility that
the assisted ejaculation technique itself may
contribute to the poor semen analysis result
through, for example, exposure of the sperm to

toxic material in rubber or plastic, urine contam-
ination or failure to stimulate emission from the
epididymal sperm store (Lim et al., 1994). The
person’s bladder management also relates to the
results of assisted ejaculation (Ohl et al., 1992;
Rutkowski et al., 1995; Brackett et al., 1998).
There are several reports that sperm obtained
from the vas deferens of men with chronic spinal
cord injury have better motility than sperm in the
semen collected by assisted ejaculation (Hovatta
and von Smitten, 1993; Buch, 1994; Brackett et al.,
2000). Brackett et al. (2000) examined sperm from
the vas deferens aspirates and from semen of men
with chronic spinal cord injury and from controls
undergoing vasectomy. The mean sperm motility
and viability (expressed as a percentage of the
sperm) in the cord-injured men’s semen were 14%
and 26% and in vas deferens aspirates were 54%
and 74%, respectively. In the control men, semen
values had 74% motility and 85% viability and vas
deferens aspirates had 78% motility and 89% vi-
ability. While the sperm motility is better in the vas
deferens of the cord-injured men than in their semen,
it is still lower than in the controls. Interestingly they
also found, in two cord-injured men, abnormalities
of the vas deferens described as edema or fibrosis.
It is known that the cauda epididymis is the
storage organ of sperm and that the seminal
vesicles do not normally contain sperm. Jarow
(1996) performed bilateral needle aspirations of
the seminal vesicles under transrectal ultrasound
guidance and confirmed that the seminal vesicles
of fertile men ejaculating regularly do not contain
significant numbers of sperm. However, with 5
days of abstinence from ejaculation the sperm
concentration in seminal vesicle aspirates was
found to range up to 10  106/ml in one-third of
the subjects. No sperm aspirated from the seminal
vesicles were actively motile. The suggestion was
that the number of sperm present might be directly
related to the duration of the abstinence. If
neuropathic anejaculation behaved as a ‘‘function-
al obstruction’’ causing a sludging of sperm in the
reproductive tract there may be reflux of sperm into
the seminal vesicles. This has been confirmed by
Ohl et al. (1999). By performing electroejaculation
after seminal vesicle aspiration, these workers
estimated that up to half the sperm collected by
433
random assisted ejaculation in men with chronic
spinal cord injury may come from the seminal
vesicles (Ohl et al., 1999).
Mechanisms of semen abnormalities
Testicular temperature
Brindley (1982, 1983a) demonstrated that intra-
scrotal temperatures were elevated in 29 men with
chronic spinal cord injury sitting in wheel chairs
with temperatures about 11C higher (36.21C) than
in clothed able-bodied controls sitting in wheel
chairs for 20 min (35.31C). He found no difference
in scrotal temperatures of cord-injured men and
controls lying in bed. In a smaller data set he
found suggestive evidence for poorer results of
electroejaculation in that only two of nine with
scrotal temperatures above 36.41C had motile
sperm compared with eight of nine with lower
scrotal temperatures. Old data also suggested
a relationship between impaired spermatogenesis,
level of lesion and impairment of sweating
(Guttmann, 1973). However, others have not been
able to confirm a clear relationship between
testicular temperature and results of assisted ejac-
ulation (Brackett et al., 1994). Bedford (1991,
1994) has reviewed the effects of elevated temper-
ature on cauda epididymal function in rats, rabbits
and hamsters and suggests there may be a parallel
in humans. In animals, the size of the tail of the
epididymis was reduced and the rate of transit of
sperm through the epididymis was increased.
Sperm in the tail of the epididymis were reduced
in number, motility and viability in the rat. Sperm
quality improved with return of normal temperature
control or with frequent ejaculation (Bedford, 1994).
We suspect genital tract stasis and disturbed
thermoregulation impairing epididymal sperm
storage are probably the main causes of the com-
mon pattern of low sperm motility and viability in
the semen of men with chronic spinal cord injury.
Stasis
We have shown that the poor quality of semen
commonly obtained by assisted ejaculation from
men with chronic spinal cord injury is consistent
434
with a rare cause of infertility in able-bodied men
known as necrospermia (Mallidis et al., 2000). The
reproductive tract stasis in spinal cord injury
caused by the neuropathic anejaculation is similar
to a functional obstruction. Recently a similar sit-
uation has been found in men with adult polycystic
kidney disease and necrospermia (Fang and Baker,
2003). These men appear to have stasis of genital
tract contents in cysts of the ejaculatory ducts,
prostate or seminal vesicles. This remarkable
association suggests that stasis from any cause:
neuropathic, partial distal genital tract obstruction
or cystic dilatation will cause necrospermia
(Mallidis et al., 2000; Fang and Baker, 2003).
Possible evolution of the sperm defects with spinal
cord injury
While multiple factors could affect sperm produc-
tion in chronic spinal cord injury and certain cord-
injured men may have spermatogenic failure,
accessory sex organ inflammation, genital tract
obstruction or sperm autoimmunity as the pre-
dominant cause, the majority of individuals do not
have such problems. Our studies suggest the fol-
lowing possible sequence in the majority of men
with spinal cord injury (Lim et al., 1994; Mallidis
et al., 1994, 2000). In the acute phase, spermato-
genesis is likely to be suppressed because of an
acute shut down of gonadotropin production as
response to critical illness including the stresses of
anesthesia, surgery, nutritional deprivation and
drug administration. Once spinal shock abates
sperm already in the epididymis can be recovered
by electroejaculation. Subsequently sperm collec-
tion depends on the resumption of spermatogen-
esis and this may take several months in some
men. Those men with spinal cord lesions that al-
low coordinated autonomic function may have
normal genital tract motility and thermoregulation
and good quality antegrade semen may be collect-
ed by either vibroejaculation or electroejaculation.
Rarely some men in this group may be able to
ejaculate without assistance. Many cord-injured
men however have abnormal genital tract motility
so they do not respond well to vibroejaculation.
There is also defective scrotal thermoregulation
with impaired cauda epididymal sperm storage
accounting for the low sperm motility in the vasa
and the rapid drop off in sperm count with re-
peated ejaculation. The disturbed genital tract
motility results in accumulation and sludging of
sperm in the ampullae of the vasa and seminal
vesicles. This produces the characteristic result
with the first electroejaculation: often discolored
semen with high sperm concentration but very low
motility. Once this material is cleared, sperm with
better motility can be obtained over the next 1 to 2
days by successive daily electroejaculations. The
fact that sperm cannot continue to be collected in
the majority of cord-injured men beyond 3–4 days
suggests that reduced sperm production, not ap-
parent on routine testicular histology, may also be
a common problem.
Management of male infertility with spinal
cord injury
The team approach
While failure of ejaculation is the major contrib-
utor to the infertility of men with spinal cord in-
jury, sperm quality is often impaired and
intracytoplasmic sperm injection may be required.
However, as some cord-injured men have normal
semen, a thorough evaluation is required to pro-
vide cost-effective management. This has been
achieved by close co-operation between the Victo-
rian Spinal Cord Service at Austin Health and
Melbourne IVF Reproductive Services at the Roy-
al Women’s Hospital. This partnership has also
facilitated clinical research. Similar team ap-
proaches have been reported from around the
world (Linsenmeyer, 2000; Biering-Sorensen and
Sonksen, 2001; Heruti et al., 2001; Shieh et al.,
2003).
Clinical evaluation
The cord-injured men undergo andrologic assess-
ment, including particularly a history of factors
that might affect testicular function, physical
examination of secondary sex characteristics and
thorough scrotal examination, serum levels of
 435

follicle-stimulating hormone and antisperm anti-
bodies and other hormonal assays and testicular
biopsy as indicated by signs of testicular failure.
The female partner is evaluated by a infertility
specialist gynecologist, has pre-pregnancy screens
for general health, rubella immunity, infections,
genetic conditions as indicated and is advised
about preventative strategies such as taking folate
and avoiding smoking and excessive alcohol
consumption. A plan of management is devised
depending on the success of assisted ejaculation
trials and on the sperm quality (Table 1).
Assisted ejaculation
The management approach is from the least inva-
sive and costly intervention that may be successful
toward the more technologically complex and
costly (Table 1). In determining the method of
assisted ejaculation, vibration stimulus is prefer-
entially chosen for men who have an intact reflex
arc for ejaculation. Vibration, if successful, may
allow home insemination, particularly when auto-
nomic dysreflexia is not a risk.
Electroejaculation is tried when vibration fails
and in men without an intact reflex ejaculation arc.
Cord-injured men with sacral spinal segment
preservation will be offered the procedure under
general anesthetic.
Both vibration ejaculation and electroejacula-
tion can be performed with Foley catheter bladder
neck tamponade if retrograde ejaculation is
expected. Both procedures are likely to precipitate
autonomic dysreflexia in those who suffer lesions
above T6. Close monitoring of the blood pressure
during the procedure is mandatory. Prophylactic
anti-hypertensive medication can be given. If
autonomic dysreflexia does develop cessation of
the procedure is usually adequate management.
With persistent blood pressure elevation, the rapid
administration of hypotensive agents sublingually
or, in rare instances, intravenously is effective.
Testicular aspiration is utilized when vibration and
electroejaculation procedures have failed.
The semen collected is analyzed by standard
methods. Sperm may be cryopreserved if not im-
mediately required for artificial insemination or
intracytoplasmic sperm injection.
Austin Health/Royal Women’s Hospital fertility
program results
Acute collection of semen is not offered routinely.
It is offered to cord-injured men who were actively
attempting to father children at the time of their
injury and to men for whom this was a major
immediate issue psychosocially. Near normal
semen was collected in 50% of cases in which acute
acquisition was attempted (Mallidis et al., 1994).

