TSR 13:2, Spring 2006

Contents
vii Foreword

viii Information for Authors

x Letters to the Editors

1 Ottawa Panel Evidence-Based Clinical Practice Guidelines for Post-Stroke

Rehabilitation
3 Method
4 Target population
4 Literature search
5 Study inclusion/exclusion criteria
5 Post-stroke rehabilitation interventions
7 Outcomes
7 Statistical analysis
9 Reviewing the guidelines
9 Results of literature search

10 Results
10 Clinical practice guidelines for therapeutic exercises
14 Summary of trials
14 Efficacy
20 Strength of published evidence compared with other guidelines
20 Clinical recommendations compared with other guidelines

21 Clinical practice guidelines for task-oriented training

24 Summary of trials
24 Efficacy
27 Strength of the published evidence compared with other guidelines
27 Clinical recommendations compared with other guidelines

28 Clinical practice guidelines for biofeedback

30 Summary of trials
30 Efficacy
32 Strength of published evidence compared with other guidelines
33 Clinical recommendations compared with other guidelines

34 Clinical practice guidelines for gait training

36 Summary of trials
37 Efficacy
40 Strength of published evidence compared to other guidelines
40 Clinical recommendations compared with other guidelines

41 Clinical practice guidelines for balance training

43 Summary of trials
43 Efficacy
45 Strength of the published evidence compared with other guidelines
45 Clinical recommendations compared with other guidelines

THOMAS LAND PUBLISHERS, INC.

 45 Clinical practice guidelines for sensory interventions
46 Summary of trials
46 Efficacy
48 Strength of the published evidence compared with other guidelines
48 Clinical recommendations compared with other guidelines

48 Clinical practice guidelines for constraint-induced movement therapy

49 Summary of trials
49 Efficacy
50 Strength of published evidence compared with other guidelines
50 Clinical recommendations compared with other guidelines

51 Clinical practice guidelines for shoulder subluxation

52 Summary of trials
52 Efficacy
53 Strength of published evidence compared with other guidelines
54 Clinical recommendations compared with other guidelines

54 Clinical practice guidelines for electrical stimulation

56 Summary of trials
57 Efficacy
59 Strength of published evidence compared with other guidelines
60 Clinical recommendations compared with other guidelines

61 Clinical practice guidelines for transcutaneous electrical nerve stimulation (TENS)

61 Summary of trials
62 Efficacy
63 Strength of published evidence compared with other guidelines
63 Clinical recommendations compared with other guidelines

64 Clinical practice guidelines for therapeutic ultrasound

64 Summary of trials
64 Efficacy
64 Strength of the published evidence compared with other guidelines
64 Clinical recommendations compared with other guidelines

65 Clinical practice guidelines for acupuncture

65 Summary of trials
66 Efficacy
67 Strength of the published evidence compared with other guidelines
67 Clinical recommendations compared with other guidelines

68 Clinical practice guidelines for intensity and organization of rehabilitation

74 Summary of trials
75 Efficacy
83 Strength of published evidence compared with other guidelines
83 Clinical recommendations compared with other guidelines

84 Discussion
86 Therapeutic exercises
87 Task-oriented training
88 Biofeedback
88 Gait training
91 Balance training
91 Sensory interventions

v
 91 Constraint-induced movement therapy
92 Shoulder subluxation
93 Electrical stimulation
94 TENS
94 Therapeutic ultrasound
95 Acupuncture
95 Intensity and organization of rehabilitation

96 Conclusion

97 Acknowledgments

97 References

119 Appendixes

All figures and tables cited in this issue are included in the CD attached to the inside
back cover. This CD also includes an electronic version of the full text of this issue
with embedded hyperlinks to the figures and tables.
Additional copies can be ordered at www.thomasland.com

vi
 Foreword
The concept of developing and implementing clinical
practice guidelines has received considerable attention
in recent years among many medical professionals, clini-
cal leaders, payers, and policy makers. It is interesting to
note that adoption of these tools has been considerably
less enthusiastic among rehabilitation programs and
practitioners than among many other disciplines and
professionals. Reasons for the slower utilization of guide-
lines are many, including concern that they might limit
the important role of creativity and ingenuity in the
clinical practice of rehabilitation, relative deficiency of
sufficient evidence supporting many prevailing rehabili-
tation practices, and an apparent lack of compelling rea-
sons for clinicians to examine the possibility of changing
existing practices. There also is limited interest in the
development and use of guidelines among rehabilita-
tion professionals and a general lack of awareness of
their usefulness among clinicians and their leaders. Al-
though there are some notable prominent exceptions,
clinical practice guidelines are in their infancy in rehabili-
tation.
There is a school of thought that asserts that the expe-
rience of developing these guidelines is as important
and beneficial as the actual implementation of these
tools. The opportunity to thoroughly and critically re-
view existing medical literature and assess its implica-
tions for clinical practice often is a unique and favorable
collaborative experience for clinicians and scholars.
However, it is in their roles in education and in pro-
moting both quality and consistency of care that clinical
practice guidelines demonstrate their greatest benefits.
For trainees and for junior clinicians with minimal practi-
cal experience, these tools can be used as an effective
training device. For all practitioners, they can be used as
a template to ensure that favorable evidence-driven
practices are being implemented. This is a method to
improve quality of care. These documents are tools and,
like other instruments used by clinicians, they serve to
support and facilitate successful rehabilitation; they do
not replace the role of originality and they do not reduce
the value of hands-on, interactive, collaborative prob-
lem-solving by clinicians and patients that often is the
hallmark of team-driven clinical stroke rehabilitation.
Topics in Stroke Rehabilitation recently has published
several Clinical Practice Guidelines (TSR 10:1 and 10:2
[2003], 11:4 [2004]). The frequency with which these
documents are being published at this time reflects the
emerging interest among thought leaders and practitio-
vii
ners in the performance and dissemination of compre-
hensive reviews of literature and in the implementation
of clinical practices that are based on strong evidence,
where it exists.
In the present issue, Dr. Lucie Brosseau and her col-
leagues in Ottawa, Ontario, Canada, have conducted an
exhaustive review of the literature, employed rigorous
grading measures to rate the quality of the research
studies, developed and applied a systematic method for
organizing key points from the manuscripts that they
reviewed, and created an extensive set of guidelines for
clinical stroke rehabilitation practices that derive directly
from these findings. The product focuses on 147 specific
recommendations concerning 13 rehabilitation inter-
ventions. Practicing according to these guidelines can
be expected to enhance the quality of care that is pro-
vided by clinicians and to improve the level of function-
ing that is experienced by patients.
Dr. Brosseau and her team are to be congratulated for
the enormous effort that was put into the production of
this document, not only for their rigorous reviews and
clear writing, but also because of the way in which the
information is presented. Because the volume of mate-
rial is potentially overwhelming, it could be somewhat
unwieldy for most practitioners. This is one reason that
the organization of the material is so relevant in order to
facilitate understanding and use of the recommendations
and their foundation. The managing editor, Mary Killion,
also should be recognized for her extensive contribution
to the layout and formatting of the data. The use of the
CD-ROM is a novel approach. Displaying information in a
tabular format is a sound way to promote understanding
and learning. The enormity of the material necessitated
an alternative additional method to disseminate the fig-
ures and tables of the reviewed literature. Using a CD-
ROM seemed a sensible method to provide this material
to you.
Your comments on the text (and the format) are wel-
come. However, the true test of the success of the prod-
uct is whether these guidelines are actually adopted in
daily practice and the extent to which the content forms
a basis of clinical practice in stroke rehabilitation.
—Elliot J. Roth, MD
Co-Editor-in-Chief
Topics in Stroke Rehabilitation
Chicago, IL, USA
January 2006
 Special Issue
Ottawa Panel Evidence-Based Clinical
Practice Guidelines for Post-Stroke
Rehabilitation
The Ottawa Panel*

Background and Purpose: The purpose of this project was to create guidelines for 13 types of physical rehabilitation
interventions used in the management of adult patients (>18 years of age) presenting with hemiplegia or hemiparesis
following a single clinically identifiable ischemic or hemorrhagic cerebrovascular accident (CVA). Method: Using Cochrane
Collaboration methods, the Ottawa Methods Group identified and synthesized evidence from comparative controlled
trials. The group then formed an expert panel, which developed a set of criteria for grading the strength of the evidence
and the recommendation. Patient-important outcomes were determined through consensus, provided that these
outcomes were assessed with a validated and reliable scale. Results: The Ottawa Panel developed 147 positive
recommendations of clinical benefit concerning the use of different types of physical rehabilitation interventions involved in
post-stroke rehabilitation. Discussion and Conclusion: The Ottawa Panel recommends the use of therapeutic exercise,
task-oriented training, biofeedback, gait training, balance training, constraint-induced movement therapy, treatment of
shoulder subluxation, electrical stimulation, transcutaneous electrical nerve stimulation, therapeutic ultrasound,
acupuncture, and intensity and organization of rehabilitation in the management of post stroke. Key words: clinical practice
guidelines, CVA, epidemiology, evidence-based practice, outcomes, physical rehabilitation, stroke

S
troke is the third cause of mortality in North
America.1 Although approximately two
thirds of stroke patients survive an initial
stroke, nearly one half of survivors have physical
disabilities as a result.2 Furthermore, while severe
stroke incidence has decreased, milder stroke
incidence with minimal and moderate deficits has
increased. The individuals surviving a stroke
require rehabilitation that includes varying degrees
of medical care, rehabilitation, nursing, and other
health professional care.3 Stroke survivors present
sensorimotor, musculoskeletal, perceptual, and
cognitive system deficits.4 Their impairments,
disabilities, and handicaps can lead to devastating
personal consequences as well as consequences for
the health care system and society at large.

*Ottawa EBCPGs Development Group: Lucie Brosseau, PhD,1
George A. Wells, PhD,2,3 Hillel M. Finestone, MD,6 Mary Egan,
PhD,1 Claire-Jehanne Dubouloz, PhD,1 Ian Graham, PhD,4 Lynn
Casimiro, MA, Vivian A. Robinson, MSc,3 Martin Bilodeau,
PhD,1 and Jessie McGowan, MLIS.3
External Panel Members: Robert Teasell, MD,10 Johanne
Desrosiers, PhD,5 Susan Barreca, MSc,8 Lucie Laferrière, MHA,9
Joyce Fung, PhD,7 Hélène Corriveau, PhD, MHA,5 Gordon
Gubitz, MD,11 Michael Sharma, MD,9 and Mr. S. U.12
Assistant Manuscript Writers: Amole Khadilkar, MD,1 Karin
Phillips, MA,1 Nathalie Jean,1 Catherine Lamothe,1 Sarah Milne,
MSc,1 and Joanna Sarnecka, MSc.1
sity of Ottawa, Ottawa, Ontario, Canada; 3Centre for Global
Health, Institute of Population Health, Ottawa, Ontario,
Canada; 4School of Nursing Sciences, Faculty of Health Sci-
ences, University of Ottawa, Ottawa, Ontario, Canada; 5Re-
search Centre on Aging and Sherbrooke University,
Sherbrooke, Québec, Canada; 6Sisters of Charity of Ottawa
Health Service, Ottawa, Ontario, Canada; 7Department of
Physical Therapy, McGill University, Montreal, Québec,
Canada; 8Hamilton Health Sciences, Hamilton, Ontario,
Canada; 9Regional Stroke Centre, Ottawa Hospital, Ottawa
(Ontario), Canada; 10University of Western Ontario, London,
Ontario, Canada; 11Division of Neurology, Dalhousie University,
Halifax (Nova Scotia), Canada; 12Patient who had a stroke.

1
School of Rehabilitation Sciences, Faculty of Health Sci-
Top Stroke Rehabil 2006;13(2):1–269
ences, University of Ottawa, Ottawa, Ontario, Canada; 2De- © 2006 Thomas Land Publishers, Inc.
partment of Epidemiology and Community Medicine, Univer- www.thomasland.com

1
2 TOPICS IN STROKE REHABILITATION/SPRING 2006
Post-stroke physical rehabilitation interventions
have been used to reduce pain and spasticity, as
well as to increase range of motion (ROM), muscle
force, mobility, walking ability, functional status,
physical fitness, and quality of life. Post-stroke
physical rehabilitation interventions are mostly
noninvasive interventions that present very few
adverse side effects and contraindications as
compared with a large number of pharmacologic
interventions.
Despite the fact that significant progress has
been made in the clinical management of stroke
over the last decade, there is an urgent need for
physicians, nurses, physiotherapists, occupational
therapists, and other rehabilitation specialists to
provide the most efficient and effective treatments
for their patients.
Evidence-based clinical practice guidelines
(EBCPGs) have been defined as systematically de-
veloped statements to help practitioners and pa-
tients with decisions about appropriate health care
for specific clinical circumstances.5 EBCPGs are a
rapidly emerging technology with considerable
potential to alter the clinical decision-making pro-
cess in fundamental ways. The appropriate use of
guidelines has been demonstrated to improve both
the process of care and patient health outcomes.6
EBCPGs allow stroke patients to benefit maximally
from the physical rehabilitation treatment that
they are receiving.
There are currently many systematic reviews
and meta-analyses on the effectiveness of post-
stroke physical rehabilitation interventions in the
scientific literature. (Summarized comparative re-
sults of these reviews are included in the Discus-
sion section.) Trials on the efficacy of the following
types of therapeutic exercises for stroke survivors
have been systematically reviewed: physical fit-
ness,7,8 therapeutic exercise,9–14 task-oriented
training,15 progressive strengthening exercise,16,17
robot-aided training,18,19 and constraint-induced
movement therapy.20–23 Four meta-analyses have
been published24–27 for the effect of the intensity of
rehabilitation following stroke, while reviews on
different aspects of gait training,28 such as the use
of the treadmill combined with body support,29–35
have also been done. Both Barclay-Goddard35 and
Pollock36 have systematically reviewed balance
training. Furthermore, the efficacy of organization
and intensity of stroke rehabilitation has been ex-
amined through systematic reviews.37–44
Several meta-analyses, systematic reviews, and
literature reviews have been conducted over the
last 7 years on the effectiveness of EMG biofeed-
back (EMG-BFB). One examined EMG-BFB for
neuromuscular reeducation,45 and others looked
at EMG-BFB for the improvement of upper ex-
tremity function.46–48 Finally, Moreland et al.49 ex-
amined the use of EMG-BFB to improve lower
extremity function after stroke. These publica-
tions, though recent, require updating because of
the rapidly growing number of scientific articles
published on the effectiveness of EMG-BFB.50,51
Moreover, one of the studies45 has been criticized
for failing to perform a sensitivity analysis on the
control group.49 Follow-up of drop-outs was also
lacking.49 EMG-BFB constitutes a small but effec-
tive part of lower extremity physical rehabilitation
in stroke patients.
Functional electrical stimulation (FES) has also
been systematically reviewed for post-stroke pa-
tients.52–60 A recent review was conducted on the
management of shoulder pain and subluxation.61–
63
Meta-analyses on post-stroke pain management
were done for acupuncture as an adjuvant therapy
in stroke rehabilitation.64–67 However, several trials
of electroanalgesia could have been added to this
review.68,69 To our knowledge, no published sys-
tematic review exists concerning the efficacy of
therapeutic electrotherapy modalities.
It is evident that the post-stroke physical reha-
bilitation literature has been exhaustively re-
viewed. However, the methodology used in these
reviews needs to be standardized (e.g., selection
criteria) and quantified (e.g., Cochrane Collabora-
tion), and the results of most of these reviews also
need to be updated in order to be included in the
development of EBCPGs.
Several multidisciplinary EBCPGs have been
published on post-stroke rehabilitation.70–78 How-
ever, the Agency for Health Care Policy and
Research’s (AHCPR’s) EBCPGs were developed for
limited clinical practice areas.70 They did not pro-
vide a clear definition of physiotherapy or review
specific physical rehabilitation interventions. They
also failed to use a rigorous grading system to

assess the evidence. Guidelines were based mainly
on committee opinions and have not been recently
updated.79 Other EBCPGs72–78 are also available for
rehabilitation specialists. Although the review of
the literature and selection of stroke topics is ex-
haustive in these guidelines, the evaluation of the
evidence is based upon descriptive conclusions of
the primary studies rather than a quantitative
analysis of the raw data. These guidelines use a
grading system that takes the research design of
the studies into account, but not the clinical sig-
nificance of the outcomes. Except for the guide-
lines from the Heart and Stroke Foundation of
Ontario,71 these EBCPGs do not base their assess-
ment of the level of evidence on a quantitative
synthesis using the raw data of the studies of inter-
est, such as proposed by the Cochrane Collabora-
tion methodology. These analyses are also not
pooled to specific outcome measures. The conclu-
sions of most of these EBCPGs concerning the
effectiveness of the selected post-stroke interven-
tions are often imprecise and difficult to apply to
the daily practice of rehabilitation practitioners.
The advantage of the proposed Ottawa Panel
EBCPGs80–82 on post-stroke physical rehabilitation
interventions is that they offer graded, quantita-
tive,83 and high-quality84 recommendations that
indicate the treatment time for which a specific
intervention is optimally effective for a specific
outcome for a particular stroke population.
The generally positive (small-to-large effect sizes
from quantitative reviews) results from the recent
meta-analyses coupled with the lack of up-to-date,
rigorously developed EBCPGs on physical post-
stroke rehabilitation interventions suggest the
need for the development of better quality EBCPGs
for these interventions. Furthermore, evidence
suggests that quality of care can be improved
through the use of EBCPGs.85–89 The purpose of
developing these guidelines is to promote the ap-
propriate use of various physical rehabilitation in-
terventions in the management of stroke survivors.
These guidelines are aimed at various users, in-
cluding physical therapists, occupational thera-
pists, physicians, and patients. This article dis-
cusses only post-stroke physical interventions
such as therapeutic exercises, task-oriented train-
ing, biofeedback, gait training, balance training,
Clinical Practice Guidelines 3
sensory interventions, constraint-induced move-
ment therapy (CIMT), treatment of shoulder sub-
luxation, electrical stimulation, transcutaneous
electrical nerve stimulation (TENS), therapeutic
ultrasound, acupuncture, and intensity and orga-
nization of rehabilitation.
Method
The development process of these EBCPGs was
similar to that of the Philadelphia Panel90 and to
previous Ottawa Panel publications,81,82 except
that a different target population was used. Briefly,
the Ottawa Methods Group (OMG), a group of 10
methodologists with experience in developing
EBCPGs, asked professional associations inter-
ested in the care of stroke patients to suggest indi-
viduals with both clinical expertise in the manage-
ment of stroke and familiarity with EBCPGs. From
among the suggestions, the OMG chose nine ex-
perts (R.T., G.G., J.D., J.F., H.C., S.B., L.L., M.S.,
S.U.) to serve as panel members. The professional
experts were recruited from multidisciplinary dis-
ciplines such as physical medicine, neurology, oc-
cupational therapy, and physical therapy. Several
experts (R.T., G.G., J.D., J.F., S.B., L.B., A.H.) are
members of the Canadian Stroke Network,91 while
some had already developed post-stroke rehabili-
tation EBCPGs (R.T., S.B.). The Ottawa Panel con-
sisted of these nine experts, in addition to all the
members of the OMG.
The OMG assembled a research and support
staff with expertise in meta-analyses, stroke reha-
bilitation interventions, research methods, or the
development and assessment of EBCPGs. The
OMG then established an a priori set of inclusion
criteria for the study designs, subject samples, in-
terventions, and outcomes to allow the research
staff to select the most relevant material as evi-
dence for the effectiveness of various rehabilitation
interventions for post-stroke patients. The OMG
also reviewed the inclusion criteria to ensure that
the approach to the study selection was reproduc-
ible and systematic. This a priori protocol guided
separate systematic reviews of the literature for
each intervention. The OMG also made sure that
the Ottawa Panel EBCPGs were methodologically
4 TOPICS IN STROKE REHABILITATION/SPRING 2006
developed at high level of quality, according to
AGREE (www.agreecollaboration.org) criteria.84
The research staff reviewed articles and created
draft evidence tables, which the nine clinical experts
received in preparation for their consensus meeting
with the OMG. These tables were used as the basis
for making the Ottawa Panel recommendations.
Target population
The target population was adult patients (>18
years of age) presenting with hemiplegia or hemi-
paresis following a single clinically identifiable is-
chemic or hemorrhagic CVA. The patients had to
be medically stable and able to follow simple in-
structions and to interpret and respond to feed-
back signals. The mean duration since stroke onset
varied from hyper-acute (the first 12 hours), acute
(first week following a stroke), subacute (from the
first to 6th week), and post-acute (from 6 weeks to
6 months) to chronic (from 6 months) as defined
by the Canadian Stroke Network (Appendix 3:
Characteristics of Included Studies).91
Patients who had been identified as having mul-
tiple CVAs, other neurological problems (e.g.,
Parkinson’s, brain tumors, traumatic brain injury),
subarachnoid hemorrhages, or subdural hemato-
mas were excluded because of the numerous and
varied associated signs and symptoms. Studies that
included patients with bilateral neurological signs
were also excluded. Further exclusion criteria in-
cluded studies whose patients presented with one
of the following conditions: (1) cancer or other
oncological conditions, (2) cardiac conditions, (3)
dermatologic conditions, (4) healthy normal sub-
jects, (5) serious cognitive deficits or severe com-
munication problems, (6) major medical problems
that could interfere with the rehabilitation pfOcess
or incapacitate functional status, or (7) psychiatric
conditions. Further inclusion and exclusion crite-
ria are exhibited in Table 1.
Literature search
The library scientist developed a structured lit-
erature search based on the sensitive search strat-
egy recommended by The Cochrane Collabora-
tion92 and modifications to that strategy proposed
by Haynes et al.93 The Cochrane Collaboration
method minimizes bias through a systematic Z -
proach to the literature search, study selection,
and data extraction and synthesis. The search was
organized around the condition and interventions
rather than the outcomes because it was an a priori
search. Thus, we had no control over the outcomes
that the authors of the primary studies decided to
measure (Appendix 1: Literature Search Results).
The library scientist expanded the search strat-
egy to identify case control, cohort, and non-ran-
domized studies and conducted the search in the
electronic databases of MEDLINE, EMBASE, Cur-
rent Contents, the Cumulative Index to Nursing
and Allied Health (CINAHL), and the Cochrane
Controlled Trials Register up to December 2004.
She also searched the registries of the Cochrane
Field of Rehabilitation and Related Therapies, the
Cochrane Musculoskeletal Group, the Physio-
therapy Evidence Database (PEDro), and the Uni-
versity of Ottawa EBCPGs Web site. Finally, she
searched the reference lists of all of the included
trials for relevant studies and contacted content
experts for additional studies.
In the first round of study inclusion or exclu-
sion, two trained independent reviewers appraised
the titles and abstracts of the literature search,
using a checklist with the a priori defined selection
criteria (Table 1). For each pair of reviewers, indi-
viduals independently read the title and abstract of
each article and created a list of all of the articles in
the database along with a reason for either includ-
ing or excluding each article. If the reviewers were
uncertain about a particular article after having
read the abstract, they ordered the article and read
it in full before making a determination. Before
deciding whether to include or exclude the article,
a comparison of their individual lists was per-
formed. A senior reviewer, a methodologist and a
clinical expert (L.B.), checked the two indepen-
dent lists of articles and the reasons for inclusion
or exclusion to determine potential inconsisten-
cies. Seven percent of the abstracts needed the
consultation of the senior reviewer and an addi-
tional review of the problematic article. For the
second round of the inclusion and exclusion pro-
cess, the pairs of reviewers retrieved articles se-
lected for inclusion from the first round and inde-

pendently assessed the full articles for inclusion or
exclusion in the study. Using predetermined ex-
traction forms, the pairs of reviewers indepen-
dently extracted from included articles data on the
population characteristics, details of the interven-
tions, trial design, allocation concealment, and
outcomes. The pairs of reviewers assessed the
methodological quality of the studies using the
Jadad Scale,83,94 a 5-point scale with reported reli-
ability and validity that assigns 2 points each for
randomization and double blinding and 1 point
for description of withdrawals. The reviewers re-
solved differences in data extraction and quality
assessment through consensus with the senior re-
viewer. This consensus served to support the reli-
ability of data obtained with the article selection
process.
Study inclusion/exclusion criteria
The inclusion/exclusion criteria were based
upon previous criteria used by the Philadelphia
Panel.90 This list of criteria, which had been cre-
ated for multiple diagnoses, was adapted and ap-
proved by the OMG for use with patients post
stroke (Table 1).
All original comparative controlled studies that
evaluated relevant physical rehabilitation interven-
tions in stroke patients were included: randomized
controlled trials (RCTs), controlled clinical trials
(CCTs),* cohort studies, and case-control studies.
Crossover studies were included, but to avoid po-
tential confounding carry-over effects the data
from only the first part of the study (before cross-
ing) was analyzed. Studies where patients served
as their own controls were excluded. No limita-
tions based on methodological quality were im-
posed a priori with regard to the selection of com-
parative controlled studies; however, the quality of
the studies was considered when grading the rec-
ommendations resulting from our analysis.
Uncontrolled cohort studies (studies with no
comparison group) and case series were excluded,
as were eligible studies with greater than a 20%
*Controlled clinical trials are considered the same as randomized
control trials (RCTs). However, according to the Jadad Scale,94 CCTs are
either not randomized or have not been appropriately randomized.
Clinical Practice Guidelines 5
drop-out rate or a sample size of fewer than 5
patients per group. Trials published in languages
other than French and English were not analyzed,
because of the additional time and resources re-
quired for translation. Abstracts were excluded if
they contained insufficient data for analysis and
additional information could not be obtained from
the authors. For further exclusion criteria, see
Table 1.
Post-stroke rehabilitation interventions
Post-stroke rehabilitation interventions were
identified as therapeutic exercises, task-oriented
training, biofeedback, gait training, balance train-
ing, sensory information, CIMT, treatment of
shoulder subluxation, electrical stimulation,
TENS, therapeutic ultrasound, and acupuncture.
Intensity and organization of rehabilitation was
also included as an intervention related to stroke
rehabilitation.
Post-stroke rehabilitation interventions related
to therapeutic exercises were identified as aerobic
training, resistance training, passive range of mo-
tion exercises, proprioceptive neuromuscular fa-
cilitation, Bobath technique, kinetron, and the use
of the overhead pulley.95 They are defined as fol-
lows. Aerobic training is considered to be activities
to increase endurance and cardiovascular func-
tion. Resistance training was defined as active exer-
cise done against a resistance. Passive range of mo-
tion exercises were defined as physiological
mobilization done by the therapist without any
effort from the patient. Proprioceptive neuromuscu-
lar facilitation was identified as the use of mostly
reflex-inhibiting patterns. Bobath technique was de-
fined as a neurodevelopmental technique using
inhibitory posture and movement to inhibit spas-
ticity and synergies, while facilitating normal
movements. Kinetron training was defined as train-
ing with this resistive lower extremity machine,
usually in isokinetic mode.
Rehabilitation interventions related to task-ori-
ented training were identified as treatments that
involved dividing activities of daily living into
component parts. Individual components of the
larger task were then practiced until the patient
was able to complete the component adequately.
6 TOPICS IN STROKE REHABILITATION/SPRING 2006
Component parts were then combined, and the
overall skill was practiced with repetition. Any
intervention that divided required tasks into indi-
vidual skills was included.95 We included tasks
such as seated reaching tasks, adapted games, and
repetitive elbow joint movements.
Post-stroke rehabilitation interventions related
to biofeedback were identified as EMG-BFB, EMG-
biofeedback-relaxation training, rhythmic posi-
tional biofeedback, audio and visual feedback,
video feedback, and force feedback. EMG-BFB was
defined as an intervention that allows a patient to
monitor his or her muscle activity through elec-
trodes with a visual or audible feedback signal.95
Rhythmic positional biofeedback was defined as a
usual biofeedback intervention with auditory or
visual stimuli aimed at increasing the rhythm of
movement. Audio and visual feedback were identi-
fied as any cue received by the patient during or
after the exercise. Video feedback was considered to
be a visual cue from a monitor after the action was
done. Force feedback was defined as feedback re-
lated to the moment of force.
Rehabilitation interventions associated with gait
training were identified as treadmill training,
overground training, body weight support train-
ing, brace-assisted walking, electrogoniometric
feedback training, FES, rhythmic auditory facilita-
tion training, and functional lower extremity train-
ing. Treadmill training was defined as ambulation
on a treadmill adjusted to patient’s comfortable
walking speed or highest speed as possible for the
patient. Overground training was defined as gait
training on an even surface with propulsion for-
ward, backward, and sideways or going up and
downstairs. Body weight support training was de-
fined as treadmill training, while an overhead har-
ness supported a percentage of the body weight.
Brace-assisted walking was defined as use of
hemibar and ankle-foot orthosis (AFO) or any
other type of brace if necessary. Electrogoniometric
feedback training was defined as auditory feedback
during gait training when patient was compensat-
ing in hyperextension or flexion. FES was defined
as electrical stimulation of a specific muscle or
nerve such as tibialis anterior or peroneal nerve for
a functional purpose – to improve swing phase or
stance phase during gait. Rhythmic auditory facilita-
tion training was defined as imposed rhythm to
improve gait rhythm and frequency. Functional
lower extremity training was defined as functional
tasks such as sitting, standing, climbing stairs,
transfers, and gait with a focus on the recovery of
stability and gait performance.
Rehabilitation interventions related to balance
training were identified as any intervention that
contributes to the enhancement of equilibrium
and balance in post-stroke patients. We included
interventions such as base of support training and
platform training.
Rehabilitation interventions related to sensory
interventions were identified as any retraining of
the sensory and visuo-spatial function to correct
posture and perceptual problems after stroke,95
such as passive vestibular stimulation, perceptual
learning exercises, and rocking chair stimulation.
Rehabilitation interventions associated with
CIMT were defined as the restriction of the
nonparetic upper extremity by a sling or hand
splint to encourage the use of the paretic limb.
Functional exercises were given to the patient to
improve the function of the affected arm.
Post-stroke rehabilitation interventions related
to treatment of shoulder subluxation were identified
as FES, supports methods, strapping, and shoul-
der positioning. FES was defined as electrical
stimulation of a specific muscle or nerve such as
supraspinatus or middle deltoid with functional
purpose resulting in the reduction of shoulder
subluxation. Support methods were defined as any
use of an external support such as orthosis or sling
to prevent shoulder subluxation. Strapping methods
were defined as strapping used to keep the gleno-
humeral joint in normal position. Shoulder position-
ing was defined as a position induced by the phys-
iotherapist to protect the structures around the
weak hemiplegic shoulder in order to avoid shoul-
der pain and shoulder subluxation.
Post-stroke rehabilitation interventions related
to electrical stimulation were identified as FES,
neuromuscular electrical stimulation (NMES),
positional feedback stimulation training, EMG-
triggered electrical muscle stimulation, and
TENS. FES was defined as the electrical stimula-
tion of a specific muscle or nerve such as tibialis
anterior and gastrocnemius for a functional pur-
pose, such as gait training. Neuromuscular electri-
cal stimulation (NMES) was defined as electrical

stimulation of a specific muscle or nerve such as
thumb flexors and extensors to help trigger ner-
vous fibers and achieve motor recovery. Positional
feedback stimulation training was defined as audi-
tory and visual feedback during training to
achieve a target position of the joint. EMG-trig-
gered electrical muscle stimulation was defined as
electrical stimulation of a muscle triggered by
EMG activity of this muscle.
TENS was defined as a form of electrical stimula-
tion that triggers nervous endings to inhibit the
message of pain. TENS is identified as being given
at high- and low-intensity levels.
Ultrasound was identified as an electrophysical
modality using an ultrasonic wave to treat a spe-
cific area, usually for pain, and indirectly for ROM.
Post-stroke rehabilitation interventions related
to acupuncture were identified as any treatment
using needles to stimulate specific anatomical
point with the hands or with electrical stimulation.
Post-stroke rehabilitation interventions related
to the intensity and organization of rehabilitation
were identified as “examining the rate, frequency
and rigor of any physical rehabilitation interven-
tion or combination of interventions in the treat-
ment of post-stroke patients.” 95 Interventions such
as stroke unit care, enhanced physical therapy,
enhanced occupational therapy, enhanced upper
extremity treatment, and intensive outpatient
physiotherapy rehabilitation were included as part
of intensity and organization of post-stroke reha-
bilitation.
Acceptable comparisons were placebo, no treat-
ment, or use of educational pamphlets. Studies
designed with a comparison of two interventions
instead of treatment versus control were included
as long as both interventions respected the inclu-
sion criteria. Concurrent therapies (such as medi-
cation) were accepted only if they were provided
to both the experimental and control groups.
Study selection was not restricted by the cost,
complexity, or general availability of equipment
and resources required to carry out the interven-
tions under investigation.
Outcomes
The outcomes were selected and based upon the
World Health Organization’s96 (WHO’s) new pro-
Clinical Practice Guidelines 7
posal of the International Classification of Function-
ing, Disability and Health, which involved the con-
cepts of body function, body structure, activities
and participation, and environmental factors. The
a priori outcomes were classified according to two
WHO concepts:
(1) Body function: pain reduction, muscle
strength, motor function/motor recovery, ROM,
postural status, balance status, gait status, cadence,
stride length, sensory status, spasticity/muscle
tone, global physician assessment, global patient
assessment, and cardiopulmonary function.
(2) Activities and participation: walking
speed, walking distance, endurance, functional
status, patient adherence, patient satisfaction,
length of stay, discharge disposition, quality of life,
and return to work.
Studies were included if any one of the afore-
mentioned outcomes was measured. A positive
recommendation was made only if a specific inter-
vention was effective for an outcome, as measured
by a validated scale. Psychological outcomes such
as depression were excluded (Table 1).
Statistical analysis
The data were analyzed using Review Manager
Software.97 Continuous data, “data with a poten-
tially infinite number of possible values along a
continuum,” 98 were analyzed using the weighted
mean differences (WMDs) between the interven-
tion and control groups at the end of the study,
where the weight is the inverse of the variance. A
WMD is “a method of meta-analysis used to com-
bine measures on continuous scales (such as
weight), where the mean, standard deviation and
sample size in each group are known.” 98 Dichoto-
mous data or data with only two classifications98
were analyzed using relative risks. According to
Cochrane, the relative risk is “the ratio of risk in
the intervention group to the risk in the control
group. The risk (proportion, probability, or rate) is
the ratio of people with an event in a group to the
total in the group.” 98
Heterogeneity (i.e., variability or difference in
estimated effects between studies) was tested using
the chi-square statistic. We tested data heterogene-
ity across the results of different included studies.
When heterogeneity was not significant, fixed-ef-
8 TOPICS IN STROKE REHABILITATION/SPRING 2006
fect models were used. A fixed-effect model is a statisti-
cal model that stipulates that the units under analysis
(e.g., participants in a meta-analysis study) are the
ones of interest and thus constitute the entire popula-
tion of units. Fixed-effect models were used to gener-
alize data across the included studies. Random-ef-
fects models include both within-study sampling
error (variance) and between-study variation in the
assessment of the uncertainty (confidence interval) of
meta-analysis results. Such random-effects models
were used when heterogeneity was significant. All
figures were created using Cochrane Collaboration
methodology (www.cochrane.org). The square in
TE-Figure 1A. illustrates the WMD between the two
groups, when they are compared for a specific out-
come of interest. The horizontal line represents the
standard deviation of the WMD. If the standard de-
viation line touches the central vertical line of the
graph, the confidence interval contains a zero and the
difference between the two groups is not statistically
significant. For example, in TE-Figure 1A that illus-
trates the comparison between aerobic training and a
control group where gait speed at end of treatment is
the outcome measure, the gait speed of the group
receiving the aerobic training is not statistically differ-
ent from that of the control. Based on previous Phila-
delphia and Ottawa Panels’ consensus,80–82 clinical
improvement for all interventions studied by the
Ottawa Panel was defined as 15% improvement,
relative to a control.
To determine clinical improvement, the abso-
lute benefit and relative difference in the change
from baseline were calculated. Absolute benefit was
calculated as the improvement in the treatment
group minus the improvement in the control
group, maintaining the original units of measure-
ment. The relative difference (RD) was calculated as
the absolute benefit divided by the baseline mean
(weighted for the intervention and control
groups). For dichotomous data, the relative per-
centage of improvement was calculated as the differ-
ence in the percentage of improvement between
the intervention and control groups.80
However, during this meta-analysis, four special
cases of calculation appeared where new formulas
were needed to calculate the RD in the change
from baseline. The first case occurred when the
baseline values were not available for an outcome.
The second scenario involved an outcome that was
measured as a change from baseline, where the
scale of measurement was known but baseline
values were absent. The third case encountered
was also where the outcome was measured as a
change from baseline but both the baseline values
and the measurement scales were either not avail-
able or were nonexistent (e.g., no measurement
scales exist for strength). Finally, in the fourth
case, where the baseline mean was given as 0, a
sum of 1 was added to all the values in the formula
of clinical relevance, based upon the following
assumption: the mean of scale + 1 = mean of
original scale + 1 and SD of scale + 1= SD of
original scale. The new formulas used to calculate
the relative difference in change from baseline for
each of the four aforementioned scenarios are
given in Appendix 2.
The recommendations were graded by their level
(I for RCTs, II for nonrandomized studies) and
strength (A, B, C+, C, or D) of evidence. Evidence
from one or more RCTs of a statistically significant,
clinically important benefit (>15%) was necessary
for a grade A recommendation. A grade B recom-
mendation was given for a statistically significant,
clinically important benefit (>15%), if the evidence
was from observational studies or CCTs. Evidence
of clinical importance (>15%) but not statistical
significance earned a grade C+ recommendation. A
grade C recommendation was given to those inter-
ventions where an appropriate outcome was mea-
sured in a study that met the inclusion criteria, but
no clinically important difference and no statistical
significance were shown. Evidence from one or
more RCTs of a statistically significant benefit fa-
voring the control group (<0%: favors controls)
resulted in a grade D recommendation. A grade of
D+ was given for evidence of clinical importance
(≤ −15% for controls) without statistical signifi-
cance. Evidence from one or more RCTs of a clini-
cally important benefit (≤ −15% for controls) that
was statistically significant, where the number of
participants in the study is equal to or higher than
100, was necessary for a grade D– recommenda-
tion (Table 2). If two or more studies had different
grades for the same outcome for the same time
period, the stronger grade was kept.
Scales demonstrated to be valid and sensitive to
change were required to support a positive recom-
mendation (A or B). Outcomes not supported in

the scientific literature by an existing validation
study but that provided useful information in the
studies were insufficient to warrant a grade A or B
recommendation.
Reviewing the guidelines
The guidelines were sent to the external experts
for review. To judge the clinical usefulness of the
guidelines, the positive recommendations were also
sent to practitioners for feedback. Practitioners were
asked four questions for each guideline: whether the
recommendation was clear, whether the practitio-
ners agreed with the recommendation, whether
they felt that the literature search on the different
intervention of rehabilitation was relevant and com-
plete, and whether the results of the trials in the
guidelines were interpreted according to the practi-
tioners’ understanding of the data. Their questions
and comments were carefully addressed to improve
the clarity of the final guidelines.
Results of literature search
The initial literature search in 2002 identified
1,533 potential articles on post-stroke rehabilita-
tion interventions. From these, 148 articles on
therapeutic exercises for post-stroke rehabilitation
were initially identified. After many updated litera-
ture searches that ended in December 2004, 151
articles were considered potentially relevant and,
of these, 71 articles met the selection criteria and
were included. From the selected articles, we cre-
ated the following five subcategories of therapeutic
exercise: therapeutic exercises, task-oriented train-
ing, CIMT, balance training, and sensory interven-
tion. In the end, 56 trials were considered for
therapeutic exercises. From these, 29 trials met the
inclusion criteria and were then included (Appen-
dix 3A) and 27 trials were excluded for different
reasons (TE-Table 3). In task-oriented training, 39
articles were considered as potentially relevant.
From these, 17 were included (Appendix 3B) and
22 were excluded (TOT-Table 37). Initially, 131
articles were identified for biofeedback. After the
last search in December 2004, 66 were considered
potentially relevant. From these, 19 were included
(Appendix 3C) and 45 were excluded for different
reasons (BFB-Table 57). Also, 433 potential ar-
Clinical Practice Guidelines 9
ticles on gait training for post-stroke rehabilitation
were initially identified. By the end of December
2004, 50 of these were considered potentially rel-
evant based upon the selection criteria. Nineteen
of these articles relating to gait training met the
selection criteria and were included (Appendix
3D). The other 23 trials were excluded from the
final selection for various reasons (Gait-Table 73).
For balance training, 16 articles were considered
after the last search. From these, 11 met the inclu-
sion criteria and were kept (Appendix 3E) and 4
were excluded (BT-Table 96). From the 23 poten-
tially relevant articles concerning sensory interven-
tions at the end of 2004, 9 were included (Appen-
dix 3F) and 14 were excluded (SI-Table 107).
CIMT included 5 articles (Appendix 3G), while
13 trials were excluded (CIMT-Table 117) from
the various updated searches leading up to Decem-
ber 2004 (N = 17 potential articles). Twenty-two
trials were considered relevant in the treatment of
shoulder subluxation. Eight were included (Appen-
dix 3H) and the other 13 were excluded for vari-
ous reasons (SH-Table 123).
The initial search found 368 studies for electro-
therapy. From the last search of 2004, 75 trials were
considered potentially relevant. Thirty studies were
included and 45 were excluded for different rea-
sons. Electrotherapy was then separated into more
specific interventions: electrical stimulation, TENS,
therapeutic ultrasound, and, finally, acupuncture.
In electrical stimulation, 41 articles were considered
potentially relevant. Eleven of them ended up being
included (Appendix 3I) and 26 were excluded (ES-
Table 135). TENS had 11 potentially relevant trials.
Six finally met the inclusion criteria and were in-
cluded (Appendix 3J) and six were excluded
(TENS-Table 148). Four articles were considered
potentially relevant for therapeutic ultrasound. How-
ever, only one of these was included (Appendix
3K) and the remaining three were excluded (US-
Table 157). Acupuncture was found to have 19
potential articles. Seven were included (Appendix
3L) and 11 were excluded (AC-Table 160). Inten-
sity and organization of rehabilitation initially had 272
potential articles. By the end of 2004, 156 were
considered as potentially relevant. Fifty-six were
finally included with respect to the inclusion criteria
(Appendix 3M) and 102 were excluded for differ-
ent reasons (IR-Table 167).
10 TOPICS IN STROKE REHABILITATION/SPRING 2006
Results
Clinical practice guidelines for therapeutic exercises
Aerobic training versus control, level I (RCT):
Grade A for cardiopulmonary function (expira-
tion per minute [VE]), muscle power, and func-
tional status (walking) at end of treatment, 10
weeks (clinically important benefit demonstrated);
grade C+ for gait speed at end of treatment, 10
weeks (clinically important benefit demonstrated
without statistical significance); grade C for car-
diopulmonary function (maximal heart rate, VO2
max, VCO2 max) and motor function at end of
treatment, 10 weeks, and for functional status
(Frenchay Activities Index [FAI]: global, social
outings, walking outside) at end of treatment, 6
months (no benefit demonstrated); grade D for
functional status (FAI: light housework activities)
at end of treatment, 6 months (no benefit demon-
strated but favoring control). Patients with sub-
acute and chronic stroke.
Aerobic individualized program training ver-
sus control, level I (RCT): Grade A for physical
fitness (highest test stage completed of the stress
test and maximal workload) and mobility (stair
climbing) at end of treatment, 8 weeks (clinically
important benefit demonstrated); grade C+ for
mobility (walking distance) at end of treatment, 8
weeks (clinically important benefit demonstrated
without statistical significance); grade C for car-
diovascular function (maximal heart rate, decrease
of resting heart rate and decrease of resting systolic
and diastolic blood pressure), gait speed, and
functional status at end of treatment, 8 weeks (no
benefit demonstrated). Patients with subacute
stroke.
Proprioceptive neuromuscular facilitation
(PNF) for upper extremity versus standard cus-
tomary muscle training, level I (RCT): Grade C
for functional status and upper extremity muscle
strength at end of treatment, 6 weeks (no benefit
demonstrated); grade D for mobility at end of
treatment, 2, 4, and 6 weeks, ROM of the wrist at
end of treatment, 6 weeks (no benefit demon-
strated but favoring control); grade D+ for ROM of
the ankle at end of treatment, 6 weeks (clinically
important benefit favoring control demonstrated
without statistical significance). Patients with post-
acute stroke.
PNF versus Bobath approach training, level I
(RCT): Grade C for mobility at end of treatment,
2, 4, and 6 weeks, ROM of the wrist and ankle at
end of treatment, 6 weeks (no benefit demon-
strated). Patients with subacute stroke.
Bobath approach versus standard customary
muscle training, level I (RCT): Grade C+ motor
function (Sodring Motor Evaluation Scale
[SMES]: upper extremity) at follow-up, 4 years,
quality of life (Nottingham Health Profile [NHP]–
global) at follow-up, 1 year and 4 years, and
quality of life (NHP–loss of energy) at end of
treatment, 3 months (clinically important benefit
demonstrated without statistical significance);
grade C for mobility at end of treatment, 2, 4,
and 6 weeks, motor function (SMES: lower ex-
tremity and trunk, balance, and gait, and Motor
Assessment Scale) at end of treatment, 3 months,
and follow-up, 1 year and 4 years, motor function
(SMES: upper extremity) at end of treatment, 3
months, and follow-up, 1 year, functional status
(Barthel Index) at follow-up, 4 years (no benefit
demonstrated); grade D for ROM of the wrist and
ankle at end of treatment, 6 weeks, functional
status (Barthel Index) at end of treatment, 3
months, and follow-up, 1 year, (no benefit dem-
onstrated but favoring control); grade D+ for
pain relief (NHP–pain) at end of treatment, 3
months (clinically important benefit favoring
control demonstrated without statistical signifi-
cance). Patients with acute and subacute stroke.
Bobath approach training versus control,
level I (RCT): Grade C+ for balance standing at
follow-up, 2 and 12 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for balance sitting at end of treat-
ment, 4 weeks (no benefit demonstrated); grade D
for balance sitting at follow-up, 2 weeks, and bal-
ance standing at end of treatment, 4 weeks (no
benefit demonstrated but favoring control); grade
D+ for balance sitting at follow-up, 12 weeks
(clinically important benefit favoring control dem-

onstrated without statistical significance). Patients
with subacute stroke.
Progressive resistance training versus active
training for the lower extremity, level I (RCT):
Grade A for functional status at end of treatment,
1 month (clinically important benefit demon-
strated). Patients with post-acute stroke.
Active training for the lower extremity versus
control (no exercise), level I (RCT): Grade D for
functional status at end of treatment, 1 month (no
benefit demonstrated but favoring control). Pa-
tients with post-acute stroke.
Progressive resistance training for the lower
extremity versus control, level I (RCT): Grade
C+ for functional status at end of treatment, 1
month (clinically important benefit demonstrated
without statistical significance). Patients with post-
acute stroke.
Progressive resistance versus no resistance
training, level I (RCT): Grade C for motor recov-
ery at end of treatment, 4 weeks and 8 weeks, and
follow-up, 6 months (no benefit demonstrated);
grade D for gait endurance at end of treatment, 4
weeks and 8 weeks, and follow-up, 6 months (no
benefit demonstrated but favoring control). Pa-
tients with subacute stroke.
Functional task training for upper extremity
versus strength training, level I (RCT): Grade C+
favoring functional task training for functional sta-
tus (FIM* self-care and FIM mobility), isometric
torque, grip strength, and palmar pinch at follow-
up, 6.5 to 8 months, and lateral pinch at end of
treatment, 4 weeks, and follow-up, 6.5 to 8
months, and grade C+ favoring strength training
for grip strength and palmar pinch at end of treat-
ment, 4 weeks (clinically important benefit dem-
onstrated without statistical significance); grade C
functional status (FIM self-care and FIM mobility)
at end of treatment, 4 weeks, ROM upper extrem-
ity, pain relief, sensory function upper extremity,
motor function upper extremity, functional status
*FIM is a trademark of Uniform Data System for Medical Rehabilita-
tion, a division of UB Foundation Activities, Inc.
Clinical Practice Guidelines 11
(Functional Test of the Hemiparetic Upper Ex-
tremity [FTHUE]) at end of treatment, 4 weeks,
and follow-up, 6.5 to 8 months, isometric torque
at end of treatment, 4 weeks (no benefit demon-
strated). Patients with subacute stroke.
Strength training versus control, level I (RCT):
Grade A for upper extremity isometric torque at
end of treatment, 4 weeks (clinically important
benefit demonstrated); grade C+ for motor func-
tion upper extremity and functional status
(FTHUE) at end of treatment, 4 weeks, palmar
pinch at end of treatment, 4 weeks, and follow-up,
6.5 to 8 months, grip strength and lateral pinch at
follow-up, 6.5 to 8 months (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for sensory function upper ex-
tremity at end of treatment, 4 weeks, and follow-
up, 6.5 to 8 months, functional status (FIM
mobility), pain relief, and grip strength at end of
treatment, 4 weeks (no benefit demonstrated);
grade D+ for lateral pinch at end of treatment, 4
weeks, functional status (FIM self-care and FIM
mobility), upper extremity isometric torque at fol-
low-up, 6.5 to 8 months (clinically important ben-
efit favoring control demonstrated without statisti-
cal significance); grade D for ROM upper
extremity at end of treatment, 4 weeks, and follow-
up, 6.5 to 8 months, functional status (FIM self-
care) at end of treatment, 4 weeks, pain relief,
motor function upper extremity, functional status
(FTHUE) at follow-up, 6.5 to 8 months (no benefit
demonstrated but favoring control). Patients with
subacute stroke.
Aerobic and strength versus aerobic training,
level I (RCT): Grade A for cardiopulmonary func-
tion and peak torque for shoulder flexors at end of
treatment, 16 weeks (clinically important benefit
demonstrated); grade C+ for peak torque for knee
flexors at end of treatment, 16 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for peak torque for shoulder
extensors and peak torque for knee extensors at
end of treatment, 16 weeks (no benefit demon-
strated). Patients with chronic stroke.
12 TOPICS IN STROKE REHABILITATION/SPRING 2006
Kinetron training for lower extremity versus
control (no Kinetron), level I (RCT): Grade D for
mobility at end of treatment, 5 weeks (no benefit
demonstrated but favoring control). Patients with
post-acute stroke.
Home-based exercise training versus control,
level I (RCT): Grade A for change in gait speed, gait
endurance, torque (change in knee isometric exten-
sors), endurance, and cardiopulmonary function at
end of treatment, 12 weeks; grade C+ for motor
function (change in Fugl-Meyer lower extremity),
change in gait speed, gait endurance, and functional
status (physical function index), strength (change in
grip strength) at end of treatment, 12 weeks (clini-
cally important benefit demonstrated without statis-
tical significance); grade C for motor function
(change in Fugl-Meyer upper extremity and lower
extremity), balance (Berg balance and change in
Berg balance), functional status (Instrumental ADL
and Barthel ADL Index), at end of treatment, 12
weeks (no benefit demonstrated); grade D+ for
balance (functional reach) at end of treatment, 12
weeks (clinically important benefit favoring control
demonstrated without statistical significance);
grade D for torque (change in ankle isometric
dorsiflexors) (no benefit demonstrated but favoring
control). Patients with post-acute stroke.
Skateboard versus overhead pulley training
for the shoulder, level I (RCT): Grade C+ for pain
relief at end of treatment, 8–10 weeks (clinically
important benefit demonstrated without statistical
significance). Patients with subacute stroke.
Overhead pulley versus control (passive ROM
training for shoulder), level I (RCT): Grade D for
pain relief at end of treatment, 8–10 weeks (no
benefit demonstrated but favoring control). Pa-
tients with subacute stroke.
Passive ROM training for shoulder versus
skateboard, level I (RCT): Grade C for pain relief
at end of treatment, 8–10 weeks (no benefit dem-
onstrated). Patients with subacute stroke.
Resisted extension versus ballistic extension
training for the hand, level I (RCT): Grade C+ for
ROM at end of treatment, 2 weeks (clinically im-
portant benefit demonstrated without statistical
significance); grade C motor function at end of
treatment, 2 weeks (no benefit demonstrated). Pa-
tients with subacute and post-acute stroke.
Resisted extension versus resisted grasp
training for the hand, level I (RCT): Grade A for
motor function (change in tapping) at end of treat-
ment, 2 weeks (clinically important benefit dem-
onstrated); grade C+ for ROM at end of treatment,
2 weeks (clinically important benefit demon-
strated without statistical significance); grade C
for motor function (change in grasp/release) at end
of treatment, 2 weeks (no benefit demonstrated).
Patients with subacute and post-acute stroke.
Resisted extension training for the hand ver-
sus control, level I (RCT): Grade A for motor
function (change in tapping) and ROM at end of
treatment, 2 weeks (clinically important benefit
demonstrated); grade D for motor function
(change in grasp/release) at end of treatment, 2
weeks (no benefit demonstrated but favoring con-
trol). Patients with subacute and post-acute stroke.
Ballistic extension versus resisted grasp
training for the hand, level I (RCT): Grade A for
motor function (change in tapping) at end of treat-
ment, 2 weeks (clinically important benefit dem-
onstrated); grade C+ for ROM at end of treatment,
2 weeks (clinically important benefit demon-
strated without statistical significance); grade C
for motor function (change in grasp/release) and
ROM at end of treatment, 2 weeks (no benefit
demonstrated). Patients with subacute and post-
acute stroke.
Ballistic extension training for the hand ver-
sus control, level I (RCT): Grade C+ for motor
function (change in tapping) at end of treatment, 2
weeks (clinically important benefit without statis-
tical significance); grade C for ROM at end of
treatment, 2 weeks (no benefit demonstrated);
grade D for motor function (change in grasp/re-
lease) at end of treatment, 2 weeks (no benefit
demonstrated but favoring control). Patients with
subacute and post-acute stroke.
Resisted grasp training for the hand versus
control, level I (RCT), Grade C+ for ROM at end of
treatment, 2 weeks (clinically important benefit
demonstrated); grade C for motor function (change
in tapping) at end of treatment, 2 weeks (no benefit
demonstrated); grade D for motor function (change
in grasp/release) at end of treatment, 2 weeks (no
benefit demonstrated but favoring control). Patients
with subacute and post-acute stroke.

Robot-aided training versus no robot-aided
training, level I (RCT) and level II (CCT): Grade A
for motor power for shoulder and elbow at end of
treatment, 5 weeks, change in motor power upper
extremity at end of treatment, 6 weeks, motor
function (Fugl-Meyer for shoulder, elbow, and co-
ordination and Motor Status Score [MSS] for
shoulder and elbow) at end of treatment, 5 weeks,
motor function (MSS for wrist and hand) at end of
treatment, 5 weeks, and motor function (change in
MSS for shoulder and elbow) at end of treatment, 6
weeks, and follow-up, 3 years, motor function
(MSS for wrist and hand) at end of treatment, 5
weeks (clinically important benefit demonstrated);
grade B for motor function (MSS for upper ex-
tremity) at end of treatment, 6 weeks (clinically
important benefit demonstrated); grade C+ for
change in motor power for shoulder and elbow at
follow-up, 3 years, motor function (Fugl-Meyer
scale for upper extremity), motor power for upper
extremity at end of treatment, 6 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for motor function (change
in Fugl-Meyer for shoulder, elbow, and coordina-
tion) at end of treatment, 6 weeks, functional sta-
tus (FIM for upper extremity) at end of treatment,
5 weeks and 6 weeks, motor function (change in
Fugl-Meyer for wrist and hand and change in MSS
for wrist and hand) at end of treatment, 6 weeks,
and follow-up, 3 years (no benefit demonstrated);
grade D for motor function (change in Fugl-Meyer
for shoulder, elbow, and coordination) at follow-
up, 3 years (no benefit demonstrated but favoring
control). Patients with subacute-chronic stroke.
Robot-assisted versus neurodevelopmental
(NDT) training, level I (RCT): Grade A for
strength (change in elbow extensors, shoulder in-
ternal rotators, abductors, adductors, and flexors
strength [%]) and functional reach (change in for-
ward medial, forward, forward lateral, and lateral
reach extent) at end of treatment, 2 months; grade
C+ for strength (change in shoulder external rota-
tors and extensors strength [%]) at end of treat-
ment, 2 months (clinically important benefit dem-
onstrated); grade C for functional status (change
in Barthel Index and change in FIM) at follow-up,
6 months, motor function (change in Fugl-Meyer
shoulder and elbow) at end of treatment, 1 month,
Clinical Practice Guidelines 13
2 months, and follow-up, 6 months, motor func-
tion (change in Fugl-Meyer hand and wrist) at end
of treatment, 1 month, 2 months, and follow-up, 6
months, (no benefit demonstrated). Patients with
chronic stroke.
Robot-aided progressive resistance training
versus robot-aided active-assisted training, level
I (RCT): Grade C for decrease of spasticity, motor
function, and strength at end of treatment, 6 weeks
(no benefit demonstrated). Patients with chronic
stroke.
Progressive-resistive robotic training versus
sensorimotor training, level I (RCT): Grade C+
for decrease of spasticity at end of treatment, 6
weeks (clinically important benefit demonstrated
without statistical significance); grade C for motor
function (Fugl-Meyer upper extremity and MSS
for shoulder and elbow and wrist and hand), and
motor power for shoulder and elbow at end of
treatment, 6 weeks (no benefit demonstrated). Pa-
tients with chronic stroke.
Music-making training versus control, level I
(RCT): Grade C+ for ROM (elbow extension) at
end of treatment, 10 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for ROM (shoulder flexion) at end
of treatment, 10 weeks (no benefit demonstrated).
Patients with post-acute stroke.
Water-based training versus control, level I
(RCT): Grade A for hip and knee extensors
strength (affected side) at end of treatment, 8
weeks (clinically important benefit demon-
strated); grade C+ for cardiopulmonary function
(VO2 max) at end of treatment, 8 weeks, muscle
power at end of treatment, 8 weeks, and gait
speed at end of treatment, 8 weeks (clinically
important benefit demonstrated without statisti-
cal significance); grade C for hip and knee exten-
sors strength (unaffected side) at end of treat-
ment, 8 weeks (no benefit demonstrated); grade
D for balance at end of treatment, 8 weeks (no
benefit demonstrated but favoring control). Pa-
tients with chronic stroke.
Agility exercise versus stretching/weight-
shifting exercise, level I (RCT): Grade C+ for step
reaction time at follow-up, 1 month (clinically
important benefit demonstrated without statistical
significance); grade C for balance, mobility, bal-
14 TOPICS IN STROKE REHABILITATION/SPRING 2006
ance confidence, and quality of life at end of treat-
ment, 10 weeks, and follow-up, 1 month, step reac-
tion time at end of treatment, 10 weeks (no benefit
demonstrated). Patients with chronic stroke.
Maximal isokinetic strengthening versus con-
trol, level I (RCT): Grade C+ for change in
strength at end of treatment, 6 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for quality of life and at end
Summary of trials
Twenty-eight RCTs and one CCT were included
that evaluated the efficacy of different kinds of
therapeutic exercises in comparison to either an-
other type of exercise therapy or to a placebo (N =
1,318). 99–127 Seven subcategories of exercise
therapies were administered: 1) aerobic exercises:
cycle ergometer for the lower extremities (n =
185) 100,109,110,121,123 and for the upper body (n =
40),100 as well as water-based aerobic training (n =
12)101; 2) therapies based on theories in neurol-
ogy: proprioceptive neuromuscular facilitation
(PNF; n = 173),102,115 the Bobath approach (n =
40),119 and approaches based on neurophysi-
ological and developmental theories (n =
47)106,116; 3) exercises for the lower extremity:
progressive resistance exercises (n = 230),107,110,118
active exercises (also for the trunk; n = 77),107 and
passive ROM (n = 20)110; 4) exercises for the
upper extremity: gross and fine movement exer-
cises and muscle strengthening (n = 92), 101,120,127
robot-aided therapy with goal-directed move-
ments (n = 97),99,105,116,122,125,126 active exercises
with a skateboard and overhead pulley and pas-
sive ROM exercises for the shoulder (n = 28),111
resisted extension of fingers, ballistic extension
and resisted grasp for the hand (n = 20)121; 5)
balance exercises (n = 88)117,119; 6) functional
training (n = 106)114,127; 7) mixture of exercise
therapies: ROM and flexibility exercises for the
upper and lower extremities and trunk, strength-
ening resistive exercises using PNF patterns for
upper and lower extremities or theraband exer-
cises, balance training, upper extremity func-
tional use, and progressive walking program/ex-
ercise on a bicycle ergometer (n = 112).103,104 The
of treatment, 6 weeks (no benefit demonstrated);
grade D for level-walking and stair-walking (no
benefit demonstrated but favoring control). Pa-
tients with chronic stroke.
Mental imagery versus standard functional
training, level I (RCT): Grade A for level of inde-
pendence in performing tasks at end of treatment,
1 week, 2 weeks, and 3 weeks (clinically important
benefit demonstrated). Patients with acute stroke.
length of each exercise session ranged from 20
minutes to 2 hours per day, and the treatment
programs’ durations ranged from 2 sessions per
week to daily for 2 weeks to 6 months or until
discharge. In Duncan et al.’s study,103 patients
were instructed to continue exercising indepen-
dently for an additional 4 weeks, while in
Langhammer’s studies, patients continued treat-
ment as outpatients112,113 (TE-Appendix 3A).
In total, 31 studies were excluded for the follow-
ing reasons: there was no control group in 10 of the
studies,128–137 healthy subjects were used in three
studies,138–140 two studies had insufficient statistical
data,141,142 two studies were not specific to stroke
patients,143,144 and another study lacked an interven-
tion.145 The remaining studies were excluded for
various other reasons146–158 (TE-Table 3).
Efficacy
Aerobic training for patients in the subacute
and chronic phases of stroke recovery versus a
control group (3 RCTs, n = 145)108,121,123 resulted in
clinically important benefits for cardiopulmonary
function (expiration per minute [VE]), muscle
power according to the maximal workload, and
functional status (walking measured through the
Adjusted Activity Score [AAS]) at the end of 10
weeks of treatment (18%, 35%, and 41% RD,
respectively). These outcomes were also statisti-
cally significant (TE-Figure 1A, Table 4A). Clini-
cally important benefits were demonstrated with-
out statistical significance for gait speed at the end
of 10 weeks of treatment (33% RD; TE-Figure 1A,
Table 4A). No benefit was demonstrated for car-
diopulmonary function as measured (maximal

heart rate, VO2 max, VCO2 max) and motor func-
tion (Fugl-Meyer index) at the end of 10 weeks of
treatment (TE-Figure 1A, Table 4A) and for func-
tional status (FAI total, social outings, walking
outside) at 6 months (TE-Figure 1B, Table 4B).
Results favored the control group for functional
status during light housework activities at the end
of 6 months of treatment (TE-Figure 1B, Table
4C). No other outcomes were measured.
For individualized aerobic program training
versus control in patients during the subacute
phase of stroke (one RCT, n = 90),109 statistically
significant and clinically important benefits were
found for physical fitness (highest test stage com-
pleted of the stress test and maximal workload)
and mobility (stair climbing) at the end of 8 weeks
of treatment (33%, 36%, and 130% RD, respec-
tively; TE-Figure 2, Table 5). Clinically important
benefits without statistical significance were dem-
onstrated for mobility (walking distance) at the
end of 8 weeks of treatment (28% RD; TE-Figure
2, Table 5). No benefit was demonstrated for
cardiovascular function (maximal heart rate, de-
crease of resting heart rate, decrease of resting
systolic and diastolic blood pressure), gait speed,
and functional status (FIM; TE-Figure 2, Table 5)
at the end of 8 weeks of treatment. No other
outcomes were measured.
For PNF versus standard customary muscle
training of the upper extremity in post-acute
stroke patients (two RCTs, n = 173),102,115 no
benefit was found for functional status (Barthel
Index) and upper extremity muscle strength
(manual muscle test) at the end of 6 weeks of
treatment (TE-Figure 3, Table 6A). Results fa-
vored the control group for mobility number
(number of patients independent in walking) at
the end of 2, 4, and 6 weeks of treatment (TE-
Figure 3, Table 6B) and for ROM of the wrist
number (number of patients with limited ROM) at
the end of 6 weeks of treatment (TE-Figure 3,
Table 6B). A clinically important benefit without
statistical significance was found that favored con-
trol for ROM of the ankle (number of patients with
limited ROM; TE-Figure 3, Table 6B). No other
outcomes were measured.
For PNF versus the Bobath approach in sub-
acute stroke patients (one RCT, n = 131), 102 no
Clinical Practice Guidelines 15
benefits were found for mobility (number of pa-
tients independent in walking) at the end of 2, 4,
and 6 weeks of treatment and ROM of the wrist
and ankle the at end of 6 weeks of treatment (TE-
Figure 4, Table 7). No other outcomes were
measured.
For the Bobath approach versus standard cus-
tomary muscle training in acute and subacute
stroke patients (three RCTs, n = 212),102,112,113 clini-
cally important benefits without statistical signifi-
cance were found for motor function (SMES upper
extremity) at the 4-year follow-up (16% RD), qual-
ity of life (NHP–global) at the 1-year and 4-year
follow-ups (26% and 22% RD, respectively), and
quality of life (NHP– loss of energy) at the end of 3
months of treatment (23% RD; TE-Figure 5,
Table 8). No benefits were demonstrated for mo-
bility (number of patients independent in walking)
at the end of 2, 4, and 6 weeks of treatments,
motor function (SMES lower extremity and trunk,
balance and gait, and Motor Assessment Scale) at
the end of 3 months of treatment and at the 1- and
4-year follow-up, motor function (SMES upper
extremity) at the end of 3 months of treatment and
at the 1-year follow-up, and functional status
(Barthel Index) at 4 years follow-up (TE-Figure 5,
Table 8). There was no benefit demonstrated, but
results favored the control therapy for ROM of the
wrist and ankle at the end of 6 weeks of treatment
and functional status (Barthel Index) at the end of
3 months of treatment and at the 1-year follow-up
(TE-Figure 5, Table 8). A clinically important ben-
efit was found without statistical significance favor-
ing the control therapy for pain relief (NHP–pain) at
the end of 3 months of treatment (TE-Figure 5,
Table 8). No other outcomes were measured.
For Bobath approach training in subacute
stroke patients versus control (one RCT, n =
40),119 a clinically important benefit without statis-
tical significance was found for standing balance at
the 2 and 12 week follow-up (22%–35% RD). No
benefit was demonstrated for sitting balance at end
of 4 weeks of treatment. Results favored the con-
trol group for sitting balance at the 2-week follow-
up and standing balance at the end of 4 weeks of
treatment. A clinically important benefit without
statistical significance that favored control was
demonstrated for sitting balance at the 12-week
16 TOPICS IN STROKE REHABILITATION/SPRING 2006
follow-up (TE-Figure 6, Table 9). No other out-
comes were measured.
For progressive resistance training versus ac-
tive training of the lower extremity in post-
acute stroke patients (one RCT, n = 77),107 a
clinically important benefit with statistical signifi-
cance was found for functional status (number of
patients improved in more than two activities of
daily living [ADL] items; 34% RD) at end of treat-
ment, 1 month (TE-Figure 7, Table 10). No other
outcomes were measured.
For active training of the lower extremity in
post-acute stroke patients versus control (no
exercise) (one RCT, n = 77), 107 no benefit was
demonstrated, but results favored control for func-
tional status (number of patients improved in
more than two ADL items) without statistical sig-
nificance at end of 1 month of treatment (TE-
Figure 8, Table 11). No other outcomes were
measured.
For progressive resistance training of the
lower extremity in post-acute stroke patients
versus control (one RCT, n = 77),107 clinically
important benefits without statistical significance
were found for functional status (number of pa-
tients improved in more than two ADL items) at
the end of 1 month of treatment (26% RD; TE-
Figure 9, Table 12). No other outcomes were
measured.
For progressive resistance in subacute stroke
patients versus no resistance training (one RCT,
n = 133),118 no benefit was shown for motor recov-
ery (Chedoke-McMaster Stroke Assessment
[CMSA]: walking and gross motor function index
section) at the end 4 weeks and 8 weeks of treat-
ment and at the 6-month follow-up. No benefit
was demonstrated, but results favored control for
gait endurance (2-minute walking test) at the end
of 4 and 8 weeks of treatment and at the 6-month
follow-up (TE-Figure 10, Table 13). No other
outcomes were measured.
For functional task training of the upper ex-
tremity versus strength training in subacute
stroke patients (one RCT, n = 60),127 clinically
important benefits without statistical significance
were shown favoring functional task training for
functional status (FIM self-care and FIM mobility;
24% and 17% RD, respectively), upper extremity
isometric torque (62% RD), grip strength (317%
RD), and palmar pinch (119% RD) at the 6.5- to 8-
month follow-up and lateral pinch at the end of 4
weeks of treatment at the 9-month follow-up (39%
and 47% RD, respectively) (TE-Figure 11, Table
14). A clinically important benefit without statisti-
cal significance was demonstrated favoring
strength training for grip strength (30% RD) and
palmar pinch (19% RD) at the end of 4 weeks of
treatment (TE-Figure 11, Table 14). No benefit
was demonstrated for upper extremity ROM, pain
relief, upper extremity sensory and motor func-
tion, and functional status (FTHUE) at the end of 4
weeks of treatment and at the 6.5- to 8-month
follow-up and isometric torque at the end of 4
weeks of treatment (TE-Figure 11, Table 14). No
other outcomes were measured.
For strength training in subacute stroke pa-
tients versus control (one RCT, n = 60),127 clini-
cally important benefits were found with statisti-
cal significance for upper extremity isometric
torque (34% RD; TE-Figure 12, Table 15) at the
end of 4 weeks of treatment. A clinically impor-
tant benefit was demonstrated without statistical
significance for motor function of the upper ex-
tremity (42% RD) and functional status, mea-
sured by the FTHUE (17% RD) at the end of 4
weeks of treatment, palmar pinch at the end of 4
weeks of treatment and at the 6.5- to 8-month
follow-up (49%–54% RD), and grip strength
(36% RD) and lateral pinch (30% RD) at the 6.5-
to 8-month follow-up (TE-Figure 12, Table 15).
No benefit was demonstrated for sensory func-
tion of the upper extremity at the end of 4 weeks
of treatment and at the 6.5- to 8-month follow-
up, functional status (FIM mobility), and pain
relief and grip strength at the end of 4 weeks of
treatment (TE-Figure 12, Table 15). Clinically
important benefits favoring control without sta-
tistical significance were demonstrated for lateral
pinch at the end of 4 weeks of treatment, func-
tional status (FIM self-care and FIM mobility),
and upper extremity isometric torque at the 6.5-
to 8-month follow-up (TE-Figure 12, Table 15).
The results favored control for upper extremity
ROM at the end of 4 weeks of treatment and at
the 6.5- to 8-month follow-up and functional
status (FTHUE) at the 6.5- to 8-month follow-up

(TE-Figure 12, Table 15). No other outcomes
were measured.
For combined aerobic and strength versus
aerobic training alone in chronic stroke patients
(one RCT, n = 40),100 clinically important benefits
with statistical significance were shown for car-
diopulmonary function (VO2 max; 18% RD) and
peak torque of shoulder flexors (47% RD) at the
end of 16 weeks of treatment. A clinically impor-
tant benefit was demonstrated without statistical
significance for peak torque of knee flexors (16%
RD) at the end of 16 weeks of treatment. No
benefit was demonstrated for peak torque of
shoulder extensors and peak torque of knee exten-
sors at the end of 16 weeks of treatment (TE-
Figure 13, Table 16). No other outcomes were
measured.
For Kinetron training of the lower extremity
in stroke patients during the post-acute phase
of recovery versus control (no Kinetron; one
RCT, n = 20),106 no benefit was found, but results
favored control for mobility (Functional
Ambulation Profile [FAP] test) at the end of 5
weeks of treatment (TE-Figure 14, Table 17). No
other outcomes were measured.
For home-based exercise training in post-
acute stroke patients versus control (two RCTs,
n = 112),103,104 clinically important benefits with
statistical significance were shown for change in
gait speed (48% RD), gait endurance (change in 6-
minute walk test, feet; 59% RD), torque (change in
knee isometric extensors; 61% RD), endurance
(change in duration of bicycle exercises; 164%
RD), and cardiopulmonary function (change in
peak VO2; 185% RD) at the end of 12 weeks of
treatment (TE-Figure 15, Table 18). A clinically
important benefit was demonstrated without sta-
tistical significance for motor function (change in
Fugl-Meyer lower extremity; 16.7% RD), change
in gait speed (94% RD), gait endurance (change in
6-minute walk test, feet; 15% RD), functional sta-
tus (physical function index; 20% RD), and
strength (change in grip strength; 17% RD) at the
end of 12 weeks of treatment (TE-Figure 15,
Table 18). No benefit was demonstrated for motor
function (change in Fugl-Meyer upper extremity
and lower extremity), balance (Berg balance and
change in Berg balance), or functional status (In-
Clinical Practice Guidelines 17
strumental ADL and Barthel ADL Index) at the end
of 12 weeks of treatment (TE-Figure 15, Table
18). A clinically important benefit without statisti-
cal significance was shown favoring control for
balance (functional reach) at the end of 12 weeks
of treatment (TE-Figure 15, Table 18). No benefit
was demonstrated, but results favored control for
torque (change in ankle isometric dorsiflexors) at
the end of 12 weeks of treatment (TE-Figure 15,
Table 18). No other outcomes were measured.
For skateboard use versus overhead pulley
training of the affected shoulder in subacute
stroke patients (one RCT, n = 28),111 a clinically
important benefit without statistical significance
was found for pain relief (number of patients with-
out shoulder pain) at the end of 8–10 weeks of
treatment (50% RD; TE-Figure 16, Table 19). No
other outcomes were measured.
Comparing the use of an overhead pulley
versus control (passive ROM training for shoul-
der) for subacute stroke patients (one RCT, n =
28),111 no benefit was found but results favored
control for pain relief (number of patients without
pain) at the end of 8–10 weeks of treatment (TE-
Figure 17, Table 20). No other outcomes were
measured.
For passive ROM training of the affected
shoulder versus skateboard use in subacute
stroke patients (one RCT, n = 28),111 no benefit was
shown for pain relief (number of patients without
pain) at the end of 8–10 weeks of treatment (TE-
Figure 18, Table 21). No other outcomes were
measured.
For resisted extension versus ballistic exten-
sion training of the hand in subacute and post-
acute stroke patients (one RCT, n = 20),121 a clini-
cally important benefit without statistical
significance was found for ROM change in finger
ROM at the end of 2 weeks of treatment (42% RD).
No benefit was demonstrated for motor function
(change in grasp/release and change in status-tap-
ping; TE-Figure 19, Table 22). No other out-
comes were measured.
For resisted extension exercises versus re-
sisted grasp training of the hand in subacute and
post-acute stroke patients (one RCT, n = 20),121 a
clinically important benefit with statistical signifi-
cance was found for motor function (change in
18 TOPICS IN STROKE REHABILITATION/SPRING 2006
tapping) at the end of 2 weeks of treatment (115%
RD). A clinically important benefit was demon-
strated without statistical significance for ROM
(change in finger ROM) at the end of 2 weeks of
treatment (50% RD). No benefit was demonstrated
for motor function (change in grasp/release; TE-
Figure 20, Table 23). No other outcomes were
measured.
For resisted extension training of the hand in
subacute and post-acute stroke patients versus
control (one RCT, n = 20),121 a clinically important
benefit with statistical significance was shown for
motor function (change in tapping) and ROM
(change in finger ROM) at the end of 2 weeks of
treatment (118% and 122% RD, respectively). No
benefit was demonstrated, but results favored con-
trol for motor function (change in grasp/release) at
the end of 2 weeks of treatment (TE-Figure 21,
Table 24). No other outcomes were measured.
For ballistic extension versus resisted grasp
training of the hand in subacute and post-acute
stroke patients (one RCT, n = 20),121 a clinically
important benefit with statistical significance was
found for motor function (change in tapping) at
the end of 2 weeks of treatment (108% RD). A
clinically important benefit was demonstrated
without statistical significance for ROM (change in
finger ROM) at the end of 2 weeks of treatment
(91%). No benefit was demonstrated for motor
function (change in grasp/release) and ROM
(change in finger ROM) at the end of 2 weeks of
treatment (TE-Figure 22, Table 25). No other
outcomes were measured.
For ballistic extension training of the hand in
subacute and post-acute stroke patients versus
control (one RCT, n = 20),121 a clinically important
benefit without statistical significance was shown
for motor function (change in tapping) and ROM
(change in finger ROM) at the end of 2 weeks of
treatment (112% and 97% RD, respectively). No
benefit was demonstrated for ROM (change in
finger ROM) at the end of 2 weeks of treatment,
and no benefit was demonstrated but results fa-
vored control for motor function (change in grasp/
release) at the end of 2 weeks of treatment (TE-
Figure 23, Table 26). No other outcomes were
measured.
For resisted grasp training of the hand in
subacute and post-acute stroke patients versus
control (one RCT, n = 20), 121 a clinically impor-
tant benefit without statistical significance was
found for ROM (change in finger ROM) at the end
of 2 weeks of treatment (84% RD). No benefit was
demonstrated for motor function (change in tap-
ping) at the end of 2 weeks of treatment. No
benefit was demonstrated, but results favored con-
trol for motor function (change in tapping and
change in grasp/release) and ROM (change in fin-
ger ROM) at the end of 2 weeks of treatment (TE-
Figure 24, Table 27). No other outcomes were
measured.
For robot-aided training in subacute-chronic
stroke patients versus no robot-aided training
(two RCTs and one CCT, n = 88),99,125,126 clinically
important benefits with statistical significance
were shown for motor power for shoulder and
elbow at the end of 5 weeks of treatment (124%
RD), change in motor power upper extremity at
the end of 6 weeks treatment (15% RD), motor
function (Fugl-Meyer for shoulder and elbow and
MSS for shoulder and elbow) at the end of 5 weeks
of treatment (18% and 43% RD, respectively), mo-
tor function (MSS for wrist and hand) at the end of
5 weeks of treatment (40% RD), and motor func-
tion (change in MSS for shoulder and elbow) at the
end of 6 weeks of treatment and at 3 years follow-
up (32.5% and 30.5% RD, respectively; TE-Figure
25A, Table 28A). A clinically important benefit
with statistical significance was demonstrated for
motor function (MSS for upper extremity) at the
end of 6 weeks of treatment (64% RD; TE-Figure
25A, Table 28A). Clinically important benefits
were demonstrated without statistical significance
for change in motor power for shoulder and elbow
at 3 years follow-up (20% RD), motor function
(Fugl-Meyer scale for upper extremity; 26% RD) at
the end of 6 weeks of treatment, and motor power
for upper extremity at the end of 6 weeks of treat-
ment (37% RD; TE-Figure 25A&B, Table 28
A&B). No benefit was demonstrated for motor
function (change in Fugl-Meyer for shoulder, el-
bow, and coordination) at the end of 6 weeks of
treatment, functional status (FIM for upper ex-
tremity) at the end of 5 weeks and 6 weeks of
treatment, and motor function (change in Fugl-
Meyer for wrist and hand and change in MSS
for wrist and hand) at the end of 6 weeks of
treatment and at 3 years follow-up (TE-Figure

25A&B, Table 28A&B). No benefit was demon-
strated, but results favored control for motor func-
tion (change in Fugl-Meyer for shoulder, elbow,
and coordination; TE-Figure 25A, Table 28A).
No other outcomes were measured.
For robot-assisted versus neurodevelopmen-
tal (NDT) training in chronic stroke patients
(one RCT, n = 27), 116 a clinically important benefit
with statistical significance was found for strength
(change in elbow extensors, shoulder internal rota-
tors, abductors, adductors, and flexors strength
[%]) and functional reach (change in forward me-
dial, forward, forward lateral, and lateral reach
extent) at the end of 2 months of treatment (270%,
99%, 134%, 107%, 160%, 270%, 175%, 229%,
and 612% RD, respectively; TE-Figure 26A&B,
Table 29A&B). A clinically important benefit was
demonstrated without statistical significance for
strength (change in shoulder external rotators and
extensors strength [%]) at the end of 2 months of
treatment (47% and 198% RD, respectively; TE-
Figure 26A, Table 29A). No benefit was demon-
strated for functional status (change in Barthel
Index) and motor function (change in FIM) at 6
months follow-up, for motor function (change in
Fugl-Meyer shoulder and elbow) at the end of 1
month and 2 months of treatment and at 6 months
follow-up, for motor function (change in Fugl-
Meyer hand and wrist) at the end of 1 month of
treatment, for strength (change in elbow flexors
and extensors, and shoulder external rotators, in-
ternal rotators, abductors, adductors, flexors and
extensors strength [%]) and functional reach
(change in forward medial, forward, forward lat-
eral, and lateral reach extent) at the end of 2
months of treatment (TE-Figure 26A&B, Table
29A&B). No other outcomes were measured.
For robot-aided progressive resistance train-
ing versus robot-aided active-assisted training
in chronic stroke patients (one RCT, n = 18),122
there was no benefit shown for decrease in spastic-
ity (measured by the Modified Ashworth scale),
increase in motor function (Fugl-Meyer scale, MSS
shoulder/elbow and hand/wrist), and improved
strength (measured with manual muscle testing
and peak force generated) at the end of 6 weeks of
treatment (TE-Figure 27, Table 30). No other
outcomes were measured.
For progressive-resistive robotic training ver-
Clinical Practice Guidelines 19
sus sensorimotor training in chronic stroke pa-
tients (one RCT, n = 20),105 clinically important
benefits without statistical significance were
shown for decrease in spasticity, measured by the
Modified Ashworth scale (20% RD) at the end of 6
weeks of treatment. No benefit was demonstrated
for motor function (Fugl-Meyer upper extremity
and MSS for shoulder/elbow and wrist/hand), and
motor power for the shoulder and elbow (Medical
Research Council [MRC] motor power score) at
the end of 6 weeks of treatment (TE-Figure 28,
Table 31). No other outcomes were measured.
For music-making training in post-acute
stroke patients versus control (RCT, n = 20),120
clinically important benefits without statistical sig-
nificance were demonstrated for ROM (elbow ex-
tension) at the end of 10 weeks of treatment (20%
RD). No benefit was demonstrated for ROM
(shoulder flexion) at the end of 10 weeks of treat-
ment (TE-Figure 29, Table 32). No other out-
comes were measured.
For water-based training in chronic stroke
patients versus control (one RCT, n = 12),101
clinically important benefits with statistical signifi-
cance were found for hip and knee extensors
strength (affected side) at the end of 8 weeks of
treatment (15% RD). Clinically important benefits
were demonstrated without statistical significance
for cardiopulmonary function (VO2 max) at the
end of 8 weeks of treatment (20% RD), muscle
power (maximal workload) at the end of 8 weeks
of treatment (22% RD), and gait speed at end of 8
weeks of treatment (16% RD). No benefit was
demonstrated for hip and knee strength (unaf-
fected side) at the end of 8 weeks of treatment, and
no benefit was demonstrated but results favored
control for balance (Berg scale) at the end of 8
weeks of treatment (TE-Figure 30, Table 33).
No other outcomes were measured.
For agility exercise versus stretching/weight-
shifting exercise in chronic stroke patients (one
RCT, n = 48),117 clinically important benefits with-
out statistical significance were shown for step
reaction time at the 1-month follow-up (24% RD).
No benefit was demonstrated for balance (Berg
scale), mobility (Timed Up & Go), balance confi-
dence (activities-specific Balance Confidence
Scale), and quality of life (NHP) at the end of 10
weeks of treatment and at the 1-month follow-up
20 TOPICS IN STROKE REHABILITATION/SPRING 2006
and for step reaction time at the end of 10 weeks of
treatment (TE-Figure 31, Table 34). No other
outcomes were measured.
For maximal isokinetic strengthening of
chronic stroke patients versus control (one
RCT, n = 20),110 a clinically important benefit
without statistical significance was found for
change in strength (composite score of paretic
side) at end of 6 weeks of treatment (112% RD).
No benefit was demonstrated for quality of life
(SF-36) at the end of 6 weeks of treatment. No
benefit was demonstrated, but results favored con-
trol for level walking (self-selected and maximal
gait speed) and stair walking (self-reported and
maximal stair climbing speed; TE-Figure 32,
Table 35). No other outcomes were measured.
For mental imagery therapy versus standard
functional training in acute stroke patients (one
RCT, n = 46), 114 clinically important benefits with
statistical significance were demonstrated for the
level of independence at the end of 1 week, 2
weeks, and 3 weeks of treatment (19%, 33%, and
36% RD, respectively; TE-Figure 33, Table 36).
No other outcomes were measured.
Strength of published evidence compared with other
guidelines
The Ottawa Panel found level I evidence (three
RCTs) for strengthening exercises and level I evi-
dence (one RCT) for a strengthening and aerobic
exercise program in patients post stroke. The
strength of the evidence has also been graded by
the Department of Veterans Affairs/Department of
Defense US Army (VA/DoD),75 which found level
III evidence (a working group consensus) for
strengthening exercises in stroke patients during
the acute phase of recovery and level II evidence
(well-designed cohort and case-control analytic
studies) for a strengthening and aerobic exercise
program in stroke rehabilitation (Appendix 4).
The Ottawa Panel also found level I evidence
(one RCT) for sensorimotor rehabilitation, while
the Heart and Stroke Foundation of Ontario71
found level I and II evidence (RCTs and cohort
trials; Appendix 4).
The Ottawa Panel found level I evidence (one
RCT) for the Bobath approach training, level I
evidence (nine RCTs) for strength training, level I
(five RCTs) and level II (one CCT) evidence for
robot-aided training for the upper extremity, and
level I evidence (two RCTs) for balance training
in patients with acute to chronic stroke. In com-
parison, St. Joseph’s Health Care London73 found
level II evidence (at least one nonexperimental
study) for the restorative (Bobath) approach,
level I evidence (two RCTs) for strength training,
level I evidence (two RCTs with conflicting evi-
dence*) for sensory-motor training with robotic
devices, and level I evidence (four RCTs) for bal-
ance training as an intervention in stroke reha-
bilitation (Appendix 4).
Last, the Ottawa Panel found level I evidence
(one RCT) for overhead pulley exercises for the
shoulder of patients with subacute stroke. For this
intervention, the strength of evidence found by the
VA/DoD75 was also level I evidence (a properly
done RCT). Both the Heart and Stroke Foundation
of Ontario71 and St. Joseph’s Health Care London73
found level I evidence (one RCT) for aggressive
range of motion therapy using overhead pulleys
for the shoulder (Appendix 4).
Clinical recommendations compared with other
guidelines
The Ottawa Panel concluded that good evidence
exists to recommend that chronic stroke patients
participate in regular strengthening and aerobic
exercise programs (grade A for cardiopulmonary
function and peak torque for shoulder flexors,
grade C+ for peak torque of knee flexors), and
poor evidence exists to support strengthening in
the rehabilitation of acute and subacute stroke
patients (grade D for ROM of wrist and ankle,
functional status, quality of life, and grade D+ for
pain relief ). These recommendations are in accor-
dance with previous guidelines by the VA/DoD;
Appendix 4).75
The Ottawa Panel can neither corroborate or
refute the findings of the Heart and Stroke Foun-
*According to the St. Joseph’s Health Care London, conflicting evi-
dence refers to a disagreement between the findings of at least two
RCTs. Where there were more than four RCTs and the results of only
one was conflicting, the conclusion was based on the results of the
majority of the studies, unless the study with conflicting results was of
higher quality.

