Europace (2004) 6, 192e198

Abnormalities of sympathetic and parasympathetic autonomic function in subjects with defaecation syncope
Louise Allana, Emma Johnsa, Mira Doshib, Rose Anne Kennya,c, Julia L. Newtona,c,)
a

Downloaded from http://europace.oxfordjournals.org/ by guest on October 1, 2012

Cardiovascular Investigation Unit, Royal Victoria Inrmary, Newcastle upon Tyne NE1 4LP, UK North Tyneside General Hospital, North Shields, UK c School of Clinical Medical Sciences, The Medical School, University of Newcastle upon Tyne, UK
b

Submitted 10 July 2003, and accepted after revision 16 January 2004

KEYWORDS
defaecation syncope; parasympathetic; autonomic nervous system; sympathetic; autonomic failure

Abstract Defaecation syncope is dened as blacking out at, or around, the time of defaecation. It is associated with increased mortality; however, patients rarely voluntarily report symptoms. We have examined autonomic function in a cohort of patients with defaecation syncope. Methods We prospectively identied all subjects referred to our unit with symptoms of defaecation syncope or presyncope on direct questioning. All subjects had autonomic function tests using beat to beat blood pressure measurement synchronized with ReR interval allowing real time assessment of autonomic function. Results Seven patients were identied who presented with defaecation syncope. Compared with age and sex matched controls, subjects had abnormalities of both sympathetic and parasympathetic autonomic function consistent with mildemoderate autonomic failure. On specic intervention syncope stopped in all subjects: 3 had culprit medication withdrawn, 3 received medication to increase blood pressure and 1 in whom cardioinhibition was demonstrated improved with permanent pacemaker insertion. Two subjects who had colonoscopy had profound haemodynamic changes during the procedure associated with syncope. Conclusions Symptoms of syncope on defaecation are associated with autonomic failure. With appropriate therapeutic intervention our subjects all improved. 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.

) Corresponding author. Cardiovascular Investigation Unit, Falls and Syncope service, C/O Elderly Ofces, Royal Victoria Inrmary, Newcastle NE1 4LP, UK. Tel.: D44-191-232-5131x24128; fax: D44-191-222-5638. E-mail address: julie.newton@nuth.northy.nhs.uk (J.L. Newton). 1099-5129/$30 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.eupc.2004.01.003

Autonomic dysfunction in defaecation syncope

193 function) maximum score 4, the Cardiovascular heart rate index (CVHRI, indicator of parasympathetic function) maximum score 3, and the Sudomotor index, maximum score 3. In addition Sandroni et al. [7] have described the progression of abnormalities of the phases of the Valsalva in worsening autonomic failure. Here we describe abnormalities of autonomic function determined using the more specic Lows criteria in a cohort of patients with defaecation syncope and/or presyncope attending our Cardiovascular Investigation Unit.

Introduction
Defaecation syncope or presyncope are symptoms that though not uncommon, are rarely voluntarily reported. It is dened as syncope occurring during or immediately after defaecation [1]. Kapoor et al. [2] reported a series of 433 syncopal patients of whom 20 had defaecation syncope. This series with defaecation syncope had a high mortality and no specic cause could be found in over half. Though the patients in Kapoor et al.s series did not have autonomic function tests, it has been suggested that a Valsalva manoeuvre may be performed during defaecation leading to decreased venous return and subsequent collapse. Alternatively, vagal stimuli during defaecation may be important in causing hypotension and/or bradycardia and syncope [3]. Ewing and Clarke [4] originally described the battery of autonomic function tests that are in routine clinical use in many units. Low [5,6] has more recently developed more sensitive, specic, reproducible non-invasive techniques for the assessment of autonomic function (Table 1). Lows criteria incorporate a composite score (maximum score 10) that gives an overall indication of the severity of autonomic abnormalities [6]. The three components that make up the composite score are, the Adrenergic index (AI, an indicator of sympathetic