Table 1. Clinical features of management in the chronic spinal cord injury patient

Clinical evaluation including

General condition stable, no recent febrile illness or urinary infection
Neurologic assessment level (upper or lower motor neuron), completeness
Andrologic assessment clinical examination, follicle stimulating hormone and sperm antibody assay, possibly testicular biopsy
Trial of assisted ejaculation vibroejaculation or electroejaculation, need for general anesthesia, autonomic dysreflexia risk
Female fertility evaluation prepregnancy checkup
Management plan artifical insemination or intracytoplasmic sperm injection
Infertility counselor information and psychological support
Management options
Home insemination if vibration ejaculation successful and safe
Repeated ejaculation for necrospermia (daily for 2–3 days) timed with ovulation for fresh intrauterine insemination
Semen cryopreservation for artificial insemination (concentration 420  106/ml, post thaw motility 420%) or intracytoplasmic sperm
injection
Fine-needle tissue aspiration testicular biopsy to obtain sperm for intracytoplasmic sperm injection if assisted ejaculation fails or
contraindicated.
Donor insemination if no sperm obtainable
436
In the chronic spinal cord injury group fifty-
eight men have entered the program over a 12-year
period. Thirty-eight men were or are engaged in
active treatment. Sixteen men have fathered twen-
ty live births. Three live births were achieved from
sperm acquired by testicular aspiration, twelve
from sperm acquired by assisted ejaculation and
six from home-based sperm acquisition and arti-
ficial insemination. No live sperm was retrievable
in six men. The drop out rate was approximately
50% after one or two treatments in couples en-
gaged in hospital-based assisted reproductive tech-
nology (ART) treatments. This is a similar figure
to the able-bodied engaged in assisted reproduc-
tive technology. Ten cord-injured men are pres-
ently engaged in active treatment with the partners
of two of them having pregnancies of 3 months
duration.
Although some groups have reported that re-
peated ejaculation does not improve semen quality
with chronic spinal cord injury, this is not the
general experience (Brindley, 1983a; Brackett,
1999; Sonksen et al., 1999; Mallidis et al., 2000;
Heruti et al., 2001). We find sperm motility can be
improved with frequent ejaculation (daily for up
to 2–3 days) in most cord-injured men with
necrospermia (Mallidis et al., 2000). This has
clinical significance for vibroejaculation in the
home setting. The men are advised to have fre-
quent ejaculation (two or more times a week) and
intravaginal artificial insemination of the semen
daily for 1–3 days around the time of ovulation
determined by symptoms or detection of the lute-
inizing hormone surge in urine. Also with electro-
ejaculation, if semen is not immediately suitable
for artificial insemination, then repeating the pro-
cedure and collection of semen on the second or
third day may allow artificial insemination to be
timed with ovulation or cryopreservation of semen
adequate for subsequent artificial insemination.
Generally the cryopreserved semen is used to pre-
pare a motile sperm suspension by density gradient
centrifugation for intrauterine insemination at the
time of ovulation (Bourne et al., 2004). Although
there are many reports of low pregnancy rates with
artificial insemination of sperm obtained by
assisted ejaculation, the results of adequately
timed inseminations can be good (Brindley, 1984;
Halstead et al., 1987; Dahlberg et al., 1995; Brack-
ett, 1999; Pryor et al., 2001).
ART including in vitro fertilization or intracy-
toplasmic sperm injection can be used if there is
coexisting female infertility, artificial insemination
fails after a reasonable number of attempts, if the
semen quality is inadequate for artificial insemi-
nation or if the sperm are obtained from the gen-
ital tract. Standard in vitro fertilization may be
possible if the semen is adequate. The results of
assisted reproductive technology seem no different
from those of other patients although a lower
pregnancy rate has been reported (Schatte et al.,
2000).
Summary
Sexual function
Treatment of sexual dysfunction in males with
spinal cord injury is often focused on the erection
disorder. Modern approaches emphasize the need
for sexual health counseling where broader issues
of intimacy, relationship and communication are
addressed. Without attention to these issues the
success of erection-promoting agents may be com-
promised. At this time all have a place clinically
but further comparative research may elucidate
which groups of cord-injured men may be more
suited to some agents than others. While the trend
is toward the evolution of further oral agents, es-
tablished options such as intracavernosal therapy,
vacuum tumescence pump, sacral anterior root
electrical stimulation and even penile prostheses
continue to have a place in clinical care.
Male infertility
An inability to ejaculate in the first few days after
acute spinal cord injury, before return of visceral
reflexes, is probably due to neural refractoriness
associated with spinal shock. Except in those with
systemic illness or multi-trauma, electroejaculation
performed 6–12 days after injury will enable col-
lection of sperm of normal or near normal quality.
Thereafter sperm quality deteriorates. The main
reason for this is not clear and may be related to

elevated scrotal temperatures, infrequent ejacula-
tion (functional obstruction) or other factors. The
fact that it occurs at approximately 2 weeks after
injury suggests that there is a neurogenic factor
and this would therefore appear to be related to loss
of input from the brain stem. However the relevant
neurogenic factor is not understood at this stage.
We have shown that most men with chronic
spinal cord injury have adequate spermatogenesis
but many have a defect of sperm motility that can
be improved by frequent ejaculation. This is
similar to the epididymal necrospermia seen in
able-bodied men. We postulate that neuropathic
anejaculation behaves as a functional ejaculatory
duct obstruction and that the distal genital tract
fills with degenerating sperm. Repeated ejaculation
clears these sperm allowing collection of sperm
with better motility. Although spermatogenesis is
not severely impaired on conventional histology, it
is possible that sperm production is abnormal as
suggested by animal studies, because we have gen-
erally been unable to continue to collect semen
daily beyond 2–4 days. Although not all studies
show significant relationships between scrotal tem-
perature and semen quality, we suspect elevated
temperature does affect spermatogenesis and par-
ticularly epididymal sperm storage. Further re-
search to examine in detail the testes and genital
tract of men with chronic spinal cord injury is
needed to confirm our hypotheses. However, the
findings that semen quality improves with repeated
ejaculation has obvious therapeutic implications as
described above. We anticipate further clinical
research will enhance our approach to male infer-
tility associated with spinal cord injury and also
benefit the able-bodied male population with
infertility through better understanding of these
interesting acquired sperm defects.
References
Aird, I.A., Vince, G.S., Bates, M.D., Hohnson, P.M. and
Lewis-Jones, I.D. (1999) Leukocytes in semen from men with
spinal cord injuries. Fertil. Steril., 72: 97–103.
Baker, H.W.G. (2001) Testicular function in systemic disease.
In: Becker K.L. (Ed.), Principles and Practice of Endocrino-
logy and Metabolism (3rd ed.), Chapter 116. Lippincott
Williams and Wilkins, Philadelphia, PA, pp. 1150–1158.
437
Basu, S., Aballa, T.C., Ferrell, S.M., Lynne, C.M. and Brack-
ett, N.L. (2004) Inflammatory cytokine concentrations are
elevated in seminal plasma of men with spinal cord injuries.
J. Androl., 25: 250–254.
Bedford, M.J. (1991) Effects of elevated temperature on the
epididymis and testis: experimental studies. In: Zorgniotti
A.W. (Ed.), Temperature and Environmental Effects on the
Testis. Plenum Press, New York, pp. 19–32.
Bedford, M.J. (1994) The status and the state of the human
epididymis. Hum. Reprod., 11: 2187–2199.
Biering-Sorensen, F. and Sonksen, J. (2001) Sexual function in
spinal cord lesioned men. Spinal Cord, 39: 455–470.
Blanco, R., Saenz De Tejada, I., Goldstein, I., Krane, R.J.,
Wotiz, H.H. and Cohen, R.A. (1988) Cholinergic neuro-
transmission in human corpus cavernosum II Acetylcholine
synthesis. Am. J. Physiol., 254: H468–H472.
Bors, E. and Comarr, A.E. (1960) Neurological disturbances of
sexual function with special reference to 529 patients with
spinal cord injury. Urol. Surv., 10: 191–222.
Bourne, H., Edgar, D.H. and Baker, H.W.G. (2004) Sperm
preparation techniques. In: Gardner D.K., Weissman A.,
Howles C.M. and Shoham Z. (Eds.), Textbook of
Assisted Reproductive Techniques Laboratory and Clinical
Perspectives (2nd ed.). Taylor and Francis, London,
pp. 79–92.
Brackett, N.L. (1999) Semen retrieval by penile vibratory stim-
ulation in men with spinal cord injury. Hum. Reprod.,
Update, 5: 216–222.
Brackett, N.L., Bloch, W.E. and Lynne, C.M. (1998) Predictors
of necrospermia in men with spinal cord injury. J. Urol., 159:
844–847.
Brackett, N.L., Davi, R.C., Padron, O.F. and Lynne, C.M.
(1996) Seminal plasma of spinal cord injured men inhibits
sperm motility of normal men. J. Urol., 155: 1632–1635.
Brackett, N.L., Lynne, C.M., Aballa, T.C. and Ferrell, S.M.
(2000) Sperm motility from the vas deferens of spinal cord
injured men is higher than from the ejaculate. J. Urol., 164:
712–715.
Brackett, N.L., Lynne, C.M., Weizman, M.S., Bloch, W.E. and
Padron, O.F. (1994) Scrotal and oral temperatures are not
related to semen quality of serum gonadotropin levels in spi-
nal cord-injured men. J Androl., 15: 614–619.
Brackett, N.L., Padron, O.F. and Lynne, C.M. (1997) Semen
quality of spinal cord injured men is better when obtained by
vibratory stimulation versus electroejaculation. J. Urol., 157:
151–157.
Brindley, G.S. (1981a) Reflex ejaculation under vibratory stim-
ulation in paraplegic men. Paraplegia, 19: 299–302.
Brindley, G.S. (1981b) Electroejaculation: its technique, neu-
rological implication and uses. J. Neurol. Neurosurg. Psych.,
44: 9–18.
Brindley, G.S. (1982) Deep scrotal temperature and the effect
on it of clothing, air temperature, activity, posture and par-
aplegia. Br. J. Urol., 54: 49–55.
Brindley, G.S. (1983a) Physiology of erection and management
of paraplegic infertility. In: Hargreave T.B. (Ed.), Male In-
fertility. Springer, Berlin, pp. 261–279.
438
Brindley, G.S. (1983b) Cavernosal alpha-blockade: a new tech-
nique for investigation and treating erectile impotence. Br. J.
Psych., 143: 332–337.
Brindley, G.S. (1984) The fertility of men with spinal cord in-
juries. Paraplegia, 22: 337–348.
Brindley, G.S. (1988) The actions of parasympathetic and sym-
pathetic nerves in human micturition, erection and seminal
emission, and their restoration in parasympathetic patients
by implanted electrical stimulation. Ferrier Lecture 1986.
Proc. R. Soc. Lond. [Biol.], 235: 111–120.
Brindley, G.S., Polkey, C.E. and Rushdon, D.N. (1982) Sacral
anterior root stimulators for bladder control in paraplegia.
J. Paraplegia, 20: 365–381.
Buch, J.P. (1994) Greatly improved sperm motility from vas
deferens sperm retrieval: a case for accessory gland related
subfertility in spinal cord injured men. Case report. Paraple-
gia, 32: 501–504.
Burns, A.S., Rivas, D.A. and Ditunno, J.F. (2001) The man-
agement of the neurogenic bladder and sexual dysfunction
after spinal cord injury. Spine, 26(Suppl. 24): 129–136.
Chapelle, P.A., Roby-Brami, A., Jondet, M., Piechaud, T. and
Bussel, B. (1993) Trophic effects on testes in paraplegics.
Paraplegia, 31: 576–583.
Dahlberg, A., Ruutu, M. and Hovatta, O. (1995) Pregnancy
results from a vibrator application, electroejaculation, and a
vas aspiration programme in spinal-cord injured men. Hum.
Reprod., 10: 2305–2307.
De Lamirande, E., Hassouna, M., Leduc, B.E., Gagnmon, C.
and Iwasaki, A. (1995) Increased Reactive Oxygen Species
Formation in Semen of Patients with Spinal Cord Injury.
Fertil. Steril., 53: 637–642.
Del Popolo, G., Li Marzi, V., Mondaini, N. and Lombardi, G.
(2004) Time/duration effectiveness of sildenafil versus ta-
dalafil in the treatment of erectile dysfunction in male spinal
cord-injured patients. Spinal Cord, 42: 643–648.
Denil, J., Ohl, D.A. and Smythe, C. (1996) Vacuum erection
device in spinal cord injured men: patient and partner sat-
isfaction. Arch. Phys. Med. Rehabil., 77: 750–753.
Derry, F., Hultling, C., Seftel, A.D. and Sipski, M.L. (2002)
Efficacy and safety of sildenafil citrate (Viagra) in men with
erectile dysfunction and spinal cord injury: a review. Urol-
ogy, 60: 49–57.
Elliott, S.P., Orejuela, F., Hirsch, I.H., Lipshultz, L.I., Lamb,
D.J. and Kim, E.D. (2000) Testis biopsy findings in the spinal
cord injured patient. J. Urol., 163: 792–795.
Fang, S. and Baker, H.W.G. (2003) Male infertility and adult
polycystic kidney disease are associated with necrospermia.
Fertil. Steril., 79: 643–644.
Guttmann, L. (1973) Spinal Cord Injuries Comprehensive
Management and Research. Blackwell, Oxford, pp. 446–477.
Halstead, L.S., VerVoort, S.M. and Seager, S.W. (1987) Rectal
probe electrostimulation in the treatment of anejaculatory
spinal cord injured men. Paraplegia, 25: 120–129.
Heruti, R.J., Katz, H., Menashe, Y., Weissenberg, R., Raviv,
G., Madjar, I. and Ohry, A. (2001) Treatment of male in-
fertility due to spinal cord injury using rectal probe electro-
ejaculation: the Israeli experience. Spinal Cord, 39: 168–175.
Hirsch, I.H., Sedor, J., Callahan, H.J. and Staas, W.E. (1992)
Antisperm antibodies in seminal plasma of spinal cord-in-
jured men. Urol., 39: 243–247.
Hovatta, O. and von Smitten, K. (1993) Sperm aspiration from
vas deferens and in-vitro fertilization in cases of non-treat-
able anejaculation. Hum. Reprod., 8: 1691–1698.
Huang, H.F., Wang, S., Molina, C.A. and Ottenweller, J.E.
(2004) Preservation of spermatogenesis in spinal cord injured
rats with exogenous testosterone. Relationship with serum
testosterone levels and cellular localization of camp respon-
sive element modulator. J. Androl., 25: 95–103.
Jarow, J.P. (1996) Seminal vesicle aspiration of fertile men.
J. Urol., 156: 1005–1007.
Levin, R.M. and Wein, A.J. (1980) Adrenergic alpha-receptors
outnumber beta-receptors in human penile corpus caverno-
sum. Invest. Urol., 18: 225–226.
Lim, T.C., Mallidis, C., Hill, S.T., Skinner, D.J., Carter, P.D.,
Brown, D.J. and Baker, H.W.G. (1994) A simple technique
to prevent retrograde ejaculation during assisted ejaculation.
Paraplegia, 32: 142–149.
Linsenmeyer, T.A. (2000) Sexual function and infertility fol-
lowing spinal cord injury. Phys. Med. Rehabil. Clin. N. Am.,
11: 141–156.
Lopez, G.J. and Kollar, W.C. (2000) Sexual function, dysfunc-
tion, orientation and the autonomic nervous system. In:
Appenzeller O. (Ed.), Handbook of Clinical Neurology, Vol.
75 (31). The Autonomic Nervous System, Part ii Dysfunc-
tions, Chapter 3, Elsevier, Amsterdam, pp. 85–103.
Lynne, C.M., Aballa, T.C., Wang, T.J., Rittenhouse, H.G.,
Ferrell, S.M. and Brackett, N.L. (1999) Serum and semen
prostate specific antigen concentrations are different in
young spinal cord injured men compared to normal con-
trols. J. Urol., 162: 89–91.
Mallidis, C., Lim, T.C., Hill, S.T., Skinner, D.J., Brown, D.J.,
Johnston, W.I.H. and Baker, H.W.G. (1994) Collection of
semen from men in acute phase of spinal cord injury. Lancet,
343: 1072–1073.
Mallidis, C., Lim, T.C., Hill, S.T., Skinner, D.J., Brown, D.J.,
Johnston, W.I.H. and Baker, H.W.G. (2000) Necrospermia
and chronic spinal cord injury. Fertil. Steril., 74: 221–227.
Marina, S., Marina, F., Alcolea, R., Nadal, J., Pons, M.C.,
Grossman, M., Esposito, R. and Vidal, J. (1999) Triplet
pregnancy achieved through intracytoplasmic sperm injection
with spermatozoa obtained by prostatic massage of a para-
plegic patient. Hum. Reprod., 14: 1546–1548.
Melman, A., Henry, D.P., Felten, D.L. and O’Connor, B.L.
(1980) Alteration of the penile corpora in patients with erec-
tile impotence. Invest. Urol., 17: 474–477.
Monga, M., Dunn, K. and Rajasekaran, M. (2001) Character-
ization of ultrastructural and metabolic abnormalities in se-
men from men with spinal cord injury. J. Spinal Cord Med.,
24: 41–46.
Naderi, A.R. and Safarinejad, M.R. (2003) Endocrine profiles
and semen quality in spinal cord injured men. Clin. End-
ocrinol. (Oxf.),, 58: 177–184.
Ohl, D.A., Denil, J., Fitzgerald-Shelton, K., McCabe, M.,
McGuire, E.J., Menge, A.C. and Randolph, J.F. (1992)