dation of Ontario, which indicate that there is
good evidence that sensorimotor training for the
upper limb should be included as an intervention
for post-stroke rehabilitation.71 The Ottawa Panel
could only assess the effectiveness of sensorimotor
training in comparison to a program of progres-
sive-resistive robotic training, not to control. Be-
tween the two interventions, there were no signifi-
cant differences for motor function or motor
power, although progressive-resistive robotic
training produced a clinically important improve-
ment in spasticity relative to sensorimotor training
(grade C+; Appendix 4).
The Ottawa Panel, based upon poor evidence
(grades C+ and D for pain relief), does not recom-
mend the inclusion of overhead pulley exercises as
an intervention for the shoulder of subacute stroke
patients. This is in line with the guidelines of the
VA/DoD,75 which also do not support overhead
pulley exercises. Similarly, the evidence found by
St. Joseph’s Health Care London73 and the Heart
and Stroke Foundation of Ontario71 is against the
use of overhead pulley (Appendix 4).
The Ottawa Panel found good evidence to rec-
ommend a home-based exercise program that in-
cludes strength training to improve the gait of
post-acute stroke patients (grades A and C+ for
gait speed and endurance). The Panel also found
good evidence for robot-assisted training for the
upper extremity (grade A for motor power, func-
Clinical practice guidelines for task-oriented training
Seated reaching task training versus control,
level I (RCT): Grade A for hand movement time
(ipsilateral and across), peak vertical ground reac-
tion force (forward and across) at end of treatment,
2 weeks (clinically important benefit demon-
strated); grade C+ for hand movement time (for-
ward) at end of treatment, 2 weeks (clinically im-
portant benefit demonstrated without statistical
significance); grade C for reaching distance and
peak vertical reaction force (ipsilateral) at end of
treatment, 2 weeks (no benefit demonstrated). Pa-
tients with chronic stroke.
Adapted game training versus control, level I
(RCT): Grade A for ROM in forearm supination at
Clinical Practice Guidelines 21
tion and strength of the shoulder and elbow) of
subacute and chronic stroke patients. Fair evi-
dence was found by the Panel to support the use of
balance training techniques such as the Bobath
approach (grade C+ for balance standing) and agil-
ity exercise training (grade C+ for step reaction
time) as interventions for acute, subacute, and
chronic post-stroke patients. Last, the Panel also
found fair evidence to recommend the Bobath ap-
proach (grade C+ for motor function of the upper
extremity and for quality of life) for acute and
subacute patients (Appendix 4).
Similarly, St. Joseph’s Health Care London73
found strong evidence that strength training im-
proves gait and sensory motor training and im-
proves functional and motor outcomes for the
shoulder and elbow but not for the wrist and hand.
They found that balance training also improves out-
comes. St. Joseph’s guidelines also indicated that
there is limited evidence that the Bobath approach
improves the quality of gait. However, in contrast to
the Ottawa Panel, limited evidence was found by St.
Joseph’s for the use of the restorative approach
(Bobath), as it resulted in increasing the number of
patients who were institutionalized (Appendix 4).
All other interventions and outcomes evaluated
by the Ottawa Panel could not be compared to
other such previously existing clinical practice
guidelines, because they were deemed not appli-
cable (Appendix 4).
end of treatment, 1 day (clinically important ben-
efit demonstrated). Patients with post-acute
stroke.
Repetitive elbow joint movement training
versus control, level II (CCT): Grade C for joint
velocity (maximal extension velocity to target and
to mouth), motor function (Fugl-Meyer) at end of
treatment, 3 weeks (no benefit demonstrated);
grade D for functional status, motor function
(Frenchay Arm Test), and joint velocity (maximal
flexion velocity to target and to mouth) at end of
treatment, 3 weeks (no benefits demonstrated but
favoring control). Patients with subacute stroke.
Functional upper extremity training versus
22 TOPICS IN STROKE REHABILITATION/SPRING 2006
control, level I (RCT): Grade A for quality of life
(NHP) at end of treatment, 12 weeks (clinically
important benefit demonstrated); grade C+ for
comfortable gait speed at end of treatment, 12 and
20 weeks, and follow-up, 6 weeks, maximum gait
speed at end of treatment, 12 and 20 weeks, and
follow-up, 6 and 32 weeks, and quality of life
(NHP) at follow-up, 6 and 32 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for quality of life (NHP) at
follow-up, 18 weeks, comfortable gait speed at
follow-up, 18 and 32 weeks, maximum gait speed
at follow-up, 18 weeks, quality of life (Sickness
Impact Profile) at end of treatment, 12 weeks, and
follow-up, 6, 18, and 32 weeks, functional status
(Frenchay Activities Index) at follow-up, 6 and 32
weeks, and mobility at end of treatment, 12 weeks,
and follow-up, 6, 18, and 32 weeks (no benefit
demonstrated); grade D for mobility at end of
treatment, 20 weeks (no benefit demonstrated but
favoring control). Patients with subacute stroke.
Functional task training for upper extremity
versus control, level I (RCT): Grade C+ for mo-
tor function upper extremity, functional status
(FTHUE), and isometric torque at end of treat-
ment, 4 weeks, grip strength and lateral pinch at
follow-up, 8 months, palmar pinch at end of
treatment, 4 weeks, and follow-up, 8 months
(clinically important benefit demonstrated with-
out statistical significance); grade C for func-
tional status (FIM mobility) at end of treatment, 4
weeks, functional status (FIM self-care and
FTHUE) and isometric torque at follow-up, 8
months, pain relief and upper extremity sensory
function at end of treatment, 4 weeks, and fol-
low-up, 8 months (no benefit demonstrated);
grade D for functional status (FIM self-care) and
lateral pinch at end of treatment, 4 weeks, func-
tional status (FIM mobility) and motor function
upper extremity at follow-up, 8 months, upper
extremity ROM at end of treatment, 4 weeks, and
follow-up, 8 months (no benefit demonstrated
but favoring control); grade D+ for grip strength
at end of treatment, 4 weeks (clinically important
benefit favoring control demonstrated without
statistical significance). Patients with acute and
subacute stroke.
Task-specific reach training versus Bobath
approach training, level I (RCT): Grade C+ for
balance sitting at follow-up, 12 weeks; grade C+
favoring Bobath approach training for balance sit-
ting at end of treatment, 4 weeks (clinically impor-
tant benefit demonstrated without statistical sig-
nificance); grade C for balance sitting at
follow-up, 2 weeks, and balance standing at end of
treatment, 4 weeks, and follow-up, 2 and 12 weeks
(no benefit demonstrated). Patients with subacute
stroke.
Task-specific reach training versus standard
rehabilitation, level I (RCT): Grade C+ for balance
sitting at follow-up, 12 weeks, and balance standing
at follow-up, 2 and 12 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for balance standing at end of
treatment, 4 weeks (no benefit demonstrated);
grade D+ for balance sitting at end of treatment, 4
weeks, and follow-up, 2 weeks (clinically important
benefit favoring control demonstrated without sta-
tistical significance). Patients with subacute stroke.
Trunk control training versus control, level I
(RCT): Grade C+ for postural status, mobility, and
unilateral neglect at end of treatment, 1 month, and
follow-up, 2 months (clinically important benefit
demonstrated without statistical significance);
grade C for functional status at end of treatment, 1
month, and follow-up, 2 months (no benefit dem-
onstrated). Patients with subacute stroke.
Trunk rotation feedback training versus con-
trol, level I (RCT): Grade A for functional status
and unilateral neglect at end of treatment, 1
month, and follow-up, 1 month (clinically impor-
tant benefit demonstrated). Patients with subacute
stroke.
Traditional functional retraining (TFR) ap-
proach versus NDT training, level I (RCT):
Grade A (favoring TFR) for gross manual dexterity
at follow-up, 6 months and 1 year (clinically im-
portant benefit demonstrated); grade C+ (favoring
TFR) for gait speed at end of treatment, 1 month,
and follow-up, 6 months and 1 year, stride length
at follow-up, 6 months and 1 year, and functional
status (time for independence: prone-to-supine,
supine-to-prone, supine-to-sit, unsupported
standing, grooming, bathing, dressing, ambulation
without device, and feeding) at end of treatment, 1
month (clinically important benefit demonstrated

without statistical significance); grade C for func-
tional status (time for independence: unsupported
sitting) and stride length at end of treatment, 1
month, functional status (FIM), fine motor coordi-
nation, and finger dexterity at end of treatment, 1
month, and follow-up, 6 months and 1 year (no
benefit demonstrated). Patients with subacute
stroke.
Functional training (dressing practice) ver-
sus control, level I (RCT): Grade C for functional
status (Nottingham stroke dressing) at end of
treatment, 3 months (no benefit demonstrated);
grade D for functional status (Rivermead ADL self-
care) at end of treatment, 3 months (no benefit
demonstrated but favoring control). Patients with
chronic stroke.
Leisure task versus functional task (ADL self-
care) training, level I (RCT): Grade C for leisure
activities and functional status at follow-up, 6
months and 1 year (no benefit demonstrated). Pa-
tients with acute, subacute, and post-acute stroke.
Leisure task training versus control, level I
(RCT): Grade C for leisure activities at follow-up,
6 months and 1 year, and functional status at
follow-up, 6 months (no benefit demonstrated);
grade D for functional status at follow-up, 1 year
(no benefit demonstrated but favoring control).
Patients with acute, subacute, and post-acute
stroke.
Functional task (ADL self-care) training ver-
sus control, level I (RCT): Grade C for functional
status (Nottingham Extended Activities of Daily
Living [EADL]) at follow-up, 6 months and 1 year
(no benefit demonstrated); grade D for functional
status (London Handicap Scale) and leisure activi-
ties at follow-up, 6 months and 1 year (no benefit
demonstrated but favoring control). Patients with
acute, subacute, and post-acute stroke.
Functional (ADL) training versus leisure ac-
tivities, level I (RCT): Grade C for functional
status (number patients improved on Barthel In-
dex [dressing and bathing] and number patients
improved on EADL Scale [cleaning, cooking, and
mobility outside]) and leisure activities (number of
patients improved on Nottingham Leisure Ques-
tionnaire Scale [shopping, gardening, and hob-
bies]) at 6-months follow-up (no benefit demon-
strated); grade D for functional status (number
Clinical Practice Guidelines 23
patients improved on Barthel Index [bed to chair
transfers] and number of patients improved on
EADL Scale [mobility on uneven ground]) and
leisure activities (number of patients improved on
Nottingham Leisure Questionnaire Scale [cooking,
entertainment, games and sports]) at 6-months
follow-up (no benefit demonstrated but favoring
control). Patients with acute, subacute, and post-
acute stroke.
Weight garments training versus control,
level I (RCT): Grade C+ for single support time
symmetry and support base width symmetry at
end of treatment, 6 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for gait speed and cadence at end
of treatment, 6 weeks (no benefit demonstrated);
grade D+ for step length symmetry and double
support time symmetry at end of treatment, 6
weeks (clinically important benefit favoring con-
trol demonstrated without statistical significance);
grade D for balance at end of treatment, 6 weeks
(no benefit demonstrated but favoring control).
Patients with chronic stroke.
Functional approach versus transfer ap-
proach training, level I (RCT): Grade A favoring
functional approach for functional status (Edmans
ADL index) and gross motor function at end of
treatment, 6 weeks (clinically important benefit
demonstrated); grade C for resource cost percep-
tion and functional status (Barthel Index) at end of
treatment, 6 weeks (no benefit demonstrated). Pa-
tients with post-acute stroke.
Bilateral arm training versus NDT approach,
level I (RCT): Grade C+ for change in shoulder
strength, Wolf Motor Arm Test (time), and self-
reported use of paretic arm at end of treatment, 6
weeks (clinically important benefit demonstrated
without statistical significance); grade C for
change in Fugl-Meyer at end of treatment, 6 weeks
(no benefit demonstrated); grade D+ for change in
Wolf Motor Arm Test (weight) and elbow strength
at end of treatment, 6 weeks (clinically important
benefit favoring control demonstrated without sta-
tistical significance). Patients with chronic stroke.
Task-specific training versus control, level I
(RCT): Grade C+ for motor function (Fugl-Meyer
subscale: upper extremity and lower extremity
control]) and balance (Berg Scale) at end of treat-
24 TOPICS IN STROKE REHABILITATION/SPRING 2006
ment, 6 weeks (clinically important benefit dem-
onstrated without statistical significance); grade C
for balance (Fugl-Meyer subscale: balance) and
gait speed at end of treatment, 6 weeks (no benefit
Summary of trials
Sixteen RCTs (n = 963) and 1 CCT (n = 27) evaluat-
ing task-oriented training in stroke were included.
119,127,159–173
The size and methodological quality of
the trials varied widely as did the treatment sched-
ules, duration of follow-up, time of stroke onset, and
materials required (Appendix 3B).
The task-oriented training exercises examined
were specific to each trial and included reaching
and balance tasks, trunk control and trunk rota-
tion training, upper extremity training, ADLs, lei-
sure activities, and other functional tasks (Appen-
dix 3B)
In total, 22 studies were excluded (TOT-Table
37). Six of these studies lacked a control group,174–
179
four studies provided insufficient statistical
data,180–183 five studies did not examine an appro-
priate study population,184–188 and two studies did
not report any outcomes of interest. 189,190 The
remaining studies were excluded for reasons de-
tailed in TOT-Table 37. 191–195
Efficacy
In comparing seated reaching task training to
control in chronic stroke patients, statistically sig-
nificant and clinically important benefits were
shown for hand movement time (ipsilateral and
across) and peak vertical ground reaction force (for-
ward and across) at the end of 2 weeks of interven-
tion (one RCT, n = 20). 159 RDs were −32%, −42%,
42%, and 54%, respectively. Clinically important
benefits without statistical significance were shown
for hand movement time in the forward direction
(−35% RD). No clinical benefit was demonstrated
for maximum reaching distance (ipsilateral, for-
ward, across) and peak vertical ground reaction
force (ipsilateral) (see TOT-Figure 34, Table 38).
No long-term follow-up was performed.
One RCT (n = 26)168 investigated the effect of
adapted game training versus control in post-
acute stroke patients with pronator spasticity. The
demonstrated); grade D for functional status
(Barthel Index) at end of treatment, 6 weeks (clini-
cal benefits favoring control). Patients with sub-
acute stroke.
adapted dice game was designed to challenge the
patient’s ability to perform functional forearm su-
pination. A statistically significant and clinically
important benefit with regard to degree of forearm
supination was found at the end of the 1-day
intervention (15% RD; TOT-Figure 35, Table
39). No other outcome measures were assessed.
The only included CCT (n = 27)161 showed no
statistically significant or clinically important out-
comes for repetitive elbow joint movement
training compared to control in subacute stroke
(TOT-Figure 36, Table 40). Specifically, no clini-
cal benefit was found for maximal joint extension
velocity (to target and to mouth) or motor function
(Fugl-Meyer Scale) at the end of 3 weeks of treat-
ment. Results favored control for maximal joint
flexion velocity (to target and to mouth), motor
function (Frenchay Arm Test), and functional sta-
tus (Barthel Index). RDs were −1%, −9%, −31%,
and −5%, respectively. The RD of −31% for the
Frenchay Arm Test suggested a clinically relevant
advantage for the control intervention.
Kwakkel et al.164,165 examined the effect of func-
tional extremity training versus control in sub-
acute stroke, finding statistically significant and
clinically important benefits for quality of life
(NHP) at the end of 12 weeks of treatment (−32%
RD). A clinically important effect without statisti-
cal significance was shown for the same outcome
measure at the 6-week and 32-week follow-ups
(18% and 22% RD, respectively), but no clinical
benefit was found at the 18-week follow-up (TOT-
Figure 37, Table 41). Another quality of life mea-
sure, the Sickness Impact Profile, yielded no clini-
cally beneficial results at any time point. Outcomes
for comfortable gait speed revealed clinically im-
portant benefits without statistical significance at
the end of treatment (12 weeks, 20 weeks) and at
6-week follow-up (40%, 40%, and 22% RD, re-
spectively), but no benefit at the 18-week and 32-
week follow-ups. Clinically important effects with-
out statistical significance were shown for

maximum gait speed at the end of treatment (12
weeks, 20 weeks) and at the 6-week and 32-week
follow-ups (30%, 38%, 25%, and 22% RD, respec-
tively), but not at the 18-week follow-up. No ben-
efit was demonstrated for functional status (FAI) at
the 6-week and 32-week follow-ups. With respect
to mobility (number of patients using walking
aids), no benefit was found at any time point and a
clinically unimportant trend favoring control was
seen at the end of 20 weeks of treatment.
In patients with acute and subacute stroke, no
statistically significant differences were found
comparing functional task training of the upper
extremity with control for any of the outcomes
measured (one RCT, n = 60) (TOT-Figure 38,
Table 42).127 Clinically important benefits without
statistical significance, however, were observed for
functional status (FTHUE) and isometric torque
and motor function at the end of the 4-week treat-
ment, for grip strength and lateral pinch at the 8-
month follow-up, and for palmar pinch at the end
of the 4-week treatment and at the 8-month follow-
up (28%, 45%, 35%, 396%, 110%, 65%, and 202%
RD, respectively). No clinical benefit was demon-
strated for functional status (FTHUE) and isometric
torque at 8 months follow-up or for pain relief and
upper extremity sensory function at the end of treat-
ment and at 8 months follow-up. Clinical benefits
were not seen in two other outcomes: FIM mobility
at the end of treatment and FIM self-care at 8
months follow-up. Outcomes favoring control were
shown for FIM self-care and lateral pinch at the end
of treatment, motor function of the upper extremity
and FIM mobility at 8 months follow-up, and upper
extremity ROM both at the end of treatment and at
follow-up. Grip strength measurement at the end of
treatment revealed a clinically relevant benefit fa-
voring control (−17% RD).
For task-specific reach training versus Bobath
approach training in subacute stroke patients, no
statistically significant differences were found (one
RCT; n = 40).119 Clinically important effects with-
out statistical significance favoring task-related
reach were only observed for sitting balance at 12
weeks follow-up (−52% RD; TOT-Figure 39,
Table 43). There was no benefit for sitting balance
at 2 weeks follow-up or for standing balance at any
time point. A clinically relevant effect favoring
Clinical Practice Guidelines 25
Bobath approach training was observed for sitting
balance at the end of the 4-week treatment in
contrast to at follow-up (20% RD).
The same RCT119 compared task-specific reach
training to standard rehabilitation and, again,
found no statistically significant differences be-
tween groups (TOT-Figure 40, Table 44). Still,
clinically important benefits favoring task-specific
reach training were shown for standing balance at
the 2-week and 12-week follow-ups and for sitting
balance at the 12-week follow-up. There was no
benefit for standing balance at the end of treat-
ment. Clinically relevant effects favoring control
were demonstrated for sitting balance at the end of
treatment and at the 2-week follow-up (20% and
18% RD, respectively).
For trunk control training using the Bon Saint
Come device compared to standard rehabilita-
tion control in subacute stroke patients, one RCT
(n = 20)160 showed no statistically significant differ-
ences (TOT-Figure 41, Table 45). Clinically im-
portant effects without statistical significance were
found for postural status (trunk control, upright
equilibrium, sitting equilibrium), mobility (func-
tional ambulation classification), and unilateral
neglect (Bell Test) at the end of treatment (4
weeks) and at 2 months follow-up. RDs were 31%,
63%, 80%, 100%, 36%, 29%, 244%, 22%, −41%,
and −24%, respectively, with end of treatment
scores paired with follow-up scores for each out-
come measure. No benefits were shown for func-
tional status (FIM) at the end of treatment or at 2
months follow-up.
In one RCT (n = 22),173 trunk rotation feedback
training in subacute stroke patients with severe
unilateral neglect syndrome produced both clini-
cally and statistically significant benefits com-
pared to control in all outcomes measured at the
end of treatment and at 1-month follow-up (TOT-
Figure 42, Table 46). RDs for functional status
(FIM) and unilateral neglect (Schekenberg Test,
Albert Test, Bell Test) were 20%, 20%, −49%,
−39%, −40%, −33%, −50%, and −50%, respec-
tively, with end of treatment and follow-up values
paired for each outcome measure.
A comparison of traditional functional retrain-
ing versus NDT training in subacute stroke pa-
tients demonstrated a statistically and clinically
26 TOPICS IN STROKE REHABILITATION/SPRING 2006
significant benefit for gross manual dexterity at the
6-month and 12-month follow- ups (97% and
94% RD, respectively; one RCT, n = 27)163 (TOT-
Figure 43, Table 47; see Appendix 3B for treat-
ment description). The remaining outcomes did
not show statistical significance. Clinically impor-
tant benefits without statistical significance were
observed for gait speed at the end of 1 month of
treatment and at the 6-month and 12-month fol-
low-ups, stride length at the 6-month and 12-
month follow-ups, and time for independence
in performing ADLs (prone-to-supine, supine-
prone, supine-to-sit, unsupported standing,
grooming, bathing, dressing, ambulation with-
out device, and feeding) at the end of treatment
(16%, 125%, 144%, 28%, 56%, −39%, −36%,
−26%, −84%, −47%, −19%, −48%, −43%, and
−84% RD, respectively). No benefits were
shown for stride length, gross manual dexter-
ity, and time for independence (unsupported
sitting, bathing, dressing, ambulation without
a device) at the end of treatment or for fine
motor coordination and finger dexterity and
functional status (FIM) at any time point.
One RCT (n = 30)172 evaluating functional
training (dressing practice) in chronic stroke pa-
tients showed no statistically significant differ-
ences compared to control (TOT-Figure 44,
Table 48). No clinical benefit was observed for
one measure of functional status (Nottingham
stroke dressing) at the end of the 3-month inter-
vention. A clinically relevant advantage favoring
control was demonstrated for another functional
status measure (Rivermead ADL self-care) at the
end of treatment (−27% RD).
Parker et al.169 investigated the effects of leisure
task training, functional task training (ADL self-
care), and no occupational therapy on three out-
come measures in acute, subacute, and post-acute
stroke patients in one RCT (n = 466). There was no
benefit of leisure task training compared to func-
tional task training (ADL self-care) with regard to
the results of the Nottingham Leisure Question-
naire, the Nottingham EADL, and the London
Handicap Scale at 6 months and 12 months follow-
up (TOT-Figure 45, Table 49). Furthermore, no
benefits were demonstrated for leisure task training
versus control or for functional task training (ADL
self-care) versus control for any of the reported
outcomes. For these same comparisons, some of the
results favored control but were without statistical
significance or clinical importance (TOT-Figures
46&47, Tables 50&51).
Logan et al.166 compared functional task train-
ing (ADL) to leisure task training in acute, sub-
acute, and post-acute stroke patients (TOT-Figure
48, Table 52). No benefit was shown for number
of patients improved on the EADL scale (cleaning,
cooking, mobility outside), number of patients im-
proved on the Barthel Index (dressing, bathing), or
number of patients improved on the Nottingham
Leisure Questionnaire (shopping, gardening, hob-
bies) at 6 months follow-up. A statistically insig-
nificant and clinically unimportant trend favoring
control was shown for number of patients im-
proved on the EADL scale (mobility on uneven
ground), number of patients improved on the
Barthel Index (bed to chair transfers), and number
of patients improved on the Nottingham Leisure
Questionnaire (cooking, entertainment, games,
sport) at 6 months follow-up.
The effect of weight garment training com-
pared to control on balance and gait in chronic
stroke patients was assessed by one RCT (n =
24).170 No statistically significant differences were
found. A clinically important benefit without sta-
tistical significance was shown for single support
time symmetry and support base width sym-
metry at the end of 6 weeks of treatment (−25%
and −35% RD, respectively; TOT-Figure 49,
Table 53). No benefit was shown for gait speed or
cadence at the end of treatment. Trends favoring
control were demonstrated for balance (Berg
Scale), step length symmetry, and double support
time symmetry, of which the latter two outcomes
were clinically relevant (-26% and 20% RD, re-
spectively). No long-term follow-up was done.
Functional approach training in post-acute
stroke patients (one RCT, n = 79)162 produced statis-
tically and clinically significant benefits compared
to transfer approach training with regard to
functional status (Edmans ADL index) and gross
motor function (Rivermead Motor Assessment) at
the end of 6 weeks of treatment (24% and 200%
RD, respectively; TOT-Figure 50, Table 54). No
benefit was observed for the Rivermead Perceptual

Assessment battery, the Barthel Index measure of
functional status, or resource cost (length of stay).
No long-term follow-up was carried out.
No statistically significant outcomes were found
for bilateral arm task training in chronic stroke
patients versus a neurodevelopmental ap-
proach (one RCT, n = 21).167 Clinically important
benefits without statistical significance were dem-
onstrated for change in shoulder strength, the
Wolf Motor Arm Test (time), and self-reported use
of paretic arm at the end of 6 weeks of treatment
(59%, −100%, and 16% RD, respectively; TOT-
Figure 51, Table 55). No benefit was found for
change in motor function (Fugl-Meyer). Notably,
changes in elbow strength and the Wolf Motor
Arm Test (weight) favored control to a clinically
relevant degree (−123% and −50% RD, respec-
tively). No long-term follow-up was done.
No statistically significant differences were dem-
onstrated between task-specific training in sub-
acute stroke patients and control for any outcome
measures at the end of treatment (one RCT, n =
27).171 Clinically important effects without statisti-
cal significance were shown for motor function
(Fugl-Meyer subscale: upper extremity and lower
extremity control) and balance (Berg Scale) (43%,
25%, and 16% RD, respectively; TOT-Figure 52,
Table 56). No benefit was found for balance (Fugl-
Meyer subscale: balance) and gait speed. A statisti-
cally insignificant and clinically unimportant trend
was seen favoring control for functional status
(Barthel Index). No long-term follow-up was done.
Strength of the published evidence compared with
other guidelines
The Ottawa Panel found level I (RCTs) and level
II (CCT) evidence evaluating various types of task-
oriented training in post-stroke rehabilitation.
Similarly, the recommendations of the VA/DoD,75
the Scottish Intercollegiate Guidelines Network,74
the Heart and Stroke Foundation of Ontario,71 St.
Joseph’s Health Care London,73 and the Stroke
Prevention and Educational Awareness Diffusion
Collaboration76 were also based on level I evidence
(RCTs). The AHCPR70 found level II evidence
(non-RCTs) to support its recommendations re-
garding task-oriented training (Appendix 4).
Clinical Practice Guidelines 27
Clinical recommendations compared with other
guidelines
The Ottawa Panel found good evidence to rec-
ommend considering task-oriented training as an
approach to stroke rehabilitation. While the find-
ings were not consistent for all outcome measures,
there is sufficient support (grade A, C+) for the
following interventions: seated reaching task train-
ing, adapted game training (forearm supination
task for stroke patients with pronator spasticity),
functional upper extremity training, functional
task training for the upper extremity, task-specific
reach training, trunk control training, trunk rota-
tion training, traditional functional retraining,
weight garment training, functional approach
training, bilateral arm training, and task-specific
training. It is unclear whether to include or ex-
clude leisure task training as a rehabilitation strat-
egy based on the available evidence (grade C). The
evidence for recommending dressing practice and
repetitive elbow movement training was found to
be poor (grade D; Appendix 4)
The Ottawa Panel guidelines are in agreement
with existing clinical practice guidelines. In-
deed, the Scottish Intercollegiate Guidelines
Network74 supported the use of task-specific
training to improve task performance. Further-
more, AHCPR,70 the Stroke Prevention and
Educational Awareness Diffusion Collabora-
tion,76 and the Heart and Stroke Foundation of
Ontario77 all encouraged functional task training
for stroke patients with functional deficits in
their affected limbs. The Ottawa Panel guide-
lines are also in concordance with St. Joseph’s
Health Care London,73 which concluded that the
benefit of leisure therapy post-stroke is uncer-
tain. The VA/DoD75 stated that there is insuffi-
cient evidence to recommend for or against NDT
compared to other approaches for motor re-
training post stroke. Similarly, while the Ottawa
Panel found some evidence (grade C+) that NDT
is inferior to traditional functional retraining,
the neurodevelopmental approach was shown to
be either more beneficial (grade D+), no differ-
ent (grade C), or less beneficial (grade C+) than
bilateral arm training depending on the specific
outcome measure (Appendix 4).
28 TOPICS IN STROKE REHABILITATION/SPRING 2006
Clinical practice guidelines for biofeedback
EMG-BFB training versus control, level I
(RCT): Grade C+ for motor function at end of
treatment, 5 weeks (clinically important benefit
demonstrated without statistical significance);
grade D for physical abilities at end of treatment, 5
weeks (no benefit demonstrated); grade D+ for
physical abilities at follow-up, 9 months, and mo-
tor function at follow-up, 9 months (clinically im-
portant benefit favoring control demonstrated
without statistical significance). Patients with post-
acute stroke.
EMG-BFB training for the wrist versus con-
trol, level II (CCT): Grade B for grip strength at
follow-up, 3 months, and motor function at end of
treatment, 3 months, and at follow-up, 3 and 9
months (clinically important benefit demon-
strated); grade C+ for grip strength at end of
treatment, 3 months (clinically important benefit
demonstrated without statistical significance);
grade D+ for grip strength at follow-up, 9 months
(clinically important benefit favoring control dem-
onstrated without statistical significance). Patients
with chronic stroke.
EMG-BFB relaxation training for the shoul-
der versus control, level I (RCT): Grade C+ for
pain relief at end of treatment, 2 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for active shoulder ROM at
end of treatment, 2 weeks (no benefit demon-
strated). Patients with subacute and post-acute
stroke.
EMG-BFB training for the upper extremity
versus control, level II (CCT): Grade A for motor
function (time to make a circle with tip of olecra-
non) at end of treatment, 6 months (clinically
important benefit demonstrated); grade C+ for
motor function (time to abduct shoulder to 90º) at
end of treatment, 6 months. Patients with chronic
stroke.
EMG-BFB training for the lower extremity
versus control, level I (RCT) and level II (CCT):
Grade A for change in timed ambulation on car-
peted surface at end of treatment, 4 weeks (clini-
cally important benefit demonstrated); grade B
for ankle ROM (change in active ankle dorsiflex-
ion) and knee ROM (change in active knee flex-
ion) at end of treatment, 3 months, ankle strength
at follow-up, 6 weeks, (clinically important ben-
efit demonstrated); grade C+ for knee ROM (de-
crease of hyperextension) at end of treatment, 4
weeks, and follow-up, 4 weeks, gait speed at end
of treatment, 4 weeks and 2 months, and follow-
up, 4 weeks, gait quality at end of treatment, 2
months, gait cycle time at end of treatment, 6
weeks, mobility at end of treatment, 3 months,
change in timed ambulation on smooth surface
and improvement in ankle angle at heel strike at
end of treatment, 4 weeks, and ankle strength at
end of treatment, 4 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for ankle ROM (active ankle
ROM) at end of treatment, 6 weeks, and follow-
up, 2 months, ankle angle at end of treatment, 6
weeks and 2 months, step length at end of treat-
ment, 4 and 6 weeks, and follow-up, 3 months,
stride width at follow-up, 3 months, decrease of
spasticity at end of treatment, 2 months, quality
of life at end of treatment, 2 months, functional
status at end of treatment, 2 months (no benefit
demonstrated); grade D+ for improvement in
lower extremity support asymmetry at end of
treatment, 4 weeks, ankle angle (improvement in
ankle angle at toe-off at end of treatment, 4 weeks
(clinically important benefit favoring control
demonstrated without statistical significance);
grade D for knee angle at end of treatment, 6
weeks, ankle ROM (change in active ROM) at end
of treatment, 4 weeks, stride width at end of
treatment, 6 weeks, ankle angle (ankle angle at
heel strike) at end of treatment, 2 months, step
length at end of treatment, 2 months, improve-
ment in lower extremity support asymmetry at
follow-up, 4 weeks, global neurological status at
end of treatment, 2 months (no benefit demon-
strated but favoring control). Patients with acute-
chronic stroke.
EMG-BFB training for the lower extremity
versus placebo, level II (CCT): Grade B for ankle

strength at end of treatment, 6 weeks, and follow-
up, 6 weeks, gait quality and ankle ROM at follow-
up, 6 weeks (clinically important benefit demon-
strated); grade C+ for gait quality and ankle ROM
at end of treatment, 6 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for step length and stride width at
end of treatment, 4 weeks, and follow-up, 3
months, and active ankle ROM at end of treatment,
6 weeks, and follow-up, 3 months (no benefit
demonstrated). Patients with subacute-chronic
stroke.
EMG-BFB training for the shoulder versus
placebo, level I (RCT): Grade C+ for upper ex-
tremity motor function at end of treatment, 6
weeks (clinically important benefit demonstrated
without statistical significance); grade C for upper
extremity motor function at follow-up, 6 weeks
(no benefit demonstrated). Patients with subacute
and post-acute stroke.
EMG-BFB training for the elbow versus pla-
cebo, level I (RCT): Grade C+ for passive elbow
ROM at end of treatment, 3 weeks (clinically im-
portant benefit demonstrated without statistical
significance); grade D for active elbow ROM at
end of treatment, 3 weeks (no benefit demon-
strated but favoring control). Patients with chronic
stroke.
EMG-BFB training for hand versus placebo,
level I (RCT): Grade C+ for functional status at
end of treatment, 4 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for motor recovery at end of treat-
ment, 4 weeks (no benefit demonstrated); grade
D+ for wrist ROM at end of treatment, 4 weeks.
Patients with post-acute stroke.
Rhythmic positional BFB training versus con-
trol, level I (RCT): Grade A for ankle ROM at end
of treatment, 12 weeks, and gait speed at end of
treatment, 12 weeks, and follow-up, 3 months (no
benefit demonstrated). Patients with chronic
stroke.
Audio and visual feedback training with gen-
eral relaxation versus control, level II (CCT):
Clinical Practice Guidelines 29
Grade C+ for active knee ROM at end of treat-
ment, 3 months (clinically important benefit dem-
onstrated without statistical significance); grade
D+ for active ankle ROM and mobility at end of
treatment, 3 months (clinically important benefit
favoring control demonstrated without statistical
significance). Patients with chronic stroke.
Video feedback training versus control, level
II (CCT): Grade B for decrease of unilateral ne-
glect (baking tray task) at end of treatment, 2 days
(clinically important benefit demonstrated); grade
C for decrease of unilateral neglect (line bisection
task) at end of treatment, 2 days (no benefit dem-
onstrated); grade D for decrease of unilateral ne-
glect (line cancellation task) at end of treatment, 2
days (no benefit demonstrated but favoring con-
trol). Patients with subacute stroke.
BFB training versus FES, level I (RCT): Grade
C+ for stride length at end of treatment, 6 weeks
(clinically important benefit demonstrated without
statistical significance); grade C for ankle and
knee angle and gait cycle time at end of treatment,
6 weeks (no benefit demonstrated). Patients with
chronic stroke.
Force-feedback training for upper extremity
versus control, level I (RCT): Grade C for motor
function at end of treatment, 6 weeks, and follow-
up, 8 weeks, and dexterity at follow-up, 8 weeks
(no benefit demonstrated); grade D for coordina-
tion at end of treatment, 6 weeks, and follow-up, 8
weeks, and for dexterity at end of treatment, 6
weeks (no benefit demonstrated but favoring con-
trol). Patients with chronic stroke.
Force-feedback training for lower extremity
versus control, level I (RCT): Grade C+ for gait
speed at end of treatment, 6 weeks, and gait endur-
ance (2-minute walking test) at follow-up, 8 weeks
(clinically important benefit demonstrated without
statistical significance); grade C for motor func-
tion and mobility (Timed Up & Go) at end of
treatment, 6 weeks, and follow-up, 8 weeks, gait
speed at follow-up, 8 weeks, and gait endurance at
end of treatment, 6 weeks (no benefit demon-
strated). Patients with chronic stroke.
30 TOPICS IN STROKE REHABILITATION/SPRING 2006
Summary of trials
Twelve RCTs and seven CCTs were found that
evaluated the efficacy of various BFB techniques in
the rehabilitation of acute to chronic post-stroke
patients (n = 444).16,50,51,196–210 The following treat-
ments were evaluated: EMG-BFB training (n =
42)197,198; EMG-BFB training for the upper extrem-
ity (n = 31),210 for the shoulder (n = 40),202 elbow (n
= 16),211 wrist (n = 18),204 and hand (n = 27)196; and
EMG-BFB training for the lower extremity (n =
68)16,50,51,199,203,206,209 versus control or placebo;
EMG-BFB relaxation training for the shoulder (n =
20)208; rhythmic positional BFB training (n =
37)205; audio and visual feedback training with
general relaxation (n = 31)210; video feedback train-
ing (n = 14)207; force-feedback training for the
upper and lower extremities (n = 25)200 versus a
control treatment; and BFB training versus FES (n
= 36).51 Treatments were given 2 to 5 times per
week for 20 to 60 minutes per session, for a total
duration of treatment of 2 days to 6 months, some-
times with follow-up at 2 weeks to 12 months
(EMG-BFB-Appendix 3C).
In total, 42 studies were excluded for the follow-
ing reasons: there were insufficient statistical data
in nine studies,212–220 the patient served as his own
control in four studies,221–224 there was no control
group for four other studies,178,225–227 while in an-
other four studies there were fewer than five sub-
jects per group.52,228–230 In three of the studies,
healthy subjects were used,231–233 while no out-
comes of interest were reported in three of the other
studies.234–236 Three studies were meta-analy-
ses,46,47,237 while two studies were not clinical tri-
als45,238 and one study lacked standard deviations.239
Finally, eight other studies were excluded for vari-
ous other reasons240–247 (EMG-BFB-Table 57).
Efficacy
For EMG-BFB training for patients with post-
acute stroke versus control (two RCTs, n =
42),197,198 clinically important benefits without sta-
tistical significance were found for motor function
(Upper Extremity Function Test) at end of treat-
ment, 5 weeks (61% RD). No benefit was demon-
strated, but results favored the control treatment
for physical abilities (Finger Oscillation Test) at
end of treatment, 5 weeks. A clinically important
benefit favoring control was demonstrated without
statistical significance for physical abilities at fol-
low-up, 9 months, and motor function at follow-
up, 9 months (EMG-BFB-Figure 53, Table 58).
No other outcomes were measured.
For EMG-BFB training for the wrist of pa-
tients with chronic stroke versus control (one
CCT, n = 18),204 clinically important benefits with
statistical significance were shown for grip
strength at follow-up, 3 months (39% RD), and
motor function (Fugl-Meyer) at end of treatment,
3 months, and follow-up, 3 and 9 months (31%–
44% RD). A clinically important benefit was dem-
onstrated without statistical significance for grip
strength at end of treatment, 3 months (25% RD).
A clinically important benefit favoring control for
grip strength was demonstrated without statistical
significance at follow-up, 9 months (EMG-BFB-
Figure 54, Table 59). No other outcomes were
measured.
For EMG-BFB relaxation training for the
shoulder for patients with subacute and post-
acute stroke versus control (one RCT, n = 20),208
clinically important benefit without statistical sig-
nificance were observed for pain relief (McGill
Pain Questionnaire) at end of treatment, 2 weeks
(18% RD). No benefit was demonstrated for active
shoulder ROM (flexion and abduction) at end of
treatment, 2 weeks (EMG-BFB-Figure 55, Table
60). No other outcomes were measured.
For EMG-BFB training for the upper extrem-
ity of patients with chronic stroke versus con-
trol (one CCT, n = 31),210 a clinically important
benefit with statistical significance was demon-
strated for motor function (time to make a circle
with tip of olecranon) at end of treatment, 6
months (912% RD). A clinically important benefit
without statistical significance was demonstrated
for motor function (time to abduct shoulder to
90º) at end of treatment, 6 months (65% RD;
EMG-BFB-Figure 56, Table 61). No other out-
comes were measured.
For EMG-BFB training for the lower extrem-
ity for patients with acute-chronic stroke versus
control (four RCTs and five CCTs, n =
205),50,51,199,203,206,209 a clinically important benefit

with statistical significance was demonstrated for
change in timed ambulation on carpeted surface at
end of treatment, 4 weeks (650% RD), and ankle
ROM (change in active ankle dorsiflexion) and
knee ROM (change in active flexion) at end of
treatment, 3 months (113% and 225% RD, respec-
tively). A clinically important benefit was observed
without statistical significance for knee ROM (de-
crease of hyperextension) at end of treatment, 4
weeks, and follow-up, 4 weeks (25% and 47% RD,
respectively); gait speed at end of treatment, 4
weeks and 2 months, and follow-up, 4 weeks
(29%–49% RD); gait quality (Basmajian Scale) at
end of treatment, 2 months (34% RD); gait cycle
time at end of treatment, 6 weeks (20% RD); mo-
bility (number of patients improved in assistive
devices necessary) at end of treatment, 3 months
(50% RD); and change in timed ambulation on
smooth surface and improvement in ankle angle at
heel strike at end of treatment, 4 weeks (752% and
960% RD, respectively). No benefit was demon-
strated for ankle ROM (active ankle ROM) at end
of treatment, 6 weeks, and follow-up, 2 months;
ankle angle (at swing phase) at end of treatment, 6
weeks and 2 months; step length at end of treat-
ment, 4 and 6 weeks, and follow-up, 3 months;
decrease in spasticity (Ashworth Scale) at end of
treatment, 2 months; quality of life (Canadian
Neurological Scale) at end of treatment, 2 months;
and functional status (Barthel Index) at end of
treatment, 2 months. A clinically important benefit
favoring control was demonstrated without statis-
tical significance for improvement in lower ex-
tremity support asymmetry (single lower extrem-
ity support) at end of treatment, 4 weeks, and for
ankle angle (improvement in ankle angle at toe-
off) at end of treatment, 4 weeks. No benefit was
demonstrated but results favored control for knee
angle (at swing phase) at end of treatment, 6
weeks; ankle ROM (change in active ROM) at end
of treatment, 4 weeks; stride width at end of treat-
ment, 6 weeks; ankle angle (ankle angle at heel
strike) at end of treatment, 2 months; step length
at end of treatment, 2 months; improvement in
lower extremity support asymmetry (single lower
extremity support) at follow-up, 4 weeks; and glo-
bal neurological status (Adams Scale) at end of
treatment, 2 months (EMG-BFB-Figures 57A-H,
Clinical Practice Guidelines 31
Tables 62A-I). No other outcomes were mea-
sured.
For EMG-BFB training of the lower extremity
in patients with subacute-chronic stroke versus
placebo (two CCTs, n = 35),50,201 clinically impor-
tant benefits with statistical significance were
shown for ankle strength (number of patients im-
proved in dorsiflexors strength) at end of treat-
ment, 6 weeks, and follow-up, 6 weeks (45% and
64% RD); and gait quality (number of patients
improved) and ankle ROM (number of patients
improved in dorsiflexion) at follow-up, 6 weeks
(45% and 45% RD, respectively). A clinically im-
portant benefit was demonstrated without statisti-
cal significance for gait quality (number of patients
improved) and ankle ROM (number of patients
improved in dorsiflexion) at end of treatment, 6
weeks (18% and 36% RD, respectively). No ben-
efit was demonstrated for step length and stride
width at end of treatment, 4 weeks, and follow-up,
3 months; and active ankle ROM at end of treat-
ment, 6 weeks, and follow-up, 3 months (EMG-
BFB-Figure 58, Table 63). No other outcomes
were measured.
For EMG-BFB training of the shoulder for
patients with subacute and post-acute stroke ver-
sus placebo, one RCT (n = 40)202 showed clinically
important benefits without statistical significance
for upper extremity motor function (Action Re-
search Arm Test and Brunnström-Fugl-Meyer
Test) at end of treatment, 6 weeks (64% and 27%
RD, respectively). No benefit was demonstrated
for upper extremity motor function (Action Re-
search Arm Test and Brunnström-Fugl-Meyer
Test) at follow-up, 6 weeks (EMG-BFB-Figure 59,
Table 64). No other outcomes were measured.
For EMG-BFB training of the elbow for pa-
tients with chronic stroke versus placebo (one
RCT, n = 16),211 clinically important benefits with-
out statistical significance were observed for pas-
sive elbow ROM (extension) at end of treatment, 3
weeks (24% RD). No benefit was demonstrated,
but results favored control for active elbow ROM
(extension) at end of treatment, 3 weeks (EMG-
BFB-Figure 60, Table 65). No other outcomes
were measured.
For EMG-BFB training of the hand in patients
with post-acute stroke versus placebo, one RCT
32 TOPICS IN STROKE REHABILITATION/SPRING 2006
(n = 27)196 illustrated clinically important benefits
without statistical significance for functional status
(scale for judging the performance of drinking
from a glass) at end of treatment, 4 weeks (21%
RD). No benefit was demonstrated for motor re-
covery (Brunnstrom’s stages of recovery) at end of
treatment, 4 weeks. A clinically important benefit
favoring control was demonstrated without statis-
tical significance for wrist ROM (active wrist exten-
sion) at end of treatment, 4 weeks (EMG-BFB-
Figure 61, Table 66). No other outcomes were
measured.
For rhythmic positional BFB training for pa-
tients with chronic stroke versus control (one
RCT, n = 37),205 a clinically important benefit with
statistical significance was shown for ankle ROM at
end of treatment, 12 weeks (28% RD), and gait
speed at end of treatment, 12 weeks, and follow-
up, 3 months (62% and 71% RD, respectively;
EMG-BFB-Figure 62, Table 67). No other out-
comes were measured.
For audio and visual feedback training with
general relaxation for patients with chronic stroke
versus control, one CCT (n = 31)210 showed a
clinically important benefit without statistical sig-
nificance for active knee ROM (change in active
flexion) at end of treatment, 3 months (2955% RD).
However, it also demonstrated a clinically impor-
tant benefit without statistical significance favoring
control for active ankle ROM (change in active dor-
siflexion) and mobility (number of patients im-
proved in assistive devices necessary) at end of treat-
ment, 3 months (EMG-BFB-Figure 63, Tables
68A&B). No other outcomes were measured.
For video feedback training for patients with
subacute stroke versus control (one CCT, n =
14),207 clinically important benefits with statistical
significance were found for decrease in unilateral
neglect (baking tray task) at end of treatment, 2
days (44% RD). No benefit was demonstrated for
unilateral neglect (line bisection task) at end of
treatment, 2 days. No benefit was demonstrated,
but results favored control for decrease in unilat-
eral neglect (line cancellation task) at end of treat-
ment, 2 days (EMG-BFB-Figure 64, Table 69).
No other outcomes were measured.
For FES versus BFB for patients with chronic
stroke (one RCT, n = 36),51 clinically important
benefits without statistical significance were ob-
served for stride length at end of treatment, 6
weeks (22% RD). No benefit was demonstrated for
ankle and knee angle (at swing phase) and gait
cycle time at end of treatment, 6 weeks (EMG-
BFB-Figure 65, Table 70). No other outcomes
were measured.
For force-feedback training for the upper ex-
tremity for patients with chronic stroke versus
control, (one RCT, n = 25),200 no benefit was
shown for motor function (Fugl-Meyer upper ex-
tremity and Test Évaluant la Performance des
Membres supérieurs des Personnes âgées
[TEMPA]) at end of treatment, 6 weeks, and fol-
low-up, 8 weeks, and dexterity (box and blocks) at
follow-up, 8 weeks. No benefit was demonstrated,
but results favored control for coordination (fin-
ger-to-nose) at end of treatment, 6 weeks, and
follow-up, 8 weeks, and dexterity at end of treat-
ment, 6 weeks (EMG-BFB-Figure 66, Table 71).
No other outcomes were measured.
For force-feedback training for the lower ex-
tremity for patients with chronic stroke versus
control, one RCT (n = 25)200 demonstrated clinically
important benefits without statistical significance
for gait speed at end of treatment, 6 weeks (24%
RD), and gait endurance (2-minute walking test) at
follow-up, 8 weeks (15% RD). No benefit was dem-
onstrated for motor function (Fugl-Meyer lower ex-
tremity), mobility (Timed Up & Go) at end of treat-
ment, 6 weeks, and follow-up, 8 weeks; gait speed
at follow-up, 8 weeks; and gait endurance at end of
treatment, 6 weeks (EMG-BFB-Figure 67, Table
72). No other outcomes were measured.
Strength of published evidence compared with other
guidelines
The Ottawa Panel found level I (12 RCTs) and
level II evidence (7 CCTs) concerning the use of
BFB in post-stroke rehabilitation. In comparison,
the VA/DoD75 found level I evidence (3 RCTs),
while the Royal College of Physicians72 found both
level I and II evidence (4 meta-analyses, 22 RCTs,
and 2 CCTs) for the use of BFB in post-stroke
rehabilitation (Appendix 4).
The Ottawa Panel found level I (one RCT) for
EMG-BFB relaxation training for the shoulder and

level I evidence (one RCT) for EMG-BFB training
for the elbow for patients in post-stroke rehabilita-
tion. The Panel also found level II evidence for
both audio and visual feedback training with gen-
eral relaxation (one CCT) and video feedback
training (one CCT). The Scottish Intercollegiate
Guidelines Network74 evidence was only based on
four meta-analyses for EMG-BFB for outcomes of
function and movement (Appendix 4).
The Ottawa Panel found level I evidence (two
RCTs) for EMG-BFB training, as well as for EMG-
BFB training of the shoulder (one RCT). Level I
evidence (one RCT) was also found for EMG-BFB
training of the hand and force-feedback training of
the upper extremity (one RCT). The Panel found
level II evidence (one CCT) for both EMG-BFB
training of the wrist and the upper extremity (one
CCT). The Heart and Stroke Foundation of
Ontario71 also found level I and II evidence (one
case series, three CCTs, and one RCT) for motor
impairment and functional motor recovery out-
comes (Appendix 4).
The Ottawa Panel found level I (one RCT) for
rhythmic positional BFB training, BFB training
versus FES (one RCT), and force-feedback training
for the lower extremity (one RCT). The Panel also
found level II evidence (two CCTs) for EMG-BFB
training of the lower extremity and level I and II
evidence (four RCTs and five CCTs) for EMG-BFB
training for the lower extremity. In comparison, St.
Joseph’s Health Care London73 found level I evi-
dence (nine RCTs) for gait and standing outcomes
(Appendix 4).
Clinical recommendations compared with other
guidelines
The Ottawa Panel agrees with the VA/DoD75 that
found insufficient evidence concerning the routine
use of BFB as an intervention for patients undergoing
post-stroke rehabilitation. However, the Panel dis-
agrees with the Royal College of Physicians that BFB
systems should be routinely excluded as an interven-
tion for post-stroke rehabilitation (Appendix 4).
The Ottawa Panel agrees with the Scottish Inter-
collegiate Guidelines Network74 that there is fair
evidence that EMG-BFB should not be used rou-
tinely in the rehabilitation of function and move-
Clinical Practice Guidelines 33
ment in patients undergoing stroke rehabilitation.
The Ottawa Panel found the following therapies
beneficial: EMG-BFB relaxation training for the
shoulder (grade C+ for pain relief) in patients with
subacute and post-acute stroke, EMG-BFB training
for the elbow (grade C+ for passive ROM) for
patients with chronic stroke, audio and visual
feedback training with general relaxation (grade
C+ for active knee ROM) for patients with chronic
stroke, and video feedback training (grade B for
decrease of unilateral neglect) for patients with
subacute stroke (Appendix 4).
The Ottawa Panel agrees with the Heart and
Stroke Foundation of Ontario71 that EMG-BFB
should be included as an intervention for post-
stroke patients who have a high level of motor
return. The Ottawa Panel found the following
treatments beneficial: EMG-BFB training (grade
C+ for motor function) for patients with post-acute
stroke, EMG-BFB training for the shoulder (grade
C+ for upper extremity motor function) for pa-
tients with subacute and post-acute stroke, EMG-
BFB training for the hand (grade C+ for functional
status) for patients with post-acute stroke, force-
feedback training for the upper extremity (grade C
for motor function and dexterity) for patients with
chronic stroke, EMG-BFB training for the wrist
(grade B for grip strength and motor function,
grade C+ for grip strength) for patients with
chronic stroke, and EMG-BFB training for the up-
per extremity (grade A and C+ for motor function)
for patients with chronic stroke (Appendix 4).
However, unlike the Heart and Stroke Founda-
tion,71 the Ottawa Panel also found control thera-
pies beneficial (grade D+ for physical abilities, mo-
tor function, wrist ROM, and grip strength;
Appendix 4).
The Ottawa Panel agrees with the St. Joseph’s
Health Care London73 guidelines that there is
strong evidence that BFB training improves gait
and standing post stroke. The Ottawa Panel found
the following treatments beneficial: rhythmic posi-
tional BFB training (grade A for ankle ROM and
gait speed) for patients with chronic stroke, BFB
training versus FES (grade C+ for stride length) for
patients with chronic stroke, force-feedback train-
ing for the lower extremity (grade C+ for gait speed
and gait endurance) for patients with chronic
34 TOPICS IN STROKE REHABILITATION/SPRING 2006
stroke, and EMG-BFB training for the lower extrem-
ity (grade A for change in timed ambulation, grade B
for ankle strength, and grade B and C+ for gait
quality and ankle ROM, grade C+ for knee ROM,
gait speed, gait cycle time, mobility, change in timed
ambulation and improvement in ankle ROM at heel
strike, and ankle strength). A control therapy (grade
D+ for improvement in lower extremity support
Clinical practice guidelines for gait training
Treadmill training and body-weight support
(BWS) combined with physiotherapy versus
treadmill training alone, level I (RCT): Grade C+
for gait speed at end of treatment, 3 weeks (clini-
cally important benefit demonstrated without sta-
tistical significance); grade C for gait speed at
follow-up, 4 months (no benefit demonstrated).
Patients with post-acute and chronic stroke.
Treadmill training and BWS versus aggres-
sive bracing-assisted walking, level I (RCT):
Grade C+ for gait endurance and length of stay at
end of treatment, 2 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for gait speed at end of treatment,
2 weeks (no benefit demonstrated). Patients with
subacute stroke.
Treadmill training versus standard custom-
ary exercises, level II (CCT): Grade B for mobility
at end of treatment, 3 weeks (clinically important
benefit demonstrated); grade C+ for gait speed
and stride length at end of treatment, 3 weeks
(clinically important benefit demonstrated without
statistical significance). Patients with subacute
stroke.
High-speed treadmill training versus low-
speed treadmill training, level I (RCT): Grade A
for fastest comfortable gait speed at end of treat-
ment, 4 weeks, and gait cadence at end of treat-
ment, 2 weeks (clinically important benefit dem-
onstrated); grade C+ for stride length at end of
treatment, 2 and 4 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for gait speed, percent body
weight support (%BWS) and upper extremity sup-
port at end of treatment, 2 and 4 weeks (no benefit
demonstrated). Patients with chronic stroke.
Low-speed treadmill training versus standard
asymmetry, and ankle angle at toe-off during gait)
for patients with acute-chronic stroke was also
found to be beneficial (Appendix 4).
All other interventions and outcomes evalu-
ated by the Ottawa Panel were not compared
because the other previously existing clinical
practice guidelines examined did not apply (Ap-
pendix 4).
gait training, level I (RCT): Grade C+ for fastest
comfortable gait speed, gait cadence, and mobility
at end of treatment, 4 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for stride length at end of treat-
ment, 2 and 4 weeks, and gait speed and gait
cadence at end of treatment, 2 weeks (no benefit
demonstrated). Patients with chronic stroke.
High-speed treadmill training versus stan-
dard gait training, level I (RCT): Grade A for
fastest comfortable gait speed and stride length at
end of treatment, 4 weeks, and gait cadence at end
of treatment, 2 and 4 weeks (clinically important
benefit demonstrated). Patients with chronic
stroke.
Treadmill training combined with overground
ambulation versus control, level I (RCT): Grade A
for step length (affected side) at end of treatment, 4
weeks, and follow-up, 3 months (clinically impor-
tant benefit demonstrated); grade C+ for gait speed
at end of treatment, 4 weeks, and follow-up, 3
months, walking distance and step length
(nonaffected side) at end of treatment, 4 weeks
(clinically important benefit demonstrated without
statistical significance); grade C for step width at
end of treatment, 4 weeks, and follow-up, 3
months, quality of life and walking distance at fol-
low-up, 3 months (no benefit demonstrated); grade
D for gait cadence at end of treatment 4 weeks, and
follow-up, 3 months, and quality of life at end of
treatment, 4 weeks (no benefit demonstrated but
favoring control). Patients with chronic stroke.
Electrogoniometric feedback training versus
control, level I (RCT): Grade C+ for change in gait
speed and maximum knee ROM (decrease of hy-
perextension) at end of treatment, 4 weeks, and
follow-up, 4 weeks (clinically important benefit

demonstrated without statistical significance). Pa-
tients with post-acute stroke.
FES of the ankle dorsiflexors versus control
for gait training, level I (RCT): Grade C+ for
functional status and level of walking indepen-
dence at end of treatment, 4 weeks, and follow-up,
4 weeks (clinically important benefit demon-
strated without statistical significance); grade D
for motor function at end of treatment, 4 weeks,
and follow-up, 4 weeks (no benefit demonstrated
but favoring control). Patients with post-acute
stroke.
FES of the ankle dorsiflexors versus control
for gait training, level I (RCT): Grade A for stride
length at end of treatment, 3 weeks (clinically
important benefit demonstrated); grade C+ for
gait cadence and gait speed at end of treatment, 3
weeks (clinically important benefit demonstrated
without statistical significance); grade C for motor
function at end of treatment, 3 weeks (no benefit
demonstrated). Patients with chronic stroke.
FES of the ankle dorsiflexors and/or gastroc-
nemius versus control for gait training, level I
(RCT): Grade A for physiological cost index (PCI)
with FES (orthosis on) and decrease of spasticity
(area beneath curve) at end of treatment, 12 weeks
(clinically important benefit demonstrated); grade
C+ for gait cycle time at end of treatment, 6 weeks,
PCI with FES (orthosis on) at end of treatment, 4
weeks (clinically important benefit demonstrated
without statistical significance); grade C for gait
speed with FES (orthosis on), PCI without FES
(orthosis off), and decrease of spasticity (relaxation
index) at end of treatment, 4 and 12 weeks, de-
crease of spasticity (area beneath curve) at end of
treatment, 4 weeks (no benefit demonstrated);
grade D for stride length and ankle and knee angle
at end of treatment, 6 weeks, gait speed without
FES at end of treatment, 4 and 12 weeks (no
benefit demonstrated but favoring control). Pa-
tients with chronic stroke.
BFB followed by FES versus control for gait
training, level I (RCT): Grade A for gait cycle time
at end of treatment, 6 weeks (clinically important
benefit demonstrated); grade C for stride length
and knee angle at end of treatment, 6 weeks (no
benefit demonstrated); grade D for ankle angle at
end of treatment, 6 weeks (no benefit demon-
Clinical Practice Guidelines 35
strated but favoring control). Patients with chronic
stroke.
Rhythmic auditory facilitation training ver-
sus control for gait training, level I (RCT): Grade
A for gait speed and stride length at end of treat-
ment, 6 weeks (clinically important benefit dem-
onstrated); grade C+ for gait symmetry at end of
treatment, 6 weeks (clinically important benefit
demonstrated without statistical significance);
grade C for gait cadence at end of treatment, 6
weeks (no benefit demonstrated). Patients with
post-acute stroke.
Step length BFB training versus control for
gait training, level 1 (RCT): Grade C for step
length at end of treatment, 4 weeks (no benefit
demonstrated). Patients with post-acute stroke.
Musical motor feedback training versus con-
trol for gait training, level I (RCT): Grade C+ for
gait speed, stride length, and gait symmetry at end
of treatment, 3 weeks (clinically important benefit
demonstrated without statistical significance);
grade C for gait cadence and heel-on–toe-off dis-
tance at end of treatment, 3 weeks (no benefit
demonstrated). Patients with post-acute stroke.
FES versus standard rehabilitation for gait
training, level I (RCT): Grade A for maximum
isometric voluntary contraction of ankle
dorsiflexors at end of treatment, 2 and 3 weeks
(clinically important benefit demonstrated); grade
C+ for maximum isometric voluntary contraction
of ankle dorsiflexors at end of treatment, 1 week,
and follow-up, 5 weeks, and for spasticity at end of
treatment, 1, 2, and 3 weeks, and follow-up, 5
weeks (clinically important benefit demonstrated
without statistical significance); grade C for walk-
ing ability at follow-up, 5 weeks (no benefit dem-
onstrated); grade D+ for walking ability at end of
treatment, 1, 2, and 3 weeks (clinically important
benefit favoring control without statistical signifi-
cance). Patients with post-acute stroke.
FES versus placebo for gait training, level I
(RCT): Grade A for maximum isometric voluntary
contraction of ankle dorsiflexors at end of treat-
ment, 3 weeks (clinically important benefit dem-
onstrated); grade C+ for maximum isometric vol-
untary contraction of ankle dorsiflexors at end of
treatment, 1 and 2 weeks, and follow-up, 5 weeks,
and for spasticity at end of treatment, 1, 2, and 3
36 TOPICS IN STROKE REHABILITATION/SPRING 2006
weeks, and follow-up, 5 weeks (clinically impor-
tant benefit demonstrated without statistical sig-
nificance); grade C for walking ability at follow-
up, 5 weeks (no benefit demonstrated); grade D+
for walking ability at end of treatment, 1 week
(clinically important benefit favoring control with-
out statistical significance); grade D- for walking
ability at end of treatment, 2 and 3 weeks (clini-
cally important benefit favoring control). Patients
with post-acute stroke.
Placebo FES versus control for gait training,
level I (RCT): Grade A for walking ability at end of
treatment, 3 weeks (clinically important benefit
demonstrated); grade C+ for maximum isometric
voluntary contraction of ankle dorsiflexors at end
of treatment, 1 and 2 weeks, and follow-up, 5
weeks, for spasticity at end of treatment, 2 and 3
weeks, and follow-up, 5 weeks, and for walking
ability at end of treatment, 2 weeks, and follow-up,
5 weeks (clinically important benefit demon-
strated without statistical significance); grade C
for maximum isometric voluntary contraction of
ankle dorsiflexors at end of treatment, 3 weeks,
and for walking ability at end of treatment, 1 week
(no benefit demonstrated); grade D+ for spasticity
at end of treatment, 1 week (clinically important
benefit favoring control without statistical signifi-
cance). Patients with post-acute stroke.
Functional lower extremity training versus
control, level I (RCT): Grade A for quality of life at
end of treatment, 12 weeks, and follow-up, 6 and
18 weeks, functional status at follow-up, 6 weeks,
quality of life at end of treatment, 12 weeks, com-
fortable gait speed and maximum gait speed at end
of treatment, 12 and 20 weeks, and follow-up, 6
weeks (clinically important benefit demonstrated);
grade C+ for comfortable gait speed and maximum
Summary of trials
Twenty-two trials were included for comparison
with a total of 923 patients.16,51,164,165,206,248–264 Eleven
trials examined the use of treadmill training in pa-
tients post stroke.248,249,252–255,260,262,263 Of these, six
trials examined the use of treadmill training and
BWS.249,252,253,257,262,263 Five randomized controlled
gait speed at follow-up, 18 and 32 weeks, and
quality of life at follow-up, 32 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for mobility at end of treat-
ment, 12 and 20 weeks, and follow-up, 26, 38, and
52 weeks, and functional status at follow-up, 32
weeks (no benefit demonstrated); grade D for func-
tional status and quality of life at follow-up, 6, 18,
and 32 weeks (no benefit demonstrated but favor-
ing control). Patients with subacute stroke.
Treadmill training and BWS versus control,
level I (RCT): Grade A for gait speed at end of
treatment 2–3 weeks and follow-up, 3 months,
balance at end of treatment, 6 weeks, gait distance
at end of treatment, 2–3 weeks, motor recovery at
end of treatment, 6 weeks, oxygen consumption at
end of treatment, 2–3 weeks (clinically important
benefit demonstrated); grade C+ for cardiovascu-
lar endurance at end of treatment, 2–3 weeks, gait
endurance at end of treatment, 6 weeks, and fol-
low-up, 3 months, and decrease in gait energy cost
at end of treatment, 2–3 weeks, motor recovery at
follow-up, 3 months (clinically important benefit
demonstrated without statistical significance);
grade C for mobility at end of treatment, 2, 3, and
10 weeks, and follow-up, 10 months, functional
status and cognitive status at end of treatment, 10
weeks, and follow-up, 10 months, motor function
at end of treatment, 10 weeks, and energy expen-
diture at end of treatment, 2 and 3 weeks (no
benefit demonstrated); grade D for motor function
at follow-up, 10 months (no benefit demonstrated
but favoring control); grade D+ for FIM locomo-
tion at end of treatment, 2–3 weeks (clinically
important benefit favoring control demonstrated
without statistical significance). Patients with
acute and post-acute stroke.
trials studied the effect of FES.51,250,251,256,264 Two
trials examined functional lower extremity train-
ing,164,165 while four trials evaluated individual inter-
ventions including electrogoniometric feedback,16
rhythmic auditory facilitation,263 step length BFB,206
and musical motor feedback.259 Sixteen trials com-
pared the intervention of interest with an active
intervention, while six trials utilized a control

group. Total treatment length varied between 2
weeks and 20 weeks, and individual treatment ses-
sions ranged between 20 minutes and 3 hours
(Gait-Appendix 3D).
Twenty-seven trials were excluded (Gait-Table
73).265–283 Reasons for exclusion included insuffi-
cient statistical data, no outcomes of interest, inap-
propriate control group, or inappropriate study
design.
Efficacy
For treadmill training and BWS versus con-
trol, clinically important benefits were demon-
strated for gait speed at end of treatment, 2–3
weeks (35% RD),253 and follow-up, 3 months
(108% RD)262; balance at end of treatment (28%
RD)262; gait distance at end of treatment, 2–3
weeks (98% RD)253; motor recovery as measured
using STREAM at end of treatment, 6 weeks (22%
RD)249; and oxygen consumption at end of treat-
ment, 2–3 weeks (35% RD)252 (Gait-Figure 68,
Table 74). Clinically important benefits without
statistical significance were demonstrated for car-
diovascular endurance at end of treatment, 2–3
weeks (18% RD)252; gait endurance at end of treat-
ment, 6 weeks, and follow-up, 3 months (97% and
114% RD, respectively)262; gait energy expenditure
at end of treatment, 2–3 weeks (95% RD) 253; mo-
tor recovery and balance at follow-up, 3 months
(18% and 21% RD, respectively) 249; and gait
speed at end of treatment, 6 weeks (34% RD). No
clinical benefits were found for mobility (Func-
tional Ambulation Category Scale) at end of treat-
ment, 2, 3, and 10 weeks, and follow-up, 10
months; functional status (FIM motor scale) at end
of treatment, 10 weeks, and follow-up, 10 months;
energy expenditure at end of treatment, 2 and 3
weeks; or motor function (Fugl-Meyer scale), bal-
ance (Berg), gait speed, and cognitive status (FIM
cognitive scale) at end of treatment, 10 weeks. No
clinical benefit was demonstrated, but results fa-
vored the control group for motor function (Fugl-
Meyer scale), balance, and cognitive status (FIM
cognitive scale) at follow-up, 10 months. In addi-
tion, a clinically important benefit favoring the
control group was found for locomotion, as mea-
sured with the FIM at end of treatment and gait
Clinical Practice Guidelines 37
speed at follow-up, 10 months. However, this re-
sult was not statistically significant. No other out-
comes were measured.
For treadmill training and BWS combined
with physiotherapy versus treadmill training
alone, one RCT (n = 28)263 found a clinically im-
portant benefit for gait speed at end of treatment, 3
weeks (24% RD; Gait-Figure 69, Table 75), while
no benefit was demonstrated for gait speed at fol-
low-up. Other outcomes were measured but no
meaningful statistical data could be extracted.
When comparing treadmill training and BWS
to aggressive brace-assisted walking, one RCT
(n = 55)254 demonstrated clinically important ben-
efits for the outcomes of gait endurance and length
of stay at end of treatment, 2 weeks (44% and
15% RD, respectively; Gait-Figure 70, Table
76). No benefit was demonstrated for gait speed at
end of treatment, 2 weeks. No other outcomes
were measured.
One CCT255 (n = 25) compared treadmill train-
ing to standard customary exercise, and found a
clinically important benefit with statistical signifi-
cance for mobility at end of treatment (39% RD;
Gait-Figure 71, Table 77) and clinically impor-
tant benefits without statistical significance for gait
speed and stride length at end of treatment, 3
weeks (63 and 26% RD, respectively; Gait-Figure
71, Table 77). No other outcomes were measured.
For high-speed treadmill training versus low-
speed treadmill training in chronic stroke pa-
tients, clinically important benefits were demon-
strated for high-speed training by Pohl258 (n = 60)
for fastest comfortable gait speed at end of treat-
ment, 4 weeks (72% RD), and gait cadence at end of
treatment, 2 weeks (31% RD; Gait-Figure 72,
Table 78). In addition, clinically important benefits
were demonstrated for high-speed treadmill train-
ing for stride length at end of treatment, 2 and 4
weeks (18%–34% RD); fastest comfortable gait
speed at end of treatment, 2 weeks (18% RD); and
gait cadence at end of treatment, 4 weeks (18% RD).
However, these results were not statistically signifi-
cant. Sullivan (n = 24) found no difference between
the two comparison groups for gait speed, % BWS
(Gait-Table 79), or upper extremity support at end
of treatment. No other outcomes were measured.
When comparing low-speed treadmill training
38 TOPICS IN STROKE REHABILITATION/SPRING 2006
to standard gait training (one RCT, n = 60),258
clinically important benefits were demonstrated
for fastest comfortable gait speed, gait cadence,
and mobility at end of treatment, 4 weeks (38%,
19%, and 24% RD, respectively; Gait-Figure 73,
Table 80). No benefits were demonstrated for
stride length at end of treatment, 2 and 4 weeks, or
fastest comfortable gait speed and gait cadence at
end of treatment, 2 weeks. No other outcomes
were measured.
For high-speed treadmill training versus stan-
dard gait training (one RCT, n = 60),258 clinically
important benefits were demonstrated for fastest
comfortable gait speed and stride length at end of
treatment, 4 weeks (112% and 71% RD, respec-
tively), and for gait cadence at end of treatment, 2
and 4 weeks (25% and 38% RD) (Gait-Figure 73,
Table 80). Clinically important benefits without
statistical significance were demonstrated for com-
fortable gait speed and stride length at end of
treatment, 2 weeks (54% and 23% RD, respec-
tively; Gait-Figure 74, Table 81). No other out-
comes were measured.
For treadmill training combined with
overground ambulation versus control (one
RCT, n = 27),248 clinically important benefits were
demonstrated for step length on the affected side at
end of treatment and follow-up (15%–20% RD)
and step length on the nonaffected side at end of
treatment, 4 weeks (17% RD; Gait-Figure 75,
Table 82). Clinically important benefits were also
demonstrated for gait speed at end of treatment
and follow-up (17% RD) and walking distance and
step length on the nonaffected side at end of treat-
ment (26% and 17% RD, respectively; Gait-Fig-
ure 75, Table 82). However, there was no statisti-
cal significance. No benefit was found for step
width at end of treatment or follow-up or for
quality of life (Sickness Impact Profile) and walk-
ing distance at follow-up, 3 months. No benefits
were demonstrated, but results favoring the control
group were found for gait cadence at end of treat-
ment and follow-up and for quality of life at end of
treatment. No other outcomes were measured.
When comparing electrogoniometric feedback
training to a control group for gait training, one
RCT (n = 26)16 demonstrated clinically important
benefits for change in gait speed and reduction of
knee hyperextension at end of treatment, 4 weeks
(155% and 25% RD, respectively), and follow-up,
4 weeks (35% and 124% RD, respectively; Gait-
Figure 76, Table 83). However, these results were
not statistically significant. No other outcomes
were measured.
FES of the ankle dorsiflexors in subacute
stroke patients (one RCT, n = 38)256 produced
clinically important benefits without statistical sig-
nificance for functional status (Barthel Index) and
level of walking independence (Massachusetts
General Hospital Functional Ambulation Classifi-
cation) at end of treatment (15% and 59% RD,
respectively) and follow-up (15% and 59% RD,
respectively; Gait-Figure 77, Table 84) when
compared to a control group. While no benefit was
demonstrated for motor function at end of treat-
ment or follow-up, results favored the control
group. No other outcomes were measured.
FES of the ankle dorsiflexors in post-acute
stroke patients (one RCT, n = 20)250 demonstrated
clinically important benefits at end of treatment, 3
weeks, for stride length (47% RD), gait cadence
(32% RD), and gait speed (92% RD; Gait-Figure
78, Table 85). However, results for gait cadence
and gait speed were not statistically significant. No
benefit was demonstrated for motor function at end
of treatment. No other outcomes were measured.
For FES of the ankle dorsiflexors and/or gastroc-
nemius in chronic stroke patients versus control
(n = 32 ),251 (n = 36),51 clinically important benefits
were established for the PCI (orthosis on) and
spasticity at end of treatment, 12 weeks (114% and
24% RD, respectively; Gait-Figure 79, Table 86).
Clinically important benefits were also established
for gait cycle time at end of treatment, 6 weeks
(25% RD)51 and for the PCI (orthosis on) at end of
treatment, 4 weeks (98% RD)251 (Gait-Figure 79,
Table 86). However, there was no statistical sig-
nificance. No benefit was established for gait speed
(orthosis on), PCI (orthosis off), or spasticity at
end of treatment, 4 or 12 weeks. No benefit was
established for stride length or ankle or knee angle
at end of treatment, 6 weeks, or for gait speed
without FES at end of treatment, 4 and 12 weeks.
However, results favored the control group. No
other outcomes were measured.
For BFB followed by FES (one RCT, n = 36),51 a
clinically important benefit was shown for gait
cycle time at end of treatment, 6 weeks (19% RD;

Gait-Figure 80, Table 87) when compared to a
control group. No benefits were demonstrated for
stride length or knee or ankle angle at end of
treatment, 6 weeks. However, the results favored
the control group for the outcome on ankle angle.
No other outcomes were measured.
Rhythmic auditory facilitation training (one
RCT, n = 20)261 demonstrated clinically important
benefits at end of treatment for gait speed (40%
RD) and stride length (37% RD; Gait-Figure 81,
Table 88) when compared to a control. A clini-
cally important benefit without statistical signifi-
cance was established for gait symmetry (19% RD;
Gait-Figure 81, Table 88). No benefit was found
for gait cadence at end of treatment. No other
outcomes were measured.
Step length BFB training in stroke patients
(one RCT, n = 14)206 demonstrated no benefit for
step length at end of treatment, 4 weeks (Gait-
Figure 82, Table 89). No other outcomes were
measured.
For musical motor feedback training versus
control (one RCT, n = 23),259 clinically important
benefits without statistical significance were estab-
lished for gait speed, stride length, and gait sym-
metry at end of treatment, 3 weeks (20%, 17%,
and 58% RD, respectively; Gait-Figure 83, Table
90). No benefit was demonstrated for gait cadence
or heel-to–toe-off distance. No other outcomes
were measured.
When comparing FES in subacute stroke pa-
tients to standard rehabilitation (one RCT, n =
38),256 clinically important benefits were found for
the maximum isometric voluntary contraction of
the ankle dorsiflexors at the end of treatment, 2
and 3 weeks (214% and 228% RD, respectively;
Gait-Figure 84, Table 91). Clinically important
benefits were also established for the maximum
isometric voluntary contraction of the ankle
dorsiflexors at the end of treatment, 1 week, and
follow-up, 5 weeks (70% and 186% RD, respec-
tively; Gait-Figure 84, Table 91), and for spastic-
ity at end of treatment 1, 2 and 3 weeks and
follow-up, 5 weeks (207%, 491%, 527% and
567% RD, respectively; Gait-Figure 83, Table
90), and walking ability (Timed Up & Go) at
follow-up, 5 weeks (33% RD; Gait-Figure 84,
Table 91). However, they were not statistically
significant. No benefit was demonstrated for walk-
Clinical Practice Guidelines 39
ing ability at end of treatment, 1, 2, or 3 weeks, but
the results favored the control group. No other
outcomes were measured.
For FES compared to a placebo (one RCT, n =
41),264 a clinically important benefit was demon-
strated for the maximum isometric voluntary con-
traction of the ankle dorsiflexors at end of treat-
ment, 3 weeks (218% RD; Gait-Figure 85, Table
92). Clinically important benefits without statisti-
cal significance were found for the maximum iso-
metric voluntary contraction of the ankle
dorsiflexors at end of treatment, 1 and 2 weeks,
and at follow-up, 5 weeks (47%, 133%, and 157%
RD, respectively; Gait-Figure 85, Table 92), and
for spasticity at end of treatment, 1, 2, and 3
weeks, and follow-up, 5 weeks (279%, 389%,
319%, and 238% RD, respectively; Gait-Figure
85, Table 92). Clinically important benefits favor-
ing the control group were found for walking abil-
ity at end of treatment, 1, 2, and 3 weeks. The
results at 2 and 3 weeks were also statistically
significant. No other outcomes were measured.
Placebo FES (one RCT, n = 41)264 produced a
clinically important benefit for walking ability at
end of treatment, 3 weeks (65% RD; Gait-Figure
86, Table 93) when compared to a control group.
Clinically important benefits without statistical
significance were also established for maximum
isometric voluntary contraction of the ankle
dorsiflexors at the end of treatment, 1 and 2 weeks,
and follow-up, 5 weeks (22%, 80%, and 31% RD,
respectively; Gait-Figure 85, Table 92); spasticity
at end of treatment, 2 and 3 weeks, and follow-up,
5 weeks (123%, 240%, and 374% RD, respec-
tively; Gait-Figure 86, Table 93); and for walking
ability at end of treatment, 2 weeks, and follow-up,
5 weeks (32% and 23% RD, respectively; Gait-
Figure 86, Table 93). No benefit was found for
maximum isometric voluntary contraction of the
ankle dorsiflexors at end of treatment, 3 weeks, or
for walking ability at end of treatment, 1 week. A
clinically important benefit favoring the control
group without statistical significance was estab-
lished for spasticity at end of treatment, 1 week.
For functional lower extremity training ver-
sus control, two RCTs (n = 89; n = 101)164,165
showed clinically important benefits for functional
status (Sickness Impact Profile) at end of treat-
ment, 12 weeks, and at follow-up, 6 and 18 weeks
40 TOPICS IN STROKE REHABILITATION/SPRING 2006
(27%, 20%, and 22% RD, respectively; Gait-Fig-
ure 87, Table 94); functional status (FAI) at fol-
low-up, 6 weeks (19% RD; Gait-Figure 87, Table
94); quality of life (NHP) at end of treatment, 12
weeks (20% RD; Gait-Figure 87, Table 94); and
comfortable gait speed and maximum gait speed at
end of treatment, 12 and 20 weeks, and follow-up,
6 weeks (36%–64% and 41%–66% RD, respec-
tively; Gait-Figure 87, Table 94). Clinically im-
portant benefits without statistical significance
were demonstrated for comfortable gait speed and
maximum gait speed at follow-up, 18 and 32
weeks (19%–23% and 18%–25% RD, respec-
tively; Gait-Figure 87, Table 94) and quality of
life (Sickness Impact Profile) at follow-up, 32
weeks (15% RD; Gait-Figure 87, Table 94). No
benefit was demonstrated for mobility (number of
patients using walking aids; Gait-Table 95) at end
of treatment, 12 and 20 weeks, or follow-up, 26,
38, and 52 weeks, and functional status at follow-
up, 32 weeks. No benefit was demonstrated, but
results favoring the control group were found for
functional status and quality of life at follow-up, 6,
18, and 32 weeks.
Strength of published evidence compared to other
guidelines
The Ottawa Panel found level I (RCT) evidence
for the use of treadmill training and BWS to in-
crease gait speed and gait distance and to improve
balance, motor recovery, and oxygen consump-
tion. The strength of the evidence has also been
graded by the Department of Veterans Affairs,75 the
Royal College of Physicians,72 and St. Joseph’s
Health Care London,73 which also reported level I
(RCT) evidence respectively for the use of partial
BWS and treadmill training. Level I (RCT) evi-
dence was also found for the use of high-speed
treadmill training. There was equally level I (RCT)
evidence to support the use of treadmill training
and overground ambulation to increase step length
on the affected side. The strength of the evidence
has also been graded by St. Joseph’s Health Care
London,73 which reported level 1 (RCT) evidence
for the use of treadmill training (Appendix 4).
The panel also reported level I (RCT) evidence
for the use of FES of the ankle dorsiflexors and/or
gastrocnemius to increase the maximum isometric
voluntary contraction of the ankle dorsiflexors,
stride length, and the PCI and to decrease spastic-
ity. Level I evidence was also found for the use of
placebo FES to improve walking ability in sub-
acute patients post stroke. The strength of the
evidence has also been graded by the Department
of Veterans Affairs75 and St. Joseph’s Health Care
London.73 The Department of Veterans Affairs re-
ported level I (RCT) evidence for the use of multi-
channel FES in severe hemiplegic patients and
level II evidence (cohort/case control studies) for
the use of FES for gait training following stroke,
while St. Joseph’s Health Care London73 reported
level I evidence for the use of FES in gait training
(Appendix 4).
Last, the Panel found good evidence to support
the use of BFB followed by FES to increase gait
cycle time and rhythmic auditory facilitation to
increase gait speed and stride length. The strength
of the evidence has also been graded by St.
Joseph’s Health Care London,73 which reported
level I evidence (9 RCTs) for the use of BFB train-
ing for gait (Appendix 4).
Clinical recommendations compared with other
guidelines
According to the Ottawa Panel, there is good
evidence that treadmill training with BWS (grade
A for gait speed, gait distance, balance, motor
recovery, and oxygen consumption; grade C+ for
cardiovascular endurance, gait endurance, gait en-
ergy cost, and motor recovery) should be included
as an intervention for stroke patients. This is in line
with the Royal College of Physicians72 and Depart-
ment of Veterans Affairs75 recommendations that
treadmill training with partial BWS should be in-
cluded as an adjunct to conventional therapy for
patients undergoing stroke rehabilitation. How-
ever, this recommendation is contrary to the rec-
ommendations by St. Joseph’s Health Care Lon-
don,73 which found conflicting evidence* on the
use of BWS and treadmill training (Appendix 4).
*According to the St. Joseph’s Health Care London, conflicting evi-
dence refers to a disagreement between the findings of at least two
RCTs. Where there were more than four RCTs and the results of only
one was conflicting, the conclusion was based on the results of the
majority of the studies, unless the study with conflicting results was
of higher quality.

The Ottawa Panel also found good evidence
(grade A for fastest comfortable gait speed, gait
cadence, and step length on the affected side) to
include treadmill training without BWS as an inter-
vention for stroke patients. Again, this recommen-
dation conflicts with that of St. Joseph’s Health Care
London,73 which found conflicting evidence* on
the use of treadmill training (Appendix 4).
The Ottawa Panel also reported that there is
good evidence to support the use of FES (grade
A for maximum isometric voluntary contraction,
stride length, PCI, and spasticity; grade C+ for
functional status, level of walking indepen-
dence, gait cadence, gait speed, and gait cycle
time) in gait training for stroke patients. This is
in line with the recommendations made by St.
Joseph’s Health Care London73 that reported
strong evidence to include FES for patients un-
Clinical practice guidelines for balance training
Balance training (sit-to-stand) with force plat-
form and auditory feedback versus control (no
feedback), level I (RCT): Grade A for time up at
follow-up, 33 months (clinically important benefit
demonstrated); grade C+ for strength (dynamic
knee extensors strength in concentric movement) at
end of treatment, 6 weeks (clinically important ben-
efit demonstrated without statistical significance);
grade C for time up, body weight distribution on
affected lower extremity in rising and sitting,
strength (dynamic knee flexors strength in eccentric
movement), physical performance, and activities of
daily living at end of treatment, 6 weeks, time down
and strength (dynamic knee extensors strength in
concentric movement) at follow-up, 33 months,
and strength (dynamic knee extensors strength in
eccentric movement) at end of treatment, 6 weeks,
and follow-up, 33 months (no benefit demon-
strated); grade D for time down, strength (dynamic
knee flexors in concentric movement), and motor
function in sit to stand at end of treatment, 6 weeks,
and body weight distribution on affected lower ex-
tremity in rising and sitting at follow-up, 33 months
(clinically important benefit favoring control dem-
onstrated). Patients with subacute and post-acute
stroke.
Clinical Practice Guidelines 41
dergoing stroke rehabilitation. In addition, it is
in partial agreement with the conflicting guide-
lines made by the Department of Veterans Af-
fairs73 that recommended FES for gait training
following stroke but also found insufficient evi-
dence to recommend multi-channel FES for se-
vere hemiplegic patients with gait impairment
(Appendix 4).
Finally, the Ottawa Panel is in agreement with
the recommendations by St. Joseph’s Health Care
London,73 which reported strong evidence for the
use of rhythmic auditory stimulation in patients
post stroke. The Ottawa Panel found good evi-
dence (grade A for gait speed and stride length;
grade C+ for gait symmetry) to support the use of
rhythmic auditory stimulation as an intervention
following stroke (Appendix 4).
Base of support training with extrinsic audi-
tory feedback versus control, level I (RCT):
Grade A for step width at end of treatment, 10
days (clinically important benefit demonstrated).
Patients with subacute stroke.
Balance training with visual feedback versus
control, level I (RCT) and level II (CCT): Grade A
for motor function (Rivermead Motor Function
Scale: global and gross motor function) at end of
treatment, 4 weeks, functional status at end of
treatment, 4 weeks, and follow-up, 8 weeks, timed
task (duration of sit-to-stand), body weight distri-
bution, rate of rise in force during sit-to-stand (%
of body weight/second), dynamic balance (end
point excursion* [right and backward], axis veloc-
ity** [right-left], and directional control*** [for-
ward-backward]) at follow-up, 6 months (clini-
*End point excursion refers to the distance traveled by the center of
gravity on the primary attempt to reach a target moving in a different
direction in 0.8 seconds.
**Axis velocity is the average speed of the subject’s center of gravity
movement in the specified direction.
***Directional control is defined as the ratio of the actual distance
traveled by the center of gravity from the center to end point excur-
sion compared with the shortest distance between those two points (a
straight line).
42 TOPICS IN STROKE REHABILITATION/SPRING 2006
cally important benefit demonstrated); grade C+
for static balance (center of gravity alignment: eyes
open, sway vision, eyes closed/sway surface, and
sway vision/sway surface) at follow-up, 6 months,
motor function (Rivermead Motor Function Scale:
lower extremity and trunk) at end of treatment, 4
weeks, dynamic balance (end point excursion* left
and forward) at follow-up, 6 months, balance
(postural sway eyes open) at end of treatment, 3
weeks (clinically important benefit demonstrated
without statistical significance); grade C for static
balance (center of gravity alignment: eyes closed,
eyes open/sway surface) at follow-up, 6 months,
functional mobility and balance (postural sway
eyes opened) at follow-up, 1 month, and balance
(Berg Scale) at end of treatment, 6 weeks, and
follow-up, 1 month, gait speed at end of treatment,
4 weeks and 6 weeks, mobility (stride length, ca-
dence, and gait cycle duration) at end of treatment,
4 weeks, balance (standing symmetry) at end of
treatment 3 weeks, static balance (maximal stabil-
ity eye open and eye closed), static balance (ankle
strategy eyes open, eyes closed, sway vision, eyes
closed/sway surface, and sway vision/sway sur-
face), dynamic balance-axis velocity** (forward-
backward) and dynamic balance-directional con-
trol*** (right-left), and timed task (duration of
stand-to-sit) at follow-up, 6 months (no benefit
demonstrated); grade D for static balance (maxi-
mal stability sway vision, eyes open/sway surface,
eyes closed/sway surface, and sway vision/sway
surface), static balance (ankle strategy eyes open/
sway surface) at follow-up, 6 months, motor func-
tion (Rivermead Motor Function Scale: gross mo-
tor function) at follow-up, 8 weeks, balance (Berg
Scale) at end of treatment, 3 and 4 weeks, func-
tional mobility at end of treatment, 4 and 6 weeks,
gait speed and balance (standing symmetry) at
follow-up, 1 month (no benefit demonstrated but
favoring control); grade D+ for balance (postural
sway eyes closed) at end of treatment, 3 weeks,
*End point excursion refers to the distance traveled by the center of
gravity on the primary attempt to reach a target moving in a different
direction in 0.8 seconds.
**Axis velocity is the average speed of the subject’s center of gravity
movement in the specified direction.
***Directional control is defined as the ratio of the actual distance
traveled by the center of gravity from the center to end point excur-
sion compared with the shortest distance between those two points (a
straight line).
and follow-up, 1 month, functional mobility and
gait speed at end of treatment, 3 weeks, motor
function (Rivermead Motor Function Scale: global
and lower extremity and trunk) at follow-up, 8
weeks (clinically important benefit favoring con-
trol demonstrated without statistical significance).
Patients with subacute and post-acute stroke.
Balance training with visual feedback versus
Bobath approach training, level I (RCT): Grade
A for standing balance at end of treatment, 4
weeks, and follow-up, 2 weeks (clinically impor-
tant benefit demonstrated); grade C+ for sitting
balance at end of treatment, 4 weeks, and follow-
up, 12 weeks, and standing balance at follow-up,
12 weeks (clinically important benefit demon-
strated without statistical significance); grade C
for sitting balance at follow-up, 2 weeks (no ben-
efit demonstrated). Patients with subacute stroke.
Balance training with visual feedback versus
task-specific training, level I (RCT): Grade A for
standing balance at follow-up, 2 weeks (clinically
important benefit demonstrated); grade C+ for
sitting balance at end of treatment, 4 weeks, and
follow-up, 12 weeks, and standing balance at fol-
low-up, 12 weeks (clinically important benefit
demonstrated without statistical significance);
grade C for sitting balance at follow-up, 2 weeks,
and standing balance at end of treatment, 4 weeks
(no benefit demonstrated). Patients with subacute
stroke.
Balance training with visual feedback versus
control, level I (RCT): Grade A for standing bal-
ance at end of treatment, 4 weeks, and follow-up,
2 weeks (clinically important benefit demon-
strated); grade C+ for sitting balance and standing
balance at follow-up, 12 weeks (clinically impor-
tant benefit demonstrated without statistical sig-
nificance); grade C for sitting balance at end of
treatment, 4 weeks (no benefit demonstrated);
grade D+ for sitting balance at follow-up, 2 weeks
(clinically important benefit favoring control dem-
onstrated without statistical significance). Patients
with subacute stroke.
Platform training versus control, level I
(RCT): Grade A for standing balance at end of
treatment, 3 weeks (clinically important benefit
demonstrated); grade C for weight distribution at
end of treatment, 3 weeks (no benefit demon-
strated). Patients with subacute stroke.