Methods
Symptoms of syncope and/or presyncope on defaecation were actively sought from all patients attending the Newcastle Cardiovascular Investigation Unit. This is a unit that specializes in the investigation of syncope and/or presyncope and receives secondary and tertiary referrals from throughout the North East of England. Subjects were recruited prospectively between April and September 2000; during that period we reviewed a total of 200 new patients with syncope. All subjects followed our units standardised investigation protocol which adheres to the European Society of Cardiology Syncope guidelines [8].
Downloaded from http://europace.oxfordjournals.org/ by guest on October 1, 2012

Table 1

Lows battery of autonomic function tests [5,6]

Parasympathetic innervation  HR response to deep breathing  Valsalva ratio  HR response to standing (30:15 ratio) Adrenergic  Beat to beat BP responses to  Valsalva  Isometric exercise  Active standing  HR response to active standing

Composite score 1e3 4e6 7e10

Degree of autonomic failure Mild Moderate Severe

Adrenergic index 1 Z Phase IIe decrease of !40 but O20 mmHg mean BP or phase IIi does not return to baseline or decrease in pulse pressure to %50% of baseline 2 Z Phase IIe decrease of !40 but O20 mmHg mean BP C phase IIi or IV absent 3 Z Phase IIe decrease of O40 mmHg C absent phases IIi and IV 4 Z Criteria for 3 C orthostatic hypotension Cardiovascular heart rate index 1 Z HR(db) or VR mildly decreased (above 50% of minimum) 2 Z HR(db) or VR decreased to !50% of minimum 3 Z Both HR(db) and VR decreased to !50% of minimum
HR(db), heart rate response to deep breathing, phases refer to phases of the Valsalva; VR, Valsalva ratio; BP, blood pressure.

194 All patients who described defaecation syncope were assessed in the morning after a light breakfast and were asked to refrain from tea, coffee and other caffeinated drinks. All medications that might interfere were discontinued for O3 half lives prior to assessments. Patients rested supine for at least 10 min before performing autonomic function tests. During the rest period ReR intervals were recorded via lead 1, and sampled at a speed of 1000 Hz. Blood pressure was monitored by digital plethysmography using Finapres (Ohmeda, Englewood, USA). Respiration was recorded with a stretch sensitive band around the chest at the level of the xiphisternum. All data were recorded using a desktop PC with facility for ofine analysis and editing of non-sinus beats by inspection of a tachogram to identify outlying ReR intervals. Mean and standard deviation of baseline ReR interval and blood pressure whilst supine were calculated from a period of 300 beats (approximately 5 min) following editing of non-sinus beats, as a measure of heart rate variability. Patients then undertook a battery of manoeuvres including isometric exercise (3 min active sit: sit with legs outstretched unsupported for 3 min), active stand, three Valsalva manoeuvres and a cycle of deep breathing at a rate of 0.1 Hz. The Valsalva manoeuvre was carried out using a peak ow meter mouthpiece attached to a transducer, which fed into the PC. Patients were asked to maintain a pressure of 40 mmHg for 15 s. Visual feedback was given via a servo-meter attached to the transducer. A small air leak in the tubing ensured that the patient had to expire with an open glottis and was not able to use mouth pressure to maintain the position of the servo-meter needle. Age and sex matched healthy controls were recruited via poster advertising. Control subjects completed the same protocol as the patients. AI and CVHRI were calculated to give an indication of the relative contributions of parasympathetic and sympathetic dysfunction in each individual. The equipment to test sudomotor function was not available at the time of this study.