Fertility of spinal cord injured males: Effect of genitourinary
infection and bladder management on results of electroejac-
ulation. J. Am. Paraplegia Soc., 15: 5359.
Ohl, D.A., Menge, A.C. and Jarow, J.P. (1999) Seminal vesicle
aspiration in spinal cord injured men: Insight into poor
sperm quality. J. Urol., 162: 2048–2051.
Ohl, D.A., Sonksen, J., Wedemeyer, G., Zaborniak, M.C.,
Dam, T.N., Menge, A.C., Putzi, M.J. and Papadopoulos,
S.M. (2001) Canine model of infertility after spinal cord in-
jury: Time course of acute changes in semen quality and
spermatogenesis. J. Urol., 166: 1181–1184.
Pryor, J.L., Kuneck, P.H., Blatz, S.M., Thorp, C., Cornwell,
C.E. and Carrell, D.T. (2001) Delayed timing of intrauterine
insemination results in a significantly improved pregnancy
rate in female partners of quadriplegic men. Fertil. Steril., 76:
1130–1135.
Rutkowski, S.B., Middleton, J.W., Truman, G., Hagen, D.L.
and Ryan, J.P. (1995) The influence of bladder management
on fertility in spinal cord injured males. Paraplegia, 33:
236–263.
Schatte, E.C., Orejuela, F.J., Lipshultz, L.I., Kim, E.D. and
Lamb, D.J. (2000) Treatment of infertility due to anejacu-
lation in the male with electroejaculation and intracytoplas-
mic sperm injection. J. Urol., 163: 1717–1720.
Scheutzow, M.H. and Bockenek, W.L. (2000) An unusual
complication during electroejaculation in an individual with
tetraplegia. J. Spinal Cord Med., 23: 28–30.
Shieh, J.Y., Chen, S.U., Wang, Y.H., Chang, H.C., Ho, H.N.
and Yang, Y.S. (2003) A protocol of electroejaculation and
systematic assisted reproductive technology achieved high
efficiency and efficacy for pregnancy for anejaculatory men
with spinal cord injury. Arch. Phys. Med. Rehabil., 84:
535–540.
439
Shirai, M., Sasaki, K. and Rikimaru, A. (1973) A histochemical
investigation of the distribution of adrenergic and cholinergic
nerves in the human male genital organs. Tohoku J. Exp.
Med., 111: 281–291.
Siosteen, A., Forssman, L., Steen, Y., Sullivan, L. and Wick-
strom, I. (1990) Quality of semen after repeated ejaculation
treatment in spinal cord injury men. Paraplegia, 28: 96–104.
Slot, O., Drewes, A., Andreasen, A. and Olsson, A. (1989)
Erectile and ejaculatory function of males with spinal cord
injury. Int. Disabil. Stud., 11: 75–77.
Sonksen, J., Ohl, D.A., Giwercman, A., Biering-Sorensen, F.,
Skakkebaek, N.E. and Kristensen, J.K. (1999) Effect of re-
peated ejaculation on semen quality in spinal cord injured
men. J. Urol., 161: 1163–1165.
Talbot, H.S. (1955) The sexual function in paraplegia. J. Urol.,
77: 157–168.
Thomas, A.J. (1983) Ejaculatory dysfunction. Fertil. Steril., 39:
445–454.
Trabulsi, E.J., Shupp-Byrne, D., Sedor, J. and Hirsch, I.H.
(2002) Leukocyte subtypes in electroejaculates of spinal cord
injured men. Arch. Phys. Med. Rehabil., 83: 31–34.
Watkins, W., Lim, T., Bourne, H., Baker, H.W.G. and
Wutthiphan, B. (1996) Testicular aspiration of sperm for in-
tracytoplasmic sperm injection: an alternative treatment to
electro-emission: case report. Spinal Cord, 34: 696–698.
Wieder, J.A., Brackett, N.L., Lynne, C.M., Green, J.T. and
Aballa, T.C. (2000) Anesthetic block of the dorsal penile
nerve inhibits vibratory-induced ejaculation in men with spi-
nal cord injuries. Urol., 55: 915–917.
Wilton, L.J., Temple-Smith, P.D., Baker, H.W.G. and de
Kretser, D.M. (1988) Human male infertility caused by de-
generation and death of sperm in the epididymis. Fertil.
Steril., 49: 1052–1058.
L.C. Weaver and C. Polosa (Eds.)
Progress in Brain Research, Vol. 152
ISSN 0079-6123
Copyright r 2006 Elsevier B.V. All rights reserved

CHAPTER 30

Female sexual function after spinal cord injury

Marca L. Sipski1,2, and Adriana Arenas2

1
Veterans Administration Rehabilitation Research and Development, Center of Excellence in Functional Recovery and
Spinal Cord Injury, Miami, FL 33101, USA
2
Department of Rehabilitation Medicine, University of Miami School of Medicine, P.O. Box 016960 (D-461), Miami,
FL 33101, USA

Abstract: Over the past 10 years, studies of the impact of spinal cord injuries on female sexuality have
expanded from questionnaire studies in small populations with unknown levels and degrees of injury to
laboratory-based analyses of women with known injury patterns. These studies have provided detailed
information on how specific injury patterns affect specific aspects of the female sexual response. Research
findings have supported the hypothesis that the sympathetic nervous system is regulatory for psychogenic
genital vasocongestion and that orgasm is a reflex response of the autonomic nervous system. Based on
these results, a new system for the classification of sexual function in women with spinal cord injury (SCI) is
proposed. Moreover, studies related to the treatment of sexual dysfunction in women with cord injury are
reviewed.