Summary of trials
Ten RCTs (n = 306) and 1 CCT (n = 42) evaluat-
ing balance training in the inpatient or outpatient
setting were included.119,284–293 The individual
training methods varied, but most of them em-
ployed supplementary visual or auditory feedback
to guide postural balance. In most of the trials,
patients were positioned on force-transducing
platforms such that they could monitor their rela-
tive weight distribution and center of gravity dur-
ing the training exercises.119,285–289,291,292 In general,
the control interventions consisted of standard
balance training without BFB or customary care
(Appendix 3E).
Treatment schedules ranged from a single 30-
minute session per day for 2 weeks to three 15-
minute sessions per day for 5 days a week and over
6 weeks. Recruited patients were in the subacute
or post-acute phase of stroke (Appendix 3E).
Altogether, three potentially relevant studies
were excluded.294–296 In two cases, insufficient sta-
tistical data were provided.289,290 In the third
study,294 the treatment period was not standard-
ized and varied between and within the treatment
and control groups (BT-Table 96).
Efficacy
In comparing balance training (sit-to-stand)
with force platform and auditory feedback to a
control training group without BFB in subacute
and post-acute stroke patients, statistically sig-
nificant and clinically important benefits were
shown for time to rise to a standing position (time
up) at 33 months follow-up (one RCT, n = 30;
BT-Figure 88, Table 97).287The RD between ex-
perimental and control groups was 90%. Clini-
cally important benefits without statistical signifi-
cance were demonstrated for strength (dynamic
knee extensors strength in concentric movement)
at the end of 6 weeks of treatment (15% RD). No
benefit was observed for time up, body weight
distribution on affected lower extremity in rising
and sitting, strength (dynamic knee flexors
strength in eccentric movement), physical perfor-
mance (Fugl-Meyer), and activities of daily living
(Barthel ADL Index) at the end of 6 weeks of
treatment. Benefits were also not seen for time
Clinical Practice Guidelines 43
down and strength (dynamic knee extensors
strength in concentric movement) at 33 months
follow-up and strength (dynamic knee extensors
strength in eccentric movement) at end of treat-
ment and follow-up. A statistically insignificant
and clinically unimportant trend favoring control
was shown for strength (dynamic knee flexors in
concentric movement) at the end of treatment
and body weight distribution on affected lower
extremity in rising and sitting at the 33 month
follow-up. Statistically and clinically significant
benefits favoring control were found for time
down (22% RD) and motor function in sit to
stand (−100% RD) at the end of treatment.
For base of support training with extrinsic
auditory feedback for subacute stroke patients
compared to control, statistically significant and
clinically important benefits were shown for step
width at the end of the 10 day treatment (41% RD;
one RCT, n = 16; BT-Figure 89, Table 98).284 No
follow-up was done.
Six RCTs (n = 196) and 1 CCT (n = 42) examined
balance training with visual feedback relative
to a control intervention without feedback for
stroke patients in the subacute and post-acute
phases of recovery.285–289,291–293 Statistically signifi-
cant and clinically important benefits were shown
for motor function (Rivermead Motor Function
Scale: global and gross motor function) at the end
of 4 weeks of treatment (40% and 42% RD, respec-
tively; BT-Figure 92, Table 100); functional sta-
tus (Nottingham 10-point ADL scale) at the end of
4 weeks of treatment and at 8 weeks follow-up
(43% and 21% RD, respectively; BT-Figure 92,
Table 100); timed task (duration of sit-to-stand)
and body weight distribution and rate of rise in
force during sit-to-stand (% of body weight/sec-
ond) at 6 months follow-up (−24%, −21%, and
88% RD, respectively; BT-Figure 90, Table 99);
dynamic balance (end point excursion [right and
backward]), axis velocity [right-left], and direc-
tional control [forward-backward]) at 6 months
follow-up (55%, 24%, 29%, and 29% RD, respec-
tively; BT-Figure 91, Table 99). Clinic a l l y i m -
portant benefits without statistical sig-
n i f i c a n c e were demonstrated for static balance
(center of gravity alignment: eyes open, sway
vision, eyes closed/sway surface, and sway vi-
44 TOPICS IN STROKE REHABILITATION/SPRING 2006
sion/sway surface) at 6 months follow-up (−18%,
−15%, −15%, and −40% RD, respectively; BT-
Figure 91, Table 99); motor function
(Rivermead Motor Function: lower extremity and
trunk) at the end of 4 weeks of treatment (37%
RD; BT-Figure 92, Table 100); dynamic balance
(end point excursion [left and forward]) at 6
months follow-up (21% and 24% RD, respec-
tively; BT-Figure 91, Table 99); and balance
(postural sway with eyes open) at the end of 3
weeks of treatment (17% RD; BT-Figure 93,
Table 100). No benefit was demonstrated for
static balance (center of gravity alignment: eyes
closed, eyes open/sway surface) at 6 months fol-
low-up, functional mobility (Timed Up & Go)
and balance (postural sway eyes opened) at 1
month follow-up and balance (Berg Scale) at the
end of 6 weeks treatment and 1 month follow-up;
gait speed at the end of treatment at 4 weeks and
6 weeks; mobility (stride length, cadence, and
gait cycle duration) at the end of 4 weeks treat-
ment; balance (standing symmetry) at the end of
3 weeks of treatment; and static balance (maxi-
mal stability eyes open and eyes closed), static
balance (ankle strategy eyes open, eyes closed,
sway vision, eyes closed/sway surface, sway vi-
sion/sway surface), dynamic balance (axis veloc-
ity [forward-backward], directional control
[right-left]), and timed task (duration of stand-to-
sit) at the end of 6 months follow-up. Statistically
insignificant and clinically unimportant trends
favoring control were found for static balance
(maximal stability sway vision, eyes open/sway
surface, eyes closed/sway surface, and sway vi-
sion/sway surface) and static balance (ankle strat-
egy eyes open/sway surface) at 6 months follow-
up, motor function (Rivermead Motor Function
Scale: gross motor function) at 8 weeks follow-
up, balance (Berg Scale) at the end of treatment at
3 and 4 weeks, functional mobility (Timed Up &
Go) at the end of treatment at 4 and 6 weeks, and
gait speed and balance (standing symmetry) at 1
month follow-up (BT-Figure 93, Table 101).
Clinically important benefits favoring control but
without statistical significance were demon-
strated for balance (postural sway with eyes
closed) at the end of 3 weeks treatment and 1
month follow-up (−74% and −90% RD, respec-
tively; BT-Figure 93, Table 102), functional mo-
bility (Timed Up & Go) and gait speed at the end
of 3 weeks treatment (RD: 55% and −48% RD,
respectively; BT-Figure 92, Table 100), and mo-
tor function (Rivermead Motor Function Scale:
global and lower extremity and trunk) at 8 weeks
follow-up (−31% RD; BT-Figure 92, Table 100).
For balance training with visual feedback
compared to Bobath approach training for sub-
acute stroke patients (one RCT, n = 40),119 statisti-
cally significant and clinically important im-
provements in standing balance were seen at the
end of 4 weeks of treatment and at the 2-week
follow-up (−34% and −36% RD, respectively;
BT-Figure 94, Table 103). There were clini-
cally important benefits without statistical sig-
nificance in sitting balance at the end of 4 weeks
of treatment and at the 12-week follow-up (−21%
and −104% RD, respectively) and standing bal-
ance at the 12-week follow-up (−31% RD). No
benefit was demonstrated for sitting balance at the
2-week follow-up.
For balance training with visual feedback
compared to task-specific training in subacute
stroke patients, statistically significant and clini-
cally important benefits were found for standing
balance at the 2-week follow-up (-30% RD; one
RCT, n = 40)119 (BT-Figure 95, Table 104). Clini-
cally important benefits without statistical signifi-
cance were demonstrated for sitting balance at the
end of the 4-week treatment and at the 12-week
follow-up (−27% and −30% RD, respectively).
Similar benefits occurred for standing balance at
the 12-week follow-up (−32% RD). No benefit was
shown for sitting balance at the 2-week follow-up
and standing balance at the end of treatment.
For balance training with visual feedback
compared to control in subacute stroke patients,
statistically significant and clinically important im-
provements in standing balance were shown at the
end of the 4-week treatment and at the 2-week
follow-up (−19% and −81% RD, respectively; one
RCT, n = 40)119 (BT-Figure 96, Table 105). There
were clinically important benefits without sta-
tistical significance for sitting balance and
standing balance at the 12-week follow-up (−52%
and −99% RD, respectively). No benefit was dem-
onstrated for sitting balance at end of the 4 weeks
of treatment. Statistically insignificant but clini-
cally important benefits favoring control were

found for sitting balance at the 2-week follow-up
(24% RD).
A comparison of platform training to control
revealed statistically and clinically significant ben-
efits with respect to standing balance, represented
by a change in the maximum movement amplitude
of the platform (RD: 66% RD; one RCT, n = 24)290
(BT-Figure 97, Table 106). No benefit was seen
for weight distribution (number of patients im-
proved on weight distribution) at the end of the 3-
week treatment. No follow-up was done.
Strength of the published evidence compared with
other guidelines
The Ottawa Panel found level I (RCTs) and level II
(CCT) evidence evaluating balance training in post-
stroke rehabilitation. The clinical practice guidelines
formulated by St. Joseph’s Health Care London73
were based on level I (RCT) evidence, although non-
RCTs were also considered (Appendix 4).
Clinical recommendations compared with other
guidelines
The Ottawa Panel found good evidence to rec-
ommend balance training as a rehabilitation in-
tervention for subacute and post-acute stroke pa-
tients. Specifically, good evidence was found for
base of support training with extrinsic auditory
feedback (grade A for step width at the end of 10
days treatment), balance training (sit-to-stand)
with force platform and auditory feedback (grade
A for time up at 33 months follow-up; grade C+
Clinical practice guidelines for sensory interventions
Feedback of eye movement versus control,
level I (RCT): Grade C for proportion of eye move-
ments and visual attention (Behavioral Inattention
Test – behavioral subtests) at end of treatment, 4
weeks (no benefit demonstrated); and grade D for
visual attention (Behavioral Inattention Test – con-
ventional subtests) at end of treatment, 4 weeks
(no benefits demonstrated but favoring control).
Patients with subacute stroke.
Passive vestibular stimulation versus control,
level II (CCT): Grade C+ for mobility at end of
Clinical Practice Guidelines 45
for strength [dynamic knee extensors strength in
concentric movement] at the end of 6 weeks of
treatment), platform training (grade A for stand-
ing balance at the end of 6 weeks of treatment),
and balance training with visual feedback (grade
A for standing balance at the end of 4 weeks of
treatment and 2 weeks of follow-up), motor func-
tion (Rivermead Motor Function Scale: global
and gross motor function) at the end of 4 weeks
of treatment, functional status at the end of 4
weeks of treatment and 8 weeks of follow-up,
timed task (duration of sit-to-stand), body weight
distribution, rate of rise in force during sit-to-
stand (% of body weight/second), dynamic bal-
ance (end point excursion [right and backward],
axis velocity [right-left], and directional control
[forward-backward]) at 6 months follow-up;
grade C+ for sitting balance and static balance
(center of gravity alignment: eyes open, sway vi-
sion, eyes closed/sway surface, and sway vision/
sway surface) at 6 months follow-up, motor func-
tion (Rivermead Motor Function: lower extremity
and trunk) at the end of 4 weeks of treatment,
dynamic balance-end point excursion (left) and
(forward) at 6 months follow-up, and balance
(postural sway eyes open) at the end of 3 weeks of
treatment (Appendix 4).
These recommendations are in agreement with
those of St. Joseph’s Health Care London,73 which
found strong evidence that balance training post
stroke improves outcomes. Other preexisting guide-
lines could not be compared to those of the Ottawa
Panel, because they did not apply (Appendix 4).
treatment, 2 weeks (clinically important benefit
demonstrated without statistical significance). Pa-
tients with subacute and post-acute stroke.
Tape-recorded imagery versus control, level I
(RCT): Grade C for motor function at end of
treatment, 4 weeks (no benefit demonstrated). Pa-
tients with chronic post-stroke.
Sensory function training for the hand versus
control, level II (CCT): Grade A for sensory func-
tion (change in location of touch, change in two-
point discrimination, change in sense of elbow
46 TOPICS IN STROKE REHABILITATION/SPRING 2006
position, and change in stereognosis) at end of
treatment, 6 weeks (clinically important benefit
demonstrated). Patients with chronic post-stroke.
Sensory motor integrative intervention using
the affected upper extremity versus functional
intervention using compensation techniques,
level I (RCT): Grade C for functional ability-meal
preparation at end of treatment, 4 weeks and 8
weeks (no benefit demonstrated); and grade D for
functional status (Barthel Index) at end of treat-
ment, 4 weeks and 8 weeks (no benefits demon-
strated but favoring control). Patients with sub-
acute stroke.
Perceptual learning exercise versus control,
level I (RCT): Grade C+ for total locus length eyes
open and enveloped area eyes open at end of treat-
ment, 2 weeks (clinically important benefit demon-
strated without statistical significance); and grade C
for total locus length eyes closed, enveloped area
eyes closed, rectangular area eyes open and eyes
closed at end of treatment, 2 weeks (no benefit
demonstrated). Patients with post-acute stroke.
Rocking chair stimulation versus control,
level I (RCT): Grade A for Brunnstrom-Fugl-
Meyer and Action Research Arm Test at follow-up,
Summary of trials
In total, seven RCTs (n = 318) and two CCTs (n =
59) examining the efficacy of sensory interventions
post stroke were included.297–305 The sensory inter-
ventions tested in the trials included passive vesti-
bular training to improve functional ambulation,
auditory feedback to reduce visual neglect, visual
attention retraining to assist driving performance,
perceptual learning exercises to improve standing
balance, mental practice of upper extremity exer-
cises using taped visual imagery, retraining of
hand sensory function, repetitive sensory-motor
training (rocking chair stimulation), and sensory-
motor integrative treatment (Appendix 3F).
Treatment schedules ranged from 5 to 10 ses-
sions per week lasting 40 minutes each, over an 8-
week period. Patients were recruited in the sub-
acute, post-acute, or chronic phases of stroke
(Appendix 3F).
Altogether, 14 studies were excluded (SI-Table
43 weeks and 5 years, Brunnstrom-Fugl-Meyer log
at follow-up 17 weeks, number of patients im-
proved more than 10% on Brunnstrom-Fugl-
Meyer log at follow-up, 17 weeks and 43 weeks
(clinically important benefit demonstrated); grade
C for Brunnstrom-Fugl-Meyer log at end of treat-
ment, 3 weeks and 6 weeks, and follow-up, 43
weeks (no benefit demonstrated); grade D+ for
Barthel Index at follow-up, 43 weeks and 5 years
(clinically important benefits favoring control
demonstrated without statistical significance). Pa-
tients with subacute stroke.
Visual attention training versus
visuoperception training, level I (RCT): Grade A
for reaction time at end of treatment 5–10 weeks
(clinically important benefit demonstrated); and
grade C for visual perception skills, visual scan-
ning/visual attention (Single Letter Cancellation
Test and Double Letter Cancellation Test), right/
left directional orientation, visual conceptual/
visuomotor tracking (Trail Making Test A and
Trail Making Test B), selective attention/visual
scanning, and visual attention processing at end of
treatment, 5–10 weeks (no benefit demonstrated).
Patients with post-acute stroke.
107). Three of these studies lacked a control
group,175,306,307 two studies provided insufficient
statistical data,308,309 two studies assigned fewer
than five patients per group,310,311 one study did
not measure any outcomes of interest,312 one study
examined a mixed population of patients with
stroke and head injury,313 one study was a case-
series,314 and in two studies the subjects served as
their own controls.315,316 Other reasons for exclu-
sion are detailed in SI-Table 107.317,318
Efficacy
One RCT (n = 18)297 evaluated the effect of eye
movement feedback versus control in subacute,
right-hemispheric stroke patients with left-sided
visual neglect. No benefit was demonstrated for
proportion of eye movements to the left/right or
visual attention (behavioral subtests of Behaviour
Inattention Test) at the end of 4 weeks of treatment
(SI-Table 108, Figure 98). A clinically unimpor-

tant and statistically insignificant trend favoring
control was observed for the conventional subtests
of the Behavioral Inattention Test.
After 2 weeks of treatment, subacute and post-
acute stroke patients who were subjected to pas-
sive vestibular stimulation showed clinically im-
portant improvements without statistical
significance in mobility as measured by the func-
tional ambulation profile (one CCT, n = 20)300 (SI-
Table 109, Figure 99). The RD was 52%. No
other outcome measures were assessed.
The use of taped visual imagery in chronic
stroke patients with right arm hemiparesis pro-
duced no benefit in motor function (% improve-
ment in Fugl-Meyer Score) at the end of 4 weeks
(one RCT, n = 16; SI-Table 110, Figure 100).98
Patients in the experimental group listened to
tape-recorded visual imagery of their affected arm
being used in various tasks. The control group, on
the other hand, listened to a tape of introductory
information about stroke that required attention
and participation.
Retraining of hand sensory function in
chronic stroke patients demonstrated statistically
significant and clinically important improvements
in localization of touch, two-point discrimination,
elbow position sense (change in elbow flexion),
and stereognosis at the end of 6 weeks of treatment
(one CCT, n = 39; 171%, 125%, 187%, and 209%
RD, respectively) (SI-Table 111, Figure 101).305
In subacute stroke, a sensory-motor integra-
tive approach showed no benefit compared to a
functional approach for functional ability (meal
preparation) at the end of 4 and 8 weeks (one RCT,
n = 90; SI-Table 112, Figure 102).301 Statistically
significant but clinically unimportant benefits fa-
voring control were shown for functional status
(Barthel Index) at 4 and 8 weeks.
One RCT (n = 26)303 evaluated the effect of a
perceptual learning exercise in which blind-
folded patients were asked to estimate the hard-
ness of various sponges placed underneath the
soles of their feet. In patients with subacute stroke,
there were clinically important benefits without
statistical significance for total locus length eyes
open and enveloped area eyes open at the end of
the 2-week treatment (−20% and −18% RD, re-
spectively; SI-Table 113, Figure 103). No benefit
was found for total locus length eyes closed, envel-
Clinical Practice Guidelines 47
oped area eyes closed, and rectangular area eyes
open and eyes closed at the end of 2 weeks.
In to included articles, based on the same
study sample, subacute stroke patients were
seated in a rocking chair with the affected arm
placed in a position contrary to the typical pat-
tern of spasticity and supported by an inflatable
long-arm splint (n = 100; (SI-Tables 114&115,
Figures 104&105).298,299 Patients were asked to
rock the chair back and forth using their heels
and affected arm. Compared to control, rocking
chair stimulation demonstrated statistically
significant and clinically important benefits for
motor function (Brunnstrom-Fugl-Meyer and
Action Research Arm Test) at 43 weeks follow-
up and 5 years follow-up (317%, ≥15%, 285%,
and ≥15% RD, respectively); motor function
(Brunnstrom-Fugl-Meyer log) at 17 weeks fol-
low-up (17% RD); and motor function (number
of patients improved more than 10% on
Brunnstrom-Fugl-Meyer) at 17 weeks follow-up
and 43 weeks follow-up (31% and 33% RD,
respectively). No benefit was demonstrated for
motor function (Brunnstrom-Fugl-Meyer log)
after 3 weeks and 6 weeks of treatment or at 43
weeks follow-up. Clinically important benefits
favoring control but without statistical signifi-
cance were shown for functional status (Barthel
Index) at 43 weeks follow-up and 5 years fol-
low-up (−22% and −22% RD, respectively).
Visual attention retraining exercises were
developed using the Useful Field of View (UFoV)
software program to target visual processing
speed, divided attention, and selective attention,
skills felt to be important for driving (one RCT, n
= 86; SI-Table 116, Figure 106).302 Compared to
a control intervention (traditional, computerized
visual perception retraining), visual attention re-
training provided statistically significant and
clinically important benefits in reaction time at
the end of treatment (20 sessions or 5–10 weeks).
The RD for this outcome was −54%. No benefits
were found for visual perception skills (Motor-
Free Visual Perception Test), visual scanning/vi-
sual attention (Single Letter Cancellation Test
and Double Letter Cancellation Test), right/left
directional orientation (Road Map Test), visual
conceptual/visuomotor tracking (Trail Making
Test A and Trail Making Test B), selective atten-
48 TOPICS IN STROKE REHABILITATION/SPRING 2006
tion (Bell Test), and visual attention processing
(Charron Test).
Strength of the published evidence compared with
other guidelines
The Ottawa Panel found level I (RCTs) and level
II (CCT) evidence evaluating sensory interventions
in post-stroke rehabilitation. The recommendations
formulated by St. Joseph’s Health Care London73
related to sensory interventions were based on level
I (RCT) evidence. The quality of the evidence for
sensory interventions has not been evaluated by
other preexisting guidelines (Appendix 4).
Clinical recommendations compared with other
guidelines
The Ottawa Panel found good evidence to rec-
ommend some types of sensory-based interven-
tions for stroke rehabilitation, including sensory
function training of the hand (grade A for change
in location of touch, change in two-point dis-
crimination, change in sense of elbow position,
and change in stereognosis at the end of 6 weeks
of treatment; chronic stroke), rocking chair
stimulation (grade A for Brunnstrom-Fugl-Meyer
and Action Research Arm Test at 43 weeks and 5
years follow-up, Brunnstrom-Fugl-Meyer log at
17 weeks follow-up, and number of patients im-
proved more than 10% on Brunnstrom-Fugl-
Meyer log at 17 weeks and 43 weeks follow-up;
subacute stroke), visual attention retraining
(grade A for reaction time at the end of 20 treat-
ment sessions; post-acute stroke), passive vesti-
bular training (grade C+ for functional
ambulation at the end of 2 weeks of treatment;
Clinical practice guidelines for constraint-induced movement therapy
Constraint-induced movement therapy
(CIMT) versus standard customary occupa-
tional therapy, level I (RCT): Grade A for mo-
tor function (Action Research Arm [ARA]-pinch
and ARA-total) at end of treatment, 2 weeks
(clinically important benefit demonstrated);
grade C+ for motor function (ARA-grasp, gross
subacute and post-acute stroke), and perceptual
learning exercises (grade C+ for total locus length
eyes open and enveloped area eyes open at the
end of 2 weeks of treatment; post-acute stroke)
(Appendix 4).
The value of taped visual imagery in promoting
upper extremity motor function is uncertain; the
evidence is poor to either include or exclude this
rehabilitative strategy in patients who have entered
the chronic phase of stroke. There is also poor
evidence to include eye movement feedback in
subacute stroke patients with unilateral visual ne-
glect. Furthermore, there is poor evidence to sug-
gest an advantage of a sensory-motor integrative
approach to a functional approach in improving
functional status (Appendix 4).
The recommendations of the Ottawa Panel are
in agreement with the recommendations outlined
by St. Joseph’s Health Care of London.73 St.
Joseph’s Health Care of London73 found strong
evidence that perceptual training interventions im-
prove perceptual functioning, which is consistent
with the Ottawa Panel’s guidelines regarding sen-
sory function training of the hand. Similarly, the
Ottawa Panel’s recommendation on visual atten-
tion retraining is consistent with St. Joseph’s
Health Care of London’s73 assessment that there is
strong evidence that visual scanning techniques
improve visual neglect and produce associated im-
provements in function. Finally, St. Joseph’s
Health Care of London73 found that there is con-
flicting evidence as to whether external sensory
interventions such as eye movement feedback are
beneficial in the treatment of visual neglect. The
Ottawa Panel also found poor evidence to recom-
mend eye movement feedback for reducing visual
neglect in affected stroke patients (Appendix 4).
motor and grip) at end of treatment, 2 weeks
(clinically important benefit demonstrated with-
out statistical significance); grade C for func-
tional status (FIM upper extremity [UE] dress-
ing ability, lower extremity dressing ability,
eating, bathing, and grooming, and Barthel In-
dex) at end of treatment, 2 weeks (no benefit

demonstrated). Patients with subacute stroke,
who can perform some active finger and wrist
extension prior to CIMT.
CIMT versus bimanual training based upon
NDT, level I (RCT): Grade C for functional status,
motor function, and quality of life at end of treat-
ment, 3 weeks, and follow-up at 6 weeks and 1
year (no benefit demonstrated). Patients with
chronic stroke who can perform some active finger
and wrist extension prior to CIMT.
CIMT for 6 hours per day versus CIMT for 3
hours per day, level I (RCT): Grade C+ for
functional status at end of treatment, 2 weeks,
and follow-up at 1 month (clinically important
benefit demonstrated without statistical signifi-
cance). Patients with chronic stroke who can per-
form some active finger and wrist extension prior
to CIMT.
Summary of trials
Five RCTs involving CIMT for stroke patients
were included (n = 136).319–323 The treatment types
given were as follows: 1) CIMT versus standard
occupational therapy (n = 20)319; 2) CIMT versus
bimanual training based upon NDT (n = 62)322; 3)
CIMT given for 6 hours per day versus CIMT given
for 3 hours per day (n = 15)321; 4) CIMT versus
control (n = 16)323; and 5) CIMT versus standard
physical therapy (n = 23320; CIMT-Appendix 3G).
CIMT consisted of the upper unaffected extrem-
ity being constrained either with an arm sling and
a hand splint321–323 or padded mittens or mittens
without thumbs,319,320 while the hemiplegic arm
was used to perform functional tasks or ADLs
(CIMT-Appendix 3G). Treatment duration
ranged from 1 to 6 hours per day. and frequency
ranged from 5 to 7 days a week, over an 8-day to 2-
week period (CIMT-Appendix 3G).
In total, 13 studies were excluded for the follow-
ing reasons. For three studies, the patient served as
his or her own control324–326; while for two other
studies, there were fewer than five subjects per
group.327,328 Two other studies lacked sufficient
numerical data,23,329 and the rest were excluded for
various other reasons21,330–334 (CIMT-Table 117).
Clinical Practice Guidelines 49
CIMT versus control, level I (RCT): Grade A
for change in functional status at end of treatment,
10 days (clinically important benefit demon-
strated); grade C for motor function at end of
treatment, 10 days (no benefit demonstrated).
Chronic stroke patients who can perform some
active finger and wrist extension prior to CIMT.
CIMT versus standard physical therapy,
level I (RCT): Grade C+ for dexterity at end of
treatment, 2 months (clinically important ben-
efit demonstrated without statistical signifi-
cance); grade C for change in arm impairment
and change in hand impairment at end of treat-
ment, 2 months (no benefit demonstrated);
grade D for change in postural control impair-
ment at end of treatment, 2 months (no benefit
demonstrated, but favoring control). Patients
with acute stroke.
Efficacy
For CIMT compared to standard customary
occupational therapy for subacute stroke patients
who could perform some active wrist and finger
extension (one RCT, n = 20),319 clinically impor-
tant benefits were shown for motor function (ARA-
pinch and ARA-total) at end of treatment, 2 weeks
(86% and 33% RD, respectively). These outcomes
were also statistically significant. (CIMT-Figure
107, Table 118) Clinically important benefits
without statistical significance were demonstrated
for motor function (ARA-grasp, gross motor, and
grip) with respective RDs of 62%, 45%, and 35%.
However, no benefit was demonstrated for func-
tional status (FIM-eating, grooming, bathing,
lower and upper extremity dressing ability, and
Barthel Index) at end of treatment, 2 weeks
(CIMT-Figure 107, Table 118). No other out-
comes were measured.
No benefit was found for CIMT versus bi-
manual training based on NDT for chronic
stroke patients who could perform some active
wrist and finger extension (one RCT, n = 62)322
with respect to the following outcomes: func-
tional status (Rehabilitation Active Profile [RAP],
personal care; RAP, occupation), Motor Activity
50 TOPICS IN STROKE REHABILITATION/SPRING 2006
Log (MAL: amount of use and quality of move-
ment in ADL), motor function (ARA-test score
and Fugl-Meyer), and quality of life (MAL-prob-
lem score) at end of treatment, 3 weeks, and
follow-up, 6 weeks and 1 year. No other out-
comes were measured (CIMT-Figure 108A-C,
Table 119).
For CIMT for 6 hours per day versus CIMT
for 3 hours per day for chronic stroke patients
with some active wrist and finger extension (one
RCT, n = 15),321 a clinically important benefit was
found for functional status (MAL: amount of use
and quality of movement in ADL) at end of treat-
ment, 2 weeks, and follow-up, 1 month (26%
and 15%–28% RD, respectively). No other out-
comes were measured (CIMT-Figure 109, Table
120).
In comparing CIMT for chronic stroke pa-
tients with some active wrist and finger extension
to control (one RCT, n = 16),323 a clinically impor-
tant benefit was demonstrated for change in func-
tional status (change in MAL-ADL assessment) at
end of treatment, 10 days (22% RD). This outcome
was also statistically significant (CIMT-Figure
110, Table 121). No benefit was demonstrated for
motor function (change in Wolf Function Test) at
end of treatment, 10 days. No other outcomes
were measured.
Clinically important benefits without statistical
significance were found for CIMT versus stan-
dard physical therapy for acute stroke patients
(one RCT, n = 23)320 with respect to arm and hand
dexterity (Action Research Arm Test [ARAT]
score) at end of treatment, 2 months (40%
RD)(CIMT-Figure 111, Table 122). No benefit
was demonstrated for change in arm and hand
impairment (Chedoke-McMaster Impairment In-
ventory [CMII] score) at end of treatment, 2
months (CIMT-Figure 111, Table 122). How-
ever, results favored control for change in postural
control impairment (CMII score) at end of treat-
ment, 2 months, but no benefit was demonstrated.
No other outcomes were measured.
Strength of published evidence compared with other
guidelines
The Ottawa Panel found level I evidence (five
RCTs)319–323 for the use of CIMT in the treatment of
subacute, acute, and chronic stroke patients. Both
the VA/DoD75 and St. Joseph’s Health Care Lon-
don73 also found level I evidence (two or more
RCTs) for the use of CIMT in stroke patients. The
strength of the evidence has not been measured by
other previous guidelines (Appendix 4).
Clinical recommendations compared with other
guidelines
The Ottawa Panel found good evidence to rec-
ommend the consideration of the inclusion of
CIMT in the treatment of subacute stroke patients
with some active finger and wrist extension prior
to CIMT (grade A for motor function [ARA pinch
and total] and grade C+ for motor function [ARA-
grasp, gross motor, and grip]). The Panel also
found good evidence to recommend that CIMT be
considered as an intervention for the treatment of
chronic stroke patients with some active finger and
wrist extension prior to CIMT (grade A for change
in functional status; Appendix 4).
These Ottawa Panel recommendations are in
agreement with the guidelines set out by the VA/
DoD75 that CIMT should be considered as a treat-
ment for a select group of patients with 20º of wrist
extension and 10º of finger extension who have no
sensory and cognitive deficits. According to the
VA/DoD,75 the only demonstrated benefit to date
occurs in individuals who receive between 6 and 8
hours of daily training for at least 2 weeks. St.
Joseph’s Health Care London73 also recommended
that constraint-induced therapies, as opposed to
traditional therapies, be included as an interven-
tion for patients undergoing stroke rehabilitation
(Appendix 4).
Other preexisting guidelines could not be com-
pared, as they did not apply to the recommenda-
tions of the Ottawa Panel.

Clinical practice guidelines for shoulder subluxation
FES versus control, level I (RCT): Grade C+
for shoulder lateral ROM (SLROM), motor func-
tion, and decrease in shoulder subluxation at end
of treatment, 4 weeks, and pain relief at end of
treatment, 4 weeks, and follow-up, 8 weeks (clini-
cally important benefit demonstrated without sta-
tistical significance); grade C for SLROM and de-
crease in shoulder subluxation at follow-up, 8
weeks, and muscle bulk at end of treatment, 4
weeks, and follow-up, 8 weeks (no benefit demon-
strated); grade D for motor function at follow-up,
8 weeks (no benefit demonstrated, but favoring
control). Patients with acute stroke.
FES versus control, level I (RCT) and level II
(CCT): Grade A for SLROM, motor function (UE
function), decrease in shoulder subluxation (v-
value* difference), and improvement of muscle
tone on affected upper extremity (UE) at end of
treatment, 6 weeks, and follow-up, 6 weeks, and
decrease in shoulder subluxation**(mm) at end of
treatment, 12 and 18 weeks (clinically important
benefit demonstrated); grade B for pain relief at
end of treatment 3, 6, 12, and 24 months, and
motor function (% of patients with recovery of
motor function) and decrease in shoulder sublux-
ation (% of decreased shoulder subluxation) at end
of treatment, 6, 12, and 24 months (clinically
important benefit demonstrated). Patients with
subacute stroke.
FES versus control, level I (RCT): Grade C for
decrease in shoulder subluxation at end of treat-
ment, 12 and 18 weeks (no benefit demonstrated).
Patients with chronic stroke.
FES of the supraspinatus versus FES of the
*v-value is calculated by using a radiological technique, where the v =
vertical component of the glenohumeral alignment is determined by
measuring the vertical distance between the acromial point and the
central point of the humeral head in millimeters.
**The degree of subluxation was determined on the x-ray by measur-
ing the shortest perpendicular distance (in mm) between tangents
drawn through the inferior border of the acromion and the most
superior aspect of the head of the humerus of the affected upper
extremity. The method was generally accepted and used for finding all
“decrease in shoulder subluxation.”
Clinical Practice Guidelines 51
middle deltoid, level I (CCT): Grade C+ for
muscle force at end of treatment, 6 weeks (clini-
cally important benefit demonstrated without sta-
tistical significance); grade C for decrease in
shoulder subluxation at end of treatment, 6 weeks
(no benefit demonstrated). Patients with chronic
stroke.
FES of the supraspinatus versus no treat-
ment, level I (CCT): Grade C+ for muscle force
and decrease in shoulder subluxation at end of
treatment, 6 weeks (clinically important benefit
demonstrated without statistical significance). Pa-
tients with chronic stroke.
FES of the middle deltoid of chronic versus
no treatment, level II (CCT): Grade C+ for
muscle force and decrease in shoulder subluxation
at end of treatment, 6 weeks. Patients with chronic
stroke.
Support methods – Bobath versus
Henderson, level II (CCT): Grade C for decrease
in shoulder subluxation at end of treatment, 1 day.
Patients with post-acute stroke.
Support methods – Henderson shoulder ring
versus no support, level II (CCT): Grade B for
decrease in shoulder subluxation at end of treat-
ment, 1 day (clinically important benefit demon-
strated). Patients with post-acute stroke.
Support methods – Bobath shoulder roll ver-
sus no support, level II (CCT): Grade B for de-
crease in shoulder subluxation at end of treatment,
1 day (clinically important benefit demonstrated).
Patients with post-acute stroke.
Strapping versus no strapping, level I (RCT):
Grade C+ for pain relief at end of treatment, 6
weeks, and follow-up, 2 months (clinically impor-
tant benefit demonstrated without statistical sig-
nificance). Patients with subacute stroke.
Shoulder positioning versus no treatment,
level I (RCT): Grade C+ for ROM (active abduc-
tion [ABD]) at end of treatment, 6 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for ROM (passive external
rotation) and pain relief at end of treatment, 6
weeks (no benefit demonstrated). Patients with
subacute stroke.
52 TOPICS IN STROKE REHABILITATION/SPRING 2006

Summary of trials treatment and at 8 weeks follow-up (126% and

189% RD, respectively) (SH-Figure 112, Table
Five RCTs (n = 224)335–339 and three CCTs (n =
124). However, no benefit was demonstrated for
163)340–342 involving various treatments for shoul-
SLROM and decrease in shoulder subluxation at 8
der subluxation in post-stroke patients were in-
weeks follow-up or for muscle bulk (upper arm
cluded. The following were the categories of treat-
circumference) at the end of 4 weeks of treatment
ment for shoulder subluxation. 1) FES was applied
and at 8 weeks follow-up. No benefit was demon-
to either the supraspinatus or middle del-
strated, but results favored control for motor func-
toid,336,338–341 where pulse frequency ranged from 1
tion (motor grading scale) at 8 weeks follow-up.
to 45 Hz, and pulse duration ranged from 300–
No other outcomes were measured.
350 µs, with an on/off period of 15 seconds each
For FES treatment of the subluxed shoulder
and a ramp up/down of 2 to 3 seconds (n = 235)
in subacute stroke patients versus control (one
compared to control. Treatment sessions lasted
RCT, n = 26),336 clinically important benefits with
between 5 and 60 minutes and were given 2 to 4
statistical significance were seen for SLROM (iso-
times a day, from 5 to 7 days a week, over a 4-week
metric difference in SLROM between affected and
to 24-month period (SH-Appendix 3H). 2) The
unaffected side), motor function (UE function),
use of support methods that consisted of the
decrease in shoulder subluxation (v-value differ-
Bobath shoulder roll made of a 1-cm strip of foam
ence*) and improvement of muscle tone on the
rubber padding and the Henderson shoulder ring
affected UE (Tone scale) at the end of 6 weeks of
made of a 1.25-cm-thick piece of polyethylene
treatment and at 6 weeks follow-up with respec-
foam warmed in an oven compared to no treat-
tive RDs of 95%–100%, 64%–81% 160%–186%,
ment for 1 day (n = 26)342 (SH-Appendix 3H). 3)
and 82%–200% (SH-Figure 113, Table 125).
The use of strapping with two main supporting
Clinically important benefits with statistical sig-
tapes compared to no treatment, 24 hours a day, 7
nificance were also found for decrease in shoulder
days a week, for 6 weeks (n = 98).337 4) Shoulder
subluxation** (mm) at the end of 12 weeks and 18
positioning compared to no treatment (n = 28),335
weeks of treatment (63%–73% RD; one RCT, n =
where prolonged positioning of the affected shoul-
16)339 (SH-Figure 113, Table 125). Pain relief (%
der was used. Positions were held for 20 minutes
of patients with no pain at the end of 3, 6, 12, and
and treatment sessions took place 5 days a week
24 months of treatment [≥ 15% RD]), motor func-
over a 6-week period (SH-Appendix 3H).
tion (% of patients with recovery of motor func-
In total, 13 studies were excluded for the follow-
tion), and decrease in shoulder subluxation (% of
ing reasons. Four studies lacked a control
decreased shoulder subluxation) at the end of 3, 6,
group,343–346 and three studies did not include an
12, and 24 months of treatment (≥ 15% and 30%–
intervention.347–349 Another two were excluded be-
60% RD) were also shown to have clinically im-
cause not enough data were available.350,351 The
portant benefits with statistical significance (one
remaining studies were not included for various
CCT, n = 120340; (SH-Figure 113, Table 125). No
reasons352–355 (SH-Table 123).
other outcomes were measured.

Efficacy
Clinically important benefits without statistical
significance were found for FES in acute stroke
patients versus control (one RCT, n = 40)335 for
SLROM (the pain free range of lateral rotation),
motor function (motor grading scale), and de-
crease in shoulder subluxation (subluxation scale)
at the end of 4 weeks of treatment (17%, 15%, and
1000% RD, respectively), as well as pain relief,
(Verbal Pain Rating Scale) at the end of 4 weeks of
*v-value is calculated by using a radiological technique, where the v =
vertical component of the glenohumeral alignment is determined by
measuring the vertical distance between the acromial point and the
central point of the humeral head in millimeters.
**The degree of subluxation was determined on the x-ray by measur-
ing the shortest perpendicular distance (in mm) between tangents
drawn through the inferior border of the acromion and the most
superior aspect of the head of the humerus of the affected upper
extremity. The method was generally accepted and used for finding all
“decrease in shoulder subluxation.”

For FES treatment of the subluxed shoulder in
chronic stroke patients versus control (one RCT,
n = 16),339 no benefit was demonstrated for decrease
in shoulder subluxation at the end of 12 weeks and
18 weeks of treatment. No other outcomes were
measured (SH-Figure 114, Table 126).
Clinically important benefits without statistical
significance were observed for FES of the su-
praspinatus muscle versus FES of the middle
deltoid muscle in chronic stroke patients (one
CCT, n = 12)341 for muscle force (maximum ABD
force) at the end of 6 weeks of treatment (31%
RD; SH-Figure 115, Table 127). However, no
benefit was demonstrated for decreasing shoulder
subluxation (with and without stress) at the end
of 6 weeks of treatment. No other outcomes were
measured.
For FES applied to the supraspinatus
muscle of the same group of chronic stroke
patients versus no treatment (one CCT, n =
12),341 clinically important benefits without sta-
tistical significance were demonstrated for
muscle force (maximum ABD force) and de-
crease in shoulder subluxation (with and with-
out stress) at the end of 6 weeks of treatment.
Relative differences were 47%, 41%, and 24%,
respectively (SH-Figure 116, Table 128). No
other outcomes were measured.
Similarly, clinically important benefits without
statistical significance were found for FES applied
to the middle deltoid muscle of chronic stroke
patients versus no treatment (one CCT, n = 11)341
for muscle force (maximum ABD force) and de-
crease in shoulder subluxation (with and without
stress) at the end of 6 weeks of treatment. Relative
differences were 26%, 45%, and 29%, respectively
(SH-Figure 117, Table 129). No other outcomes
were measured.
No benefit was found for Bobath shoulder roll
versus Henderson shoulder ring use with re-
spect to decreasing shoulder subluxation among
post-acute stroke patients at the end of 1 day of
treatment (one CCT, n = 26342; SH-Figure 118,
Table 130). No other outcomes were measured.
The clinically important benefits of using a
Henderson shoulder ring versus no support
(one CCT, n = 26)342 were observed in decreasing
shoulder subluxation among post-acute stroke pa-
Clinical Practice Guidelines 53
tients at the end of the 1 day treatment (25% RD).
This outcome was also statistically significant (SH-
Figure 119, Table 131). No other outcomes were
measured.
Clinically important benefits of using a Bobath
shoulder roll versus no support (one CCT, n =
26)342 were found for decreasing shoulder sublux-
ation in post-acute stroke patients at the end of 1
day of treatment (28% RD). This outcome was also
statistically significant (SH-Figure 120, Table
132). No other outcomes were measured.
For strapping of the subluxed shoulder of
subacute stroke patients versus no strapping (one
RCT, n = 98),337 clinically important benefits with-
out statistical significance were found for pain re-
lief (pain on dressing) at the end of 6 weeks of
treatment and at 2 months follow-up (172%–
235% RD; SH-Figure 121, Table 133). No other
outcomes were measured.
Clinically important benefits without statistical
significance were observed for shoulder position-
ing of subacute stroke patients versus no treat-
ment (one RCT, n = 28)335 with respect to ROM
(active ABD) at the end of 6 weeks of treatment
(111% RD) (Figure 122, Table 134). No benefits
were demonstrated for ROM (passive external ro-
tation) and pain relief (pain on dressing) at the end
of 6 weeks of treatment. No other outcomes were
measured.
Strength of published evidence compared with other
guidelines
The Ottawa Panel found both level I (three
RCTs) 336,338,339 and level II evidence (two
CCTs)340,341 for the use of FES in reducing shoul-
der subluxation. St. Joseph’s Health Care London73
also found level I (two or more RCTs) evidence for
the use of FES for the shoulder (Appendix 4).
According to the Ottawa Panel, there is level II
evidence (one CCT)342 for the use of support meth-
ods such as Bobath and Henderson. St. Joseph’s
Health Care London73 also found level II (CCT)
evidence for support methods (Appendix 4).
The Ottawa Panel found level I (RCT) evidence for
both prolonged shoulder positioning335 and strap-
ping.337 St. Joseph’s Health Care London73 also found
level I (RCT) evidence for both of these interventions.
54 TOPICS IN STROKE REHABILITATION/SPRING 2006
The strength of the evidence has not been graded by
other preexisting guidelines (Appendix 4).
Clinical recommendations compared with other
guidelines
According to the Ottawa Panel, there is good
evidence that FES should be included as an inter-
vention for the reduction of shoulder subluxation
in subacute stroke patients (grade A for SLROM,
motor function, decrease in shoulder subluxation,
and increase in muscle tone, and grade B for pain
relief and motor function). This recommendation
concurs with the Heart and Stroke Foundation of
Ontario71 and the St. Joseph‘s Health Care London
guidelines73 (Appendix 4).
Based upon fair evidence, the Ottawa Panel rec-
ommends that the Bobath shoulder roll and the
Henderson shoulder ring be considered as an in-
tervention for the reduction of shoulder sublux-
ation (grade B for decrease in shoulder sublux-
ation). However, as the Ottawa Panel also found
poor evidence (grade C for decrease in shoulder
subluxation) to support either the inclusion or
exclusion of shoulder slings as an intervention for
post-acute stroke patients, it is also in partial
agreement with the Heart and Stroke Foundation
of Ontario71 and St. Joseph’s Health Care Lon-
don,73 which found that there is limited evidence
that shoulder slings influence clinical outcomes
(Appendix 4).
Clinical practice guidelines for electrical stimulation
FES of quadriceps and rectus femoris com-
bined with positional feedback stimulation
training versus control for the knee, level I
(RCT): Grade C+ for ROM (selective knee exten-
sion: extension of the knee without any help of the
synergistic muscles) at end of treatment, 2 and 4
weeks, and knee extension torque and ROM (syn-
ergistic ROM extension: extension of the knee with
possible help of synergistic muscles) at end of
treatment, 1, 2, 3, and 4 weeks (clinically impor-
tant benefit demonstrated without statistical sig-
nificance); grade C was given for ROM (selective
The Ottawa Panel found fair evidence to support
the use of shoulder strapping for patients with
subacute stroke (grade C+ for pain relief). This is
in partial agreement with St. Joseph’s Health Care
London,73 which concluded that shoulder strap-
ping does not result in significant reductions in
pain or improve upper limb function or ROM
(Appendix 4).
Finally, the Ottawa Panel found fair evidence
regarding the use of positioning of the shoulder as
an intervention for patients with post-acute stroke
(grade C+ for ROM, active ABD, and grade C for
ROM, passive external rotation, and pain relief).
This is in partial agreement with the finding of St.
Joseph’s Health Care London73 that prolonged po-
sitioning does not result in significant reductions
in pain or improve upper limb function or range of
motion. The Ottawa Panel is however in agree-
ment with the consensus opinion* of St. Joseph’s
Health Care London that proper positioning of the
hemiplegic shoulder does indeed help prevent
subluxation (Appendix 4).
Other preexisting guidelines concerning shoul-
der subluxation did not apply to those of the Ot-
tawa Panel.
*According to St. Joseph’s Health Care London, consensus opinion
refers to the agreement of a group of experts on an appropriate course
of treatment in the absence of evidence. It is regarded as the lowest
form of evidence and can be considered as no evidence at all.
extension) at end of treatment, 1 week (no benefit
demonstrated); grade D+ for ROM (selective ex-
tension) at end of treatment, 3 weeks (clinically
important benefit favoring control demonstrated
without statistical significance). Patients with post-
acute stroke.
Positional feedback stimulation training of
wrist extensors versus control, level I (RCT):
Grade A for torque and ROM at end of treatment,
2, 3, and 4 weeks. Patients with subacute and post-
acute stroke.
Neuromuscular electrical stimulation of the

wrist and finger extensors versus control, level I
(RCT): Grade A for grip strength (grip in kilo-
grams) at end of treatment, 8 weeks, torque (mo-
ment of extension at 15º) at end of treatment, 8
weeks, and follow-up, 24 weeks, and motor func-
tion (ARAT total) at end of treatment, 8 weeks
(clinically important benefit demonstrated); grade
C+ for ROM (resting wrist angle) at follow-up, 24
weeks, ROM (maximum active extension) at end
of treatment, 8 weeks, and follow-up, 24 weeks,
torque (moment of extension at 0º) at end of treat-
ment, 8 weeks, and follow-up, 24 weeks, motor
function (ARAT-grasp) at end of treatment, 8
weeks, and follow-up, 24 weeks, motor function
(ARAT-grip, pinch, gross movement, and total
score) at end of treatment, 8 weeks, functional
status at follow-up, 24 weeks, and motor function
(ARAT total) at follow-up, 24 weeks (clinically
important benefit demonstrated without statistical
significance); grade C was given for torque (mo-
ment of extension at 30o), ROM (maximum pas-
sive extension of the wrist and resting wrist angle),
functional status at end of treatment, 8 weeks,
motor function (ARAT-grip, pinch, and gross
movement) at follow-up, 24 weeks, decrease of
visuospatial neglect at end of treatment, 8 weeks,
and follow-up, 24 weeks, decrease of spasticity at
end of treatment, 8 weeks, and follow-up, 24
weeks (no benefit demonstrated); grade D was
given for ROM (maximum passive extension of the
wrist) at follow-up, 24 weeks (no benefit demon-
strated but favoring control); grade D+ was given
for grip strength at follow-up, 24 weeks (clinically
important benefit favoring control demonstrated
without statistical significance). Patients with sub-
acute and post-acute stroke.
FES of anterior tibialis and gastrocnemius
combined with EMG-BFB versus FES of ante-
rior tibialis and gastrocnemius for gait training,
level I (RCT): Grade A for stride length at end of
treatment, 6 weeks (clinically important benefit
demonstrated); grade C for ankle and knee angle
at end of treatment, 6 weeks; grade D for gait cycle
time at end of treatment, 6 weeks (no benefit
demonstrated but favoring control). Patients with
chronic stroke.
EMG-triggered electrical muscle stimulation
training versus control, level I (RCT): Grade A
Clinical Practice Guidelines 55
for active ankle ROM at end of treatment, 5 weeks
(clinically important benefit demonstrated); grade
C+ for active wrist ROM and functional status at
end of treatment, 5 weeks (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for decrease in spasticity at end of
treatment, 5 weeks (no benefit demonstrated). Pa-
tients with subacute stroke.
FES of the posterior deltoid and supraspina-
tus for the shoulder versus control, level I
(RCT): Grade A for active ROM at end of treat-
ment, 6 weeks, decrease in shoulder subluxation at
end of treatment, 6 weeks, and difference in active
ROM between affected and nonaffected shoulder
at end of treatment, 6 weeks, and follow-up, 6
weeks (clinically important benefit demonstrated);
grade C+ for active ROM at end of treatment, 3
weeks, and follow-up, 6 weeks, shoulder muscle
tone at end of treatment, 3 and 6 weeks, and
follow-up, 6 weeks (clinically important benefit
demonstrated without statistical significance). Pa-
tients with chronic stroke.
NMES of chronic stroke patients for the hand
versus placebo, level I (RCT): Grade C+ for
strength, functional status, and motor function
(turning pages, small objects, and stacking) at end
of treatment, 3 weeks (clinically important benefit
demonstrated without statistical significance);
grade C for dexterity at end of treatment, 3 weeks
(no benefit demonstrated); grade D+ for motor
function (feeding, tracking accuracy) at end of
treatment, 3 weeks (clinically important benefit
favoring control, but demonstrated without statis-
tical significance); grade D for motor function
(light cans, heavy cans) at end of treatment, 3
weeks (no benefit demonstrated but favoring con-
trol). Patients with chronic stroke.
FES of the finger thumb extensors and flex-
ors for the hand versus control, level I (RCT):
Grade A for functional status (average successful
trials for higher and lower functioning groups) at
end of treatment, 3 weeks, and follow-up, 3, 10,
and 23 weeks, coordination for lower functioning
group at follow-up, 3, 10, and 23 weeks, decrease
in spasticity for higher functioning group at fol-
low-up, 23 weeks, and functional status (% of
maximum score for higher and lower functioning
group) at follow-up, 23 weeks (clinically impor-
56 TOPICS IN STROKE REHABILITATION/SPRING 2006
tant benefit demonstrated); grade C+ for coordi-
nation for lower functioning group at end of treat-
ment, 3 weeks; grade C for decrease in spasticity
for lower functioning group at end of treatment, 3
weeks, and follow-up, 3, 10, and 23 weeks (no
benefit demonstrated). Patients with subacute and
post-acute stroke.
Blocked practice combined with NMES of the
extensor communis digitorium/extensor carpi
ulnaris/triceps brachii/anterior and middle del-
toid versus random practice combined with
NMES of extensor communis digitorium/exten-
sor carpi ulnaris/triceps brachii/anterior and
middle deltoid, level I: Grade A for motor reaction
time (time starting directly after pre-motor time and
ended with movement initiation at bilateral peak
force) favoring random practice at end of treatment,
2 weeks (clinically important benefit demon-
strated); grade C+ for pre-motor reaction time
(time from stimulus onset until the EMG activity of
the muscles reaches 30% of unilateral peak reac-
tion) favoring blocked practice at end of treatment,
2 weeks (clinically important benefit demonstrated
without statistical significance); grade C for pre-
motor bilateral reaction time and unilateral motor
reaction time at end of treatment, 2 weeks, (no
benefit demonstrated). Patients with chronic stroke.
Blocked practice combined with NMES of the
extensor communis digitorium/extensor
carpiulnaris/triceps brachii/anterior and
middle deltoid versus passive/active ROM, level
Summary of trials
Eleven RCTs involving FES of stroke patients
compared to placebo or no treatment as control (n
= 385) were included.51,356–365 The types of FES
treatment given were as follows. 1) Electrical
stimulation of the quadriceps and rectus femoris
was combined with positional feedback stimula-
tion training (PFST) for the knee and compared to
control (n = 40).365 2) PFST of the wrist extensors
was compared to control (n = 30).357 3) NMES of
the wrist and finger extensors was compared to
control (n = 60).363 4) NMES for the hand was
compared to placebo (n = 16).362 5) FES was ap-
I (RCT): Grade C+ for unilateral pre-motor reac-
tion time and bilateral motor reaction time at end
of treatment, 2 weeks (clinically important benefit
demonstrated without statistical significance);
grade C for bilateral pre-motor reaction time at
end of treatment, 2 weeks (no benefit demon-
strated); grade D for unilateral motor reaction
time at end of treatment, 2 weeks (no benefit
demonstrated but favoring control). Patients with
chronic stroke.
Random practice combined with NMES of the
extensor communis digitorium/extensor carpi
ulnaris/triceps brachii/anterior and middle del-
toid versus passive/active ROM, level I (RCT):
Grade C+ for bilateral motor reaction time at end of
treatment, 2 weeks (clinically important benefit
demonstrated without statistical significance);
grade C for bilateral pre-motor reaction time and
unilateral pre-motor reaction time at end of treat-
ment, 2 weeks (no benefit demonstrated but favor-
ing control). Patients with chronic stroke.
Electrical stimulation of extensors versus al-
ternating electrical stimulation of flexors and
extensors, level I (RCT): Grade A for decrease in
spasticity at end of treatment, 6 weeks (clinically
important benefit demonstrated); grade C for de-
crease in spasticity at follow-up, 6 weeks, and
manual dexterity, motor impairment, grip muscle
force, and wrist mobility at end of treatment, 6
weeks, and follow-up, 6 weeks (no benefit demon-
strated). Patients with chronic stroke.
plied to the posterior deltoid and supraspinatus for
the shoulder versus control (n = 89),356,360 as well as
the finger and/or thumb flexors and extensors for
the hand versus control (n = 28).363 6) FES of the
anterior tibialis and gastrocnemius was also com-
bined with EMG-BFB for gait training and com-
pared to FES of the anterior tibialis and gastrocne-
mius (n = 16). 51 Transcutaneous electrical
stimulation of the arm, forearm, knee flexors, and
ankle extensors triggered by EMG was compared
to control (n = 28).361 8) blocked practice was
combined with neuromuscular stimulation of ex-
tensor communis digitorium/extensor carpi
ulnaris/triceps brachii/anterior and middle deltoid

and compared to random practice combined with
NMES of extensor communis digitorium/extensor
carpi ulnaris/triceps brachii/anterior and middle
deltoid (n = 28).358 Blocked practice combined
with NMES was also compared to active and pas-
sive ROM (n = 20), as was random practice com-
bined with NMES (n = 20).358 9) Electrical stimula-
tion of extensors was compared to alternating
electrical stimulation of flexors and extensors (n =
28)359 (ES-Appendix 3I).
During treatment, the frequency of electrical
stimulation ranged from 36 to 80 Hz, with a pulse
duration varying from 0.3 to 200 ms. Intensity
varied from 20 mA to 60 mA. Treatment sessions
lasted from 20 to 60 minutes, with the number of
sessions ranging from 8 to 112, over a 3- to 8-week
period (ES-Appendix 3I).
Twenty-six trials were excluded for the follow-
ing reasons. Ten trials were excluded because they
lacked a control group.366–375 Another eight trials
were excluded because there were fewer than five
subjects per group.376–383 Two studies were elimi-
nated because they were literature reviews.59,60 The
remaining studies were excluded for other rea-
sons384–389 (ES-Table 135).
Efficacy
For FES of the quadriceps and rectus femoris
combined with positional feedback stimulation
training for the knee of post-acute stroke patients
versus control (one RCT, n = 40),365 a clinically
important benefit without statistical significance
was shown for ROM (selective knee extension) at
end of treatment, 2 and 4 weeks (18%–41% RD),
knee torque (extension) and ROM (synergistic ex-
tension) at end of treatment, 1, 2, 3, and 4 weeks,
with relative differences (RDs) of 96%–199% and
21%–31% respectively. No benefit was demon-
strated for ROM (selective extension) at end of
treatment, 1 week. A clinically important benefit
without statistical significance favoring control
was found for ROM (selective extension) at end of
treatment, 3 weeks. No other outcomes were mea-
sured (ES-Figure 123, Table 136).
Clinically important benefits were found for
PFST of the wrist extensors of subacute and
post-acute stroke patients versus control (one
Clinical Practice Guidelines 57
RCT, n = 30)357 with respect to change in torque
(change in 30º torque flexion and change in 30º
torque extension) and ROM (change in selective
ROM, change in synergistic ROM) at end of treat-
ment, 2, 3, and 4 weeks. RDs were 96%–143%,
87.5%–137%, 124%–150%, and 107%–133%, re-
spectively. These outcomes were also statistically
significant (ES-Figure 124A&B, Table
137A&B). No other outcomes were measured.
For neuromuscular stimulation of the wrist
and finger extensors in subacute stroke patients
versus control (one RCT, n = 60),364 clinically
important benefits were demonstrated for grip
strength (grip in kilograms) at end of treatment, 8
weeks (102% RD); motor function (ARAT total) at
the end of 8 weeks of treatment (272% RD); and
torque (moment of extension at 15º) at end of
treatment, 8 weeks, and follow-up, 24 weeks
(1179%–1878% RD, respectively). These out-
comes were also statistically significant (ES-Figure
125A, Table 138A). Clinically important benefits
without statistical significance were found for
ROM (resting angle) at 24 weeks follow-up (58%
RD); and torque (moment of extension at 0º) at the
end of 8 weeks of treatment and at the 2-week
follow-up with respective RDs of 222% and 284%.
Clinically important benefits were also found for
ROM (maximum active extension of the wrist) at
end of treatment, 8 weeks, and follow-up, 24
weeks, where RD ranged from 17%–22%; motor
function (ARAT-grasp) at end of treatment, 8
weeks, and follow-up, 24 weeks (400%–200%
RD); motor function (ARAT-grip-pinch-gross
movement-total score) at end of treatment, 8
weeks ( 15% RD for all); motor function (ARAT
total) at follow-up, 24 weeks (160% RD); and
functional status measured by the Barthel Index at
follow-up, 24 weeks (RD 40%) (ES-Figure
125A,C,D, Table 138A,C,D). No benefit was
demonstrated for ROM (maximum passive exten-
sion and resting wrist angle) and functional status
(Barthel Index) at the end of 8 weeks of treatment,
torque (moment of extension at 30º) at the end of
8 weeks of treatment and at 24 weeks follow-up,
motor function (ARAT-grip, pinch, gross move-
ment) at 24 weeks follow-up, decrease of
visuospatial neglect (Stars Cancellation Test) at the
end of 8 weeks of treatment and at 24 weeks
58 TOPICS IN STROKE REHABILITATION/SPRING 2006
follow-up, and decrease of spasticity at the end of
8 weeks of treatment and at 24 weeks follow-up
(ES-Figure 125B&D, Table 138B&D). There
was no clinical benefit, but results favored control
for ROM (maximum passive extension of the
wrist) at 24 weeks follow-up (ES-Figure 125C,
Table 138C). Clinically important benefits with-
out statistical significance favoring control were
shown for grip strength at 24 weeks follow-up
(ES-Figure 125E, Table 138E). No other out-
comes were measured.
For FES of the anterior tibialis and gastrocne-
mius combined with EMG-BFB versus FES of
the anterior tibialis and gastrocnemius of chronic
stroke patients for gait training (one RCT, n =
16),51 clinically important benefits were demon-
strated for stride length at end of treatment, 6
weeks (17% RD). This outcome was also statisti-
cally significant (ES-Figure 126, Table 139). No
clinically important benefit was demonstrated for
ankle and knee angle (swing phase) at end of
treatment, 6 weeks. However, results favored con-
trol without clinical benefit for gait cycle time at
end of treatment, 6 weeks. No other outcomes
were measured.
Clinically important benefits with statistical sig-
nificance were found for EMG-triggered electri-
cal muscle stimulation training of the upper
arm, forearm, knee flexors, and ankle extensors
of patients with subacute stroke versus control
(one RCT, n = 28)361 for active ankle ROM (dorsi-
flexion) at end of treatment, 5 weeks (61% RD)
(ES-Figure 127, Table 140). Clinically important
benefits without statistical significance were dem-
onstrated for active wrist ROM (extension) and
functional status (Barthel Index) at end of treat-
ment. 5 weeks (62% and 23% RD, respectively;
ES-Figure 127, Table 140). No benefits were
demonstrated for decrease of spasticity (upper and
lower extremities) at end of treatment, 5 weeks.
No other outcomes were measured.
Clinically important benefits were found for
FES of the posterior deltoid and supraspinatus
of subacute and post-acute stroke patients for the
shoulder versus a control of standard physio-
therapy (two RCTs, n = 89)356,360 with respect to
the following outcomes: active ROM at end of
treatment, 6 weeks (667% RD); difference in active
ROM between affected and nonaffected shoulder
at the end of 6 weeks of treatment and at the 6-
week follow-up (67% and 73% RD, respectively);
and decrease in shoulder subluxation at the end of
6 weeks of treatment (44% RD). These outcomes
were also statistically significant (ES-Figure
128A&B, Table 141A&B). Clinically important
benefits without statistical significance were dem-
onstrated for active ROM at end of treatment, 3
weeks, and follow-up, 6 weeks, with respective
RDs of 93% and 480%; difference in active ROM
between affected and nonaffected shoulder at fol-
low-up, 6 weeks (73% RD); and shoulder muscle
tone measured by the Ashworth Scale at end of
treatment, 3 and 6 weeks, and follow-up, 6 weeks
(162%–200% RD; ES-Figure 128A&B, Table
141A&B). No other outcomes were measured.
Clinically important benefits without statistical
significance were found for NMES of chronic
stroke patients for the hand versus placebo (one
RCT, n = 16)362 with respect to the following out-
comes: strength at end of treatment, 3 weeks (23%
RD); functional status measured by the MAL indi-
cating the amount of use and quality of movement
at end of treatment, 3 weeks (34% and 28% RD,
respectively); and motor function measured by the
Jebsen Taylor Hand Test involving page turning,
small objects, and stacking at end of treatment, 3
weeks. RDs were 19%, 20%, and 76%, respec-
tively (ES-Figure 129A&B, Table 142). No ben-
efit was demonstrated for dexterity (box/block) at
end of treatment, 3 weeks. However, clinically
important benefits without statistical significance
were demonstrated for control with respect to mo-
tor function (feeding, tracking, and accuracy) at
end of treatment, 3 weeks. Similarly, for motor
function measured through the Jebsen Taylor
Hand Test involving heavy and light cans at the
end of the 3-week treatment, results favored con-
trol though no clinical benefit was demonstrated.
No other outcomes were measured.
For FES of the finger and thumb extensors and
flexors of subacute and post-acute stroke patients
versus functional tasks without FES for the
hand (one RCT, n = 14),363 clinically important
benefits were found for the following outcomes:
functional status for higher and lower functioning
groups at end of treatment, 3 weeks, and follow-

up, 3, 10, and 23 weeks, measured through the
average number of successful trials in the Upper
Extremity Function Test (RDs were 155%–273%
and ≥ 15% RD, respectively); coordination for the
higher and lower functioning groups measured
through the drawing test at end of treatment, 3
weeks, and follow-up, 3, 10, and 23 weeks (ES-
Figure 130A&B, Table 143). RDs were 32%–
38% for the higher group and 63%–73% for the
latter group (ES-Figure 130A&B, Table 143).
Clinically important benefits were also found for
the decrease in spasticity of the higher function-
ing group measured by the Ashworth Scale at
follow-up, 23 weeks (32% RD); and functional
status for higher and lower functioning group
measured by the percentage maximum score on
the MAL at follow-up, 23 weeks (209%–237%
and ≥ 15% RD, respectively; ES-Figure
130A&B, Table 143). Clinically important ben-
efits without statistical significance were demon-
strated for decrease in spasticity for lower func-
tioning group (Ashworth Scale) at follow-up, 23
weeks. No other outcomes were measured (ES-
Figure 130A&B, Table 143).
For blocked practice combined with NMES of
the extensor communis digitorium/extensor
carpi ulnaris/triceps brachii/anterior and
middle deltoid versus random practice com-
bined with NMES of extensor communis
digitorium/extensor carpi ulnaris/triceps brachii/
anterior and middle deltoid of chronic stroke pa-
tients (one RCT, n = 28),358 clinical benefits favor-
ing random practice were demonstrated for bilat-
eral motor reaction time at end of treatment, 2
weeks (16% RD). This outcome was also statisti-
cally significant (ES-Figure 131, Table 144). A
clinically important benefit without statistical sig-
nificance favoring blocked practice was demon-
strated for unilateral pre-motor reaction time at
end of treatment, 2 weeks (ES-Figure 131, Table
144). No other outcomes were measured.
Clinically important benefits without statistical
significance were found for blocked practice
combined with NMES of the extensor commu-
nis digitorium/extensor carpi ulnaris/triceps
brachii/anterior and middle deltoid of chronic
stroke patients versus passive/active ROM (one
RCT, n = 20),358 with respect to unilateral and
Clinical Practice Guidelines 59
bilateral pre-motor reaction time at end of treat-
ment, 2 weeks (15% and 16% RD, respectively;
ES-Figure 132, Table 145). No benefit was dem-
onstrated for unilateral pre-motor reaction time
at end of treatment, 2 weeks. However, results
also favored control for unilateral motor reaction
time at end of treatment, 2 weeks, although no
benefit was demonstrated. No other outcomes
were demonstrated.
For random practice combined with NMES of
the extensor communis digitorium/extensor
carpi ulnaris/triceps brachii/anterior and
middle deltoid of chronic stroke patients versus
passive/active ROM (one RCT, n = 20),358 clini-
cally important benefits without statistical signifi-
cance were found for bilateral motor reaction time
at end of treatment, 2 weeks (34% RD) (ES-Figure
133, Table 146). No benefits were demonstrated
for bilateral and unilateral pre-motor reaction time
at end of treatment, 2 weeks. No other outcomes
were measured.
For the electrical stimulation of extensors
versus alternating electrical stimulation of flex-
ors and extensors of chronic stroke patients (one
RCT, n = 28),359 clinically important benefits were
found for decrease in spasticity measured by the
Ashworth Scale at end of treatment, 6 weeks
(100% RD). This outcome was also statistically
significant (ES-Figure 134A&B, Table 147). No
benefit was demonstrated for decrease in spasticity
at follow-up, 6 weeks; manual dexterity (ARAT);
grip muscle force (hand ratio); motor impairment
(Motricity Index); and wrist mobility (active
ROM) at end of treatment, 6 weeks, and follow-up,
6 weeks. No other outcomes were measured.
Strength of published evidence compared with other
guidelines
The Ottawa Panel found level I (one RCT) evi-
dence for the use of FES for gait training.51 The
strength of the evidence was also found to be level
I (RCT) by both St. Joseph‘s Health Care London,73
whose findings for FES for gait training were based
upon the results of two or more RCTs, and the
Royal College of Physicians.72 However, the VA/
DoD75 found level II (one CCT) evidence for the
use of FES in gait training and level I (one RCT)
60 TOPICS IN STROKE REHABILITATION/SPRING 2006
evidence for the use of multiple channel FES for
gait impairment75 (Appendix 4).
The Ottawa Panel also found level I (two
RCTs)356,360 evidence for FES for the shoulder. The
VA/DoD,75 the Heart and Stroke Panel on the Man-
agement of the Hemiplegic Arm and Hand,71 and
St. Joseph’s Health Care London73 also found level
I (RCT) evidence for FES used for shoulder sub-
luxation (Appendix 4).
Although the Ottawa Panel found level I
(RCT) evidence for FES for the knee and hand,
positional feedback stimulation training of the
wrist and finger extensors, NMES for the hand,
and EMG-BFB triggered electrical muscle stimu-
lation, the strength of the evidence of these in-
terventions has not been graded by other guide-
lines (Appendix 4).
Clinical recommendations compared with other
guidelines
According to the Ottawa Panel, there is good
evidence to recommend that FES of the ankle
dorsiflexors be considered as an intervention for
gait training in chronic stroke patients (grade A for
stride length). This recommendation is in agree-
ment with the clinical practice guidelines of both
the VA/DoD75 and St. Joseph’s Health Care Lon-
don73 that also recommend the use of FES for gait
training. However, the Panel disagrees with the
finding of the VA/DoD that there is insufficient
evidence* as to whether multi-channel FES should
or should not be included as an intervention for
*According to VA/DoD, a grading of insufficient evidence (I) indicates
that one cannot determine from the evidence the net benefit of the
intervention, either for or against, and therefore clinicians must use
their judgment in either including or excluding it in the treatment of
the patient.
severe hemiplegic patients with gait impairment
(Appendix 4).
With respect to the use of FES in the treatment
of the shoulder, the Ottawa Panel is in agreement
with the VA/DoD,75 the Heart and Stroke Panel on
the Management of the Hemiplegic Arm and
Hand71 and St. Joseph’s Health Care London73 that
FES be included as an intervention for the treat-
ment of the shoulder in subacute and post-acute
stroke patients (grade A for active ROM, decrease
in shoulder subluxation, and grade C+ for active
ROM, difference in active ROM between affected
and nonaffected shoulder and shoulder muscle
tone; Appendix 4).
According to the Ottawa Panel, there is good
evidence to recommend the use of FES applied to
the finger and thumb flexors and extensors as an
intervention for the hands of subacute and post-
acute stroke patients (grade A for functional status,
decrease in spasticity, and coordination for the
higher and/or lower functioning groups). Based
upon good evidence, the Ottawa Panel also recom-
mends the consideration of the inclusion of EMG-
triggered electrical muscle stimulation of the upper
arm, forearm, knee flexors, and ankle extensors
(grade A for active ankle ROM, and grade C+ for
active wrist ROM and functional status). These
guidelines are in line with the recommendation
made by the Heart and Stroke Foundation of
Ontario71 that EMG-BFB and electrical stimulation
of the wrist and the forearm be included as an
interventions to reduce motor impairment and im-
prove functional motor recovery (Appendix 4).
Finally, the Ottawa Panel disagrees with the
Royal College of Physicians72 that FES should not
be routinely considered as an intervention for
stroke patients. Other preexisting guidelines con-
cerning electrical stimulation do not apply to those
of the Ottawa Panel (Appendix 4).

Clinical practice guidelines for transcutaneous electrical nerve stimulation (TENS)
Low-intensity TENS applied to lower extrem-
ity versus placebo, level I (RCT): Grade A for
decrease in spasticity for lower extremity at end of
treatment, 2 and 3 weeks (clinically important
benefit demonstrated); grade C+ for dorsiflexor
strength at end of treatment, 3 weeks (clinically
important benefit demonstrated, without statisti-
cal significance). Patients with chronic stroke.
Low-intensity TENS applied to the elbow and
ankle versus placebo, level I (RCT): Grade A for
functional status at end of treatment, 8 weeks
(clinically important benefit demonstrated). Pa-
tients with post-acute stroke.
Low-intensity TENS applied to the shoulder
versus placebo, level I (RCT): Grade C for passive
shoulder ROM at end of treatment, 4 weeks, and at
follow-up, 4 weeks (no benefit demonstrated). Pa-
tients with post-acute stroke.
Low-intensity TENS applied to the upper ex-
tremity versus control, level I (RCT): Grade C
for motor function (Fugl-Meyer) and functional
status at end of treatment, 3 months, and follow-
up, 3 years, as well as motor function (change in
Fugl-Meyer) and decrease in spasticity (elbow flex-
ors) at end of treatment, 3 months (no benefit was
demonstrated); grade D+ for decrease in spasticity
(wrist flexors) at follow-up, 3 years (clinically im-
portant benefit favoring control, demonstrated
without statistical significance); grade D for de-
crease in spasticity (wrist flexors) at end of treat-
ment, 3 months. Patients with chronic stroke.
Summary of trials
Five RCTs68,69,390–392 and one CCT (n = 264),393
involving TENS with either a placebo or no treat-
ment as control, were included. Three types of
TENS were prescribed: 1) low-intensity TENS
(1.7–50 Hz, 125–200 µs) applied to the upper and
lower extremity, elbow, ankle, hand, and foot ver-
sus placebo or control (n = 244)68,69,390–393; 2) high-
intensity TENS (100 Hz) applied to the shoulder
versus placebo (n = 40)68; 3) high-intensity TENS
versus low-intensity TENS applied to the shoulder
(n = 40).68 Therapeutic applications of TENS
Clinical Practice Guidelines 61
High-intensity TENS applied to the shoulder
versus placebo, level 1 (RCT): Grade A for pas-
sive shoulder ROM at end of treatment, 4 weeks,
and follow-up, 4 weeks (clinically important ben-
efit demonstrated). Patients with post-acute
stroke.
High-intensity TENS versus low-intensity
TENS applied to the shoulder, level I (RCT):
Grade A for passive shoulder ROM (extension,
abduction, external rotation) at end of treatment, 4
weeks, and follow-up, 4 weeks, and passive shoul-
der ROM (flexion) at follow up, 4 weeks (clinically
important benefit demonstrated); grade C for pas-
sive shoulder ROM (flexion) at end of treatment, 4
weeks (no benefit demonstrated). Patients with
post-acute stroke.
Low-intensity TENS applied to the hand ver-
sus placebo, level II (CCT): Grade B for sensation
at end of treatment, 3 weeks (clinically important
benefit demonstrated); grade C for motor function
and gait speed at end of treatment, 3 weeks (no
benefit demonstrated). Patients with chronic
stroke.
Low-intensity TENS applied to the foot ver-
sus placebo, level II (CCT): Grade B for sensation
at end of treatment, 3 weeks (clinically important
benefit demonstrated); grade C for motor function
and gait speed at end of treatment, 3 weeks (no
benefit demonstrated). Patients with chronic
stroke.
ranged from 20 and 60 minutes and the number of
sessions ranged from 9 to 60, over a 3- to 12-week
period (TENS-Appendix 3J).
Five trials were excluded for the following rea-
sons. One trial was excluded due to a lack of a
control group.394 Another study was excluded be-
cause no intervention was used.395 One trial was
excluded because more than 20% of participants
were recurrent stroke victims, 396 while another
had fewer than five subjects in the control
group.397 Finally, one trial was excluded because
healthy subjects were used in the control group398
(TENS-Table 148).
62 TOPICS IN STROKE REHABILITATION/SPRING 2006
Efficacy
For low-intensity TENS applied to the lower
extremity versus placebo (one RCT, n = 17),390 a
clinically important benefit was demonstrated for
the decrease in spasticity of the lower extremity at
end of treatment, 2 and 3 weeks (15%–21% RD),
and this outcome was statistically significant for
chronic stoke patients (TENS-Figure 135, Table
149). Clinically important benefits were demon-
strated for dorsiflexor muscle force at end of treat-
ment, 3 weeks (19% RD) (TENS-Figure 135,
Table 149). However, there was no statistical sig-
nificance. No other outcomes were measured.
For low-intensity TENS applied to the elbow
and lower extremity versus placebo (one RCT, n =
60),392 clinically important benefits were shown for
the functional status of post-acute stroke patients,
measured by the Barthel Index, at end of treatment,
8 weeks (91% RD). This outcome was statistically
significant (TENS-Figure 136, Table 150). No
other outcomes were measured.
No clinical benefit was found for low-intensity
TENS applied to the shoulder versus placebo
(one RCT, n = 40)68 for passive shoulder ROM
(flexion, extension, abduction, external rotation) of
post-acute stroke patients, at end of treatment, 4
weeks, and follow-up, 4 weeks (TENS-Figure 137,
Table 151). No other outcomes were measured.
Low-intensity TENS applied to the upper ex-
tremity of chronic stroke patients versus con-
trol (two RCTs, n = 49)69,391 demonstrated no clini-
cal benefit for motor function, measured by the
Fugl-Meyer scale, and functional status measured
by the Barthel Index at end of treatment, 3 months,
and follow-up, 3 years. Equally, there was no clini-
cal benefit for motor function (change in Fugl-
Meyer) and decrease in spasticity of the elbow
flexors at end of treatment, 3 months (TENS-
Figure 138, Table 152). However, clinically im-
portant benefits favoring control without statistical
significance were demonstrated for decrease of
spasticity in the wrist flexors at follow-up, 3 years
(TENS-Figure 138, Table 152). No clinical ben-
efit was demonstrated for decrease in spasticity of
wrist flexors at the end of treatment, 3 months. No
other outcomes were measured.
For high-intensity TENS applied to the shoul-
der of post-acute stroke patients versus placebo
(one RCT, n = 40),68 clinically important benefits
were found for passive shoulder ROM (flexion,
extension, abduction, external rotation) at end of
treatment, 4 weeks, and follow-up, 4 weeks. RDs
were 16%–24%, 25%–29%, and 30%–34%, re-
spectively. These outcomes were also statistically
significant (TENS- Figure 139, Table 153). No
other outcomes were measured.
High-intensity TENS versus low-intensity
TENS applied to the shoulder of post-acute
stroke patients (one RCT, n = 40)68 demonstrated
clinically important benefits for passive shoulder
ROM (extension, abduction, and external rotation)
at end of treatment, 4 weeks, and follow-up, 4
weeks. RDs were 22%–27%, 37%–42%, and
24%–30%, respectively. These outcomes were also
statistically significant (TENS- Figure 140, Table
154). A clinically important benefit with statistical
significance (20% RD) was also shown for passive
shoulder ROM (flexion) at follow-up, 4 weeks
(TENS-Figure 140, Table 154). However, no
clinical benefit was demonstrated for passive
shoulder ROM (flexion) at end of treatment, 4
weeks. No other outcomes were measured.
A clinically important and statistically significant
benefit was shown for sensation at end of treatment,
3 weeks, when low-intensity TENS versus pla-
cebo (one CCT, n = 51)393 was applied to the hands
of chronic stroke patients. The RD for this outcome
was 37% (TENS-Figure 141, Table 155). How-
ever, no benefit was demonstrated for either the
motor function, measured by the MAS, or gait speed
(10 m walking time) at end of treatment, 3 weeks.
No other outcomes were measured.
For low-intensity TENS versus placebo ap-
plied to the foot of chronic stroke patients (one
CCT, n = 27),393 a clinically important benefit was
found for sensation at end of treatment, 3 weeks
(28% RD). This outcome was also statistically sig-
nificant (TENS-Figure 142, Table 156). How-
ever, no benefit was demonstrated for motor func-
tion, measured through the Modified Motor
Assessment Scale, and gait speed (10 m walking
time) at end of treatment, 3 weeks. No other out-
comes were measured.

Strength of published evidence compared with other
guidelines
The Ottawa Panel found level I evidence (five
RCTs)68,69,390–392 and level II evidence (one CCT)393
for the use of low-intensity TENS in the treatment
of stroke patients. There was also level I evidence
(one RCT)68 for the use of high-intensity TENS.
The strength of the evidence has also been graded
by the Royal College of Physicians,72 which also
reported level I evidence (RCTs) for high- and low-
intensity TENS. St. Joseph’s Health Care London73
also found level I evidence for TENS. The strength
of the evidence has not been reported by other
guidelines (Appendix 4).
Clinical recommendations compared with other
guidelines
According to the Ottawa Panel, there is good
evidence that low-intensity TENS applied to the
lower extremity (grade A for decrease in spasticity,
and grade C+ for dorsiflexor muscle force) and
elbow and wrist (grade A for functional status)
should be included as an intervention for stroke
patients. Based upon fair evidence, the Panel also
recommends that low-intensity TENS for the hand
and foot be specifically considered as an interven-
tion for sensation (grade B). The Ottawa Panel
Clinical Practice Guidelines 63
found good evidence to recommend the use of
high-intensity TENS applied to the shoulder
(grade A for passive shoulder ROM) as an interven-
tion for stroke patients. This is in line with the
Royal College of Physicians72 recommendation
that high-intensity TENS should be included as an
intervention for patients undergoing stroke reha-
bilitation, if nonsteroidal anti-inflammatory anal-
gesia has no significant effect (Appendix 4).
However, the Ottawa Panel disagrees with the
recommendation of the Royal College of Physi-
cians72 that TENS should not be used as a routine
intervention for improving muscle control for pa-
tients undergoing stroke rehabilitation in the con-
text of ongoing trials. Finally, the Ottawa Panel
agrees with the St. Joseph’s Health Care London73
finding that there is conflicting evidence* that
TENS improves a variety of outcomes, including
motor recovery, spasticity, and ADL (Appendix 4).
Other preexisting guidelines concerning TENS
did not apply to those of the Ottawa Panel.
*According to the St. Joseph’s Health Care London, conflicting evi-
dence refers to a disagreement between the findings of at least two
RCTs. Where there were more than four RCTs and the results of only
one was conflicting, the conclusion was based on the results of the
majority of the studies, unless the study with conflicting results was of
higher quality.
64 TOPICS IN STROKE REHABILITATION/SPRING 2006
Clinical practice guidelines for therapeutic ultrasound
Ultrasound versus placebo applied to the
shoulder, level I (RCT): Grade C+ for ROM (flex-
ion and external rotation) at end of treatment, 4
weeks, where a clinically important benefit was
demonstrated without statistical significance;
grade C for ROM (abduction in internal rotation
and abduction in external rotation) at end of treat-
ment, 4 weeks (no benefit was demonstrated).
Patients with chronic stroke.
Summary of trials
One RCT that compared therapeutic ultra-
sound to both a placebo and a control was in-
cluded (n = 33).399 The treatment group received
therapeutic ultrasound to the paravertebral and
shoulder joint areas for 5 minutes prior to exer-
cise (n = 10). The control group (n = 10) received
only the concurrent treatment (US-Appendix
3K), while the placebo group (n = 10) received
ultrasound with the energy turned off prior to
exercises. One treatment was given per session,
and sessions took place 7 days a week for 4 weeks
(US-Appendix 3K).
Three trials were excluded for the following rea-
sons. One provided insufficient statistical data,400
whereas another combined the results for both
groups.401 A third trial was excluded because it
included patients with periarthritis post trauma,
fractures, and sclerosis402 (US-Table 157).
Efficacy
For ultrasound applied to the shoulder of
chronic stroke patients versus placebo (one RCT,
n = 20),399 a clinically important benefit without
statistical significance was demonstrated for ROM
(flexion and external rotation) at the end of the 4-
week treatment (20% and 57% RD, respectively;
US-Figure 143; Table 158). No benefit was dem-
onstrated for ROM (abduction in internal and ex-
ternal rotation) at the end of treatment. No other
outcomes were measured.
No benefits were shown for therapeutic ul-
trasound applied to the shoulder versus con-
trol in chronic stroke patients (one RCT, n =
23). 399 At the end of 4 weeks of treatment, the
Ultrasound versus control applied to the
shoulder, level I (RCT): Grade D for ROM (flex-
ion, abduction in internal rotation, abduction in
external rotation) at end of treatment, 4 weeks (no
benefit was demonstrated, but results favored con-
trol); grade D+ for ROM (external rotation) at end
of treatment, 4 weeks (clinically important benefit
favoring control demonstrated without statistical
significance). Patients with chronic stroke.
results favored control with respect to ROM
(flexion and abduction in internal and external
rotation), although these results were without
statistical significance or clinical importance.
Notably, there was a clinically important trend
without statistical significance favoring ROM
(external rotation) at the end of treatment (-
93% RD; US-Figure 144, Table 159). No
other outcomes were measured.
Strength of the published evidence compared with
other guidelines
The Ottawa Panel found level I evidence (one
RCT) for the effects of ultrasound on post-stroke
patients. St. Joseph’s Health Care London72 also
based their findings on level I evidence (a single
RCT). The strength of evidence for ultrasound
has not been evaluated by other guidelines (Ap-
pendix 4).
Clinical recommendations compared with other
guidelines
Based on the evidence of this trial, (grade C+ for
ROM [flexion and external rotation], grade C for
ROM [abduction in internal rotation and abduc-
tion in external rotation], grade D for ROM [flex-
ion, abduction in internal rotation, abduction in
external rotation], and grade D+ for ROM [external
rotation]), the Ottawa Panel is in partial agreement
with the St. Joseph’s Health Care London73 finding
that adding ultrasound therapy to ROM exercises
does not change outcomes. Other preexisting rec-
ommendations regarding ultrasound did not apply
to those of the Ottawa Panel (Appendix 4).