L. Allan et al.

Results
Seven subjects were prospectively identied attending our unit who complained of symptoms of syncope or presyncope on defaecation. All subjects described multiple episodes of syncope and presyncope at, or around, the time of defaecation. Three patients (43%) had been referred for investigation by gastroenterologists, 1 by a cardiologist and 3 by other medical specialties. Four patients (57%) were referred directly with symptoms of syncope or presyncope on defaecation. The remaining 3 were referred for investigation of syncope but did not voluntarily report defaecation symptoms. Twelve lead electrocardiograms were normal in all subjects; in those aged over 50 carotid sinus massage was carried out and was normal. Head up tilt tests were carried out in 6 of the 7 subjects (the nal subject was physically unable to stand for the period of the test) and these were all normal. Autonomic function tests were carried out on all subjects with defaecation syncope or presyncope. Clinical details and outcome in each subject are shown in Table 2. Autonomic function tests were also carried out on 7 age and sex matched controls.

Downloaded from http://europace.oxfordjournals.org/ by guest on October 1, 2012

Autonomic function tests examining sympathetic function

The most striking abnormalities in haemodynamic responses to the battery of autonomic function tests between subjects and controls were seen with the Valsalva manoeuvre. Change in blood pressure in all phases that have been associated with progressive autonomic failure [7] i.e. phase IIe, phase III and phase IV was all signicantly lower in subjects compared with controls (Table 3, p!0:001, p!0:001, p!0:001, respectively). Time to peak value of phase IV was also signicantly longer in the patients (Z 3:3, p 0:001). A representative Valsalva from a subject with defaecation syncope (subject 4) is shown in Fig. 1. Isometric exercise was performed as an active sit (see Methods). The change in systolic blood pressure after 3 min was signicantly lower in the patient group (Z 3:297, p 0:028) compared with controls. The patient group also demonstrated a signicantly greater fall in systolic blood pressure within 3 min of standing (Z 2:417, p 0:016). These ndings suggest that subjects with defaecation syncope have abnormalities of the

Statistics
Clinical characteristics are expressed as mean values and standard deviation. t Tests were used to investigate differences between groups. Autonomic indices were not normally distributed and results are therefore expressed as median (range). Non-parametric tests of signicance have been used (Wilcoxon signed ranks test). A signicant value has been taken when p!0:05.

Autonomic dysfunction in defaecation syncope

Table 2 Id 1 2 3 Age 48 81 54 Case histories of subjects with defaecation syncope/presyncope (ND, not done; M, male; F, female) Sex F M F Signicant comorbidity Nil Parkinsons Treated Arnolde Chiari malformation and syringomyelia NIDDM Nil Nil Primary biliary cirrhosis Culprit medications Atenolol None None Colonoscopy ND ND Cardioinhibitory response on insufation ND ND ND Hypotensive response on insufation with BP fall from 170 to 84 systolic with reproduction of presyncopal symptoms Outcome

195

4 5 6 7

32 59 66 76

F M F F

None None Atenolol Bendrouazide

Symptoms stopped when atenolol discontinued Symptoms controlled on udrocortisone and midodrine Symptoms stopped after insertion of permanent pacemaker Symptoms controlled on udrocortisone Symptoms controlled on udrocortisone Symptoms stopped after stopping atenolol Symptoms stopped after stopping bendrouazide

Downloaded from http://europace.oxfordjournals.org/ by guest on October 1, 2012

sympathetic function consistent with mild to moderate sympathetic failure.

Autonomic function tests examining parasympathetic function

Parasympathetic function indicated that subjects also had associated abnormalities of parasympathetic function. The heart rate response to standing

was signicantly lower in the patient group (Z 3:181, p 0:001). The mean change in heart rate during a 1 min cycle of deep breathing at 0.1 Hz was signicantly lower in the patient group (Z 2:934, p 0:003). The Valsalva ratio (ratio of the longest ReR interval shortly after the Valsalva to the shortest ReR interval during the Valsalva manoeuvre) was also signicantly lower in the patient group (Z 3:29, p 0:001).