Our understanding of the impact of spinal cord
injury (SCI) on the female sexual response and
sexuality is probably greater than that of any other
neurologic disorder. Evolving from a stage in
which only questionnaire studies were available to
address the effects of cord injury on female sex-
uality, there is now a relatively large body of lit-
erature that documents the impact of various types
of cord injuries on the female sexual response.
Additionally, methods of treatment have begun to
be explored. The goal of this chapter is to review
the literature pertaining to the impact of SCI on
the female sexual response, to discuss the issue of
diagnosis of sexual dysfunction in the population
of women with cord injury and to discuss potential
treatment methods currently being evaluated.
The impact of SCI on female sexuality
Prior to the 1990s, a number of questionnaire
studies documented the effects of SCI on women’s
Corresponding author.; E-mail: m.sipski@worldnet.att.net
sexuality (Charlifue et al., 1992; Sipski and
Alexander, 1993; Jackson and Wadley, 1999;
Fisher et al., 2002). These studies, although lack-
ing appropriate controls, definition of the injury
and validation of the questionnaires, showed
that the frequency of sexual activity and sexual
satisfaction is diminished in women after SCI
(Charlifue et al., 1992; Sipski and Alexander, 1993;
Fisher et al., 2002). Sexual activities engaged in by
women after cord injury (Sipski and Alexander,
1993) are unaltered; however, a surprisingly low
percentage of women with cord injury masturbate.
Effect of SCI on female sexual arousal
The effect of cord injury on female sexual arousal
has been evaluated via multiple laboratory-based
trials (Sipski et al., 1995, 1996, 1997, 2001). Two
separate pathways control female genital sexual
arousal: a psychogenic pathway and a reflex path-
way. Thus, the impact of SCI on sexual response
depends on which of these pathways is altered by
the neurologic injury. Women with complete cord

DOI: 10.1016/S0079-6123(05)52030-2 441

442
injury at or above the level of the 6th thoracic
segment (T6) were evaluated in the laboratory to
see the effects of psychogenic, and psychogenic
combined with manual genital stimulation on
heart rate, respiratory rate, blood pressure, vagi-
nal pulse amplitude (Lann and Everaerd, 1998)
(as a measure of genital arousal) and subjective
sexual arousal. A 78 min protocol was used that
consisted of 6 min baseline periods alternating with
two 12 min periods of audiovisual erotic stimula-
tion and two 12 min periods of audiovisual erotic
combined with manual genital stimulation. Women
with complete upper motor neuron injuries affect-
ing their sacral spinal cord had significant increas-
es in subjective sexual arousal with audiovisual
erotic stimulation alone, but no increase in vaginal
pulse amplitude. Adding manual genital stimula-
tion to the audiovisual stimulation resulted in
augmentation in vaginal pulse amplitude without
continuation of the increased subjective sexual
arousal. These results were interpreted as a dem-
onstration of the presence of reflex genital vaso-
congestion in women with complete SCI.
The neurologic control of psychogenic arousal
was also studied in the laboratory in women with a
wide range of levels of SCI (Sipski et al., 2001).
Using the identical 78 min research protocol de-
scribed above, the ability to achieve psychogenic
genital arousal was shown to be related to the
preservation of the combined ability to perceive
pinprick and light touch sensation in the T11–L2
dermatomes (Sipski et al., 2001). Women with
combined scores of 24–32 in these dermatomes as
determined by the ASIA standards (Sipski et al.,
2001), were significantly more likely to demon-
strate psychogenic genital vasocongestion than
women with scores of 9–23. In turn, women with
scores of 9–23 were significantly more likely to
achieve psychogenic genital vasocongestion than
women with scores of 0–8. This information was
interpreted to be evidence for a role of the sym-
pathetic preganglionic neurons (with cell bodies at
the T11–L2 level) in the control of psychogenic
genital arousal.
The control of reflex arousal was also further
studied in the laboratory. Women with incomplete
SCI who were first subjected to psychogenic, then
to manual genital stimulation showed further
increases in their level of genital arousal with
manual stimulation, regardless of whether they
had concomitant increases in level of subjective
arousal (Sipski et al., 1997). This was interpreted
as evidence for the maintenance of reflex genital
arousal in women with upper motor neuron inju-
ries affecting their sacral spinal segments. Further
research compared the effect of the addition of
manual to psychogenic stimulation in women with
upper versus lower motor neuron injuries affecting
their sacral segments (Sipski et al., 2001). The
subjects with upper versus lower motor neuron
injury did not differ in vaginal pulse amplitude,
but vaginal pulse amplitude tended to increase
with upper motor neuron injuries, as would be
expected in the presence of reflex genital vasocon-
gestion. Despite the lack of significance in these
results, the overall psychophysiologic data tend to
validate the hypothesis that reflex lubrication is
maintained in women with SCI and upper motor
neuron injuries affecting their sacral segments, and
that women with cord injury and lower motor
neuron incomplete injuries should still have partial
preservation of reflex lubrication. It follows that
the only subset of women who should not have
the potential for reflex lubrication should be
those women with complete lower motor neuron
injuries.
In another laboratory-based study (Komisaruk
et al., 1997) of the effects of SCI on sexual re-
sponse, the authors hypothesized that vaginal and/
or cervical self-stimulation will not produce per-
ceptual responses in women with complete SCI (as
by the ASIA standards) at or above the highest
level of entry of the hypogastric nerves (T10–T12),
but will produce perceptual responses if the injury
is below T10. The authors studied 6 women with
complete SCI (T10 and/or above), 10 women with
complete lower cord injury (below T10) and 5 un-
injured women as a control group. Perceptual re-
sponse to vaginal and/or cervical self-stimulation
was quantified as the magnitude of analgesia pro-
duced by a calibrated finger compressive force ap-
plied to the vagina or cervix for a period of 12 min
interspersed with non-stimulation control periods.
Significant analgesia was noted in all the groups
including the lower cord injury group. The authors
interpreted the finding of analgesia in the group