Clinical practice guidelines for acupuncture
Needle acupuncture without stimulation ver-
sus placebo, level I (RCT): Grade C for ankle
spasticity at end of treatment 4 weeks (no benefit
demonstrated). Patients with chronic stroke.
Needle acupuncture with electrical stimula-
tion on paretic side versus placebo, level I
(RCT): Grade C+ for ROM at end of treatment, 1
month (clinically important benefit demonstrated
without statistical significance). Patients with post-
acute stroke.
Needle acupuncture with manual and electri-
cal stimulation versus placebo acupuncture,
level 1 (RCT): Grade C for functional status at end
of treatment, 3 months and 10 months (no benefit
demonstrated). Patients with acute stroke.
Needle acupuncture with manual and electri-
cal stimulation versus no acupuncture, level I
(RCT): Grade A for motor function at end of
treatment, 1 month, for walking mobility and for
balance at end of treatment, 1 month and 3
months, for motor function (NHP–physical move-
ment) at follow-up, 12 months, for functional sta-
tus (Sunnaas Scale) at follow-up, 12 months, and
for quality of life (number of patients without loss
of energy) at follow-up, 12 months (clinically im-
portant benefit demonstrated); grade C+ for mo-
tor function (NHP–physical movement, and MAS)
at end of treatment, 6 weeks, for motor function
(NHP–physical movement) at follow-up, 10
months, pain relief at end of treatment, 6 weeks,
and follow-up, 10 months, and for motor function
Summary of trials
Seven RCTs (n = 417) were included that investi-
gated the effects of acupuncture compared to ei-
ther no acupuncture or a placebo control following
stroke.403–409 The outcomes assessed in the various
studies included functional status, motor function,
walking mobility, balance, level of energy, ROM,
relief of spasticity, and relief of pain (AC-Appen-
dix 3L).
There was substantial variation in acupuncture
technique. Three RCTs404–406 reported manual
Clinical Practice Guidelines 65
at end of treatment, 3 months (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for quality of life (number of
patients without loss of energy) at end of treat-
ment, 6 weeks, and for functional status (Sunnaas
Scale) at end of treatment, 6 weeks (no benefit
demonstrated); grade D for functional status
(change in neurological scale, change in Barthel
Index, change in Sunnaas Scale) at end of treat-
ment, 3 months, and follow-up, 10 months (no
benefit demonstrated but favoring control). Pa-
tients with acute and subacute stroke.
Surface acupuncture with electrical stimula-
tion versus no acupuncture, level I (RCT): Grade
A for walking mobility at end of treatment, 2
weeks (clinically important benefit demonstrated);
grade C for motor function of upper and lower
extremity and functional status at end of treat-
ment, 2 weeks (no benefit demonstrated). Patients
with subacute stroke.
Needle acupuncture with electrical stimula-
tion on paretic side versus needle acupuncture
without stimulation, level I (RCT): Grade A for
decrease of spasticity at end of treatment, 10 and
15 days (clinically important benefit demon-
strated); grade C+ for decrease of spasticity at end
of treatment, 5 days (clinically important benefit
demonstrated without statistical significance);
grade C for decrease of spasticity at end of treat-
ment, 1 day (no benefit demonstrated). Patients
with post-acute stroke.
stimulation of inserted needles. In addition, elec-
trical stimulation was used in seven RCTs404–409
and was typically applied to selected points on the
paretic side. There was little consistency in the
choice of stimulation frequency from one study to
the next (range, 1–50 Hz). While stimulation set-
tings were adjusted to induce visible muscle con-
tractions in some studies, lower thresholds were
used in other studies (AC-Appendix 3L).
Except for one study,403 experimental and con-
trol groups received standard rehabilitative care
concurrently. Three trials used sham or placebo
66 TOPICS IN STROKE REHABILITATION/SPRING 2006
acupuncture as the control intervention.403,404,408
The duration of most individual acupuncture ses-
sions was 30 minutes, and the frequency of ses-
sions ranged from 2 to 5 per week and the total
treatment interval ranged from 4 to 10 weeks.
Overall, the number of sessions allocated to each
patient varied between 8 and 24 sessions (AC-
Appendix 3L).
Twelve articles were excluded, the reasons for
which are listed in AC-Table 160. 410–421 Two stud-
ies were reviews, two were case series, one did not
involve acupuncture, one lacked a control group,
one contained fewer than five subjects in the con-
trol group, one was a duplicate of an already in-
cluded study, one was written in a foreign lan-
guage, and in two studies more than 20% of the
study sample was represented by patients with
recurrent stroke.
Efficacy
In one RCT (n = 25),403 needle acupuncture
without manual or electrical stimulation was
compared to placebo acupuncture in chronic
stroke patients with spastic equinovarus deformity
of the leg (AC-Figure 145, Table 161). At the end
of 4 weeks of treatment, no benefits were observed
for ankle spasticity (modified Ashworth Scale).
Other outcomes were not assessed.
Needle acupuncture with low frequency (1-2
Hz) electrical stimulation produced clinically
important, but statistically nonsignificant, benefits
for ROM (number of patients that improved more
than 10% on ROM tests) compared to placebo
acupuncture in patients with post-acute, left hemi-
spheric stroke (one RCT, n = 16).408 The RD at the
end of 1 month of treatment was 40% (AC-Figure
146, Table 162). No other outcomes were mea-
sured.
Patients in the acute phase of stroke who were
unable to walk without support or perform ADLs
without assistance were randomized to receive
needle acupuncture with manual and electri-
cal stimulation or placebo acupuncture (one
RCT, n = 104).404 In the experimental group,
needles inserted on the nonparetic side were
stimulated manually and needles introduced on
the paretic side were stimulated electrically at low
frequency (2 Hz). The control group received
placebo acupuncture, in which four shortened
needles were placed superficially just under the
skin – one at each limb – and left without further
stimulation (AC-Appendix 3L). No benefits
were demonstrated for functional status (change
in Neurological Scale, change in Barthel Index,
and change in Sunnaas Scale) at the end of 3
months of treatment or at 10 months follow-up
(AC-Figure 147, Table 163).
Needle acupuncture with electrical and
manual stimulation was compared to no acu-
puncture in acute and subacute stroke patients
(three RCTs, n = 223404–406; AC-Figure 148,
Table 164). In one of the RCTs, statistically
significant and clinically important benefits
were shown for motor function (NHP–physical
movement), functional status (Sunnaas Scale),
and quality of life (number of patients with no
loss of energy) at 12 months follow-up (one
RCT, n = 41406; −4%, 35%, and 35% RD, respec-
tively). In another study, similarly positive ef-
fects were shown for motor function, balance,
and walking mobility at the end of 1 month of
treatment and for balance, walking mobility,
and functional status (Barthel Index) at the end
of 3 months of treatment (one RCT, n = 78; 36%,
34%, 112%, 29%, 112%, and 40% RD, respec-
tively).405 Important clinical benefits that did
not achieve statistical significance were found
for motor function (NHP–physical movement,
and MAS) at the end of 6 weeks of treatment and
at 12 months follow-up, as well as for pain relief
(number of patients with no pain) at the end of 6
weeks of treatment and at 12 months follow-up
(−35%, −44%, 20%, and 30% RD, respec-
tively). 406 Additionally, there were clinically
meaningful effects on motor function at the end
of 3 months of treatment (25% RD),405 but no
improvement of quality of life (number of pa-
tients with no loss of energy) or functional status
(Sunnaas Scale) at the end of 6 weeks of treat-
ment.406 Clinically and statistically nonsignifi-
cant changes in functional status (change in
neurological scale, change in Barthel Index,
change in Sunnaas Scale) favoring control were
seen at the end of 3 months of treatment and at
10 months follow-up.404

A unique technique was used in one RCT (n =
118),409 where acupuncture points located on
the affected extremities of subacute stroke pa-
tients were stimulated electrically via surface elec-
trodes instead of through percutaneous needles.
The intensity of stimulation was sufficient to
achieve muscle contraction and the frequency was
adjusted to 20–25 Hz. Compared to no acupunc-
ture, statistically significant and clinically impor-
tant benefits were shown for walking mobility at
the end of 2 weeks of treatment (AC-Figure 149,
Table 165). The RD was 47%. No benefits were
found for functional status (FIM) or for motor
function (Brunnstrom) of the upper and lower
extremity.
The additional benefit of electroacupuncture
in post-acute stroke patients with elbow spastic-
ity was investigated in one RCT (n = 35),407
where needle acupuncture with electrical stimu-
lation of selected points on the paretic side was
compared to needle acupuncture without elec-
trical stimulation. Statistically significant and
clinically meaningful decreases in spasticity
(modified Ashworth Scale) were observed at the
end of 10 and 15 days of treatment (−27% and −
27% RD, respectively; AC-Figure 150, Table
166). Clinically important decreases in spastic-
ity also occurred after 5 days of treatment, but
without statistical significance (−15% RD). No
benefits were seen in terms of degree of spastic-
ity at the end of 1 day of treatment. No other
outcomes were assessed.
Strength of the published evidence compared with
other guidelines
The Ottawa Panel found level I (RCTs) evidence
evaluating acupuncture in stroke patients. The
clinical practice guidelines developed by the Royal
College of Physicians72 and St. Joseph’s Health
Care London73 were based on level I (RCT) evi-
dence as well (Appendix 4).
Clinical Practice Guidelines 67
Clinical recommendations compared with other
guidelines
The Ottawa Panel recommends that there is good
scientific evidence to consider including acupunc-
ture as an adjunct to standard stroke rehabilitation
to improve walking mobility (grade A for walking
mobility), ROM (grade A for number of patients
improved on ROM tests), and balance (grade A for
balance). The evidence for the effects of acupunc-
ture on spasticity, motor function, and functional
status are conflicting. One study reported no benefit
of needle acupuncture without stimulation for ankle
spasticity in chronic stroke patients with spastic
equinovarus deformity (grade C). However, another
study revealed statistically significant and clinically
important benefits of electroacupuncture on elbow
spasticity compared to standard acupuncture in
post-acute stroke. With regard to motor function,
surface acupuncture with electrical stimulation was
not effective, but needle acupuncture with manual
and electrical stimulation showed positive results.
The evidence from three RCTs that assessed func-
tional status was also inconsistent, with two RCTs
suggesting no effect and the other RCT suggesting
statistically significant and clinically important ben-
efits (Appendix 4).
The Ottawa Panel recommendations are gener-
ally consistent with the recommendations of St.
Joseph’s Health Care London,73 which found that
“there is conflicting evidence that acupuncture is
effective for improving stroke outcomes.” How-
ever, the Ottawa Panel wishes to emphasize that
the available evidence is not conflicting with re-
spect to all stroke outcomes. The Royal College of
Physicians72 concluded that the evidence was in-
sufficient to make definitive statements about the
role of acupuncture. They expressed a cautious
statement based on consensus opinion that “acu-
puncture should only be used as an intervention
for patients undergoing post-stroke rehabilitation
in the context of ongoing trials” (Appendix 4).
68 TOPICS IN STROKE REHABILITATION/SPRING 2006
Clinical practice guidelines for intensity and organization of rehabilitation
Stroke unit versus general ward, level I (RCT)
and level II (CCT): Grade A for length of stay at
end of treatment, 2 weeks, and follow-up, 3 years
(clinically important benefit demonstrated); grade
B for mobility (number of patients able to walk
long distances independently) at end of treatment,
6 weeks (clinically important benefit demon-
strated); grade C+ for functional status (number of
patients with mild-moderate impairment) at fol-
low-up, 1 year (clinically important benefit dem-
onstrated without statistical significance); grade C
for functional status (number of patients deterio-
rated in ADL score) at end of treatment, 3 months,
and follow-up, 6 months, functional status (num-
ber of patients independent in ADL) at end of
treatment, 6 weeks, length of stay (in days) at end
of treatment, 6 weeks and 3 months, mobility
(number of patients independent in walking mo-
bility) at end of treatment, 16 weeks, mobility
(number of patients independent in walking
[Barthel Index], number of patients independent
in indoor and outdoor walking [Rivermead ADL
Scale–self-care], number of patients independent
in walking outside and climbing stairs [Rivermead
ADL Scale–household]) at end of treatment, 3 and
6 months, motor function (number of patients
independent in upper extremity and lower extrem-
ity function) at end of treatment, 16 weeks, effi-
ciency at end of treatment, 2 weeks, postural status
(number of patients sitting and standing) at fol-
low-up, 6 weeks, functional status (ADL score) at
end of treatment, 3 months, and follow-up, 9
months, and resource cost (total cost per patient in
thousands of dollars) at follow-up, 1 year (no ben-
efit demonstrated); grade D for mobility (number
of patients able to walk short distances with or
without aids) at end of treatment, 6 weeks, func-
tional status (number of functional patients) at
follow-up, 1 year, and resource cost (number of
physician visits per patient) at follow-up, 1 year
(no benefit demonstrated but favoring control);
grade D+ for functional status (ADL score) at end
of treatment, 1 week (clinically important benefit
favoring control demonstrated without statistical
significance). Patients with acute, subacute, post-
acute, and chronic stroke.
Stroke unit versus home care, level I (RCT):
Grade A for functional status (number of patients
independent with Barthel) at follow-up, 9 months
(clinically important benefit demonstrated); grade
C for functional status (number of patients indepen-
dent with Modified Rankin score) at end of treat-
ment, 3 months, and follow-up, 9 months, and
functional status (number of patients independent
with Barthel) at end of treatment, 3 months (no
benefit demonstrated). Patients with acute stroke.
Stroke unit (large artery stroke) versus gen-
eral ward (large artery stroke), level I (RCT):
Grade A for functional status at follow-up, 1 year,
and quality of life at follow-up, 3 months (clini-
cally important benefit demonstrated); grade C for
functional status at follow-up, 3 months, and qual-
ity of life at follow-up, 3 months (no benefit dem-
onstrated); grade D for length of stay at follow-up,
3 months (no benefit demonstrated but favoring
control); grade D- for efficiency at follow-up, 3
months (clinically important benefit favoring con-
trol demonstrated with statistical significance). Pa-
tients with acute stroke.
Stroke unit (small artery stroke) versus gen-
eral ward (small artery stroke), level I (RCT):
Grade A for quality of life at follow-up, 3 months
(clinically important benefit demonstrated); grade
C for quality of life at follow-up, 1 year (no benefit
demonstrated); grade D for functional status at fol-
low-up, 3 months, and 1 year (no benefit demon-
strated but favoring control); grade D- for length of
stay and efficiency at follow-up, 3 months (clinically
important benefit favoring control demonstrated
with statistical significance). Patients with acute
stroke.
Intensive outpatient physiotherapy rehabili-
tation program versus control group, level I
(RCT): Grade C+ for quality of life at end of
treatment, 3 months (clinically important benefit
demonstrated without statistical significance);
grade C for functional status at end of treatment, 3
months (no benefit demonstrated); grade D for
functional status at follow-up, 6 months (no ben-
efit demonstrated but favoring control). Patients
with chronic stroke.
Six days/week versus seven days/week treat-
ment, level I (RCT): Grade C+ for mobility at end
of treatment, 3 weeks (clinically important benefit

demonstrated without statistical significance);
grade C for functional status and length of stay at
end of treatment, 3 weeks (no benefit demon-
strated). Patients with post-acute stroke.
Enhanced occupational therapy versus stan-
dard customary occupational therapy, level I
(RCT): Grade A for functional status (number of
patients improved in ADL) at end of treatment, 8
weeks and 6 months, life habit/leisure (overall lei-
sure score) at end of treatment, 3 and 6 months,
life habit/leisure (total leisure score) at end of treat-
ment, 6 months, mobility (Nottingham EADL
score for mobility) at end of treatment, 3 and 6
months, functional status (Nottingham EADL
score) at end of treatment, 8 weeks, 3 months, and
6 months (clinically important benefit demon-
strated); grade C+ for quality of life (number of
patients living independently) and functional sta-
tus (FIM for upper extremity and lower extremity
dressing) at end of treatment, 3 weeks, and func-
tional status (EADL total score) at follow-up, 3
months (clinically important benefit demonstrated
without statistical significance); grade C for life
habit/leisure (total leisure score) and mobility
(NHP score for mobility) at end of treatment, 6
months, functional status (Barthel Index score) at
follow-up, 3 months (no benefit demonstrated);
grade D for mobility (NHP score for mobility) at
end of treatment, 3 months (no benefit demon-
strated but favoring control); grade D+ for pain
relief at end of treatment, 3 months and 6 months,
and quality of life (General Health Questionnaire)
at follow-up, 3 months (clinically important ben-
efit favoring control demonstrated without statisti-
cal significance). Patients with subacute stroke.
Enhanced occupational therapy versus no
therapy, level I (RCT): Grade A for mobility (NHP
score for mobility and Nottingham EADL for mo-
bility) at end of treatment, 3 months and 6
months, life habit/leisure (overall leisure score and
total leisure activity) at end of treatment, 3 and 6
months, functional status (number of patients im-
proved in ADL) at follow-up, 6 months (clinically
important benefit demonstrated); grade C+ for
activity involvement (Katz Adjustment Index:
number of patients satisfied with their walking) at
follow-up, 13 weeks, activity involvement (num-
ber of patients satisfied with their work in the
Clinical Practice Guidelines 69
yard) at end of treatment, 5 weeks, and functional
status (EADL) at follow-up, 1 year (clinically im-
portant benefit demonstrated without statistical
significance); grade C for functional status
(Barthel ADL Index) at follow-up, 6 months and 1
year, functional status (EADL) at follow-up, 6
months, activity involvement (number of patients
satisfied with their work in and around the house
and number of patients satisfied with their walk-
ing) at end of treatment, 5 weeks, and activity
involvement (number of patients satisfied with
their work in the yard) at follow-up, 13 weeks (no
benefit demonstrated); grade D for activity in-
volvement (number of patient satisfied with their
work in and around house) at follow-up, 13 weeks
(no benefit demonstrated but favoring control);
grade D+ for pain relief at end of treatment, 3 and
6 months (clinically important benefit favoring
control demonstrated without statistical signifi-
cance). Patients with subacute stroke.
Standard customary occupational therapy
versus no therapy, level I (RCT): Grade A for
functional status (EADL score) at follow-up, 1
month (clinically important benefit demon-
strated); grade C+ for mobility (NHP for mobility)
at end of treatment, 3 and 6 months, for pain relief
at end of treatment, 3 months (clinically important
benefit demonstrated without statistical signifi-
cance); grade C for life habit/leisure (overall lei-
sure score and total leisure activity score) at end of
treatment, 3 and 6 months, for functional status
(Barthel Index score) at follow-up, 1 month, pain
relief at end of treatment, 6 months (no benefit
demonstrated); grade D for quality of life at fol-
low-up, 1 month (no benefit demonstrated but
favoring control); grade D+ for functional status
(EADL score for mobility) at end of treatment, 3
months and 6 months (clinically important benefit
favoring control demonstrated without statistical
significance). Patients with subacute stroke.
Enhanced physiotherapy (60-minute treat-
ment of physiotherapy) versus standard cus-
tomary physiotherapy care (30 minutes), level I
(RCT): Grade C for quality of life at end of treat-
ment, 6 weeks (no benefit demonstrated); grade D
for quality of life at end of treatment, 6 months (no
benefit demonstrated but favoring control). Pa-
tients with chronic stroke.
70 TOPICS IN STROKE REHABILITATION/SPRING 2006
Enhanced physiotherapy versus standard cus-
tomary physiotherapy treatment, level I (RCT):
Grade A for motor function (ARAT) at follow-up,
21 weeks (clinically important benefit demon-
strated); grade C+ for functional status (Barthel
Index) at follow-up, 3 and 16 weeks (clinically
important benefit demonstrated without statistical
significance); grade C for functional status (Barthel
Index) at end of treatment, 5 weeks, and follow-up,
21 weeks, functional status (EADL score) at follow-
up, 16 weeks and 21 weeks, motor function
(Rivermead Motor Assessment–upper extremity) at
end of treatment, 5 weeks, and follow-up, 21
weeks, motor function (Ten-Hole Peg Test) at end
of treatment, 5 weeks, and follow-up, 16 weeks,
and motor function (ARAT) at end of treatment, 5
weeks (no benefit demonstrated); grade D for func-
tional status (EADL score) at end of treatment, 5
weeks (no benefit demonstrated but favoring con-
trol); grade D+ for motor function (Rivermead Mo-
tor Assessment–gross function) at follow-up, 3
weeks, and grip strength (maximum grip) at follow-
up, 16 weeks (clinically important benefit favoring
control demonstrated without statistical signifi-
cance); grade D- for motor function (Rivermead
Motor Assessment-upper extremity) at follow-up, 3
weeks and 16 weeks, motor function (ARAT) at
follow-up, 16 weeks, motor function (Rivermead
Motor Assessment–gross function) at end of treat-
ment, 5 weeks, and follow-up, 16 weeks, and grip
strength (maximum grip) at end of treatment, 5
weeks, and follow-up, 3 weeks (clinically important
benefit favoring control demonstrated with statisti-
cal significance). Patients with subacute stroke.
Enhanced assistant physiotherapist versus
standard customary physiotherapy treatment,
level I (RCT): Grade A for motor function
(Rivermead Motor Assessment-upper extremity) at
follow-up, 21 weeks, motor function (ARAT) at
end of treatment, 5 weeks, and follow-up, 21
weeks (clinically important benefit demonstrated);
grade C+ for functional status (Barthel Index) at
follow-up, 3 and 16 weeks, functional status
(EADL) at end of treatment, 5 weeks, and follow-
up, 21 weeks, motor function (ARAT) at follow-
up, 16 weeks (clinically important benefit demon-
strated without statistical significance); grade C
for functional status (Barthel Index) at end of treat-
ment, 5 weeks, functional status (EADL score) at
follow-up, 16 weeks, motor function (Rivermead
Motor Assessment-upper extremity) at end of
treatment, 5 weeks, and follow-up, 16 weeks, mo-
tor function (Rivermead Motor Assessment–gross
function) at follow-up, 3 and 16 weeks, motor
function (Ten-Hole Peg Test) at end of treatment,
5 weeks, and follow-up, 16 weeks (no benefit
demonstrated); grade D for functional status
(Barthel Index) at follow-up, 21 weeks (no benefit
demonstrated but favoring control); grade D+ for
motor function (Rivermead Motor Assessment-up-
per extremity) at follow-up, 3 weeks, and grip
strength (maximum grip) at end of treatment, 5
weeks, and follow-up, 16 weeks (clinically impor-
tant benefits favoring control demonstrated with-
out statistical significance); grade D- for motor
function (Rivermead Motor Assessment–gross
function) at end of treatment, 5 weeks, and grip
strength (maximum grip) at follow-up, 3 weeks
(clinically important benefit favoring control dem-
onstrated with statistical significance). Patients
with subacute stroke.
Enhanced physiotherapy versus enhanced as-
sistant physiotherapist treatment, level I (RCT):
Grade A for motor function (Rivermead Motor
Assessment–gross function) at end of treatment, 5
weeks, grip strength (maximum grip) at follow-up,
3 weeks, and motor function (ARAT) at follow-up,
21 weeks (clinically important benefit demon-
strated); grade A favoring enhanced assistant phys-
iotherapist for functional status (Barthel Index) at
end of treatment, 5 weeks, motor function
(Rivermead Motor Assessment–upper extremity)
at follow-up, 16 weeks and 21 weeks, functional
status (EADL score) at follow-up, 21 weeks, motor
function (ARAT) at end of treatment, 5 weeks, and
follow-up, 16 weeks (clinically important benefit
demonstrated); grade C+ for motor function
(Rivermead Motor Assessment–gross function) at
follow-up, 16 weeks (clinically important benefit
demonstrated without statistical significance);
grade C+ favoring enhanced assistant physio-
therapist for functional status (Barthel Index) at
follow-up, 3 and 16 weeks, functional status
(EADL) at end of treatment, 5 weeks, motor func-
tion (Rivermead Motor Assessment–upper extrem-
ity) at follow-up, 3 weeks, grip strength at end of

treatment, 5 weeks, and follow-up, 16 weeks
(clinically important benefit demonstrated without
statistical significance); grade C for functional sta-
tus (Barthel Index) at follow-up, 21 weeks, func-
tional status (EADL score) at follow-up, 16 weeks,
motor function (Rivermead Motor assessment–up-
per extremity) at end of treatment, 5 weeks, and
motor function (Ten-Hole Peg Test) at end of treat-
ment, 5 weeks, and follow-up, 16 weeks (no ben-
efit demonstrated). Patients with subacute stroke.
Enhanced upper-extremity treatment versus
interdisciplinary treatment, level I (RCT): Grade
A for motor function (Frenchay Arm Test) and
functional status (Barthel Index) at follow-up, 18
weeks (clinically important benefit demonstrated);
grade C for motor function (Upper Limb Motricity
Index and Frenchay Arm Test) at follow-up, 6
weeks, functional status (Nottingham EADL) at
follow-up, 18 weeks, and functional status
(Barthel Index) at follow-up, 6 weeks (no benefit
demonstrated); grade D for motor function (Up-
per Limb Motricity Index) at follow-up, 18 weeks,
functional status (Nottingham EADL) at follow-
up, 6 weeks (no benefit demonstrated but favoring
control); grade D+ for motor function (ARAT) at
follow-up, 6 and 18 weeks (clinically important
benefit favoring control demonstrated without sta-
tistical significance). Patients with subacute stroke.
Enhanced therapy (severe stroke) versus
standard customary care (severe stroke), level I
(RCT): Grade D for mobility at end of treatment, 6
months (no benefit demonstrated). Patients with
acute stroke.
Enhanced therapy (mild stroke) versus stan-
dard customary care (mild stroke), level I (RCT):
Grade D for mobility at end of treatment, 6
months (no benefit demonstrated). Patients with
acute stroke.
Home therapy (physiotherapy) versus stan-
dard customary care without home visits, level I
(RCT): Grade C+ for functional status (Instrumen-
tal Activities of Daily Living [IADL]–domestic ac-
tivities) at follow-up, 6 months (clinically impor-
tant benefit demonstrated without statistical
significance); grade C for discharge disposition,
functional status, mobility, and life habits/leisure
at follow-up, 6 months (no benefit demonstrated).
Patients with chronic stroke.
Clinical Practice Guidelines 71
Home-based physiotherapy at high-intensity
versus low-intensity physiotherapy control,
level I (RCT): Grade A for motor function
(STroke REhabilitation Assessment of Movement
[STREAM] lower extremity) at follow-up, 11
weeks (clinically important benefit demon-
strated); grade C+ functional status and motor
function (STREAM–upper extremity) at follow-
up, 11 weeks (clinically important benefit dem-
onstrated without statistical significance); grade
C for mobility at follow-up, 11 weeks, functional
status and motor function (STREAM–upper ex-
tremity) at follow-up, 22 weeks (no benefit dem-
onstrated); grade D for mobility at follow-up, 22
weeks (no benefit demonstrated but favoring
control); grade D+ for motor function
(STREAM–lower extremity) at follow-up, 22
weeks (clinically important benefit favoring con-
trol demonstrated without statistical signifi-
cance). Patients with chronic stroke.
Home-based rehabilitation versus hospital-
based rehabilitation, levels I and II (RCT, CCT):
Grade C for resource cost at follow-up, 1 year, and
mobility and functional status at end of treatment,
3 months (no benefit demonstrated). Patients with
post-acute and chronic stroke.
Rehabilitation versus no rehabilitation
group, level II (CCT): Grade D for motor function
at end of treatment, 6 months (no benefit demon-
strated but favoring control); grade D+ for func-
tional status at end of treatment, 6 months (clini-
cally important benefit favoring control
demonstrated without statistical significance). Pa-
tients with subacute stroke.
Home therapy (physician) versus standard
customary care without home visits, level I
(RCT): Grade A for decrease in hospital readmis-
sions and functional status (FAI) at follow-up, 6
months (clinically important benefit demon-
strated); grade C for functional status (Functional
Quality of Movement Scale [FQM]) motor perfor-
mance, Barthel Index, EADL–personal daily care,
IADL–domestic activities), mobility and life habits/
leisure at follow-up, 6 months (no benefit demon-
strated); grade D for functional status (FQM qual-
ity of movement) at follow-up, 6 months (no ben-
efit demonstrated but favoring control). Patients
with chronic stroke.
72 TOPICS IN STROKE REHABILITATION/SPRING 2006
Home therapy (physician) versus home
therapy (physiotherapy), level I (RCT): Grade
C+ favoring home therapy (physiotherapy) for
functional status (IADL–domestic activities) at fol-
low-up, 6 months (clinically important benefit
demonstrated without statistical significance);
grade C for decrease in hospital readmission,
functional status (FAI, FQM motor performance,
FQM quality of movement, Barthel Index, and
EADL–personal daily care), mobility and life hab-
its/leisure at follow-up, 6 months (no benefit dem-
onstrated). Patients with chronic stroke.
Extended stroke unit service with early sup-
ported discharge versus ordinary stroke unit
service, level I (RCT): Grade A for functional
status (number of patients with Rankin score less
than 2) at follow-up, 6 months (clinically impor-
tant benefits demonstrated); grade C+ for physical
mobility at follow-up, 3 weeks and 1 year (clini-
cally important benefits demonstrated without sta-
tistical significance); grade C for functional status
(number of patients with Barthel Index score
higher than 95) at follow-up, 3 weeks, 6 months,
and 1 year, for discharge status and preventing
mortality at follow-up, 1 year, for energy level and
pain at follow-up, 3 weeks, 6 months, and 1 year,
for global health status at follow-up, 1 year, and
functional status (number of patients with Rankin
score less than 2) at follow-up, 3 weeks and 1 year
(no benefit demonstrated); grade D for global
health status at follow-up, 3 weeks and 6 months
(no benefit demonstrated but favoring control);
grade D+ for physical mobility at follow-up, 6
months (clinically important benefits demon-
strated without statistical significance favoring
control). Patients with acute stroke.
Early physiotherapy intervention versus con-
trol group, level I (RCT): Grade C+ for functional
status (Modified Barthel Index) and motor func-
tion (Fugl-Meyer score for lower extremity) at end
of treatment, 1 month, and follow-up, 5 months,
and motor function (Fugl-Meyer score for upper
extremity) at end of treatment, 1 month (clinically
important benefits demonstrated without statisti-
cal significance); grade D for motor function
(Fugl-Meyer score for upper extremity) at follow-
up, 5 months (no benefit demonstrated but favor-
ing control). Patients with acute stroke.
Home-based exercise training versus control,
level I (RCT): Grade A for change in gait speed,
gait endurance, torque (change in knee isometric
extensors), endurance, and cardiopulmonary
function at end of treatment, 12 weeks; grade C+
for motor function (change in Fugl-Meyer lower
extremity), change in gait speed, gait endurance,
functional status (Physical Function Index), and
strength (change in grip strength) at end of treat-
ment, 12 weeks (clinically important benefit dem-
onstrated without statistical significance); grade C
for motor function (change in Fugl-Meyer upper
extremity and lower extremity), balance (Berg bal-
ance and change in Berg balance), functional status
(IADL and Barthel ADL Index) at end of treatment,
12 weeks (no benefit demonstrated); grade D+ for
balance (Functional reach) at end of treatment, 12
weeks (clinically important benefit favoring con-
trol demonstrated without statistical significance);
grade D for torque (change in ankle isometric
dorsiflexors; no benefit demonstrated but favoring
control). Patients with post-acute stroke.
Outpatient versus home exercise group, level I
(RCT): Grade C+ favoring home exercise group for
gait speed at end of treatment, 6 months, and fol-
low-up, 3 months (clinically important benefit
demonstrated without statistical significance);
grade C for single support time at end of treatment,
6 months, and follow-up, 3 months (no benefit
demonstrated). Patients with chronic stroke.
Outpatient versus control, level I (RCT):
Grade D for gait speed at follow-up, 3 months (no
benefit demonstrated but favoring control); grade
D+ for gait speed at end of treatment, 6 months,
and single support time at end of treatment, 6
months, and follow-up, 3 months (clinically im-
portant benefit demonstrated without statistical
significance). Patients with chronic stroke.
Outpatient therapy versus home therapy,
level I (RCT): Grade C for grip strength on af-
fected side and motor assessment, functional sta-
tus, and gait speed at end of treatment, 6 weeks
and 3 months (no benefit demonstrated). Patients
with subacute stroke.
Home physiotherapy versus day-hospital
group, level I (RCT): Grade A for resource cost at
end of treatment, 8 weeks (clinically important ben-
efit demonstrated). Patients with subacute stroke.

Home-exercise versus control group, level I
(RCT): Grade C for gait speed at end of treatment,
6 months, and follow-up, 3 months (no benefit
demonstrated); grade D for single support time at
end of treatment, 6 months, and follow-up, 3
months (no benefit demonstrated but favoring
control). Patients with chronic stroke.
Home therapy versus control group, level II
(CCT): Grade C for motor function at end of
treatment, 9 weeks (no benefit demonstrated). Pa-
tients with subacute stroke.
Combined outpatients and home exercise
versus control group, level I (RCT): Grade C for
gait speed at end of treatment, 6 months, and
follow-up, 3 months (no benefit demonstrated);
grade D for single support time at end of treat-
ment, 6 months, and follow-up, 3 months (no
benefit demonstrated but favoring control). Pa-
tients with chronic stroke.
Early care versus standard customary care in
stroke unit, level I (RCT): Grade A for length of
stay at end of treatment, 6 weeks (clinically impor-
tant benefit demonstrated); grade C for mobility
and functional status at end of treatment, 6 weeks
(no benefit demonstrated). Patients with acute
stroke.
Early supported discharge with home reha-
bilitation versus standard customary rehabilita-
tion, level I (RCT): Grade A for functional status
(Older Americans Resource Scale for Instrumental
ADL) at end of treatment, 1 and 3 months, and
length of stay at follow-up, 3 months (clinically
important benefit demonstrated); grade C for mo-
tor function and mobility at end of treatment, 1
and 3 months, quality of life at follow-up, 6
months, functional status (Barthel Index) and pain
relief at end of treatment, 3 months (no benefit
demonstrated); grade D for functional status
(Nottingham EADL score) at end of treatment, 3
months, and follow-up, 6 months, quality of life at
end of treatment, 3 months, pain relief and func-
tional status (Barthel Index) at end of treatment, 1
month (no benefit demonstrated but favoring con-
trol). Patients with acute stroke.
Early supported discharge versus standard
customary rehabilitation, level I (RCT): Grade A
for length of stay and quality of life (Dartmouth
Coop Global Health Status–total) at end of treat-
Clinical Practice Guidelines 73
ment, 3 months, and functional status
(Nottingham EADL) at follow-up, 3 months (clini-
cally important benefit demonstrated); grade C for
motor function and functional status (Rivermead
ADL score) at follow-up, 9 months, gait speed and
functional status (Barthel ADL Index) at end of
treatment, 3 months, and follow-up, 9 months,
and quality of life (Dartmouth Coop Global Health
Status–physical fitness, daily activities, and pain
relief) at end of treatment, 3 months (no benefit
demonstrated); grade D for motor function and
functional status (Nottingham ADL) at end of
treatment, 3 months (no benefit demonstrated but
favoring control); grade D+ for functional status
(Nottingham ADL) at follow-up, 9 months (clini-
cally important benefit favoring control demon-
strated without statistical significance). Patients
with acute and post-acute stroke.
Home intervention (rehabilitation and nurs-
ing services) versus standard customary care,
level I (RCT): Grade A for resource cost at follow-
up, 3 months (clinically important benefit demon-
strated); grade C for quality of life at follow-up, 3
months (no benefit demonstrated). Patients with
subacute stroke.
Nursing early activation program versus no
therapy, level I (RCT): Grade C for mobility and
functional status at follow-up, 1 year (no benefit
demonstrated). Patients with acute stroke.
Integrated care pathway versus standard cus-
tomary multidisciplinary team care, level I
(RCT): Grade C for functional status (number of
patients with no problems with self-care, number
of patients with no problems with usual activities)
and pain relief at end of treatment, 6 months (no
benefit demonstrated); grade D for mobility, func-
tional status (number of patients with some prob-
lems with usual activities), and pain relief at end of
treatment, 6 months (no benefits demonstrated
but favoring control); grade D+ for functional sta-
tus (number of patients with some problems with
washing and dressing) at end of treatment, 6
months (clinically important benefit favoring con-
trol demonstrated without statistical significance).
Patients with acute stroke.
Full-time integrated treatment (FIT) versus
standard customary rehabilitation, level II
(CCT): Grade C for functional status at end of
74 TOPICS IN STROKE REHABILITATION/SPRING 2006
treatment, 4 weeks and 6 weeks, and length of stay
at end of treatment, 4 weeks (no benefit demon-
strated). Patients with acute stroke.
Critical path method versus standard cus-
tomary care method, level I (RCT): Grade C for
functional status (FIM motor subscale) at end of
treatment, 1 month (no benefit demonstrated);
grade D for functional status (FIM raw score) and
length of stay at end of treatment, 1 month (no
benefit demonstrated but favoring control). Pa-
tients with acute stroke.
Summary of trials
Forty-eight RCTs and eight CCTs were found
that evaluated the efficacy of varying intensities
and organizations of rehabilitation for acute to
chronic stroke patients (n = 8,932).103,104,279,422–474
The following treatments were evaluated: 1) stroke
unit versus general ward (n = 3,146)431,437–439,448–
450,452,463–465
; 2) stroke unit versus home care (n =
447)449; 3) stroke unit (large artery stroke) versus
general ward (large artery stroke) (n = 267)434; 4)
stroke unit (small artery stroke) versus general
ward (small artery stroke) (n = 267)434; 5) intensive
outpatient physiotherapy rehabilitation program
versus control group (n = 54)279; 6) 6 days/week
versus 7 days/week treatment (n = 113)462; 7) en-
hanced occupational therapy versus standard
customary occupational therapy (n = 492)
410,432,433,441,454
; 8) enhanced occupational therapy
versus no therapy (n = 481)430,432,433,441,447; 9) stan-
dard customary occupational therapy versus no
therapy (n = 378)431,432,471; 10) enhanced physio-
therapy (60-minute treatment of physiotherapy)
versus standard customary physiotherapy care (30
minutes;n = 114)458; 11) enhanced physiotherapy
versus standard customary physiotherapy treat-
ment (n = 564)453,457; 12) enhanced assistant phys-
iotherapist versus standard customary physio-
therapy treatment (n = 564)453,457; 13) enhanced
physiotherapy versus enhanced assistant physio-
therapist treatment (n = 564)453,457; 14) enhanced
upper extremity treatment versus interdisciplinary
treatment (n = 626)459; 15) enhanced therapy (se-
vere stroke) versus standard customary care (se-
vere stroke; n = 137)468; 16) enhanced therapy
Structured nursing intervention versus con-
ventional rehabilitation, level I (RCT): Grade C
for functional status (FIM and IADL) and self-
perception of health at end of treatment, 3 months,
and follow-up, 3 months (no benefit demon-
strated). Patients with acute stroke.
Home hospitalization versus standard stroke
care (inpatient), level I (RCT): Grade D for neu-
rological status and functional status at follow-up,
6 months (no benefit demonstrated but favoring
control). Patients with acute stroke.
(mild stroke) versus standard customary care
(mild stroke; n = 137)468; 17) home therapy (phys-
iotherapy) versus standard customary care without
home visits (n = 310)423,424; 18) home-based at
high-intensity versus low-intensity physiotherapy
control (n = 19)451; 19) home-based rehabilitation
versus hospital-based rehabilitation (n =
205)425,445; 20) rehabilitation versus no rehabilita-
tion group (n = 43)442; 21) home therapy (physi-
cian) versus standard customary care without
home visits (n = 310)423,424; 22) home therapy
(physician) versus home therapy (physiotherapy)
(n = 310)423,424; 23) extended stroke unit service
with early supported discharge versus an ordinary
stroke unit service (n = 702)426,439,446; 24) early
physiotherapy intervention versus control group
(n = 128)436; 25) home-based exercise training ver-
sus control (n = 112)103,104; 26) outpatient versus
home exercise group (n = 20)472; 27) outpatient
versus control (n = 20)472; 28) outpatient therapy
versus home therapy (n = 100) 428; 29) home phys-
iotherapy versus day-hospital group (n = 95)474;
30) home-based exercise versus control group (n =
20)472; 31) home therapy versus control group (n =
22)470; 32) combined outpatients and home exer-
cise versus control group (n = 20)472; 33) early care
versus standard customary care in stroke unit (n =
30)444; 34) early supported discharge with home
rehabilitation versus standard customary rehabili-
tation (n = 196)427,455; 35) early supported dis-
charge versus standard customary rehabilitation (n
= 323)460,461; 36) home intervention (rehabilitation
and nursing services) versus standard customary
care (n = 114)469; 37) nursing early activation pro-
gram versus no therapy (n = 59)443; 38) integrated

care pathway versus standard customary
multidisciplinary team care (n = 152)467; 39) full-
time integrated treatment (FIT) versus standard
customary rehabilitation (n = 106)466; 40) critical
path method versus standard customary care
method (n = 121)435; 41) structured nursing inter-
vention versus conventional rehabilitation (n =
155) 456; and 42) home hospitalization versus stan-
dard stroke care (inpatient) (n = 120)422 session or
even all day in one study,460 for a total duration of
treatment of 1 week to 6 months, with follow-up at
2 weeks up to 3 years (Appendix 3M).
In total, there were 96 studies excluded. Among
these, 17 were excluded due to insufficient statisti-
cal data.25,475–490 For 14 other studies, there was
more than a 20% subject drop-out rate.491–505 An-
other major reason for exclusion was the participa-
tion of subjects with either multiple conditions or
multiple strokes, resulting in the exclusion of 19
more studies.506–523 The remaining studies were
excluded for various other reasons outlined in IR-
Table 167.524–569
Efficacy
For stroke unit rehabilitation for patients
with acute, subacute, post-acute, and chronic
stroke versus general ward rehabilitation (eight
RCTs and three CCTs, n = 3,146), 429,431,437,438,448–
450,452,463–465
clinically important benefits with statis-
tical significance were shown for length of stay
(days) at end of treatment, 2 weeks, and follow-up,
3 years (19% and 46% RD, respectively). A clini-
cally important benefit with statistical significance
was demonstrated for mobility (number of pa-
tients able to walk long distances independently)
at end of treatment, 6 weeks (18% RD). A clinically
important benefit without statistical significance
was demonstrated for functional status (number of
patients with mild-moderate impairment) at fol-
low-up, 1 year (16% RD). No benefit was demon-
strated for functional status (number of patients
deteriorated in ADL score) at end of treatment, 3
months, and follow-up, 6 months; functional sta-
tus (number of patients independent in ADL) at
end of treatment, 6 weeks, and length of stay
(days) at end of treatment, 6 weeks and 3 months;
mobility (number of patients independent in walk-
ing mobility) at end of treatment, 16 weeks; mobil-
Clinical Practice Guidelines 75
ity (number of patients independent in walking
[Barthel Index], number of patients independent
in indoor and outdoor walking [Rivermead EADL–
self-care], walking outside, and climbing stairs
[Rivermead EADL household]) at end of treat-
ment, 3 and 6 months; motor function (number of
patients independent in upper extremity and
lower extremity function) at end of treatment, 16
weeks; efficiency (number of patient in-hospital
fatalities, number of patients independent upon
discharge, number of patients with medical com-
plications) at end of treatment, 2 weeks; postural
status (number of patients sitting and standing) at
follow-up, 6 weeks; functional status (ADL score)
at end of treatment, 3 months, and follow-up, 9
months; and resource cost (total cost per patient in
thousands of dollars) at follow-up, 1 year. No
benefit was demonstrated, but results favored the
control treatment for mobility (number of patients
able to walk short distances with or without aids)
at end of treatment, 6 weeks; functional status
(number of functional patients) at follow-up, 1
year; and resource cost (number of physician visits
per patient) at follow-up, 1 year. A clinically im-
portant benefit favoring control was demonstrated
without statistical significance for functional status
(ADL score) at end of treatment, 1 week. No other
outcomes were measured (IR-Figures 151A-F,
Tables 168A-J).
For stroke unit rehabilitation for patients with
acute stroke versus home care (one RCT, n =
447),449 clinically important benefits with statistical
significance were observed for functional status
(number of patients with Barthel score between 15
and 20) at follow-up, 9 months (15% RD). No
benefit was demonstrated for functional status
(number of patients with Modified Rankin score
from 0 to 3) at end of treatment, 3 months, and
follow-up, 9 months; and functional status (number
of patients with Barthel score between 15 and 20) at
end of treatment, 3 months. No other outcomes
were measured (IR-Figure 152, Table 169).
For stroke unit (large artery stroke) rehabili-
tation for patients with acute stroke versus gen-
eral ward (large artery stroke; one RCT, n =
267),434 clinically important benefits with statisti-
cal significance were seen for functional status
(number of patients with Barthel score 15–20) at
follow-up, 1 year (22% RD); and quality of life
76 TOPICS IN STROKE REHABILITATION/SPRING 2006
(Euroqol score) at follow-up, 3 months (22% RD).
No benefit was demonstrated for functional status
(number of patients with Barthel score 15–20) at
follow-up, 3 months, and quality of life (Euroqol
score) at follow-up, 3 months. No benefit was
demonstrated, but there was favor for the control
rehabilitation for length of stay (in days) at follow-
up, 3 months. A clinically important benefit favor-
ing control was demonstrated with statistical sig-
nificance for efficiency (duration of physiotherapy
visit, hours) at follow-up, 3 months. No other
outcomes were measured (IR-Figure 153, Tables
170A&B).
For stroke unit (small artery stroke) rehabili-
tation for patients with acute stroke versus gen-
eral ward (small artery stroke; one RCT, n =
267),434 clinically important benefits with statisti-
cal significance were shown for quality of life
(Euroqol score) at follow-up, 3 months (15% RD).
No benefit was demonstrated for quality of life
(Euroqol score) at follow-up, 1 year. No benefit
was demonstrated, but the control therapy was
favored for functional status (number of patients
with Barthel score 15–20) at follow-up, 3 months
and 1 year. A clinically important benefit favoring
control was demonstrated with statistical signifi-
cance for length of stay (in days) and efficiency
(duration of physiotherapy visit, hours) at follow-
up, 3 months. No other outcomes were measured
(IR-Figure 154, Tables 171A&B).
For intensive outpatient physiotherapy reha-
bilitation for patients with chronic stroke ver-
sus control group (one RCT,n = 54),279 clinically
important benefits without statistical significance
were observed for quality of life (Sickness Index
Profile) at end of treatment, 3 months (28% RD).
No benefit was demonstrated for functional status
(FIM) at end of treatment, 3 months. No benefit
was demonstrated, but results favored the control
therapy for functional status (Sickness Index Pro-
file) at follow-up, 6 months. No other outcomes
were measured (IR-Figure 155, Table 172).
For 6 days/week of rehabilitation for patients
with post-acute stroke versus 7 days/week
treatment, one RCT (n = 113)462 showed clinically
important benefits without statistical significance
for mobility (ambulation section of Functional Re-
covery Scale) at end of treatment, 3 weeks (19%
RD). No benefit was demonstrated for functional
status (dressing section of Functional Recovery
Scale) and length of stay (days). No other out-
comes were measured (IR-Figure 156, Table
173).
For enhanced occupational therapy for pa-
tients with subacute stroke versus standard
customary occupational therapy (five RCTs, n =
492),431,432,440,441,454 clinically important benefits
with statistical significance were demonstrated for
functional status (number of patients improved in
ADL) at end of treatment, 8 weeks and 6 months
(23%–18% RD); life habit/leisure (overall leisure
score) at end of treatment, 3 and 6 months (15%–
24%); life habit/leisure (total leisure activity score)
at end of treatment, 6 months (23% RD); mobility
(Nottingham EADL score for mobility) at end of
treatment, 3 and 6 months (56%–58%); and func-
tional status (Nottingham EADL score) at end of
treatment, 8 weeks, 3 months, and 6 months
(16%, 91%, and 28% RD, respectively). Clinically
important benefits without statistical significance
were demonstrated for quality of life (number of
patients living independently) at end of treatment,
3 weeks (28%); functional status (FIM for upper
extremity and lower extremity dressing; 41% and
50% RD, respectively) at end of treatment, 3 weeks
(28% RD); and functional status (EADL total
score) at follow-up, 3 months (28% RD). No ben-
efit was demonstrated for life habit/leisure (total
leisure score) and mobility (Nottingham EADL for
mobility) at end of treatment, 6 months, or func-
tional status (Barthel Index score) at follow-up, 3
months. No benefit was demonstrated, but the
results favored the control therapy for mobility
(NHP score for mobility) at end of treatment, 3
months. Clinically important benefits favoring the
control therapy were demonstrated without statis-
tical significance for pain relief (NHP score for
pain) at end of treatment, 3 and 6 months, and
quality of life (General Health Questionnaire) at
follow-up, 3 months. No other outcomes were
measured (IR-Figures 157A-D, Tables 174A-D).
For enhanced occupational therapy for pa-
tients with subacute stroke versus no therapy
(five RCTs, n = 481),430,431,433,441,447 clinically impor-
tant benefits with statistical significance were dem-
onstrated for mobility (NHP for mobility and

EADL for mobility) at end of treatment, 3 months
and 6 months (49%–62% and 39%–40% RD, re-
spectively); life habit/leisure (overall leisure score
and total leisure activity) at end of treatment, 3 and
6 months (24%–30% and 20%–30% respectively);
and functional status (number of patients im-
proved in ADL) at follow-up, 6 months (19% RD).
Clinically important benefits were demonstrated
without statistical significance for activity involve-
ment (Katz Adjustment Index: number of patients
satisfied with their walking) at follow-up, 13
weeks (20% RD); activity involvement (number of
patients satisfied with their work in the yard) at
end of treatment, 5 weeks (15% RD); and func-
tional status (EADL) at follow-up, 1 year (40%
RD). No benefit was demonstrated for functional
status (Barthel ADL Index) at follow-up, 6 months
and 1 year; functional status (EADL) at follow-up,
6 months; activity involvement (number of pa-
tients satisfied with their work in and around the
house and number of patients satisfied with their
walking) at end of treatment, 5 weeks; and activity
involvement (number of patients satisfied with
their work in the yard) at follow-up, 13 weeks. No
benefit was demonstrated, but results favored the
control for activity involvement (number of pa-
tient satisfied with their work in and around
house) at follow-up, 13 weeks. Clinically impor-
tant benefits favoring control were demonstrated
without statistical significance for pain relief (NHP
for pain) at end of treatment, 3 and 6 months. No
other outcomes were measured (IR-Figures
158A&B, Tables 175A-D).
For standard customary occupational therapy
for patients with subacute stroke versus no
therapy, three RCTs (n = 378)431,433,471 showed
clinically important benefits with statistical signifi-
cance for functional status (EADL score) at follow-
up, 1 month (48% RD). Clinically important ben-
efits were demonstrated without statistical
significance for mobility (NHP for mobility) at end
of treatment, 3 and 6 months (32%–38% RD), and
for pain relief (NHP) at end of treatment, 3 months
(34% RD). No benefit was demonstrated for life
habit/leisure (overall leisure score and total leisure
activity) at end of treatment, 3 and 6 months; for
functional status (Barthel Index score) at follow-
up, 1 month; and pain relief at end of treatment, 6
Clinical Practice Guidelines 77
months. No benefit was demonstrated, but there
was favor for the control therapy for quality of life
(General Health Questionnaire) at follow-up, 1
month. A clinically important benefit favoring
control was demonstrated without statistical sig-
nificance for functional status (EADL score for
mobility) at end of treatment, 3 months and 6
months. No other outcomes were measured (IR-
Figures 159A&B, Tables 176A&B).
For enhanced physiotherapy (60 minutes of
physiotherapy) for patients with chronic stroke
versus standard customary physiotherapy (30
minutes of physiotherapy) care group, good evi-
dence (one RCT, n = 114)458 showed no benefit for
quality of life (profiles of recovery) at end of treat-
ment, 6 weeks. No benefit was demonstrated, but
results favored the control therapy for quality of
life (profiles of recovery) at end of treatment, 6
months. No other outcomes were measured (IR-
Figure 160, Table 177).
For enhanced physiotherapy for patients with
subacute stroke versus standard customary
physiotherapy (two RCTs, n = 564),452,457 clini-
cally important benefits with statistical signifi-
cance were demonstrated for motor function
(ARAT) at follow-up, 21 weeks (18% RD). A clini-
cally important benefit was demonstrated without
statistical significance for functional status (Barthel
Index) at follow-up, 3 and 16 weeks (15% RD). No
benefits were demonstrated for functional status
(Barthel Index) at end of treatment, 5 weeks, and
follow-up, 21 weeks; functional status (EADL
score) at follow-up, 16 weeks and 21 weeks; motor
function (Rivermead Motor Assessment–upper ex-
tremity) at end of treatment, 5 weeks, and follow-
up, 21 weeks; motor function (Ten-Hole Peg Test)
at end of treatment, 5 weeks, and follow-up, 16
weeks; and motor function (ARAT) at end of treat-
ment, 5 weeks. No benefit was demonstrated, but
results favored the control therapy for functional
status (EADL score) at end of treatment, 5 weeks.
A clinically important benefit favoring control was
demonstrated without statistical significance for
motor function (Rivermead Motor Assessment–
gross function) at follow-up, 3 weeks, and grip
strength (maximum grip) at follow-up, 16 weeks.
Clinically important benefits favoring control were
demonstrated with statistical significance for mo-
78 TOPICS IN STROKE REHABILITATION/SPRING 2006
tor function (Rivermead Motor Assessment–upper
extremity) at follow-up, 3 weeks and 16 weeks;
motor function (ARAT) at follow-up, 16 weeks;
motor function (Rivermead Motor Assessment–
gross function) at end of treatment, 5 weeks, and
follow-up, 16 weeks; and grip strength (maximum
grip) at end of treatment, 5 weeks, and follow-up,
3 weeks. No other outcomes were measured (IR-
Figures 161A&B, Table 178).
For enhanced assistant physiotherapy for pa-
tients with subacute stroke versus standard
customary physiotherapy treatment, two RCTs
(n = 564)452,457 showed clinically important ben-
efits with statistical significance for motor function
(Rivermead Motor Assessment–upper extremity)
at follow-up, 21 weeks (20% RD); and motor func-
tion (ARAT) at end of treatment, 5 weeks, and
follow-up, 21 weeks (17% and 33% RD, respec-
tively). A clinically important benefit was demon-
strated without statistical significance for func-
tional status (Barthel Index) at follow-up, 3 and 16
weeks (≥15% RD); functional status (EADL) at end
of treatment, 5 weeks, and follow-up, 21 weeks
(34% and 28% RD, respectively); and motor func-
tion (ARAT) at follow-up, 16 weeks (≥15% RD).
No benefit was demonstrated for functional status
(Barthel Index) at end of treatment, 5 weeks; func-
tional status (EADL score) at follow-up, 16 weeks;
motor function (Rivermead Motor Assessment–
upper extremity) at end of treatment, 5 weeks, and
follow-up, 16 weeks; motor function (Rivermead
Motor Assessment–gross function) at follow-up, 3
and 16 weeks; and motor function (Ten-Hole Peg
Test) at end of treatment, 5 weeks, and follow-up,
16 weeks. No benefit was demonstrated, but re-
sults favored the control therapy for functional
status (Barthel Index) at follow-up, 21 weeks.
Clinically important benefits favoring control were
demonstrated without statistical significance for
motor function (Rivermead Motor Assessment–
upper extremity) at follow-up, 3 weeks, and grip
strength (maximum grip) at end of treatment, 5
weeks, and follow-up, 16 weeks. A clinically im-
portant benefit favoring the control therapy was
demonstrated with statistical significance for mo-
tor function (Rivermead Motor Assessment–gross
function) at end of treatment, 5 weeks, and grip
strength (maximum grip) at follow-up, 3 weeks.
No other outcomes were measured (IR-Figures
162A&B, Table 179).
For enhanced physiotherapy for patients with
subacute stroke versus enhanced assistant
physiotherapist treatment (two RCTs, n =
564),452,457 clinically important benefits with statis-
tical significance were shown for motor function
(Rivermead Motor Assessment–gross function) at
end of treatment, 5 weeks (100% RD); grip
strength (maximum grip) at follow-up, 3 weeks
(≥15% RD); and motor function (ARAT) at follow-
up, 21 weeks (33% RD). Clinically important ben-
efits with statistical significance favoring enhanced
assistant physiotherapist were found for functional
status (Barthel Index) at end of treatment, 5 weeks
(19% RD); motor function (Rivermead Motor As-
sessment–UE) at follow-up, 16 weeks, and 21
weeks (200% and 20% RD, respectively); func-
tional status (EADL score) at follow-up, 21 weeks
(27% RD); and motor function (ARAT) at end of
treatment, 5 weeks, and follow-up, 16 weeks (19%
and ≥15% RD, respectively). A clinically important
benefit was demonstrated without statistical sig-
nificance for motor function (Rivermead Motor
Assessment–gross function) at follow-up, 16
weeks (250% RD). A clinically important benefit
was demonstrated without statistical significance
favoring enhanced assistant physiotherapist for
functional status (Barthel Index) at follow-up, 3
and 16 weeks (15% RD), functional status (EADL)
at end of treatment, 5 weeks (18% and 43% RD,
respectively); motor function (Rivermead Motor
Assessment–upper extremity) at follow-up, 3
weeks (100%); and grip strength at end of treat-
ment, 5 weeks, and follow-up, 16 weeks (≥15%
RD). No benefit was demonstrated for functional
status (Barthel Index) at follow-up, 21 weeks;
functional status (EADL score) at follow-up, 16
weeks; motor function (Rivermead Motor Assess-
ment-upper extremity) at end of treatment, 5
weeks; and motor function (Ten-Hole Peg Test) at
end of treatment, 5 weeks, and follow-up, 16
weeks. No other outcomes were measured (IR-
Figures 163A&B, Table 180).
For enhanced upper-limb treatment for pa-
tients with subacute stroke versus interdiscipli-
nary treatment, one RCT (n = 626)459 showed a
clinically important benefit with statistical signifi-

cance for motor function (Frenchay Arm Test) and
functional status (Barthel Index) at follow-up, 18
weeks. No benefit was demonstrated for motor
function (Upper Limb Motricity Index and
Frenchay Arm Test) at follow-up, 6 weeks; func-
tional status (Nottingham EADL) at follow-up, 18
weeks; and functional status (Barthel Index) at
follow-up, 6 weeks. No benefit was demonstrated,
but results favored the control treatment for motor
function (Upper Limb Motricity Index) at follow-
up, 18 weeks, and functional status (Nottingham
EADL) at follow-up, 6 weeks. Clinically important
benefits favoring the control treatment were also
demonstrated without statistical significance for
motor function (ARAT) at follow-up, 6 and 18
weeks. No other outcomes measured (IR-Figure
164, Table 181).
For enhanced therapy (severe stroke) for pa-
tients with acute stroke versus standard cus-
tomary care (severe stroke), one RCT (n =
137)468 showed no benefit but the results favored
the control therapy for mobility (number of pa-
tients independent in walking) at end of treatment,
6 months. No other outcomes were measured (IR-
Figure 165, Table 182).
For enhanced therapy (mild stroke) for pa-
tients with acute stroke versus standard cus-
tomary care (mild stroke; one RCT, n = 137),468
no benefit was shown but results favored the con-
trol therapy for mobility (number of patients inde-
pendent in walking) at end of treatment, 6
months. No other outcomes were measured (IR-
Figure 166, Table 183).
For home visit (physiotherapy) for patients
with chronic stroke versus standard customary
care without home visit groups (two RCTs, n =
310),423,424 a clinically important benefit without
statistical significance was observed for functional
status (IADL–domestic activities) at follow-up, 6
months. No benefit was demonstrated for discharge
disposition (number of readmissions into hospital
because of stroke-related condition) and functional
status (FQM motor performance, FQM quality of
movement, Barthel Index, FAI, EADL–personal
daily care), mobility (EADL–mobility), and life hab-
its/leisure (IADL–leisure/social activities) at follow-
up, 6 months. No other outcomes were measured
(IR-Figures 167A&B, Tables 184A&B).
Clinical Practice Guidelines 79
For high-intensity home-based physiotherapy
for patients with chronic stroke versus low-
intensity physiotherapy (one RCT, n = 19),451
clinically important benefits with statistical signifi-
cance were shown for motor function (STREAM–
lower extremity) at follow-up, 11 weeks (25%
RD). Clinically important benefits were demon-
strated without statistical significance for func-
tional status (Barthel Index) and motor function
(STREAM–upper extremity) at follow-up, 11
weeks (17% and 21% RD, respectively). No ben-
efit was demonstrated for mobility (STREAM–ba-
sic mobility) at follow-up, 11 weeks, and func-
tional status (Barthel Index) and motor function
(STREAM–upper extremity) at follow-up, 22
weeks. No benefit was demonstrated but results
favored the control therapy for mobility
(STREAM–basic mobility) at follow-up, 22 weeks.
A clinically important benefit favoring control was
demonstrated without statistical significance for
motor function (STREAM–lower extremity) at fol-
low-up, 22 weeks. No other outcomes were mea-
sured (IR-Figure 168, Table 185).
For home-based rehabilitation for patients
with post-acute and chronic stroke versus hos-
pital-based rehabilitation (one RCT and one
CCT, n = 205),425,445 no clinically important differ-
ence was shown for resource cost (total cost) at
follow-up, 1 year, mobility (number of patients
able to walk 10 m without an aid), functional
status (number of patients independent in Barthel
ADL Index and number of patients independent in
IADL) at end of treatment, 3 months. No other
outcomes were measured (IR-Figure 169, Tables
186A&B).
For rehabilitation for patients with subacute
stroke versus no rehabilitation group, one CCT
(n = 43)442 showed no benefit but the results fa-
vored the control rehabilitation for motor function
(Brunnstrom Upper Extremity Scale, Brunnstrom
Lower Extremity Scale, Upper Extremity Func-
tional Test) at end of treatment, 6 months. A clini-
cally important benefit favoring control was dem-
onstrated without statistical significance for
functional status (Barthel ADL Index) at end of
treatment, 6 months. No other outcomes were
measured (IR-Figure 170, Table 187).
For home therapy (physician) for patients
80 TOPICS IN STROKE REHABILITATION/SPRING 2006
with chronic stroke versus standard customary
care without home visits (two RCTs, n =
310),423,424 clinically important benefits with statis-
tical significance were shown for decrease in hos-
pital readmissions (number of readmissions into
hospital because of stroke-related condition) and
functional status (FAI) at follow-up, 6 months
(18% and 20% RD, respectively). No benefit was
demonstrated for functional status (FQM motor
performance, Barthel Index, EADL–personal daily
care, IADL–domestic activities), mobility (EADL–
mobility), and life habits/leisure (IADL–leisure/so-
cial activities) at follow-up, 6 months. No benefit
was demonstrated, but results favored the control
therapy for functional status (FQM quality of
movement) at follow-up, 6 months. No other out-
comes were measured (IR-Figures 171A&B,
Tables 188A&B).
For home therapy given by physician to pa-
tients with chronic stroke versus home therapy
given by a physiotherapist (one RCT, n =
310),423,424 a clinically important benefit without
statistical significance was shown favoring home
physiotherapy (physiotherapy) of home visit (phy-
sician) for functional status (IADL–domestic ac-
tivities) at follow-up, 6 months (23% RD). No
benefit was demonstrated for decrease in hospital
readmission (number of readmissions into hospi-
tal because of stroke-related condition) and func-
tional status (FAI, FQM motor performance, FQM
quality of movement, Barthel Index, and EADL–
personal daily care), mobility (EADL–mobility),
and life habits/leisure (IADL–leisure/social activi-
ties) at follow-up, 6 months. No other outcomes
were measured (IR-Figures 172A&B, Tables
189A&B).
For extended stroke unit service with early
discharge for patients with acute stroke versus
ordinary stroke unit service (three RCTs, n =
702),426,439,446 a clinically important benefit with
statistical significance was demonstrated for func-
tional status (number of patients with Rankin
score less than 2) at follow-up, 6 months (18%
RD). A clinically important benefit without statisti-
cal significance was demonstrated for physical mo-
bility (NHP) at follow-up, 3 weeks and 1 year
(30% and 23% RD). No benefit was demonstrated
for functional status (number of patients with
Barthel Index score higher than 95) at follow-up, 3
weeks, 6 months, and 1 year; discharge status
(number of patients discharged at home); prevent-
ing mortality (number patients deceased) at fol-
low-up, 1 year; energy level and pain (NHP) at
follow-up, 3 weeks, 6 months, and 1 year; global
health status score (NHP) at follow-up, 1 year; and
functional status (number of patients with Rankin
score less than 2) at follow-up, 3 weeks and 1 year.
No clinically important benefit was demonstrated,
but results favored the control treatment for global
health status score (NHP) at follow-up, 3 weeks
and 6 months. A clinically important benefit with-
out statistical significance favoring control was
demonstrated for physical mobility (NHP) at fol-
low-up, 6 months. No other outcomes were mea-
sured (IR-Figures 173A-C, Tables 190A&B).
For early physiotherapy intervention for pa-
tients with acute stroke versus control group
(one RCT, n = 128),436 a clinically important ben-
efit without statistical significance was demon-
strated for functional status (Modified Barthel In-
dex) and motor function (Fugl-Meyer score for
lower extremity) at end of treatment, 1 month, and
follow-up, 5 months (27%–39% and 21%–15%
RD, respectively) and motor function (Fugl-Meyer
for upper extremity) at end of treatment, 1 month
(18% RD). No benefit was demonstrated, but re-
sults favored the control group for motor function
(Fugl-Meyer score for upper extremity) at follow-
up, 5 months. No other outcomes were measured
(IR-Figure 174, Table 191).
For home-based exercise program training
for patients with post-acute stroke versus con-
trol (two RCTs, n = 112),103,104 clinically impor-
tant benefits with statistical significance were ob-
served for change in gait speed (48% RD), gait
endurance (change in 6-minute walk test, feet;
59% RD), torque (change in knee isometric ex-
tensors; 61% RD), endurance (change in duration
of bicycle exercies; 164% RD), and cardiopulmo-
nary function (change in peak VO2; 185% RD) at
end of treatment, 12 weeks. A clinically impor-
tant benefit was demonstrated without statistical
significance for motor function (change in Fugl-
Meyer lower extremity; 17% RD), change in gait
speed (94% RD), gait endurance (change in 6-
minute walk test, feet; 15% RD), functional status

(physical function index; 20% RD), and strength
(change in grip strength; 17% RD) at end of
treatment, 12 weeks. No benefit was demon-
strated for motor function (change in Fugl-Meyer
upper extremity and lower extremity), balance
(Berg balance and change in Berg balance), func-
tional status (IADL and Barthel ADL Index) at
end of treatment, 12 weeks. A clinically impor-
tant benefit favoring the control was demon-
strated without statistical significance for balance
(functional reach) at end of treatment, 12 weeks.
No benefit was demonstrated, but results favored
the control for torque (change in ankle isometric
dorsiflexors) at end of treatment, 12 weeks. No
other outcomes were measured (IR-Figures
175A-C, Table 192).
For outpatient therapy for patients with
chronic stroke versus home-exercise group (one
RCT,n = 20),472 clinically important benefits with-
out statistical significance favoring home exercise
group were observed for gait speed at end of treat-
ment, 6 months, and follow-up, 3 months (16%–
24% RD). No benefit was demonstrated for single
support time at end of treatment, 6 months, and
follow-up, 3 months. No other outcomes were
measured (IR-Figure 176, Table 193).
For outpatient therapy for patients with
chronic stroke versus control group, one RCT (n
= 20)472 showed no benefit but favored the control
group for gait speed at follow-up, 3 months. Clini-
cally important benefits favoring control were
demonstrated without statistical significance for
gait speed at end of treatment, 6 months, and
single support time at end of treatment, 6 months,
and follow-up, 3 months. No other outcomes were
measured (IR-Figure 177, Table 194).
For outpatient therapy for patients with sub-
acute stroke versus home therapy, good evidence
was found (one RCT, n = 100)428 that showed no
benefits for grip strength on affected side and motor
assessment, gait speed (10 m walking speed), or
functional status (Barthel Index) at end of treat-
ment, 6 weeks and 3 months. No other outcomes
were measured (IR-Figure 178, Table 195).
For day-hospital for patients with subacute
stroke versus home physiotherapy group, one
RCT (n = 95)474 showed a clinically important
benefit with statistical significance for resource
Clinical Practice Guidelines 81
cost (total cost in pounds) at end of treatment, 8
weeks. No other outcomes were measured (IR-
Figure 179, Table 196).
For home exercise for patients with chronic
stroke versus control, another RCT (n = 20)472
showed no benefit for gait speed at end of treat-
ment, 6 months, and follow-up, 3 months. No ben-
efit was demonstrated, but results favored the con-
trol for single support time at end of treatment, 6
months, and follow-up, 3 months. No other out-
comes were measured (IR-Figure 180, Table 197).
For home therapy for patients with subacute
stroke versus control group, one CCT (n =
22)470 demonstrated no benefit for motor func-
tion (Peg Test Time) at end of treatment, 9 weeks.
No other outcomes were measured (IR-Figure
181, Table 198).
For combined outpatient and home exercise
for patients with chronic stroke versus control
group, (one RCT, n = 20)470 no benefit was illus-
trated for gait speed. No benefit was demon-
strated, but results favored the control group for
single support time at end of treatment, 6 months,
and follow-up, 3 months. No other outcomes were
measured (IR-Figure 182, Table 199).
For early care for patients with acute stroke
versus standard customary care in stroke unit
(one RCT, n = 30),444 a clinically important benefit
with statistical significance was shown for length
of stay (days). No benefit was demonstrated for
functional status (ADL score) and mobility (ambu-
latory ability) at end of treatment, 6 weeks. No
other outcomes were measured (IR-Figure 183,
Table 200).
For early supported discharge with home re-
habilitation for patients with acute stroke ver-
sus standard customary rehabilitation, one RCT
(n = 196)427,455 showed a clinically important ben-
efit with statistical significance for functional status
(Older Americans Resource Scale for Instrumental
ADL) at end of treatment, 1 and 3 months (16%
and 15% RD, respectively) and length of stay at
follow-up, 3 months (24% RD). No benefit was
demonstrated for motor function (STREAM); mo-
bility (Timed Up & Go) at end of treatment, 1
month and 3 months; quality of life at follow-up, 6
months; and functional status (Barthel Index) and
pain relief at end of treatment, 3 months. No ben-
82 TOPICS IN STROKE REHABILITATION/SPRING 2006
efit was demonstrated, but results favored the con-
trol rehabilitation for functional status (Nottigham
EADL score) at end of treatment, 3 months, and
follow-up, 6 months; quality of life at end of treat-
ment, 3 months; and pain relief and functional
status (Barthel ADL Index) at end of treatment, 1
month. No other outcomes were measured (IR-
Figures 184A-D, Tables 201A-C).
For early supported discharge for patients
with acute and post-acute stroke versus rou-
tine rehabilitation (two RCTs, n = 323),459,461 a
clinically important benefit with statistical signifi-
cance was observed for length of stay and quality
of life (Dartmouth Coop Global Health Status–
total) at end of treatment, 3 months (40%–49%
and 40% RD, respectively) and functional status
(Nottingham EADL) at follow-up, 3 months
(35% RD). No benefit was demonstrated for mo-
tor function (Motricity Index) and functional sta-
tus (Rivermead ADL score) at follow-up, 9
months; gait speed (5 m timed walk) and func-
tional status (Barthel ADL Index) at end of treat-
ment, 3 months, and follow-up, 9 months; and
quality of life (Dartmoyth Coop Global Health
Status–physical fitness, daily activities, and pain
relief) at end of treatment, 3 months. No benefit
was demonstrated, but results favored the control
rehabilitation for motor function and functional
status (Nottingham ADL score) at end of treat-
ment, 3 months. A clinically important benefit
favoring control was demonstrated without sta-
tistical significance for functional status
(Nottingham ADL score) at follow-up, 9 months.
No other outcomes were measured (IR-Figures
185A-C, Tables 202A&B).
For home intervention (rehabilitation and
nursing services) for patients with subacute
stroke versus standard customary care, one RCT
(n = 114)469 showed a clinically important benefit
with statistical significance for resource cost (total
cost) at follow-up, 3 months (35% RD). No benefit
was demonstrated for quality of life (SF-36) at fol-
low-up, 3 months. No other outcomes were mea-
sured (IR-Figure 186, Table 203).
For nursing early activation program for pa-
tients with acute stroke versus no therapy, one
RCT (n = 59)443 showed no benefits for mobility
(number of patients independent in walking) and
functional status (number of patients independent
in household work and ADL Index score) at fol-
low-up, 1 year. No other outcomes were measured
(IR-Figure 187, Tables 204A&B).
For integrated care pathway for patients with
acute stroke versus standard customary
multidisciplinary team care, one RCT (n =
152)467 showed no benefit for functional status
(number of patients with no problems with self-
care, number of patients with no problems with
usual activities) and pain relief at end of treatment,
6 months. No benefit was demonstrated, but re-
sults favored the control treatment for mobility
(number of patients with some problems walking
and with no problems walking), functional status
(number of patients with some problems with
usual activities), and pain relief at end of treat-
ment, 6 months. A clinically important benefit
favoring the control treatment was demonstrated
without statistical significance for functional status
(number of patients with some problems with
washing and dressing) at end of treatment, 6
months. No other outcomes measured (IR-Figure
188, Table 205).
For full-time integrated treatment for pa-
tients with acute stroke versus standard cus-
tomary rehabilitation, one CCT (n = 106)466
showed no benefit for functional status (FIM mo-
tor subscale and total) at end of treatment, 4 weeks
and 6 weeks, and for length of stay at end of
treatment, 4 weeks. No other outcomes were mea-
sured (IR-Figure 189, Table 206).
For the critical path method for patients with
acute stroke versus the standard customary
care method, one RCT (n = 121)435 showed no
benefits for functional status (FIM motor subscale)
at end of treatment, 1 month. No benefit was
demonstrated, but there was favor for the control
method for functional status (FIM total raw score)
and length of stay (days) at end of treatment, 1
month. No other outcomes were measured (IR-
Figure 190, Table 207).
For structured nursing intervention for pa-
tients with acute stroke versus conventional
rehabilitation, the evidence (one RCT, n = 155)456
showed no benefit for functional status (FIM and
IADL) and self-perception of health care at end of
treatment, 3 months, and follow-up, 3 months. No

other outcomes were measured (IR-Figure 191,
Table 208).
For home hospitalization for patients with
acute stroke versus standard stroke care (inpa-
tient), one RCT (n = 120)422 showed no benefit,
but results demonstrated a favor for the control
treatment for neurological status (Neurological
Impairment Score) and functional status (Func-
tional Impairment Score) at follow-up, 6 months.
No other outcomes were measured (IR-Figure
192, Table 209).
Strength of published evidence compared with other
guidelines
The Ottawa Panel found level I evidence (2
RTCs) for multidisciplinary team care, early care
(3 RCTs), early supported discharge (6 RCTs),
outpatient therapy (6 RCTs), physiotherapy (9
RCTs), occupational therapy (13 RCTs), and en-
hanced therapy (24 RCTs) in post-stroke rehabili-
tation. The Ottawa Panel also found level I and
level II evidence for conventional or standard cus-
tomary rehabilitation (15 RCTs and 3 CCTs),
stroke unit and general ward care (15 RCTs, 3
CCTs), care pathways (2 RCTs, 1 CCT), and home
therapy (18 RCTs, 2 CCTs).
In comparison, the VA/DoD75 also found level
I evidence for early rehabilitation and inpatient
versus outpatient stroke rehabilitation. The
Scottish Intercollegiate Guidelines Network74
equally found level I evidence for the treatment
of acute stroke patients in a multidisciplinary
unit and level III evidence for occupational
therapy treatment. The European Stroke Initia-
tive78 found evidence for the referral to a home
management program. However, the type of evi-
dence (RCT or CCT) was not reported. The
Royal College of Physicians72 found level I evi-
dence concerning the expertise of specialists in a
stroke unit and expertise of care in home ser-
vices. Last, St. Joseph’s Health Care London73
also found level I evidence for the intensity of
therapy, care pathways in rehabilitation, reha-
bilitation in specialized stroke units, functional
outcomes from rehabilitation, and additional
home-based therapy and level II evidence for
early admission to rehabilitation.
Clinical Practice Guidelines 83
Clinical recommendations compared with other
guidelines
The Ottawa Panel supports early care (grade A
for length of stay) and early physiotherapy inter-
vention (grade C+ for functional status and motor
function for the lower and upper extremities) for
patients with acute stroke. This is in line with the
VA/DoD75 that found good evidence that early re-
habilitation therapy should be provided as soon as
the patient’s medical status is stable (Appendix 4).
Similarly, St. Joseph’s Health Care London73 found
fair evidence that an early admission to post-stroke
rehabilitation should be considered for medically
stable stroke patients (Appendix 4).
The Panel found good evidence supporting
treatment in a stroke unit (grade A for length of
stay, functional status, and quality of life; grade B
for mobility; and grade C+ for functional status)
and poor evidence supporting treatment in a gen-
eral ward (grade D+ for functional status) for
patients with acute to chronic stroke. This is in
line with the Royal College of Physicians,72 which
found good evidence that acute care for patients
undergoing post-stroke rehabilitation in a hospi-
tal should be delivered in a ward or stroke unit
with specialists with expertise in stroke manage-
ment. As articulated by the St. Joseph’s Health
Care London 73 guidelines, the Ottawa Panel
found that stroke unit care versus care in a gen-
eral ward was better for functional status out-
comes (grades A and C+) for patients with acute
to chronic stroke (Appendix 4).
The Ottawa Panel found evidence in favor for an
interdisciplinary treatment approach (grade D+ for
arm motor function) for patients with subacute
stroke and in favor for a standard customary
multidisciplinary team care (grade D+ for func-
tional status) for patients with acute stroke. This is
in line with the Scottish Intercollegiate Guidelines
Network74 that found good evidence that stroke
patients in the acute phase should be treated in a
multidisciplinary stroke unit (Appendix 4).
The Ottawa Panel found poor evidence for the
use of care pathways in rehabilitation such as an
integrated care pathway (grade C for functional
status and pain relief, and grade D for mobility,
functional status, and pain relief) and full-time inte-
84 TOPICS IN STROKE REHABILITATION/SPRING 2006
grated treatment (FIT) (grade C for functional status
and length of stay) and for a critical path method
(grade C for functional status) for patients with
acute stroke. Similarly, St. Joseph’s Health Care
London73 found good evidence that care pathways
do not improve post-stroke rehabilitation outcomes
and do not reduce hospital costs or decrease the
length of hospital stays for patients undergoing
post-stroke rehabilitation (Appendix 4).
The Ottawa Panel also agrees with the VA/
DoD75 that there is insufficient evidence as to
whether inpatient stroke rehabilitation care dem-
onstrates superiority in outcomes, when com-
pared to outpatient stroke rehabilitation. The
Panel found that stroke unit rehabilitation was
preferred (grade A and C for functional status)
over home rehabilitation and standard inpatient
stroke care was preferred (grade D for neurologi-
cal status and functional status) instead of home
hospitalization for patients with acute stroke.
However, there is also support for home-based
rehabilitation (grade C for resource cost, mobil-
ity, and functional status) versus hospital-based
rehabilitation for patients with post-acute and
chronic stroke and for home physiotherapy
(grade A for resource cost) versus day-hospital for
patients with subacute stroke (Appendix 4).
The Ottawa Panel found good evidence to rec-
ommend early supported discharge with home
rehabilitation (grade A for functional status) for
patients with acute stroke, while the European
Stroke Initiative78 found poor evidence as to
whether care services that are able to provide ad-
equate and flexible support within 24 hours for
stroke patients should be considered for a home
management program (Appendix 4).
The Panel found good evidence that treatment
in a stroke unit was preferred over home care
(grade A for functional status) for patients with
acute stroke. This is in line with the Royal College
of Physicians,72 which found that there is good
evidence that stroke patients should only be man-
aged at home, if the services delivered at home
are part of a stroke service provided by special-
ists. Otherwise, these patients should be admitted
to the hospital for initial care and assessment
(Appendix 4).
For enhanced therapies, the Panel found evi-
dence in favor of extended stroke unit service with
early supported discharge (grade A for functional
status, and grade C+ for physical mobility) for
patients with acute stroke, enhanced physio-
therapy (grade A for motor function, and grade C+
for functional status), and enhanced occupational
therapy (grade A for functional status, life habit/
leisure, and mobility; and grade C+ for quality of
life and functional status) for patients with sub-
acute stroke. Similarly, St. Joseph’s Health Care
London73 found good evidence that a faster recov-
ery and an earlier discharge from the hospital can
result from intense therapies delivered in short
periods of time within the post-stroke rehabilita-
tion process (Appendix 4).
The Ottawa Panel found some support for home
therapy (physiotherapy or exercise training; grades
C+ and C) and home-based rehabilitation (grade
C) for patients with post-acute and chronic stroke,
for home therapy (provided by a physician; grades
A and C) for patients with chronic stroke, and for
early supported discharge with home rehabilita-
tion (grades A and C) for the outcome of functional
status for patients with acute stroke. This is in
partial agreement with St. Joseph’s Health Care
London,73 which found good evidence that addi-
tional home-based therapy does not improve the
overall functional outcome scores for patients un-
dergoing post-stroke rehabilitation (Appendix 4).
All other interventions and outcomes evaluated
by the Ottawa Panel were not compared, because
they did not apply to other preexisting clinical
guidelines (Appendix 4).
Discussion
For the past 5 years, evidence-based practice in
the area of stroke rehabilitation has been gaining
popularity. The Ottawa Panel conducted a system-
atic review of the literature, which revealed a num-
ber of physical interventions in post-stroke reha-
bilitation whose clinical effectiveness was
supported by one or more comparative controlled
studies (CCSs). Based on this systematic review,
EBCPGs were developed, resulting in 147 positive
recommendations with grades of A, B, or C+.
Other interventions in stroke rehabilitation re-
quire further evidence to prove their effectiveness

with regard to specific outcomes, as the Ottawa
Panel also came up with 272 negative recommen-
dations with grades of C (n = 148), D (n = 79), D+
(n = 41), and D- (n = 4). Consequently, the Ottawa
Panel does not recommend the inclusion of these
interventions for the management of stroke survi-
vors for specific outcomes
The strength of the Ottawa Panel’s post-stroke
guidelines lies in the rigorous methodology used
to develop them. The evidence for the Ottawa
Panel’s EBCPGs was based on systematic reviews
and meta-analyses that were conducted according
to Cochrane Collaborations methods. A systematic
method of grading the evidence designed by an
earlier expert panel (Philadelphia Panel)90 was
used, in which the direction, statistical signifcance,
and clinical importance of the reported outcomes
and the overall study design (RCT vs. CCT) were
taken into account. To establish both the clarity
and practicality of the guidelines, several practitio-
ners sat on the Ottawa Panel and participated in
feedback surveys regarding the relevance of these
guidelines in their daily practice. Finally, the Ot-
tawa Panel guidelines were evaluated according to
the Appraisal of Guidelines Research and Evalua-
tion (AGREE) criteria84 (www.agreecollaboration.
org), resulting in the following scores. On dimen-
sions 1 (purpose), 2 (stakeholder involvement), 4
(clarity), and 6 (editorial independence), the
guidelines received excellent scores. Lower scores
were obtained on dimensions 3 (rigor of develop-
ment) and 5 (applicability). Inadequate reporting
of side effects and risks in the primary trials low-
ered the rigor of the development score. In identi-
fying cost implications, potential organizational
barriers, and methods of applying and monitoring
the guidelines, the EBCPGs’ applicability was rated
as low. Exact scores and a decision aid tool are
available on the University of Ottawa School of
Rehabilitation Sciences’ Web page (www.health.
uottawa.ca/rehabguidelines).
In comparing the Ottawa Panel guidelines with
other existing guidelines on post-stroke rehabilita-
tion, it is fair to say that, in general, there is relative
consistency (Appendix 4). However, some dis-
agreements exist. For example, when the Ottawa
Panel assigned a grade of C+ for shoulder strap-
ping and therapeutic ultrasound, considered a
Clinical Practice Guidelines 85
positive recommendation by the Panel, other
guidelines either did not recommend the interven-
tion or found the evidence to be insufficient to
make a recommendation. Second, several guide-
lines used a more descriptive approach to obtain
consensus recommendations,70,72–76 while others,
like the Ottawa Panel, used a more quantitative
approach.71 These differences could potentially ex-
plain some of the variation in guideline recom-
mendations.
It is important to point out that the Ottawa Panel
EBCPGs are not without limitations. First of all,
the strength of clinical practice guidelines depends
upon the quality of the primary studies found in
the literature. The clinical studies that met the
Ottawa Panel’s selection criteria rarely exceeded 3
out of 5 on the Jadad scale,94 and the sample sizes
were generally small. These methodological issues
limit the reliability of the reported outcomes and
the overall quality of the evidence. For example, it
is often difficult to achieve adequate blinding with
physical treatments that produce cutaneous sensa-
tion.570 However, all guidelines developers face
these same issues with regard to methodological
considerations. Of additional note, heterogeneity
with respect to interventions, treatment schedules,
study populations, outcome measures, and com-
parators was frequently encountered, which re-
duced the comparability of individual trials. As a
result, quantitative pooling of data through meta-
analysis was not appropriate in most cases.
Equally, the findings were sometimes inconsistent
from one study or outcome measure to the next.
Weighing the evidence in such situations inevita-
bly involves value judgments and is subject to
interpretation. Due to the absence of a clear con-
sensus with regard to the relative importance of
specific, validated outcome measures, individual
study findings were not weighted according to the
type of outcome assessed or measurement scale
used.
The Ottawa Panel also faced other limitations
with regard to the development of these guide-
lines. Articles in the scientific literature were only
considered if they were written in English or
French due to the additional time and resources
required for translation. Moreover, the categoriza-
tion of studies according to the type of interven-
86 TOPICS IN STROKE REHABILITATION/SPRING 2006
tion examined was not always straightforward, be-
cause in some cases a particular study could be
applied to several categories. A decision was made
as to which category of intervention a particular
study best belonged in order to avoid duplication.
This decision was inherently subjective and could
contribute to potential variation in the Ottawa
Panel’s recommendations with other published
clinical practice guidelines.
With regard to the calculation of treatment ben-
efit, the Ottawa Panel considered a 15% improve-
ment relative to control as clinically important.
However, this criterion remains somewhat arbi-
trary and may not be applicable to all rehabilita-
tion interventions or outcome measures. Interven-
tions that showed clinically important benefits
without statistical significance for validated out-
comes (grade C+) were interpreted as worthy of
consideration in the rehabilitation of stroke pa-
tients and were given positive recommendations.
Most of the existing EBCPGs on stroke rehabilita-
tion did not consider clinical significance in syn-
thesizing the evidence, which may further account
for any differences in recommendations made by
other guideline development groups. In the calcu-
lations of clinical relevance, difficulties also arose
when the variance of data was not directly pro-
vided in the published articles. As a result, the
Ottawa Methods Group, working closely with a
senior biostatistician (G.A.W.), developed a stan-
dardized methodology to estimate the variance of
data (Appendix 2). This was the best conservative
approximation that could be used to produce the
Ottawa Panel recommendations.
Finally, the Ottawa Panel did not formally assess
the cost-effectiveness of the various interventions
studied. It is recognized, however, that cost and
resource availability are important factors in the
individual clinician’s descion-making process.
Therapeutic exercises
According to the Ottawa Panel, there is strong
evidence suggesting that aerobic training, strength
training (either combined with aerobic exercise or
alone), robot-aided training, water-based training,
and mental imagery should be included as inter-
ventions in post-stroke rehabilitation. The Ottawa
Panel also recommends progressive resistance
muscle training, functional task training, active
shoulder ROM using a skateboard, resisted and
ballistic finger resisted extension training, agility
exercise training, music-making training, and
maximal isokinetic strengthening depending on
the specific characteristics of individual stroke pa-
tients. However, the Ottawa Panel does not recom-
mend the following: 1) PNF for increasing ankle
ROM in post-acute stroke, especially in presence
of spasticity; 2) the use of overhead pulley in the
subacute phase of stroke, especially if the shoulder
is subluxed (see Shoulder Management section);
and 3) home-based exercise for improving bal-
ance, because patients need greater supervision to
ensure their safety.
The Ottawa Panel agrees with other clinical
guidelines70–76 in not fully recommending the use
of the Bobath approach/NDT at large. The Ottawa
Panel recommends the use of the Bobath ap-
proach/NDT for at least some outcomes, such as
for improving motor function and quality of life
over the long term. However, we recommend that
the same approach be excluded in the manage-
ment of short-term pain and sitting balance among
stroke survivors who are in the acute and subacute
phases of recovery. Over time, NDT has tended to
integrate more functional components in their ap-
proach to treat stroke patient, but these elements
do not seem to translate into functional outcomes
as reflected by validated outcomes measures used
in stroke rehabilitation.
The Ottawa Panel’s recommendations are consis-
tent with recent systematic reviews on therapeutic
exercise.14 Meek et al.571 conducted a systematic
review of three studies that were included by the
Ottawa Panel and found similar results. In another
review, van der Lee572 did not find enough evidence
to determine which form of therapeutic exercises
produces the best outcomes in post-stroke rehabili-
tation. The Ottawa Panel is generally in concor-
dance with other existing reviews on the efficacy of
post-stroke general therapeutic exercise,13,478,573
aerobic training,7,8 progressive strengthening exer-
cise,17,573 and robot-aided training.13,18,19 Post-stroke
aerobic exercise, strengthening, functional training
(either combined or alone; see Task-Oriented sec-
tion), robot-aided training, water-based training,

mental imagery,574–578 and progressive resistance
muscle training all improve motor unit recruitment,
which is limited in hemiparesis,200 and promote
muscle strength and endurance. In their review,
Potempa et al. also found that aerobic exercise after
stroke improves tolerance to ADLs by increasing the
maximal rate of oxygen consumption and lowering
heart rate and systolic blood pressure.156 Aerobic
training plays an important role in the reduction of
cardiovascular risk factors and in maximizing the
patient’s physical endurance to fully participate in
community life.
The advantage of resistive training in compari-
son to passive exercise is that active participation
better promotes central nervous system activity
and neural reorganization.579,580 It is based on the
overload principle that states that a threshold of
stress must be exceeded for muscle adaptation to
occur. Muscle strengthening in the context of
functional, task-oriented training improves
strength not only in paretic muscles, but in all of
the muscle groups necessary to function as well.
The key to a successful program is to understand
the patient’s needs and develop functional objec-
tives.581 Resisted and ballistic extension for the
hand were all favored compared to resisted grasp
and control for improvements in finger tapping.124
Subjects were given immediate feedback. Atten-
tion, practice, and feedback are all components of
motor learning and may indicate that practice of a
specific movement with concentration is more im-
portant in regaining motor control than recruit-
ment of motor units. This could explain why the
ballistic and resisted extension demonstrated
greater benefits in finger tapping. Exercises prac-
ticed at different specific speeds using a Kinetron
may contribute to improve muscle contraction and
may reflect what happens in a functional con-
text.582 However, not all clinical settings have ac-
cess to costly equipment such as Kinetron and
robot-aided devices. Furthermore, patients may
not be able to reproduce these sophisticated exer-
cises in the home setting.
The Ottawa Panel agrees with previous reviews
that more research is needed to establish the opti-
mal exercise intensity, frequency, duration, dosage,
and content according to different patient character-
istics. More details are urgently needed on which
Clinical Practice Guidelines 87
type of TE treatment may achieve the most optimal
health outcomes under the safest conditions in this
heterogeneous population.8,478,555
Task-oriented training
The Ottawa Panel strongly recommends the
consideration of task-oriented training as a thera-
peutic strategy in stroke rehabilitation. The follow-
ing interventions are supported by the available
evidence: seated reaching task training, functional
upper extremity training, functional task training
for the upper extremity, task-specific reach train-
ing, trunk control training, trunk rotation training,
traditional functional retraining, weight garment
training, functional approach training, bilateral
arm training, and task-specific training.
The Ottawa Panel guidelines are in agreement
with existing clinical practice guidelines, which
consistently made favorable recommendations for
task-specific training.70,71,74,76 The Ottawa Panel
guidelines are also in concordance with recent
systematic reviews,12,13,332,583 which drew similar
conclusions about task-oriented training. Interest-
ingly, van Peppen et al.12 suggested that task-ori-
ented training is of particular value when applied
intensively and early after stroke onset. The task-
oriented approach aims to rehabilitate stroke pa-
tients by engaging them in functional tasks that are
more meaningful to everyday life and for which
there may be greater motivation to perform.584 Re-
petitive functional or task-specific training affects
the adaptive reorganization of the central nervous
system, leading to improved functional recov-
ery.263,580,585,586 This type of training appears to
make a difference, regardless of when it has been
administered, as the brain remains plastic
throughout life. Therefore, it is beneficial to con-
tinue functional task training even years after
stroke onset, instead of limiting rehabilitation to
the few months following stroke.
Mulder and Hochstenbach587 stated that variable
but structured stimulation, such as varying the
context or the characteristics of the movements
being practiced, enables patients to adapt their
learned skills to the changing environment of ev-
eryday life. Also, improvement of motor control
after stroke can occur with meaningful rehabilita-
88 TOPICS IN STROKE REHABILITATION/SPRING 2006
tion consisting of functional tasks important to the
daily life of the patient.155,579,588 For example,
stroke patients seemed more concerned with walk-
ing quickly than with maintaining a symmetrical
gait pattern, because gait speed is a more meaning-
ful and practical outcome (for example when
crossing a street).472 Standardized exercises may be
less appealing to the patient compared to more
functional activities, which might explain why
they showed less functional improvement.579,584
Motivation is a very important element involved in
the adoption, adherence, and maintenance of
physical activity and is necessary for individuals
with chronic diseases to obtain long-term benefi-
cial effects from therapeutic exercise.579,589
Biofeedback
The Ottawa Panel recommends the use of sev-
eral forms of EMG-BFB training for the upper (UE)
and lower (LE) extremities at all stages of stroke
recovery. The available evidence provides strong
support for video feedback training for unilateral
neglect reduction (subacute stroke), general facili-
tation EMG-BFB training for the UE (subacute and
chronic stroke) and the LE (acute to chronic
stroke), and LE rhythmic positional BFB training
(chronic stroke) and LE audio and visual feedback
training with general relaxation (chronic stroke).
There is also sufficient evidence for the value of LE
force-feedback training in chronic stroke and UE
EMG-BFB relaxation training in subacute and
post-acute stroke.
The Ottawa Panel agrees with the Heart and
Stroke Foundation of Ontario guidelines,71 which
found that EMG-BFB should be included as an
intervention for post-stroke patients who have a
high level of motor return. The Ottawa Panel also
agrees with St. Joseph’s Health Care London73 that
there is strong evidence that BFB training improves
gait and standing post-stroke. However, the Ot-
tawa Panel disagrees with several other guidelines,
which state that BFB systems should not be rou-
tinely included as an intervention for stroke survi-
vors undergoing rehabilitation.72,74,75 These dis-
crepancies could be partially attributed to the fact
that only the Ottawa Panel and the Heart and
Stroke Foundation of Ontario71 based their recom-
mendations on a quantitative analysis of relevant
trials. The majority of previous post-stoke rehabili-
tation guidelines based their recommendations on
descriptive conclusions provided by the authors of
the included studies. The use of the term routine
may be a further source of disagreement. While
BFB systems are recognized as being effective, they
are not widely available in clinical and commu-
nity-based settings and are not convenient to
transport. At present, busy clinicians use other
forms of immediate feedback in their daily prac-
tice. For instance, clinicians frequently use “verbal
cues” to guide postural/alignment corrections and
facilitate the equalization of weight distribution.
Perhaps for these reasons, previous guidelines
have not supported the “routine” use of sophisti-
cated and costly EMG equipment. Still, EMG-BFB
constitutes a relevant, albeit small, part of physical
rehabilitation for the stroke patient.
The Ottawa Panel’s guidelines are consistent
with previous systematic reviews examining the
efficacy of EMG-BFB. 45–49,590 According to
Moreland and Thomson, EMG-BFB was found to
provide subjects with useful information about
their muscle activity, helping them to execute
more controlled movements using the appropriate
groups of muscle in the correct sequence.46 It has
been suggested that feedback, particularly pro-
vided through music, has a positive impact on the
patients’ motivation and enjoyment, resulting in
important clinical gains.590 This may account for
the Ottawa Panel finding that rhythmic auditory
facilitation (using a metronome or special music
tapes) was effective at improving gait speed, stride
length, and swing symmetry. A provocative hy-
pothesis that deserves further exploration is that
constant external feedback may not encourage
subjects to develop their own intrinsic feedback
mechanisms and therefore limits their learning.292
Gait training
The Ottawa Panel recommends the use of sev-
eral types of gait training methods during the post-
acute to chronic stages of stroke recovery. There is
strong evidence in favor of treadmill training (sub-
acute stroke), high-speed treadmill (chronic
stroke), treadmill training combined with

overground ambulation (chronic stroke), treadmill
training and BWS in acute and post-acute stroke,
FES of the ankle dorsiflexors (post-acute or
chronic stroke), FES of the ankle dorsiflexors and/
or gastrocnemius (chronic stroke), BFB followed
by FES (chronic stroke), rhythmic auditory facili-
tation training (post-acute stroke), and functional
lower extremity training (subacute stroke). The
available evidence also provides sufficient support
for BWS in post-acute and chronic stroke, low-
speed treadmill training in chronic stroke,
electrogoniometric feedback training in post-acute
stroke, FES of the ankle dorsiflexors in post-acute
stroke, and musical motor feedback training in
post-acute stroke. However, the Ottawa Panel does
not recommend the following: FES of the ankle
dorsiflexors for walking abilities outcome in post-
acute stroke, placebo FES in post-acute stroke for
reduction of spasticity, and treadmill training and
BWS for FIM locomotion in acute and post-acute
stroke.
The Ottawa Panel‘s recommendations are in
agreement with the Royal College of Physicians72
and the VA/DoD75 that treadmill training with par-
tial BWS should be included as an adjunct to
conventional therapy for patients undergoing
stroke rehabilitation. Of note, this recommenda-
tion is contrary to the recommendations of St.
Joseph’s Health Care London,73 which found con-
flicting evidence* on the use of BWS combined
with treadmill training and treadmill alone. The
Ottawa Panel agrees with St. Joseph’s Health Care
London73 that there is strong evidence to use FES
as well as rhythmic auditory stimulation in pa-
tients undergoing stroke rehabilitation. The Ot-
tawa Panel also agrees with the VA/DoD75 that
recommended FES for gait training following
stroke but disagrees with their finding that there is
insufficient evidence to recommend for or against
the use of multi-channel FES for severe hemiplegic
*According to the St. Joseph’s Health Care London, conflicting evi-
dence refers to a disagreement between the findings of at least two
RCTs. Where there were more than four RCTs and the results of only
one was conflicting, the conclusion was based on the results of the
majority of the studies, unless the study with conflicting results was
of higher quality.
Clinical Practice Guidelines 89
patients with gait impairment. Again, the partial
disagreement with the Ottawa Panel recommenda-
tions may be related to the fact that previous post-
stoke rehabilitation guidelines formulated consen-
sus recommendations based on the descriptive
conclusions provided by the included studies. The
quantitative analysis used by the Ottawa Panel has
the important advantage of being more precise and
being able to resolve conflicting evidence.
The Ottawa Panel’s guidelines are generally con-
sistent with the conclusions of previous reviews on
the efficacy of treadmill training combined with
BWS.12,28–33 However, Moseley29 chose to make dif-
ferent comparisons than the Ottawa Panel by pool-
ing studies together in a different way. First, when
they compared treadmill training and BWS to
other treatment, they combined studies that used
different comparators such as general physio-
therapy at two different intensities and bracing.
The primary outcome used in their review was
independence with walking, while our primary
outcome was gait speed. Furthermore, some of the
studies included in their review did not meet our
selection criteria for reasons such as insufficient
sample size or inappropriate study population
(e.g., patients with multiple infarcts). Finally,
Moseley29 only presented statistically significant
results, whereas our review considered clinical sig-
nificance as well.
High-speed treadmill training was favored over
low-speed training for improving gait cadence,
walking speed, and stride length. However, low-
speed treadmill training was favored for decreasing
the percentage of BWS required. This may be ex-
plained in part by the greater balance required at
faster speeds of walking. Intensive lower extremity
training produced clinically important increases in
walking speed and statistically significant im-
provements in walking ability, comfortable gait
speed, and maximal gait speed. Recovery of func-
tional walking depends in part on the strength of
the lower extremities.33,591 Therefore, it is not sur-
prising that an intensive leg muscle performance
training program results in improvements in gait
speed.
Gait training on a treadmill has consistent posi-
tive results according to the literature.33,580 During
a gait training program, the walking speed must be
90 TOPICS IN STROKE REHABILITATION/SPRING 2006
raised to a level that adequately challenges the
patient in order to promote the activation of
weight-bearing muscles, decrease energy cost, re-
duce dependence on walking aids, increase walk-
ing speed, and improve cardiovascular fitness.30,249
Endurance training is essential for helping stroke
patients achieve a functional walking ability and
perform ADLs.33,249 Studies have found that endur-
ance training, using a treadmill or cycling, in-
creases the cardiovascular and energy demands of
walking.33 A major disadvantage of over-ground
ambulation and treadmill training is the amount of
physical effort required by the physiotherapists to
obtain a satisfactory walking pattern.590 BWS pro-
vided through a harness is therefore very helpful at
reducing the unnecessary physical effort required
from therapists.590
It is theorized that the recovery of functional
walking ability after stroke greatly depends upon
the adaptive reorganization of cortical and subcor-
tical structures, plasticity of synapses, and forma-
tion of new circuits through sprouting in both
damaged and unaffected areas of the brain.591,592
Motor tasks increase the activity of neurons lead-
ing to increased tissue blood flow to the motor
cortex.592 The location and amount of activity in
the brain depends on the characteristics (force,
amplitude, rate, complexity) of the motor task.592
Plasticity of the central nervous system and subse-
quent improvement of gait is positively influenced
by training in stroke rehabilitation.579,592 Training
must be specific, suggesting that improvements in
gait are best achieved with direct gait training so
that the sensory stimuli are the same as those
produced during walking.33,591 Gait training facili-
tates trunk posture and hip extension by providing
sensory inputs of loading249,579 and activates gait
pattern generators in the spine and brain.30 Prac-
tice is necessary for learning.33 Gait training should
take place under varying conditions so that differ-
ent sensory inputs can be engaged (e.g., walking
on different surfaces, with obstacles, up and down
ramps, stairs).33,249,579 Recovery of walking contin-
ues up to 2 years after stroke.33
BWS with treadmill training can help stroke
patients with poor coordination, major muscle
weakness, poor trunk stabilization, and anxiety to
practice many repetitions of the full gait cycle
without compensatory movements.33,590,591 Prac-
tice of the whole task or gait cycle is very important
to improve coordination.33 BWS with treadmill
walking minimizes compensations, which occurs
with other assistive walking devices, by forcing the
use of the paretic lower extremity.249 BWS is better
than walking on a treadmill without BWS or walk-
ing within parallel bars, because it reduces com-
pensations made by the upper extremities and
non-paretic lower extremities, resulting in a more
symmetrical gait pattern.249 However, to promote
weight bearing and appropriate muscle activity, the
patient’s body weight should not be supported
more than 30%.33 The BWS provided should be
progressively decreased with training to challenge
the patient’s posture by improving activation of
weight-bearing muscles.30,249 Exercises such as
walking and strengthening in the weight-bearing
position help prevent muscle shortening and stiff-
ness.33,249 Richards et al.33 suggested that treadmill
training with partial BWS, as needed, improves gait
post-stroke. In addition, Barbeau et al.,590 in their
meta-analysis, found that BWS with treadmill walk-
ing increased gait, balance, speed of walking, lower
limb motor recovery, and walking endurance in
people with neurological conditions and produced
more benefits than conventional or gait training in
full weight bearing. Treadmill training with BWS or
no BWS was also more effective on outcomes of
interest than gait training over-ground.591 BWS with
treadmill training resulted in greater functional ben-
efits, such as increased walking speed, than over-
ground ambulation.30,249,591 Hesse and Werner
found that treadmill training with BWS versus stan-
dard and usual customary care in chronic
nonambulatory hemiparetic patients restores gait
and increases gait velocity. They state that partial
BWS with treadmill training after stroke is consid-
ered a task-oriented repetitive approach.30
Gait training in stroke patients is a difficult topic
to research for clinical and methodological rea-
sons. Although a specific treatment may appear to
be ineffective in one type of patient, similar results
cannot be assumed for other types of patients.
Researchers must ensure that when trials are con-
ducted, proper outcome measures are used. Fu-
ture investigations on the effects of the phase of
recovery, intensity of treatment, and treatment du-
ration would also be useful.
In the clinical setting, sophisticated and costly

equipment such as a BWS system and treadmill
may not be appropriate for every stroke patient.
The BWS system combined with treadmill training
requires significant financial and human re-
sources. For instance, it involves two or three reha-
bilitation staff to train a stroke patient with a BWS
system and treadmill: one physiotherapist has to
facilitate the proper trunk alignment and weight
transfer to maximize the quality of the gait pattern,
one physiotherapist needs to assist the leg swing,
and another person must operate the control
panel.
Balance training
The Ottawa Panel strongly recommends balance
training as a rehabilitation intervention for sub-
acute and post-acute stroke patients. Indeed, grade
A recommendations were found for all the follow-
ing post-stroke balance interventions: balance
training (sit-to-stand) with force platform and au-
ditory feedback in subacute and post-acute stroke,
base of support training with extrinsic auditory
feedback in subacute stroke, balance training with
visual feedback in subacute and post-acute stroke,
balance training with visual feedback in subacute
stroke, and platform training in subacute stroke.
There are, however, exceptions related to two spe-
cific long-term outcomes. The Ottawa Panel does
not recommend balance training with visual feed-
back in subacute and post-acute stroke to improve
postural sway with eyes closed or balance training
with visual feedback to improve sitting balance in
subacute stroke. The Ottawa Panel’s recommenda-
tions are in agreement with St. Joseph’s Health
Care London73 that overall there is strong evidence
that balance training post-stroke improves out-
comes.
The results of a systematic review of balance
training35 that force platform feedback (visual or
auditory) improved stance symmetry but not sway
in standing are in concordance with our recom-
mendations. However, the Ottawa Panel recom-
mendations differ from the conclusions of another
systematic review that suggested that there is in-
sufficient evidence that any one physiotherapy
treatment approach is more effective than another
in promoting the recovery of postural control36;
this review did not include recent trials, which
Clinical Practice Guidelines 91
may partly explain the different conclusions.
It is well-recognized that stroke survivors may
have sensori-motor, cognitive, and perceptual im-
pairments that affect their mobility and balance.593
Indeed, stroke is considered to be one of greatest
risk factors for falls in older adults; 23% to 73% of
community-dwelling older adults in the chronic
phase of stroke have been reported to fall over a
period of 4–6 months.594 Because of the fear of
falling, the balance scores of people who have falls
or restricted their activity after falls tend to be
reduced.595 Consequently, balance training is a key
element in post-stroke gait rehabilitation591 and is
essential in preventing falls46,596,597 and other asso-
ciated complications.598
Sensory interventions
The Ottawa Panel recommends the use of sev-
eral types of sensory training during the subacute
to chronic stages of stroke recovery. The available
evidence provides strong support for sensory func-
tion training of the hand in chronic stroke, rocking
chair stimulation in subacute stroke, and visual
attention training in post-acute stroke. There is
also sufficient support for passive vestibular stimu-
lation in the subacute and post-acute stages of
stroke and perceptual learning exercises in the
post-acute stage of stroke. The Ottawa Panel’s rec-
ommendation regarding sensory function training
of the hand is consistent with St. Joseph’s Health
Care of London’s guidelines,73 which reported
strong evidence that perceptual training interven-
tions improve perceptual functioning. Similarly,
the Ottawa Panel’s recommendation on visual at-
tention retraining is consistent with St. Joseph’s
Health Care of London’s73 assessment that there is
strong evidence to indicate that visual scanning
techniques improve visual neglect and produce
associated functional improvements.
To our knowledge, there are no other systematic
reviews published on sensory interventions for
post-stroke rehabilitation.
Constraint-induced movement therapy
According to the Ottawa Panel, there is suffi-
cient evidence to recommend the use of CIMT
during the acute, subacute, or chronic phases of
92 TOPICS IN STROKE REHABILITATION/SPRING 2006
rehabilitation for improving dexterity, motor func-
tion, and functional status in stroke patients ca-
pable of some active finger and wrist extension.
The Ottawa Panel is in agreement with several
existing guidelines that addressed CIMT,73,74,75 in-
cluding St. Joseph’s Health Care London, the Scot-
tish Intercollegiate Guidelines Network, and VA/
DoD. The Ottawa Panel also agrees with the exist-
ing reviews13,22,585 that recommended the use of
this innovative rehabilitation intervention.
CIMT was derived directly from basic research
on monkeys with somatosensory deafferenta-
tion.22,23,585 The large improvements in motor func-
tion produced by CIMT correlate with marked
alterations in brain organization.
The Ottawa Panel’s recommendations for CIMT
are based on RCTs with relatively small sample
sizes and variability in methodological quality. The
results of a large, ongoing, multicenter study (Ex-
tremity Constraint-Induced Therapy Evaluation
[EXCITE]; www.excite.org) is expected to provide
stronger evidence regarding this promising inter-
vention. Ethical considerations and patient selec-
tion factors must be taken into account before
prescribing CIMT.
Shoulder subluxation
According to the Ottawa Panel, there is strong
evidence to recommend the use of FES in subacute
stroke and the Henderson shoulder ring or Bobath
shoulder roll in post-acute stroke for the manage-
ment of shoulder subluxation. Furthermore, there is
sufficient evidence to recommend the use of FES
during the acute and chronic phases of stroke as
well as shoulder strapping during the subacute
phase of stroke. The Ottawa Panel’s recommenda-
tion of FES in the management of shoulder sublux-
ation is consistent with the guidelines of the Heart
and Stroke Foundation of Ontario71 and St. Joseph’s
Health Care.73 This recommendation is in partial
agreement with the Heart and Stroke Foundation of
Ontario71 and St. Joseph’s Health Care London73
that found limited evidence that shoulder slings
influence clinical outcomes. The Ottawa Panel,
which supports the use of shoulder strapping and
positioning of the shoulder, disagrees with St.
Joseph’s Health Care London73 that shoulder strap-
ping and prolonged positioning do not result in
significant improvements in pain, upper limb func-
tion, and range of motion. However, the consensus
opinion* of St. Joseph’s Health Care London is that
proper positioning of the hemiplegic shoulder does
indeed help prevent subluxation.
A recent review599 was conducted on the efficacy
of FES in the management of shoulder pain and
subluxation, and the Ottawa Panel recommenda-
tions fully agree with the results of this review.
However, there are some discrepancies between
the conclusions of the Ottawa Panel and those of
Price and Pandyan54 related to the use of FES in the
management of post-stroke shoulder pain. In con-
trast to the Ottawa Panel guidelines, Price and
Pandyan stated that the evidence from RCTs does
not confirm whether FES should be included or
excluded in the management of post-stroke shoul-
der pain. These differences are attributed to the
following factors: their systematic review is not up-
to-date; our review also included CCTs, which
lead to grade B recommendations; and their data
analysis included the use of standardized mean
difference (SMD), which allowed for the pooling of
data from RCTs using different scales for the same
outcome.
The overhead pulley exercises are not recom-
mended by the Ottawa Panel and others,63,71 be-
cause they do not provide adequate stabilization of
the shoulder girdle during scapular rotation and
humeral external rotation. Rather, they passively
elevate the arm in a movement of pure abduction
that can become painful for the hemiplegic shoul-
der due to soft tissue impingement between the
acromion and greater tuberosity. Passive ROM ex-
ercises provide a greater stabilization of the
scapula and allow the humerus to direct move-
ments using a combination of external rotation
and flexion, rather than abduction alone.600 Passive
ROM exercises performed on the shoulder of the
stroke survivor by a qualified rehabilitation practi-
tioner are favored over overhead pulley and skate-
*According to St. Joseph’s Health Care London, consensus opinion
refers to the agreement of a group of experts on an appropriate course
of treatment in the absence of evidence. It is regarded as the lowest
form of evidence and can be considered as no evidence at all.