Table 3 Blood pressure responses to autonomic function tests assessed using continuous measurements of blood pressure (ns, not signicant) Controls n Mean age (95% CI) Resting systolic blood pressure mean (95% CI) Resting diastolic blood pressure mean (95% CI) Active stand median change in systolic BP (range) Active stand median change in diastolic BP (range) Median 30:15 ratio (range) Valsalvadphase IIe median change in SBP (range) Valsalvadphase III median change in SBP (range) Valsalvadphase IV median change in SBP (range) Median Valsalva ratio (range) Median time to overshoot (range) 7 60 (50e69) 144 (117e170) 83 (65e100) 16 (37 to C23) 12 (87 to C11) 1.27 (1.06e1.48) 24 (4 to 37) 27 (58 to C6) 45 (23e67) 1.73 (1.49e2.07) 23 (20e30) Subjects 7 62 (50e73) 128 (106e150) 59 (38e80) 30 (85 to C23) 16 (31 to C7) 1.07 (1.01e1.20) 48 (6 to 83) 63 (94 to 13) 33 (12e56) 1.39 (1.05e2.16) 25 (20e96) Signicance, p ns ns ns 0.016 ns !0.05 !0.001 !0.002 !0.001 !0.001 !0.001

196

L. Allan et al.

Downloaded from http://europace.oxfordjournals.org/ by guest on October 1, 2012

Figure 1 (a) Valsalva manoeuvre in a normal control subject assessed using continuous beat to beat recording of blood pressure. Phase I: forced expiration against a closed glottis leads to an increase in intrathoracic and intraabdominal pressuredcausing a brief rise in arterial pressure and heart rate. Phase IIe: rise in intrathoracic and intraabdominal pressure prevents venous return to the heart and blood pressure falls. Phase IIi: a rise in blood pressure to baseline levels before the end of expiration due to changes in sympathetic tone in the splanchnic vasculature which causes a surge of blood from the encapsulated reservoirs of the spleen and liver. Phase III: at the end of expiration a second fall in blood pressure is seen due to a fall in intrathoracic and abdominal pressure. Phase IV: this is rapidly corrected by the continued increase in sympathetic tone. Blood pressure continues to rise, resulting in an overshoot above the baseline in healthy subjects. (b) Valsalva manoeuvre in a subject with defaecation syncope illustrating defects in sympathetic function.

Relative contributions of sympathetic and parasympathetic autonomic abnormalities

Adrenergic (AI) and Cardiovascular heart rate index (CVHRI) were calculated for each subject [6]

(Table 1) and compared with the abnormalities that would have been detected using the conventional Ewing battery of tests (Table 4). The predominant autonomic abnormality in those with defaecation syncope was of sympathetic function.

Autonomic dysfunction in defaecation syncope

197

Table 4 Comparing Lows criteria [5,6]; Adrenergic index (AI) and Cardiovascular heart rate index (CVHRI) with the number of single abnormal tests in the standard Ewing battery [4] (O3 abnormal tests are regarded as clinically relevant) Subjects AI 1 2 3 4 5 6 7 Median 3 2 0 4 1 2 2 2 CVHRI 0 0 1 1 1 1 1 1 Ewing battery 1 0 1 0 3 4 2 1 Controls AI 0 0 1 0 1 1 1 1 CVHRI 0 0 0 0 0 0 0 0 Ewing battery 1 1 0 0 0 1 1 1

Outcome after intervention in subjects with defaecation syncope/presyncope

Two subjects had colonoscopy in view of an associated change in bowel habit. Both of these subjects were monitored during the procedure using continuous beat to beat assessments of heart rate and blood pressure and had profound haemodynamic responses to insufation. In all subjects defaecation symptoms improved on specic therapeutic intervention. Three of the seven subjects (43%) improved on discontinuation of culprit medication (see Table 2). A similar number (43%) improved with medication aimed at reducing hypotensive episodes by raising blood pressure i.e. udrocortisone or midodrine. In the nal patient, in whom cardioinhibition was demonstrated during insufation at colonoscopy symptoms disappeared after insertion of a permanent pacemaker.