with lower injury as support for their original hy-
pothesis. The unexpected finding of analgesia in
the group with injury above T10 was proposed as
evidence for a genital afferent pathway that by-
passes the spinal cord, such as the vagus nerve.
However, women with upper cord injury may also
experience menstrual discomfort, awareness of
vaginal and/or cervical stimulation and orgasms.
Thus, one must realize that the definition of injury
as ‘‘complete’’ is based solely on preservation of
voluntary rectal contraction and/or preservation
of anal sensation and does not provide any other
information with respect to remaining autonomic
function. Moreover, the results of this study must
be considered in view of the fact that the protocol
in which subjects were asked to participate re-
quired them to stimulate their cervix at a specific
intensity with a device constructed as a diaphragm
with a handle on it. In addition, the subjects were
not alone during this test, but had an investigator
in the room with them administering both tactile
and pain testing. Under these circumstances, it is
possible that the alterations in sensory perception
found in these subjects were related to a distrac-
tion effect of the study methodology.
Effects of SCI on orgasm
The ability to achieve orgasm in women with SCI
has also been assessed via questionnaire studies
and in the laboratory (Sipski et al., 1995; Whipple
et al., 1996). In the largest laboratory-based series
to date (Sipski et al., 2001), the ability of women
with traumatic SCI to achieve orgasm was assessed
both historically and in the laboratory. Sixty-two
women with SCI and 21 able-bodied control sub-
jects participated. They were set up with monitors
for heart rate, respiratory rate and blood pressure,
brought into the laboratory, provided with an
erotic video and asked to stimulate themselves to
orgasm any way they would like. Women with all
levels and degrees of SCI were significantly less
likely than able-bodied control subjects to achieve
orgasm. Fifty-five percent of women with SCI re-
ported the ability to achieve orgasm compared to
100% of able-bodied control subjects. In the
443
laboratory, 44% of cord-injured subjects were or-
gasmic compared to 100% of able-bodied con-
trols. The characteristics of orgasm in able-bodied
versus cord-injured subjects were also analyzed.
Despite previous reports that non-genital stimula-
tion is often used as a means to achieve orgasm,
only one woman in this study (Sipski et al., 2001)
chose non-genital stimulation in combination with
genital stimulation. The average latency to orgasm
was significantly greater in cord-injured versus
able-bodied subjects (26 min versus 16 min). The
heart rate, systolic blood pressure and respiratory
rate at orgasm compared to baseline were signif-
icantly greater for both cord-injured and able-
bodied subjects; however, there were no significant
differences between the two groups at any time
period. Diastolic blood pressure was similar in
both groups of subjects at orgasm and baseline;
additionally there was no significant increase in
diastolic blood pressure at orgasm versus baseline.
When two investigators were blinded to subjects’
descriptions of orgasms, they were unable to de-
termine whether the women had complete or in-
complete SCI or were able-bodied control subjects.
The women’s ability to achieve orgasms were
also compared between various groups of women
with spinal cord injuries. No statistically signifi-
cant differences were observed in orgasmic ability
based upon grouping of subjects according to re-
maining sensation at the T11–L2 or S2–S5 derma-
tomes, completeness of injury, or upper or lower
motor neuron injury affecting their sacral cord
segments. Subjects with complete lower motor
neuron injuries affecting S2–S5 ðn ¼ 6Þ were sig-
nificantly less likely to report the historical ability
to achieve orgasm as compared to those with all
other patterns and degrees of SCI (n ¼ 56; 17%
lower motor neuron complete, 59% all other cord
injuries combined; w2 ¼ 3:91; p ¼ 0:048). In the
laboratory, however, there were no significant dif-
ferences in the ability of these women to achieve
orgasm (17% lower motor neuron complete; 46%
all other cord injuries combined; w2 ¼ 1:96;
p ¼ 0:16). Based on the belief that the historical
reporting of orgasm is more accurate than the
women’s performance in the laboratory, these re-
sults were taken as evidence that the occurrence of
orgasm depends on the presence of an intact sacral
444
reflex arc. These authors hypothesized that orgasm
is a reflex response of the autonomic nervous sys-
tem that can be either facilitated or inhibited by
cerebral input.
Recent research points to the fact that orgasm
may be associated with a pattern generator in the
spinal cord. The suggestion has been made that an
ejaculation generator is present in the spinal cord
(Truitt and Coolen, 2002) of male rats. These re-
searchers documented activation of a subset of
lumbar spinothalamic neurons after copulatory
behavior in male but not in female rats (Truitt
et al., 2003). This finding in male rats is similar to
that of the urogenital reflex that is found in anest-
hetized animals with spinal cord transections
above T9. The reflex consists of rhythmic firing
of the hypogastric, pelvic and pudendal motor
nerves in response to self-stimulation (McKenna
et al., 1991; Chung et al., 1988). The peripheral
activity displayed during the urogenital reflex
strongly resembles that seen during human or-
gasm (Bohlen et al., 1982). Vaginal, uterine and
anal sphincter rhythmic contractions are present in
both the urogenital reflex and orgasm, and both
are relatively insensitive to gonadal hormones.
Thus, a pattern of neural activity similar to the
urogenital reflex may underlie the response ob-
served in women with SCI.
The authors hypothesized that if the neurologic
potential to achieve orgasm exists in approximate-
ly 50% of women with all levels of SCI (except
those with complete lower motor neuron injury of
S2–S5), then those women who did not achieve
orgasm must have some intervening variable pre-
cluding them from achieving orgasm. Lack of ed-
ucation and interfering psychologic issues were
mentioned as possible problems in addition to the
overall negative viewpoint in the medical literature
that has previously existed (Money, 1960; Fitting
et al., 1978). Other possible contributing factors
include medications that are commonly taken by
women with SCI such as antidepressant and anti-
spasticity medications that could diminish sexual
responsiveness. Based upon research findings, the
authors recommended that women with SCI
should be educated that longer and potentially
more intense genital stimulation is necessary to
achieve orgasm. Furthermore, they recommended
the development of treatment methods to remedy
orgasmic dysfunction.
Another group of investigators also performed
laboratory studies of the effects of SCI on orgasm.
Sixteen women with complete cord injury at or
below the level T6 were studied during self-stim-
ulation of the anterior vaginal wall and cervix with
a device constructed as a diaphragm with a handle
on it, designed for laboratory studies (mentioned
above) (Whipple et al., 1996; Komisaruk et al.,
1997). Nine-minute control periods alternated with
12 min stimulation periods. The stimulation was at
a specific intensity that was monitored by the sub-
jects. Three of the subjects with SCI reported or-
gasms from cervical stimulation and one subject
also reported orgasm from vaginal stimulation.
Moreover, two of the three subjects had multiple
orgasms during the experimental sessions. These
authors hypothesized that the neural pathway ac-
counting for these subjects’ ability to achieve or-
gasm was the vagus nerve, citing laboratory
studies of animals as the reason for this hypoth-
esis (Ortega-Villalobos et al., 1990; Cueva-Rolon
et al., 1996; Komisaruk et al., 1996; Komisaruk
et al., 2004;). In one of these animal studies, an-
algesia and pupil dilatation in response to genital
stimulation of rats persisted after genital deaff-
erentation by bilateral pelvic, hypogastric and pu-
dendal neurectomy (Cueva-Rolon et al., 1996).
These two residual responses were abolished after
bilateral vagotomy; thus, the authors concluded
that the pelvic, hypogastric and pudendal nerves
are not the only vaginocervical afferent pathways
and that the vagus pathway could remain intact
and functional after SCI. The authors further sur-
mised that this pathway could account, at least in
part, for the reports of perceptual responses to
vaginal or cervical self-stimulation in women with
complete SCI.
This last hypothesis was recently tested in a pilot
study by Komisaruk (Komisaruk et al., 2004) us-
ing functional magnetic resonance imaging to as-
certain whether the region of the brainstem to
which the sensory component of the vagus nerves
project, the nucleus tracti solitarii in the medulla
oblongata, is activated by vaginal/cervical self-
stimulation in women with complete SCI at or
above T10. They found an overall increase in

activation at orgasm of multiple brain regions in-
cluding the nucleus tractus solitarius and the
hypothalamus. While this observation is sugges-
tive, it does not constitute proof. Proof would re-
quire the reversible loss of activation of the
nucleus tractus solitarius by reversible block of
the vagus nerve.
Documentation of sexual dysfunction in women
with SCI
The above documentation of the impact of SCI on
sexual response provides a framework for under-
standing how the injury affects sexual response.
But it does not give us any information about
whether a woman with a cord injury has sexual
dysfunction. According to the International Con-
sensus Development Conference on Female Sexual
Dysfunction (Basson et al., 2000), sexual dysfunc-
tion implies personal distress. Therefore a woman
with a cord injury who has alterations in her sexual
response related to her injury, but does not com-
plain of distress, does not have sexual dysfunction.
Conversely, a woman with a SCI who has no in-
jury-related alterations in her sexual response but
complains of sexual distress, suffers from sexual
dysfunction. In order to remedy this lack of a
means of documentation, the Female Spinal Sex-
ual Function Classification (FSSFC) was proposed
(Sipski et al., 2002). This classification system re-
lies on previous research to define four categories
of sexual function after SCI, document their pres-
ence and associated characteristics and determine
which aspects of the neurologic examination
should be used to determine the likely capacity
for sexual response. Based upon the performance
of the neurologic examination and detailed histo-
ry, one should be able to document the expected
effects of the injury on specific components of
sexual response and also document whether the
subject reports any sexual dysfunction. This latter
issue becomes especially important when the issue
of clinical trials for remedying sexual dysfunction
after SCI is addressed. The FSSFC is currently
being utilized in a study of women with SCI and
multiple sclerosis, to assess its utility for docu-
mentation of the remaining sexual function and
445
presence or absence of sexual dysfunction in wom-
en with SCI.
Improving sexual responsiveness
A number of studies have begun to test therapies
to improve sexual responsiveness in women with
SCI. The majority of these studies have used treat-
ments previously used in able-bodied women, ex-
cept one drug study that tested the efficacy of
medications utilized in men (Sipski et al., 2000a).
The first series of therapies can be described as
cognitive. False positive feedback was used in a
laboratory-based study (Sipski et al., 2000b) to
increase the level of sexual arousal in a sample of
women with SCI. It was unknown whether or not
these women complained of sexual dysfunction.
False positive feedback was shown to increase
psychogenic arousal in women with both complete
or incomplete SCI; however, genital arousal was
only increased in women with incomplete injuries
who had preservation of sensory function in the
T11–L2 dermatomes. This study concluded that
cognitively based therapies might be useful to im-
prove function in this subset of women with SCI.
In another study, the same authors studied the
impact of an anxiety-provoking video on sexual
arousal (Sipski et al., 2004). Subjects viewed two
erotic videos, one of which was preceded by a
neutral video and another that was preceded by an
anxiety-provoking video. In subjects with impaired
genital responsiveness to psychogenic erotic stim-
ulation (T11–L2 combined ASIA scores less than
23), anxiety pre-exposure resulted in a small in-
crease in genital responsiveness to erotic stimula-
tion compared to neutral pre-exposure. In subjects
who had mostly intact genital responsiveness (cord
injury subjects with T11–L2 combined ASIA
scores 24–32) and able-bodied subjects, anxiety
pre-exposure resulted in decreased genital respon-
siveness compared to neutral pre-exposure. On the
assumption that anxiety is associated with sympa-
thetic activation, the authors concluded that there
is a therapeutic benefit in manipulating the sym-
pathetic nervous system in cord-injured subjects
with impaired, but not absent, ability to achieve
psychogenic genital vasocongestion.
446
Only one study has been published on the effects
of medications on sexual responsiveness after SCI
in females. In a laboratory-based, double-blind
crossover design study (Sipski et al., 2000a) the
effects of sildenafil 50 mg versus placebo were
compared on vaginal pulse amplitude, subjective
arousal and autonomic function. Participants un-
derwent a 78 min protocol, in which they were
subject to audiovisual erotic stimulation and au-
diovisual erotic combined with self-applied man-
ual stimulation alternating with baseline periods.
A statistically significant increase in subjective
arousal was noted with the use of the medication.
A borderline significant effect of drug administra-
tion was noted on vaginal pulse amplitude. In both
treatment conditions, sildenafil and placebo, both
visual and visual plus manual stimulation resulted
in small increases in blood pressure (
3–5 mmHg)
above baseline. Regardless of the type of stimula-
tion, however, the mean blood pressure of subjects
on sildenafil was 4–5 mmHg lower compared to
placebo. Similarly, regardless of the type of stim-
ulation, there was a slight increase in baseline
heart rate (4–5 beats/min) with sildenafil compared
to placebo.
Although the above studies begin to outline
possible methods of treatment for sexual arousal
dysfunction for women with SCI, data are still
preliminary and have yet to be translated into
clinical utility. Other studies currently in progress
are examining the effect of the use of medications
to improve sexual response. As new medications,
such as the testosterone patch, become available,
their effects on sexual desire and arousal dysfunc-
tion after SCI will be assessed.
No controlled clinical trials have been designed
to treat orgasmic dysfunction after SCI. Based
upon the assumption that orgasm is a reflex re-
sponse of the autonomic nervous system and that
a spinal pattern generator exists, it may be possible
to train the pattern generator. This concept is
similar to work being done with ambulation train-
ing (Wernig et al., 1995, 1998) and is the founda-
tion for a study to test the efficacy of EROS
(clitorical vacuum stimulation procedure) therapy
(Billups et al., 2001) versus vibratory stimulation
that is currently underway. Both of these therapies
are designed to stimulate a reflex response and
could be useful to treat sexual dysfunction in
women with SCI.
Compared to other neurologic injuries, knowl-
edge of the impact of SCI on the female sexual
response is relatively advanced. A study of women
with predominantly spinal multiple sclerosis has
recently been initiated to determine if the impact
of spinal multiple sclerosis lesions on sexual re-
sponse in women will be similar to that of trau-
matic SCI. It is hoped that knowledge about the
impact of SCI on the female sexual response can
serve as a model to study not only multiple scle-
rosis but also other neurologic disabilities.
Acknowledgments
This work was supported in part by funds from
NIH R01 HD 30149
References
Basson, R., Berman, J., Burnett, A., Derogatis, L., Ferguson,
D., Fourcroy, J., Goldstein, I., Graziottin, A., Heiman, J.,
Laan, E., Leiblum, S., Padma-Nathan, H., Rosen, R., Se-
graves, K., Segraves, R.T., Shabsigh, R., Sipski, M., Wagner,
G. and Whipple, B. (2000) Report of the international con-
sensus development conference on female sexual dysfunction:
definitions and classification. J. Urol., 163: 888–893.
Billups, K.L., Berman, L., Berman, J., Metz, M.E., Glennon,
M.E. and Goldstein, I. (2001) A new non-pharmacological
vacuum therapy for female sexual dysfunction. J. Sex Marital
Ther., 27: 1435–1441.
Bohlen, J.G., Held, J.P., Sanderson, M.O. and Anderson, M.O.
(1982) The female orgasm: pelvic contractions. Arch. Sex
Behav., 11: 367–386.
Charlifue, S.W., Gerhart, K.A., Menter, R.R., Whiteneck,
G.G. and Manley, M.S. (1992) Sexual issues of women with
spinal cord injuries. Paraplegia, 30: 192–199.
Chung, S.K., Mc Vary, K.T. and Mc Kenna, K.E. (1988) Sex-
ual reflexes in male and female rats. Neurosci. Lett., 94:
343–348.
Cueva-Rolon, R., Sane, G., Bianca, R., Gomez, L., Beyer, C.,
Whipple, B. and Komisaruk, B.R. (1996) Vagotomy blocks
responses to Vaginocervical stimulation after genitospinal
neurectomy in rats. Physio. Behav., 60: 19–24.
Fisher, T.L., Laud, P.W., Byfield, M.G., Brown, T.T., Hayat,
M.J. and Fiedler, I.G. (2002) Sexual health after spinal cord
injury: a longitudinal study. Arch. Phys. Med. Rehabil., 83:
1043–1051.
Fitting, M.D., Salisbury, S., Davies, N.H. and Mayclin, D.K.
(1978) Self-concept and sexuality of spinal injured women.
Arch. Sex Behav., 7: 143–156.