board exercises. The quality of the shoulder mo-
tion can be better controlled by an experienced
therapist and thus can be beneficial to avoid un-
desired movements that could further potentiate
pain and damage of the hemiplegic shoulder.
Electrical stimulation
The Ottawa Panel recommends the use of elec-
trical stimulation as a rehabilitation intervention
for post-acute to chronic stroke patients. Clini-
cally, electrical stimulation is used to facilitate
muscle activity in situations where voluntary mo-
tor unit recruitment is insufficient. The disuse at-
rophy of fast twitch units that may be observed
following stroke is associated with such decreases
in voluntary muscle recruitment. Strong evidence
was found for positional feedback stimulation
training of the wrist extensors in subacute and
post-acute stroke, NMES of the wrist and finger
extensors in subacute and post-acute stroke, FES
of the anterior tibialis and gastrocnemius com-
bined with EMG-BFB in chronic stroke, EMG-
triggered electrical muscle stimulation training in
subacute stroke, FES of the finger thumb extensors
and flexors for the hand in subacute and post-
acute stroke, blocked practice combined with
NMES of the extensor communis digitorium/ex-
tensor carpi ulnaris/triceps brachii/anterior and
middle deltoid in chronic stroke, and electrical
stimulation of wrist and fingers extensors in
chronic stroke.
Recommendations with sufficient evidence were
also found for FES of the quadriceps and rectus
femoris combined with positional feedback stimu-
lation training in post-acute stroke, FES of the
posterior deltoid and supraspinatus for the shoul-
der in chronic stroke, NMES of chronic stroke
patients for the hand, and random practice com-
bined with NMES of the extensor communis
digitorium/extensor carpi ulnaris/triceps brachii/
anterior and middle deltoid in chronic stroke.
However, the Ottawa Panel does not recommend
FES of the quadriceps and rectus femoris com-
bined with positional feedback stimulation train-
ing for improving selective ROM in post-acute
stroke or NMES for feeding, tracking accuracy,
and hand motor function in chronic stroke.
Clinical Practice Guidelines 93
The Ottawa Panel’s recommendation of FES for
gait training in the chronic phase of stroke is in
agreement with the guidelines of both the VA/
DoD75 and St. Joseph’s Health Care London.73
However, the Panel disagrees with the finding of
the VA/DoD75 that there is insufficient evidence*
as to whether multichannel FES should or should
not be included as an intervention for severe
hemiplegic patients with gait impairment.
The Ottawa Panel agrees with the Heart and
Stroke Panel on the Management of the Hemiple-
gic Arm and Hand,71 St. Joseph’s Health Care Lon-
don,73 and the VA/DoD75 that FES should be in-
cluded in the management of shoulder
subluxation during the subacute and post-acute
stages of stroke. The Ottawa Panel recommenda-
tions are also in line with the recommendation
made by the Heart and Stroke Foundation of
Ontario71 for considering electrical stimulation of
the wrist and the forearm and EMG-BFB to reduce
motor impairment and improve functional motor
recovery. Finally, the Ottawa Panel disagrees with
the Royal College of Physicians that FES should
not be routinely considered as an intervention for
stroke patients.
FES for post-stroke patients has been exten-
sively reviewed in the literature.53,55–59,590,600 Each of
the existing meta-analyses found that FES was ef-
fective for improving muscle strength. Barbeau et
al.590 also reported improvements in walking speed
following intramuscular stimulation. FES is
thought to work by facilitating the activation of
appropriate weight-bearing muscles required dur-
ing the act of walking.
Neuromuscular stimulation is one of the thera-
peutic modalities available to enhance muscle per-
formance in patients with CVA. Stroke survivors
may not be able to voluntarily recruit motor units
to the level required for performing the high-in-
tensity exercises needed to enhance muscle func-
tion.601 The disuse atrophy of fast twitch motor
fibers observed following stroke further contrib-
*According to VA/DoD, a grading of insufficient evidence (I) indicates
that one cannot determine from the evidence the net benefit of the
intervention, either for or against, and therefore the clinician must use
his or her judgment in either including or excluding the intervention
in the treatment of the patient.
94 TOPICS IN STROKE REHABILITATION/SPRING 2006
utes to muscle weakness. Clinically, neuromuscu-
lar stimulation may be used in these situations.
Future RCTs should aim to clarify the relation-
ship between specific patient characteristics and dif-
ferent stimulation protocols.602,603 Age, gender,
phase of stroke, extent of disuse atrophy, device
characteristics, electrical stimulation protocol,
waveform, amplitude and duration of stimulation
trains, total number of cycles delivered, modulation
of frequency, and concurrent treatments are impor-
tant variables that require further investigation.
TENS
The Ottawa Panel recommends the use of several
types of TENS applications during the post-acute to
chronic stages of stroke recovery. There is strong
evidence to support the use of low-intensity TENS
applied to the hand, foot, and lower extremity in
chronic stroke; low-intensity TENS applied to the
elbow and ankle in post-acute stroke; and high-
intensity TENS applied to the shoulder in post-
acute stroke. However, the Ottawa Panel does not
recommend low-intensity TENS applied to the up-
per extremity of stroke patients in the chronic phase
of recovery to reduce spasticity in the wrist flexors.
The Ottawa Panel recommendations are in line
with the Royal College of Physicians’72 recommen-
dation that high-intensity TENS should be in-
cluded as an intervention for patients undergoing
stroke rehabilitation, if nonsteroidal anti-inflam-
matory analgesia has no significant effect. How-
ever, the Ottawa Panel disagrees with the recom-
mendation of the Royal College of Physicians72 that
TENS should only be used as a routine interven-
tion for improving muscle control for patients un-
dergoing stroke rehabilitation in the context of
ongoing trials. Finally, the Ottawa Panel agrees
with the St. Joseph’s Health Care London’s73 find-
ing that there is conflicting evidence* that TENS
improves a variety of outcomes, including motor
recovery, spasticity, and ADL.
*According to the St. Joseph’s Health Care London, conflicting evi-
dence refers to a disagreement between the findings of at least two
RCTs. Where there were more than four RCTs and the results of only
one was conflicting, the conclusion was based on the results of the
majority of the studies, unless the study with conflicting results was of
higher quality.
The various application modes of TENS repre-
sent effective therapeutic options in post-stroke
rehabilitation depending on the practitioners’
treatment goals and target outcomes. Indeed, on
one hand, low-intensity TENS seems to be more
effective for decreasing spasticity, improving sen-
sation, and increasing functional status at different
stages of stroke recovery. On the other hand, high-
intensity TENS seems to be more effective for in-
creasing shoulder passive ROM in the post-acute
stage of stroke recovery, which may be indirectly
related to pain relief. TENS, depending upon its
intensity and frequency, stimulates nerves and
muscle fibers and may improve muscle tone and
strength, reduce shoulder stiffness, and dampen
the perception of pain.60
The neuroregulatory peripheral and central pain
relieving effects604–607 of TENS have been proposed
to be more effective at higher intensity applica-
tions.82,608–610 This effect was captured in the study
involving the application of higher intensity, acu-
puncture-like TENS to the shoulder compared
with placebo. However, both conventional and
acupuncture-like TENS excite afferent fibers in the
A-alpha-beta range.611 Their pain-relieving effects
are thought to be mediated by the activation of
intrinsic pain-suppressive systems612,613 and the
concomitant release of opiate observed in both
animals614 and humans.605 That the stimulation
parameters in TENS analgesia are important has
been shown in both animal studies and human
research. Changes in frequency recruit different
opioid receptors, for example, and hence an
awareness of the parameters used during TENS
treatments is essential.615,616 Several investigators
have recommended that vibrator stimulation be
part of TENS application, especially when TENS is
being applied for relief of chronic pain.617–619
To establish clear evidence for the use of TENS
in stroke rehabilitation, greater clarity is needed
about the optimal treatment duration, stimulation
parameters, and application site based on indi-
vidual patient characteristics.602,603
Therapeutic ultrasound
The Ottawa Panel recommends considering the
application of therapeutic ultrasound to the shoul-

der in the chronic phase of stroke. However, the
Ottawa Panel agrees with St. Joseph’s Health Care
London73 that adding ultrasound therapy to ROM
exercises does not change functional outcomes.
According to clinical practice guidelines on
rheumatoid arthritis developed by the Ottawa
Panel, therapeutic ultrasound has the potential to
reduce shoulder pain and increase ROM in rheu-
matoid arthritis.82 The use of continuous ultra-
sound is supported by its documented physiologi-
cal effects.602,620,621 The mechanical effect of either
pulsed or continuous ultrasound leads to an in-
crease in skin permeability and facilitates the soft-
tissue healing process. Furthermore, both pulsed
and continuous ultrasound are known to reduce
nerve conduction velocity in nociceptive fibers.
Additional thermal effects produced by continu-
ous ultrasound help reduce muscle spasm and
pain. The vasodilatory response associated with
these thermal effects also enhances the excretion of
chronic inflammatory cells.
To be able to adequately compare and repro-
duce the results of published RCTs, it is crucial
that the following details be clearly described:
characteristics of the device (such as size of the
head and calibration), characteristics of the thera-
peutic application (including pulsed or continu-
ous ultrasound, frequency, intensity, pulse ratio,
area treated, and schedule of treatment) and char-
acteristics of the study population (age, gender,
acute or chronic stage of disease).602,603
Acupuncture
The Ottawa Panel recommends the use of acu-
puncture as an adjunct to improve specific out-
comes during the acute and subacute stages of
stroke recovery. The available evidence provides
support for surface acupuncture with electrical
stimulation in subacute stroke and needle acupunc-
ture with electrical stimulation in acute and sub-
acute stroke for improving walking mobility, ROM,
and balance. The evidence for the effects of acu-
puncture on spasticity, motor function, and func-
tional status are however conflicting. One study
reported no benefit of needle acupuncture without
stimulation for ankle spasticity in chronic stroke
patients with spastic equinovarus deformity.403
However, another study revealed statistically signifi-
Clinical Practice Guidelines 95
cant and clinically important benefits of
electroacupuncture on elbow spasticity compared
to standard acupuncture in post-acute stroke.407
With regard to motor function, surface acupuncture
with electrical stimulation was not effective, but
needle acupuncture with manual and electrical
stimulation showed positive results. The evidence
from three RCTs that assessed functional status was
also inconsistent, with two RCTs suggesting no ef-
fect and the other RCT suggesting statistically sig-
nificant and clinically important benefits.
The Ottawa Panel recommendations are gener-
ally consistent with the recommendations of St.
Joseph’s Health Care London,73 which found that
“there is conflicting evidence that acupuncture is
effective for improving stroke outcomes.” How-
ever, the Ottawa Panel wishes to emphasize that
the available evidence is not conflicting with re-
spect to all stroke outcomes. The Royal College of
Physicians72 concluded that the evidence was in-
sufficient to make definitive statements about the
role of acupuncture. They expressed a cautious
statement based on consensus opinion that “acu-
puncture should only be used as an intervention
for patients undergoing post-stroke rehabilitation
in the context of ongoing trials.”
Existing meta-analyses64–67 have all concluded
that acupuncture represents a safe and promising
adjuvant therapeutic modality for post-stroke pain
management, but they suggest that high-quality
RCTs need to be conducted before they recom-
mend its inclusion in stroke rehabilitation pro-
grams. Generally, methodological issues such as
randomization method, quality of double-blind-
ing, sample size, study duration, and selection of
outcome have limited the strength and reliability
of results reported in the literature. Clearly, the
treatment application protocol differs between
studies that used acupuncture and studies that
used electroacupuncture. This fundamental differ-
ence exemplifies the lack of standardization in
application methods and contributes to the diffi-
culty of pooling data.622
Intensity and organization of rehabilitation
Stroke rehabilitation in the acute care setting
should be provided by an organized and
multidisciplinary team that specializes in stroke
96 TOPICS IN STROKE REHABILITATION/SPRING 2006
management. The Ottawa Panel guidelines pro-
vide strong evidence in favor of extended stroke
unit service with early supported discharge during
the acute phase of stroke as well as specialized
stroke unit rehabilitation at all other phases of
recovery. Physiotherapy at high-intensity, en-
hanced physiotherapy, enhanced assistant physio-
therapist physiotherapy, enhanced upper-extrem-
ity treatment, and enhanced occupational therapy
have all been shown to be effective interventions
during the subacute phase of stroke. The available
evidence also supports the use of home-based ex-
ercise training in post-acute stroke, home therapy
(physician) in chronic stroke, home intervention
(rehabilitation and nursing services) in subacute
stroke, early care and early physiotherapy inter-
vention in acute stroke, and intensive outpatient
physiotherapy rehabilitation program in chronic
stroke. Several innovative post-stroke rehabilita-
tion interventions were not recommended by the
Ottawa Panel because their effects did not appear
to be sustained over time.
The Ottawa Panel’s recommendations are in con-
cordance with the VA/DoD,75 which found good
evidence that early rehabilitation therapy should be
provided as soon as the patient’s medical status is
stable and insufficient evidence as to whether inpa-
tient stroke rehabilitation care is superior to outpa-
tient stroke rehabilitation. Like St. Joseph’s Health
Care London,73 the Ottawa Panel found that struc-
tured, stroke unit care leads to better functional
outcomes than care in a general ward. The Ottawa
Panel is also in agreement with the Royal College of
Physicians,72 which found good evidence that acute
hospital care for stroke patients should be delivered
in a ward or stroke unit by specialists with expertise
in stroke management. The Ottawa Panel recom-
mendation for considering early supported dis-
charge with home rehabilitation in patients with acute
stroke is in concordance with the European Stroke
Initiative.78 The conclusions drawn by other previously
published systematic reviews on the organization of
stroke rehabilitation services are also consistent with
the Ottawa Panel guidelines, despite some differences
in methodology and study selection criteria.37–43
With regard to rehabilitation intensity, the Ot-
tawa Panel agrees with St. Joseph’s Health Care
London73 that there is good evidence to show that
intensive rehabilitation therapies delivered in short
periods of time lead to faster recovery and earlier
hospital discharge. Several existing reviews24–27,623
also suggest that rehabilitation programs of greater
intensity have better effects on functional outcomes.
According to Wall,472 the successful acquisition of
new motor skills by healthy subjects requires much
more than 2 hours of practice per week. Compared
to healthy individuals, stroke patients may require
an even higher level of exercise intensity to reach
new levels of motor performance.
Unfortunately, there is no clear evidence in the
current literature about the precise amount of
therapy per day necessary to optimize the efficacy of
therapeutic exercises. In clinical practice, approxi-
mately 2 hours of therapy per day (occupational
therapy, 40.8 min/day; and physical therapy, 54.8
min/day) is usually offered.555
The Ottawa Panel found some support for home-
based exercise training and home therapy (physi-
cian), which is in disagreement with St. Joseph’s
Health Care London73 that found good evidence
that additional home-based therapy does not im-
prove the overall functional outcome scores for pa-
tients undergoing post-stroke rehabilitation. The
Ottawa Panel recommends considering the use of
enhanced occupational therapy in subacute stroke
patients, although the Scottish Intercollegiate
Guidelines Network74 found poor evidence to sup-
port occupational therapy treatment.
Conclusion
Despite certain methodological limitations in
the primary studies underlying these physical re-
habilitation EBCPGs, the Ottawa Panel was able to
analyze and include an impressive number of ac-
ceptable, well-designed trials that satisfied the ini-
tial selection criteria. A large number of positive
recommendations were formulated to help practi-
tioners make decisions concerning the use of evi-
dence-based rehabilitation interventions in their
practice. The recommendations of the Ottawa
Panel cannot however replace clinical judgment,
which is critical for applying the available evidence
appropriately to the care of individual patients
under specific circumstances. A careful consider-
ation of patient values is always essential.62

The publication of these Ottawa Panel EBCPGs
will help lead to their dissemination and implemen-
tation, while directing the development of future
RCTs in areas of rehabilitation practice where poor
or insufficient scientific evidence currently exists.
The ultimate goal of the Ottawa Panel EBCPGs is
to enable stroke patients to receive current and
optimal care so that they can achieve their highest
possible level of recovery. A related objective is to
identify ineffective treatments that lead to delays in
improvement and unnecessary cost to the health
care system.
Acknowledgments
This project has been financially supported by
the Heart and Stroke Foundation of Ontario, the
REFERENCES
Introduction
1. Petrasovits A, Nair C. Epidemiology of stroke in
Canada. Health Reports. 1994;6:39–44.
2. Anderson CS, Linto J, Stewart-Wynne EG. A popu-
lation-based assessment of the impact and burden
of caregiving for long-term stroke survivors.
Stroke. 1995;26:843–849.
3. Wolf PA, D’Agostino RB, O’Neal A, Sytkowski P,
Case CD, Belanger AJ, Kannal WB. Secular trends
in stroke incidence and mortality: the
Framingham Study. Stroke. 1992;23:1551–1555.
4. Wade DT, Lewer RL, Skilbeck CE, David RM.
Stroke: A Critical Approach to Diagnosis, Treatment
and Management. London: Chapman and Hall;
1992.
5. Woolf SH. Practice guidelines: a new reality in
medicine I. Recent developments. Arch Intern Med.
1990;150:1811–1818.
6. Grimshaw JM, Freemantle N, Wallace S, et al.
Developing and implementing clinical practice
guidelines. Quality Health Care. 1995;4:55–64.
7. Saunders DH, Greig CA, Young A, Mead GE. Physi-
cal fitness training for stroke patients. Cochrane
Database Syst Rev. 2004;(1):CD00316.
8. Rimmer JH, Wang E. Aerobic exercise training in
stroke survivors. Top Stroke Rehabil. 2005;12(1):
17–30.
9. Gordon NF, Gulanick M, Costa F, Fletcher G,
Franklin BA, Roth EJ, Shephard T. American Heart
Association Council on Clinical Cardiology, Sub-
committee on Exercise, Cardiac Rehabilitation,
and Prevention; the Council on Cardiovascular
Nursing; the Council on Nutrition, Physical Activ-
Clinical Practice Guidelines 97
Ontario Ministry of Health and Long-Term Care
(career scientist salary support), University of Ot-
tawa for the University Research Chair (research
staff salary support), Faculty of Health Sciences,
University of Ottawa (conference financial sup-
port), and the Ministry of Human Resources, Sum-
mer Students Program (salary support).
The Ottawa EBCPGs Development Group is in-
debted to Dr. Antoine Hakim, Dr. Peter Tugwell,
Ms. Patsy McNamara, Ms. Marian Thorton, Ms.
Tracey Dyks, and Mrs. Michelle Morin for their
useful comments. The Ottawa EBCPGs Develop-
ment Group is also thankful to Ms. Isabelle Viau,
Ms. Judith Lambert, Ms. Karine Maltais, Ms. Maria
Delgadillo, Ms. Isabelle Mayer, and Magali
Scheubel for their technical support and their help
in extraction of data.
ity, and Metabolism; and the Stroke Council.
Physical activity and exercise recommendations
for stroke survivors: an American Heart Association
scientific statement from the Council on Clinical
Cardiology, Subcommittee on Exercise, Cardiac
Rehabilitation, and Prevention; the Council on
Cardiovascular Nursing; the Council on Nutrition,
Physical Activity, and Metabolism; and the Stroke
Council. Stroke. 2004;35(5):1230–1240.
10. Shepherd RB. Exercise and training to optimize
functional motor performance in stroke: driving
neural reorganization? Neural Plasticity. 2001;8(1-
2):121–129.
11. van der Lee JH, Snels IA, Beckerman H, Lankhorst
GJ, Wagenaar RC, Bouter LM. Exercise therapy for
arm function in stroke patients: a systematic re-
view of randomized controlled trials. Clin Rehabil.
2001;15(1):20–31.
12. van Peppen RP, Kwakkel G, Wood-Dauphinee S,
Hendriks HJ, Van der Wees PJ, Dekker J. The impact
of physical therapy on functional outcomes after
stroke: what’s the evidence? Clin Rehabil.
2004;18(8):833–862.
13. Stein J. Motor recovery strategies after stroke. Top
Stroke Rehabil. 2004;11(2):12–22.
14. Brosseau L, Sarnecka J, Démosthènes P, et al. Effi-
cacy of post-stroke therapeutic exercises: a meta-
analysis. Topics in Stroke Rehabil. Manuscript sub-
mitted for publication.
15. Macko RF, DeSouza CA, Tretter LD, et al. Treadmill
aerobic exercise training reduces the energy ex-
penditure and cardiovascular demands of
hemiparetic gait in chronic stroke patients. A pre-
liminary report. Stroke. 1997;28:326–330.
98 TOPICS IN STROKE REHABILITATION/SPRING 2006
16. Morris ME, Matyas TA, Bach TM, Goldie PA.
Electrogoniometric feedback: its effect on genu
recurvatum in stroke. Arch Phys Med Rehabil.
1992;73:1147–1154.
17. Riolo L, Fisher K. Is there evidence that strength
training could help improve muscle function and
other outcomes without reinforcing abnormal
movement patterns or increasing reflex activity in
a man who has had a stroke? Phys Ther.
2003;83(9):844–851.
18. Dobkin BH. Strategies for stroke rehabilitation.
Lancet Neurol. 2004;3(9):528–536.
19. Volpe BT, Krebs HI, Hogan N. Robot-aided sen-
sorimotor training in stroke rehabilitation. Adv
Neurol. 2003;92:429–433.
20. Taub E, Uswatte G. Constraint-induced move-
ment therapy: bridging from the primate labora-
tory to the stroke rehabilitation laboratory. J
Rehabil Med. 2003;(suppl 41):34–40.
21. Taub E, Crago JE, Uswatte G. Constraint-induced
movement therapy: a new approach to treatment
in physical rehabilitation. Rehabil Psychol.
1998;43(2):152–170.
22. Taub E, Morris DM. Constraint-induced move-
ment therapy to enhance recovery after stroke.
Curr Atherosclerosis Rep. 2001;3(4):279–286.
23. Morris DM, Crago JE, DeLuca SC, Pidikiti RD, Taub
E. Constraint-induced movement therapy for mo-
tor recovery. NeuroRehabilitation. 1997;9:29–43.
24. Kwakkel G, van Peppen R, Wagenaar RC, et al.
Effects of augmented exercise therapy time after
stroke: a meta-analysis. Stroke. 2004;35(11):
2529–2539.
25. Kwakkel G, Wagenaar RC, Koelman TW, Lankhorst
GJ, Koetsier JC. Effects of intensity of rehabilitation
after stroke. Stroke. 1997;28:1550–1556.
26. Langhorne P, Wagenaar R, Partridge C. Physio-
therapy after stroke: more is better? Physiother Res
Int. 1996;1(2):75–87.
27. Weir RP. Rehabilitation of cerebrovascular disorder
(stroke): early discharge and support: a critical
appraisal of the literature. NZ Health Technol As-
sess. 1999;2:1–53.
28. Malouin F, Richards CL, McFadyen B, Doyon J.
[New perspectives of locomotor rehabilitation af-
ter stroke]. M S-Medecine Sciences. 2003;19(10):
994–998.
29. Moseley AM, Stark A, Cameron ID, Pollock A.
Treadmill training and body weight support for
walking after stroke [review]. Cochrane Database
Syst Rev. 2003;(3):CD002840.
30. Hesse S, Werner C. Partial body weight supported
treadmill training for gait recovery following
stroke. Adv Neurol. 2003;92:423–428.
31. Hesse S, Werner C, von Frankenberg S, Bardeleben
A. Treadmill training with partial body weight sup-
port after stroke. Phys Med Rehabil Clin North Am.
2003;14(suppl 1):S111–S123.
32. Hesse S, Werner C, Bardeleben A, Barbeau H. Body
weight-supported treadmill training after stroke.
Curr Atherosclerosis Rep. 2001;3(4):287–294.
33. Richards CL, Malouin F, Dean C. Gait in stroke:
assessment and rehabilitation. Clin Geriatr Med.
1999;15(4):833–855.
34. Hesse S. Recovery of gait and other motor func-
tions after stroke: novel physical and pharmaco-
logical treatment strategies. Restorative Neurol
Neurosci. 2004;22(3-5):359–369.
35. Barclay-Goddard R, Stevenson T, Poluha W,
Moffatt MEK, Taback SP. Force platform feedback
for standing balance training after stroke [review].
Cochrane Database Syst Rev. 2004;(4):CD004129.
36. Pollock A, Baer G, Pomeroy V, Langhorne P. Phys-
iotherapy treatment approaches for the recovery
of postural control and lower limb function follow-
ing stroke [review]. Cochrane Database Syst Rev.
2003;(2):CD001920. DOI: 10.1002/14651858.
37. Kwan J, Sandercock P. In-hospital care pathways
for stroke [review]. Cochrane Database Syst Rev.
2004;(4):CD002924. DOI: 10.1002/14651858.
38. Early Supported Discharge Trialists. Services for
reducing duration of hospital care for acute stroke
patients [review]. Cochrane Database Syst Rev.
2005;(2):CD000443.
39. Outpatient Service Trialists. Therapy-based reha-
bilitation services for stroke patients at home [re-
view]. Cochrane Database Syst Rev. 2003;(1):
CD002925. DOI: 10.1002/14651858.
40. Stroke Unit Trialist’s Collaboration. Organized in-
patient (stroke unit) care for stroke [review].
Cochrane Database Syst Rev. 2001;(3):CD000197.
DOI: 10.1002/14651858.
41. Langhorne P, Cadilhac D, Feigin V, Grieve R, Liu
M. How should stroke services be organised? Lan-
cet Neurol. 2002;1(1):62–68.
42. Brady BK, McGahan L, Skidmore B. Systematic
review of economic evidence on stroke rehabilita-
tion services. Int J Technol. 2005;21(1):15–21.
43. Legg L, Langhorne P, Outpatient Service Trialists.
Rehabilitation therapy services for stroke patients
living at home: systematic review of randomised
trials. Lancet. 2004;363(9406):352–356.
44. Treib J, Grauer MT, Woessner R, Morgenthaler M.
Treatment of stroke on an intensive stroke unit: a
novel concept. Intens Care Med. 2000;26(11):
1598–1611.
45. Schleenbaker RE, Mainous AG. Electromyographic
biofeedback for neuromuscular reeducation in the
hemilegic stroke patients: a meta-analysis. Arch
Phys Med Rehabil. 1993;74:1301–1304.
46. Moreland J, Thomson MA. Efficacy of electromyo-
graphic biofeedback compared with conventional
physical therapy for upper-extremity function in
patients. Phys Ther. 1994;74(6):534–547.
47. Glanz M, Klawansky S, Stason W, Berker C, Shah
N, Phan H, Chalmers TC. Biofeedback therapy in
poststroke rehabilitation: a meta-analysis of the
randomized controlled trials. Arch Phys Med
Rehabil. 1995;76:508–515.
48. Woldag H, Hummelsheim H. Evidence-based
physiotherapeutic concepts for improving arm
and hand function in stroke patients: a review. J
Neurol. 2002;249(5):518–528.
49. Moreland J, Thomson MA. Electromyographic bio-
feedback to improve lower extremity function af-
ter stroke: a meta-analysis. Arch Phys Med Rehabil.

1988;79:134–140.
50. Bradley L, Hart BB, Mandana S, Flowers K, Riches
M, Sanderson P. Electromyographic biofeedback
for gait retraining after stroke. Clin Rehabil.
1998;12:11–22.
51. Cozean CD, Pease WS, Hubbel SL. Biofeedback
and functional electric stimulation in stroke reha-
bilitation. Arch Phys Med Rehabil. 1988;69:401–
405.
52. Francisco G, Chae J, Chawla H, Kirshblum S,
Zorowitz R, Lewis G, Pang S. Electromyogram-
triggered neuromuscular stimulation for improv-
ing the arm function of acute stroke survivors: a
randomized pilot study. Arch Phys Med
Rehabil. 1988;79:570–575.
53. Glanz M, Klawansky S, Stason W, Berkey C,
Chalmers TC. Functional electrostimulation in
poststroke rehabilitation: a meta-analysis of the
randomized controlled trials. Arch Phys Med
Rehabil. 1996;77:549–553.
54. Price CIM, Pandyan AD. Electrical stimulation for
preventing and treating post-stroke shoulder pain
[review]. Cochrane Database Syst Rev.
2000;(4):CD001698. DOI: 10.1002/14651858.
55. Pomeroy VM, Pollock A. Electrostimulation for
promoting recovery of movement or functional
ability after stroke [protocol]. Cochrane Database
Syst Rev. 2001;(3):CD003241. DOI: 10.1002/
14651858.
56. Aoyagi Y, Tsubahara A. Therapeutic orthosis and
electrical stimulation for upper extremity hemiple-
gia after stroke: a review of effectiveness based on
evidence. Top Stroke Rehabil. 2004;11(3):9–15.
57. Kottink AI, Oostendorp LJ, Buurke JH, Nene AV,
Hermens HJ, IJzerman MJ. The orthotic effect of
functional electrical stimulation on the improve-
ment of walking in stroke patients with a dropped
foot: a systematic review. Artificial Organs.
2004;28(6):577–586.
58. Wilder RP, Wind TC, Jones EV, Crider BE, Edlich RF.
Functional electrical stimulation for a dropped
foot. J Long-Term Effects Med Implants.
2002;12(3):149–159.
59. Granat MH. Functional electrical stimulation and
rehabilitation. Curr Opin Orthopedics. 1994;5(VI):
90–95.
60. Binder-Macleod SA, Lee SCK. Assessment of the
efficacy of functional electrical stimulation in pa-
tients with hemiplegia. Top Stroke Rehabil.
1997;3(4):88–89.
61. Ada L, Foongchomcheay A. Efficacy of electrical
stimulation in preventing or reducing subluxation
of the shoulder after stroke: a meta-analysis. Aust J
Physiother. 2002;48(4):257–267.
62. Goulding R, Thompson D, Beech C. Caring for
patients with hemiplegia in an arm following a
stroke. Br J Nursing. 2004;13(9):534–539.
63. Turner-Stokes L, Jackson D. Shoulder pain after
stroke: a review of the evidence base to inform the
development of an integrated care pathway. Clin
Rehabil. 2002;16(3):276–298.
64. Ernst E, White AR. Acupuncture as an adjuvant
therapy in stroke rehabilitation? Wiener
Clinical Practice Guidelines 99
Medizinische Wochenschrift. 1996;146(21-
22):556–558.
65. Zhang SH, Liu M, Asplund K, Li L. Acupuncture for
acute stroke [review]. Cochrane Database Syst Rev.
2005;(2):CD00317.
66. Tang JL, Wu HM, Lin XP, Zhang YL, Lau J, Leung
PC, Woo J, Li YP. Acupuncture for stroke rehabilita-
tion [protocol]. Cochrane Database Syst Rev.
2003;(2):CD004131. DOI: 10.1002/14651858.
67. Park J, Hopwood V, White AR, Ernst E. Effectiveness
of acupuncture for stroke: a systematic review. J
Neurol. 2001;248(7):558–563.
68. Leandri M, Parodi CI, Corrieri N, Rigardo S. Com-
parison of TENS treatments in hemiplegic shoul-
der pain. Scand J Rehabil Med. 1990;22(2):69–71.
69. Sonde L, Kalimo H, Fernaeus SE, Viitanen M. Low
TENS treatment on post-stroke paretic arm: a
three-year follow-up. Clin Rehabil. 2000;14(1):14–
19.
70. Agency for Health Care Policy and Research. Clini-
cal Practice Guideline Number 16: Post-Stroke Reha-
bilitation. Rockville, MD: US Department of Health
and Human Services; 1995. AHCPR Publication
No. 95-0662.
71. Heart and Stroke Foundation of Ontario. Manage-
ment of the Post Stroke Arm and Hand Treatment
Recommendations of 2001 Consensus Panel.
Hamilton, Ontario, Canada: Consensus Panel on
the Management of the Hemiplegic Arm and
Hand; 2002.
72. Royal College of Physicians. National Clinical
Guidelines for Stroke. London, England: 2004.
Available at: http://www.rcplondon.ac.uk
73. St. Joseph’s Health Care London. Evidence-Based
Review of Stroke Rehabilitation. London, Ontario,
Canada: Ministry of Health and Long-Term Care
and the Heart and Stroke Foundation of Ontario;
2004.
74. Scottish Intercollegiate Guidelines Network. Man-
agement of Patients with Stroke: Rehabilitation, Pre-
vention and Management of Complications, and
Discharge Planning-A National Clinical Guideline.
Edinburgh, Scotland: SIGN, Royal College of Phy-
sicians; 2002.
75. Department of Veterans Affairs/Department of
Defense. Clinical Practice Guideline for the Manage-
ment of Stroke Rehabilitation in the Primary Care
Setting. Washington, DC: Management of Stroke
Rehabilitation Working Group; 2003.
76. SPREAD: Stroke Prevention and Educational
Awareness Diffusion Collaboration. The Italian
Guidelines for Stroke Prevention. Neurol Sci.
2000;21:5–12.
77. American Heart Association. Physical activity and
exercise recommendations for stroke survivors.
Stroke. 2004;35:1229–1239.
78. European Stroke Initiative. Recommendations for
stroke management. Cerebrovasc Dis. 2003;
16:311–337.
79. Trombly C. Clinical practice guidelines for post-
stroke rehabilitation and occupational therapy
practice. Am J Occup Ther. 1995;49:711–714.
80. The Philadelphia Panel. Evidence-based clinical
100 TOPICS IN STROKE REHABILITATION/SPRING 2006
practice guidelines on musculoskeletal rehabilita-
tion interventions: overview and methodology.
Phys Ther. 2001;81(10):1629–1640.
81. The Ottawa Panel (Ottawa Methods Group:
Brosseau L, Wells G, Tugwell P, Egan M, Dubouloz
CJ, Casimiro L, Robinson V, Pelland L, McGowan J,
and Lamb M). Ottawa Panel evidence clinical
practice guidelines for therapeutic exercises and
manual therapy in the treatment of rheumatoid
arthritis. Phys Ther. 2004;84:934–981.
82. The Ottawa Panel (Ottawa Methods Group:
Brosseau L, Wells G, Tugwell P, Egan M, Dubouloz
CJ, Casimiro L, Robinson V, Pelland L, McGowan J,
and Lamb M). Ottawa Panel evidence clinical
practice guidelines for electrotherapy and thermo-
therapy interventions in the treatment of rheuma-
toid arthritis. Phys Ther. 2004;84:1016–1043.
83. Clark HD, Wells GA, Huet C, et al. Assessing the
quality of randomized trials: reliability of the Jadad
scale. Control Clin Trials. 1999;20:448–452.
84. Cluzeau F, Littlejohns P. Appraising clinical prac-
tice guidelines in England and Wales: the develop-
ment of a methodologic framework and its appli-
cation to policy. Joint Comm J Qual Improv.
1999;25:514–521. Available at: http://
sghms.ac.uk/phs/hceu/
85. Davis DA, Taylor-Vaisey A. Translating guidelines
into practice. A systematic review of theoretic con-
cepts, practical experience and research evidence
in the adoption of clinical practice guidelines. Can
Med Assoc J. 1997;157:408–416.
86. Browman GP, Levine MN, Mohide EA, et al. The
practice guidelines development cycle: a concep-
tual tool for practice guidelines development and
implementation. J Clin Oncol. 1995;13:502–512.
87. Levine MN, Browman G, Newman T, et al. The
Ontario cancer treatment practice guidelines ini-
tiative. Oncology Suppl. 1996;10:19–22.
88. Rogers EM. Lessons for guidelines form the diffu-
sion of innovations. Joint Comm J Qual Improv.
1995;21:324–328.
89. Woolf SH, Grol R, Hutchinson A, Eccles M,
Grimshaw J. Potential benefits, limitations and
harms of clinical guidelines. BMJ. 1999;318:527–
530.
Method
90. The Philadelphia Panel (Ottawa Methods Group:
Wells G, Tugwell P, Brosseau L, et al.). Philadelphia
Panel evidence-based clinical practice guidelines
on selected rehabilitation interventions: overview
and methodology. Phys Ther. 2001;81:1629–
1640.
91. Canadian Stroke Network. Stroke Expert Confer-
ence To Develop Best Practice Recommendations for
Upper and Lower Extremity Rehabilitation: Work-
book. Toronto, Ontario, Canada: Score Stroke Ex-
pert Conference; January 2004:21–24.
92. Dickersin K, Scherer R, Lefebvre C. Identifying rel-
evant studies for systematic reviews. BMG.
1994;309(6964):1286–1291.
93. Haynes R, Wilczynski N, McKibbon KA, Walker CJ.
Developing optimal search strategies for detecting
clinically sound studies in MEDLINE. J Am Med
Inform Assoc. 1994;1:447–458.
94. Jadad AR, Moore RA, Carroll D, et al. Assessing the
quality of reports of randomized trials: is blinding
necessary? Control Clin Trials. 1996;17:1–12.
95. Harrison M. Physiotherapy in Stroke Management.
Edinburgh: Churchill Livingstone; 1995:337.
96. World Health Organization. ICIDH-2: International
Classification of Functioning and Disability. Geneva:
World Health Organization; 2001.
97. Mulrow CD, Oxman A, eds. How to Conduct a
Cochrane Systematic Review [San Antonio
Cochrane Collaboration, version 3.0.2]. Oxford,
United Kingdom: The Cochrane Library, The
Cochrane Collaboration; 1997.
98. The Cochrane Reviewers’ Handbook glossary.
Available at: http://www.cochrane.dk/cochrane/
hanbook/hbookCOCHRANE_REVIEWERS_
HANBOOK_GLOS.htm. Accessed June 4, 2003.
Results
Therapeutic exercises
99. Aisen ML, Krebs HI, Hogan N, McDowell F, Volpe
BT. The effect of robot-assisted therapy and reha-
bilitative training on motor recovery following
stroke. Arch Neurol. 1997;54:443–446.
100. Carr M, Jones J. Physiological effects of exercise on
stroke survivors. Top Stroke Rehabil. 2003;9(4):57–
64.
101. Chu KS, Eng JJ, Dawson AS, Harris JE, Ozkaplan A,
Gylfadottir S. Water-based exercise for cardiovas-
cular fitness in people with chronic stroke: a ran-
domized controlled trial. Arch Phys Med Rehabil.
2004;85:870–874.
102. Dickstein R, Hocherman S, Pillar T, Shaham R.
Stroke rehabilitation: three exercise therapy ap-
proaches. Phys Ther. 1986;66:1233–1238.
103. Duncan P, Richards L, Wallace D, et al. A random-
ized, controlled pilot study of a home-based exer-
cise program for individuals with mild and moder-
ate stroke. Stroke. 1998;29:2055–2060.
104. Duncan P, Studenski S, Richards L, et al. Random-
ized clinical trial of therapeutic exercise in sub-
acute stroke. Stroke. 2003;34:2173–2180.
105. Fasoli SE, Krebs HI, Stein J, Frontera WR, Hogan N.
Effects of robotic therapy on motor impairment
and recovery in chronic stroke. Arch Phys Med
Rehabil. 2003;84:477–482.
106. Glasser L. Effects of isokinetic training on the rate
of movement during ambulation in hemiparetic
patients. Phys Ther. 1986;66(5):673–676.
107. Inaba M, Edberg E, Montgomery J, Gillis MK. Ef-
fectiveness of functional training active exercise,
and resistive exercise for patients with hemiplegia.
Phys Ther. 1973;53(1):28–35.
108. Katz-Leurer M, Carmeli E, Shochina M. The effect of
early aerobic training on independence six months
post stroke. Clin Rehabil. 2003;17:735–741.

109. Katz-Leurer M, Shochina M, Carmeli E, Friedlander
Y. The influence of early aerobic training on the
functional capacity in patients with cerebrovascu-
lar accident at the subacute stage. Arch Phys Med
Rehabil.. 2003;84:1609–1614.
110. Kim CM, Eng JJ, MacIntyre DL, Dawson AS. Effects
of isokinetic strength training on walking in per-
sons with stroke. A double-blind controlled pilot
study. J Stroke Cerebrovascular Dis. 2001;10(6):
265–273.
111. Kumar R, Metter J, Mehta AJ, Chew T. Shoulder
pain in hemiplegia: the role of exercise. Am J Phys
Med Rehabil. 1990;69(4):205–208.
112. Langhammer B, Stanghelle JK. Bobath or motor
relearning programme? A comparison of two dif-
ferent approaches of physiotherapy in stroke reha-
bilitation: a randomized controlled study. Clin
Rehabil. 2000;14:361–369.
113. Langhammer B, Stanghelle JK. Bobath or motor
relearning programme? A follow-up one and four
years post stroke. Clin Rehabil. 2003;17:731–734.
114. Liu KP, Chan C, Lee T, Hui-Chan C. Mental imag-
ery for promoting relearning for people after
stroke: a randomized controlled trial. Arch Phys
Med Rehabil. 2004;85:1403–1408.
115. Logigian MK, Samuels MA, Falconer J, Zagar R.
Clinical exercise trial for stroke patients. Arch Phys
Med Rehabil. 1983;64:364–367.
116. Lum PS, Burgar CG, Shor PC, Majmundar M, Van
der Loos M. Robot-assisted movement training
compared with conventional therapy techniques
for the rehabilitation of upper-limb motor func-
tion after stroke. Arch Phys Med Rehabil.
2002;83:952–959.
117. Marigold DS, Eng JJ, Dawson AS, Inglis JT, Harris
JE, Gylfadottir S. Exercise leads to faster postural
reflexes, improved balance and mobility, and re-
duced falls in older persons with chronic stroke. J
Am Geriatr Soc. 2005;53(3):416–423
118. Moreland JD, Goldsmith CH, Huijbregts MP, et al.
Progressive resistance strengthening exercises af-
ter stroke: a single-blind randomized controlled
trial. Arch Physl Med Rehabil. 2003;84:1433–1440.
119. Mudie MH, Winzeler-Mercay U, Radwan S, Lee L.
Training symmetry of weight distribution after
stroke: a randomized controlled pilot study com-
paring task-related reach, Bobath and feedback
training approaches. Clin Rehabil. 2002;16:582–
592.
120. Paul S, Ramsey D. The effects of electronic music-
making as a therapeutic activity for improving
upper extremity active range of motion. Occup
Ther Int. 1998;5(3):223–237.
121. Potempa K, Lopez M, Braun LT, Szidon JP, Fogg L,
Tincknell T. Physiological outcomes of aerobic ex-
ercise training in hemiparetic stroke patients.
Stroke. 1995;26:101–105.
122. Stein J, Krebs HI, Frontera WR, Fasoli SE, Hughes R,
Hogan N. Comparison of two techniques of robot-
aided upper limb exercise after stroke. Am J Phys
Med Rehabil.. 2004;83(9):720–728.
123. Teixeira-Salmela LF, Olney SJ, Nadeau S, Brouwer
B. Muscle strengthening and physical condition-
Clinical Practice Guidelines 101
ing to reduce impairment and disability in chronic
stroke survivors. Arch Phys Med Rehabil. 1999;
80:1211–1218.
124. Trombly CA, Thayer-Nason L, Bliss G, Girard CA,
Lyrist LA, Brexa-Hooson A. The effectiveness of
therapy in improving finger extension in stroke
patients. Am J Occup Ther. 1986;40(9):612–617.
125. Volpe BT, Krebs HI, Hogan N, Edelsteinn L, Diels
CM, Aisen ML. Robot training enhanced motor
outcome in patients with stroke maintained over 3
years. Neurology. 1999;53:1874–1876.
126. Volpe BT, Krebs HI, Hogan N, Edelstein L, Diels C,
Aisen M. A novel approach to stroke rehabilitation.
Neurology. 2000;54:1938–1944.
127. Winstein CJ, Rose DK, Tan SM, Lewthwaite R, Chui
HC, Azen SP. A randomized controlled compari-
son of upper-extremity rehabilitation strategies in
acute stroke: a pilot study of immediate and long-
term outcomes. Arch Phys Med Rehabil.
2004;85:620–628.
128. Altschuler EL, Wisdom SB, Stone L, Foster C,
Galasko D, Llewellyn DME, Ramachandran VS. Re-
habilitation of hemiparesis after stroke with a mir-
ror. Lancet. 1999;353:2035–2036.
129. Eng JJ, Chu KS, Kim CM, Dawson AS, Carswell A,
Hepburn KE. A community-based group exercise
program for persons with chronic stroke. Med Sci
Sports Exerc. 2003;35(8):1271–1278.
130. Harada N, Chiu V, Fowler E, Lee M, Reuben DB.
Physical therapy to improve functioning of older
people in residential care facilities. Phys Ther.
1995;75(9):830–838.
131. Hesse S, Jahnke MT, Schaffrin A, Lucke D, Reiter F,
Konrad M. Immediate effects of therapeutic facili-
tation on the gait of hemiparetic patients as com-
pared with walking with and without a cane. Neu-
rophysiology. 1998;109:515–522.
132. Macko RF, DeSouza CA, Tretter LD, et al. Treadmill
aerobic exercise training reduces the energy ex-
penditure and cardiovascular demands on
hemiparetic gait in chronic stroke patients: a pre-
liminary report. Stroke. 1997;28:326–330.
133. Nugent JA, Schurr KA, Adams RD. A dose-response
relationship between amount of weight-bearing
exercise and walking outcome following cere-
brovascular accident. Arch Phys Med Rehabil.
1994;75:399–402.
134. Smith GV, Macko RF, Silver KHC, Goldberg AP.
Treadmill aerobic exercise improves quadriceps
strength in patients with chronic hemiparesis fol-
lowing stroke: a preliminary report. J Neuro
Rehabil. 1998;12:111–117.
135. Trombly CA, Quintana LA. The effects of exercise
on finger extension of CVA patients. Am J Occup
Ther. 1983;37(3):1952.
136. Ferraro M, Palazzolo JJ, Krol J, Krebs HI, Hogan N,
Volpe BT. Robot-aided sensorimotor arm training
improves outcome in patients with chronic stroke.
Neurology. 2003;61(11):1604–1607.
137. Lum PS, Burgar CG, Shor PC. Evidence for im-
proved muscle activation patterns after retraining
of reaching movements with the MIME robotic
system in subjects with post-stroke hemiparesis.
102 TOPICS IN STROKE REHABILITATION/SPRING 2006
IEEE Trans Neural Syst Rehabil Eng. 2004;12(2):
186–194.
138. Fujitani J, Ishikawa T, Akai M, Kakurai S. Influence
of daily activity on changes in physical fitness for
people with post-stroke hemiplegia. Am J Phys
Med Rehabil. 1999;78(6):540–544.
139. Marras WS, Mirka GA. Trunk strength during
asymmetric trunk motion. Human Factors.
1989;31(6):667–677.
140. Newham DJ, Hsino SF. Knee muscle isometric
strength, voluntary activation and antagonist co-
contraction in the first six months after stroke.
Disabil Rehabil. 2001;23(9):379–386.
141. Fasoli SE, Krebs, HI, Stein J, Frontera WR, Hughes
R, Hogan N. Robotic therapy for chronic motor
impairments after stroke: follow-up results. Arch
Phys Med Rehabil. 2004;85:1106–1111.
142. Stern PH, McDowell F, Miller JM, Robinson M.
Effects of facilitation exercise techniques in stroke
rehabilitation. Arch Phys Med Rehabil. 1970;51(9):
526–531.
143. Krebs DE, Jette AM, Assmann SF. Moderate exer-
cise improves gait stability in disabled elders. Arch
Phys Med Rehabil. 1998;79:1489–1495.
144. Topp R, Mikesky A, Wigglesworth J, Holt W,
Edwards JE. The effect of a 12-week dynamic resis-
tance strength training program on gait velocity
and balance of older adults. Gerontologist.
1993;33(4):501–509.
145. Barrett JA, Watkins C, Dickinson H, et al. The
COSTAR wheelchair study: a two-centre pilot
study of self-propulsion in a wheelchair in early
stroke rehabilitation. Clin Rehabil. 2001;15:32–41.
146. Basmajian JV, Gowland CA, Brandstater ME, Trot-
ter JE. Intergral behavioral and physical therapy in
the rehabilitation of the post-stroke arm. Int J
Rehabil Res. 1985;8(1):89–90.
147. Engardt M, Knutsson E, Jonsson M, Sternhag M.
Dynamic muscle strength training in stroke pa-
tients: effects on knee extension torque, elec-
tromyographic activity, and motor function. Arch
Phys Med Rehabil. 1995;75:419–425.
148. Fasoli SE, Drebs HI, Hogan N. Robotic technology
and stroke rehabilitation: translating research into
practice. Top Stroke Rehabil. 2004;11(4):11–19.
149. Kondo I, Hosokawa K, Soma M, Iwata M, Maltais
D. Protocol to prevent shoulder-hand syndrome
after stroke. Arch Phys Med Rehabil. 2001;82:
1619–1623.
150. Kopp B, Kunkel A, Muhlnickel W, Villringer K, Taub
E, Flor H. Plasticity in the motor system related to
therapy-induced improvement of movement after
stroke. NeuroReport. 1999;10:807–810.
151. Hesse S, Schmidt H, Werner C, Bardeleben A.
Upper and lower extremity robotic devices for
rehabilitation and for studying motor control. Curr
Opin Neurol. 2003;16(6):701–710.
152. Liston R, Mickelborough J, Harris B, Hann AW,
Tallis RC. Conventional physiotherapy and tread-
mill re-training for higher-level gait disorders in
cerebrovascular disease. Age Ageing. 2000;29:
311–318.
153. Lord JP, Hall K. Neuromuscular reeducation versus
traditional programs for stroke rehabilitation. Arch
Phys Med Rehabil. 1986;67:88–91.
154. Nilsson M, Nordholm LA. Physical therapy in
stroke rehabilitation: bases for Swedish physio-
therapists’ choice of treatment. Physiother Theory
Practice. 1992;8:49–55.
155. Pohl PS, Winstem CJ. Practice effects on the less-
affected upper extremity after stroke. Arch Phys
Med Rehabil. 1999;80:668–675.
156. Potempa K, Braun LT, Tinknell T, Popovich J. Ben-
efits of aerobic exercise after stroke. Sports Med.
1996;21(5):337–346.
157. Sharp SA, Brouwer BJ. Isokinetic strength training
of the hemiparetic knee: effects on function and
spasticity. Arch Phys Med Rehabil. 1997;78:1231–
1236.
158. Stein J. Motor recovery strategies after stroke. Top
Stroke Rehabil. 2004;11(2):2–22.
Task-oriented training
159. Dean CM, Shepherd RB. Task-related training im-
proves performance of seated reaching tasks after
stroke. Stroke. 1997;28(4):722–728.
160. De Sèze M, Wuart L, Bon-Saint-Côme A, et al.
Rehabilitation of postural disturbances of hemiple-
gic patients by using trunk control retraining dur-
ing exploratory exercises. Arch Phys Med Rehabil.
2001;82:793–800.
161. Dickstein R, Heffes Y, Laufer Y, Abulaffio N. Repeti-
tive practice of a single joint movement for en-
hancing elbow function in hemiparetic patients.
Perceptual Motor Skills. 1997;85:771–785.
162. Edmans JA, Webster J, Lincoln NB. A comparison
of two approaches in the treatment of perceptual
problems after stroke. Clin Rehabil. 2000;14:230–
243.
163. Gelber DA, Josefczyk PB, Herman D, Good DC,
Verhulst SJ. Comparison of two therapy ap-
proaches in the rehabilitation of the pure motor
hemiparetic stoke patient. J Neurol Rehabil.
1995;9(4):191–196.
164. Kwakkel G, Wagenaar RC, Twisk JW, Lankhorst GJ,
Koetsier JC. Intensity of leg and arm training after
primary middle-cerebral-artery stroke: a
randomised trial. Lancet. 1999;354:191–196.
165. Kwakkel G, Kollen BJ, Wagenaar RC. Long term
effects of intensity of upper and lower limb train-
ing after stroke: a randomised trial. J Neurol
Neurosurg Psychiatry. 2002;72:473–479.
166. Logan PA, Gladman JRF, Drummond AER, Radford
KA. A study of interventions and related outcomes
in a randomized controlled trial of occupational
therapy and leisure therapy for community stroke
patients. Clin Rehabil. 2003;17:249–255.
167. Luft AR, McCombe-Waller S, Whitall J, et al. Re-
petitive bilateral arm training and motor cortex
activation in chronic stroke. JAMA. 2004;292(15):
1853–1861.
168. Nelson DL, Konosky K, Fleharty K, et al. The effects
of an occupationally embedded exercise on bilat-
erally assisted supination in persons with hemiple-

gia. Am J Occup Ther. 1996;52(8):639–646.
169. Parker CJ, Gladman JRF, Drummond AER, et al. A
multicentre randomized controlled trial of leisure
therapy and conventional occupational therapy
after stroke. Clin Rehabil. 2001;15:42–52.
170. Pomeroy VM, Evans B, Falconer M, Jones D, Hill E,
Giakas G. An exploration of the effects of
weighted garments on balance and gait stroke of
stroke patients with residual disability. Clin Rehabil.
2001;15:390–397.
171. Richards CL, Malouin F, Wood-Dauphinee S, Will-
iams JI, Bouchard JP, Brunet D. Task-specific physi-
cal therapy for optimization of gait recovery in
acute stroke patient. Arch Phys Med Rehabil.
1993;74:612–620.
172. Walker MF, Drummond AER, Lincoln NB. Evalua-
tion of dressing practice for stroke patients after
discharge from hospital: a crossover design study.
Clin Rehabil. 1996;10:23–31.
173. Wiart L, Bon Sait Come A, Debelleix X, Petit H,
Joseph PA, Mazaux JM, Barat M. Unilateral neglect
syndrome rehabilitation by trunk rotation and
scanning training. Arch Phys Med Rehabil.
1997;78:424–429.
174. Bergego C, Azouvi P, Deloche G, Samuel C, Louis-
Dreyfus A, Kaschel R, Willmes K. Rehabilitation of
unilateral neglect: a controlled multiple-baseline-
across-subjects trial using computerized training
procedures. Neuropsychological Rehabil.
1997;7(4):279–283.
175. Byl N, Roderick J, Mohamed O, et al. Effectiveness
of sensory and motor rehabilitation of the upper
limb following the principles of neuroplasticity:
patients stable poststroke. Neurorehabil Neural Re-
pair. 2003:17(3):176–191.
176. Hsieh CL, Nelson DL, Smith DA, Peterson CQ. A
comparison of performance in added-purpose oc-
cupations and rote exercise for dynamic standing
balance in persons with hemiplegia. Am J Occup
Ther. 1996;50(1):10–16.
177. Trombly CA, Wu CY. Effect of rehabilitation tasks
on organization of movement after stroke. Am J
Occup Ther. 1999;53(4):333–344.
178. Whitall J, Waller SM, Silver KHC, Macko RF. Repeti-
tive bilateral arm training with rhythmic auditory
cueing improves motor function in chronic
hemiparetic stroke. Stroke. 2000;31:2390–2395.
179. Wu CY, Wong MK, Lin KC, Chen HC. Effects of task
goal and personal preference on seated reaching
kinematics after stroke. Stroke. 2001;32:70–76.
180. Buterfisch C, Hummelsheim H, Denzler P, Mauritz
KH. Repetitive training of isolated movements im-
proves the outcome of motor rehabilitation of the
centrally paretic hand. J Neurol Sci. 1995;130:59–
68.
181. Engardt M, Knutsson E. Dynamic thigh muscle
strength after auditory feedback training of body
weight distribution in stroke patients. Physiother
Theory Practice. 1994;10:103–112.
182. Maulucci RA, Eckhouse RH. Retraining reaching in
chronic stroke with real-time auditory feedback.
Neurorehabilitation. 2001;16:171–182.
183. Pollock AS, Durward BR, Rowe PJ, Paul JP. The
Clinical Practice Guidelines 103
effect of independent practice of motor tasks by
stroke patients: a pilot randomized controlled trial.
Clin Rehabil. 2002:16:473–480.
184. Hong DA, Corcos DM, Gottlieb GL. Task depen-
dent patterns of muscle activation at the shoulder
and elbow for unconstrained arm movements. J
Neurophysiol. 1994;71(3):1261–1265.
185. Michaelsen SM, Luta A, Roby-Brami A, Levin MF.
Effect of trunk restraint on the recovery of reach-
ing movements in hemiparetic patients. Stroke.
2001;32:1875–1883.
186. Paolucci S, Antonucci G, Guariglia C, Magnotti L,
Pizzamiglio L, Zoccolotti P. Facilitatior effect of
neglect rehabilitation on the recovery of left
hemiplegic stroke patients: a cross-over study. J
Neurol. 1996;243:308–314.
187. Paolucci S, Antonucci G, Grasso MG, Pizzamiglio
L. The role of unilateral spatial neglect in rehabili-
tation of right brain-damaged ischemic stroke pa-
tients: a matched comparison. Arch Phys Med
Rehabil. 2001;82:743–749.
188. Robertson IH, Gray JM, Pentland B, Waite LJ. Mi-
crocomputer-based rehabilitation for unilateral
left visual neglect: a randomized controlled trial.
Arch Phys Med Rehabil. 1990;71:663–668.
189. Asberg KH. Orthostatic tolerance training of stroke
patients in general medical wards an experimental
study. Scand J Rehabil Med. 1989;21:179–185.
190. Hanlon RE. Motor learning following unilateral
stroke. Arch Phys Med Rehabil. 1996;77:811–815.
191. Wu SH, Huang HT, Lin CF, Chen MH. Effects of a
program on symmetrical posture in patients with
hemiplegia: a single-subject design. Am J Occup
Ther. 1996;50(1):17–23.
192. Beer RF, Given JD, Dewald JPA. Task-dependent
weakness at the elbow in patients with hemipare-
sis. Arch Phys Med Rehabil. 1990;80:766–772.
193. Dolecheck JR, Schkade JK. The extent dynamic
standing endurance is effected when CVA subjects
perform personally meaningful activities rather
than nonmeaningful tasks. Occup Ther J Res.
1999;19(1):40–54.
194. Aruin AS, Hanke T, Chaudhuri G, Harvey R, Rao N.
Compelled weight bearing in persons with hemi-
paresis following stroke: the effect of a lift insert
and goal-directed balance exercise. J Rehabil Res
Develop. 2000;37(1):65–72.
195. Smedley RR, Fiorino AJ, Soucar E, Reynolds D,
Smedley WP, Aronica MJ. Slot machines: their use
in rehabilitation after stroke. Arch Phys Med
Rehabil. 1986;67:546–549.
Biofeedback
196. Armagan O, Tascioglu F, Oner C. Electromyo-
graphic biofeedback in the treatment of the
hemiplegic hand: a placebo-controlled study. Am J
Phys Med Rehabil. 2003;82(11):856–861.
197. Basmajian JV, Gowland C, Brandstater ME,
Swanson L, Trotter J. EMG feedback treatment of
upper limb in hemiplegic stroke patients: a pilot
study. Arch Phys Med Rehabil. 1982;63:613–616.
104 TOPICS IN STROKE REHABILITATION/SPRING 2006
198. Basmajian JV, Gowland CA, Finlayson AJ, et al.
Stroke treatment: comparison of integrated be-
havioral-physical therapy vs traditional physical
therapy programs. Arch Phys Med Rehabil.
1987;68:267–272.
199. Binder SA, Moll CB, Wolf SL. Evaluation of elec-
tromyographic biofeedback as an adjunct to
therapeutic exercise in treating the lower extremi-
ties of hemiplegic patients. Phys Ther. 1981;61(6):
886–893.
200. Bourbonnais D, Bilodeau S, Lepage Y, Beaudoin N,
Gravel D, Forget R. Effect of force-feedback treat-
ments in patients with chronic motor deficits after
a stroke. Am J Phys Med Rehabil. 2002;81(12):890–
897.
201. Burnside IG, Tobias HS, Bursill D. Electromyo-
graphic feedback in the rehabilitation of stroke
patients: a controlled trial. Arch Phys Med Rehabil.
1982;63:217–222.
202. Crow JL, Lincoln NB, Nouri RM, DeWeerdt W. The
effectiveness of EMG biofeedback in the treatment
of arm function after stroke. Int Disabil Studies.
1989;11(4):155–160.
203. Intiso D, Santilli V, Grasso MG, Rossi R, Caruso I.
Rehabilitation of walking with electromyographic
biofeedback in foot-drop after stroke. Stroke.
1994;25(6):1189–1192.
204. Kraft GH, Fitts SS, Hammond MC. Techniques to
improve function of the arm and hand in chronic
hemiplegia. Arch Phys Med Rehabil. 1992;73:220–
227.
205. Mandel AR, Nymark JR, Balmer SJ. Electromyo-
graphic versus rhytmic positional biofeedback in
computerized gait retraining with stroke patients.
Arch Phys Med Rehabil. 1990;71:649–654.
206. Montoya R, Dupui Ph., Pages B, Bessou P. Step-
length biofeedback device for walk rehabilitation.
Med Biol Eng Comput. 1994;32:416–420.
207. Tham K, Tegner R. Video feedback in the rehabili-
tation of patients with unilateral neglect. Arch Phys
Med Rehabil. 1997;78:410–413.
208. Williams JM. Use of electromyographic biofeed-
back for pain reduction in the spastic hemiplegic
shoulder: a pilot study. Physiother Canada. 1982;
34(6):327–333.
209. Wolf SL, Binder-Macleod SA. Electromyographic
biofeedback applications to the hemiplegic pa-
tient: changes in lower extremity neuromuscular
and functional status. Phys Ther. 1983;63(9):
1404–1413.
210. Wolf SL, Binder-Macleod SA. Electromyographic
biofeedback applications to the hemiplegic pa-
tient: changes in upper extremity neuromuscular
and functional status. Phys Ther. 1983;63(9):
1393–1403.
211. Wolf SL, Catlin PA, Blanton S, Edelman J, Lehrer N,
Schroeder D. Overcoming limitations in elbow
movement in the presence of antagonist hyperac-
tivity. Phys Ther. 1994;74(9):826–835.
212. Cauraugh JH, Kim S. Two coupled motor recovery
protocols are better than one: electromyogram-
triggered neuromuscular stimulation and bilateral
movements. Stroke. 2002;33:1589–1594.
213. Greenberg S, Fowler Jr RS. Kinesthetic biofeed-
back: a treatment modality for elbow range of
motion in hemiplegia. Am J Occup Ther.
1980;34(11):738–743.
214. Inglis J, Donald MW, Monga TN, Sproule M,
Young MJ. Electromyographic biofeedback and
physical therapy of the hemiplegic upper limb.
Arch Phys Med Rehabil. 1984;65:755–759.
215. Llorca R, Rodriguez LP, Ponce PJ. Biofeedback and
its use in treatments for controlling spasticity.
Europa Medicophysica. 1999;35(1):3–9.
216. Prevo AJH, Visser SL, Vogelaar TW. Effect of EMG
feedback on paretic muscles and abnormal co-
contraction in the hemiplegic arm, compared
with conventional physical therapy. Scand J
Rehabil. 1982;14:121–131.
217. Mulder T, Van der Meer J. EMG feedback and the
restoration of motor control. Am J Phys Med.
1986;65(4):173–189.
218. Skelly AM, Kenedi RM. EMG biofeedback therapy
in the re-education of the hemiplegic shoulder in
patients with sensory loss. Physiotherapy. 1982;
68(2):34–38.
219. Smith KN. Biofeedback in strokes. Austral J
Physiother. 1979;25(4):155–160.
220. Wolf SL, Baker MP, Kelly JL. EMG biofeedback in
stroke: a 1-year follow-up on the effect of patients
characteristics. Arch Phys Med Rehabil.
1980;61:351–355.
221. Hurd WW, Pegram V, Nepomuceno C. Compari-
son of actual and simulated EMG biofeedback in
the treatment of hemiplegic patients. Am J Phys
Med. 1980;59(2):73–82.
222. John J. Failure of electrical myofeedback to aug-
ment the effects of physiotherapy in stroke. Int J
Rehabil Res. 1986;9(1):35–45.
223. Lee K-H, Hill E, Johnston R, Smiehorowski T.
Myofeedback for muscle retraining in hemiplegic
patients. Arch Phys Med Rehabil. 1976;57:588–
591.
224. Mathieu PA, Sullivan SJ. Changes in the
hemiparetic limb with training I. Torque output.
Electromyogr Clin Neurophysiol. 1995;35:491–513.
225. Ring H, Tsur A, Vashdi Y. Long-term clinical and
electromyographical (EMG) follow-up of
hemiplegic’s shoulder. Eur J Phys Med Rehabil.
1993;3(4):137–140.
226. Smith LE. Restoration of volitional limb movement
of hemiplegics following patterned functional
electrical stimulation. Percept Motor Skills.
1990;71:851–861.
227. Wannstedt GT, Herman RM. Use of augmented
sensory feedback to achieve symmetrical stand-
ing. Phys Ther. 1978;58(5):553–559.
228. Balliet R, Levy B, Blood KMT. Upper extremity
sensory feedback therapy in chronic cerebrovascu-
lar accident patients with impaired expressive
aphasia and auditory comprehension. Arch Phys
Med Rehabil. 1986;67:304–310.
229. Davis AE, Lee RG. EMG biofeedback in patients
with motor disorders: an aid for co-ordinating
activity in antagonistic muscle groups. Can J
Neurol Sci. 1980;7(3):199–206.