Discussion
During the six months when symptoms on defaecation were actively sought from patients attending our unit, 7 individuals with defaecation syncope or presyncope were identied. Considering the total number of patients with syncope seen in our unit during that time our prevalence is comparable with that found by Kapoor et al. [2]. Although this is a small and heterogenous group, compared with age and sex matched controls, subjects with symptoms of syncope or presyncope on defaecation appear consistently to have abnormalities of autonomic nervous function compatible with an associated early autonomic failure. These abnormalities appear to affect both the parasympathetic and sympathetic nervous systems but most

predominately sympathetic function and are consistent with mild to moderate autonomic failure. In this current study we were unable to derive a sudomotor score; however, despite this, all of our subjects had mild to moderate abnormalities suggesting that if combined with a sudomotor index the composite autonomic score would imply moderate to severe autonomic failure in those with defaecation syncope. The patients included in this study suggest that the underlying aetiology of syncopal symptoms of defaecation may be related either to central abnormalities of autonomic nervous system modulation, or to local abnormalities of autonomic innervation of capacitance vessels. The splanchnic vasculature has a large capacitance and hence small abnormalities in innervation may lead to large changes in total circulating blood volume. This may result in hypotensive episodes potentially leading to reduced cerebral perfusion and subsequent syncope. Such episodes may be aggravated by vasoactive medication; however, in the patients studied, medications were discontinued prior to autonomic function testing and therefore the data are not contaminated by culprit medication. It is interesting to speculate how disturbances of autonomic function which are presumably static relate to a situational syncope. It is important to acknowledge that although in this cohort we have demonstrated both parasympathetic and sympathetic abnormalities of the extrinsic autonomic nervous system we have not examined the intrinsic/enteric nervous system. The abnormalities that we have found in the phases of the Valsalva in our patients with defaecation syncope raise questions as to the role of the encapsulated intra-abdominal organs in maintenance of blood pressure. It could be hypothesized that the cardiovascular reexes used to dene autonomic abnormalities are, in fact, in this condition

Downloaded from http://europace.oxfordjournals.org/ by guest on October 1, 2012

198 detecting specic abnormalities of innervation or blood supply to the liver and spleen in humans. Reassuringly, with specic interventions, such as withdrawal of culprit medication, medication to maintain blood pressure or in the most extreme case permanent pacemaker insertion, in all cases, symptoms were obliterated or improved, suggesting that this is a symptom that is worth ascertaining, investigating and treating. Further detailed assessment of the precise physiological mechanisms underlying this distressing condition and longer term outcome are required.

L. Allan et al.

References
[1] Pathy MS. Defaecation syncope. Age Ageing 1978;7:233e6. [2] Kapoor WN, Peterson J, Karpf M. Defaecation syncope. A symptom with multiple etiologies. Arch Intern Med 1986; 146:2377e9. [3] Brophy CM, Evans L, Sumpio BE. Defaecation syncope secondary to functional inferior caval obstruction during a Valsalva maneuver. Ann Vasc Surg 1993;7:374e7. [4] Ewing DJ, Clarke BF. Diagnosis and management of autonomic neuropathy. Br Med J 1982;285:916e8. [5] Low P. Autonomic nervous system function. J Clin Neurophysiol 1993;10:14e27. [6] Low P. Composite autonomic scoring scale for laboratory quantication of generalised autonomic failure. Mayo Clin Proc 1993;68:748e52. [7] Sandroni P, Benarroch EE, Low PA. Pharmacologic dissection of components of the Valsalva maneuver in adrenergic failure. J Appl Physiol 1991;71:1563e7. [8] Brignole M, Alboni P, Benditt D, et al. Task force on syncope: parts 1 and 2. Europace 2001;3:253e68.

Acknowledgements
JN thanks the Digestive Disorders Foundation for support.

Downloaded from http://europace.oxfordjournals.org/ by guest on October 1, 2012