Jackson, A.B. and Wadley, V. (1999) A multicenter study of
women’s self-reported reproductive health after spinal cord
injury. Arch. Phys. Med. Rehabil., 80: 1420–1428.
Komisaruk, B.R., Bianca, R., Sansone, G., Gomez, L.E.,
Cueva-Rolon, R., Beyer, C. and Whipple, V. (1996) Brain-
mediated responses to vaginocervical stimulation in spinal
cord-transacted rats: role of the vagus nerves. Brain Res.,
708: 128–134.
Komisaruk, B.R., Gerdes, C.A. and Whipple, B. (1997) ‘‘Com-
plete’’ spinal cord injury does not block percepetual respons-
es to genital self-stimulation in women. Arch. Neurol., 544:
1513–1520.
Komisaruk, B.R., Whipple, B., Crawford, A., Liu, W.C.,
Kalnin, A. and Mosier, K. (2004) Brain activation during
Vaginocervical self-stimulation and orgasm in women with
complete spinal cord injury: FMRI evidence of mediation by
the vagus nerves. Brain Res., 1024: 1513–1520.
Lann, E. and Everaerd, W. (1998) Physiological measures of
vaginal vasocongestion. Int. J. Impot. Res., 10: 107–110.
McKenna, K.E., Chung, S.K. and McVary, K.T. (1991)
A model for the study of sexual function in anesthetized
male and female rats. Am. J. Physiol., 30: R1276–R1285.
Money, J. (1960) Orgasm in the dreams of paraplegic men and
women. Arch. Gen. Psychiatry., 3: 373–382.
Ortega-Villalobos, M., Garcia-Bazan, M., Solano-Flores, L.,
Ninomiya-Alarcon, J., Guvara-Guzman, R. and Wayne, M.
(1990) Vagus nerve afferent and efferent innervation of the
rat uterus: an electrophysiological and HRP study. Brain
Res. Bull., 25: 365–371.
Sipsk, M.L., Rosen, R.C., Alexander, C.J. and Gomez-Marin,
O. (2004) Sexual responsiveness in women with spinal cord
injuries: differential effects of anxiety-eliciting stimulation.
Arch. Sex Behav., 33(3): 295–302.
Sipski, M.L. and Alexander, C.J. (1993) Sexual activities, re-
sponse and satisfaction in women pre- and post-spinal cord
injury. Arch. Phys. Med. Rehabil., 74: 1025–1029.
Sipski, M.L. and Alexander, C.J. (2002) Documentation of the
impact of spinal cord injury on female sexual function: the
female spinal sexual function classification. Top Spinal Cord
Inj. Rehabil., 8(1): 63–73.
Sipski, M.L., Alexander, C.J. and Rosen, R.C. (1995) Orgasm
in women with spinal cord injuries: a laboratory-based as-
sessment. Arch. Phys. Med. Rehabil., 76: 1097–1102.
447
Sipski, M.L., Alexander, C.J. and Rosen, R.C. (1995) Physio-
logic parameters associated with psychogenic sexual arousal
in women with complete spinal cord injuries. Arch. Phys.
Med. Rehabil., 76: 811–818.
Sipski, M.L., Alexander, C.J. and Rosen, R.C. (1996) Physio-
logic parameters associated with the performance of a dis-
tracting task and genital self-stimulation in women with
complete spinal cord injuries. Arch. Phys. Med. Rehabil., 77:
419–424.
Sipski, M.L., Alexander, C.J. and Rosen, R.C. (1997) Physio-
logic parameters associated with psychogenic sexual arousal
in women with incomplete spinal cord injuries. Arch. Phys.
Med. Rehabil., 78: 305–313.
Sipski, M.L., Alexander, C.J. and Rosen, R.C. (2001) The ne-
urologic basis of sexual arousal and orgasm in women: effects
of spinal cord injury. Ann. Neurol., 49: 35–44.
Sipski, M.L., Rosen, R.C., Alexander, C.J. and Hamer, R.M.
(2000a) Sildenafil effects on sexual and cardiovascular re-
sponses in women with spinal cord injury. Urology, 55:
812–815.
Sipski, M.L., Rosen, R., Alexander, C.J. and Hamer, R.
(2000b) A controlled trial of positive feedback to increase
sexual arousal in women with spinal cord injuries. Neuro.
Rehabil., 15: 145–153.
Truitt, W.A. and Coolen, L.M. (2002) Identification of a po-
tential ejaculation generator in the spinal cord. Science,
297(5586): 1566–1569.
Truitt, W.A., Shipley, M.T., Veening, J.G. and Coolen, L.M.
(2003) Activation of a subset of lumbar spinothalamic neu-
rons after copulator behavior in male but not female rats.
J. Neurosci., 23: 325–331.
Wernig, A., Muller, S., Nanassy, A. and Cagol, E. (1995) La-
ufband therapy based on ‘‘Rules of spinal locomotion’’ is
effective in spinal cord injured persons. Eur. J. Neurosci., 7:
823–829.
Wernig, A., Nanassy, A. and Muller, S. (1998) Maintenance of
locomotor abilities following Laufband (treadmill) therapy in
para- and tetraplegic persons: follow-up studies. Spinal Cord,
36: 744–749.
Whipple, B., Gerdes, C.A. and Komisaruk, B.R. (1996) Sexual
response to self-stimulation in women with complete spinal
cord injury. J. Sex Res., 33: 231–240.
 Subject Index
adrenergic 52–54, 61–63, 65
adrenergic antagonists 429
adrenoceptor 62, 238–239, 241, 289, 292–296, 350, 352
afferent 53, 249, 266
afferent arbor 249, 251–252, 255, 259, 300, 305–306, 309
afferent neurons 59, 63, 75–76, 100–101, 361, 363, 404,
409
anal canal 207–209, 321, 345–346, 351–352, 360, 375,
380–381
anatomy 27–28, 52, 60, 65, 164, 317, 346, 360
anejaculation 433–434, 437
anorectal 326, 336, 338–339, 341–342, 347, 365, 373–376,
378–380
anorectum 346–347, 349, 355
anti-spasticity 444
anticholinergic 53, 164, 171, 178, 327
areflexia 73, 119, 341
areflexic rectum 339
areflexive bladder 164
arousal 388–392, 394, 396, 422, 441–442, 445–446
arrhythmias 256, 275–285, 329, 341, 397
arterial pressure 39–40, 42, 226, 231, 245–246, 269–270,
275, 278, 281, 283–284, 289, 299
autonomic 42, 73, 237, 241, 265–266, 272, 281–282, 292,
296, 392, 396
autonomic dysreflexia 11, 13, 27, 35, 39–42, 74, 76,
147–148, 166, 173, 178, 223–226, 233, 235, 237–241,
245–249, 251–252, 255–256, 259, 265–266, 268–272,
275, 278, 280–281, 289–290, 299–302, 304–305,
307–310, 328, 336, 341, 354, 369, 387, 390–391,
393, 395–397, 408, 421, 423, 432, 435
barrier function 54, 137, 140
bladder 51–56, 59–65, 68–78, 85–86, 88–90, 92, 97–104,
106–109, 117–124, 126–127, 130–132, 135–137,
139–143, 147–156, 158, 163–188, 195, 200–201,
205–211, 213, 215, 217, 225–226, 236, 238–239,
241, 246, 251, 259, 265–267, 290, 293, 317, 325,
328–330, 336–337, 341–342, 345, 354–355, 359–362,
367, 387–390, 392, 402, 406, 420–422, 430–431, 433,
435
bladder afferent neurons 59, 63–64, 75–76, 103, 108, 111,
155, 337
bladder cholinergic 143
bladder control 51, 119, 165, 176, 184, 186–188
449
bladder detrusor 118, 121, 126, 139, 195, 206–207
bladder dyssynergia 147, 327
bladder guarding reflex 213
bladder hyperactivity 158
bladder hyperreactivity 143
bladder hyperreflexia 97, 107, 158, 205
bladder hyporeflexia 136
bladder incontinence 155, 184
bladder inflammation 142
bladder interneurons 154
bladder preganglionic 182, 363
bladder primary afferents 147, 149–150, 155, 158
bladder reflex 74, 77, 97, 99–101, 107, 121, 126,
200, 213
bladder urothelium 137, 143
bladder–external urethral sphincter 119, 124–126, 130,
132
bladder–sphincter 97, 99
bladder–sphincter dyssynergia 85, 97, 107, 130, 154, 422
bladder, urethra 69
blood pressure 11, 41–42, 45, 172, 223–227, 231–241,
246, 267, 275, 278–281, 289–290, 299–301, 304–308,
392, 396, 435, 443, 446
bowel 19, 53, 172, 188, 195, 200–201, 205–211, 213, 215,
217, 225–226, 234, 241, 246, 265–266, 317–318,
320–321, 323–330, 335–337, 339–342, 345, 354–355,
362, 369, 375, 378, 382, 387–392, 396
bowel dysfunction 315, 317, 335, 340, 342, 345
bowel guarding reflex 213
bowel incontinence 391, see also fecal incontinence
bulbospongiosus 359, 361–363, 365–368, 415–417, 420
cardiovascular 11, 17, 27, 39–45, 59, 104, 221, 223–225,
227–228, 231, 235, 238, 245, 275–276, 279–281, 285,
328, 387, 390–392, 396
cell replacement 201–202
cholinergic 11, 17, 53–54, 60, 62, 65, 78, 92, 107, 211,
271, 291, 320–321, 323, 337, 345, 351, 429
colon 206, 208, 210, 246–247, 251, 256, 259, 267, 281,
290, 300–301, 304–305, 308, 318, 320–323, 325–329,
337, 341, 346–350, 354, 360, 363–364, 379, 406, 408
colonic 266, 268, 271, 318, 320–329, 337, 339–342, 346,
349–350, 354–355
colonic transit times 345, 354, see also bowel transit
times
450
constipation 210, 318, 321, 323–329, 336–337, 339–342,
345, 354–355, 360, 374–375
defecation 174, 177, 206, 209, 234, 318, 322–323,
325–326, 329, 335, 339–341, 345–347, 349–350,
352–356, 360, 363–364, 366–367, 369, 373–375,
377, 416, 418, 422
detrusor 51–54, 60, 65, 73–74, 76, 78, 98, 107, 119, 122,
126–128, 135, 139, 143, 148, 155, 164, 168, 170–176,
178–179, 184–188, 206, 208, 210–211, 213, 360,
367
detrusor areflexia 74, 341
detrusor hyperactivity 52, 142
detrusor hyperreflexia 53, 73–74, 76–78, 135, 164, 166,
170–171, 176, 211, 213, 215, 367
detrusor–external urethral sphincter 119, 123, 125, 127,
130
detrusor–sphincter 53, 73, 74, 76
detrusor–sphincter dyssynergia 74, 76, 112, 155, 148,
164, 166–167, 170, 178–179, 185, 188, 354, 359,
367
dorsolateral funiculus 12, 45, 67, 91, 104, 149, 199
dysreflexia 13, 39, 42, 45, 73, 237, 241, 265–266, 272,
281–282, 292, 296, 301, 392, 396, 432
dyssynergia 52–54, 73–74, 76, 85, 90, 92, 99, 360,
367
dyssynergy 52, 54
ejaculation 239, 387, 391–392, 394–397, 402,
404–405, 407–409, 415–423, 427–428,
430–437, 444
ejaculation vibroejaculation 435
electrical stimulation 90, 119, 163, 168, 170–171,
173–174, 176–178, 181–183, 185–188, 213, 277,
284, 295, 365, 378, 418, 429, 431, 436
electroejaculation 391, 409, 422–423, 427–428, 431–436
electromicturition 341
electrophysiology 36, 275, 278, 280, 282, 284–285, 379
enkephalin 17, 21, 72, 100, 111
enteric 206, 208, 317–318, 320, 323, 327, 330, 337, 345,
347, 349–350, 353, 363, 373
enteric reflexes 322–323, 339
erection 12, 174, 177, 235, 366–369, 387, 389–390,
392–396, 402, 404, 408, 416–417, 420–423,
427–430, 436
eureflexic 300
external sphincter 52
external urethral sphincter 61–62, 67–68, 71, 77, 85, 88,
99–100, 117–123, 125–128, 130, 132, 148, 155, 164,
169, 174, 176–177, 182, 184, 362, 367, 416
external urethral sphincter motoneurons 85, 87
fecal incontinence 205, 318, 325–326, 328–329, 335,
338–342, 345–346, 355, 376, see also bowel
incontinence
female infertility 436
female sexual function 392, 441
flaccid bladder 52, 166
functional magnetic resonance imaging 373, 381, 444
gallbladder 325
ganglion 20, 31, 33, 60–61, 63, 75, 102, 104, 106, 108,
112, 150, 152, 271, 277, 289–292, 323, 351, 361, 393,
402–404
ganglionectomy 402
ganglionic 137, 164, 207, 235, 239, 241, 248, 272, 278,
290–292, 323, 348, 361, 429
gastric emptying 324, 328, 373–374
gastrointestinal 102, 104, 235, 239, 246, 299, 317–320,
323–325, 327, 330, 335, 340, 345, 348–349, 354–355,
360, 366, 373–374, 379–380, 382
gastrointestinal transit times 354–355, see also colonic
transit times
gene therapy 369
genitalia 369, 388–389, 401, 406–407
glutamic acid 13
guarding reflex 205, 208–210, 211, 213, 215
human spinal cord 45, 373, 389, 397, 415
humoral 323
hyperactivity 51, 53, 148
hyperreactivity 289, 294–296, 367
hyperreflexia 11, 53, 59, 76, 97, 103, 112, 135, 164, 173,
210–211, 213, 238, 359, 366–367
hyperreflexic bladder detrusor 119
hyperreflexive bladder 74, 171–173, 176, 178, 180, 183,
186
hypertension 11, 39, 42, 54, 73, 223, 224–225, 231–232,
235, 237–239, 241, 245–246, 265–271, 275, 278,
280–281, 290, 299–300, 328, 390, 396
hypogastric 52, 60–62, 97, 207–208, 241, 267, 270, 321,
347, 360–361, 366, 401–403, 405, 416, 429, 442, 444
hyporeflexive bladders 167
hypotension 223
immunocytochemistry 252
implantation 154–155, 166, 170, 173–178, 181, 183, 185,
187, 199–200
incontinence 11, 51, 53, 55, 117, 130, 143, 164, 166,
171–172, 175, 177, 184–185, 205, 209, 211, 325–329,
336, 339, 341, 345, 374–376, 390
inflammation 103, 107, 111, 135, 137, 140, 152, 155, 246,
254–255, 259, 299, 306, 375, 432, 434