230. Mroczek N, Halpern D, McHugh R. Electromyo-
graphic feedback and physical therapy for neuro-
muscular retraining in hemiplegia. Arch Phys Med
Rehabil. 1978;59:258–267.
231. Cheng HS, Ju MS, Lin CK. Improving elbow torque
output of stroke patients with assistive torque con-
trolled by EMG signals. J Biomech Eng. 2003;
125:881–886.
232. Engardt M. Rising and sitting down in stroke pa-
tients. Scand J Rehabil Med. 1994;(suppl. 31):3–57.
233. Horak FB, Esselman P, Anderson ME Lynch MK. The
effects of movement velocity, mass displaced, and
task certainty on associated postural adjustments
made by normal and hemiplegic individuals. J
Neurol Neurosurg Psychiatry. 1984;47:1020–1028.
234. Bate PJ, Matyas TA. Negative transfer of training
following brief practice of elbow tracking move-
ments with electromyographic feedback from
spastic antagonists. Arch Phys Med Rehabil.
1992;73:1050–1058.
235. Fowler V, Carr J. Auditory feedback: effects on
vertical force production during standing up fol-
lowing stroke. Int J Rehabil Res. 1996;19:265–269.
236. Takebe K, Kukulka CG, Narayan MG, Basmajian JV.
Biofeedback treatment of foot drop after stroke
compared with standard rehabilitation technique
(part 2 ): effects on nerve conduction velocity and
spasticity. Arch Phys Med Rehabil. 1976;57:9–11.
237. Moreland JD, Thomson MA, Fuoco AR. Elec-
tromyographic biofeedback to improve lower ex-
tremity function after stroke: a meta-analysis. Arch
Phys Med Rehabil. 1998;79:134–140.
238. Baikoushev ST, Rathkolb O, Baikousheva V. Spas-
ticity: kinetic versus static synergias in the reeduca-
tion of hemiparetics by EMG biofeedback. Folia
Medica (Plovdiv). 1985;27:5–17.
239. Basmajian JV, Kukulka CG, Narayan MG, Takebe K.
Biofeedback treatment of drop-foot after stroke
compared with standard rehabilitation technique:
effects on voluntary control and strength. Arch
Phys Med Rehabil. 1975;56:231–236.
240. Amato A, Hermsmeyer CA, Kleinman KM. Use of
electromyographic feedback to increase inhibitory
control of spastic muscles. Phys Ther.
1973;53(10):1063–1066.
241. Brudny J, Korein J, Grynbaum BB, Belandres PV,
Gianutsos JG. Helping hemiparetic to help them-
selves - sensory feedback therapy. J Am Med Assoc.
1979;241(8):814–818.
242. Ince LP, Leon MS, Christidis D. EMG biofeedback
with upper extremity musculature for relaxation
training: a critical review of the literature. J Behav
Ther Exp Psychiatry. 1985;16(2):133–137.
243. Lee SJ, Ahn YP. Clinical effect of electromyo-
graphic biofeedback treatment in hemiplegic pa-
tients. J Catholic Med Coll. 1985;38(1):387–394.
244. Marinacci AA, Hornade M. Electromyographic in
neuromuscular re-education. Bull Los Angeles
Neurol Soc. 1960;25(2):57–71.
245. Olney SJ, Nymark JR, Zee BC-Y. Computer-assisted
feedback and gait re-education in stroke patients:
a two-centre randomized control trial. Canadian J
Rehabil.1993;7(1):31–33.
Clinical Practice Guidelines 105
246. Rakos M, Freudenschub B, Girsch W, et al. Elec-
tromyogram-controlled functional electrical
stimulation for treatment of the paralysed upper
extremity. Artificial Organs. 1999;23(5):466–469.
247. Wolf SL, LeCraw DE, Barton LA. Comparison of
motor copy and targeted biofeedback training
techniques for restitution of upper extremity func-
tion among patient s with neurologic deficit. Phys
Ther. 1989;69(9):719–735.
Gait training
248. Ada L, Dean C, Hall JM, Bampton J, Crompton S. A
treadmill and overground walking program im-
proves walking in persons residing in the commu-
nity after stroke: a placebo-controlled, random-
ized trial. Arch Phys Med Rehabil. 2003:84:
1486–1491.
249. Barbeau H, Visintin M. Optimal outcomes ob-
tained with body-weight support combined with
treadmill training in stroke patients. Arch Phys Med
Rehabil. 2003;84:1458–1465.
250. Bogataj U, Gros N, Klajic M, Acimovic R, Malezic
M. The rehabilitation of gait in patients with
hemiplegia: a comparison between conventional
therapy and multichannel functional electrical
stimulation therapy. Phys Ther. 1995;75:490–502.
251. Burridge JH, Taylor PN, Hagan SA, Wood DE,
Swain ID. The effects of common peroneal stimu-
lation on the effort and speed of walking: a ran-
domized controlled trial with chronic hemiplegic
patients. Clin Rehabil. 1997;11:201–210.
252. da Cunha Filho IT, Lim PAC, Quershy H, Henson H,
Monga T, Protas EJ. A comparison of regular reha-
bilitation with supported treadmill ambulation for
acute stroke patients. J Rehabil Res Develop.
2001;38(2):245–255.
253. da Cunha IT, Lim PA, Quershy H, Henson H,
Monga T, Portas EJ. Gait outcomes after acute
stroke rehabilitation with supported treadmill
ambulation training: a randomized controlled pi-
lot study. Arch Phys Med. 2002:83:1258–1265.
254. Kosak MC, Reding MJ. Comparison of partial body
weight-supported treadmill gait training versus
aggressive bracing assisted walking post stroke.
Neurorehabil Neural Repair. 2000;14(1):13–19.
255. Laufer Y, Dickstein R, Chefez Y, Marcovitz E. The
effects of treadmill training on the ambulation of
stroke survivors in the early stages of rehabilita-
tion: a randomized study. J Rehabil Res Develop.
2001;38(1):69–78.
256. Macdonell RAL, Triggs WJ, Leikauskas J, Bourque
M, Robb K, Day BJ, Shahani BT. Functional electri-
cal stimulation to the affected lower limb and
recovery after cerebral infarction. J Stroke
Cerebrovasc Dis. 1994;4(3):155–160.
257. Nilsson L, Carlsson J, Danielsson A, et al. Walking
training of patients with hemiparesis at an early
stage after stroke: a comparison of walking train-
ing on a treadmill with body weight support and
walking training on the ground. Clin Rehabil.
2001;15:515–527.
106 TOPICS IN STROKE REHABILITATION/SPRING 2006
258. Pohl M, Mehrholz J, Ritschel C, Rückriem S. Speed-
dependant treadmill training in ambulatory
hemiparetic stroke patients: a randomized con-
trolled trial. Stroke. 2002;33:553–558.
259. Schauer M, Mauritz KH. Musical motor feedback
(MMF) in walking hemiparetic stroke patients:
randomized trials of gait improvement. Clin
Rehabil. 2003;17:713–722.
260. Sullivan KJ, Knowlton BJ, Dobkin BH. Step training
with body weight support: effect of treadmill
speed and practice paradigms on post-stroke lo-
comotor recovery. Arch Phys Med Rehabil.
2002;83:683–691.
261. Thaut MH, McIntosh GC, Rice RR. Rhythmic facili-
tation of gait training in hemiparetic stroke reha-
bilitation. J Neurol Sci. 1997;151:207–212.
262. Visintin M, Barbeau H, Korner-Bitensky N, Mayo
NE. A new approach to retrain gait in stroke pa-
tients through body weight support and treadmill
stimulation. Stroke. 1998;29:1122–1128.
263. Werner C, Bardeleben A, Mauritz K H, Kirker S,
Hesse S. Treadmill training with partial body
weight support and physiotherapy in stroke pa-
tients: a preliminary study. Eur J Neurol.
2002;9:639–644.
264. Yan T, Hui-Chang WY, Li SW. Functional electrical
stimulation improves motor recovery of the lower
extremity and walking ability of subjects with first
acute stroke. Stroke. 2005:36:80–85
265. Barbeau H, Norman K, Fung J, Visintin M,
Ladouceur M. Does neurorehabilitation play a role
in the recovery of walking in neurological popula-
tions? Ann NY Acad Sci. 1998;860:377–392.
266. Basmjian JV, Kukulka BS, Narayan MG, Takebe K.
Biofeedback treatment of foot-drop after stroke
compared with standard rehabilitation technique:
effects on voluntary control and strength. Arch
Phys Med Rehabil. 1975;56:231–236.
267. Beckerman H, Becher J, Lankhorst GJ, Verbeek
ALM. Walking ability of stroke patients: efficacy of
tibial nerve blocking and a polyprophylene ankle-
foot orthosis. Arch Phys Med Rehabil.
1996;6(5):141–144.
268. Ceceli E, Dursun E, Çakçi A. Comparison of joint-
position biofeedback and conventional therapy
methods in recurvatum after stroke—6 months’
follow-up. Eur J Phys Med Rehabil. 1996;6(5):141–
144.
269. Danielsson A, Sunnerhagen KS. Oxygen consump-
tion during treadmill walking with and without
body weight support in patients with hemiparesis
after stroke and in healthy subjects. Arch Phys Med
Rehabil. 2000;81:953–957.
270. Dean CM, Richards CL, Malouin F. Task-related
circuit training improves performance of locomo-
tor tasks in chronic stroke: a randomized, con-
trolled trial. Arch Phys Med Rehabil. 2000;81:409–
417.
271. Duncan PW. Stroke disability. Phys Ther.
1994;74(5):399–407.
272. Hesse S, Bertlelt C, Schaffrin A, Malezic M, Mauritz
KH. Restoration of gait in nonambulatory
hemiparetic patients by treadmill training with
partial body-weight support. Arch Phys Med
Rehabil. 1994;75:1087–1093.
273. Iwata M, Kondo I, Sato Y, Satoh K, Soma M. An
ankle-foot orthosis with inhibitor bar: effect on
hemiplegic gait. Arch Phys Med Rehabil. 2003;84:
924–927.
274. Mulder T, Hulstijn W, Van Der Meer J. EMG feed-
back and the restoration of motor control. Am J
Phys Med. 1986;65(4):173–189.
275. Nakamura R, Hosokawa T, Yamada Y, Mojica JA.
Application of computer-assisted gait training
(CAGT) program for hemiparetic stroke patients: a
preliminary report. Tohoku J Exp Med.
1988;156:101–107.
276. Shumway-Cook A, Anson D, Haller S. Postural
sway biofeedback: its effect on reestablishing
stance stability in hemiplegic patients. Arch Phys
Med Rehabil. 1988;69:395–400.
277. Smith GV, Silver KH, Goldberg AP, Macko RF.
“Task-oriented” exercise improves hamstring
strength and spastic reflexes in chronic patients.
Stroke. 1999;30:2112–2118.
278. Taylor PN, Burridge JH, Dunkerley AL, Wood DE,
Norton JA, Singleton C, Swain ID. Clinical use of
the Odstock dropped foot stimulator: its effect on
the speed effort of walking. Arch Phys Med Rehabil.
1999;80:1577–1583.
279. Werner RA, Kessler S. Effectiveness of an intensive
outpatient rehabilitation program for postacute
stroke patients. Am J Phys Med Rehabil.
1996;75(2):114–120.
280. Werner C, von Frankenberg S, Treig T, Konrad M,
Hesse S. Treadmill training with partial body
weight support and an electromagnetic gait
trainer for restoration of gait in subacute stroke
patients. Stroke. 2002;33(12):2895–2901.
281. Wong AM, Su TY, Tang FT, Cheng PT, Liaw MY.
Clinical trial of electrical acupuncture on hemiple-
gic stoke patients. Am J Phys Med Rehabil.
1999:117–122.
282. Wissel J, Ebersback G, Gutjahr L, Dahlke F. Treat-
ing chronic hemiparesis with modified biofeed-
back. Arch Phys Med Rehabil. 1989;70:612–617.
283. Yavuzer G, Ergin S. Effect of an arm sling on gait
pattern in patients with hemiplegia. Arch Phys Med
Rehabil. 2002;83:960–963.
Balance training
284. Aruin AS, Hanke TA, Sharma A. Base of support
feedback in gait rehabilitation. Int J Rehabil Res.
2003;26:309–312.
285. Chen IC, Cheng PT, Chen CL, Chen SC, Chung
CY, Yeh TH. Effects of balance training on
hemiplegic stroke patients. Chang Gung Med J.
2002;25(9):583–590.
286. Cheng PT, Wu SH, Liaw MY, Wong AMK, Tang FT.
Symmetrical body-weight distribution training in
stroke patients and its effects on fall prevention.
Arch Phys Med Rehabil. 2001;82:1650–1654.
287. Engardt M. Long term effect of auditory feedback
training on relearned symmetrical body weight

distribution in stroke patients. Scand J Rehabil Med.
1994;26:65–69.
288. Geiger RA, Allen JB, O’Keefe J, Hicks R. Balance and
mobility following stroke: effects of physical
therapy interventions with and without biofeed-
back/forceplate training. Phys Ther.
2001;81(4):995–1005.
289. Grant T, Brouwer B, Culham E, Vandervoort A.
Balance retraining following acute stroke: a com-
parison of two methods. Can J Rehabil.
1997;11(2):69–73.
290. Hocherman S, Dickstein R, Pillar T. Platform train-
ing and postural stability in hemiplegia. Arch Phys
Med Rehabil 1984;65:588–592.
291. Sackley CM, Lincoln NB. Single blind randomized
controlled trial of visual feedback after stroke: ef-
fects on stance symmetry and function. Disabil
Rehabil. 1997;19(12):536–546.
292. Walker C, Brouwer BJ, Culham EG. Use of visual
feedback in retraining balance following acute
stroke. Phys Ther. 2000;80(9):886–895.
293. Winstein CJ, Gardner ER, McNeal DR, Barto PS,
Nicholson DE. Standing balance training: effect on
balance and locomotion in hemiparetic adults.
Arch Phys Med Rehabil. 1989;70:755–762.
294. Barreca S, Sigouin C.S., Lambert, C., Ansley, B.
Effects of extra training on the ability of stroke
survivors to perform an independent sit-to-stand:
a randomized controlled trial. J Geriatr Phys Ther.
2004;27:59–64.
295. Bonan IV, Yelnik AP, Colle FM, et al. Reliance on
visual information after stroke. Part II: effectiveness
of a balance rehabilitation program with visual cue
deprivation after stroke: a randomized controlled
trial. Arch Phys Med Rehabil. 2004;85:274–278.
296. Engardt M, Ribbe T, Olsson E. Vertical ground
reaction force feedback to enhance stroke pa-
tients’ symmetrical body-weight distribution while
rising/sitting down. Scand J Rehabil Med.
1993;25:41–48.
Sensory interventions
297. Fanthome Y, Lincoln NB, Drummond A, Walker
MF. The treatment of visual neglect using feed-
back of eye movements: a pilot study. Disabil
Rehabil. 1995;17(8):413–417.
298. Feys H, De Weerdt W, Verbeke G, et al. Early and
repetitive stimulation of the arm can substantially
improve the long-term outcome after stroke: a 5-
year follow-up study of a randomized trial. Stroke.
2004;35:924–929.
299. Feys HM, De Weerdt WJ, Selz BE, et al. Effect of a
therapeutic intervention for the hemiplegic upper
limb in the acute phase after stroke. Stroke.
1998;29:785–792.
300. Fiebert IM, Brown E. Vestibular stimulation to im-
prove ambulation after a cerebral vascular acci-
dent. Phys Ther. 1979;59(4):423–426.
301. Jongbloed L, Stacey S, Brighton C. Stroke rehabili-
tation: sensorimotor integrative treatment versus
functional treatment. Am J Occup Ther.
Clinical Practice Guidelines 107
1989;43:391–397.
302. Mazer BL, Sofer S, Korner-Bitensky N, Gelinas I,
Hanley J, Wood-Dauphinee S. Effectiveness of a
visual attention retraining program on the driving
performance of clients with stroke. Arch Phys Med
Rehabil. 2003;84:541–550.
303. Morioka S, Yagi F. Effects of perceptual learning
exercises on standing balance using a hardness
discrimination task in hemiplegic patients follow-
ing stroke: a randomized controlled pilot study.
Clin Rehabil. 2003;17:600–607.
304. Page SJ. Imagery improves upper extremity motor
function in chronic stroke patients: a pilot study.
Occup Ther J Res. 2000;20(3):200–215.
305. Yekutiel M, Guttman E. A controlled trial of the
retraining of the sensory function of the hand in
stroke patients. Neurol Neurosurg Psychiatry.
1993;56:241–244.
306. Conforto AB, Kaelin-Lang A, Cohen LG. Increase in
hand muscle strength of stroke patients after so-
matosensory stimulation. Ann Neurol. 2002;51(1):
122–125.
307. Hommel M, Peres B, Pollak P, Memin B, Besson G,
Gaio JM, Perret J. Effects of passive tactile and
auditory stimuli on left visual neglect. Arch Neurol.
1990;47:573–576.
308. Garland SJ, Hayes KC. Effects of brushing on elec-
tromyographic activity and ankle dorsiflexion in
hemiplegic subjects with foot drop. Physiother
Canada. 1987;39(4):239–247.
309. Weinberg J, Piasetsky E, Diller L, Gordon W. Treat-
ing perceptual organization deficits in
nonneglecting RBD stroke patients. J Clin
Neuropsychol. 1982;4(1):59–75.
310. Carey LM, Matyas TA, Oke LE. Sensory loss in
stroke patients: effective training of tactile and
proprioceptive discrimination. Arch Phys Med
Rehabil. 1993;74:602–611.
311. Hajek VE, Kates MH, Donnelly R, McGree S. The
effect of visuo-spatial training in patients with
right hemisphere stroke. Can J Rehabil.
1993;6:175–186.
312. Giaquinto S, Fraioli L. Enhancement of the soma-
tosensory N140 component during attentional
training after stroke. Clin Neurophysiol.
2003;114:329–335.
313. Lincoln NB, Whiting SE, Cockburn J, Bhavnani G.
An evaluation of perceptual retraining. Int Rehabil
Med. 1985;7:99–110.
314. Webster JS, Jones S, Blanton P, Gross R, Beissel GF,
Wofford JD. Visual scanning training with stroke
patients. Behav Ther. 1984;15:129–143.
315. Ageranioti SA, Hayes KC. Effects of vibration on
hypertonia and hyperreflexia in the wrist joint of
patients with spastic hemiparesis. Physiother Can.
1990;42(1):24–33.
316. Robertson IH, Nico D, Hood BM. Believing what
you feel: using proprioceptive feedback to reduce
unilateral neglect. Neuropsychology.
1997;11(1):53–58.
317. Taylor MM, Schaeffer JN, Blumenthal FS, Grisell JL.
Perceptual training in patients with left hemiple-
gia. Arch Phys Med Rehabil. 1971;52(4):163–169.
108 TOPICS IN STROKE REHABILITATION/SPRING 2006
318. Magnusson M, Johansson K, Johansson BB. Sen-
sory stimulation promotes normalization of pos-
tural control after stroke. Stroke. 1994;25:1176–
1180.
Constraint-induced movement therapy
319. Dromerick AW, Dorothy FE, Hahn M. Does the
application of constraint-induced movement
therapy during acute rehabilitation reduce arm
impairment after ischemic stroke? Stroke.
2000;31:2984–2988.
320. Ploughman M, Corbett D. Can forced-use therapy
be clinically applied after stroke? An exploratory
randomized controlled trial. Arch Phys Med
Rehabil. 2004;85:1417–1423.
321. Sterr A, Elbert T, Berthold I, Kölbert S, Rockstroh B,
Taub E. Longer versus shorter daily constraint-
induced movement therapy of chronic hemipare-
sis: An exploratory study. Arch Phys Med Rehabil.
2002;83:1374–1377.
322. van der Lee JH, Wagenaar RC, Lankhorst GJ,
Vogelaar TW, Devillé WL, Bouter LM. Forced used
of the upper extremity in chronic stroke patients.
Stroke. 1999;30:2369–2375.
323. Wittenberg GF, Chen R, Ishii K, Bushara KO, Taub
E, Gerber LH, Hallett M, Cohen LG. Constraint-
induced therapy in stroke: magnetic–stimulation
motor maps and cerebral activation. Neurorehabil
Neural Repair. 2003;17(1):48–57.
324. Kunkel A, Bauder H, Sommer M, Dettmers C, Taub
E. Constraint-induced movement therapy for mo-
tor recovery in chronic stroke patients. Arch Phys
Med Rehabil. 1999;80:624–628.
325. Liepert L, Miltner WHR, Bauder H, Sommer M,
Dettmers C, Taub E, Weiller C. Motor cortex plas-
ticity during constraint-induced movement
therapy in stroke patients. Neurosci Lett.
1998;250:5–8.
326. Miltner WHR, Bauder H, Sommer M, Dettmers C,
Taub E. Effects of constraint-induced movement
therapy on patients with chronic motor deficits
after stroke. Stroke. 1999;30:586–592.
327. Page SJ, Sisto SA, Levine P, McGrath RE. Efficacy of
modified constraint-induced movement therapy
in chronic stroke: a single-blinded randomized
controlled trial. Arch Phys Med Rehabil.
2004;85:14–18.
328. Taub E, Miller NE, Novack TA, et al. Technique to
improve chronic motor deficit after stroke. Arch
Phys Med Rehabil. 1993;74:347–354.
329. Winstein CJ, Miller JP, Blanton S, et al. Methods for
a multisite randomized trial to investigate the ef-
fect of constraint-induced movement therapy in
improving upper extremity function among adults
recovering from cerebrovascular stroke.
Neurorehabil Neural Repair. 2003;17:137–152.
330. Ostendorf CG, Wolf SL. Effect of forced use of the
upper extremity of a hemiplegic patient on
changes in function. Phys Ther. 1981;67(7):1022–
1028.
331. Ottenbacher KJ, Jannerl S. The results of clinical
trials in stroke rehabilitation research. Arch Neurol.
1993;50:37–44.
332. Rose DK, Winstein CJ, Bimanual training after
stroke: are two hands better than one? Top Stroke
Rehabil. 2004;11(4):20–30.
333. Tremblay F, Tremblay LE. Constraint-induced
movement therapy: evidence for its applicability
in the context of a home rehabilitation interven-
tion for subacute stroke. Physiother Can.
2002:116–121,131.
334. Wolf SL, Lecraw DE, Barton LA, Jann BB. Force use
of hemiplegic upper extremities to reverse the
effect of learned nonuse among chronic stroke
and head-injured patients. Exp Neurol.
1989;104:125–132.
Shoulder subluxation
335. Dean CM, Mackey FH, Katrak P. Examination of
shoulder positioning after stroke: a randomized
controlled pilot trial. Aust J Physiother. 2000;46(1):
35–40.
336. Faghri PD, Rodgers MM, Glaser RM, Bors JG, Ho C,
Akuthola P. The effects of functional electrical
stimulation on shoulder subluxation, arm function
recovery, and shoulder pain in hemiplegic stroke
patients. Arch Phys Med Rehabil. 1994;75:73–79.
337. Hanger HC, Whitewood P, Brown G, Ball MC,
Harper J, Cox R, Sainsbury R. A randomized con-
trolled trial of strapping to prevent stroke shoulder
pain. Clin Rehabil. 2000;14(4):370–380.
338. Linn SL, Granat MH, Lees KR. Prevention of shoul-
der subluxation after stroke with electrical stimula-
tion. Stroke. 1999;30(5):963–968.
339. Wang RY, Chan RC, Tsai MW. Functional electrical
stimulation on chronic and acute hemiplegic
shoulder subluxation. Am J Phys Med Rehabil.
2000;79(4):385–390.
340. Chantraine A, Baribeault A, Uebelhart D, Gremion
G. Shoulder pain and dysfunction in hemiplegia:
effects of functional electrical stimulation. Arch
Phys Med Rehabil. 1999;80(3):328–331.
341. Kobayashi H, Onishi H, Ihashi K, Yagi R, Handa Y.
Reduction in subluxation and improved muscle
function of the hemiplegic shoulder joint after
therapeutic electrical stimulation. J Electromyogr
Kinesiol. 1999;9(5):327–336.
342. Williams R, Toffs L, Minuk T, Evaluation of two
support methods for the subluxated shoulder of
hemiplegic patients. Phys Ther. 1988;68(8):1209–
1214.
343. Brooke MM, de Lateur BJ, Diana-Rigby GC,
Questad KA. Shoulder subluxation in hemiplegia:
effects of three different supports. Arch Phys Med
Rehabil. 1991;72(8):582–586.
344. Hurd MM, Farrell KH, Waylonis GW. Shoulder
sling for hemiplegia: friend or foe? Arch Phys Med
Rehabil. 1974;55:519–522.
345. Yu DT, Chae J, Walker ME, Hart RL, Petroski GF.
Comparing stimulation-induced pain during per-
cutaneous (intramuscular) and transcutaneous
neuromuscular electric stimulation for treating

shoulder subluxation in hemiplegia. Arch Phys Med
Rehabil. 2001;82:756–760.
346. Zorowitz RD, Idank D, Ikai T, Hughes MB,
Johnston MV. Shoulder subluxation after stroke: a
comparison of four supports. Arch Phys Med
Rehabil. 1995;76:763–771.
347. Bohannon RW, Andrews AW. Shoulder sublux-
ation and pain in stroke patients. Am J Occup Ther.
1990;44(6):507–509.
348. Fitzgerald-Finch OP, Gibson II. Subluxation of the
shoulder in hemiplegia. Age Ageing. 1975;4:16–
18.
349. Hakuno A, Sashika H, Ohkawa T, Itoh R.
Arthrographic findings in hemiplegic shoulders.
Arch Phys Med Rehabil. 1984;65(11):706–711.
350. Jensen EM. The hemiplegic shoulder. Scand J
Rehabil Med. 1980;(suppl 7):113–119.
351. Porter SF, Hershey SL. Supporting the subluxated
shoulder joint with a pajama-bottom sling. Nurs-
ing. 1979;9(4):56–57.
352. Jackson D, Turner-Stokes L, Khatoon A, Stern H,
Knight L, O’Connell A. Development of an inte-
grated care pathway for the management of
hemiplegic shoulder pain. Disabil Rehabil.
2002;24(7):390–398.
353. Hanger HC, Whitewood P, Brown G, Ball MC,
Harper J, Marson R. Strapping the stroke shoulder.
J Clin Neurosci. 1999;6(3):286.
354. Braun RM, West F, Mooney V, Nickel VL, Roper B,
Caldwell C. Surgical treatment of the painful
shoulder contracture in the stroke patient. J Bone
Joint Surg Am. 1971;53(7):1307–1312.
355. Healey WE, Falconer JA. Aggressive gait training in
inpatient geriatric rehabilitation: a pilot study. Is-
sues On Aging. 1996;19(2):3–9.
Electrical stimulation
356. Baker LL, Parker K. Neuromuscular electrical
stimulation of the muscles surrounding the shoul-
der. Phys Ther. 1986;66(12):1930–1937.
357. Bowman BR, Baker LL, Waters RL. Positional feed-
back and electrical stimulation: an automated
treatment for the hemiplegic wrist. Arch Phys Med
Rehabil. 1979;60:497–501.
358. Cauraugh JH, Kim SB. Stroke motor recovery: ac-
tive neuromuscular stimulation and repetitive
practice schedules. J Neurol Neurosurg Psychiatry.
2003;74:1562–1566.
359. De Kroon JR, Ijzerman MJ, Lankhorst GJ, Zilvold G.
Electrical stimulation of the upper limb in stroke:
stimulation of the extensors of the hand vs. alter-
nate stimulation of flexors and extensors. Am J
Phys Med Rehabil. 2004;83(8):592–600.
360. Faghri PD, Rodgers MM. The effects of functional
neuromuscular stimulation-augmented physical
therapy program in the functional recovery of
hemiplegic arm in stroke patients. Clin Kinesiol J
Am Kinesiother Assoc. 1997;51(1):9–15.
361. Heckmann J, Mokrusch T, Kröckel A, Warnke S,
von Stockert T, Neundörfer B. EMG-triggered
electrical muscle stimulation in the treatment of
Clinical Practice Guidelines 109
central hemiparesis after a stroke. Eur J Phys Med
Rehabil. 1997;7(5):138–141.
362. Kimberley TJ, Lewis SM, Auerbach EJ, Dorsey LL,
Lojovich JM, Carey JR. Electrical stimulation driv-
ing functional improvements and cortical changes
in subjects with stroke. Exp Brain Res.
2004;154:450–460.
363. Popovic MB, Popovic DB, Sinkjaer T, Stefanovic A,
Schwirtlich L. Clinical evaluation of functional
electrical therapy in acute hemiplegic subjects. J
Rehabil Res Dev. 2003;40(5):443–454.
364. Powell J, Pandyan DA, Granat M. Cameron M,
Scott DJ. Electrical stimulation of wrist extensors in
poststroke hemiplegia. Stroke. 1999;30(7):1384–
1389.
365. Winchester P, Montgomery J, Bowman B, Hislop
H. Effects of feedback stimulation training and
cyclical electrical stimulation on knee extension in
hemiparetic patients. Phys Ther. 1983;63(7):
1096–1103.
366. Baker LL, Yeh C, Wilson D, Waters RL. Electrical
stimulation of wrist and fingers for hemiplegic
patients. Phys Ther. 1979;59(12):1495–1499.
367. Faghri PD. The effects of neuromuscular stimula-
tion-induced muscle contraction versus elevation
on hand edema in CVA patients. J Hand Ther.
1997;10:29–34.
368. Fields WR. Electromyographically triggered elec-
tric muscle stimulation for chronic hemiplegia.
Arch Phys Med Rehabil. 1987;68:407–414.
369. Hendricks HT, Ijzerman MJ, de Kroon JR, Zilvold G.
Functional electrical stimulation by means of the
“ness handmaster orthosis” in chronic stroke pa-
tients: an exploratory study. Clin Rehabil.
2001;15:217–220.
370. Hummelsheim H, Maier-Loth ML, Eickhof C. The
functional value of electrical muscle stimulation
for the rehabilitation of the hand in stroke pa-
tients. Scand J Rehabil Med. 1997;29:3–10.
371. Packman-Braun R. Relationship between func-
tional electrical stimulation duty cycle and fatigue
in wrist extensor muscles of patients with hemi-
paresis. Phys Ther. 1988;68(1):51–56.
372. Pandyan AD, Grant MH, Stott DJ. Effects of electri-
cal stimulation on flexion contractures in the
hemiplegic wrist. Clin Rehabil. 1997;11:123–130.
373. Potisk KP. Gregoric M, Vodovnik L, Ind D. Effects
of transcutaneous electrical nerve stimulation
(TENS) on spasticity in patients with hemiplegia.
Scand J Rehabil Med. 1995;27(3):169–174.
374. Renzenbrink GJ, Ijzerman M. Percutaneous neuro-
muscular electrical stimulation (P-NMES) for treat-
ing shoulder pain in chronic hemiplegia. Effects on
shoulder pain and quality of life. Clin Rehabil.
2004;18:359–365.
375. Wang RY, Chan RC, Tsai MW. Effects of thoraco-
lumbar electric sensory stimulation on knee exten-
sor spasticity of persons who survived cerebrovas-
cular accident (CVA). J Rehabil Res Dev.
2000;37(1):73–79.
376. Cauraugh J, Light K, Sangbum K, Thigpen M,
Behrman A. Chronic motor dysfunction after
stroke. Stroke. 2000;31:1360–1364.
110 TOPICS IN STROKE REHABILITATION/SPRING 2006
377. Chae J, Hart R. Comparison of discomfort associ-
ated with surface and percutaneous intramuscular
electrical stimulation for persons with chronic
hemiplegia. Am J Phys Med Rehabil. 1998;77(6):
516–522.
378. Daly JJ, Ruff RL. Electrically induced recovery of
gait components for older patients with chronic
stroke. Am J Phys Med Rehabil. 2000;79:349–360.
379. Dimitrijevic MM, Soroker N. Mesh-glove. 2.
modulation of residual upper limb motor control
after stroke with whole hand electric stimulation.
Scan J Rehabil Med. 1994;26:187–190.
380. Hines AE, Crago PE, Billian C. Functional electrical
stimulation for the reduction of spasticity in the
hemiplegic hand. ISA. 1993;0067-8856:259–266.
381. Popovic MB, Popovic DB, Sinkjaer T, Stefanovic A,
Schwirtlich L. Restitution of reaching and grasping
promoted by functional electrical therapy. Artificial
Organs. 2002;26(3):271–275.
382. Teng EL, McNeal DR, Kralj A, Waters RL. Electrical
stimulation and feedback training: effects on the
voluntary control of paretic muscles. Arch Phys
Med Rehabil. 1976;57:228.
383. Yu DT, Chae J, Walker ME, Fang ZP. Percutaneous
intramuscular neuromuscular electric stimulation
for the treatment of shoulder subluxation and pain
in patients with chronic hemiplegia: a pilot study.
Arch Phys Med Rehabil. 2001;82:20–25.
384. Chae J, Bethoux F, Dobos L, Davis T, Friedl A.
Neuromuscular stimulation for upper extremity
motor and functional recovery in acute hemiple-
gia. Stroke. 1998;29:975–979.
385. Freed ML, Freed L, Chatburn RL, Christian M.
Electrical stimulation for swallowing disorders
caused by stroke. Respiratory Care. 2001;46(5):
466–474.
386. Lagassé PP, Roy MA. Functional electrical stimula-
tion and the reduction of co-contraction in spastic
biceps brachii. Clin Rehabil. 1989;3:111–116.
387. Linn SL, Granat MH, Lees KR. Electrical stimulation
prevents subluxation and associated pain when
applied soon after stroke. Aust J Physiother.
2000;46:240–241.
388. Petersen T, Klemar B. Electrical stimulation as a
treatment of lower limb spasticity. J Neuro Rehabil.
1988;2(3):103–108.
389. Waters RL, McNeal DR, Faloon W, Clifford B. Func-
tional electrical stimulation of the peroneal nerve
for hemiplegia. J Bone Joint Surg. 1985;67-
A(5):792–793.
TENS
390. Levin MF, Hui-Chan CWV. Relief of hemiparetic
spasticity by TENS is associated with improvement
in reflex and voluntary motor functions.
Electroencephalogr Clin Neurophysiol.
1992;85(2):131–142.
391. Sonde L, Gip C, Fernaeus SE, Nilsson CG, Viitanen
M. Stimulation with low frequency (1.7 HZ) trans-
cutaneous electric nerve stimulation (low-TENS)
increases motor function of the post-stroke paretic
arm. Scand J Rehabil Med. 1998;30(2):95–99.
392. Tekeoolu Y, Adak B, Goksoy T. Effects of transcuta-
neous electrical nerve stimulation (TENS) on
Barthel Activities of Daily Living (ADL) Index score
following stroke. Clin Rehabil. 1998;12(4):277–
280.
393. Peurala SH, Pitkanen K, Sivenius J, Tarkka IM. Cu-
taneous electrical stimulation may enhance sen-
sorimotor recovery in chronic stroke. Clin Rehabil.
2002;16(7):709–716.
394. Hui-Chan CWY, Levin MF. Stretch reflex latencies
in spastic hemiparetic subjects are prolonged after
transcutaneous electrical nerve stimulation. Le
Journal Canadien des Sciences Neurologiques.
1993;20:97–106.
395. Lafosse C, Kerckhofs E, Troch M, Vandenbussche.
Upper limb exteroceptive somatosensory and
proprioceptive sensory afferent modulation of
hemispatial neglect. J Clin Exp Neuropsychol.
2003;25(3):308–323
396. Johansson BB, Haker E, von Arbin M, et al. Acu-
puncture and transcutaneous nerve stimulation in
stroke rehabilitation. Stroke. 2001;32(3):707–713.
397. Sonde L, Bronge L, Kalimo H, Viitanen M. Can the
site of brain lesion predict improved motor func-
tion after low-TENS treatment on the post-stroke
paretic arm. Clin Rehabil. 2000;15(5):545–551.
398. Perennou DA, Leblond C, Amblard B, Micallef JP,
Herisson C, Pelissier Y. Transcutaneous electric
nerve stimulation reduces neglect related postural
instability after stroke. Arch Phys Med Rehabil.
2001;82:440–448.
Ultrasound
399. Inaba MK, Piorkowski M. Ultrasound in treatment
of painful shoulders in patients with hemiplegia.
Phys Ther. 1972;52(7):737–741.
400. Bansil CK, Joshi JB. Effectiveness of shortwave di-
athermy and ultrasound in treatment of osteo-
arthritis of the knee joint. Med J Zambia.
1975;9(5):138–139.
401. Falconer J, Hayes KW, Chang RW. Effect of ultra-
sound on mobility in osteoarthritis of the knee.
Arthritis Care Res. 1992;5(1):29–35.
402. Lehmann JF, Erickson DJ, Martin GM, Krusen FH.
Comparison of ultrasonic and microwave di-
athermy in the physical treatment of periarthritis
of the shoulder. Arch Phys Med Rehabil.
1954;35(10):627–634
Acupuncture
403. Fink M, Rollnik JD, Bijak M, et al. Needle acupunc-
ture in chronic poststroke leg spasticity. Arch Phys
Med Rehabil. 2004;85:667–672.
404. Gosman-Hedstrom G, Claesson L, Klingenstierna
U, et al. Effects of acupuncture treatment on daily
life activities and quality of life. Stroke. 1998;
29:2100–2108.

405. Johansson K, Lindgren I, Widner H, Wiklund I,
Johansson BB. Can sensory stimulation improve
the functional outcome in stroke patients? Neurol-
ogy. 1993;43:2189–2192.
406. Kjendahl A, Sällström S, Osten Egil P, Stanghelle
Kvalvik J, Borchgrevink CF. A one year follow-up
study on the effects of acupuncture in the treat-
ment of stroke patients in the subacute stage: a
randomized, controlled study. Clin Rehabil.
1997;11:192–200.
407. Moon SK, Whang YK, Park SU, Ko CN, Kim YS, Bae
HS, Cho KH. Antispastic effect of
electroacupuncture and moxibustion in stroke pa-
tients. Am J Chin Med. 2003;31(3):467–474.
408. Naeser M, Alexander MP, Stiassny-Eder D, Galler
V, Hobbs J, Bachman D. Real versus sham acu-
puncture treatment of paralysis in acute stroke
patients: a CT scan lesion site study. J Neuro
Rehabil. 1992;6:163–173.
409. Wong Alice MK, Su T-Y, Tang F-T, Liaw M-Y. Clini-
cal trial of electrical acupuncture on hemiplegic
stroke patients. Am J Phys Med Rehabil.
1999;78:117–122.
410. Chen CH, Chen TW, Weng MC, Wang WT, Wang
YL, Huang MH. The effect of electroacupuncture
on shoulder subluxation for stroke patients.
Kaohsiung J Med Sci. 2000;16(10):525–532.
411. Hopwood V, Lewith G. The effect of acupuncture
on the motor recovery of the upper limb after
stroke. Physiotherapy. 1997;83(12):614–619.
412. Jianfei C, Meifang Y, Jia W. Hemorrheological
study on the effect of acupuncture in treating
cerebral infarction. J Trad Chin Med. 1988;8(3):
167–172.
413. Johansson BB, Haker E, von Arbin M, et al. Acu-
puncture and transcutaneous nerve stimulation in
stroke rehabilitation. A randomized, controlled
trial. Stroke. 2001;32:707–713.
414. Johansson K, Lindgren I, Widner H, Wiklund I,
Johansson BB. Can sensory stimulation improve
the functional outcome in stroke patients? Neurol-
ogy. 1993;43:2189–2192.
415. Liyi Q, Zhonghui Z, Yuping L, Bangquan L, Guilan
Z, Shixi H, San W. Acupuncture treatment of cere-
brovascular occlusion and changes in
hemorrheological indices during treatment; a
clinical analysis of 100 cases. J Trad Chin Med.
1986;6(2):105–110.
416. Magnusson M, Johansson K, Johansson BB. Sensory
stimulation promotes normalization of postural
control after stroke. Stroke. 2002;25:186–194.
417. Park J, Hopwood V, White AR, Ernst E. Effectiveness
of acupuncture for stroke: a systematic review. J
Neurol. 2001;248:558–563.
418. Sällström S, Kjendahl A, Osten PE, Stanghelle JK.
Acupuncture in the treatment of stroke patients in
the subacute stage. Chin J Rehabil Theory Pract.
1995;1:5–9.
419. Sallstrom S, Kjendahl A, Osten PE, Stanghelle JH,
Borchgrevink CF. Acupuncture in the treatment of
stroke patients in the subacute stage: a random-
ized, controlled study. Complement Ther Med.
1996; 4:193–197.
Clinical Practice Guidelines 111
420. Smith GV, Forrester LW, Silver KHC, Macko RF.
Effects of treadmill training on translational bal-
ance perturbation responses in chronic
hemiparetic stroke patients. J Stroke Cerebrovasc
Dis. 2000;9(5):238–245
421. Sze K-H F, Wong E, Xiang Y, Woo J. Does acupunc-
ture have additional value to standard poststroke
motor rehabilitation? Stroke. 2002;33:186–194.
Intensity and organization of rehabilitation
422. Ahrens J. Italian study concludes home hospitaliza-
tion benefits stroke patients. Caring.
2004;23(8):40–42.
423. Andersen HE, Schultz-Larsen K, Kreiner S,
Forchhammer BH, Eriksen K, Brown A. Can read-
mission after stroke be prevented? Results of a
randomized clinical study: a postdischarge follow-
up service for stroke survivors. Stroke.
2000;31:1038–1045.
424. Andersen HE, Eriksen K, Brown A, Schultz-Larsen
K, Forchhammer BH. Follow-up services for stroke
surviviors after hospital discharge- a randomized
control study. Clin Rehabil. 2002;16:593–600.
425. Andersson A, Levin LA, Oberg B, Mansson L.
Health care and social welfare costs in home-
based and hospital-based rehabilitation after
stroke. Scand J Caring Sci. 2002;16:386–392.
426. Askim T, Rohweder G, Lydersen S, Indreavik B.
Evaluation of an extended stroke unit service with
early supported discharge for patients living in a
rural community. A randomized controlled trial.
Clin Rehabil. 2004;8:238–248.
427. Bautz-Holter E, Sveen U, Rygh J, Rodgers H, Bruun
Wyller T. Early supported discharge of patients
with acute stroke: a randomized controlled trial.
Disabil Rehabil. 2002;24(7):348–355.
428. Baskett JJ, Broad JB, Reekie G, Hocking C, Green G.
Shared responsibility for ongoing rehabilitation: a
new approach to home-based therapy after
stroke. Clin Rehabil. 1999;13:23–33.
429. Claesson L, Gosman-Hedström G, Johannesson M,
Fagerberg B, Blomstrand C. Resource utilization
and costs of stroke unit care intergrated in a care
continuum: a 1-year controlled, prospective, ran-
domized study in elderly patients. Stroke.
2000;31:2569–2577.
430. Corr S, Bayer A. Occupational therapy for stroke
patients after hospital discharge – a randomized
controlled trial. Clin Rehabil. 1995;9:291–296.
431. Drummond AER, Miller N, Colquohoun M, Logan
PC. The effect of a stroke unit on activities of daily
living. Clin Rehabil. 1996;10:12–22.
432. Drummond AER, Walker MF. A randomized con-
trolled trial of leisure rehabilitation after stroke.
Clin Rehabil. 1995;9:283–290.
433. Drummond A, Walker M. Generalisation of the
effects of leisure rehabilitation for stroke patients.
Br J Occup Ther. 1996;59(7):330–334.
434. Evans A, Harraf F, Donaldson N, Kalra L. Random-
ized controlled study of stroke unit care versus
stroke team care in different stroke subtypes.
112 TOPICS IN STROKE REHABILITATION/SPRING 2006
Stroke. 2002;33:449–455.
435. Falconer JA, Roth EJ, Sutin JA, Strasser DC, Chang
RW. The critical path method in stroke rehabilita-
tion: lessons from an experiment in cost contain-
ment and outcome improvement. QRB. 1993:8-
16.
436. Fang Y, Chen X, Li H, Lin J, Huang R, Zeng J. A
study on additional early physiotherapy after
stroke and factors affecting functional recovery.
Clin Rehabil. 2003;17:608–617.
437. Feigenson JS, Gitlow HS, Greenberg SD. The disabil-
ity oriented rehabilitation unit – A major factor influ-
encing stroke outcome. Stroke. 1979;10(1):5–7.
438. Feldman DJ, Lee PR, Unterecker J, Lloyd K, Rusk
HA, Toole A. A comparison of functionally ori-
ented medical care and formal rehabilitation in
the management of patients with hemiplegia due
to cerebrovascular disease. J Chronic Dis.
1961;15:297–310.
439. Fjaertoft H, Indredavik B, Lydersen S. Stroke unit
care combined with early supported discharge:
long-term follow-up of a randomized controlled
trial. Stroke. 2003:34:2687–2692.
440. Gibson JW, Schkade JK. Occupational adaptation
intervention with patients with cerebrovascular
accident: a clinical study. Am J Occup Ther.
1997;51(7):523–529.
441. Gilbertson L, Langhorne P, Walker A, Allen A,
Murray GD. Domiciliary occupational therapy for
patients with stroke discharged from hospital: a
randomized controlled trial. Br Med J.
2000;320:603–606.
442. Gürsel Y, Dinçer G, Dalyan M, Arasil T. Rehabilita-
tion outcome in stroke patients in Turkey— an
observational study of treated and untreated pa-
tients. Eur J Phys Med Rehabil. 1998;8(3):87–90.
443. Hamrin E. One year after stroke: a follow-up of an
experimental study. Scand J Rehabil Med.
1982;14:111–116.
444. Hayes SH, Carroll SR. Early intervention care in the
acute stroke patient. Arch Phys Med Rehabil.
1986;67:319-321.
445. Holmqvist WL, von Koch L, Kostulas V, et al. A
randomized controlled trial of rehabilitation at
home after stroke in southwest Stokholm. Stroke.
1998;29:591–597.
446. Indredavik B, Fjaertoft H, Ekeberg G, Loge AD,
Morch B. Benefit of an extended stroke unit ser-
vice with early supported discharge: a random-
ized, controlled trial. Stroke. 2000:31(12):2989–
2994.
447. Jongbloed L, Morgan D. An investigation of in-
volvement in leisure activities after a stroke. Am J
Occup Ther. 1991;45(5):420–427.
448. Kalra L, Eade J. Role of stroke rehabilitation units in
managing severe disability after stroke. Stroke.
1995;26(11):2031–2034.
449. Kalra L, Evans A, Perez I, Knapp M, Donaldson N,
Swift CG. Alternative strategies for stroke care: a
prospective randomized controlled trial. Lancet.
2000;356:894–899.
450. Krespi Y, Gurol ME, Coban O, Tuncay R, Bahar S.
Stroke unit versus neurology ward: a before and
after study. J Neurol. 2003;250:1363–1369.
451. Lin JH, Hshieh CL, Kai Lo S, Chai HM, Liao LR.
Preliminary study of the effect of low-intensity
home-based physical therapy in chronic stroke
patients. Kaohsiung J Med Sci. 2004;20(1):18–22.
452. Lincoln NB, Willis BA, Philips SA, Judy LC, Berman
P. Comparison of rehabilitation practice on hospi-
tal wards for stroke patients. Stroke. 1996;27:18–
23.
453. Lincoln NB, Parry RH, Vass CD. Randomized, con-
trolled trial to evaluate increased intensity of phys-
iotherapy treatment of arm function after stroke.
Stroke. 1999;30:573–579.
454. Logan PA, Ahern J, Gladman JRF, Lincoln NB. A
randomized controlled trial of enhanced social ser-
vice occupational therapy for stroke patients. Clin
Rehabil. 1997;11:107–113.
455. Mayo NE, Wood-Dauphinee S, Cote R, Gayton D,
Carlton J, Buttery J, Tamblyn R. There’s no place
like home: an evaluation of early supported dis-
charge for stroke. Stroke. 2000;31:1016–1023.
456. Nir Z, Zolotogorsky Z, Sugarman H. Structured
nursing intervention versus routine rehabilitation
after stroke. Am J Phys Med Rehabil.
2004;83(7):522–529.
457. Parry RH, Lincoln NB, Vass CD. Effect of severity of
arm impairment on response to additional physio-
therapy early after stroke. Clin Rehabil.
1999;13:187–198.
458. Partridge C, Mackenzie M, Edwards S, Reid A,
Jayawardena S, Guck N, Potter J. Is dosage of
physiotherapy a critical factor in deciding of recov-
ery from stroke: a pragmatic randomized con-
trolled trial. Physiother Res Int. 2000;5(4):230–
240.
459. Rodgers H, Mackintosh J, Price C, Wood R,
McNamee P, Fearon T, Marritt A, Curless R. Does
an early increased-intensity interdisciplinary upper
limb therapy programme following acute stroke
improve outcome? Clin Rehabil. 2003;17:579–
589.
460. Rodgers H, Soutter J, Kaiser W, Pearson P, Dobson
R, Skilbeck C, Bond J. Early supported hospital
discharge following acute stroke: pilot study re-
sults. Clin Rehabil. 1997;11:280–287.
461. Rudd AG, Wolfe CDA, Tilling K, Beech R. Random-
ized controlled trial to evaluate early discharge
scheme for patients with stroke. Br Med J.
1997;315:1039–1044.
462. Ruff RM, Yarnell S, Marinos JM. Are stroke patients
discharges sooner if in-patient rehabilitation ser-
vices are provided seven vs six days per week? Am
J Med Rehabil. 1999;78(2):143–146.
463. Sivenius J, Pyörälä K, Heinonen OP, Salonen JT,
Riekkinen P. The significance of intensity of reha-
bilitation of stroke – a controlled trial. Stroke.
1985;16(6):928–931.
464. Smith DS, Goldenberg E, Ashburn A, et al. Reme-
dial therapy after stroke: a randomized controlled
trial. Br Med J. 1981;282:517–520.
465. Smith ME, Garraway WM, Smith DL, Akhtar AJ.
Therapy impact on functional outcome in a con-
trolled trial of stroke rehabilitation. Arch Phys Med

Rehabil. 1982;63:21–24.
466. Sonada S, Saitoh E, Nagai S, Kawakita M, Kanada
Y. Full-time integrated treatment program, a new
system for stroke rehabilitation in Japan: compari-
son with conventional rehabilitation. Am J Phys
Med Rehabil. 2004;83:88–83.
467. Sulch D, Melbourn A, Perez I, Kalra L. Integrated
care pathways and quality of life on a stroke reha-
bilitation unit. Stroke. 2002;33:1600–1604.
468. Sunderland A, Tinson DJ, Bradley EL, Fletcher D,
Hewer RL, Wade DT. Enhanced physical therapy
improves recocery of arm function after stroke. A
randomized trial. J Neurol Neurosurg Psychiatry.
1992;55:530–535.
469. Teng J, Mayo NE, Latimer E, Hanley J, Wood-
Dauphinee S, Côté R, Scott S. Costs and caregiver
consequences of early supported discharge for
stroke patients. Stroke. 2003;34:528–536.
470. Turton A, Fraser C. The use of home therapy pro-
grams for improving recovery of the upper limb
following stroke. Br J Occup Ther. 1990;53(11):
458–463.
471. Walker MF, Gladman JRF, Lincoln NB, Siemonsma
P, Whiteley.T. Occupational therapy for stroke pa-
tients not admitted to hospital: a randomized con-
trolled trial. Lancet. 1999;354:278–280.
472. Wall JC, Turnbull GI. Evaluation of out-patient
physiotherapy and a home exercise program in
the management of gait asymmetry in residual
stroke. J Neurol Rehabil. 1987;1(3):115–123.
473. Wikander B, Ekelund P, Milsom I. An evaluation of
multidisciplinary intervention governed by func-
tional independence measure (FIM) in incontinent
stroke patients. Scand J Rehabil Med. 1998;30:15–21.
474. Young J, Forster A. Day hospital and home physio-
therapy for stroke patients: a comparative cost-
effectiveness study. J Royal Coll Physicians Lond.
1993;27(3):252–258.
475. Aitken PD, Rodgers H, French JM, Bates D, James
OFW. General medical or geriatric unit care for
stroke? A controlled trial. Age Ageing.
1993;22(2):4–5.
476. Goldberg G, Segal ME, Berk SN, Schall RR,
Gershkoff AM. Stroke transition after inpatient re-
habilitation. Top Stroke Rehabil. 1997;4(1):64–79.
477. Gompertz P, Pound P, Briffa J, Ebrahim S. How
useful are non-random comparisons of outcomes
and quality of care in purchasing hospital stroke
services. Age Ageing. 1995;24:137–141.
478. Gordon N, Gulanick M, Costa F, et al. Physical
activity and exercise recommendations for stroke
survivors. Stroke. 2004;35:1229–1239
479. Hamrin EKF, Lindmark B. The effect of systemic care
planning after stroke in general hospital medical
wards. J Adv Nursing. 1990;15:1146–1153.
480. Indredavik B, Bakke F, Slordahl SA. Stroke unit
treatment 10-years follow-up. Stroke.
1999;30:1524–1527.
481. Lincoln NB, Walker MF, Dixon A, Knights P. Evalu-
ation of a multiprofessional community stroke
team: a randomized controlled trial. Clin Rehabil.
2004;18:40–47.
482. McCann BC, Culbertson RA. Comparison of two
Clinical Practice Guidelines 113
system for stroke rehabilitation in a general hospi-
tal. J Am Geriatr Soc. 1976;24(5):211–216.
483. McNamee P, Christensen J, Soutter J, Rodgers H,
Craig N, Pearson P, Bond J. Cost analysis of early
supported hospital discharge for stroke. Age Age-
ing. 1998;27:345–351.
484. Reding MJ, McDowell FH. Focused stroke rehabili-
tation programs improve outcome. Arch Neurol.
1989;46:700–701.
485. Ricauda NA, Bo M, Molaschi M, Massaia M,
Salerno D, Amati D, Tibaldi V, Fabris F. Home
hospitalization service for acute uncomplicated
first ischemic stroke in eldery patients: a random-
ized trial. J Am Geriatr Soc. 2004:52:278–283.
486. Ronning OM, Guldvog B. Stroke unit versus gen-
eral medicine wards, II: Neurological deficits and
activities of daily living; a quasi-randomized con-
trolled trial. Stroke. 1998;29:586–590.
487. Sulch D, Perez I, Melbourn A, Kalra L. Randomized
controlled trial of intergrated (managed) care
pathway for stroke rehabilitation. Stroke.
2000;31:1929–1934.
488. Sulch D, Evans E, Melbourn A, Kalra L. Does an
integrated care pathway improve processes of
care in stroke rehabilitation? A randomized con-
trolled trial. Age Ageing. 2002;31:175–179.
489. Summers D, Soper PA. Implementation and evalu-
ation of stroke clinical pathways and the impact
on cost of stroke care. J Cardiovasc Nursing.
1998;13:69–87.
490. Wolcott LE, Wheeler PC, Ballard P, Crumb CK,
Miles G, Mueller A. Home-care vs. institutional
rehabilitation of stroke: a comparative study.
Rehabil Stroke. 1966;63(9):722–724.
491. Corr S, Phillips CJ, Walker M. Evaluation of a pilot
service designed to provide support following
stroke: a randomized cross-over design study. Clin
Rehabil. 2004;18:69–75.
492. Garraway WM, Akhtar AJ, Prescott RJ, Hockey L.
Management of acute stroke in the elderly: pre-
liminary results of a controlled trial. Br Med J.
1980;281:827–829.
493. Gladman JRF, Lincoln NB. Follow-up of a con-
trolled trial of domiciliary stroke rehabilitation
(DOMINO study). Age Ageing. 1994;23:9–13.
494. Granger CV, Greer DS, Liset E, Coulombe J,
O’Brien E. Measurement of outcomes of care for
stroke patients. Stroke. 1975;6:34–41.
495. Hamrin E. Early activation in stroke: does it make a
difference? Scan J Rehabil Med. 1982;14:101–109.
496. Indredavik B, Slordahl SA, Bakke F, Rokseth R,
Hahein LL. Stroke unit treatment. Stroke.
1997;28:1861–1866.
497. Indredavik B, Bakke F, Slordahl SA, Rokset R,
Haheim LL. Stroke unit treatment improves long-
term quality of life; a randomized controlled trial.
Stroke. 1998;29:895–899.
498. Lindmark B, Hamrin E. Instrumental activities of
daily living in two patient populations, three
months and one year after a stroke. Scand J Caring
Sci. 1989;3(4):161–168.
499. Parry RH, Lincoln NB, Appleyard MA. Physio-
therapy for the arm and hand after stroke. Physio-
114 TOPICS IN STROKE REHABILITATION/SPRING 2006
therapy. 1999;85(8):417–425.
500. Ricauda NA, Fiorio Pla L, Marinello R, Molaschi M,
Fabris F. Feasibility of an acute stroke home care
service for elderly patients. Arch Gerontol Geriatry.
1998;6:17–22.
501. Ronning OM, Guldvog B. Outcome of subacute
stroke rehabilitation: a randomized controlled
trial. Stroke. 1998;29:779–784.
502. Stevens RS, Ambler NR, Warren MD. A random-
ized controlled trial of a stroke rehabilitation ward.
Age Ageing. 1984;13:65–75.
503. Sunderland A, Fletcher D, Bradley L, Tinson D,
Hewer RL, Wade DT. Enhanced physical therapy
for arm function after stroke: a one year follow up
study. J Neurol Neurosurg Psychiatry. 1994;57:
856–858.
504. Wade DT, Langton-Hewer R, Skilbeck CE, Bainton
D, Burns-Cox C. Controlled trial of a home-care
service for acute stroke patients. Lancet.
1985;1(8424):323–326.
505. Wolfe CD, Tilling K, Rudd AG. The effectiveness of
community-based rehabilitation for stroke pa-
tients who remain at home: a pilot randomized
trial. Clin Rehabil. 2000;14:563–569.
506. Eagle DJ, Guyatt GH, Patterson C, Turpie I, Sackett
B, Singer J. Effectiveness of a geriatric day hospital.
Can Med Assoc J. 1991;144(6):699–704.
507. Evans RL, Hendricks RD. Comparison of subacute
rehabilitative care with outpatient primary medi-
cal care. Disabil Rehabil. 2001;23(12):531–538.
508. Hui E, Lum CM, Woo J, Or KH, Kay RLC. Outcome
of elderly stroke patients, day hospital versus con-
ventional medical management. Stroke.
1995;26:1616–1619.
509. Kalra L, Dale P, Crome P. Improving stroke reha-
bilitation: a controlled study. Stroke. 1993;24:
1462–1467.
510. Moseley BC. Rehabilitation effectiveness among
long term nursing home stroke residents. Phys
Occup Ther Geriatr. 1996;14(4):27–41.
511. Strand T, Asplund K, Eriksson S, Hägg E, Lithner F,
Wester PO. A non-intensive stroke unit reduces
functional disability and the need for long-term
hospitalization. Stroke. 1985;16(1):29–34.
512. Tucker MA, Davison JG, Ogle SJ. Day hospital
rehabilitation—effectiveness and cost in the eld-
erly: a randomised controlled trial. Br Med J.
1984;289:1209–1212.
513. Von Arbin M, Britton M, de Faire U, Helmers C,
Miah K, Murray U. A study of stroke patinets
treated in a non-intensive stroke unit or in general
wards. Acta Medica Scandinavica. 1980;208:81–
85.
514. Wade DT, Collen FM, Robb GF, Warlow CP. Phys-
iotherapy intervention late after stroke and mobil-
ity. Br Med J. 1992;304:609–613.
515. Anderson C , Mhurchu CN, Rubenach S, Clank M,
Spencer C, Winsor A. Home or hospital for stroke
rehabilitation? Results of a randomized controlled
trial: II: cost minimization analysis at 6 months.
Stroke. 2000;31(5):1032–1037.
516. Glader EL, Stegmayr B, Johansson L, Hulter-Asberg
K, Wester PO. Differences in long-term outcome
between patients treated in stroke units and in
general wards. Stroke. 2001;32:2124–2130.
517. Holmqvist LW, von Koch L, Kostulas V, Holm M,
Widsell G, Tegler H, Johansson K, Almazan J, de
Pdro-Cuesta J. A randomized controlled trial of
rehabilitation at home after stroke in southwest
Stockholm. Stroke. 1998;29:591–597.
518. Jorgensen HS, Nakayama H, Raaschou HO, Larsen
K, Hubbe P, Olsen TS. The effect of a stroke unit:
reduction in mortality, discharge rate to nursing
home, length of hospital stay, and cost: a commu-
nity-based study. Stroke. 1995;26:1178–1182.
519. Özdemir F, Birtane M, Tabatabaei R, Kokino S,
Ekuklu G. Comparing stroke rehabilitation out-
comes between acute inpatient and non intense
home settings. Arch Phys Med Rehabil.
2001;82:1375–1379.
520. Roderick P, Low J, Day R, Peasgood T, Mullee MA,
Turnbull JC, Villar T, Raftery. Stroke rehabilitation
after hospital discharge: a randomized trial com-
paring domiciliary and day-hospital care. Age Age-
ing. 2001;30:303–310.
521. Von Koch L, Widen Holmqvist L, Kostulas V,
Almazan J, de Pedro-Cuesta J. A randomized con-
trolled trial of rehabilitation at home after stroke in
southwest Stockholm: outcome at six months.
Scand J Rehabil Med. 2000;32:80–86.
522. Wood-Dauphinee S, Shapiro S, Bass E, Fletcher C,
Georges P, Hensby V, Mendelsohn B. A random-
ized trial of team care following stroke. Stroke.
1984;15(5):864–872.
523. Young JB, Forster A. The Bradford community
stroke trial: eight week results. Clin Rehabil.
1991;5:283–292.
524. Antonucci G, Guariglia C, Judica A, Magnotti L,
Paolucci S, Pizzamiglio L, Zoccolotti P. Effective-
ness of neglect rehabilitation in a randomized
group study. J Clin Exp Neuropsychol. 1995;17(3):
383–389.
525. Brocklehurst JC, Andrews K, Richards B, Laycock
PJ. How much physical therapy for patients with
stroke. Br Med J. 1978;1:1307–1310.
526. Bryant NH, Candland L, Loewenstein R. Compari-
son of care and cost outcomes for stroke patients
with and without home care. Stroke. 1974;5:54–
59.
527. Burch S, Longbottom J, McKay M, Borland C. A
randomized controlled trial of day hospital and
day center therapy. Clin Rehabil. 1999;13:105–
112.
528. De Weerdt W, Nuyens G, Feys H, et al. Group
physiotherapy improves time use by patients with
stroke in rehabilitation. Austr J Physiother.
2001;47:53–61.
529. Donald IP, Baldwin RN, Bannerjee M. Gloucester
hospital-at-home: a randomized controlled trial.
Age Ageing. 1995;24:434–439.
530. Fagerberg B, Claesson L, Gosman-Hedström G,
Blomstrand C. Effect of acute stroke unit care inte-
grated with care continuum versus conventional
treatment: a randomized 1-year study of elderly
patients. Stroke. 2000;31:2578–2584.
531. Friedman PJ. Stroke outcome in elderly people

living alone. Disabil Rehabil. 1995;17(2):90–93.
532. Gladman JRF. A randomized controlled trial of
domiciliary stroke rehabilitation: results at six and
12 months. Clin Rehabil. 1992;6:13–14.
533. Gladman JRF, Lincoln NB, Barer DH. A randomized
controlled trial of the domiciliary and hospital-
based rehabilitation for stroke patients after dis-
charge from hospital. J Neurol Neurosurg Psychia-
try. 1993;56:960–966.
534. Gladman J, Whynes D, Lincoln N. Cost compari-
son of domiciliary and hospital-based stroke reha-
bilitation. Age Ageing. 1994;23:241–245.
535. Gladman J, Forster A, Young J. Hospital-and-
home-based rehabilitation after discharge from
hospital for stroke patients: analysis of two trials.
Age Ageing. 1995;24(1):49–53.
536. Gosman- Hedström G, Claesson L, Blomstrand C.
Assistive devices in elderly people after stroke: a
longitudinal, randomized study—The Göteborg
70+ Stroke Study. Scand J Occupl Ther.
2002;9:109–118.
537. Gosman- Hedström G, Claesson L, Blomstrand C,
Fagerberg B, Lundgren-Lindquist B. Use and cost
of assistive technology the first year after stroke.
Int J Technol Assess Health Care. 2002;18(3):520–
527.
538. Hackett M, Anderson C, Vandal A, Rubenach S, et
al. One year follow-up of a randomized controlled
trial of accelerated hospital discharge and home-
based stroke rehabilitation. Stroke. 2000;31:2817–
2818.
539. Hamilton-Bruce MA, Black AB, Rowe CR, Lee R,
Florance JM, Chivell J, Thomas EM, Massy-
Westropp MG, Tomlinson L. Home stroke – early
discharge for stroke patients. J Clin Neurosci.
1999;6(3):284–285.
540. Holmqvist LW, Pedro-Cuesta J, Holm M, Kostulas
V. Intervention design for rehabilitation at home
after stroke: a pilot feasibility study. Scand J Rehabil
Med. 1995;27(1):43–50.
541. Indredavik B, Bakke F, Solberg R, Rokseth R,
Haaheim LL, Holme I. Benefit of a stroke unit: a
randomized controlled trial. Stroke. 1991;22:
1026–1031.
542. Indredavik B, Slordahl SA, Bakke F, Rokseth R,
Haaheim LL Stroke unit care improves long-term
survival and function. Cardiol Rev. 1999;16(1):24–
27.
543. Indredavik B, Bakke F, Slordahl SA, Rokseth R,
Haheim LL. Treatment in a combined acute and
rehabilitation stroke unit: which aspects are most
important? Stroke. 1999;30:917–923.
544. Jorgensen HS, Kammersgaard LP, Nakayama H,
Raaschou HO, Larsen K, Hubbe P, Olsen TS. Treat-
ment and rehabilitation on a stroke unit improves
5-year survival. Stroke. 1999;30(5):930–933.
545. Jorgensen Hs, Kammersgaard LP, Houth J,
Nakayam H, Raaschou HO, Larsen K, Hubbe P,
Olsen TS. Who benefits from treatment and reha-
bilitation in a stroke unit? Stroke. 2000;31(2):434–
439.
546. Juby LC, Lincoln NB, Berman P, Drummond A,
Miller N, Colquhoum M, Clarke P. The effect of a
Clinical Practice Guidelines 115
stroke rehabilitation unit on functional outcome
and psychological outcome: a randomized con-
trolled trial. Cerebrovas Dis. 1996;6(2):106–110.
547. Kalra L. The influence of stroke unit rehabilitation
on functional recovery from stroke. Stroke.
1994;25:821–825.
548. Kalra L, Dale P, Crome P. Stroke rehabilitation
units – do elderly stroke patients benefit?
Cerebrovasc Dis. 1994;4:146–151.
549. Keith RA, Wilson DB, Gutierrez P. Acute and sub-
acute rehabilitation for stroke: a comparison. Arch
Phys Med Rehabil. 1995;76:495–500.
550. Landi F, Gambassi G, Cocchi A, Bernabei R.
Therapy for stroke patients living at home; outpa-
tient therapy trialists. Lancet. 1999;354:1731.
551. Langhorne P, Williams BO, Gilchrist W, Howie K.
Do stroke units save lives? Lancet. 1993;342:395–
398.
552. Lincoln NB, Drummond AER, Berman P. Percep-
tual impairment and its impact on rehabilitation
outcome. Disabil Rehabil. 1997;19(6):231–234.
553. Lincoln NB, Husbands S, Trescoli C, Drummond
AER, Gladman JRF, Bertman P. Five year follow up
a randomized controlled trial of a stroke rehabilita-
tion unit. Br Med J. 2000;320:549.
554. Löfgren B, Nyberg L, Mattsson M, Gustafson Y.
Three years after in-patient stroke rehabilitation: a
follow-up study. Cerebrovasc Dis. 1999;9(3):163–
170.
555. MacKay-Lyons MJ, Makrides L. Cardiovascular
stress during a contemporary stroke rehabilitation
program: Is the intensity adequate to induce a
training effect? Arch Phys Med Rehabil.
2002;83:1378–1383.
556. Mackey F, Ada L, Heard R, Adams R. Stroke reha-
bilitation: Are highly structured units more condu-
cive to physical activity than less structured units?
Arch Phys Med Rehabil. 1996;77:1066–1070.
557. Mant J, Carter J, Wade DT, Winner S. Family sup-
port for stroke: a randomized controlled trial. Lan-
cet. 2000;356:808–813.
558. Millikan CH. Stroke intensive care units: objectives
and results. Stroke. 1979;10(3):235–237.
559. Morten Ronning O, Guldvog B. Stroke units versus
general medical wards, I: Twelve- and eighteen-
month survival: a randomized controlled trial.
Stroke. 1998;29:58–62.
560. Ni CM, Fu J, Liu CY. Influence of early rehabilita-
tion on walking function of the patients with acute
stroke. Chin J Rehabil. 1997;12:164–165.
561. Oseley CB. Rehabilitation effectiveness among
long term nursing home stroke residents. Phys
Occup Ther Geriatr. 1996;14(4):27–40.
562. Paolucci S, Antonucci G, Grasso MG, Morelli D,
Troisi E, Coiro P, Bragoni M. Early versus delayed
inpatient stroke rehabilitation: a matched com-
parison conducted in Italy. Arch Phys Med Rehabil.
2000;81:695–700.
563. Patel M, Potter J, Perez I, Kalra L. The process of
rehabilitation and discharge planning in stroke: a
controlled comparison between stroke units.
Stroke. 1998;29:2484–2487.
564. Pitner SE, Mance CJ. An evaluation of stroke inten-
116 TOPICS IN STROKE REHABILITATION/SPRING 2006
sive care: results in a municipal hospital. Stroke.
1973;4:737–741.
565. Rubenach S, Anderson CS, Clark MS, Russell M,
Spencer C, Winsor A. Early supportive discharge
and rehabilitation trial (ESPRIT) in stroke. J Clin
Neurosci. 1999;6(3):277–287.
566. Stavem K, Ronning OM. Survival of unselected
stroke patients in a stroke unit compared with
conventional care. QJM. 2002;95:143–152.
567. Von Koch L, de Pedro-Cuesta J, Kostulas V,
Almazan J, Widen Holmqvist L. Randomized con-
trolled trial of rehabilitation at home after stroke:
one-year follow-up of patient outcome, resource
use and cost. Cerbrovasc Dis. 2001;12:131–138.
568. Webb DJ, Fayad PB, Wilbur C, Thomas A, Brass
LM. Effects of a specialized team on stroke care:
the first two years of the Yale Stroke Program.
Stroke. 1995;26:1353–1357.
569. Young JB, Forster A. The Bradford community
stroke trial: results at six months. Br Med J.
1992;304:1085–1089.
Discussion
570. Deyo RA, Walsh NE, Schoenfeld LS, Ramamurthy
S. Can trials of physical treatments be blinded: the
example of transcutaneous electrical nerve stimu-
lation for chronic pain. Am J Phys Med Rehabil.
1990;69:6–10.
571. Meek C, Pollock A, Potter J, Langhorne P. A sys-
tematic review of exercise trials post stroke. Clin
Rehabil. 2003;17:6–13.
572. van der Lee JH. Constraint-induced therapy for
stroke: more of the same or something completely
different? Curr Opin Neurol. 2001;14(6):741–744.
573. Morris SL, Dodd KJ, Morris ME. Outcomes of pro-
gressive resistance strength training following
stroke: a systematic review. Clin Rehabil.
2004;18(1):27–39.
574. Clark S, Tremblay F, Ste-Marie D. Differential
modulation of corticospinal excitability during ob-
servation, mental imagery and imitation of hand
actions. Neuropsychologia. 2004;42(1):105–112.
575. Tremblay F, Tremblay LE, Colcer DE. Modulation
of corticospinal excitability during imagined knee
movements. J Rehabil Med. 2001;33(5):230–234.
576. Malouin F, Belleville J, Richards CL, Desrosiers J,
Doyon J. Working memory and mental practice
outcomes after stroke. Arch Phys Med Rehabil.
2004; 85:177–183.
577. Jackson PL, Lafleur MF, Malouin F, Richards C,
Doyon J. Potential role of mental practice using
motor imagery in neurological rehabilitation. Arch
Phys Med Rehabil. 2044;82:1133–1141.
578. Pomeroy VM, Clark CA, Millar SG, Baron JC,
Markus HS, Tallis RC. The potential for utilizing the
“mirror neurone system” to enhance recovery of
the severely affected upper limb early after stroke:
a review and hypothesis. Neurorehabil Neural Re-
pair. 2005;19(1):4–14.
579. Pomeroy V, Tallis R. Neurological rehabilitation: a
science struggling to come of age. Physiother Res
Int. 2002;7(2):76–89.
580. LeMura LM, von Duvillard SP. Clinical Exercise
Physiology. Philadelphia: Lippincott Williams &
Wilkins; 2004.
581. Barnes WS. The relationship of motor-unit activa-
tion to isokinetic muscular contraction at different
contractile velocities. Phys Ther. 1980;60(9):
1152–1158.
582. Macko RF, Ivey FM, Forrester LW. Task-oriented
aerobic exercise in chronic hemiparetic stroke:
training protocols and treatment effects. Top
Stroke Rehabil. 2005;12(1):45–57.
583. Bach-y-Rita P. Rehabilitation versus passive recov-
ery of motor control following general nervous
system lesions. In: Desmedt JE, ed. Motor Control
Mechanisms in Health and Disease. New York, NY:
Raven Press; 1983:1085–1092.
584. Taub E, Uswatte G, Pidikiti R. Constraint-induced
movement therapy: a new family of techniques
with broad application to physical rehabilitation: a
clinical review. J Rehabil Res Dev. 1999;36(3):237–
251.
585. Teasell R. Stroke recovery and rehabilitation.
Stroke. 2003;34(2):365–366.
586. Green J, Forster A, Bogle S, Young J. Physiotherapy
for patients with mobility problems more than 1
year after stroke: a randomised controlled trial.
Lancet. 2002;359:199–203.
587. Mulder T, Hochstenbach J. Adaptability and flex-
ibility of the human motor system: implications for
neurological rehabilitation. Neural Plasticity.
2001;8(1-2):131–140.
588. Sallis J, Owen N. Ecological models. In: Glanz K,
Lewis F, Rimer B, eds. Health Behaviour and Health
Education: Theory, Research, and Practice. San
Fransisco, CA: Jossey-Bass; 1997:403–424.
589. Mauritz KH. Gait training in hemiplegia. Eur J
Neurol. 2002;9(suppl. 1):23–29.
590. Barbeau H, Fung J. The role of rehabilitation in the
recovery of walking in the neurological popula-
tion. Curr Opin Neurol. 2001;14(6):735–740.
591. Thirumala P, Hier DB, Patel P. Motor recovery after
stroke: lessons from functional brain imaging.
Neurol Res. 2002;24:453–458.
592. Mayo, N. E., N. Korner-Bitensky, et al. Relationship
between response time and falls among stroke
patients undergoing physical rehabilitation. Int J
Rehabil Res. 1990;13(1):47–55.
593. Jorgensen L, Engstad T., Jacobsen BK. Higher inci-
dence of falls in long-term stroke survivors than
population controls: depressive symptoms predict
falls after stroke. Stroke. 2002;33:542–547.
594. Mackintosh SF, Goldie P, Hill K. Falls incidence and
factors associated with falling in older, commu-
nity-dwelling, chronic stroke survivors (> 1 year
after stroke) and matched controls. Aging Clin Exp
Res. 2005;17:74–81.
595. Harris JE, Eng JJ, Marigold DS, Tokuno CD, Louis
CL. Relationship of balance and mobility to fall
incidence in people with chronic stroke. Phys Ther.
2005;85(2):150–158.
596. Perell KL. Nelson A. Goldman RL. Luther SL. Prieto-
Lewis N. Rubenstein LZ. Fall risk assessment mea-

sures: an analytic review. J Gerontol Series A-Biol Sci
Med Sci. 2001;56(12):M761–766.
597. Perennou D, El Fatimi A, Masmoudi M, Benaim C,
Loigerot M, Didier JP, Pelissier J. Incidence, cir-
cumstances and consequences of falls in patients
undergoing rehabilitation after a first stroke.
Annales de Readaptation et de Medecine Physique.
2005;48(3):138–145.
598. Ada L, Foongchomcheay A, Canning C. Support-
ive devices for preventing and treating sublux-
ation of the shoulder after stroke [review].
Cochrane Database Syst Rev. 2005;(1):CD003863.
599. Griffin R, Reddin G. Shoulder pain in patients with
hemiplegia: a literature review. Phys Ther.
1981;61:1041–1045.
600. Binder-Macleod SA, Lee SCK. Assessment of the
efficacy of functional electrical stimulation in pa-
tients with hemiplegia. Top Stroke Rehabil.
1997;3(4):88–98.
601. McComas AJ, Sica REP, Upton ARM, Aguilera N.
Functional changes in motoneurones of
hemiparetic patients. J Neurol Neurosurg Psychia-
try. 1973;36:183–193.
602. Bélanger A. Evidence-Based Guide to Therapeutic
Physical Agents. Philadelphia, PA: Lippincott Will-
iams & Wilkins; 2002.
603. Morin M, Brosseau L. A theoretical framework on
low level laser therapy (classes I, II, III) application
for the treatment of OA and RA [abstract]. In:
Proceedings of the Canadian Physiotherapy Associa-
tion Annual Congress, May 31, 1996. Victoria, BC:
Canadian Physiotherapy Association; 1996.
604. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Con-
tribution of central neuroplasticity to pathological
pain: review of clinical and experimental evidence.
Pain. 1993;52:259–285.
605. Han JS, Chen XH, Sun SL, et al. Effect of low- and
high-frequency TENS on Met-enkephalin-Arg-Phe
and dynorphin A immunoreactivity in human lum-
bar CSF. Pain. 1991;47(3):295–298.
606. Ridding MC, Brouwer B, Milnes TS, et al. Changes
in muscle responses to stimulation of motor cortex
induced by peripheral nerve stimulation in human
subjects. Exp Brain Res. 2000;131:135–143.
607. Willer JC. Relieving effect of TENS on painful
muscle contraction produced by an impairment of
reciprocal innervation: an electrophysiological
analysis. Pain. 1988;32(3):271–274.
608. Abelson K, Langley GB, Sheppeard H, et al.
Trascutaneous electrical nerve stimulation in rheu-
matoid arthritis. N Z Med J. 1983;96:156–158.
609. Brosseau L, Yonge K, Marchand S, Robinson V,
Wells G, Tugwell P. Efficacy of transcutaneous
electrical nerve stimulation (TENS) for rheumatoid
arthritis: a systematic review. Phys Ther Rev.
2002;7:199–208.
610. Brosseau L, Yonge K, Marchand S, Robinson V,
Clinical Practice Guidelines 117
Osiri M, Wells G, Tugwell P. Efficacy of transcuta-
neous electrical nerve stimulation for osteoarthritis
of the lower extremities: a meta-analysis. Phys Ther
Rev. 2004;9:213–233.
611. Levin MF, Hui-Chan CWY. Conventional and acu-
puncture-like transcutaneous electrical nerve
stimulation excites similar afferent fibers. Arch Phys
Med Rehabil. 1993;74:54–60.
612. Cheng R, Pomeranz B. Electroacupuncture analge-
sia could be mediated by at least two pain-reliev-
ing mechanisms: endorphin and non-endorphin
systems. 1980;1957–1962.
613. Sjolund BH. Peripheral nerve stimulation suppres-
sion of C-fiber-evoked flexion reflex in rats. Part 1:
parameters of continuous stimulation. J Neurosurg.
1985;63(4):612–616.
614. Gopalkrishnan P, Sluka KA. Effect of varying fre-
quency, intensity, and pulse duration of transcuta-
neous electrical nerve stimulation on primary hy-
peralgesia in inflamed rats. Arch Phys Med Rehabil.
2000;81(7):984–990.
615. Sluka KA, Deacon M, Stibal A, Strissel S, Terpstra A.
Spinal blockade of opioid receptors prevents the
analgesia produced by TENS in arthritic rats. J
Pharmacol Exp Ther. 1999;289(2):840–846.
616. Sluka KA, Judge MA, McColley MM, Reveiz PM,
Taylor BM. Low frequency TENS is less effective
than high frequency TENS at reducing inflamma-
tion-induced hyperalgesia in morphine-tolerant
rats. Eur J Pain. 2000;4(2):185–193.
617. Lundeberg T. Long-term results of vibratory
stimulation as pain relieving for chronic pain. Pain.
1984;20:13–23.
618. Guieu R, Tardy-Gervet MF, Roll JP. Analgesic ef-
fects of vibration and transcutaneous electrical
nerve stimulation applied separately and simulta-
neously to patients with chronic pain. Can J Neurol
Sci. 1991;18:113–119.
619. Tardy-Gervet MF, Guieu R, Ribot-Ciscar E, Roll JP.
Les vibrations mécaniques transcutanées: effets
antalgiques et mécanismes antinociceptifs. Rev
Neurol. (Paris). 1993;149:177–185.
620. Hartley A. Therapeutic Ultrasound. 2nd ed.
Etobicoke, Ontario: Anne Hartley Agency; 1993.
621. Ministre des Approvisionnements et Services
Canada. Principes d’utilisation des ultrasons. 1re
partie: applications médicales et paramédicales. Ot-
tawa: Centre d’édition du gouvernement du
Canada; 1989.
622. Casimiro L, Brosseau L, Robinson V, Milne S, Judd
M, Wells G, Tugwell P. Therapeutic ultrasound for
the treatment of rheumatoid arthritis. Cochrane
Database Syst Rev. 2002;(3):CD003787. DOI:
10.1002/14651858.
623. Koetsier JC. Effects of intensity of rehabilitation
after stroke: a research synthesis. Stroke.
1997;28(8):1550–1556.
Appendixes
120 APPENDIX 1. Literature Search Results
122 APPENDIX 2. Special Formulas for Clinical Relevance
123 APPENDIX 3A. Characteristics of Included Studies for Therapeutic Exercise
145 APPENDIX 3B. Characteristics of Included Studies for Task-Oriented Training
158 APPENDIX 3C. Characteristics of Included Studies for Biofeedback
177 APPENDIX 3D. Characteristics of Included Studies for Gait Training
192 APPENDIX 3E. Characteristics of Included Studies for Balance Training
199 APPENDIX 3F. Characteristics of Included Studies for Sensory Interventions
204 APPENDIX 3G. Characteristics of Included Studies for Constraint-Induced Movement Therapy
207 APPENDIX 3H. Characteristics of Included Studies for Shoulder Subluxation
212 APPENDIX 3I. Characteristics of Included Studies for Electrical Stimulation
219 APPENDIX 3J. Characteristics of Included Studies for Transcutaneous Electrical Nerve Stimulation
(TENS)
222 APPENDIX 3K. Characteristics of Included Studies for Ultrasound
223 APPENDIX 3L. Characteristics of Included Studies for Acupuncture
227 APPENDIX 3M. Characteristics of Included Studies for Intensity and Organization of Rehabilitation
263 APPENDIX 4. Existing Guidelines on Stroke Rehabilitation

119
APPENDIX 1. Literature Search Results
120

Data base: Medline <1966—latest update (AutoAlert service here)>

SET Search Results
SET Search Results
37 (retrospective or case-control).tw. 80997
1 exp cerebrovascular disorders/ 131948
38 or/23-37 899180
2 (stroke or cerebrovascular or cerebral vasc 54754
STUDY FILTER (LINES 23–38)
3 (tia or transient ischemic attach$ or trans 1510
4 (cerebral hemorrhage or cerebral haemorrhag 1885
39 22 and 38 1482
5 (intracerebral hemorrhage or intracerebral 2130
COMBINATION OF STROKE, PHYSIO/REHAB
6 cva.tw. 717
AND STUDY FILTER (LINE 39)
7 hemiplegia.tw,sh. 7741
8 (cerebr$ or cerebellar of brain$ or vertebr 435874
40 gait.tw,sh. 10495
9 (inract or schaemia or schemia or thrombo 206602
41 ambulation disorders$.tw. 4
10 8 and 9 26108
42 locomot$ disorders.tw. 70
11 (cerebral or intracerebral or intracranial 428049
43 cadence.tw. 388
12 (hemorrhage or haemorrhage or hematoma or b 108140
44 exp locomotion/ 22364
13 11 and 12 21288
45 or/40-44 31287
14 or/1-7 162100
46 39 and 45 122
TOPICS IN STROKE REHABILITATION/SPRING 2006

15 10 or 13 or 14 172152
47 activities of daily living/ 18150
STROKE TERMS (LINES 1–15)
48 task-oriented.tw. 204
49 task-related.tw. 573
16 exp physical therapy/ 71533
50 ability focus$.tw. 72
17 rh.fs 77687
51 adl.tw. 1540
18 (physical therapy or physiotherap$). Tw. 8037
52 (activities adj2 living) .tw. 3731
19 rehabilitation/ 7823
53 or/47-52 20806
20 rehabilit$.tw. 38359
54 39 and 53 433
21 or/16-20 162841
TOTAL FOR GAIT = 433 (LINE 54)
PHYSIO AND REHABILITATION TERMS (LINES 16–21)
55 “biofeedback (psychology)” / 3629
22 15 and 21 6299
56 biofeedback.tw. 2613
23 clinical trial . Pt. 284041
57 electromyography/ 37921
24 randomized controlled trial.pt. 131937
58 (electromyography or emg).tw. 15673
25 random$.tw. 192625
59 muscle contraction.tw. 5615
26 (double adj blind$).tw. 53060
60 muscle relax$.tw. 6196
27 placebo$.tw. 61787
61 spastic$ control.tw. 12
28 meta-analysis.pt,sh. 7154
62 motor skills/ 9717
29 (meta-anal : or metaanal:).tw. 5900
63 motor control.tw. 1894
30 (quantitativ: review: or quantitativ: overv 152
64 neuromuscular.tw. 19228
31 (methodologic: review: or methodologic: ove 102
65 or/55-64 82304
32 (systematic: review: or systematic: overvie 1761
66 39 and 65 131
33 review.pt. And medline.tw. 3462
TOTAL FOR BIOFEEDBACK = 131 (LINE 66)
34 exp cohort studies/ 366379
35 (cohort or longitudinal or prospective).tw. 159080
36 exp case-control studies/ 169449 continues
APPENDIX 1. Continued

SET Search Results SET Search Results

94 cryotherapy.tw,sh. 2594
67 exp electric stimulation therapy/ 9148 95 ice.tw 5933
68 ((electric$ adj nerve) or therapy).tw. 450623 96 sensory stimulation.tw. 826
69 (electric$ adj (stimulation or muscle)).tw. 23153 97 (tactile or brush$ or touch$ or cutaneous). 72059
70 electrostimulation.tw. 1686 98 physical stimulation/ 7702
71 electroanalgesia.tw. 156 99 postural control.tw. 594
72 (tens or altens).tw. 1601 100 acoustic stimulation/ 14745
73 electroacupuncture.tw. 769 101 photic stimulation 15537
74 neuromusc$ electric$.tw. 65 102 or/94-101 115884
75 (high volt or pulsed or current).tw. 204406 103 39 and 102 23
76 (electromagnetic or electrotherap$).tw. 6920 TOTAL FOR CYROTHERAPY AND STIMULATION = 23
77 ionophoresis.tw. 231 (LINE 103)
78 or/67-77 669977
79 39 and 78 368 104 constraint.tw. 2633
TOTAL FOR PAIN CONTROL USING TENS, 105 (functional training or functional retraining) 57
ELECTROSTIMULATION ETC. = 368 (LINE 79) 106 functional therap$.tw. 220
107 or/104-106 2910
80 length of stay/ 20344 108 39 and 107 6
81 length stay.tw. 5446 TOTAL FOR FUNCTIONAL TRAINING = 6
82 (program or session).tw. 107270 (LINE 108)
83 or/80-82 128546
84 39 and 83 272 109 exp exercise therapy/ 11322
TOTAL FOR LOS = 272 (LINE 84) 110 (therap$ adj2 exercise$).tw. 1771
111 aerobics.tw. 20486
85 exp leg/ 69577 112 (endurance adj2 therap$).tw. 16
86 (leg or ankle or foot or feet).tw. 61944 113 breathing exercise$.tw. 293
87 85 and 86 17708 114 or/109-113 32424
88 39 and 87 16 115 39 and 114 70
TOTAL FOR LOWER EXTREMITIES = 15 TOTAL FOR EXERCISE THERAPY = 70
(LINE 88 ) (LINE 115)

89 facilitation.tw. 9233 116 exp arm/ 62019

90 range of motion, articular/ 7624 117 shoulder.sh,tw. 16097
91 range motion.tw. 4502 118 arm.tw,sh. 43127
92 or/89-91 19681 119 wrist.tw,sh 10814
93 39 and 92 49 120 hand.tw,sh 121408
TOTAL FOR FACILITATION AND ROM = 49 121 forearm.tw,sh 15901
(LINE 93) 122 elbow.tw,sh 8974
Appendixes

123 or/116-122 209310

124 39 and 123 166
TOTAL FOR UPPER EXTREMITIES = 166
(LINE 124)
121
122 TOPICS IN STROKE REHABILITATION/SPRING 2006

APPENDIX 2. Special Formulas for Clinical Relevance

Regular case
Clinical relevance formula:
(et – bt) – (ec – bc)
( (bt * nt) + (bc * nc) ) / (nt + nc)
Legend: et = end of study value for treatment
bt = baseline value for treatment
ec = end of study value for control
bc = baseline value for control
nt = number of patients in treatment group
nc = number of patients in control group

Special case #1
When the baseline value was not available for an outcome, a formula was used to calculate the relative
difference in the change from baseline:
et – ec
(nt * et + nc * ec) / (nt + nc)
Legend: et = end of study value for treatment
ec = end of study value for control
nt = number of patients in treatment group
nc = number of patients in control group

Special case #2
When the outcome measure was given as a change from baseline value and the baseline value itself was
not given, but the scale was known, a formula was used to calculate the relative difference in change
from baseline:
Δt – Δc
Legend: Δt = change in the value from treatment
Δc = change in the value from control

Special case #3
When the outcome measure was given as a change from baseline value, where the baseline value was not
given and the scale was unknown or nonexistent, a formula was used to calculate the relative difference
in the change from baseline:
Δt – Δc
(nt * Δt + nc * Δc) / (nt + nc)
Legend: Δt = change in the value from treatment
Δc = change in the value from control
nt = number of patients in treatment group
nc = number of patients in control group

Special case #4
When the baseline mean was 0, we added a sum of 1 to all the values in the formula of clinical relevance
based on the assumption that:
Mean of “Scale + 1” = Mean of original scale + 1 and SD of “Scale + 1” = SD of original scale.
APPENDIX 3A. Characteristics of Included Studies for Therapeutic Exercise

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Aisen 199799 Total: 20 Inclusion: Patients Gr1: 2.8 wk Gr1: 58.5 Gr1: Patients’ hands and Gr2: Weekly to All patients received Gr1: 4 to 5 None 0, 1, 1
Gr1: 10 with hemiplegia SD: 1.1 wk SD: 8.3 wrists were held in a rigid biweekly contact standard customary hr/wk until
Gr2: 10 admitted to the support affixed to the robotic with the robotic care. discharge
same medical ward Gr2: 3.3 wk Gr2: 63.3 arm. Therapy consisted device. Patients
and team of Burke SD: 1.2 wk SD: 10.6 of flexion, extension, and actively moved the
Rehabilitation Subacute rotation movements across robotic arm and
Hospital, White elbow and shoulder joints. were able to observe
Plains, NY, 3 wk; These arm movements are the response on the
and who had SD of goal-directed and robot- video monitor. The
1 wk after a single assisted, and the therapy uses robotic device was
stroke. a computer-generated video also used to record
Hemiplegia: program that provides visual strength and quality
Gr1: 3 R; 7 L and auditory feedback. If of movement.
Gr2: 1 R; 9 L the arm was paralyzed, the
Treated area: robot initiated the movement
Affected upper passively. As function
extremity returned, the patient initiated
more often.

Carr 2003100 Total: 40 Stroke survivors Greater than 30–82 yr Gr1: Same aerobic protocol Gr2: 20 min of N/A 3 days/wk for N/A 1, 0, 0
whose ages ranged 6 months as Gr2; 8 different strength aerobic exercise on 16 wk
from 30 to 82 yr training exercises: chest a upper and lower
press, seated row, leg press, body ergometer.
leg curl, triceps press down,
biceps curl, shoulder front
raise; 2 sets of 10 repetitions;
free weights and isokinetic
machines; and flexibility
exercises.
Appendix 3A
123

continues
APPENDIX 3A. Continued
124

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Chu 2004101 Total: 12 At least 1 yr Gr1: 3.0 yr Gr1: 61.9 Gr1: Main objective: improve Gr2: Main objective: N/A 8 wks, 3 days/ N/A 2, 0, 1
Gr1: 7 poststroke from SD: 2.0 yr SD: 9.4 cardiovascular fitness in improve upper wk, 1 hr/session
Gr2: 5 a single CVA; stroke patients after 8 wk extremity function.
independent in Gr2: 4.2 yr Gr2: 63.4 of intensive water-based Each session: 5-min
walking (with or SD: 2.1 yr SD: 8.4 exercise in chest-level water. warm-up of active
without assistive Intervention: 10 min of upper extremity
device); medically land-based stretching, 5 min movement, 6-station
stable; no previous of light aerobic warm-up (7-min stations)
myocardial in water (marching on the circuit focused
infarction; no spot, single- and double- on gross upper
significant legged hopping holding limb movement
musculoskeletal onto the pool edge), 30 (reaching), fine motor
problems from min of moderate to high movement (adjusting
conditions other aerobic activities (shallow small screws and
than stroke. water walking, running, side bolt), and muscle
TOPICS IN STROKE REHABILITATION/SPRING 2006

Hemiplegia: stepping) at target heart rate strengthening of

Gr1: 4 R; 3 L prescribed for that wk (50%
Gr2: 3 R; 2 L to 70%, 75% and 80% heart
rate reserve +/– 5 beats/min
for wk 1–2, 3–5, and 6–8,
respectively), 5 min light cool
down (marching on the spot),
and 10 min gentle stretching
in the water.
the upper extremity
(using hand putty,
theraband, weights),
5-min cool-down
from upper extremity
exercises.
Dickstein Total: 131 Hemiplegic patients 16 days 70.5 Gr1: Proprioceptive Gr3: Standard N/A At least 5 Follow- 2, 0, 0
1986102 Gr1: 36 who had had a SD: 7.65 neuromuscular facilitation customary care: Gait sessions/wk for ups at 2
Gr2: 38 recent stroke. (PNF): Reflexes (most training exercises 6 wk wk (mid-
Gr3: 57 Hemiplegia: commonly the stretch reflex) performed in treatment),
62 L; 66 R frequently were used to elicit anatomical planes. 4 wk (mid-
3 bilateral movements. Mass-movement Progression was treatment),
patterns, such as diagonal encouraged either and 6 wk
and spiral patterns and total by gradual increase (end of
patterns of the development in number of joints treatment)
sequence formed an integral involved or by
part of the exercises. increased resistance
Gr2: Bobath: First step to a requested
of treatment sessions was movement. Passive
geared toward the inhibition movements were
of abnormal muscle tone, administered to
usually through application of immobile joints.
appropriate reflex-inhibiting Option of using
patterns. This was performed pulleys, suspensions,
concurrently with an effort to or weights. Practice
initiate normal movements of ADL training for
(automatic and voluntary). independence began
Through “key points of as early as possible.
control” in patients’ bodies. Each treatment
Imposition activity such session lasted 30–45
as effort to impose normal min.
sensations of posture and
movements in which weight-
bearing exercises played an
important role. Facilitation
of postural activity by touch
and proprioceptive stimuli
for low or flaccid muscle
tone. Resistive exercises,
mass movements, and use of
simple and abnormal reflexes
were forbidden. Progress
generally followed the normal
developmental sequence.
For Gr1 and Gr2, each
treatment session lasted
30–45 min.
Appendix 3A
125

continues
APPENDIX 3A. Continued
126

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Duncan Total: 20 Inclusion: Gr1: 66 days Gr1: 67.3 Gr1: Home-based exercise Gr2: Usual care as Concurrent Gr1: 3 sessions/ None 2, 0, 1
1998103 Gr1: 10 30–90 days after Gr2: 56 days SD: 9.6 training provided by PT. 10- prescribed by their treatment: Speech wk for 8 wk;
Gr2: 10 onset, minimal min warm-up (stretching and physicians (6 patients therapy if needed. 1.5 hr/session.
or moderately Gr2: 67.8 flexibility); 4 blocks: had home health Patients
impaired SD: 7.2 1) Assistive and resistive visits, 4 patients instructed to
sensorimotor exercise using PNF patterns had outpatient continue on
function, to upper and lower physiotherapy). Visit their own for
ambulatory with extremities or theraband included balance additional 4
supervision and/or exercises to major muscle training, progressive wk.
assistive device, groups of upper and lower resistive exercises, Gr2:
living at home extremities. Progression: bimanual activities, Patients had
within 50 miles When patient was able to and facilitative approximately
of the University complete 2x20 repetitions, exercises. 39 visits,
of Kansas Medical increased theraband resistance approximatively
Center. or increased PNF manual 44 min per
TOPICS IN STROKE REHABILITATION/SPRING 2006

Exclusion: Medical resistance. visit.

condition that 2) 15 min of balance
interfered with exercises.
outcome assessment 3) Patients encouraged to use
or limited affected upper extremity in
participation functional activities.
in submaximal 4) Progressive walking
exercise program; program/exercise on a bicycle
Mini-Mental State ergometer x 20 min.
score <18; and
receptive aphasia
that interfered with
the ability to follow
a 3-step command.
Hemiplegia:
Gr1: 6 L; 4 R
Gr2: 5 L; 4 R
Duncan Total: 92 Stroke within 30 Gr1: 77.5 Gr1: 68.5 Gr1: Exercise program: Gr2: Standard N/A 12–14 wk N/A 2, 1, 1
2003104 Gr1: 44 to 150 days; ability SD: 28.7 days SD: 9.0 ROM and flexibility for customary care:
Gr2: 48 to ambulate 25 shoulder, elbow, wrist, Home visits by
feet independently; Gr2: 73.5 Gr2: 70.2 fingers, hip, ankle, and research staff every
mild to moderate SD: 27.1 days SD: 11.4 trunk. Strengthening resistive 2 wk for health
stroke deficits exercises, 2 sets of 10 education, vital signs,
defined by a Fugl- repetitions, balance training, and test of oxygen
Meyer score of 27 upper extremity functional saturation.
to 90 for upper and use, endurance training (up to
lower extremity; 30 min on bike), 36 sessions
an Orpington of 90 min.
Prognostic Scale
score of 2.0 to
5.2 and palpable
wrist extension
on the involved
side; and Folstein
Mini-Mental Status
examination score
> 16.
Hemiplegia:
Gr1: 18 R; 22 L
Gr2: 22 R; 22 L
Appendix 3A
127

continues
APPENDIX 3A. Continued
128

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Fasoli Total: 20 Diagnosis of a 31 months 55.5 Gr1: The robot provided Gr2: Same goal- Robotic therapy 1 hr, 3 x/wk for Follow-up 1, 0, 0
2003105 Gr1: 13 single, unilateral SD: SD: 17.2 movement assistance. directed, planar delivered with 6 wk Gr2 at 6 wk
Gr2: 7 stroke within the 12.1 months reaching tasks while the MIT-MANUS.
past 1 to 5 yr moving against Hemiparetic
verified by brain an opposing force arm placed in
imaging; sufficient generated by the a customized
cognitive and robot. The magnitude arm support and
language abilities of the opposing force patient was asked
to understand and was determined and to perform goal-
follow instructions; modified according directed, planar
and stroke-related to patient’s muscle reaching tasks
impairments in strength. that emphasized
muscle strength shoulder and elbow
of the affected movements while
shoulder and elbow computer screen
TOPICS IN STROKE REHABILITATION/SPRING 2006

between grades provided visual

2 and 4 on the feedback.
Medical Research
Council (MRC)
motor power score.
Hemiplegia:
6 R; 14 L
Treated area:
Hemiplegic upper
extremity

Glasser Total: 20 Inclusion: 3–6 months 40–75 Gr1: Exercise program: Gr2: Control: Therapeutic 2 sessions/day, Follow-up 1, 0, 0
1986106 Gr1: 10 Patients who have During the first week, patients Exercise program exercise program 5 days/wk for at end of
Gr2: 10 hemiparesis due received 1 hr and 50 min without Kinetron. that consisted of 5 wk treatment (5
to stroke and who of therapeutic exercise and techniques based on wk)
are 3 to 6 months independently exercised for neurophysiological
onset. 10 min on the Kinetron. and developmental
Hemiplegia: An additional 5 min of theories.
Gr1: 5 R; 5 L Kinetron was added weekly:
Gr2: 5 R; 5 L 25 repetitions of both lower
Treated area: Lower extremities was followed by
extremity rest for 1 min.
Inaba 1973107 Total: 77 Hemiplegic patients 3 months or 56 Gr1: Active exercise: Gr3: Control: No Functional training 4 to 8 wk Follow-ups 1, 0, 1
Gr1: 23 with CVA secondary less Gr1: 56.1 Consisted of bilateral hip and exercise treatment and appropriate at 1 month
Gr2: 28 to thrombosis, Gr2: 55.9 knee flexion and extension was given. stretching was (end of
Gr3: 26 embolus, or Gr3: 56.9 in the supine and side lying given if contracture treatment)
intracerebral positions; hip abduction that interfered and 2
hemorrhage; able and adduction; lower limb with function was months
to follow verbal coordination exercises; and present. Functional (end of
or demonstrated trunk flexion, extension, and training consisted treatment)
directions; and rotation. Reciprocal exercise of bed activities;
able to push a 1.1 without resistance was assuming and
kg weight on the accomplished by use of the maintaining sitting
Elgin table with Restorator for 15 min/day. position; operating
the involved lower Gr2: Progressive resistive a wheelchair;
extremity. Also exercise: Included mass transferring to and
unable to walk extension of the involved from bed, toilet, car,
independently. lower limb in the supine tub or shower; gait
Hemiplegia: position. Extension was training on various
Gr1: 8 L; 15 R initiated from 90° of knee surfaces; and
Gr2: 14 L; 14 R flexion to full knee extension getting up from the
Gr3: 13 L; 13 R ground. Stretching
Treated area: Lower was done for hip
extremity and knee flexion
contractures.
Appendix 3A
129

continues
APPENDIX 3A. Continued
130

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Katz-Leurer Total: 90 Patients who were Subacute Gr1, 2: 63 Gr1: Aerobic exercise Gr2: Control group: Regular therapy Gr1: 8 wk; first 8 wk 2, 0, 1
2003a108 Gr1: 46 hospitalized up stage SD: 11 training: Trained on lower Regular therapy in in rehab (physical part lasted 5 (end of
Gr2: 44 to 48 hr after the extremity cycle ergometer rehab and 5 days/wk therapy, OT, and days/wk for 2 treatment)
initiation of clinical with individualized exercise of group activity for speech therapy). wk; second part
signs of their first program (on the basis of general exercises. lasted 3 x/wk
stroke. initial stress test results). for 30 min
Hemiplegia: N/A Training divided into 2 parts:
first part lasted 5 days/wk for
2 wk: started with multiple
2-min intervals, according to
patient tolerance, with 1-min
resting period (up to 10 min
of work in first day). Add 1
min to one or more interval
working periods each day
TOPICS IN STROKE REHABILITATION/SPRING 2006

so that by end of second wk

could work continuously
for 20 min at low level.
Second part lasted next 6 wk:
exercised 3 x/wk for 30 min.
Intensity was limited to 60 %
of heart rate reserve.
Katz-Leurer Total: 90 Patients who were Subacute Gr1: 62 Gr1: Exercise-training: Gr2: Control group: N/A Gr1: 8 wk; first Follow-up 2, 0, 1
2003b109 Gr1: 46 hospitalized up stage SD: 11 Trained on lower extremity No individualized part lasted 5 at 8 wk
Gr2: 44 to 48 hr after the cycle ergometer with exercise program. days/wk for (end of
initiation of clinical Gr2: 65 individualized exercise 2 wk; second treatment)
signs of their first SD: 11 program (on the basis of part: 30-min
stroke. initial stress test results). sessions,3 x/wk
Hemiplegia: N/A Training divided into 2 parts:
First part: lasted 5 days/wk for
2 wk: started with multiple
2-min intervals, according
to patient tolerance, with
1-min resting period (up to
10 min of work in first day).
Added 1 min to one or more
interval working periods
each day so that by end of
second wk patient could work
continuously for 20 min at
low level. Second part: lasted
next 6 wk: exercised 3 x/wk
for 30 min. Intensity was
limited to 60 % of heart rate
reserve.
Appendix 3A
131

continues
APPENDIX 3A. Continued
132

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Kim 2001110 Total: 20 Patients aged 50 Gr1: 4.9 Gr1: 60.4 Gr1: Isokinetic strengthening: Gr2: Control: Same N/A Three 45-min Follow-up 2, 2, 1
Gr1: 10 yr or more; history SD: 3.3 yr SD: 9.5 5 min warm-up (5 repetitions warm-up and cool sessions/wk for at 6 wk
Gr2: 10 of a single stroke of active alternative flexion down. 3 sets of 6 wk (end of
at least 6 months Gr2: 3.2 Gr2: 61.9 and extension of the hip, passive ROM were treatment)
before participating SD: 1.2 yr SD: 7.5 knee, and ankle sitting in performed for each
in the study; a chair followed by 5 min joint of paretic lower
ability to walk mild stretching of the paretic extremity using
independently for upper extremity) and 5 min Kin-Com Isokinetic
a minimum of 40 cool down (mild stretching). Dynamometer.
meters with rest 3 sets of 10 repetitions of Participants were
intervals, with or maximal effort concentric instructed to relax
without assistive hip flexion/extension, knee the extremity as
device; achievement flexion/extension, and ankle it was moved into
of a minimum dorsiflexion/plantiflexion were flexion and extension
of stage 3 for leg performed using Kin-Com by the dynamometer.
TOPICS IN STROKE REHABILITATION/SPRING 2006

and foot on the Isokinetic Dynamometer for

Chedoke-McMaster approximately 30 min. Rest
Stroke Assessment; break if necessary.
activity tolerance
of 45 min with
rest intervals; and
non-participation in
any formal therapy
program.
Hemiplegia:
Gr1: 6 L; 4 R
Gr2: 2 L; 8 R
Kumar Total: 28 Patients who 14.5 64.9 Gr1: Overhead pulley Gr3: Passive ROM: Gait and ADL 5 sessions/wk N/A 1, 0, 1
1990111 Gr1: 12 had had a stroke SD: 2.52 days SD: 18.2 exercise: Patients were seated An OT passively training
Gr2: 8 recently. Range, in wheelchair between the ranged the patient’s
Gr3: 8 Hemiplegia: Gr1: 14.5 43–90 pulley ropes. The upper affected arm. The
12 L; 16 R SD: 1.51 days extremity was positioned OT supported the
Treated area: Gr1: 63.2 in 45° of abduction and patient’s elbow and
Shoulder Gr2: 15.1 was attached to one end of forearm with one
SD: 3.4 days Gr2: 70.1 the rope with a mitten. The hand, and upper
patient pulled the other end of arm with the other.
Gr3: 13.8 Gr3: 60.8 the ropes and moved the arm The shoulder was
SD: 2.36 days through a range of 130–150° ranged 140–150°
of abduction always parallel to of abduction. Care
the scapula. was taken to keep
Gr2: Skateboard: Patients the upper extremity
were seated at a 32-in. high in external rotation
table. A skateboard with a during abduction.
figure eight on it was centered The shoulder was
in front of the patient who ranged of external
was instructed to follow the and internal rotation.
figure eight with the affected Elbow, wrist, and
arm. The arm was moved fingers were also
through approximately ranged through full
90° of forward flexion, 90° range of extension
of abduction, and 45° of and flexion.
adduction.