intermediolateral cell column 12–14, 17, 20–21, 40,
258–259, 267, 269, 272, 365–366, 416
interneuron 11–12, 17, 21–22, 28–34, 77, 86–87, 89–92,
101, 118, 148–149, 153–154, 181–182, 209, 223, 249,
266, 268–271, 305, 363, 365, 404–405, 417–418, 423
interstitial cells 322, 348–350
intraurethral 60, 163, 167, 169, 179–180, 182, 185
lateral funiculus 34, 39–40, 42–43, 45, 67, 69, 415, 419
lower urinary tract 51, 59–60, 62, 65–67, 69, 72, 74, 78,
90, 92, 97–99, 101–103, 107–108, 111, 112, 117–122,
125, 126–127, 130, 142–143, 155, 163–165, 168–169,
175, 184–186, 188, 195, 200–202, 206–207, 211, 232
lumbosacral 59–60, 63, 73, 97–101, 103–104, 107, 109,
117, 126, 128, 147, 152–154, 158, 195–198, 200, 202,
205, 251, 265–271, 338–339, 354, 359, 361, 365–367,
369, 402–404, 407, 415, 417–418, 420, 423
lumbosacral preganglionic 103
lumbosacral propriospinal 268–269, 271
lumbosacral propriospinal projections 265
lumbosacral sympathetic 207
lumbosacral ventral root avulsion 198
male infertility 428, 434, 436–437
male sexual function 369, 389, 393, 415
mechanical devices 163, 165, 168, 185, 329
medulla 266
megacolon 327
mesenteric 318, 321, 323, 347
mesenteric ganglion 321
microstimulation 181–182, 187, 418–420
microstimulation, urethral 163
micturition 59–60, 62, 65, 69–74, 76, 85–86, 88–92,
97–100, 103, 111, 117–118, 122, 124, 128, 130–131,
137, 147–149, 152–156, 158, 164, 168–170, 172, 174,
176–178, 181–183, 195, 200, 206–208, 213, 234, 359,
365, 367–368, 416, 420, 422
micturition reflex 52, 59, 70, 72, 74–77, 90, 97–100, 103,
107–108, 111–112, 119, 130, 147–149, 152–153, 154,
155, 158
motoneurons 28, 31–32, 65, 67, 70, 73, 77, 85–92, 99,
101, 112, 126, 128–130, 164, 170, 177, 182, 196–201,
208–209, 393, 405, 416, 420–421, 429
motoneurons, sympathetic preganglionic neurons 27
mouse genetics 299, 308
myenteric 101–102, 320–321, 326, 328, 337, 341,
347–348, 351
necrospermia 427–428, 432, 434–437
nerve growth factor 76, 108–112, 140, 150–152, 155–156,
158, 199, 265–271, 290
451
neurogenic bladder 51, 54–55, 72, 163, 165, 168,
170–171, 175, 211, 390
neurogenic detrusor 73, 164, 172
neurons 249, 265–266, 299, 402
neuropeptide 13, 15, 17, 21, 63, 72, 100, 111, 143, 237,
296, 320, 337
neurophysiology 27–30, 34, 36, 168, 205–207, 209,
215–217, 378, 389, 395, 397
neuroprosthesis 171–172, 176, 183, 186, 188
neuroprosthetic 174, 187
neuroprotection 195, 199, 202, 308–309
neurotrophic factors 76, 108, 111, 112, 150, 152, 155,
157, 200, 259, 337
neurotrophin 108, 136, 152, 154, 156, 199
neurovascular 295–296
neurovascular transmission 292
non-cholinergic 60, 429
noradrenaline 14, 19, 87–88, 232–233, 235–238, 247,
292–295, 353, 429
noradrenergic 69
orgasm 388–389, 391–392, 394–396, 408, 422, 427–428,
430, 441, 443–446
orgasm/ejaculation 390
orgasmic 388–389, 392, 443–444, 446
orthostatic hypotension 223, 225–227, 231–236
parasympathetic 52, 60–62, 64–65, 68–70, 76, 78,
97–102, 104–105, 108, 118, 125–128, 142, 148, 151,
154, 164, 173, 177–183, 197, 200–201, 206–208,
210, 215, 226, 231–232, 268, 277–280, 282, 284,
320, 323–324, 326, 329, 336, 340, 347, 351,
353–354, 359, 361–366, 369, 382, 389, 393–394,
401, 404, 416, 429
parasympathetic defecation reflex 363, 365
parasympathetic ganglia 62
parasympathetic micturition 70
parasympathetic postganglionic 78, 101
parasympathetic preganglionic 21–22, 66–67, 69–70, 77,
101, 148–149, 173–174, 180, 184, 198, 369, 416, 429
parasympathetic reflex 65
paroxysmal hypertension 231, 235, 241, 266
pelvic 22, 52, 59–63, 69–70, 73, 86, 97–98, 101–102, 140,
148, 154–155, 163–164, 168–169, 172–173, 177, 181,
184–185, 197, 200–201, 205–211, 215–217, 252, 265,
267, 318, 320, 323, 326, 336, 338–340, 347, 360–362,
364–367, 374–377, 379, 388, 393, 401–408, 416,
418–423, 429, 444
pelvic floor 52, 169, 170–171, 355, 359, 363, 367, 393
pelvic ganglion 21–22, 196, 198, 361
pelvic reflex 213, 323
pelvic somato-visceral reflexes 205
452
pelvic sphincters 215
pelvic sympathetic 33
pelvic urethra 416
pelvic viscera 11, 21–22, 86, 101, 206–207, 235, 239, 265,
267, 270–272, 361, 379, 401
penis 100, 171, 361, 367–368, 393–395, 401, 403–409,
415, 417–423, 429–430, 432
perineal 86–87, 89–90, 148, 154, 174, 178, 195, 361, 393,
402–404, 406, 409, 415–420, 422–423
periurethral 60
perivascular nerves 293, 295
piloerection 246
postganglionic 61, 78, 98, 277–278, 289–292, 295–296,
321, 420, 429
postganglionic parasympathetic 320
postganglionic sympathetic 28, 419
preganglionic 18, 40, 52, 60–61, 67, 69–72, 87, 100–101,
103, 149, 154, 164, 197, 200–201, 207, 231, 236,
248–249, 251, 258–259, 265–266, 271, 277, 289–293,
295–296, 299, 321, 337, 363, 365, 393, 402, 404, 420
preganglionic neurons 18, 28, 31–32, 118, 266, 271, 308,
321
preganglionic parasympathetic 118, 172, 196–201, 348
preganglionic sympathetic 269, 361
preproneuropeptide 17
presynaptic 423
presynaptic inhibition 89–90, 92, 418, 420, 423
primary afferents 27–28, 35–36, 90, 92, 149–152, 155,
196, 248, 267–268
proenkephalin 155
propriospinal 33–35, 265, 268, 270–272, 369
propriospinal interneurons 268
psychogenic 394
pudendal 52, 54, 60, 62, 64, 68, 86–90, 97, 164, 169, 171,
181–184, 207–210, 213, 215, 338, 347, 360–363,
365–366, 393, 395, 401, 403–405, 415, 417, 419–421,
423, 429, 431, 444
pudendal motoneuron 67, 364, 418, 420–422
pudendal motor nuclei 418
pudendal primary afferents 417
pudendal reflex 210, 213, 415, 417–422
pudendal sympathetic 415, 419, 421
rat 20–21, 33–36, 40, 60, 62–63, 66–72, 74–78, 88–89, 92,
98, 100–103, 107, 117–120, 122, 124, 127, 131,
151–154, 158, 196–200, 238, 246–253, 256–259,
265–271, 276, 278–279, 281–284, 289–290,
293–296, 301–303, 306, 309, 360–361, 363–368,
416–417, 420, 422–423, 433
rectal sensation 338, 340–341, 375, 379, 381
rectum 206–208, 210, 215, 235, 239, 318, 320–321, 323,
326, 329, 339–342, 345–348, 350–355, 360–361,
363–364, 373, 375–376, 379–381, 402, 422, 431
reflex 51–52, 55, 60, 64–65, 69, 74, 76, 78, 85, 89, 91–92,
97, 101, 103, 111, 119, 121, 127, 130, 135, 137, 143,
148–149, 153, 168, 173, 195, 205, 207–211, 215, 224,
233, 235–236, 239, 241, 245–246, 248–249, 252, 256,
268, 276, 289–290, 292, 296, 301, 318, 323, 326, 328,
336–341, 346–347, 353, 355, 359, 363, 367–369,
373–375, 377, 393–395, 403, 405, 408–409, 416–419,
421–423, 428, 432, 435, 441–442, 444, 446
reflex arousal 442
reflex bladder 74, 76, 97, 117, 120–121, 128, 131
reflex colonic 323, 328
reflex defecation 174, 178, 340, 355
reflex detrusor 55, 73
reflex ejaculation 435
reflex erections 174, 178, 393–394
reflex micturition 148, 178
reflex sympathetic 290
reflex vasoconstrictor 226
reflexes 27, 52, 59, 65, 69, 76–77, 88, 91, 98–100,
118–119, 164, 183, 195, 205–206, 208–211, 213, 215,
224–225, 231, 233, 235–236, 245–246, 248–249,
289–290, 292, 296, 299, 301, 320–321, 323–324,
330, 338, 341, 349, 359–360, 363, 365–367, 369, 373,
376, 397, 430, 436
reflexive bladder 122, 164, 171, 173, 178
reflexogenic 394, 404, 408
reflexogenic erection 393–394
reinnervation 13, 195, 200–202, 291–292, 302
renal sympathetic 28–30, 33–36, 248
reproductive 366, 387, 401–402, 404–409, 416, 436
secondary injury 302, 369
semaphorin 3A 265, 267–268, 270
sensory 51, 136
serotonergic 69
serotonin 14–15, 87, 88, 117–118, 125, 130, 258, 322,
365–366, 420, 423
sexual dysfunction 195, 201, 259, 317, 345, 389, 408, 415,
428, 436, 441, 445–446
sexual rehabilitation 388–390, 397
sexual response 388, 391–392, 416, 423, 441–442, 445–446
sexuality 387–390, 392, 397, 427, 441
somato-visceral 207, 210, 213, 216
somatosympathetic 98
spasticity 52–53, 170, 177, 225, 359–360, 366–368, 430
sphincter 11, 51–55, 59, 64, 67–68, 70, 73–74, 77–78,
85–86, 88–92, 97–99, 121–122, 135, 147, 154, 178,
197, 205–209, 211, 213, 215, 241, 318, 320–321,
323–329, 335, 338–342, 345–347, 350–355, 359–361,
 453