Langhammer Total: 24 Patients with Not specified 78 Gr1: Motor relearning Gr2: Bobath: No Comprehensive, 1 session/day, Follow- 1, 2, 1
2000112 Gr1: 53 first-ever stroke SD: 9 program: No specification. specification. multi-disciplinary 5 days/wk ups at 2
Gr2: 29 with hemiparesis, Range, Treatments for both treatment for stroke until discharge wk (mid-
verified clinically 49–95 groups given for at from doctors, (followed by treatment)
and by CT scan. least 40 min/session. nurses, OTs, and out patient and 3
Hemiplegia: speech therapists. treatments). months
Gr1: 16 L; 17 R (end of
Gr2: 11 L;17 R treatment)

Langhammer Total: 28 Patients who agreed Not specified 78 Gr1: Motor relearning Gr2: Bobath: No Comprehensive, 1 session/day, Follow-ups 1, 0, 1
2003113 Gr1: 27 to participate in program: No specification. specification. multi-disciplinary 5 days/wk at 1 year
Gr2: 21 the first study Treatments for both treatment for stroke until discharge and 4 yr
were invited to groups given for at from doctors, (followed by
participate in a least 40 min/session. nurses, OTs, and out patient
follow-up study. speech therapists. treatments)
Appendix 3A
133

continues
APPENDIX 3A. Continued
134

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Liu 2004114 Total: 46 1) Diagnosed Gr1: 12.3 Gr1: 71.0 Mental imagery program: In the functional Patients in both 3 wk with five N/A 2, 0, 1
Gr1: 26 as having had a days SD: 6.0 Patients were trained in the retraining program, groups were trained 1-hr sessions
Gr2: 20 first unilateral SD: 5.3 days technique of mental imagery the demonstration- to perform 3 sets each week
cerebral infarction Gr2: 72.7 to practice specific tasks. In then-practice method of daily tasks.
as confirmed by Gr2: SD: 9.4 the first wk, the focus was on was adopted. Patient There were 5
a CT scan; 2) age 15.4 days analyzing task sequences to were required to tasks in each set
60 yr or older; 3) SD: 12.2 days facilitate motor planning and practice the same including mobility
independent in problem identification process tasks following functioning,
performing daily using computer-generated a sequence and balance, or upper
activities before pictures and movies. In the training schedule limb coordination.
admission; 4) able second wk, patients identified similar to that of The difficulty level
to communicate their own problems for the mental imagery was organized in
effectively, as rectification through the use program. However, ascending order,
screened by the of mental imagery. In the the problems with the easiest
Cognistat, and; 5) third wk, the focus was on encountered by tasks covered in
TOPICS IN STROKE REHABILITATION/SPRING 2006

having given their practicing the rectified task patients were the 1st wk. Patients
voluntary consent. perfomance using mental rectified with the in both groups
imagery and actual practice. help of therapists. also received 1-
hr physiotherapy
sessions for training
in walking and
general muscle
strengthening 5
days/wk.
Logigian Total: 42 Patients with stroke 7 wk or less Total: 61.6 Facilitation technique: Standard customary Physiotherapy Daily until N/A 1, 0, 0
1983115 Gr1: 21 documented by CT SD: 21 Treatment included bilateral care emphasized and OT exercises discharge (time
Gr2: 21 scan within 7 wk weight-bearing and weight- strengthening the in addition to not specified)
of onset. Medically shifting exercises; utilization developing motion ADL and speech
stable and fit to of reflex inhibiting patterns; and maintaining therapy. Extra ROM
participate in a and tactile, vibratory, and full passive motion. exercises if patients
nonrestrictive vestibular stimulation Treatment techniques wished.
program as activities. included passive,
determined by the assistive, active,
attending physician. and progressive
resistance exercises
and employed the
use of upper limb
skateboard, weighted
sanders, reciprocal
pulleys, and springs.
Each patient received
between 60 to 90
min of treatment a
day.
Appendix 3A
135

continues
APPENDIX 3A. Continued
136

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Lum 2002116 Total: 27 Inclusion: Diagnosis Time since Gr1: 63.2 Gr1: Patient seated in a Gr2: Conventional Concurrent 1 hr per session Follow-up 2, 2, 1
Gr1: 12M/ of a single CVA, stroke onset, SD: 3.6 wheelchair, torso movement treatment based on treatment: Home- for 24 sessions at 6 months
1F more than 6 months: limited, affected limb strapped neurodevelopmental based exercise
Gr2: 8M/6F months post-CVA, Gr1: 30.2 Gr2: 65.9 to forearm splint, robot therapy: 10 min regimen or
and obvious deficit SD: 6.2 SD: 2.4 manipulator attached to the establishing physical community-based
in upper-extremity splint. 4 exercise modes: postural base of stroke programs
motor function as a Gr2: 28.8 1) Passive: Robot moved arm support coupled they were enrolled
result of CVA. SD: 6.3 toward target. with assessing and in at the time of
Exclusion: Upper 2) Active-assisted: Patient facilitating the intake into the
extremity joint pain triggered initiation of alignment of the study.
or ROM limitations movement with volitional shoulder; and 35 min
that would limit force toward the target and graded application
their ability to worked with the robot as it of the arm’s use in
complete the moved the extremity. ADLs. Emphasis
protocols; unstable 3) Active-constrained: Robot on reeducation
TOPICS IN STROKE REHABILITATION/SPRING 2006

cardiovascular/ provided viscous resistance of muscles using

orthopedic/ in the direction of the desired sensorimotor
neurologic movement and springlike approach to control
conditions; and forces in all other directions motor output. 5-min
unable to cooperate as the patient attempted to exposure to robot.
with the study reach toward the target with
tasks. maximal effort.
Hemiplegia: 4) Bimanual: Patient
Gr1: 9 L; 4 R attempted a bimanual mirror-
Gr2: 10 L; 4 R image movement while robot-
assisted affected extremity.
Emphasis placed on target
reaching movement, starting
close to the body and ending
further away. 4 point-to-point
reaching directions trained:
1) forward medial,
2) directly forward,
3) forward lateral,
4) directly lateral.
12 targets located at shoulder
or eye level. Progression from
easiest exercise mode to most
challenging.
Marigold Total: 48 Patients aged 50 Gr1: 3.8 Gr1: 67.5 5-min warm up: walking Same warm up and N/A 1 hr/session, 3 Follow-up 2, 0, 1
2005117 Gr1: 26 yr or more; single SD: 2.4 yr SD: 7.2 and light stretching and cool down as Gr1. x/wk for 10 wk at 1 month
Gr2: 22 stroke; at least 12 5-min cool down: light Challenged dynamic
months from onset; Gr2: 3.6 Gr2: 68.1 stretching. Slow low-impact balance and tasks
and able to walk SD: 1.8 yr SD: 9.0 movements consisting of were progressively
with or without an stretching and weight shifting. increased in difficulty.
assistive device for Weight shifting: Tai chi-like Emphasis on agility
a minimum of 10 movements and reaching and multisensory
meters and have an tasks. Stretching: Major approach. Tasks
activity tolerance muscle groups while standing included standing
of 60 min with rest and while on floor mats. in various postures
intervals. and walking with
various challenges.
Additional
exercises: sit-to-
stand movements,
rapid knee raise
while standing,
and standing
perturbations. Eyes
closed conditions and
foam surfaces were
incorporated.

Moreland Total: 133 Patients who were Gr1: 36.8 Gr1: 69.1 Progressive resistance Same as Gr1 Standard customary 30 min/session, Follow-up 2, 0, 1
2003118 Gr1: 68 less than 6 months SD: 27.8 days SD: 14.8 exercises with weight at except no external care. 3 x/wk at 8 months
Gr2: 65 post stroke; able waist or on lower extremities resistance.
to understand and Gr2: 38.1 Gr2: 72.0 in functional patterns of
follow instructions; SD: 25.6 days SD: 12.1 movement.
motor recovery of
the upper extremity
at stages 3, 4, or 5,
as defined by the
CMSA stages; motor
recovery of the foot
at stages 2, 3, 4, 5
or 6; and informed
consent given by
the participant or
substitute decision
maker.
Appendix 3A
137

continues
APPENDIX 3A. Continued
138

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Mudie Total: 40 Patients had Within 2–6 72.4 Gr1: Portable computer-like Gr3: Bobath-trained N/A 2 wk of daily Follow-ups 2, 2, 1
2002119 Gr1: 10 suffered a recent wk SD: 9.01 BPM feedback console used staff physiotherapists training at 2 and 12
Gr2: 10 stroke and to provide awareness of focused on increasing sessions wk
Gr3: 10 consistently bore weight distribution during trunk and pelvic
Gr4: 10 majority of weight training in sitting; patient ROM, normalizing
on one side. required to touch target with trunk muscle
unaffected hand at various tone, maintaining
heights and distances and appropriate balance
patients attempted to return responses during
to symmetrical position after reaching, and
reaching; 2 wk of training; 30 improving the
min session. patient’s ability to
Gr2: Task-related reach move in and out
training: Patient seated on of asymmetric
adjustable plinth, grocery posture; verbal and
TOPICS IN STROKE REHABILITATION/SPRING 2006

items at about 140% of manual facilitation

upper extremity length were
retrieved with unaffected
upper extremity and placed
on shelves at various heights
and distances to allow
reaching to extremes of seated
base of support.
by therapist during
seated reaching or
lying.
Gr4: Control:
Standard customary
care in physiotherapy
and OT.
Paul 1998120 Total: 20 Diagnosis of 93.4 days 61.75 Gr1: Music-making activity Gr2: Physical exercise N/A 30 min/session, Follow-ups 1, 0, 0
Gr1: 10 unilateral cerebral SD: 49.5 days SD: 5.1 with upper extremity group conducted by 2 x/wk at 10 wk
Gr2: 10 hemiplegia; had movement; participants had recreation therapists.
reached their to find a rhythm or beat Patients were
maximum capacity that was expressive and encouraged to move
of physical function comfortable for them. their extremities in
and subsequently various directions
discharged from and positions and to
occupational raise their affected
and PT; had extremities as high
sufficient verbal as they could in
comprehension to different directions
participate in the (involved shoulder
study, evaluated flexion and elbow
by the ability to extension).
follow 2- to 3-
step directions;
had at least 10°
of limitation in
active shoulder
flexion and elbow
extension in the
involved upper
extremity; had
at least stage IV
Brunnstrom motor
recovery in the
affected upper
extremity; were
able to hold a
drumstick that is 5
cm in diameter and
weighed 8 grams
in their affected
hand; and free from
cardiac conditions.
Hemiplegia:
12 R; 8 L
Appendix 3A
139

continues
APPENDIX 3A. Continued
140

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Potempa Total: 42 Patients who More than 6 Range, Gr1: Exercise: Exercised on Gr2: Control: Given N/A 3 sessions/wk Follow-up 1, 0, 1
1995121 Gr1: 19 had had a stroke months 43–72 yr an adapted cycle ergometer passive exercise for 10 wk at 10 wk
Gr2: 23 more than 6 for 30 min. During first 4 wk, for ROM to body (end of
months before training load was gradually joints in a systematic treatment)
randomization. increased from a workload procedure for 30
Medically representing 30%–50% of min.
stable and had maximal effort to the highest
completed a formal level attainable by the patient.
rehab program. The highest load was then
Hemiplegia: maintained for the final 6 wk
Gr1: 8 L; 11 R of training.
Gr2: 15 L; 8 R

Stein 2004122 Total: 18 Patient who Gr1: Gr1: 53.3 All patients received 6 wk Gr2: During active- N/A 3/wk for 1 hr N/A 2, 0, 1
Gr1: 9 had had a single 27 months SD: 16.4 of robot-aided exercises. assisted training, sessions, for a
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr2: 9 previous stroke, SD: All evaluation and training the robot provided total of 18 hr
residual paresis 12.7 months Gr2: 52.5 exercises consisted of reaching assistance in reaching of robot-aided
with average SD: 14.4 tasks in the horizontal plane each target if the exercise training
strength in the Gr2: that involved shoulder and patient was unable to per patient
upper limb 27.3 months elbow movements. Patients reach independently.
(measured at SD: were asked to move between a
shoulder and 11.4 months center target and 8 peripheral
elbow flexors targets arranged in circular
and extensors) display. 60 reps of each round
between 2 and 4 of moving to the set of target
on the Medical were performed.
Research Council Gr1: During resistance
grading system. training, the robot provided
All the patients resistance to the movement of
were required to reaching.
have conclued
any conventional
physical or OT
before enrollment
in the study.
Hemiplegia:
Gr1: 4 L; 5 R
Gr2: 6 L; 3 R
Teixeira- Total: 13 Patients with Gr1: 9.15 Gr1: 65.87 Gr1: Exercise: Supervised Gr2: Control: No N/A 3 x/wk for 10 Follow-up 1, 0, 1
Salmela Gr1: 6 unilateral stroke SD: 12.7 yr SD: 10.16 sessions included a warm exercise treatment wk at 10 wk
1999123 Gr2: 7 with residual up (5–10 min) consisting of was given. (end of
weakness, spasticity, Gr2: 6.4 Gr2: 69.42 calisthenics, mild stretching, treatment)
or both, of the SD: 6.2 yr SD: 8.85 and ROM exercises. Aerobic
affected lower exercises consisting of
extremity. At least 9 graded walking plus stepping
months post stroke; or cycling at a gradually
independently increased target heart rate
ambulatory for 15 of 70% maximal heart rate;
min with or without strength training; and a
assistive devices; cool down period (5–10
with an activity min) consisting of muscular
tolerance of 45 min relaxation and stretching.
with rests; and no Patients were also given
comprehensive exercises to do at home and
aphasia. encouraged to do 3x/wk. Each
Hemiplegia: session lasted 60 to 90 min.
Gr1: 3 L; 3 R
Gr2: 4 L; 3 R
Appendix 3A
141

continues
APPENDIX 3A. Continued
142

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Trombly Total: 20 Patients who Gr1: 5.2 wk Gr1: 67.2 Gr2: Resisted extension: Gr1: Control: No N/A 1 session/ Follow-up 2, 0, 1
1986124 Gr1: 5 could grasp a Gr2: 3.4 wk Gr2: 60.8 Rubber bands were placed exercise treatment day for at 2 wk
Gr2: 5 2.5 cm cylinder, Gr3: 6 wk Gr3: 75.8 over the middle phalanges was given. approximately (end of
Gr3: 5 understand Gr4: 11.1 wk Gr4: 63.4 and resistance was offered Gr4: Resisted 2 wk treatment)
Gr4: 5 directions, and who by the maximum number grasp: A Jamar
were free of pain on of rubber bands that the dynamometer was
the affected upper patient could extend fingers set up so the handles
extremity and against and hold for 6 s. The were 2.5 cm apart
medically cleared pronated forearm and palm or a Hand Helper
to participate. rested on a board designed to exercise aid was
Hemiplegia: stabilize the wrist. loaded with the
Gr1: 2 R; 3 L Gr3: Ballistic extension: A maximum number of
Gr2: 2 R; 3 L ping-pong ball was placed in elastics the patients
Gr3: 2 R; 3 L front of the patient’s flexed could squeeze. The
Gr4: 3 R; 2 L fingers (middle finger) and the maximum grasp was
TOPICS IN STROKE REHABILITATION/SPRING 2006

Treated area: Hand patient was told to flick the
ball to knock down the paper
cup placed 30–45 cm away
from the patient’s hand.
held for 6 s at each
trial. The forearm
was in mid-position
with the elbow
resting on the table
for support. For
every group verbal
encouragement was
given. Each trial
was repeated 10
times with a 5-s rest
between trials.
Volpe Total: 12 Inclusion: Gr1: 1098.5 Gr1: 54 Gr1: Robot treatment. 25 Gr2: Control: Standard customary 1 hr/day, 5 Follow-up 1, 0, 1
1999125 Gr1: 6 Patients who had days SD: 7 sessions, 1500 repetitions Patients received care in PT and OT days/wk at 3 yr
Gr2: 6 hemiparesis or SD: of goal-directed movement similar initial post stroke.
hemiplegia of the 137.2 days Gr2: 66 to a target. Patients moved exposure to the robot
upper extremity SD: 5 the handle of the tip of a except that half the
or lower extremity, Gr2: 933.8 robot that then moved a trials were performed
and who could days cursor on a screen. A video with the unaffected
follow simple SD: screen provided visual upper extremity,
instructions. 191.1 days feedback through targets and when patients
Hemiplegia: Chronic identical to those drawn on could not perform
Gr1: 2 L; 4 R the support board in front of the task with the
Gr2: 1 L; 5 R the patient and that tracked affected limb, they
Treated area: the movement of the robot used the unaffected
Gr1: Shoulder and handle. Auditory feedback extremity or
elbow (affected) indicated correct movement. technician assistance
Gr2: Affected and If patients did not respond, to complete the
unaffected upper robot guided their hand to the task. The robot
extremity target. Focused on shoulder, never actively
elbow movement pattern moved the patients’
organized in 3 batches, limbs. Patients were
consisting of 20 repetitions exposed to the robot
each. 1 hr/wk.

Volpe Total: 56 Inclusion: Patients Stroke Gr1: 62 Gr1: Robot treatment: 25 Gr2: Control: Similar standard 1 hr/day, 5 None 1, 0, 1
2000126 Gr1: 30 with hemiparesis onset/initial SD: 10.95 sessions, 1500 reps of goal- Patients received customary care in days/wk
Gr2: 26 or hemiplegia of pretreatment directed movement to a target. similar initial PT and OT post
the upper extremity evaluation: Gr2: 67 Patients moved the handle exposure to the robot stroke.
or lower extremity Gr1: SD: 10.20 of the tip of a robot that then except that half the
who were able 22.5 days moved a cursor on a screen. A trials were performed
to follow simple SD: 7.12 days video screen provided visual with the unaffected
instructions. feedback through targets upper extremity, and
Hemiplegia: Gr2: identical to those drawn on when patients could
Gr1: 13R; 17 L 26.0 days the support board in front of not perform the task
Gr2: 12 R; 14 L SD: 7.14 days the patient and that tracked with the affected
Treated area: Acute the movement of the robot extremity, they used
Gr1: Shoulder and handle. Auditory feedback unimpaired limb or
elbow (affected) indicated correct movement. technician assistance
Gr2: Affected and If patients did not respond, to complete the task.
unaffected upper the robot guided their hand to The robot never
extremity the target. Focus on shoulder, actively moved the
elbow movement pattern patients’ extremity.
organized in 3 batches, Patients were
Appendix 3A

consisting of 20 repetitions exposed to the robot

each. 1 hr/wk.
143

continues
APPENDIX 3A. Continued
144

Age, yr Session
Author and Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)
Winstein 60 First-time stroke N/A Gr1: Gr1: Standard customary Gr3: Strengthening N/A 1 hr/day, 5 Follow-up 2, 0, 1
2004127 Gr1: 20 from infarction <35 yr: 2; care delivered by an OT; and motor control days/wk for at 6.5 to 8
Gr2: 20 in the anterior 35–75 yr: muscle facilitation exercises; training; eccentric, 4 wk months
G3: 20 circulation 18; neuromuscular electric isometric, or
confirmed by MRI >75 yr: 0 stimulation for shoulder concentric muscle
or computed axial subluxation; stretching contractions
tomography scan; Gr2: exercise; ADL; and self-care. performed in a
onset of stroke from <35 yr: 0; Gr2: Functional training: gravity-lessened
2 to 35 days before 35–75 yr: Systematic and repetitive position or against
study entry; and 19; practice of tasks arranged in gravity; progressed
a FIM instrument > 75 yr: 1 proximal to distal recovery against resistance
total score at patterns of reaching and using free-weights,
admission of 40 Gr3: grasping. theraband, or grip
to 80. <35 yr: 0; devices for the
Hemiplegia: 35–75 yr: fingers; alternate
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr1: 8 L; 12 R 20; days for 3 days/wk,

Gr2: 7 L; 13 R >75 yr: 0 on other days, same
Gr3: 4 L; 16 R exercises with less
resistance and greater
speed.

Note: ADL = activity of daily living; BPM = balance performance monitor; CMSA = Chedoke-McMaster Stroke Assessment; CVA = cerebrovascular accident; F = female; Gr = group; L = left; M = male; N/A = not available; OT
= occupational therapy; PNF = proprioceptive neuromuscular facilitation; PT = physical therapy; Quality R, B, W = randomization, blinding, withdrawals; R = right; ROM = range of motion; x/wk = times per week.
APPENDIX 3B. Characteristics of Included Studies for Task-Oriented Training

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Dean Total: 20 Inclusion: Patients Gr1: 6.7 yr Gr1: 68.2 Gr1: To improve sitting balance. Gr2: Patients N/A 10 sessions N/A 2, 2, 1
1997159 Gr1: 10 who were hemiplegic SD: 5.8 yr SD: 8.2 Emphasis: Appropriate loading performed of 30 min
Gr2: 10 at least 12 months of the affected leg while cognitive- each for 2 wk
ago and had been Gr2: 5.9 yr Gr2: 66.9 practicing reaching tasks using manipulative tasks
discharged from SD: 2.9 yr SD: 8.2 the unaffected hand to grasp while seated in
all formal rehab objects beyond arm’s length. a chair with arm
services; who can The reaching task was done and back supports
walk short distances under systematic conditions. and forearms
within the home Variations: Location or weight resting on a table.
and understand of object, movement speed, Patients performed
instructions; who distance, direction, extent manipulative tasks
have no orthopedic of thigh support on the seat, using the unaffected
problems that would and height. The number of hand over small
interfere with the repetitions and complexity of distances (less than
ability to perform the tasks were increased over 50% of arm length).
seated reaching the 2 wk. Feedback was given. The number of
tasks; and who can repetitions and
sit unsupported for cognitive difficulty
20 min. of the tasks
Hemiplegia: increased. Patients
Gr1: 5 R; 5 L did an equal
Gr2: 6 R; 4 L number of reaching
Treated area: Affected movements as Gr1.
leg Feedback was given.
Appendix 3B
145

continues
APPENDIX 3B. Continued
146

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

De Sèze Total: 20 Inclusion: Gr1: 36.8 days Gr1: 63.5 Gr1: 1 hr rehab with the Gr2: 2 hr standard Standard rehab: 2 hr each wk Follow-up at 2, 0, 1
2001160 Gr1: 10 Hemiplegia SD: 25 days SD: 17 Bon Saint Côme Device + rehab only Bobath-inspired for 4 wk 8 wk (during
Gr2: 10 caused by a single 1 hr standard rehab (4 wk) approach and the wk both
supratentorial Gr2: 27.7 days Gr2: 67.7 then standard rehab (8 wk) functional therapy groups still
ischemic or SD: 15 days SD: 15 Bon Saint Côme device: 4 received
hemorrhagic stroke parts interconnected: trunk standard
that had occurred orthosis, 2 hemipanels (boards) rehab)
at least 1 month corresponding to the halves of
previously; static extrapersonal space with targets,
imbalance of the 1 keyboard, and a pointer
trunk resulting from guided by trunk movements.
the stroke. When pointer touches a target,
CVA type (ischemic/ visual and auditory signals are
hemorrhagic): elicited. Progression: decreased
Gr1: 3/7 cueing intensity; horizontal
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr2: 4/6 then vertical and diagonal

Hemiplegia: exploration; alternation in
Gr1: 1 R; 9 L rhythm and sensorial modality
Gr2: 4 R; 6 L involved; and patient sitting
then standing when adequate
trunk control gained.

Dickstein Total: 27 Inclusion: Patients 4.9 wk 70.5 Gr1: Repetitive training. Gr2: Control. 10 N/A 8 sessions N/A 0, 0, 1
1997161 Gr1: 15 who can flex the SD: 1.4 wk SD: 8.4 Practicing 100 elbow flex min every other every other
Gr2: 12 paretic elbow for movements, 10 series of 10 day. Composed of a day for
at least 5º on an movements during each session. variable repertoire 19–21 days
exercise board Total: 800 repeated elbow of exercises to the
(constructed for the movements in maximal 80º of paretic UE aimed at
study), follow simple amplitude divided in 8 equal improving function.
directions, and sessions every other day.
who consented to Patients were seated and
participate in study. asked to flex the paretic elbow
Hemiparetia: repeatedly as fast as possible
Gr1: 6 R; 9 L until the index finger reached
Gr2: 7 R; 5 L the target on the board and then
Treated area: Paretic promptly return to the starting
elbow position. Oral encouragement
was given and rest periods
provided between series.
Edmans Total: 79 Inclusion: Patients Gr1: Gr1: 69.75 Gr2: Perceptual training with Gr1: Transfer General OT 2.5 hr/wk for N/A 2, 0, 1
2000162 Gr1: 40 well enough to 37.68 days SD: 9.10 functional approach for 2.5 training approach treatment 6 wk
Gr2: 39 be assessed on SD: hr/wk for 6 wk for 2.5 hr/wk for
the Rivermead 16.60 days Gr2: 67.85 6 wk
Perceptual SD: 11.38
Assessment Battery, Gr2:
with sufficient 31.15 days
functional use of 1 SD:
hand to complete 10.13 days
the RPAB and to
carry out perceptual
treatment activities;
medically stable;
able to transfer with
a maximum of 2
nurses; no discharge
date planned; able
to tolerate 30-min
treatment sessions;
able to do 2 out of
4 specified activities
(eat, drink, wash
their face, toilet
themselves) prior to
stroke.
Hemiparetic:
Gr1: 24 R; 16 L
Gr2: 21 R; 19 L
Appendix 3B
147

continues
APPENDIX 3B. Continued
148

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Gelber Total: 27 Inclusion: Gr1: 11.3 days Gr1: 73.8 Gr1: Neurodevelopmental Gr2: Traditional N/A N/A Follow-ups 1, 0, 0
1995163 Gr1: 15 Patients who were SD: 1.1 days SD: 2.0 technique (NDT). This functional at 26 wk and
Gr2: 12 hemiparetic and had treatment philosophy stresses retraining approach 52 wk
an ischemic stroke Gr2: 13.8 days Gr2: 69.8 inhibiting abnormal muscle (TFR). Practicing
(maximum time from SD: 2.7 days SD: 2.9 tone, initiating normal (good functional tasks as
stroke to entry 1 quality) motor movement early as possible
month). with progression through even in the presence
Hemiplegia: developmental sequences of spasticity or
Gr1: 8 R; 7 L before advancing to functional abnormal posture.
Gr2: 5 R; 7 L activities, and encourages Techniques: Passive
Treated area: Affected patients to use their affected ROM in anatomic
side side. Resistive exercises and use planes, progressive
of abnormal reflexes and mass resistive exercises,
movements were avoided. early use of assistive
TOPICS IN STROKE REHABILITATION/SPRING 2006

device and bracing,

and allowing
patients to use their
affected side to
perform functional
tasks.
Kwakkel Total: 89 Inclusion: Patients Gr1: 7.5 days Gr1: 64.1 Gr2: Upper extremity training. Gr1: Immobilization Concurrent 5 x/wk for Follow-up at 2, 0, 1
1999164 Gr1: 34 who had stroke SD: 2.9 days SD: 15.0 Functional exercises that of paretic arm and treatment: 15 20 wk 6 wk
Gr2: 29 diagnosis (WHO facilitated forced arm and leg by inflatable min/day lower
Gr3: 26 definition); had Gr2: 7.2 days Gr2: 69.0 hand activity such as leaning, splint applied with extremity rehab,
primary first- SD: 2.8 days SD: 9.8 punching a ball, and grasping the patient supine 15 min/day upper
ever stroke in the and moving objects. for 30 min. extremity rehab,
territory of the Gr3: 7.0 days Gr3: 64.5 Gr3: Lower extremity training. 1.5 hr/wk ADL
middle cerebral SD: 2.5 days SD: 9.7 Sitting, standing, weight- training
artery; were 30–80 bearing exercises while
yr old; had impaired standing and walking, with
motor function of emphasis on achieving stability
the arm and leg; and improving gait velocity.
unable to walk at Treadmill training where
first assessment; had equipment available.
no complicating Gr2 and 3 training was applied
medical history; had by OT/PT for 30 min.
no severe deficits
in communication,
memory, or
understanding;
gave written/spoken
informed consent;
were motivated to
participate in the
research project.
Hemiplegia:
Gr1: 13 R; 24 L
Gr2: 16 R; 17 L
Gr3: 13 R; 18 L
Appendix 3B
149

continues
APPENDIX 3B. Continued
150

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Kwakkel Total: 101 Inclusion: Primary, Gr1: 7.5 days Gr1: 64.1 Gr2 (upper extremity): Focused Gr1 (control): 15 min lower 30 min/day, Follow-ups 1, 0, 1
2002165 Gr1: 37 first-ever stroke in SD: 2.9 days SD: 15.0 on improvement in disabilities Immobilization of extremity rehab, 5 days/wk for at 3 and 6
Gr2: 33 the territory of the involving the hemiplegic upper the paretic upper 15 min upper 20 wk months
Gr3: 31 middle cerebral Gr2: 7.2 days Gr2: 69.0 extremity (grasping, reaching, extremity and lower extremity rehab,
artery as shown by SD: 2.8 days SD: 9.8 leaning, dressing). extremity by means 1.5 hr/wk ADL
CT or MRI; between Gr3 (lower extremity): Focused of an inflatable training
30–80 yr of age; had Gr3: 7.0 days Gr3: 64.5 on functional recovery of pressure splint.
an impaired motor SD: 2.5 days SD: 9.7 balance (sitting, standing
function of upper balance), transfers (such
as well as lower as turning over) and gait
extremities; unable (performance and climbing
to walk at first stairs).
assessment; had no
complicating medical
TOPICS IN STROKE REHABILITATION/SPRING 2006

history such as
cardiac/pulmonary/
other neurological
disorders; and had
no severe deficits
in communication/
memory/
understanding.
Hemiplegia:
Gr1: 13R; 24 L
Gr2: 16 R; 17 L
Gr3: 13 R; 18 L

Logan Total: 466 All patients were N/A N/A Gr1: Treated with ADL activities Gr2: Treated with N/A Gr1 & Gr2: Follow-up at 1, 0, 0
2003166 Gr1: 156 recruited after a to achieve ADL goals. leisure activities 10 therapy 6 months
Gr2: 153 hospital admission (activities done for sessions,
Gr3: 157 with stroke. pleasure) to achieve each between
Hemiplegia: N/M leisure goals. 30 and 60
Gr3: Control group min duration
(not reported in
analysis).
Luft Total: 21 Inclusion: Patients Gr1: 75 Gr1: 63.3 Gr1: BATRAC consisted of Gr2: DMTE N/A 3 x/wk for N/A 2,0,1
2004167 Gr1: 9 with residual months SD: 15.3 hr-long therapy sessions (four was based on 6 wk
Gr2: 12 upper extremity Gr2: 45.5 5-min movement periods neurodevelepmental
spastic hemiparesis months Gr2: 59.6 interspered with 10-min rest principles and
following a SD: 10.5 periods). Upon auditory cues at included thoracic
single cortical individually determined rates spine mobilization,
or subcortical of 0.67 to 0.97 Hz, participants scapular
ischemic stroke. pushed and pulled bilaterally, in mobilization, weight
All patients had synchrony or alternation, 2T-bar bearing with the
the ability to move handles sliding in the transverse paretic arm, and
the affected limb plane. opening a closed
and had completed fist.
3 to 6 months of
conventional rehab.
Adequate language
and neurocognitive
function to
understand
instructions.
Hemiplegia:
Gr1: 6 R; 3 L
Gr2: 8 R; 4 L
Appendix 3B
151

continues
APPENDIX 3B. Continued
152

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Mudie Total: 40 Patients had suffered Within 2–6 72.4 Gr1: Portable computer-like Gr3: Bobath- N/A 30-min N/A 2, 2, 1
2002119 Gr1: 10 a recent stroke and wk SD: 9.01 yr balance performance monitor trained staff sessions for
Gr2: 10 consistently bore (BPM) feedback console used physiotherapists 2 wk
Gr3: 10 majority of weight on to provide awareness of weight focused on
Gr4: 10 one side. distribution during training increasing trunk
in sitting; patients required to and pelvic ROM,
touch target with nonparetic normalizing
hand at various heights and trunk muscle
distance; patients attempted to tone, maintaining
return to symmetrical position appropriate balance
after reaching. responses during
Gr2: Patient seated on reaching and
adjustable plinth; grocery items improving the
placed at about 140% of arm’s patient’s ability to
length were retrieved with move in and out
TOPICS IN STROKE REHABILITATION/SPRING 2006

nonparetic extremity and placed of asymmetric

on shelves at various heights posture; verbal and
and distances to allow reaching manual facilitation
to the extremes of the seated by therapist during
base of support. seated reaching or
lying.
Gr4: Control:
Standard PT and
OT.
Nelson Total: 26 Inclusion: Unilateral 53.5 days 68.4 Apparatus: Handle—when Gr2 (rote Preliminary 1 visit N/A 2, 0, 1
1996168 Gr1: 14 hemiplegia caused SD: 51.9 SD: 11.2 rotated to 40º first die falls exercises): Cap exercises: 3 reps of
Gr2: 12 by a first and only noisily and the last (3rd) die on handle weighs reaching between
M/F: 14/12 CVA; duration since falls at 90º. Rotation of the approximately the knees to floor,
onset of the stroke handle met with little resistance. same as the 3 dice. reaching forward,
between 9 days and Gr1: Instructions given: Instruction: Exercise reaching toward
7 months; age of at Exercise your arm by playing your arm by using uninvolved side,
least 45 yr; pronator a dice game. Patients were the handle. reaching toward
spasticity; full passive encouraged to get doubles or involved side,
supination after brief 3 in a row. 2 trials of manual moving wrist from
warm-up exercises; guidance– therapist assisted ulnar deviation to
no functional patient in grasping firm bilateral radial deviation.
supination; ability to grasp of the handle and
grasp a 3-cm dowel manually guided the patient
bilaterally with through the motion until all
overlapping fingers; 3 dice dropped. Slow and
no contraindication steady movement emphasized.
to supination Then patient had 10 trials by
exercises; and themselves, a 3-min break and
sufficient visual then 10 additional trials.
perception and
comprehension.
Hemiplegia:
15 R; 11 L

Parker Total: 466 Inclusion: Acute Less than 6 Gr1: Gr1: Leisure. Patients practiced Control: No OT N/A Minimum of Follow-up at 2, 0, 1
2001169 Gr1: 153 stroke with onset of months Median: 72 the leisure tasks and any ADLs treatment within the 10 sessions 26 wk and
Gr2: 156 less than 6 months; SD: 7 tasks necessary to achieve the trial. lasting not 52 wk
Gr3: 157 participating in a leisure objective. less than 30
rehab program. Gr2: Gr2: ADLs. Patients practiced min each
Hemiplegia: Median: 71 the independent self-care tasks
Gr1: 71 R; 69 L SD: 6 yr (preparing a meal or walking
N/B: 13 outdoors).
Gr2: 75 R; 65 L Gr3:
N/B: 16 Median: 72
Gr3: 67 R; 74 L SD: 6.5 yr
N/B: 16
Treated area:
Functional
Appendix 3B
153

continues
APPENDIX 3B. Continued
154

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Pomeroy Total: 24 Inclusion: 6 months At least 6 N/A Gr1: “Weight garments”: Simple Gr2: Control. N/A Gr1: 6 wk N/A 2, 0, 0
2001170 Gr1: 12 post stroke; not months wrist, biceps, thigh, and ankle After the 6-wk with the
Gr2: 12 participating in other bands, in addition to complex intervention phase, garments for
physical rehab; could bands around pelvis and the 12 control intervention
walk 10 meters with shoulder girdle. Worn on the patients were given phase and
or without assistance paretic side only. Patients were the garments and nothing
or walking aid; had instructed to adjust the weights asked to complete for control
no receptive aphasia and garment on a daily basis the logbooks. phase.
(scoring 18 or above according to how able they felt During wk 8–12 Gr2: First 6
on Body part and and were asked to wear the they were contacted wk control
13 or above on garments all day. An instruction by telephone. then 6 wk of
Commands section); sheet was given. During wk 2 treatment.
had no visual to 6, each patient was contacted
unilateral neglect; by telephone once a week.
could don and doff
TOPICS IN STROKE REHABILITATION/SPRING 2006

the garments alone

or with help; had
laundry access;
had no neuro- or
musculoskeletal
disorder; had no
severe heart disease;
and were not taking
sedative during
treatment time.
Hemiparetic:
8 R; 16 L
Richards Total: 27 Inclusion: Patients Gr1: Gr1: 69.6 Gr1: Experimental: Patients Gr2: Standard Standard 2 sessions/ Follow-ups 1, 0, 0
1993171 Gr1: 10 who met the 8.3 days SD: 7.4 started as early as possible customary care. customary day, 5 days/ at 6 wk and
Gr2: 8 following criteria: SD: 1.4 days after admission to the study; Started later; hospital care wk, for 5 wk 18 wk
Gr3: 9 were resident within Gr2: 67.3 they were provided with an performed
50 km of Quebec Gr2: SD: 11.2 intensive and focused approach nonintensive
City; were between 8.8 days to therapy that incorporated techniques similar
40–80 yr old; had SD: 1.5 days Gr3: 70.3 the use of a tilt table and a to those provided
onset of first stroke SD: 7.3 limb-load monitor; resistance to Gr3.
within past 0–7 days Gr3: exercises with a Kinetron Gr3: Early standard
(first episode of TIA 13 days isokinetic device and a customary care.
lasting < 24 hr not SD: 2.8 days treadmill. Started early; more
considered a first intensive than
stroke); and had experimental group
clinically identifiable but contained
middle cerebral more standard
artery syndrome of approaches.
thrombo-embolic
origin involving
subcortical structures
confirmed by CAT
scan.
Hemiplegia:
Gr1: 2 R; 8 L
Gr2: 6 R; 2 L
Gr3: 3 R; 6 L

Walker Total: 30 Inclusion: Patients 6 months after Gr1: 65.9 Gr1: Dressing practice was Gr2: No treatment. All other rehab End of N/A 1, 0, 1
1996172 Gr1: 15 who had been stroke SD: 8.16 given on a regular basis with continued as treatment at
Gr2: 15 discharged from the the amount of therapy at the usual. 3 months;
general medical care Gr2: 70.2 therapist’s discretion. median of
of the elderly and SD: 10.35 Treatment: Teaching patients 8 visits at
stroke unit. and caregivers appropriate the patient’s
Hemiplegia: techniques such as dressing home
Gr1: 6 R; 9 L the affected extremity first,
Gr2: 10 R; 5 L conserving energy, using a
Treated area: Affected red alignment of buttons, and
and unaffected choosing clothing. Relatives
extremity were encouraged to continue
the dressing practice between
sessions.
Appendix 3B
155

continues
APPENDIX 3B. Continued
156

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Wiart Total: 22 Patient with severe Gr1: 35 days Gr1: 66 Gr1: Experimental group: Part Gr2: Control: 3 N/A 1 hr every Follow-up at 2, 0, 1
1997173 Gr1: 11 left unilateral neglect SD: 9 days SD: 8 1: Patient wore a vest. The –4 hr of traditional day for 20 4 wk
Gr2: 11 syndrome who had Motor extremity of the pointer was 1.5 rehab each day. days followed
suffered a stroke less impairment: Gr2: 72 meters in front and above the by 2–3 hr of
than 3 months ago. hemiplegic SD: 6 patient’s head and forced him traditional
Patient with good or (6), to make an axial rotation of the rehab: (2 hr
poor trunk control. hemiparesis trunk under visual control to PT and 1
Hemiplegia: Right for (3), none (2) displace the pointer laterally hour OT)
both groups and explore the spatial field.
Treated area: Trunk Gr2: 30 days Part 2: A series of targets of
SD: 4 days different geometric form,
Motor house an electrical circuit, are
impairment: connected to a light bulb and
hemiplegic to a buzzer, are attached to a
(8), mobile wooden panel placed
TOPICS IN STROKE REHABILITATION/SPRING 2006

hemiparesis in front of the patient. When

(2), none (1) pointer is in contact with target
= audible and luminous signals
(biofeedback).
Winstein Total: 60 First-time stroke N/A Gr1: Gr1: Standard care delivered Gr3: Strengthening N/A 1 hr/day, 5 Follow-up at 2, 0, 1
2004127 Gr1: 20 from infarction in the <35 yr: 2 by an OT; muscle facilitation and motor control days/wk, 4 9 months
Gr2: 20 anterior circulation SD: 10 exercises; neuromuscular training; eccentric, wk
G3: 20 confirmed by MRI 35–75 yr: 18 electric stimulation for shoulder isometric, or
or CAT scan; onset SD: 85 subluxation; stretching exercise; concentric muscles
of stroke from 2 >75 yr: 0 ADL; and self-care. contractions
to 35 days before SD: 0 Gr2: Functional training: performed in a
study entry; and Systematic and repetitive gravity-lessened
an FIM instrument Gr2: practice of tasks arranged in position or against
total score upon <35 yr: 0 proximal to distal recovery gravity; progressed
admission of 40 to SD: 0 patterns of reaching and against resistance
80. 35–75 yr: 19 grasping. using free-weights,
SD: 95 theraband, or grip
>75 yr: 1 devices for the
SD: 5 fingers; alternate
days for 3 days/wk;
Gr3: on other days,
<35 yr: 0 same exercises with
SD: 0 less resistance and
35–75 yr: 20 greater speed.
SD: 100
>75 yr: 0
SD: 0

Note: ADLs = activities of daily living; CVA = cerebrovascular accident; Gr = group; L = left; N/A = not available; N/B = neither/bilateral; OT = occupational therapy; PT = physiotherapy; Quality R, B, W = randomization,
blinding, withdrawals; R = right; ROM = range of motion; TIA = transient ischemic attack; UE = upper extremity; x/wk = times per week.
Appendix 3B
157
APPENDIX 3C. Characteristics of Included Studies for Biofeedback
158

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Armagan Total: 27 Inclusion: Patients Gr1: 4.43 months Gr1: 57.00 Gr1: Exercise program Gr2: Same as Gr1 Exercise program 5 x/wk for N/A 2, 1, 0
2003196 Gr1: 14 had to be able to SD: 1.09 months SD: 10.53 according to Brunnstrom’s except the patients according to 20 min; 20
Gr2: 13 communicate, have approach and EMG-BFB. received a placebo Brunnstrom’s treatment
full comprehension, Gr2: 4.77 months Gr2: 57.92 Surface electrodes were EMG-BFB. neurophysiologic sessions in
no visual or SD: 1.30 months SD: 11.27 placed over the musculus approach with total
auditory defects, extensors, carpi radialis, a duration of
have significant and extensor digitorum 45 min/day for
motivation, have communis of the paretic a period of 20
stable health status, arm. treatments.
and have stage 2 or 3
hemiparesis according
to the modified
Brunnstrom’s scale for
the hand.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Hemiplegia:
Gr1: 7 R; 7 L
Gr2: 7 R; 6 L
Treated area: Upper
extremity

Basmajian Total: 13 Patients must have Gr1: 3.5 months Gr1: 65.00 Gr1: Experimental group: Gr2: Control N/A 3 x/wk, 40 N/A 2, 0, 1
1982197 Gr1: 8 had a stroke 2 to SD: 1 month SD: 9.75 Integrated behavioral group: Standard min each
Gr2: 5 5 months before physical therapy exercise physical session for
(only inclusion, be referred Gr2: 2.8 months Gr2: 62.00 including EMG-BFB. therapy program 5 wk
the mild by family physicians, SD: 0.9 months SD: 6.50 Experimental technique using a general
group) and show a residual of regular physical neurophysiological
defect in upper (all 37 patients) (all 37 therapy as practiced in approach.
extremity function. patients) the hospital with the
Treated area: Upper addition of EMG-BFB.
extremity
Basmajian Total: 29 Patients had to Gr1: 16.38 wk Gr1: 60.77 Gr1: Integrated Gr2: Standard N/A 45 min/ Follow-up at 8 2, 0, 1
1987198 Gr1: 13 have: no previous SD: 7.60 wk SD: 8.50 behavioral and physical therapeutic session, 3 months
Gr2: 16 history of stroke; an therapy (EMG-BFB): exercise: x/wk for 5 wk
age of 80 yr or less Gr2: 16.00 wk Gr2: 63.77 1) Conceptualization: Techniques of
with an obstruction SD: 11.74 wk SD: 13.07 Demonstrations and facilitation and
involving the territory dialogue, patient active inhibition are
with distribution participant in treatment. used with selected
centering on the 2) Skill acquisition: EMG- sensory input
middle cerebral BFB goals are learned and to bring about
artery; impaired the patient is taught how automatic, high-
motor function of to direct cognitive skills. quality motor
the upper extremity 3) Repeated skill: Home output.
but with an ability to practice of the specific
extend wrist/finger; skills is emphasized and
have had their stroke the results are carefully
<1 year before; and monitored and discussed
comprehension of with the patient. 4)
simple commands Skill transfer: Using
and the purpose the Cyborg BL 900
of the treatment. dual processor: pairs of
Patients also had to standard surface BFB
be well informed, electrodes to the patient’s
sufficiently motivated, skin over muscles that are
and willing to being targeted for either
participate, have the active recruitment or
approval of their active inhibition.
physician, a relatively
uncomplicated
medical history,
and mild motor
involvement of the
hemiplegic upper
extremity (> or = 20
UEFT). If it had been
> or = to 4 months
since their stroke,
the mild group: or
less than 4 months
post-stroke with <20
(UEFT).
Hemiplegia:
Gr1: 9 R; 4 L
Appendix 3C

Gr2: 7 R; 9 L
Treated area: Upper
extremity
159

continues
APPENDIX 3C. Continued
160

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Binder Total: 10 Patients had to be 16 months post N/A Gr2: Experimental Gr1: Control: N/A 4-wk N/A 1, 0, 1
1981199 Gr1: 5 at least 16 months onset (therapeutic exercise Standardized treatment
Gr2: 5 post-onset; have regimen combined therapeutic exercise
intact proprioception; with EMG-BFB): EMG- regimen.
be able to walk on BFB was used to help The last patient of
level surfaces with the patient initiate a each group agreed
or without assistive contraction when he to receive 12 more
devices; have residual was unable voluntarily treatments over an
lower extremity to produce muscle additional 4 wk.
motor deficits, activity, maintain an
including dorsiflexion active muscle contraction
paresis; have no when the strength of
signs of receptive a muscle contraction
aphasia; no previous began to improve, relax
TOPICS IN STROKE REHABILITATION/SPRING 2006

EMG-BFB training; a muscle contraction in a

and no concurrent hypertonic antagonist or
engagement in a in an agonist following
physical therapy active contraction, and
program. improve maintenance of
Treated area: Lower motivation by providing
extremity EMG-BFB to the patient
as soon as an activity was
performed.
The frequency of audio
feedback (audio tone)
increased with increasing
muscle activity. A pair of
small length and meter
displays provided visual
feedback.
Bourbonnais Total: 25 Resident within 35 Gr1: 37.3 months Gr1: 47.2 Gr1: Using visual Gr2: A total of N/A 3 x/wk for Follow-up at 2, 0, 0
2002200 Gr1: 13 km of the Institute, SD: 14.3 months SD: 13.9 feedback, the patients 16 directions 6 wk 8 wk
Gr2: 12 aged between 21–70 were asked to control were selected
yr, single history Gr2: 34.7 months Gr2: 44.6 combinations of moments corresponding to
of hemiplegia or SD: 16.1 months SD: 14.1 of force in two directions predetermined
thrombo-embolic of either a single or two combinations of
origin, and stage 3, joints or to combine a flexion-extension
4, 5, or 6 of the arm moment of force in one movements starting
section on Chedoke direction with a handgrip with the hip and
McMaster stroke force. A sequence of 16 knee. The patient’s
assessment. combinations of moments task was to exert
of force, progressing from submaximal efforts
proximal to distal, was in each of the 16
developed to increase the directions chosen
level of difficulty. by using a cursor
The Gr1 was also control that is displaced
group for the lower in proportion to
extremity intervention. the magnitude of
Phase 2: Standard PT (see effort provided and
control group treatment) in the direction in
for 4 wk. which the effort is
The knee monitor was produced. The Gr2
not used in phase 2 in was also control
order to evaluate transfer group for the
of training to performance upper extremity
without EMG-BFB. intervention.
Importance of visual
input in maintaining a
stable standing posture
has been shown in
healthy adults. Other
activities: targeting,
reaching, stride standing,
and stepping.
Appendix 3C
161

continues
APPENDIX 3C. Continued
162

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Bradley Total: 13 Patients had to be 35.6 days post Gr1: Gr1: EMG-BFB group: Gr2: Same Standard 6-wk Follow-up at 3 0, 2, 1
199850 Gr1: 7 medically stable stroke (for all the 72.4 yr Patients were encouraged technique with the customary care treatment months
Gr2: 6 and have unilateral groups) SD: N/A to facilitate or inhibit surface electrodes.
(severe weakness. abnormal muscle tone via EMG-BFB machine
stroke Hemiplegia: Gr2: auditory or visual signals was turned off and
only) Gr1: 5 R; 2 L 68.0 yr transmitted from surface faced away from
Gr2: 4 R; 2 L SD: N/A electrodes placed over the the patient and
appropriate muscles (on therapist to control
the motor point of the the placebo effect.
affected side). Treatment No visual/auditory
also encouraged normal feedback given
alignment of the trunk/ (standard
hip, active control of the customary care).
hip/knee/ankle, weight EMG 200 (Biodata,
TOPICS IN STROKE REHABILITATION/SPRING 2006

transfer through the Manchester)

affected side, and gait
reeducation.
Burnside Total: 22 Patients had had a Disease duration Gr1: 71.45 Gr2: 15 min of exercises Gr1: 15 min of Therapeutic 6-wk Follow-up at 0, 2, 0
1982201 Gr1: 11 CVA at least 3 months Gr1: 5.2 yr SD: N/A with EMG feedback exercises similar exercise treatment 6 wk
Gr2: 11 before the study and SD: N/A regarding their efforts. to those used in
had residual foot Gr2: 4.45 yr Gr2: 68.45 Electrodes were attached Gr2. Electrodes
dorsiflexion paresis SD: N/A SD: N/A over the muscle being were placed over
and a minimum trained at that time. the tibialis anterior
passive dorsiflexion If there was spasticity, and the machine
from complete the 2nd machine was switched off
plantar flexion to was connected to the (eventually, patients
neutral position (with gastrocnemius while were given gait
or without a cane/ the 1st was connected training with or
brace). to the tibialis anterior. without EMG-BFB
Hemiplegia: Patients were encouraged depending of their
Gr1: 5 R; 6 L to practice walking as grouping).
Gr2: 6 R; 5 L much as possible between
Treated area: Lower sessions.
extremity The patients sat on
an adjustable chair in
front of the trainer with
their feet on the dual
force platform and they
rose up and sat down
as symmetrically as
possible, again by using
the trainer’s visual and
auditory signals. They
repeated the movement
for 20 min under the
therapist’s supervision.
Appendix 3C
163

continues
APPENDIX 3C. Continued
164

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Cozean Total: 36 Medically stable No. of patients Gr1: Gr2: EMG-BFB: Gr1: Passive and Rehabilitation 6-wk Follow-up at 1, 1, 1
198851 Gr1: 8 patients referred more than 1 year mean: 62 Electromyographic active ROM for all program: Details treatment 6 wk
Gr2: 8 to the center for post-CVA signals were received major joints and N/A
Gr3: 8 rehab following a Gr1: 5 Gr2: with surface electrodes muscle groups of
Gr4: 8 CVA, who had the Gr2: 4 mean: 51 over the muscle surface. both legs.
cognitive ability to Gr3: 4 Patient was asked to Gr3: Electrical
cooperate and give Gr4: 3 Gr3: produce a 30 s maximal stimulation of the
informed consent and mean 52 contraction followed by a tibialis anterior
the physical ability 1-min rest. during the swing
to ambulate a short Gr4: phase and the
distance with the mean: 56 gastrocnemius
assistance of a single during the stance
therapist. phase of the gait
Hemiplegia: cycle.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr1: 5 R; 3 L Gr4: 15 min of

Gr2: 6 R; 2 L treatment with
Gr3: 4 R; 4 L EMG-BFB + 15-min
Gr4: 3 R; 5 L treatment of FES
Treated area: Lower (the first 2 wk).
extremity The next 4 wk: 5
min of FES + 30
min of independent
use.
Crow Total: 40 Patients had had a Between 2 and 8 Gr1: 67.4 Gr1: EMG-BFB group: Gr2: Control Standard 6-wk Follow-up at 2, 0, 1
1989202 Gr1: 20 stroke between 2–8 wk after stroke SD: 10.45 Biodata EMG-120 group: There was customary care treatment 6 wk
Gr2: 20 wk before inclusion, biofeedback system, no skin or electrode
had some arm Gr2: 68.05 silver-silver chloride preparation and
function (i.e., at least SD: 9.53 Beckmann-style the system was
a flicker of activity electrodes were used with switched on but
around the shoulder saline conducting gel. The auditory and visual
girdle), had not electrodes’ placement was feedback were
already spontaneously based on muscle activity not switched on.
recovered, or did not at each treatment session The system was
have near normal (i.e., not a predetermined positioned near the
movement. Patients position for each muscle). patient to allow for
were also living in the Auditory and visual placebo effect and
Nottingham area and feedback were switched electrodes were
did not have global on. placed over the
aphasia or dementia. upper fibers of the
Hemiplegia: trapezius, pectoral,
Gr1: 12 R; 8 L or deltoid muscle.
Gr2: 14 R; 6 L
Treated area: Upper
extremity
Appendix 3C
165

continues
APPENDIX 3C. Continued
166

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Intiso Total: 16 Patients were 40–85 Gr1: 11.3 months Gr1: 61.3 Gr1: Experimental group Gr2: Control: Only Standard physical 2 months N/A 1, 0, 1
1994203 Gr1: 8 yr old, had had a SD: 12.6 months SD: 12.3 Phase 1: Contraction physiotherapy therapy
Gr2: 8 first ischemic stroke, of the anterior tibial (standard exercises (no details)
and had a foot-drop Gr2: 8.3 months Gr2: 53.5 muscle to obtain acoustic according to the
deficit or paretic SD: 6.0 months SD: 18.5 BFB. 2 baseline sessions Bobath method for
extremity. about the technique and dorsiflexion) (only
Hemiplegia: learning to contract the for the control
Gr1: 4 R; 4 L anterior tibial. For the group).
Gr2: 5 R; 3 L next 15 sessions: 20 Sensory outputs to
Treated area: Lower isotonic contractions obtain automatic
extremity lasting 5 s followed by high-quality motor
regular intervals of 30 s output.
of rest (flexed knee at 30º
and relaxed foot). Phase
TOPICS IN STROKE REHABILITATION/SPRING 2006

2: 2 baseline sessions to
determine the “threshold
value” and 15 training
sessions with acoustic
feedback activated every
time the contraction
value of the anterior tibial
muscle was higher than
the threshold value of the
baseline session.
Audio monitoring of the
EMG signals: Patient
instructed to activate
dorsal flexion of foot
during swing phase of
gait cycle. Patient errors
< 20%: increase the
threshold of acoustic
feedback 0.30 µV; if
20%–50%: increase to
0.10 µV.
Kraft Total: 18 At least 1 year Gr1: 26.0 months Gr1: 59.5 Gr1: EMG-BFB initiated Gr2: Bias/Balance Home rehab Depends on Follow-up 12 0, 0, 1
1992204 Gr1: 6 post-CVA, clinically SD: 23.4 months SD: 6.2 electrical stimulation stimulation: program the treatment months post-
Gr2: 4 stable, possessed of wrist extensors: 3 Respond II (no details) treatment
Gr3: 3 postural stability, and Gr2: 36.8 months Gr2: 64.8 sessions/wk, 36 1-hr electrical
Gr4: 5 had consented to SD: 19.8 months SD: 11.6 sessions. Transcutaneous stimulator: 5 30-
participate according electrical stimulation, min sessions/wk
to procedures. Gr3: 14.3 months Gr3: 67.0 triggered by a low- for 3 months,
Hemiplegia: SD: 2.5 months SD: 3.6 level of voluntary EMG low-intensity
Gr1: 2 R; 4 L activity began as low as stimulation was
Gr2: 2 R; 2 L Gr4: 24.2 months Gr4: 63.2 5 µV and was increased applied to wrist
Gr3: 2 R; 1 L SD: 6.0 months SD: 12.3 gradually by the therapist extensors muscles
Gr4: 1 R; 4 L as voluntary recruitment in 0.3 ms square-
improved. In addition, wave pulses at
EMG-BFB was used rates of 30–90
over paretic finger or Hz. Stimulation
elbow extensors, forearm was applied at
pronators or supinators, an intensity that
or shoulder elevators increased the
or abductors according patient’s voluntary
to the patient’s abilities. range of wrist
Auto move stimulator in extension without
0.2 ms, biphasic square- producing any
wave pulses at rates of visible movement
30–90 Hz, constant at rest. Patients
current of 20–60 µV for were instructed
10 s. EMG µV threshold to perform 3 sets
set at the median of the 5 (30 contractions
contractions minus 10%. or until fatigue)
of voluntary wrist
extension exercises
while stimulation
was applied each
session.
Gr3: PNF:
Treatment
encompassed
the whole upper
extremity, including
wrist extension.
Gr4: No treatment.
Appendix 3C
167

continues
APPENDIX 3C. Continued
168

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Mandel Total: 37 Inclusion: Patients Time since stroke Gr1: 56.8 Gr2: Auditory and visual Gr1: No treatment N/A Twice a wk Follow-up at 3 1, 0, 1
1990205 Gr1: 11 who met the onset: SD: 12.8 feedback of calf and for 12 wk months
Gr2: 13 following criteria: pretibial muscle activity
Gr3: 13 were a chronic Gr1: 32 months Gr2: 54.7 during active ankle
hemiparetic; were SD: 31 months SD: 13.9 movements. Minimum
at least 6 months 5 reps of maximum
post-stroke, with Gr2: 17 months Gr3: 57.5 contraction/trial,
various classifications SD: 10 months SD: 14.2 maximum 240 repetitions
of lower extremity per session.
deficits; and were no Gr3: 47 months Gr3: EMG-BFB for first
longer receiving any SD: 32 months half of study, followed
treatment aimed at by rhythmic positional
improving gait. BFB for last half. RP-BFB:
Hemiplegia: Feedback of dorsiflexion
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr1: 7 R;4 L and plantar flexion

Gr2: 6 R; 7 L ROM. Use of auditory
Gr3: 8 R; 5 L and visual pacing stimuli
aimed at increasing the
rate of dorsi- and plantar
flexion alternation. While
walking, feedback at
precise points of heel-off
and swing through to
reinforce correct timing
events.
Montoya Total: 14 Inclusion: Patients Gr1: < 6 months Gr1: 64 Gr1: BFB group: Initial Gr2: Reference N/A 45 min/ N/A 1, 0, 0
1994206 Gr1: 9 who had had a first- SD: 2.75 6-meter walk with group: Patients session, 2
Gr2: 5 ever stroke that had Gr2: < 6 months locometer analysis to followed the same sessions/wk
occurred less than Gr2: 60 obtain initial values of standard rehab for 4 wk
6 months before SD: 2.25 spatial and temporal program and walk
the beginning of parameters. Patients were session without
the experimental told to secure the best BFB. Description of
protocol; who walking performance. a session:
had motor ability Five 6-meter walks with 1) 6-meter walk
allowing them to BFB: the imposed right with locometer
walk 50 meters and left step lengths were analysis;
without any help; the initial step length 2, 3) 5 and 10 6-
who had an absence measures in the initial meter walks on the
of comprehension 6-meter walk. Patient had walkway without
problems as rated to pay attention to the the BFB;
by a Mini Mental task and place the foot 4) a final gait
Status of 24; who on the target. No effort analysis under the
had an absence was required to correct same conditions as
of biomechanical step length asymmetry there were in the
malfunction of or to increase step length BFB group.
the legs; who on the deficient side.
had an absence 10 walks with the BFB
of hemianopsia where the imposed right
as stated by a and left step lengths
complete ophthalmic were identical in order to
examination correct the asymmetry.
including visual field
measurement; and
who had an absence
of unilateral neglect
syndrome as rated
by a neurological
examination showing
that there was no
sensory trouble and
astereognosia.
Treated area: Lower
extremity
Appendix 3C
169

continues
APPENDIX 3C. Continued
170

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Morris Total: 26 Patients had suffered Gr1: 79.0 days Gr1: 64.2 Gr2: Experimental Gr1: Control group: N/A 8 wk N/A 0, 0, 1
199216 Gr1: 13 a CVA up to 4 SD: 41 days SD: 10.3 group: Phase 1: Electro- Standard physical
Gr2: 13 months previously, goniometric feedback as therapy that was
were medically stable, Gr2: 45.0 days Gr2: 64.4 an adjunct to physical based on the
were able to provide SD: 23 days SD: 11.0 therapy for 4 wk. EMG- Motor Relearning
informed consent, BFB during standing and Program and
and demonstrated gait training: when the incorporated the
the ability to safely patient hyperextended use of instruction,
walk 10 meters 4 the knee, the auditory explanation,
times without aids or signal was activated demonstration,
orthoses. (the pitch of the signal manual guidance,
Hemiplegia: was proportional verbal feedback,
Gr1: 8 R; 6 L to hyperextension visual feedback,
Gr2: 8 R; 5 L angle). A fixed high and whole-part
TOPICS IN STROKE REHABILITATION/SPRING 2006

frequency warning practice.

signal sounded when Phase 1: 4 wk
knee flexion exceeded Phase 2: 4 wk
75º to discourage
overcompensation with
excessive knee flexion
(during normal gait
training, no signal).
Phase 2: Standard PT (see
control group treatment)
for 4 wk.
The knee monitor was
not used in phase 2 in
order to evaluate transfer
of training to performance
without EMG-BFB.
Importance of visual
input in maintaining a
stable standing posture
has been shown in
healthy adults.
Other activities: targeting,
reaching, stride standing,
and stepping.
Tham Total: 14 Patients had Gr1: 18.1 days Gr1: 67.3 Gr1: Video EMG- Gr2: Conventional N/A 2 days N/A 0, 0, 1
1997207 Gr1: 7 suffered from right SD: 7.67 days SD: 10.42 BFB group: Therapist BFB group: The
Gr2: 7 hemisphere CV videotaped the patients therapist gave the
lesions 6 to 42 days Gr2: 21.6 days Gr2: 68.6 when performing the patients verbal and
before testing SD: 12 days SD: 12.33 baking tray test (BTT). visual guidance to
Treated area: Upper To draw the patients’ help them see the
extremity Acute phase post- attention to their own neglected left half
stroke neglect behavior, of the “baking tray.”
they were given the
opportunity of seeing
their performance on
the monitor directly
after the testing. When
they missed placing the
“buns” on the left side of
the tray, they could see
this on the right side of
the TV monitor. After
video confrontation,
patients were asked to
describe and comment on
their BTT. The therapist
commented on the
performance and the
results. The patients were
asked to formulate their
own strategies to perform
better on tasks like
the BTT. The therapist
discussed possible
compensatory strategies
with the patients and gave
instructions on how to
use tactile discrimination
with the right hand to
find out where the left
edge on the tray was
and then systematically
place the buns from left
to right, upside down.
The therapist presented
the same compensatory
Appendix 3C

strategy in both video and

conventional groups.
171

continues
APPENDIX 3C. Continued
172

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Tham 1997 Lines: The number of

(continued) marked lines (%) was
recorded.
Test 3: Figure Copying
Task: The severity of
neglect was rated as
+ if objects on the left
side of the drawing
were omitted, as ++ if
the entire left side was
omitted, and as +++ if
some objects in the right
side were omitted.
Test 4: Line Bisection:
TOPICS IN STROKE REHABILITATION/SPRING 2006

Patients were presented

with 15 lines in random
order, 5 lines for each
length of 25, 100, and
200 mm. Deviations from
the true midpoint were
measured to the nearest
millimeter, and a mean
error was calculated for
each patient across 15
lines.
Computer corresponded
to a preset target. To
encourage weight shifting,
the patients moved their
CoG and observed the
corresponding cursor
movement on the
computer screen.
Williams Total: 20 Patients had Between 3–16 wk Mean: 63.5 Gr1: Instructions in Gr2: 30 min on 2 Standard Gr1: 5 x/wk Follow-ups 1, 0, 0
1982208 Gr1: 10 a diagnosis of SD: 11.8 yr relaxation given with consecutive days customary care for 1 wk at 1 wk (end
Gr2: 10 hemiparesis or the first two EMG-BFB at beginning of program (no Gr2: 2 x/wk of treatment)
hemiplegia secondary treatments; 20–25 treatment week; details): 1 hr for 1 wk and 2 wk (end
to vascular disease, min of EMG-BFB on Jacobson method of given by various of treatment
were between 3–16 5 consecutive days; 2 relaxation taught in physiotherapists => after cross-
wk post stroke, had surface electrodes placed the sitting position in the over).
a painful shoulder over target muscles’ with involved department.
on hemiplegic side, upper tendinous area of upper extremity
had no previous latissimus dorsi and teres supported on
history of shoulder major; ground electrode a pillow; and a
pain prior to stroke, on acromion; and patient reflex-inhibiting
had ability to follow was encouraged to relax pattern (RIP) was
instructions and did to decrease electrical encouraged.
not show any signs activity of muscles.
of receptive aphasia,
could understand
English or French,
and were not taking
any medication to
reduce spasticity.
Hemiplegia:
8 R; 12 L
Appendix 3C
173

continues
APPENDIX 3C. Continued
174

Age, yr Session
Author and Sample (mean, SD Concurrent frequency Follow-up Quality
year size Population details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Wolf Total: 37 Patients who had Gr1: 2.2 yr Gr1: 51.2 Gr1–3: All but one of Gr4: Control: All patients 2–3 sessions/ N/A 0, 0, 1
1983a209 Gr1: 6 sustained a CVA SD: 1.2 yr SD: 13.7 these patients received 60 Without had undergone wk around 3
Gr2: 6 at least 1 yr before EMG-BFB sessions. The intervening previous rehab months
Gr3: 6 their participation, Gr2: 2.2 yr Gr2: 55.0 remaining patient elected treatment. (no defined
Gr4: 6 undergone previous SD: 1.2 yr SD: 12.4 to discontinue after 40 Gr5: EMG-BFB number of
Gr5: 16 rehab, and who did treatments. EMG-BFB treatment for the sessions)
Gr6: 8 not have receptive Gr3: 2.4 yr Gr3: 53.8 provided by Hyperion involved UE. LE
aphasia. SD: 1.2 yr SD: 13.8 4080 Biocoonditioner was also examined
Treated area: Lower or the Cyborg J53. for the generalizing
extremity Gr4: 2.6 yr Gr4: 57.2 Protocol: Feedback to effect of upper
SD: 2.2 yr SD: 11.9 relax hyperactive muscles extremity treatment
to recruiting weakened upon lower
Gr5: 4.1 yr Gr5: 56.1 antagonist muscles in a extremity activity
SD: 3.8 yr SD: 13.8 proximal (hip) to distal changes.
TOPICS IN STROKE REHABILITATION/SPRING 2006

(ankle) manner. Initial Gr6: General

Gr6: 3.1 yr Gr6: 53.7 treatment in a supine relaxation group:
SD: 3.2 yr SD: 19.2 position with progression 20 relaxation
to sitting, standing, sessions: Patients
and finally, ambulatory were taught
activities. autogenic phases
Examination at 3 months. and underwent
tense-relax
procedures for
specific muscle
groups while
receiving audio
BFB proportional
to activity from the
target muscles.
Examination at 2
months.
Wolf Total: 31 Patients who had Gr1: 3.8 yr Gr1: 55.3 Gr1: Experimental group: Gr2: Control N/A 6 months Examinations 0, 0, 1
1983b210 Gr1: 22 sustained a CVA at SD: 3.6 yr SD: 14.8 Patients were trained group: No rehab at 3, 5, or
Gr2: 9 least 1 yr before this in 45 to 60 sessions to treatment. 8 months
study, who had no Gr2: 2.4 yr GR2: 55.9 relax specific hyperactive (approximately
previous EMG-BFB SD: 1.6 yr SD: 10.9 muscles through the 6 months)
training, and no provision of audio
evidence of receptive and visual EMG-BFB
aphasia. detected with surface
electrodes (attempts
to recruit weakened
antagonist muscles).
Patients received
feedback from spastic
muscles to reinforce
previous training directed
toward their inhibition.
Biofeedback proceeded
in a proximal to distal
direction with each
patient progressing from
isolated joint movements
to manipulative efforts
requiring voluntary
stabilization of proximal
musculature.
Appendix 3C
175

continues
APPENDIX 3C. Continued
Author and Sample
year size Population details
Wolf Total: 16 Patients who
1994211 Gr1: 8 had hemiplegia
Gr2: 8 secondary to a CVA,
a time since onset
of 1 yr or longer, an
involved extremity in
Brunnström stage 3
or 4 for longer than
6 months since the
patient’s most recent
session of PT or OT
for UE, an absence
of receptive aphasia,
visual fields deficits
and proprioceptive
deficits in the
involved elbow, and
an ability to obtain at
least 60º of passive
shoulder flexion and
abduction.
Hemiplegia:
Gr1: 3 R; 5 L
Gr2: 5 R; 3 L
Note: CoG = center of gravity; CVA = cardiovascular accident; EMG-BFB: electromyographic biofeedback; FES = functional electrial stimulation; Gr = group; L = left; LE = lower extremity; N/A = not available; N/M = not
measured; OT = occupational therapy; PFST = positional feedback stimulation training; PNF = proprioceptive neuromuscular facilitation; PT = physical therapy; Quality R, B, W = randomization, blinding, withdrawals; R =
right; ROM = range of motion; UEFT = upper extremity function test; UE = upper extremity; x/wk = times per week.
176
Age, yr Session
(mean, SD Concurrent frequency Follow-up Quality
Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)
Gr1: Gr1: 63.88 Gr1: EMG-BFB group: Gr2: Non-feedback N/A 25 min/ N/A 1, 0, 1
32.63 months SD: 10.88 Uptraining of the triceps group: Same task session, 2–4
SD: 16.37 months muscle during the elbow performance and sessions/wk
Gr2: 62.00 extension task, using progression, except for a total of
Gr2: SD: 14.39 one EMG channel of that EMG-BFB 10 sessions
65.50 months the Bioprompt program. was not used (the
SD: 39.54 months Patients were encouraged monitor was turned
to increase the output of off and the program
the triceps muscle based was not engaged).
on the visual and auditory
feedback provided by the
computer. The patients
worked to increase the
ROM of elbow extension
TOPICS IN STROKE REHABILITATION/SPRING 2006
within the context of a
task. Patients progressed
in subsequent sessions
to tasks of greater
difficulties based on
ability to complete the
task. Regardless of ability,
all patients advanced to
the task of next highest
difficulty after three
sessions.
APPENDIX 3D. Characteristics of Included Studies for Gait Training

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Ada Total: 27 First stroke between 6 Gr1: 28 months Gr1: 66 Gr1: 30 min of Gr2: Home exercise N/A 3 x/wk for Follow- 2, 1, 1
2003248 Gr1: 13 months and 5 yr previous SD: 17 months SD: 11 treadmill walking and program to carry 4 wk up at 4
Gr2: 14 to inclusion; hemiparesis; overground walking out 3 x/wk for months
between 50 and 80 yr of Gr2: 26 months Gr2: 66 with the proportion 4 wk, exercises
age; walking 10 meters SD: 20 months SD: 11 of treadmill walking to lengthen and
independently with a decreasing by 10% strengthen lower-
speed of less than 1.2 each wk, starting at limb muscles as
m/s. 80% in wk 1 and well as to train
Hemiplegia: decreasing to 50% in balance and
Gr1: 8 R; 5 L wk 4. coordination.
Gr2: 6 R; 8 L

Barbeau Total: 79 Admitted to the Jewish Gr1: 68.1 days Gr1: 66.5 Gr1: Treadmill Gr2: Treadmill Both groups: 4 x/wk for Follow- 2, 0, 1
2003249 Gr1: 43 Rehab Hospital, Laval, SD: 26.5 days SD: 12.8 training while a % training without Maximum of 3 trials, 6 wk up at 3
Gr2: 36 QC, between Oct. ’92 of their body weight BWS, harness was not more than 20 months
and Jan. ’95 for physical Gr2: 78.4 days Gr2: 66.7 was supported by an worn for security. min, walking on
rehab after stroke. SD: 30 days SD: 10.01 overhead harness, the treadmill was
Hemiplegia: up to 40% BWS initiated at 0.0 km/
Gr1: 20 R; 30 L at the beginning hr and increased by
Gr2: 29 R; 21 L and progressively increments of 0.15
decreased. km/hr.
Appendix 3D
177

continues
APPENDIX 3D. Continued
178

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Bogataj Total: 20 Inclusion: Patients who
1995250 Gr1: 10 met the following criteria:
Gr2: 10 had cardiovascular
capacity sufficient to
sustain therapy without
harm; could stand
independently or with
therapist’s assistance
(required substantial
weightbearing support
of one or two therapists);
had sustained perceptual
and intellectual abilities
post CVA; and had
Gr1: 116 days Gr1: 53.4 Gr2: Multichannel Gr1: Physical N/A 5 x/wk for End of 1, 0, 0
SD: 66 SD: 11.5 FES to the peroneal therapy for 1 to 2 3 wk treatment
nerve (dorsiflexors), hr a day involved
Gr2: 104 days Gr2: 59.1 plantar flexors, a passive approach
SD: 62 SD: 9.0 hamstrings, that focused on
quadriceps femoris, preserving the
gluteus maximus, and ROM of the joints
triceps brachii (for and sensory
reciprocal arm swing). input through
Treatments lasted 30 modalities such as
min to 1 hr; pulse icing, heating, and
frequency at 30 Hz brushing; active
lasting 200 µs. methods facilitated
functional
TOPICS IN STROKE REHABILITATION/SPRING 2006

sufficient functional movement and

response to stimulation. proprioceptive
Hemiplegia: neuromuscular
Gr1: 8 R; 2 L facilitation and used
Gr2: 1 R; 9 L BFB exercises.
Area treated: Lower
extremity
Burridge Total: 32 Stroke causing a Gr1: 3.58 yr Gr1: 52.25 Gr1: Experimental. Gr2: Control. Both groups Two End of 2, 0, 1
1997251 Gr1: 16 hemiplegia of at least SD: SD: 14.25 Odstock Dropped Concurrent received ten 1-hr treatments/wk treatment;
Gr2: 16 6 months duration; 7.17 months Foot stimulator (FES) treatment only physiotherapy for 4–5 wk follow-up
single drop foot; Gr2: 61.25 to the tibialis anterior sessions during at 8–9 wk
sufficient dorsiflexion Gr2: 4.92 yr SD: 8.5 muscle and common the first month of
of the ankle; no mental SD: peroneal nerve treatment. Time
impairment; no other 9.83 months proximal to head of was spent on arm
medical conditions that fibula. Duration of posture, trunk
impaired walking; no treatment was not control, balance, and
hypersensitivity to the reported. functional activities.
sensation of stimulation;
and who could stand
unsupported, walk
10 meters (walking
aids permitted but not
another person), stand
from sitting without
help, and walk 50 meters
independently prior to
stroke.
Hemiplegia:
Gr1: 9 R; 7 L
Gr2: 8 R; 8 L
Appendix 3D
179

continues
APPENDIX 3D. Continued
180

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Cozean Total: 36 Inclusion: Medically Number of Gr1: 62 Gr2: BFB. Gr1: PROM and Rehabilitation 3 treatments/ End of 1, 1, 1
198851 Gr1: 8 stable patients referred patients with Gr2: 51 Electromyographic AROM for all major program wk for 6 wk treatment
Gr2: 8 to the center for rehab less than 1 year Gr3: 52 signals were recorded joints and muscle
Gr3: 8 following a CVA, and post-CVA Gr4: 56 with surface groups of both legs
Gr4: 8 who had cognitive electrodes over the in a 30-min session.
ability to cooperate, Gr1: 5 muscle belly. Patient Gr3: Electrical
give informed consent, Gr2: 4 was asked to produce stimulation of the
and physical ability to Gr3: 4 a 30-s maximal tibialis anterior
ambulate a short distance Gr4: 3 contraction followed during swing phase
with the assistance of a by a 1-min rest in a and gastrocnemius
single therapist. 30-min session. during stance phase
Hemiplegia: of gait cycle in a
Gr1: 5 R;3 L 30-min session.
Gr2: 6 R; 2 L Gr4: 15-min
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr3: 4 R; 4 L treatment with

Gr4: 3 R; 5 L BFB and 15-min
treatment with
FES for the first 2
wk. For the next
4 wk, 5 min of
FES and 30 min of
independent use.

da Cunha Total: 12 Patients who had had Time since Gr1: 59.67 Gr2: Body-supported Gr1: Only Concurrent 5 days/wk for None 2, 0, 1
Filho Gr1: 6 a recent stroke (< 6 stroke onset SD: 13.58 treadmill training concurrent treatment (1 hr 2–3 wk
2001252 Gr2: 6 wk ago) secondary (instead of the 20- treatment each):
to CVA based on Gr1: 14.33 Gr2: 57.83 min gait training the PT: Strengthening,
clinical presentation days SD: 5.56 control group received function mobility
(hemiparesis) or MRI; SD: 6.06 during concurrent activities (including
and who displayed a gait treatment). Training gait training).
speed ≤36 m/min and Gr2: 15.67 started with support Kinesiotherapy: To
a score of 0–2 on FAC days of up to 30% of body increase strength and
scale. SD: 7.66 weight, progressively endurance.
Hemiplegia: decreased as patient OT: ADL training.
Gr1: 4 R; 4 L became more capable
Gr2: 2 R; 5 L of more self-support.
da Cunha Total: 13 Recent stroke; significant N/A Gr1: 58.9 Gr1: Regular rehab: Gr2: STAT N/A 5 x/wk, until N/A 2, 0, 1
2002253 Gr1: 7 gait deficit as evidenced SD: 12.9 Daily sessions of (supported discharge
Gr2: 6 by a gait speed of 36 physical therapy, treadmill (average of 3
m/min or less and a score Gr2: 57.8 kinesiotherapy, and ambulation wk)
of 0, 1, or 2 on the FAC SD: 5.5 OT, 3 hr/day, 1 hr/day training) same as
scale; sufficient cognition; Gr1 but the gait
ability to stand with or training was done
without assistance; and on a treadmill
stable medical condition. with an overhead
Hemiplegia: harness. Training
Gr1: 4 R; 3 L started with
Gr2: 5L and 1 brainstem support up to 30%
of body weight
support and was
gradually decreased
aiming for zero
BWS. Subjects were
trained daily for 20
min.

Kosak Total: 55 No prior stroke; Gr1: 39 days Gr1: 74 Gr1: Patient asked Gr2: Hemibar and Concurrent 45 min/day a N/A 1, 0, 1
2000254 Gr1: 21 independent with SD: 14.07 days SD: 9.38 to bear as much KAFO if necessary treatment: Physio plus 45 min of
Gr2: 34 ambulation before of own weight as (least restrictive that was functionally physiotherapy,
current stroke; no active Gr2: 40 days Gr2: 70 possible. Affected brace needed). oriented and 5 days/wk for
angina pectoris or SD: 23.32 days SD: 11.66 ankle wrapped with Therapist prevented incorporated a ~2 wk
orthostatic hypotension; elastic bandage into patient’s pelvis variety of motor
no neurologic or mild dorsiflexion from shifting away facilitation and
orthopedic disorder that position to avoid toe from the bar. Once motor control
might preclude normal drag if needed. Patient patient was able techniques, often
walking; FIM walking walked until fatigued; to walk 3.7 to including bracing
subscore of 3 or less; and speed increased and 7.6 meters at the and walking assistive
hemiparesis of the lower BWS decreased as hemibar, device devices.
extremity and whose tolerated. Therapist was changed
iliopsoas strength is provided assistance for hemiwalker/
graded as 3 or less on the with weight shifting, quadruped cane.
Medical Research Council leg advancement, and
scale (0–5). foot placement.
Hemiplegia:
Gr1: 8 R; 12 L
Bilateral: 2
Gr2: 12 R; 16 L
Bilateral: 6
Appendix 3D
181

continues
APPENDIX 3D. Continued
182

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Kwakkel Total: 89 Inclusion: Patients who Time since Gr1: 69.0 Gr1: Upper extremity Gr3: Gr1&2: 15 min/day 5 x/wk for Follow-up 2, 0, 1
1999164 Gr1: 29 met the following criteria: stroke onset SD: 9.8 training: Functional Immobilization leg rehab; 20 wk at 6 wk
Gr2: 26 had stroke diagnosis Gr1: 7.2 days exercises that of affected upper 15 min/day
Gr3: 34 (WHO definition); had SD: 2.8 days Gr2: 64.5 facilitated forced-arm extremity and upper extremity
primary first-ever stroke SD: 9.7 and hand activity lower extremity rehabilitation;
in the territory of the Gr2: 7.0 days such as leaning, by inflatable splint 1.5 hr/wk ADL
middle cerebral artery as SD: 2.5 days Gr3: 64.1 punching a ball, and applied with the training
revealed by CT scan or SD: 15.0 grasping and moving patient supine for
MRI; are 30–80 yr old; Gr3: 7.5 days objects. 30 min.
have impaired motor SD: 2.9 days Gr2: Lower
function of the arm and extremity training:
leg; have no complicating sitting, standing,
medical history; have weightbearing
no severe deficits in exercises while
TOPICS IN STROKE REHABILITATION/SPRING 2006

communication, memory, standing and walking,

or understanding; gave with emphasis on
written/spoken informed achieving stability
consent; are motivated to and improving gait
participate in the research velocity. Treadmill
project. training where
Hemiplegia: equipment available.
Gr1: 16 R; 17 L Gr1 and 2 training
Gr2: 13 R; 18 L was applied by OT/PT
Gr3: 13 R; 24 L for 30 min.
Kwakkel Total: 101 Inclusion: Primary, Time since Gr1: 69.0 Gr1 Upper Gr3 (Control): Gr1, Gr2, & Gr3: 15 30 min/day, Follow-up 1, 0, 1
2002165 Gr1: 33 first-ever stroke in the onset (days) SD: 9.8 extremity: Focused Immobilization of min lower extremity 5 days/wk for at 3 and 6
Gr1: 31 territory of the middle Gr1: 7.5 on improvement in the affected upper rehabilitation; 15 20 wk months
Gr3: 37 cerebral artery as shown SD: 2.9 days Gr2: 64.5 disabilities involving extremity and lower min upper extremity
by CT or MRI; between SD: 9.7 the affected upper extremity by means rehabilitation; and
30–80 yr of age; had an Gr2: 7.2 extremity (grasping, of an inflatable 1.5 hr/wk ADL
impaired motor function SD: 2.8 days Gr3: 64.1 reaching, leaning, or pressure splint. training.
of upper as well as lower SD: 15.0 dressing).
limbs; unable to walk at Gr3: 7.0 Gr2 Lower extremity:
first assessment; had no SD: 2.5 days Focused on functional
complicating medical recovery of balance
history such as cardiac, (sitting or standing
pulmonary, or other balance), transfers
neurological disorders; (such as turning
had no severe deficits in over), and gait
communication, memory, (performance &
or understanding. climbing stairs).
Hemiplegia:
Gr1: 16 R; 17 L
Gr2: 13 R; 18 L
Gr3: 13 R; 24 L
Appendix 3D
183

continues
APPENDIX 3D. Continued
184

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Laufer Total: 25 Inclusion: Patients Gr1: 35.8 days Gr1: 69.3 Gr2: Ambulating on a Gr1: Ambulation Concurrent Training 5 N/A 0, 0, 1
2001255 Gr1: 12 who met the following SD: 17.3 days SD: 8.1 treadmill adjusted to on floor surface at treatment: PT once x/wk for 3 wk.
Gr2: 13 conditions: had first Gr2: 32.6 days patient’s comfortable comfortable speed a day, 5 x/wk based Total walking
supratentorial strokes SD: 21.2 days Gr2: 66.6 walking speed. using walking aids; on Bobath approach; time: 4 min/
in ant brain circulation SD: 7.2 Training 5 x/wk. assistance and OT and speech day for first
as evidenced by CT; Same walking time as resting periods as therapy if needed. wk, 6 min/day
had no additional control group. needed. for 2nd wk,
neurological and/or 8 min/day for
orthopedic deficiencies 3rd wk.
impairing ambulation;
no cardiac/respiratory/
medical condition that
could interfere with
protocol; and no severe
TOPICS IN STROKE REHABILITATION/SPRING 2006

cognitive/communication
impairment that could
hamper understanding
simple instructions; had
onset of stroke no more
than 90 days prior to
beginning of study; and
could walk on treadmill
at speed of at least 0.2
km/hr with minimal to
moderate assistance for 2
min without rest.
Hemiplegic:
Gr1: 5 R; 7 L
Gr2: 5 R; 8 L
Macdonell Total: 38 Patients likely to remain
1994256 Gr1: 20 as inpatients for the
Gr2: 18 duration of the study (4
wk).
Location of hemiplegia
was NR.
Treated area: Lower
affected extremity
Montoya Total: 14 Inclusion: Patients
1994206 Gr1: 9 who met the following
Gr2: 5 criteria: had first stroke;
whose stroke occurred
less than 6 months
before the beginning
of the experimental
protocol; had motor
ability allowing 50-m
walking without any
help; and had absence
of understanding
problems, biomechanical
malfunction of the
legs, hemianopsia, or
unilateral neglect.
Gr1: 25 days Gr1: 65 Gr1: FES and Gr2: Control: No Both groups received CES: 5 days/ 4 wk post 1, 0, 0
SD: 6.5 days SD: 9 Cyclical Electrical FES-CES a self-exercise wk for 4 wk; treatment,
Stimulation (CES): program 5 days/wk, FES: 3 days/ 8 wk from
Gr2: 26 days Gr2: 68 Patients received CES and exercise or wk for 4 wk; baseline
SD: 6.5 days SD: 9 5 days/wk for 20-min functional activities Self-exercise
sessions and FES 3 3 days/wk for program: 5
x/wk for 20 min to 20-min sessions. days/wk for
the peroneal nerve at Passive, active, 4 wk;
the knee to produce and active-assisted Functional
dorsiflexion of the exercises and activities: 3
ankle in the affected functional activities days/wk for
leg. FES was given were performed. 4 wk
at a frequency of
30–50 Hz and pulse
duration of 0.3 ms,
whereby the intensity
was adjusted to the
patient’s tolerance.
Gr1: Less than Gr1: 64.0 Gr1: BFB. Initial 6-m Gr2: Control. Standard rehab 2 sessions/wk End of 1, 0, 0
6 months SD: 2.75 walk with locomotor. Patients followed program and walk for 4 wk treatment
Gr2: Less than Patients were told the same standard sessions
6 months Gr2: 60.0 to secure the best rehab program
SD: 2.25 walking performance. and walk sessions
Five 6-m walks with but without BFB.
BFB. The imposed Session lasted 45
right and left step min.
lengths were the
initial step-length
measures in the initial
6-m walk. To correct
asymmetry, 10 walks
with the BFB where
the imposed right and
left step lengths were
identical. A last 6-m
walk with locomotor
analysis to evaluate
short-term gain of
step-cycle parameters.
Session lasted 45 min.
Appendix 3D
185
continues
APPENDIX 3D. Continued
186

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Morris Total: 26 Inclusion: Patients who Gr1: 79.0 days Gr1: 64.2 Gr2: Phase 1: Gr1: Control. Physiotherapy 5 sessions/wk Follow-up 1, 0, 1
199216 Gr1: 13 had suffered a CVA up SD: 41 days SD: 10.3 Electrogoniometric Standard PT based for 4 wk at end of
Gr2: 13 to 4 months previously; feedback during 4 on the Motor treatment
were medically stable; Gr2: 45.0 days Gr2: 64.4 wk. EMG-BFB during Relearning Program; (4 wk)
could provide informed SD: 23 days SD: 11.0 standing and gait approximately 30-
consent and demonstrate training. When the min sessions.
ability to safely walk 10 patient hyperextended
m four times without aids the knee, the
or orthoses. auditory signal was
Hemiplegia: activated. A fixed high
Gr1: 8 R; 6 L frequency warning
Gr2: 5 R; 8 L signal sounded when
knee flexion exceeded
75° to discourage
TOPICS IN STROKE REHABILITATION/SPRING 2006

over compensation
with excessive
knee flexion.
Approximately 30-
min sessions.