363–367, 373, 375–376, 378, 416, 418, 422–423, 430,
444
sphincter dyssynergia 74, 85, 90–91, 155, 205,
211, 213
sphincter guarding reflexes 211
sphincter hyperreflexia 359
sphincter motoneuron 67, 70, 85–92
sphincter reflex 77, 85, 92, 205, 215
sphincter spasticity 52, 360
sphincterotomy 54, 148
spinal sympathetic interneurons 27–28, 30
spinal voiding reflexes 99
sprouting 35, 76, 99, 128, 147–148, 150–151, 153–155,
157–158, 213, 224, 245, 248, 251–252, 258–259,
265–271, 289–292, 296, 305, 307–308, 369
sprouting primary afferents 147
stimulation 278, 282–283
stoma 329, 342
substance P 17, 20, 77, 87, 100, 111, 136, 149, 266–267,
320, 337, 405
sudden cardiac death 278
sympathetic 17, 19, 21–22, 27–30, 32–33, 36, 39–40, 45,
52, 60–62, 69, 73, 97–99, 101, 118, 164, 206–208,
223–226, 231–236, 238–239, 246, 248–249, 258,
265–267, 270–272, 275–280, 282, 284–285, 289–293,
295–296, 299, 308, 320–321, 323–326, 328, 336–339,
347, 351–354, 359, 361, 366, 382, 389–390, 393–394,
401–402, 416, 420–421, 423, 429–430, 441, 445
sympathetic activation 445
sympathetic adrenergic 54
sympathetic ganglia 239, 289, 291, 361
sympathetic interneurons 27–36
sympathetic neurons 278
sympathetic pathways 61
sympathetic postganglionic 31, 419–421
sympathetic preganglionic 18–22, 27–36, 39–40, 66, 101,
223–226, 246, 248, 265, 267–268, 270–272, 299, 301,
305, 308, 366, 416, 429, 442
sympathetic reflex 27, 33, 36, 69, 246, 266, 300
sympathoadrenal preganglionic neurons 21
synaptic plasticity 301, 304, 307
transducer function 136
transganglionic 100, 402
transit times 324, 326, 336–337, 341, 345, 350, 354–355
transmission 295–296
transurethral 177
tyrosine hydroxylase 14
ultrastructure 137, 139–140
urethra 51–52, 55, 59–65, 68–72, 74, 76, 78, 85–86,
88–90, 92, 97, 100, 103, 121–122, 135, 150, 154,
163–167, 174–175, 177, 180–183, 185, 187, 207–209,
211, 239, 361, 416–418, 420, 422–423, 430–431
urethra sphincter motoneurons 88
urethral sphincter 62, 65, 68, 70, 73, 78, 85–86, 97, 99,
118,135, 346
urethral sphincter reflex 86
urinary 51–55, 135–136
urinary bladder 19, 21, 51–52, 55, 59–60, 64–65, 67–68,
70, 73, 75, 76, 97–100, 102–104, 107–108, 111, 112,
118,135, 136, 140–141, 143, 163, 207–208, 235, 237,
239, 241, 246, 299, 354, 360, 431
urinary bladder areflexia 99
urinary bladder detrusor 99
urinary bladder hyperreflexia 101
urinary bladder urothelium 142
urinary tract 62, 68
urothelium 54, 63–64, 135–137, 140–143, 166
vasoconstrictor 73, 233, 237, 272, 289, 291, 293, 295–296,
393
vasoconstrictor preganglionic 271
vasomotor 45
vasomotor pathways 39, 42, 44–45, 225
ventral root avulsion 197–200
ventrolateral medulla 13–14, 266, 365, 415
vesicosympathetic reflex 69
vesicoureteric 166
vibroejaculation 428, 432, 434, 436
viscera 140, 265, 408
viscero-sympathetic reflex 265

tachy-arrhythmias 280 wallerian degeneration 299, 302, 304

testicular function 430, 434 wlds 299, 302, 304–308