Nilsson Total: 60 Inclusion: Patients who Gr1: 22 days Gr1: 54 Gr1: Walking training Gr2: Standard gait Concurrent PT: 30 min/day of Follow- 2, 0, 1
2001257 Gr1: 28 met the following criteria: Gr2: 17 days Gr2: 56 on treadmill with training. Approach Transfers, ROM treatment plus up at 10
Gr2: 32 were <70 yr, had first BWS. 2 therapists according to a exercises, techniques 30 min/day of months
stroke with residual assisted patient’s leg motor relearning to improve motor PT, 5 x/wk for
hemiparesis, and were and trunk movement. program for function in paretic approximately
within 8 wk of stroke BWS gradually stroke by Carr and side. 10 wk
onset. decreased and speed Shepherd.
Hemiplegia: increased as fast as
Gr1: 21 R; 11 L possible (adjusted to
Brainstem: 4 the improvement).
Gr2: 18 R; 14 L
Brainstem: 5
Pohl Total: 60 Inclusion: Patients who
2002258 Gr1: 20 met the following criteria:
Gr2: 20 had hemiparesis caused
Gr3: 20 by R or L supratentorial
ischemic stroke or
intracerebral hemorrhage;
had impaired gait; had
hemiparesis lasting for
>4 wk; had no or slight
spasticity (Ashworth
score 0 or 1); could
walk without personal
assistance; and whose
time required to walk 10
m >5 and <60 s.
Hemiplegia:
Gr1: 4 R; 16 L
Gr2: 4 R; 16 L
Gr3: 5 R; 15 L
Gr1: 16.10 wk Gr1: 61.6 Gr2: 30 min of Gr1: Had 45 min of Standard customary 3 x/wk for N/A 2, 0, 1
SD: 18.5 wk SD: 10.6 limited progressive control gait training care in PT (not 4 wk
treadmill training; using PNF and described)
Gr2: 16.8 wk Gr2: 57.1 speed increased by Bobath concepts. 8 times for 45 min
SD: 20.5 wk SD: 13.9 no more than 5%
of maximum initial
Gr3: 16.2 wk Gr3: 58.2 walking speed each
SD: 16.4 wk SD: 10.5 wk.
Gr3: 30 min of
speed-dependent
treadmill training;
maximum overground
walking speed
was determined,
then halved for
5-min warm-up
on treadmill; over
1–2 min, belt speed
increased to highest
speed at which patient
could walk safely
(Vt1) and was held
for 10 s. Recovery
period —during
which patient’s pulse
returned to resting
level— occurred
before Vt2 phase,
which saw an increase
of 10% over Vt1.
Patient tried to hold
speed for 10 s then
had recovery time,
and so on until Vt5
(during each phase
speed increased by
10%). If speed was
not kept up for 10 s,
speed decreased by
10% in next phase.
Appendix 3D
187
continues
APPENDIX 3D. Continued
188

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Schauer Total: 23 Inclusion: Patients who Gr1: 53 days Gr1: 59 Gr1: Practiced Gr2: Control: 20- Neurodevelopmental 5 days/wk, 20 N/A 2, 0, 1
2003259 Gr1: 11 suffered from stroke in Gr2: 67 days SD: 12 walking with music. min training session therapy (45 min/ min each day
Gr2: 12 the middle cerebral artery Music motor feedback each day: warming day)
distribution and could Gr2: 61 device: sensor insoles up and common Total: 15
execute at least task 7 SD: 12 that detect the exercises led by a sessions
but failed at task 11 or ground contact of the therapist (such as
higher of the Rivermead heels and portable slow walking with
Motor Assessment music player. Music support of parallel
(RMA), able to walk 20 was played at an bars and hand rails,
min without any assistive adjustable speed that stepping sideways
device, did not exhibit was estimated from and backwards,
any severe cognitive the time interval etc.).
or communication between 2 consecutive
impairment, no signs of heel strikes.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Parkinson’s disease or
Huntington’s disease,
and willing to walk with
music.
Hemiplegia:
Gr1: 4 R; 7 L
Gr2: 7 R; 5 L
Sullivan Total: 24 Inclusion: Patients who Gr1: Gr1: 70.9 Up to 40% body- Gr1: 0.5 mph N/A 4 x 5 min N/A 1, 0, 1
2002260 Gr1: 8 have unilateral stroke 22.51 months SD: 9.8 weight-support was walking/
Gr2: 8 within the middle SD: provided initially session; 12
Gr3: 8 cerebral artery or basilar 3.6 months Gr2: 66.5 & progressively sessions over
artery distribution SD: 13.9 decreased as the 4–5 wk
resulting in unilateral Gr2: 27.6 subject increased
hemiparesis, time months Gr3: 64.4 activity tolerance and
since stroke onset > 6 SD: 21.5 SD: 13.4 could maintain proper
months, living within months limb kinematics
the community, able throughout stance
to ambulate 10 m with Gr3: 27.2 & swing. Assistance
or without an assistive months of 1 therapist and
device and no more SD: 13.7 1 aide to provide
than standby physical months stability, assistance,
assistance, and walking and monitor posture/
speed reported to be weight shift/limb
slower than before stroke. position. Rest periods
Exclusion: Subjects allowed at any time;
were excluded if no orthosis.
they were receiving
any physical therapy, Gr2: 0.5, 1.0, 1.5, 2.0
had musculoskeletal mph
impairments that limited
full knee extension or Gr3: 2.0 mph
ankle plantarflexion to
neutral, and any other
neurologic condition
other than unilateral
stroke.
Hemiplegia:
Gr1: 3 R; 5 L
Gr2: 3 R; 5 L
Gr3: 2 R; 6 L
Appendix 3D
189
APPENDIX 3D. Continued
190

Age, yr Session
Author Sample Symptom (mean, SD frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group Concurrent therapy and duration duration R, B, W

Thaut Total: 20 Inclusion: Patients Gr1: 15.7 days Gr1: 72 Gr2: Rhythmic Gr1: Gait training Concurrent 2 sessions a N/A 1, 0, 1
1997261 Gr1: 10 within 3 wk post CVA SD: 4.0 days SD: 8.0 auditory stimulation for same amount treatment: day, 5 x/wk
Gr2: 10 who could complete 5 (RAS) using of time as PT based on for 6 wk
strides with hand-held Gr2: 16.1 days Gr2: 73 metronome or experimental group neurodevelopmental
assistance. SD: 4.0 days SD: 7.0 specially prepared and with equivalent treatment approach.
Hemiplegia: music tapes; 1–2 instructions but
Gr1: 5 R; 5 L min warm-up walk, without RAS.
Gr2: 5 R; 5 L matched rhythm
frequency for first
quarter of the session,
5%–10% rhythm
frequency increase
during second
and third quarter,
TOPICS IN STROKE REHABILITATION/SPRING 2006

last quarter RAS

intermittently faded.

Visintin Total: 100 Inclusion criteria not Gr1: 68.1 days Gr1: 66.5 Gr1: Gait training Gr2: As Concurrent 4 x/wk for Follow- 2, 0, 1
1998262 Gr1: 50 mentioned. SD: 26.5 days yr on treadmill with experimental group treatment: PT to 6 wk, with up at 3
Gr2: 50 Exclusion: Patients who SD: 12.8 yr BWS up to 40% but with no BWS. maximize function. a maximum months
met the following criteria: Gr2: 78.4 days (progressively duration of
who had normal gait SD: 30.0 days Gr2: 66.7 decreased as walking 20 min per
pattern, severe cardiac yr ability improved). session
problem, or cerebellar/ SD: 10.1 yr Assistance for proper
bilateral/brainstem CVA; trunk alignment,
whose treadmill training weight shifting,
is contraindicated; who stepping, and limb
are unable to understand control. Speed
simple commands; who increased as walking
were readmitted during ability improved.
study period and not
ambulating before stroke;
whose anticipated length
of stay < 4 wk; and
who had onset CVA > 6
months.
Stroke location:
Gr1: 20 R; 30 L
Gr2: 29 R; 21 L
Werner Total: 28 Inclusion: Subjects had Gr1: 4.2 54.7 yr Gr1: 30 treadmill Gr2: Received 15 Both groups Gr1: 30 Follow- 2, 0, 1
2002263 Gr1: 14 to be within 8 wk post months training sessions with treadmill training participated in treadmill up at 4
Gr2: 14 stroke and no more Range, 2.5–7.5 BWS (progressively sessions with BWS comprehensive training months
than 9 months after decreased as soon as with no additional rehab program session: 30
a first supratentorial Gr2: 5.1 possible to enable PT. for another 3 wk. min/session,
stroke; nonambulatory months full load of the lower After return to their 5 x/wk for a
(required continuous or Range, 2.5–8.2 limbs. 10 subjects homes they received total of 3 wk,
intermittent support); needed 5%–15% BWS outpatient PT and PT for 40
participation in a until the end). 2–3 x/wk. Patients min/session,
comprehensive 9 wk Added to single and family were 5 x/wk for
inpatient rehab program; session of instructed to practice a total of 30
able to understand physiotherapy: gait as much as sessions
simple instructions and gait preparatory possible. Gr2: 15
the meaning of the study; maneuvers in treadmill
no other orthopedic or sitting and standing sessions, each
neurological diseases following Bobath 30 min long
impairing mobility. principles and the for 3 wk
No exclusion criteria practice of gait on
were mentioned floor or on the stairs.
Hemiplegia:
Gr1: 5 R; 9 L
Gr2: 7 R; 7 L

Yan Total: 41 Inclusion: First acute Gr1: 8.7 days Gr1: 68.2 Gr1: 2 dual-channel Gr2: Stimulation Standard rehab 60 Gr1: 30 min/ 5 wk 2, 0, 1
2005264 Gr1: 13 stroke, unilateral stroke SD: 5.8 days SD: 7.7 stimulation connected from electrical min day, 5 days/wk
Gr2: 15 within the corotid artery with a timer to form stimulation device PT based on NDT x 3 wk
Gr3: 13 system according to CT, Gr2: 10.1 days Gr2: 73.3 one stimulating with disconnected and OT focused on
aged 45 to 85 yr old, SD: 2.8 days SD: 8.1 unit for FES. circuit. ADL x 5 wk. Gr2: 60 min/
independent in daily Surface electrodes Gr3: Standard day, 5 days/wk
activities before stroke. Gr3: 9.1 days Gr3: were applied rehab, 60 min x 3 wk
SD: 3.5 days 70.4 yr on quadriceps, physiotherapy
SD: 7.6 yr hamstrings, tibialis based on NDT and
anterior and medial OT focused on ADL
gastrocnemius; x 5 wk.
0.3-ms pulses at 30
HZ, max tolerance
intensity using an
activation sequence
that mimicked normal
gait; treated 3 days
after being transferred
from acute hospital.
Appendix 3D

Note: ADL = activity of daily living; AROM = active range of motion; BFB = biofeedback; BWS = body weight support; CVA = cerebrovascular accident; EMG = electromyographic; FAC = functional ambulatory category; FES
= functional electrical stimulation, Gr = group; KAFO = knee-ankle-foot orthosis; N/A = not available; NDT = neurodevelopmental therapy; NR = not reported; OT = occupational therapy; PNF = proprioceptive neuromuscular
facilitation; PROM = passive range of motion; PT = physiotherapy; Quality R, B, W = randomization, blinding, withdrawals; x/wk = times per week.
191
APPENDIX 3E. Characteristics of Included Studies for Balance Training
192

Age, yr Session
Author Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)

Aruin Total: 16 Ability to stand and 18.0 days 65.34 Gr1: Feedback information on Gr2: Standard Gr1&2: 10 days, 2 N/A 1, 0, 0
2003284 Gr1: 8 walk up to 4.5–6 m SD: 1.8 days SD: 3.4 base of support as an adjunct customary gait Standard x/day for 25
Gr2: 8 with assistance of to conventional gait therapy. training (pregait customary gait min each
physical therapist Device: 2 sensors that were training focused training time
and ability to strapped to the subject’s lower on weight shifting,
understand and extremities below the knees stepping, trunk
follow verbal and next to tibial tuberosity stabilization, and
instructions. with Velcro tape and a main facilitation of
unit that provides distinct muscle in the lower
signals (tone of 500 Hz) every extremity), followed
time the distance between the 2 by 3–4 ambulations
sensors is less than previously that did not exceed
established threshold. 15–18 m with rest
periods of 2–3 min in
TOPICS IN STROKE REHABILITATION/SPRING 2006

between.

Chen Total: 41 Ambulatory stroke Gr1: Gr1: 58.7 Gr1: Visual feedback balance Gr2: Control group: Gr1&2: 20 min/day, 6 months 1, 0, 0
2002285 Gr1: 23 patients admitted 3.00 months SD: 0.19 training with the Smart Idem to concurrent Standard 5 days/wk for
Gr2: 18 to the rehab ward. Gr2: Balance Master. Encouraged to treatment. customary care: 2 wk
Hemiplegia: 3.78 months Gr2: 5.33 maintain their posture steadily PT and OT
Gr1: 13 R; 10 L SD: 1.78 and to maintain symmetric programs.
Gr2: 7 R; 11 L weight bearing while adapting
to different static sensory
conditions through verbal or
tactile cues. Dynamic function
training: Practice controlling
their weight shifts by tracing
the moving targets on the
screen in every main direction
while the limits of stability
condition was set at 50%.
Cheng Total: 54 Medically stable; Gr1: Gr1: 63.1 Gr2: Training group: Standing Gr1: Control group: Gr1 & Gr2: 3-wk Follow- 1, 0, 1
2001286 Gr1: 24 could adequately 2.9 months SD: 7.8 postural symmetry and Standard stroke Standard program up at 6
Gr2: 30 comprehend SD: repetitive sit-to-stand training rehab program customary care months
our instructions; 1.2 months Gr2: 62.3 through use of a standing (neuromuscular
could stand up SD: 8.0 BFB trainer. Patients stood in facilitation
independently Gr2: front of the table with their techniques, functional
and walk with or 2.8 months feet on the force platform and electric stimulation,
without a cane. SD: their paretic hand fixed by an mat exercises, and
Hemiplegia: 1.4 months elastic bandage to the handle other therapeutic
Gr1: 13 R; 11 L of a weight loaded box and exercises).
Gr2: 13 R; 17 L they pushed and pulled the
box. They were instructed to
maintain a symmetric upright
posture according to the
trainer’s visual and auditory
signals. After a 15-min rest,
patients performed a sit-to-
stand movement.

Engardt Total: 30 Inclusion: Able to 1 wk to 3 Gr1: 67 Gr1: BFB group: The patient Gr2: Control group: Gr1&2: 15 min/ 33 months 1, 0, 1
1994287 Gr1: 16 understand and months SD: 6.05 sat in a standardized position Training program Individual session
Gr2: 14 follow instructions, on an adjustable armless of rising and sitting standard 3 sessions/
have adequate Gr2: 65 chair with a back support. down with vertical customary care day
hearing, and SD: 8.46 The patient placed one foot ground reaction force 5 days/wk
able to stand up firmly on each platform right without BFB. for 6 wk
independently. above the strain gauge force treatment
Exclusion: Normal transducers. The patient was duration
motor function instructed to stand-up, to
in the lower stand for 1 min and to sit
extremities; stood down again. The patients
up with ataxia or supported by the auditory BFB
who had severe signal received knowledge
cognitive deficits of performance immediately
Hemiplegia: when the load of the paretic leg
Gr1: 6 R; 10 L reached the preset body-weight
Gr2: 8 R; 6 L distribution.
Appendix 3E
193

continues
APPENDIX 3E. Continued
194

Age, yr Session
Author Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)

Geiger Total: 13 Hemiplegia as a Gr1: Gr1: 61.8 Gr1: EMG-BFB group: The Gr2: PT interventions Gr1&2: 4-wk N/A 2, 0, 1
2001288 Gr1: 7 result of stroke; 99.9 days SD: 16.9 training session begins with a to improve muscle Received the treatment
Gr2: 6 able to maintain a SD: warm-up period, a period of force, ROM, balance, same physical
stationary standing 96.0 days Gr2: 58.7 stationary standing with eyes and mobility. These therapy
position with SD: 14.8 open and eyes closed. The interventions include intervention
or without an Gr2: subjects were instructed to mat activities,
assistive device for 133.8 days maintain or shift their weight, stretching and
a minimum of 2 SD: as appropriate, to make the strengthening,
consecutive min 203.4 days representation of their center weight-bearing
without manual of gravity reach the target(s) or shifting and
assistance; and presented visually. Emphasis standing lower-
able to follow was placed on anterior, extremity exercise
instructions. posterior, lateral, and diagonal in parallel bars, and
Hemiplegia: weight shifts to the subjects’ balance activities.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr1: 2 R; 5 L affected side as well as weight Interventions were

Gr2: 3 R; 3 L shifting sequentially to the individualized
targets. for each subjects
based on his-her
impairments. 50-min
session
Grant Total: 16 Inclusion: Patients 33 days 65 Gr1: Visual feedback balance Gr2: Standard Gr1&2: Gr1 & Gr2: Follow-up 2, 0, 1
1997289 Gr1: 8 who can follow SD: 5 days SD: 3 training on the Balance customary balance Standard 30-min at 4 wk
Gr2: 8 directions (scores post-stroke Master. Standing with each training. Based on customary balance
>21 points on the foot on a force platform activity. Progression: physiotherapy. training
Modified Folstein (23 x 46 cm) and shifting Symmetrical weight Details: N/A (either
Mini-Mental State) a cursor representing the distribution, weight standard or
and who have center of gravity (CoG) shifting, reaching, experimental)
no neurological, within the theoretical limit and performing per day for
vestibular, or of stability (LoS). Goal: To functional activities 19 sessions;
orthopedic problem attain symmetrical weight that challenged 5 days/wk
that could affect distribution, shifting the CoG balance. Tasks as inpatient
balance. toward the perimeter of the became more (3 wk
Hemiplegia: NR LoS, and stabilizing the CoG difficult: changing minimum)
position within predefined base of support (BoS), and 2
targets spatially oriented within increasing the speed days/wk as
the LoS. Progression: Reaching, of performance, outpatient
stepping in place, having or altering visual (maximum 8
targets placed close to the LoS, input. Functional wk)
and increasing the required balance activities
speed of weight shifts. included locomotion
with an altered BoS
and stepping over
obstacles.

Hocherman Total: 24 Inclusion: Patients N/A N/A Gr1: Platform training. Each Gr2: Control. Gr1&2: 15 sessions None 1, 0, 1
1984290 M/F who suffered a session comprised two 5-min No training on Standard over 3 wk
Gr1: 7/6 CVA in the anterior sessions. In the first 5 min, the platform and customary
Gr2: 6/5 cerebral circulation patients stood parallel to the concurrent treatment physical
10–21 days before movement axis; in the second, only. therapy
entry to the study, they stood perpendicular to the Details: N/R
can stand on the movement axis.
platform for a 2- Apparatus was a 60 x 60 cm
min period with the platform that could move back
platform moving and forth along a horizontal
back and forth, and line. Movement velocity
have appropriate profile was sinusoidal, and the
judgment. frequency was 0.5 Hz. Peak-to-
Exclusion: peak amplitude was adjustable
Completely flaccid within a range of 1–16 cm.
or spastic patients.
Hemiplegia:
Gr1: 5 R; 8 L
Appendix 3E

Gr2: 5 R; 6 L
195

continues
APPENDIX 3E. Continued
196

Age, yr Session
Author Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)

Mudie Total: 40 Subjects had Within 2–6 72.4 Gr1: Portable computer-like Gr3: Bobath-trained N/A All groups: N/A 2, 2, 1
2002119 Gr1: 10 suffered a recent wk SD: 9.01 BPM feedback console used to staff physiotherapists 2 wk;
Gr2: 10 stroke and provide awareness of weight focused on increasing one 30-min
Gr3: 10 consistently bore distribution during training trunk & pelvic session/day
Gr4: 10 majority of weight in sitting; subject required to ROM, normalizing
on one side. touch target with non paretic trunk muscle
hand at various heights and tone, maintaining
distances; subjects attempted to appropriate balance
return to symmetrical position responses during
after reaching. reaching, and
Gr2: Task-related reach improving the
training: tests extremes of subject’s ability
seated base of support. Subject to move in and
seated on plinth; grocery out of asymmetric
TOPICS IN STROKE REHABILITATION/SPRING 2006

items of approximately 140% posture; verbal &

of upper extremity’s length
were retrieved with unaffected
extremity and placed on
shelves at various heights &
distances to allow reaching.
manual facilitation
by therapist during
seated reaching or
lying.
Gr4: Control:
Standard customary
PT and OT.
Sackley Total: 26 Able to stand for 1 Gr1: 20.1 wk Gr1: 60.8 Gr1: Experimental training Gr2: The patients N/A 4-wk N/A 2, 0, 1
1997291 Gr1: 13 min and displayed SD: 15.8 wk SD: 12.3 from a therapist using practiced similar treatment
Gr2: 13 an abnormal stance EMG-BFB program of the activities to the
symmetry (outside Gr2: 18.8 wk Gr2: 67.9 Nottingham Balance Platform treatment group but
2 SD, allowing for SD: 19.3 wk SD: 9.2 (NBP). BFB signals displaying without the visual
age and gender). weight distribution and weight BFB of at least one
Hemiplegia: shift activity were continuously therapist. Therefore,
Gr1: 4 R; 9 L presented to the patient in the the targeting and
Gr2: 8 R; 5 L form of 2 vertical red columns. balancing column
Each column moved upwards exercise were
with an increase in weight on performed using
the corresponding foot. When the subjective
the columns were within 5% impressions of the
of each other, a red triangle patient and therapist.
appeared confirming that
stance symmetry had been
achieved.

Walker Total: 46 Inpatient with Duration of Gr1: 65.4 Gr1: Visual EMG-BFB training: Gr2: Standard N/A 2 hr/day, 5 Follow- 2, 0, 1
2000292 Gr1: 16 hemiparesis disease: SD: 13.8 Use of the Balance Master customary care: days/wk until up at 1
Gr2: 16 secondary to a first Gr1: consisting of 2 force plates Extension of the discharge month
Gr3: 14 stroke, within 4 65.4 days Gr2: 62.4 positioned side by side. Tactile regular rehab
months of their SD: SD: 13.3 and verbal cues were provided program. Symmetrical
stroke; could stand 19.5 days as necessary to ensure proper weight distribution
unassisted 1 min; Gr3: 65.8 alignment and stability of the was encouraged
in need of balance Gr2: SD: 9.9 hips, knees, and trunk (erect through verbal and
training according 35.1 days posture with no observable tactile cues and was
to the judgment SD: leaning to one side). The task made more difficult
of their senior 22.3 days progressed through to the by the addition of
physical therapist. addition of an upper extremity arm activities or
Gr3: activity or introduction of actions requiring
Hemiplegia: 42.0 days trunk rotation. To increase trunk rotation. Use
Gr1: 7 R; 9 L SD: weight bearing on the affected of stool to increase
Gr2: 7 R; 9 L 13.4 days extremity, subjects were weight bearing on the
Gr3: 5 R; 9 L instructed to shift their weight affected side. Weight
until the bars on the computer shifting.
corresponded to a preset target. Gr3: Control:
Standard customary
PT program and no
additional balance
training.
Appendix 3E
197

continues
APPENDIX 3E. Continued
198

Age, yr Session
Author Sample Symptom (mean, SD Concurrent frequency Follow-up Quality
and year size Population details duration for control) Treatment Comparison group therapy and duration duration (R, B, W)

Winstein Total: 42 Inclusion: Time since Gr1: 53.8 Gr2: Standing feedback Gr1: Standard Concurrent Training: None 0, 0, 1
1989293 Gr1: 21 Patients who have stroke onset: SD: 3.2 training. Static standing customary standing treatment: 30–45 min/
Gr2: 21 been diagnosed Gr1: 6.3 wk progressed to stepping: 1) balance and weight- Physical day, 5 x/wk
with unilateral SD: 0.7 wk Gr2: 51.9 equal weight distribution, shifting training. therapy for 3–4 wk
hemiparesis; are SD: 3.4 2) sitting to standing with exercises
medically stable; Gr2: 7.7 wk equal weight distribution, (sitting balance,
have the potential SD: 1.0 wk 3) lateral weight shifting, coordination,
for functional 4) anteroposterior weight motor control,
locomotion with shifting with affected extremity strengthening);
rehab; show ability forward, and 5) stepping in functional
to stand without place. training
external support (transfers,
for at least 30 s; walking); and
and were to have at functional
TOPICS IN STROKE REHABILITATION/SPRING 2006

least 3 wk inpatient electric muscle

stay. stimulation
Hemiplegia: when indicated.
Gr1: 9 R; 12 L 90 min/day, 5
Gr2: 6 R; 15 L x wk.

Note: ADL = activity of daily living; BFB = biofeedback; BPM = balance performance monitor; CVA = cerebral vascular accident; EMG-BFB = electromyography-biofeedback; F = female; Gr = group; L = left; M = male; N/A =
not available; N/R = not reported; OT = occupational therapy; PT = physical therapy; Quality R, B, W = randomization, blinding, withdrawals; R = right; ROM = range of motion; x/wk = times per week.
APPENDIX 3F. Characteristics of Included Studies for Sensory Interventions

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and Year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Fanthome Total:18 Subjects were not Gr1: 1.0 months Gr1: 66.3 Gr1: Feedback group: The Gr2: Control: Did N/A 2 hr and 40 N/A 1, 0, 1
1995297 Gr1: 9 blind; under 80 SD: 0.7 months SD: 10.7 eye movement is detected not receive any min/wk for
Gr2: 9 yr old; had no with glasses connected treatment for their 4 wk at 40
history of dementia Gr2: 0.6 month Gr2: 71.1 to a feedback device that visual inattention min/day; 20
or psychiatric SD: 1.0 months SD: 7.6 delivered auditory signals or other perceptual min in the
problems; were in the form of a continual deficits for 4 wk. morning and
not too ill to be beep (if the subject did 20 min in
assessed; were not look left within a afternoon
right-handed; had a fixed interval [15 s]). Eye
score of more than movements were recorded
6 on Abbreviated while looking at slides: 1)
Mental Test; right Practice slide with 3 letter
hemisphere stroke; As; 2–3) With 6 letter
and scored <130 As; 4) Pictures from the
on Behavioral Western Aphasia Battery;
Inattention 5–6) Two short passages.
Test (BIT). Left Subject was instructed to
hemiplegic. count the number of As
Treated area: Eyes, and was asked questions
visual neglect about these slides.
Appendix 3F
199

continues
APPENDIX 3F. Continued
200

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and Year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Feys Total: Inclusion: Diagnosis Gr1: 21.40 days Gr1: 65.62 Therapeutic intervention: Control group: Standard customary 30 min a Follow- 2, 0, 1
1998299 100 of ischemic SD: 5.94 days SD: 11.81 Patient was in a rocking Patient was in a care in rehab. session, 5 ups at 17
Gr1: 50 brain damage chair with the affected arm rocking chair with days/wk for wk and 43
Gr2: 50 or intracerebral Gr2: 24.02 days Gr2: 62.78 in inflatable splint. Rocking the arm rested 6 wk wk
hemorrhage; SD: 6.98 days SD: 12.03 movement performed. on a cushion
obvious motor on patient’s lap,
deficit of the no additional
upper extremity stimulation.
(Brunnstrom-Fugl- Placebo: Shortwave
Meyer score on diathermy.
the subscale of the
upper extremity
lower than 46);
ability to sit
TOPICS IN STROKE REHABILITATION/SPRING 2006

independently or
with a minimum
of support; and
ability to perform
the experimental
treatment
independently.
Hemiplegia:
Gr1: 22 R; 28 L
Gr2: 20 R; 30 L
Treated area: Upper
extremity

Feys Total: 62 A diagnosis of N/A Gr1: 62.34 Gr1: Patients were Gr2: Patients also Patients also 30 min/ Follow-up 1, 0, 1
2004298 Gr1: 33 ischemic brain SD: 11.33 positioned in a rocking positioned in a received standard treatment at 5 yr
Gr2: 29 damage or chair with the arm rocking chair; customary care once/day
intracerebral Gr2: 58.14 supported by an inflatable they received fake performed by same for 6 wk,
hemorrhage, an SD: 10.86 long-arm splint in a short-wave therapy therapists in both totaling 30
obvious motor position contrary to of the shoulder groups. sessions
deficit of the upper the typical pattern of during rocking.
extremity, and the spasticity. Patients were
ability to perform asked to perform rocking
the experimental movements, pushing with
treatment the affected arm.
independently.
Fiebert Total: 20 Had had a CVA 3 Between 3 wk Gr1: 71.9 Gr1: Passive vestibular Gr2: Control: No Standard customary Once a day N/A 0, 0, 1
1979300 Gr1: 10 wk to 6 months and 6 months SD: 7 training: Patients were vestibular training. care in rehab, for 2 wk
Gr2: 10 before study. strapped into a rotary chair, 1x/day for 2 wk:
Hemiparesis: Gr2: 71.2 but their heads were not ROM exercises, eye
Gr1: 5 R; 4 L SD: 6.5 secured. The chair was tracking, brushing
Bilateral: 1 rotated to the right, at a of the involved
Gr2: 5 R; 4 L velocity of 8 rotations in 20 extremities, and
Bilateral: 1 s. After 20 s of continued balance activities.
Treated area: rotation, the chair was
Vestibular system stopped abruptly. A rest
period of at least 20 s
followed. The chair was
than rotated to the left for 8
rotations in 20 s and once
again stopped abruptly.
Another rest period of 20 s
followed.

Jongbloed Total: 90 Admitted to the 40 days 71.32 Gr1: Sensory integrative Gr2: Functional Standard customary 40 min/day, NA 1, 2, 1
1989301 Gr1: 43 hospital within SD: 42 days SD: 9.07 approach approach: care (nursing, 5 days/wk for
Gr2: 47 12 wk after a first 1. Patient preparation: A) Emphasized medical, and PT) 8 wk
CVA; a weakness normalized tone (brushing, the practice of
in the UE + LE joint compression, neutral particular tasks
on one side of the warmth, vibration). B) (ADL: dressing,
body on admission; approximation to improve grooming, bathing,
were not admitted equal weight bearing and toileting, mobility,
to an extended trunk rotation activity. and homemaking).
care setting before 2. Treatment: 1) compensation,
the stroke; and A) provided sensory ex: taught one-
were not severely stimulation: patient had handed techniques
aphasic. to be actively involved and assistive
Hemiplegia: (bilateral activities, devices to ensure
44 R; 46 L activities with numbers independent
Treated area: or letters, cards, pick up function.
Upper and lower objects from different 2) adaptation:
extremities places). modification of
B) made activity environment.
meaningful, used cues, and A) participated
facilitated developmental in functional
sequence. activities to
increase function,
tolerance, and
Appendix 3F

balance.
B) splinting.
201

continues
APPENDIX 3F. Continued
202

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and Year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Mazer Total: 86 Hemispheric stroke Gr1: 91.2 days Gr1: 65.5 Gr1: Received a visual Gr2: Received Concurrent 2–4x/wk for N/A 2, 1, 1
2003302 Gr1: 41 that occurred SD: 51.8 days SD: 11.4 information-processing visuoperceptual treatment: All 20 sessions
Gr2: 45 within the previous training program using the retraining with subjects from both
6 months, were Gr2: 66.7 days Gr2: 66.5 UFOV, a software program commercially groups received
licensed to drive SD: 28.2 days SD: 8.9 presented on a large available computer 4 sessions of
before the referent touch-screen computer software, such as physical training
stroke, had driven that measure 3 aspects of Tetris, Mastermind, on the Baltimore
in the 6 months visual attention: speed of Othello, and Therapeutic
before the stroke, visual processing, divided Jigs@w puzzle, Equipment
and had a desire to attention, and selected on the same large work simulator.
return to driving. attention. screen computer Simulation of
Hemiplegia: as the UFOV. They turning the steering
Gr1: 22 R; 25 L were selected so wheel and pressing
Gr2: 25 R; 25 L that they targeted the gas and brake
TOPICS IN STROKE REHABILITATION/SPRING 2006

perceptual and pedals was provided

cognitive skills but to refine motor
not the elements skills and familiarize
of speed of visual with the adaptive
processing. equipment. 20
sessions, 30–60
min.

Morioka Total: 26 Patients who Gr1: 65.4 days Gr1: 62.6 Gr1: Perceptual learning Gr2: Control Rehab program Gr1: N/A 2, 2, 1
2003303 Gr1: 12 were receiving SD: 18.6 days SD: 13.3 exercise: Exercises to group: No exercise. (PT that included Everyday for
Gr2: 14 hospital rehab and discriminate the hardness ordinary postural 10 days over
whose standing Gr2: 61.9 days Gr2: 61.3 of sponge rubber placed control exercises, 2 wk
maintenance SD: 20.8 days SD: 11.0 under the sole of the foot. such as maintenance
was becoming Three 30-cm² (5,10,15 of standing, shift of
independent. mm thickness) sponge the weight loads to
Hemiplegia: rubbers of identical shape the non-affected and
Gr1: 6 R; 6 L and material composition affected side of the
Gr2: 9 R; 5 L were placed under the sole foot on the health
of the foot. Subjects were meter, and OT).
in standing posture and
were blindfolded. They
were instructed to estimate
hardness of sponge.
Page Total: 16 Stroke onset 1.8 yr 63.2 Gr1: Tape-recorded Gr2: Listened to Both groups 4 wk N/A 2, 0, 1
2000304 Gr1: 8 between 1–3 yr. Range, 1–3 yr SD: 4 imagery intervention: 5 a 20-min tape received OT 3
Gr2: 8 Patients >55 yr of min of relaxation followed with instructions x/wk in 30-min
age. by 10 min of suggestions and information outpatient sessions
Right arm for external cognitive requiring the for 4 wk. Treatment
hemiparesis. visual images related to patient’s attention consisted of 40%
Treated area: using the affected arm and participation, neurodevelopmental
Affected arm in weight-bearing and provided techniques and
functional tasks that were with recorded 60% compensatory
being practiced during information on strategies using
OT sessions and finally 5 the causes and the ipsilateral,
min of refocusing into the pathology of unaffected
room. stroke. The course extremity.
of stroke and the
outcomes that they
could anticipate
from regular
participation.

Yekutiel Total: 39 2 or more yr after a 6.2 yr Gr1: 64 Gr1: Feedback group: 45- Gr2: Control: No N/A 45-min N/A 0, 0, 1
1993305 Gr1: 20 major stroke with SD: 4 yr SD: 9.25 min sessions, 3 sessions/wk treatment for any sessions, 3
Gr2: 19 persisting sensory Chronic for 6 wk. Identify the possible learning sessions/wk
deficit in the hand, Gr2: 67 number of touches or lines effect of repeating for 6 wk
no communication SD: 9.25 and of numbers and letters the sensory
problems or drawn on the arm and examination.
significant cognitive hand. Blindfolded “find
or emotional your (paretic) thumb.”
disturbance. Discrimination of shape,
Hemiplegia: weight, and texture of
Gr1: 10 R; 10 L objects or materials placed
Gr2: 9 R; 10 L in their hand. Passive
Treated area: Sound drawing: Patient is unable
and paretic arm, to see his hand. The
hand therapist held his hand and
a pencil and drew with it.
Patient had to identify the
figure drawn by choosing
from 4 figures on a card.
Difficulty increased as
proprioception improved.
They alternated between
functional and paretic
Appendix 3F

hands.

Note: ADL = activity of daily living; BIT = Behavioral Inattention Test; CVA = cerebrovascular accident; Gr = group; L = left; LE = lower extremity; N/A = not available; OT = occupational therapy; PT = physiotherapy; Quality
R, B, W = randomization, blinding, withdrawals; R = right; ROM = range of motion; UE = upper extremity; UFOV = useful field of view.
203
APPENDIX 3G. Characteristics of Included Studies for Constraint-Induced Movement Therapy
204

Age, yr Session/
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Dromerick Total: 20 Inclusion: Admission to Equal or less Gr1: 61.5 Gr1: CIMT circuit training Gr2: Standard Gr1&Gr2: Treatment N/A 2, 0, 1
2000319 Gr1: 11 inpatient rehab within 14 than 14 days SD: 13.7 encouraged the use customary Routine Gr1 and Gr2:
Gr2: 9 days of ischemic stroke. of the hemiplegic arm OT treatment interdisciplinary 2 hr a day,
Persistent hemiparesis leading Gr2: 71.4 with a variety of UE and that included stroke rehab. 5 x/wk for
to impaired upper extremity SD: 5.3 functional tasks. Subjects compensatory 2 wk
function (score of 1 or 2 on wore padded mitten for at techniques
the motor arm item of the least 6 hr/day to discourage for ADL, UE
National Institute of Health them from using their strength, and
Stroke Scale). Evidence of unaffected hand outside ROM and
preserved cognitive function of therapy. OT treatment traditional
(0 or 1 on the consciousness, focused on ADLs and UE positioning.
communication, and neglect training. Patients also
items of the NIHSS). Presence participated in a
of a protective response circuit-training
TOPICS IN STROKE REHABILITATION/SPRING 2006

(score of > or = to 3 on the program,

upper-arm item of the Motor allowing them to
Assessment Scale), and no perform bilateral
upper extremity injury or self-ROM and
condition that limited use functional
before the stroke. Exclusion: activities in
Hemorrhagic stroke. a supervised
Hemiplegia: setting.
Gr1: 8 R; 3 L
Gr2: 6 R; 3 L
Ploughman Total: 23 Inclusion: N/A Gr1: 57.8 Gr1: Wearing a thumbless Gr2: N/A N/A N/A 2, 0, 1
2004320 Gr1: 10 1) first ischemic or SD: 10.65 mitten that extended Conventional
Gr2: 13 hemorragic stroke, from the fingertips to just treatment
2) currently receiving Gr2: 61.62 below the elbow, on the involving
physical rehab at least 2x/wk SD: 5.68 uninvolved hand. The facilitation
as inpatient or outpatient, mitten discouraged use of proximal
3) no more than 16 wk of the uninvolved arm motor control
postroke, 4) motor control and hand but permitted progressing
of upper extremity not more bilateral activities and to skilled-task
than stage 6 on Chedoke- use of sound arm to training. Subjects
McMaster. stabilize when walking. also received
Hemiplegia: The protocol of wear was strength and
Gr1: 4 L; 6 R progressive, beginning with endurance
Gr2: 9 L; 4 R 1 hr/day and increasing to training,
6 hr/day by wk 2 of rehab functional
and continuing at that level electric
for the remaining rehab stimulation, gait
period. training, and
education, as
appropriate.

Sterr Total: 15 Capability of extending the 4.8 yr Gr1: 49.9 Gr1: CIMT standard Gr2: Same as N/A 5x/wk for Follow- 1, 0, 1
2002321 Gr1: 7 affected hand against gravity SD: 4.7 yr SD: 18.5 protocol with constraint Gr1 except that 2 wk ups at 3,
Gr2: 8 (wrist: 20º; fingers: 10º); of the unaffected arm the shaping 4, 5, and
minimal balance problems; Gr2: 68.4 for a target of 90% of procedure was 3 6 wk
only minor spasticity of the SD: 7.0 waking hr and 6 hr of hr/day.
affected hand; no receptive daily training by using a
aphasia; Mini-Mental State shaping procedure. Every
Examination score > 20; and weekday during 2 wk.
post-stroke interval > 12 On the weekend, only the
months. constraint was worn and
Hemiplegia: no training was provided.
Gr1: 1 R; 6 L 2 types of constraints were
Gr2: 3 R; 5 L used: Resting hand splint
and arm sling together (for
patients without balance
problem) or a specially
designed half glove (for
patients with balance
problems).
Appendix 3G
205

continues
APPENDIX 3G. Continued
206

Age, yr Session/
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

van der Lee Total: 62 Single stroke at least 1 yr Gr1: 3.4 yr Gr1: 59.6 Gr1: Constraint-induced Gr2: Control N/A 5x/wk for 4 wk, 1 yr 2, 0, 1
1999322 Gr1: 31 before the start of the study, SD: 2.45 yr SD: 6 training of the upper group. Patients 2 wk
Gr2: 31 the stroke resulting in extremity. Patients had their were treated
hemiparesis on the dominant Gr2: 2.7 yr Gr2: 62.8 healthy UE immobilized according to
side, a minimum of 20º of SD: 2.8 yr SD: 8 in a splint and a closed the NDT. All
active wrist extension and 10º arm sling. The duration of activities were
of finger extension. Age 18– immobilization was 6 hr/ performed
80 yr. Able to walk indoors day. Training of the affected bimanually, and
without a walking aid, no arm focused on functional if necessary the
major balance problems, tasks for 10 sessions. affected arm
no severe aphasia, and no Therapist’s assistant was supported
severe cognitive impairment. was there to give verbal with the
Sensory disorders and feedback and hands-on unaffected hand.
hemineglect were recorded. facilitation and to inhibit Symmetrical
TOPICS IN STROKE REHABILITATION/SPRING 2006

Hemiplegia: inappropriate muscle posture and

Gr1: 26 R; 5 L contraction. Attention was inhibiting
Gr2: 25 R; 6 L paid to avoiding associated inappropriate
proximal movement and “synergistic”
to relaxing through verbal movements were
guidance. emphasized.

Wittenberg Total: 16 Exclusion: Patients without Gr1: 34 Gr1: 65 Gr1: CIMT group. Restraint Gr2: Control N/A Gr1: 8 days N/A 2, 0, 1
2003323 Gr1: 9 voluntary extension of at least months Range, of unaffected upper group. Passive of 6 hr and
Gr2: 7 10º of the affected fingers or Range, 16– 41–81 extremity (using hand therapy 2 weekend
20º of the wrist. 86 months splint and sling ensemble) (stretching and days of 4 hr.
Hemiplegia: N/A Gr2: 63 during waking hr and heat) to the Gr2: 8 days
Gr2: 28 Range, task-oriented therapy of affected upper of 3 hr of
months 50–75 affected upper extremity. extremity for therapy and
Range, 12– Progressively improving 1 hr during 2 weekend
48 months motor task performance by weekday days of rest.
a successive approximation sessions.
procedure during
combined physical,
occupational, and
recreational therapy.

Note: ADL = activity of daily living; CIMT = constraint-induced movement therapy; Gr = group; L = left; N/A = not available; NDT = neurodevelopmental treatment; OT = occupational therapy; Quality R, B, W =
randomization, blinding, withdrawals; R = right; ROM = range of motion; UE = upper extremity; x/wk = times per week.
APPENDIX 3H. Characteristics of Included Studies for Shoulder Subluxation

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Chantraine Total: 120 Participants were Patients were Gr1: 52.7 Gr1: FES; 4 channels, Gr2: Control: No Conventional 3 sequences/ N/A 0, 0, 1
1999340 Gr1: 60 diagnosed with a followed in SD: 12.6 4 electrodes; current treatment. rehab therapy day for 24
Gr2: 60 subluxated and the rehab rectangular biphasic wave; according to the months
painful hemiplegic service starting Gr2: 52.3 pulse width: 350 µs; duty Bobath concept
shoulder. between the SD: 14.4 cycle: 1:5; 3 sequences: 1st
Hemiplegia: 2nd and 4th wk = 8Hz, 90 min; 2nd = 40Hz,
75 L; 55 R after the causal 30 min; 3rd = 1Hz, 130
lesion was min for the 1st wk. In 2nd
diagnosed. and 3rd wk, sequence 1 and
2 increases of 5 min each,
same in 4th and 5th wk.
Participants were followed
for 24 months as inpatients
during their hospitalization
and then as outpatients after
discharge.

Dean Total: 28 Inclusion criteria: Gr1: 32.1 days Gr1: 58.1 Gr1: Prolonged positioning Gr2: Control: No Concurrent 5 days/wk for None 2, 0, 1
2000335 Gr1: 14 Fewer than 10 wk SD: 13.4 days SD: 12.5 of the affected shoulder, treatment. therapy: 6 wk
Gr2: 14 from the onset of 20 min in each position: Multidisciplinary
stroke; score of less Gr2: 35.3 days Gr2: 58.2 1) supine, shoulder in therapy (PT,
than 5 on the upper- SD: 11.9 days SD: 10.5 max tolerable ABD and psychology, OT,
arm function item of ext rotation, elbow flexed; social work, and
the Motor Assessment 2) supine, shoulder ABD speech therapy
Scale for stroke; no to 90°, max tolerable ext as indicated).
pre-morbid shoulder rotation, elbow flexed; 3) Active training
pain; no pre-morbid sitting, shoulder forward of reaching and
restriction of shoulder flexed at 90°, elbow ext, manipulation
movement; passive wrist ext, cylinder in hand determined
ROM shoulder to provide a web space by the treating
abduction and flexion stretch. therapist.
>90°; and able to
comprehend and use
VAS for pain.
Hemiplegia:
Gr1: 6 L; 4 R
Appendix 3H

Gr2: 6 L; 7 R
207

continues
APPENDIX 3H. Continued
208

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Faghri Total: 26 Participants had Gr1: 16 days Gr1: 65 Gr1: FES: 1.5–6 hr/day, Gr2: Control: No Conventional 6 wk, 7 days/ Post- 1, 0, 0
1994336 Gr1: 13 recent hemiplegic SD: 5 days SD: 13 7 days/wk, for 6 wk; treatment. physical therapy wk treatment
Gr2: 13 stroke or patients active electrode over follow-up
with shoulder muscle Gr2: 17 days Gr2: 69 posterior deltoid and done at
flaccidity/paralysis. SD: 4 days SD: 12 passive electrode over 6 wk
Side of hemiplegia: supraspinatus; frequency
left: 17 of 35 Hz; contraction/
Gr1: 8 relaxation ratio progressively
Gr2: 9 increased (10/12 s to 30/2 s
ON-OFF); intensity set to
obtain the desired motion
of humeral elevation
with some abduction and
extension to pull the head
TOPICS IN STROKE REHABILITATION/SPRING 2006

of the humerus into the

glenoid cavity.

Hanger Total: 98 Participants had Gr1: 12.2 days Gr1: 79.1 Gr1: Strapping: 6 wk Gr2: Control: No N/A 7 days/wk Follow- 2, 0, 1
2000337 Gr1: 49 acute hemiplegic SD: 7.9 days SD: 8.2 or until able to achieve strapping. and 24 hr/ ups at wk
Gr2: 49 stroke (excluding abduction to 90° against day for 6 wk 6 and 14
subarachnoid Gr2: 12.4 days Gr2: 77.8 gravity for 2 s or until
hemorrhage) in the SD: 6.9 days SD: 7.3 discharged from hospital.
preceding 4 wk. Replaced every 2–3
Side of hemiplegia: days. Technique: 2 main
Gr1: 27 L; 22 R supporting tapes, 5 cm
Gr2: 24 L; 25 R above elbow, moving arm
front and back crossing at
top of shoulder; posterior
tape anchored down past
clavicle, whereas the tape
from the anterior aspect
came across the shoulder
and down past the spine of
scapula; and 3rd main tape
applied from medial 1/3
of clavicle, around surgical
neck of humerus and along
spine of scapula (medial
2/3).
Kobayashi Total: 17 Patients with Gr1: 60.3 wk Gr1: 59.3 Gr1: FES to the Gr3: Control: No Conventional 2 times/day, 0, 0, 1
1999341 Gr1: 6 hemiplegia as a result SD: 57.1 wk SD: 13.1 supraspinatus: active treatment. physical therapy 5 days/wk for
Gr2: 6 of a stroke, who electrode: 5 cm from 6 wk
Gr3: 5 showed downward Gr2: 95.0 wk Gr2: 69.3 acromion, on the
subluxation on a SD: 90.5 wk SD: 7.4 supraspinatus fossa; passive
stress X-ray. electrode: acromion.
Side of hemiplegia: Gr3: 190.2 wk Gr3: 53.2 Gr2: FES on the middle
Gr1: 3 R; 3 L SD: 65.5 wk SD: 9.2 deltoid: active electrode: 5
Gr2: 5 R; 1 L cm distal from acromion on
Gr3: 2 R; 3 L the middle deltoid; passive:
posterior part of the axilla.
Pulse duration: 0.3 ms,
15 s ON, 15 s OFF; ramp
up/down: 3 s/2 s; negative
monophasic rectangular
pulses, frequency: 20 Hz,
5–15 min

Linn Total: 40 Participants had no N/A Gr1: 71 Gr1: FES, asymmetrical Gr2: Control: No Conventional PT 4 times/day Follow-up 1, 0, 1
1999338 Gr1: 20 previous pathology Gr2: 73 biphasic pulses, 4 times/day, treatment. and OT for 4 wk at 8 wk
Gr2: 20 to the shoulder; CVA with a minimum of 2 hr treatment
had to have resulted between sessions, 30 min in and 8 wk of
in significant motor wk 1, 45 min in wk 2–3, 60 concurrent
deficit of upper limb min in wk 4, 2 electrodes therapy only
with a grade of ≤ 2 on supraspinous fossa and
on Manual Muscle posterior aspect of upper
Test; adequate arm, 300 µs, 30 Hz, 15 s
communication ON, 15 s OFF, ramp up: 3 s,
ability; no cardiac ramp down: 3 s
pacemaker or metal
in situ; no women
of childbearing age;
and recruitment
and all initial
measurements must
have been completed
and treatment
commenced within
48 hr of admission.
Side of hemiplegia:
Gr1: 4 R
Gr2: 5 R
Appendix 3H
209

continues
APPENDIX 3H. Continued
210

Age, yr Session
Author Sample Time since (mean, SD Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Wang Total: 32 Participants Gr1: 15.9 days Gr1: 56.1 Gr1&3: FES for 6 wk, Gr2&4: Control: N/A 5 days/wk; N/A 1, 0, 0
2000339 Gr1: 8 were required to SD: 3.2 days SD: 7.4 routine therapy for 6 No treatment. 6 wk with
Gr2: 8 demonstrate a wk, FES for 6 wk. FES: FES, 6 wk
Gr3: 8 minimum of 9.5 mm Gr2: 14.9 days Gr2: 56.4 Asymmetric biphasic without, and
Gr4: 8 of acromiohumeral SD: 2.7 days SD: 8.4 waveform stimulation of 6 wk with it
distance. 2 groups supraspinatus and posterior again
short duration: Gr3: 427 days Gr3: 58.1 deltoid, active electrode on
symptoms lasting <21 SD: 45.5 days SD: 6.0 supraspinatus, 6 hr, 10–24
days; long duration: pulses/s, frequency set to
>365 days. Gr4: 433.6 days Gr4: 58.4 create a tetanized muscle
Side of hemiplegia: SD: 47.4 days SD: 6.7 contraction; duty cycle: 1:
Gr1: 4 R 3; increases of ON time by
Gr2: 5 R 2 s every 1 or 2 days, when
Gr3: 3 R a 24-s decrease of OFF time
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr4: 3 R by 2 s every 1 or 2 days

until 2 s.
Williams Total: 26 Participants had 64.9 days 66.1 Gr1: Bobath shoulder roll: Gr2: Henderson N/A 1 session (1 N/A 0, 0, 0
1988342 a diagnosis of SD: 30.0 days SD: 9.4 A strip of foam-rubber shoulder ring: day)
hemiplegia, impaired padding 1-cm thick and 1.25-cm thick piece
motor function of 14-cm wide secured into of polyethylene
the affected upper a roll 5.5 cm in diameter foam modified to
extremity, clinically placed inside a piece of the size of subjects,
subluxated GHJ of stockinette 4-cm wide and warmed in oven
that extremity as long enough to pass under until flexible. With
determined by finger both axillae and form a teardrop extension
palpation. figure eight across the back. lying across chest
Side of hemiplegia: wall, the foam was
16 L; 10 R molded around
the shoulder.
Shorter end
contoured under
paralyzed axilla;
longer straight
section molded
over superior and
posterior aspects of
shoulder, joining
the section from
axilla to form a
ring.

Note: CVA = cerebrovascular accident; FES = functional electrical stimulation; Gr = group; L = left; N/A = not available; OT = occupational therapy; PT = physiotherapy; Quality R, B, W = randomization, blinding,
withdrawals; R = right; ROM = range of motion.
Appendix 3H
211
APPENDIX 3I. Characteristics of Included Studies for Electrical Stimulation
212

Age, yr Session
Author Sample (mean, SD Comparison Concurrent frequency and Follow-up Quality
and year size Population details Time since onset for control) Treatment group therapy duration duration (R, B, W)

Baker Total: 63 All subjects were Gr1: 46 days Gr1: 55 Gr2: NMES of the Gr1: Standard NA 5 sessions/wk Follow- 1, 0, 0
1986356 Gr1: 32 required to demonstrate SD: 51 days SD: 12 posterior deltoid and customary for 6 wk up at 3
Gr2: 31 a minimum of 5 mm of supraspinatus muscles. “hemi-slings” months post
shoulder subluxation Gr2: 49 days Gr2: 56 or wheelchair treatment
in their involved upper SD: 32 days SD: 13 arm supports
extremity, as compared when sitting or
with radiographs standing.
of their uninvolved
extremity.
Hemiplegia:
Gr1: 18 L; 14 R
Gr2: 16 L; 15 R
Treated area:
Hemiplegic shoulder
TOPICS IN STROKE REHABILITATION/SPRING 2006

Bowman Total: 30 Patients had: Ranged from 3 wk N/A Gr2: Received a Gr1: Control: Both groups 10 sessions/wk N/A 2, 0, 0
1979357 Gr1: 15 1) unilateral hemiplegia to 4 months combination of Standard received standard for 4 wk
Gr2: 15 as a result of either a positional feedback and customary customary care
thrombotic or embolic electrical stimulation physical of the hand and
CVA ranging from 3 (referred to as PFST) therapy (no wrist that, while
wk to 4 months; 2) a to the wrist extensor PFST). individualized,
minimum of 5° and muscles, twice daily generally
a maximum of 30° for 30 min, while involved PROM
active extension at sitting. The stimulator exercises,
the wrist from normal had a duty cycle of active resistive
posturing position, 6–8 on, and 20-s exercise, classic
with full PROM; and 3) rest. The stimulation neuromuscular
sufficient cognition to would result in wrist facilitation. and
follow instructions. extension, which the training for ADL.
Location of hemiplegia subject was instructed Physical therapy
was not reported to repeat between 20 was given 5 days/
Area treated: and 100 times, based wk for 4 wk.
Hemiplegic wrist upon his/her level of
endurance.
Cauraugh Total: 34 Inclusion: Diagnosis Gr1: 3.18 yr Gr1: 65.15 Gr1: Surface electrodes Gr3: Did N/A Gr1 & Gr2: 1, 0, 0
2003358 Gr1: 14 of no more than 2 Gr2: 3.29 yr Gr2: 67.28 were attached to 3 not receive 2 days of 90
Gr2: 14 strokes, a lower limit Gr3: 3.03 yr Gr3: 65.82 sets of muscles in any active min training
Gr3: 6 of 10º of voluntary the impaired upper stimulation. for each of 2
wrist/finger extension extremity: extensor Each joint/set wk with at
from a 90º wrist flexed communis digitorum/ of muscles least 24 hr of
position, an 80% extensor carpi ulnaris, were passively rest between
upper limit of motor triceps brachii, moved and sessions. 1
recovery as assessed anterior/middle deltoid. subjects session = 3
by rectified EMG For each movement, attempted to sets of 30
activation patterns and subjects voluntarily voluntarily successful
force generation while generated a target execute active
directly comparing threshold level of EMG wrist/finger neuromuscular
the impaired and activity. Automove EMG extension, stimulation
unimpaired limbs, facilitator provided an elbow trials with 3
absence of other electrical stimulation extension, and movements
neurological deficits, that assisted the shoulder joint executed 10
currently not muscles through full abduction. times/set.
participating in another range. Settings for Gr3: 90 min/
upper extremity rehab stimulation: 1 s ramp day over 4
protocol. up, 5 s of biphasic days.
Hemiplegia: stimulation at 50 Hz,
Gr1: 8 R; 6 L 1 s ramp down, initial
Gr2: 7 R; 7 L threshold 50 microV,
Gr3: 4 R; 2 L 25 s of rest between
Treated area: Upper trials. 10 consecutive
extremity movement trials
(blocked practice)
were executed before
the electrodes were
disconnected and
replaced with electrodes
for another set of
muscles.
Gr2: Same as Gr1
except that electrodes
for different muscles
were changed after each
trials (random practice).
Appendix 3I
213

continues
APPENDIX 3I. Continued
214

Age, yr Session
Author Sample (mean, SD Comparison Concurrent frequency and Follow-up Quality
and year size Population details Time since onset for control) Treatment group therapy duration duration (R, B, W)

Cozean Total: 32 Medically stable (No of subjects Gr1: 62 Gr2: BFB to the anterior Gr1: Control: Standard 3 sessions/wk Follow-up 1, 1, 1
198851 Gr1: 8 patients; cognitive greater than 1 Gr2: 51 tibialis: subject was Standard customary care for 6 wk at 4 wk
Gr2: 8 ability to give consent; year post stroke) Gr3: 52 asked to repeatedly physical as prescribed for from end of
Gr3: 8 ability to ambulate a Gr1: 5 yr Gr4: 56 produce a maximum therapy for 30- each patient. treatment/
Gr4: 8 short distance with the Gr2: 4 yr SD: N/A contraction for 30 s, min session. 10 wk from
assistance of a single Gr3: 4 yr followed by a 1-min Passive baseline
therapist; common Gr4: 3 yr rest during 30-min and active
dynamic gait problem sessions. ROM, and
of spastic; equinus Gr3: FES to produce strengthening
posturing of the the contraction of exercises were
affected leg; and had the anterior tibialis practiced with
suffered cortical or during swing phase all major joints
subcortical infarction or and gastrocnemius and muscle
hemorrhage. during the stance phase. groups of both
TOPICS IN STROKE REHABILITATION/SPRING 2006

Hemiplegia: Treatments lasted 30 legs. Special

Gr1: 5 R; 3 L min and were given by attention
Gr2: 6 R; 2 L a Medtronic Respond was paid to
Gr3: 4 R; 4 L II stimulator, while ankle and
Gr4: 3 R; 5 L pulse frequency was foot control
Treated area: Lower set at a level where in the affected
hemiplegic leg maximum contraction limb. Specific
was achieved within ambulation
tolerable levels of training was
discomfort. also conducted
GR4: BFB and FES: to overcome
Within the 30-min dynamic gait
sessions, 15 min abnormalities.
were devoted to
each modality. In the
subsequent wk, FES
was used for 5 min,
with 30 min provided
for independent use
of the stimulator post
treatment sessions.
De Kroon Total: 28 Patients had interval of Gr1: 14.7 months Gr1: 58 Extensors and Extensors only: N/A 3 x/wk a Follow-up 2, 0, 1
2004359 Gr1: 13 more than 6 months SD: 11.8 months SD: 17.3 flexors: Splint was Same as GR1 starting at 6 wk
Gr2: 15 since unilateral stroke prepared in which the except that the with 20
in territory of middle Gr2: 21.4 months Gr2: 61.7 electrode position was exercise/open min/session,
cerebral artery; between SD: 16.1 months SD: 9.7 individually adjusted mode was used and during
18–80 yr old; impaired to evoke optimal finger and subjects first 10 days,
function of the upper movements according received stimulation
extremity due to spastic to randomization; NESS stimulation of time was
paresis; voluntary Handmaster was used: the extensor gradually
extension of wrist (at 5 surface electrodes muscles only. increased to a
least 10°) and fingers; with external control maximum of 1
and stable general box connected to the hr/session
health status. splint with a cable.
Hemiplegia: Frequency: 36 Hz, duty
Gr1: 3 R; 10 L cycle: 40%; exercise
Gr2: 8 R; 7 L mode was used.
Treated area: Affected Alternating stimulation
upper extremity of extensor and flexor
muscles for 6 wk.

Faghri Total: 26 Recent hemiplegic Gr1: 17 days Gr1: 69 Gr2: Functional Gr1: Standard 7 sessions/wk Follow-up 1, 0, 0
1997360 Gr1: 13 stroke patients with SD: 4 days SD: 12 NMES of the posterior, Control: No customary care: for 6 wk at 6 wk post
Gr2: 13 shoulder muscle deltoid muscle, and neuromuscular 1) bed mobility; treatment
flaccidity/paralysis. Gr2: 16 days Gr2: 65 supraspinatus muscle stimulation. 2) head/trunk
Hemiplegia: SD: 5 days SD: 13 for 1.5 to 6 hr/day. control;
Gr1: 9 L; 4 R Electrodes were placed 3) sitting balance;
Gr2: 8 L; 5 R over the deltoid muscle 4) using a tilt
Treated area: and the supraspinatus table for a sense
Hemiplegic shoulder muscle, utilizing a of verticality;
configuration that 5) active assisted
minimizes activation ROM to the
of the upper trapezius sound side.
muscle.
Appendix 3I
215

continues
APPENDIX 3I. Continued
216

Age, yr Session
Author Sample (mean, SD Comparison Concurrent frequency and Follow-up Quality
and year size Population details Time since onset for control) Treatment group therapy duration duration (R, B, W)

Heckmann Total: 28 Right-handed Gr1: 56.1 days Gr1: 50.1 Gr1: EMG-BFB: Gr2: Standard N/A 4-wk N/A 1, 0, 0
1997361 Gr1: 14 stroke patient, large SD: 24 days SD: 14 Transcutaneous customary treatment
Gr2: 14 supra tentorial electrical stimulation care only:
cerebrovascular lesions Gr2: 61.6 days Gr2: 54 was triggered by Treatments
SD: 40 days SD: 11 voluntary EMG-activity based on
in the target muscles Bobath’s
Acute phase post- (upper arm/forearm/ principles; 45
stroke hand extensors, knee min/session.
flexors, and ankle Also OT at
extensors) to produce least 3 hr/wk
joint movement; and up to 2
parameters of hr of group
stimulation were 0,3 ms therapy.
biphasic sinus-waved
TOPICS IN STROKE REHABILITATION/SPRING 2006

pulses at 80 Hz, and

a constant current of
20–60 mA for 1 s; 15
stimulations/session for
each group of muscle.
Also standard
customary care:
See control group
description.
Kimberley Total: 16 Subjects at least 6 Gr1: Range, Gr1: Range, Gr1: NMES: First visit, Gr2: Same N/A 6 hr/day for 10 Follow-up 1, 1, 1
2004362 Gr1: 8 months post stroke 11–58 months 33–75 patients were instructed instructions days over the at 3 wk
Gr2: 8 with at least 10º of SD: 11.75 months SD: 10.5 in the operation of the as Gr1 but the course of 3 wk (end of
active flexion/extension electrical stimulator. machine did (total 60 hr) treatment)
movement at the Gr2: Range, 7–74 Gr2: Range, Subjects maintained not deliver
metacarpophalangeal months 41–78 records of use to show any current.
joint of the index SD: 16.75 months SD: 9.25 compliance. Half of Half the hr
finger; must have the treatment time was of treatment,
score of at least 25 spent with active effort the subjects
out of a possible 30 on the subject’s part, were asked to
in Mini-Mental State requiring him or her lift the hand
Examination. to trigger a stimulated actively when
Hemiplegia: response, while the the light came
Gr1: 4 R; 4 L other half of the time on and the
Gr2: 4 R; 4 L was spent with the “stimulation”
Treated area: machine automatically started.
Hemiplegic forearm stimulating the muscles
to contract cyclically
without any volitional
trigger from the patient.

Popovic Total: 28 Had >2 wk and <6 Gr1: Range, 4–11 Gr1: Range, Gr1: The exercise Gr2: 30 min Standard 30-min long Follow-ups 2, 0, 1
2003363 Gr1: 8 months following the wk 50–68 was assisted with a of supervised customary care. exercise with at 3,10, 23
Gr2: 8 first CVA that was Gr2: Range, 4–8 Gr2: Range, neural prosthesis that daily exercise the paretic wk post-
Gr3: 6 caused by ischemia wk 53–72 controlled the opening, with the arm and hand treatment
Gr3: 6 or hemorrhagia and Gr3: Range, 4–11 Gr3: Range, grasping, and releasing paretic arm every day for
confirmed by MRI or wk 37–75 functions by mimicking and hand. 3 wk
CT; were above age Gr4: Range, 5–11 Gr4: Range, natural movement. Four Same tasks
18; were able to give wk 45–72 channels of electrical as the FES
informed consent; and stimulation were group without
were able to understand applied over the finger a neural
how to apply electrical flexors, finger extensors, prosthesis.
stimulation for thumb extensor, and Gr4: Same as
controlling the grasp. the thenar muscle Gr2.
Hemiplegia: group. Frequency: 50
Gr1: 3 L; 5 R Hz, pulse duration:
Gr2: 1 L; 7 R T = 200 µs, intensity:
Gr3: 1 L; 5 R 20mA–45mA. During
Gr4: 2 L; 4 R the sessions, subjects
were instructed to try
to functionally use a
toothbrush, comb,
Appendix 3I

telephone receiver, pen,

small food, etc.
Gr3: Same as Gr1.
217

continues
APPENDIX 3I. Continued
218

Age, yr Session
Author Sample (mean, SD Comparison Concurrent frequency and Follow-up Quality
and year size Population details Time since onset for control) Treatment group therapy duration duration (R, B, W)

Powell Total: 60 Hemiparesis due to Gr1: 22.9 days Gr1: 66.4 Gr2: FES of wrist and Gr1: Control: Standard 3 half-hour Follow- 2, 0, 1
1999364 Gr1: 30 acute stroke with SD: 5.5 days SD: 12.2 finger extensors. No FES. treatment from periods daily up at 24
Gr2: 30 Medical Research Standard ward rehab team for 8 wk wk post
Council power of wrist Gr2: 23.9 days Gr2: 69.0 customary PT. (e.g., contact with treatment/
extension grade of 4/5 SD: 7.7 days SD: 10.8 physiotherapists 32 wk from
or worse; and at 2 to 4 and occupational baseline
wk after onset. therapists).
Hemiplegia:
Gr1: 20 L; 10 R
Gr2: 18 L; 12 R
Area treated:
Hemiplegic wrist

Winchester Total: 40 1) Minimum of 5° of Gr1: 59 days Gr1: 60 Gr2: 1) PFST of the Gr1: Control: Physical therapy PFST: 7 N/A 1, 0, 1
TOPICS IN STROKE REHABILITATION/SPRING 2006

1983365 Gr1: 20 active knee extension SD: 40 days SD: 10 quadriceps femoris No electrical 5 days/wk, 4 sessions/wk for
Gr2: 20 in antigravity position. muscle and rectus stimulation times daily. 4 wk;
2) No greater than Gr2: 45 days Gr2: 57 femoris muscle for and PFST. It included Electrical
a 25° knee flexion SD: 38 days SD: 13 30 min. Standard neuromuscular stimulation: 28
contracture. 3) Absence 2) Cyclical electrical customary PT. facilitation sessions/wk for
of severe quadriceps stimulation, involving techniques, 4 wk
femoris muscle tone, the quadriceps femoris progressive
as determined by muscle. resistive exercises,
observation of the and weight-
patient’s response to bearing activities.
a quick stretch of the
quadriceps femoris
muscle.
Hemiplegia:
Gr1: 8 L; 12 R
Gr2: 7 L; 13 R
Area treated:
Hemiplegic leg

Note: ADL = activities of daily living; BFB = biofeedback; CES = cyclical electrical stimulation; CVA = cerebrovascular accident; EMG = electromyography; FES = functional electrical stimulation; Gr = group; L = left; N/A =
not available; NMES = neuromuscular electrical stimulation; OT = occupational therapy; PFST = positional feedback stimulation training; PROM = passive range of motion; PT = physiotherapy; Quality R, B, W = randomization,
blinding, withdrawals; R = right; ROM = range of motion; x/wk = times per week.
APPENDIX 3J. Characteristics of Included Studies for Transcutaneous Electrical Nerve Stimulation (TENS)

Age, yr Session
Author Sample Population (mean, SD Concurrent frequency Follow-up Quality
and year size details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Leandri Total: 60 Patients suffering from hemiplegic Gr1: 3.42 months Gr1: 67.90 Gr1: High intensity Gr3: Placebo: Standard 3 sessions/wk Follow- 1, 2, 0
199068 Gr1: 20 shoulder pain following ischemic SD: 2.17 SD: 7.01 TENS, high Patients undergoing customary for 4 wk up done
Gr2: 20 stroke; all were affected by a frequency (100 Hz) sham stimulation care. Basic 1 month
Gr3: 20 discrete loss of motor function, Gr2: 2.72 months Gr2: 65.65 on the shoulder were connected physical post
but were able to stand or walk if SD: 1.88 months SD: 4.94 muscles and the to the stimulator treatment treatment
assisted. glenohumeral joint. whose output every day in
Hemiplegia: Gr3: 3.17 months Gr3: 64.30 Gr2: Low intensity was loaded with the morning
Gr1: 7 L; 13 R SD: 2.72 months SD: 5.40 TENS, high 3 kOhm resistor, (no other
Gr2: 6 L; 14 R frequency (100 Hz) while the circuit details
Gr3: 7 L; 13 R on the shoulder was interrupted. available).
Treated area: Hemiplegic shoulder muscles and the Patients received no
glenohumeral joint. stimulus, but display
showed appropriate
Intensity: N/A values and current.

Levin Total: 13 Spastic hemiparesis due to CVA; Gr1: 26.4 months Gr1: 58.5 Gr1: TENS to the Gr2: Placebo: N/A 5 days/wk for N/A 1, 0, 1
1992390 Gr1: 7 spasticity in lower extremity; SD: 21.9 months SD: 14.7 common peroneal Subthreshold TENS. 3 wk
Gr2: 6 Minimum 10° passive ankle nerve, posterior
dorsiflexion; no history of Gr2: 29.2 months Gr2: 64.7 to the head of
previous neurological disorder; SD: 17.2 months SD: 10.6 the fibula on the
and no pain or major sensory hemiparetic lower
impairment in lower limb. extremity, 60 min/
Hemiplegia: day.
Gr1: 7 R; 4 L Frequency: 99 Hz.
Gr2: 3 R; 3 L Intensity: 125 µs
Treated area: Hemiparetic lower
extremity
Appendix 3J
219

continues
APPENDIX 3J. Continued
220

Age, yr Session
Author Sample Population (mean, SD Concurrent frequency Follow-up Quality
and year size details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Peurala Gr1: 32 Hemiplegia: 3.3 yr 54.4 Gr1: Cutaneous Gr3: Placebo N/A 3 wk N/A 0, 0, 2
2002393 Gr2: 19 14 L; 35 R SD: 10 stimulation stimulation was
Gr3: 8 delivered to delivered in the
affected upper same way but no
extremity with a current was applied.
glove electrode.
The intensity
was adjusted
to just below
sensory threshold.
Monophasic
constant current
twin pulses at 50
Hz. 20 min each
TOPICS IN STROKE REHABILITATION/SPRING 2006

time for 3 wk.

Gr2: Cutaneous
stimulation
delivered to affected
lower extremity
with a sock
electrode as Gr1.
Intensity: N/A

Sonde Total: 44 Nondemented patients who had Gr1: 8.3 months Gr1: 73 Gr2: Low intensity Gr1: Control (no Standard 5 sessions/wk Follow-up 1, 0, 0
1998391 Gr1: 18 a paretic upper extremity (score SD: 2.1 months SD: 3.5 TENS (1.7 Hz) TENS). customary for 3 months done 3
Gr2: 26 of 0–50 in the Fugl-Meyer motor for 60 min on the care at a day (total = 60 months
performance scale: 0 points Gr2: 9.1 months Gr2: 71 elbow extensors care center sessions) post
means no motor function, 66 SD: 2.2 months SD: 6.0 and wrist extensors. twice a wk. treatment
points means normal motor 80% had electrodes (treatment
function), following their first- placed over the group
ever stroke occurring 6–12 elbow extensors only)
months previous to the study. and the shoulder
Hemiplegia: abductors.
Gr1: 7 L; 11 R Intensity: N/A
Gr2: 10 L; 16 R
Treated area: Hemiplegic upper
extremity
Sonde Total: 28 Patients who had participated Gr1: 47 months Gr1: 70.4 Gr1: Low TENS to Gr2: Control: no Daily rehab 7 sessions/wk Follow-up 1, 0, 1
2000397 Gr1: 18 in a randomized trial of daily SD: 7 months SD: 6.1 post-stroke paretic TENS. (no other for 12 wk at 3 yr
Gr2: 10 treatment with low frequency (1.7 arm: elbow flexor information
Hz) TENS on the paretic arm for Gr2: 47 months Gr2: 72.1 muscle, wrist flexor available).
3 months starting 6 months after SD: 7 months SD: 3.7 muscle.
stroke.
Location of hemiplegia was not
reported
Treated area: Post-stroke paretic
arm

Tekeoolu Total: 60 1) Stroke with hemiplegia Gr1: 40.8 days Gr1: 55.9 Gr1: 100 Hz TENS, Gr2: Placebo TENS: Todd Davies 40 sessions N/A 1, 1, 0
1998392 Gr1: 30 or hemiparesis; 2) diagnosis SD: 11.4 days SD: 7.0 200 µs to extensor sham stimulation: exercise over 8 wk
Gr2: 30 determined by physical and muscles of elbow patients were program—a
laboratory examination including Gr2: 44.3 days Gr2: 52.2 and common connected to basic neuro-
radiological examination, CT, and SD: 13.1 days SD: 5.4 peroneal nerve stimulator with a physiological
blood screen; 3) patients affected posterior to the resistor and display treatment
by discrete loss of motor function head of the fibula that showed results; program.
but who were able to stand or on the hemiparetic however, the
walk if assisted. lower extremity, for patients received no
Hemiplegia: half an hour a day. current.
Gr1: 16 R; 14 L
Gr2: 17 R; 13 L
Treated area: Hemiparetic upper
and lower extremity

Note: CVA = cerebrovascular accident; Gr = group; N/A = not available; L = left; Quality R, B, W = randomization, blinding, withdrawals; R = right; TENS = transcutaneous electrical stimulation.
Appendix 3J
221
APPENDIX 3K. Characteristics of Included Studies for Ultrasound
222

Age, yr Session
Author Sample Time since (mean, SD Comparison frequency Follow-up Quality
and year size Population details onset for control) Treatment group Concurrent therapy and duration duration (R, B, W)

Inaba Total: 33 Subjects chosen Gr1: 7 months Gr1: 57 Gr2: Ultrasound to Gr1: Control All groups received ROM exercises 7 sessions/wk N/A 1, 2, 1
1972399 Gr1: 13 were patients with SD: 9 months SD: 8 the paravertebral (only and arm positioning. Self-ROM for 4 wk
Gr2: 10 hemiplegia who and shoulder concurrent exercises for shoulder flexion,
Gr3: 10 had shoulder pain Gr2: 3 months Gr2: 59 joint areas for one treatment). abduction, and external rotation were
occurring within the SD: 2 months SD: 10 5-min treatment Gr3: Placebo: performed with pulley equipment 3x/
range of 0 to 90° of prior to exercise. Ultrasound day, repeating each motion five times
flexion or abduction Gr3: 4 months Gr3: 58 with the during each session. Positioning
of the arm. SD: 3 months SD: 11 energy turned programs were attempted 24 hr/day.
Hemiplegia: off prior to The arm was positioned in maximum
Gr1: 9 L; 4 R exercises. pain-free abduction and external
Gr2: 6 L; 4 R rotation through the use of pillows
Gr3: 3 L; 7 R or slings when the patient was in
Treated area: bed, and with troughs, overhead
Hemiplegic suspension slings, or pillows when
shoulder he was in a wheelchair.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Note: Gr = group; L = left; N/A = not available; Quality R, B, W = randomization, blinding, withdrawals; R = right; ROM = range of motion.
APPENDIX 3L. Characteristics of Included Studies for Acupuncture

Age, yr Session
Author Sample Population (mean, SD Concurrent frequency Follow-up Quality
and year size details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Fink Total: 25 All patients had Gr1: 66.5 months Gr1: 56.0 Gr1: Acupuncture: Verum Gr2: Placebo: To avoid N/A 2 treatments/ N/A 1, 0, 0
2004403 Gr1: 13 hemiparesis SD: 50.2 months SD: 11.8 needles were inserted transdermal stimulation, wk, 4 wk
Gr2: 12 and spastic at acupuncture points placebo needles were
equinovarus Gr2: 64.2 months Gr2: 61.3 GB34, GB39, LR3, LI4 of inserted at defined
deformity of SD: 48.3 months SD: 8.4 the affected limbs, and, nonacupoints. The tip
the affected depending on additional of the needle is blunt,
leg. Threshold symptoms, according and when it touches
Modified to traditional Chinese the skin, patients feel
Ashworth Scale medicine criteria at ST36 a pricking sensation,
(MAS) score and LI10 of the affected simulating puncturing of
required for limbs, and at SP6 and LU9 the skin. A cube-shaped
admittance to this bilaterally. In addition, elastic foam was used
study was 1 or GV20 was needled in 10 to fix the needle on the
more. of 13 patients. The needles skin. It is not visible that
(maximum of 15 needles the blunt placebo needle
per patient and treatment) is not inserted into
were left in place for 30 deeper tissue layers, but
min after insertion without the blunt tip on the skin
further manipulation. may be felt.

Gosman- Total: Patients 40 and Less than Mean age: Deep acupuncture: 10 Placebo control Standard 2 x/wk for 10 months 2, 0, 1
Hedstrom 104 over with acute 1 wk before Gr1: acupuncture points (superficial) customary care. 10 wk
1998404 Gr1: 37 focal ischemic randomization M: 77 were used according acupuncture): 4 short No details
Gr2: 34 nonhemorrhagic F: 75 to traditional Chinese needles (15 mm) placed provided.
Gr3: 33 lesion. Stroke medicine. The needles just under the skin in
onset less than Gr2: were 30-mm long; on the each extremity and left
1 wk before M: 76 nonparetic side they were for 30 min.
randomization. F: 82 stimulated manually every
Patient could not 5 min, until the “de chi” No acupuncture
walk without Gr3: sensation was achieved. control: No acupuncture
support and/or M: 74 On the paretic side, an treatments.
could not eat/dress F: 78 electrical stimulation of 2
without assistance. Hz was used and increased
Patient able to gradually to achieve
cooperate mentally muscle contraction. Each
and willing to treatment last 30 min.
Appendix 3L

participate.
223

continues
APPENDIX 3L. Continued
224

Age, yr Session
Author Sample Population (mean, SD Concurrent frequency Follow-up Quality
and year size details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Johansson Total: 78 Patient able to Wthin 10 days of Gr1: 76 Acupuncture: Treatment Control: No Daily PT and 2 x/wk for NA 1, 0, 1
1993405 Gr1: 40 cooperate during onset began 4 to 10 days after acupuncture treatment. OT. 10 wk
Gr2: 38 examination and Gr2: 75 stroke. Acupuncture
tests. Patients’ was given on both sides,
paresis had to 10 points according
extend to the traditional Chinese
point that patient medicine acupuncture.
could not walk Needles were left in place
without support and stimulated for 30 min,
and could not eat manually and electrically.
and dress without Electrical stimulation was
help; they had to performed on 4 needles
be independent on the paretic side at
for ADL before the 2–5 Hz. The intensity of
stroke. electrical stimulation was
TOPICS IN STROKE REHABILITATION/SPRING 2006

Location of stroke: adjusted to induce muscle

35 L; 43 R contraction.

Kjendahl Total: 41 Patients with Gr1: Gr1: Acupuncture: Sites were Control: No Individually 30 min/ 12 months 2, 0, 1
1997406 Gr1: 21 hemiparesis, median: 39 days Median: 57 selected for the treatment acupuncture treatment. adapted, mul- session, 3–4
Gr2: 20 following a first SD: 22.5 days SD: 17 of stroke in compliance tidisciplinary sessions/wk
ever stroke. with traditional Chinese rehab program. for 6 wk
Patients with Gr2: Gr2: medicine. Manual No details pro-
global aphasia, median: 43 days Median: 58 stimulation of needles was vided.
only if they SD: 22.5 days SD: 16.5 performed in most cases.
had a sufficient However, if indicated,
understanding for low-frequency electrical
participating in stimulation (2–4 Hz) was
the study. applied or moxibustion
Location of stroke: was added. Contralateral
Gr1: 13 L; 8 R motor and sensory areas of
Gr2: 11 L; 9 R the scalp were treated with
high frequency electrical
acupuncture.
Moon Total: 35 Patients with Gr1: 3.7 months Gr1: 58.2 Gr1: Electro acupuncture Gr3: Routine All 3 groups: 3–4 x/wk N/A 1, 0, 0
2003407 Gr1: 15 elbow spasticity SD: 3.7 months SD: 10.8 group: Electrical acupuncture therapy Same
Gr3: 10 who were more stimulation was applied for stroke and ROM acupuncture
than 5 wk out Gr3: 2.7 months Gr3: 65.1 every other day for 15 exercises. therapy for
Gr2 from the onset of SD: 1.4 months SD: 7.9 days (total of 8 sessions). stroke and ROM
is not stroke. Frequency of 50 Hz was exercises once a
defined Affected side: given to 4 needles on day, which had
here Gr1: 8 R; 7 L the paretic side for 30 been ongoing
because Gr2: 5 R; 5 L min. Amplitude strong since admission.
we did Gr3: 4 R; 6 L enough for patients to Given on both
not use feel stimulation but not paretic and
values to elicit visible muscle nonparetic side
of this contractions. during 30 min
group each time.
(moxi-
bustion).

Naeser Total: 16 Patients who 1–3 months Gr1: 61.25 15 to 16 needles of 34 Placebo: Points of Daily physical 5 days/wk for N/A 1, 2, 1
1992408 Gr1: 10 had suffered a SD: 8.6 gauge were inserted into normal electrical therapy. 4 wk
Gr2: 6 left-hemispheric standard acupuncture resistance on the
infarction and Gr2: 61 points on the scalp and nonparalyzed upper and
had significant SD: 8 body and left in place for lower extremities were
hemiparesis. 20 min. Low-frequency identified with a Fluke
Patients had electrical stimulation (1–2 Ohm meter. Needles
to have greatly Hz) was used on selected were inserted into these
reduced arm and points on the paralyzed normal resistance points
leg power with size. and left in place for 20
reduced or no min. Clips were attached
voluntary isolated to the needles but they
finger movement were not connected to
the electrical stimulation
machine.
Appendix 3L
225

continues
APPENDIX 3L. Continued
226

Age, yr Session
Author Sample Population (mean, SD Concurrent frequency Follow-up Quality
and year size details Time since onset for control) Treatment Comparison group therapy and duration duration (R, B, W)

Wong Total: First stroke, 10 to 14 days Gr1: 60.4 Electrical acupuncture: Control: No Standard 5 x/wk for N/A 1, 0, 1
1999409 118 complicated by SD: 11.1 was given at 8 acupuncture treatment. customary rehab 2 wk
Gr1: 59 physical symptoms acupuncture points via program with
Gr2: 59 of hemiplegia, Gr2: 60.6 adhesive surface electrodes multidisciplinary
physically stable, SD: 10.8 of 4 x 5 cm squared at approaches,
with a clear 4 points on the upper including OT
consciousness and extremity and 4 points on and PT for
no complications the lower extremity on the more than 2 hr
during medical paretic side. Stimulation every day until
course. was of 20 to 25 Hz, to discharge.
Affected side: induce muscle contraction,
Gr1: 31 L; 28 R treatment lasting 30
Gr2: 25 L; 31 R min. The magnitude of
stimulation was set at a
level sufficient to induce
TOPICS IN STROKE REHABILITATION/SPRING 2006

muscle contraction (10–20

mV)

Note: ADL = activities of daily living; F = female; Gr = group; L = left; M = male; N/A = not available; OT = occupational therapy; PT= physiotherapy; Quality R, B, W = randomization, blinding, withdrawals; R = right; ROM =
range of motion; x/wk = times per week.
APPENDIX 3M. Characteristics of Included Studies for Intensity and Organization of Rehabilitation

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Ahrens Total: 120 Age 70 or older, N/A 70 or older Gr1: Patients were Gr2: Patients N/A N/A 6 months 1, 0, 0
2004422 Gr1: 60 resided in the study immediately transported received standard
Gr2: 60 catchment area, home after being stroke care and
and had a caregiver stabilized. This rehab services
available. group then received in the general
services at home from ward of the same
multidisciplinary team hospital that
of geriatricians, nurses, provided home
therapists, and social hospitalization
workers who specialize services for the
in stroke care. Such experimental
care included daily group.
physician, nurse, and
physical therapist visits.
A physician and a nurse
remained on call 24 hr
a day. Rehab focused on
participants’ individual
interests, and caregivers
were encouraged to
participate.
Appendix 3M
227

continues
APPENDIX 3M. Continued
228

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Andersen Total: 155 Admitted with acute 3 months Gr1: 69.8 Gr1: Visits focused Gr3: Standard N/A Gr1: Three Follow- 2, 0, 1
2000423 Gr1: 51 stroke as defined Acute SD: 9.9 on early detection aftercare; 1-hr sessions up at 6
Gr2: 44 by WHO criteria; and treatment of outpatient rehab at 2, 6, and months
Gr3: 43 discharge to own Gr2: 74.1 complications, by hospital 12 wk post- (end of
home; impaired motor SD: 11.4 maintenance of physician or GP discharge. treatment)
capacity (SSS score of functional capacity, and and home care Gr2: An
<58 at discharge). Gr3: 68.3 psychological and social to compensate average of
Hemiplegia: SD: 12.3 adjustment to a new life. for disability (no three 1-hr
Gr1: 21 R; 28 L Discussions, medical follow-up home sessions
Gr2: 26 R; 20L exam if needed; patients visits). over a 6-wk
Gr3: 22 R; 19L and caregivers were period.
provided with info on Gr3: N/A
stroke-related issues.
Gr2: Instruction and
TOPICS IN STROKE REHABILITATION/SPRING 2006

reeducation by hospital
PT; evaluation of range
of functions related to
indoor and outdoor
mobility and ADLs.
Instruction given to
relatives and caregivers.
Andersen Total: 155 Admitted with acute N/A Gr1: 69.8 Gr1: Visits focused Gr3: Standard N/A Gr1: 3 6 months 2, 0, 1
2002424 Gr1: 51 stroke as defined Outcomes were SD: 9.9 on early detection customary sessions x 1 end of
Gr2: 44 by WHO criteria; taken 6 months and treatment of aftercare; hr at 2, 6, and treatment
Gr3: 43 discharge to own after discharge Gr2: 74.1 complications, outpatient rehab 12 wk post-
home; impaired motor SD:11.4 maintenance of functional by hospital discharge.
capacity (SSS score of capacity, psychological physician or GP Gr2: 3
<58 at discharge) Gr3: 68.3 and social adjustment and home care sessions
Hemiplegia: SD: 12.3 to a new life with to compensate average x 1
Gr1:19 R; 14 L stroke related disability. for disability (no hr over 6-wk
1 cerebellum Discussion concerning follow-up home period.
1 brainstem health conditions/stroke- visits). Gr3: N/A
Gr2:15 R; 24 L related symptoms/
0 cerebellum functional capacity,
5 brainstem medical exam if needed;
Gr3:15 R; 17 L patients and caregivers
1 cerebellum, 1 were provided info on
brainstem stroke related issues.
Lesion type: Gr2: Instruction and
Haemorrhage/ reeducation by hospital
infarction/no lesion PT; evaluation of range
Gr1: 7/28/12 of functions related to
Gr2: 7/37/6 indoor and outdoor
Gr3: 6/28/8 mobility and ADLs.
Instruction to relatives
and caregivers.
Appendix 3M
229

continues
APPENDIX 3M. Continued
230

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Andersson Total: 124 Patients admitted N/A Gr1: 79 Gr1: Patients received Gr2: The acute N/A N/A Follow-up 0, 0, 1
2002425 Gr1: 56 either to the University SD: 6 treatment from all ward was located at 1 year
Gr2: 68 Hospital in Linkoping categories of personnel in one hospital
between November Gr2: 76 that are requested in and the rehab
27, 1996 and January SD: 9 rehab (PT, OT, nurse, ward located in
8, 1998, or to Eksjo hospital social worker, another hospital,
County Hospital speech therapist). They but under the
between September spent their initial acute same management.
15, 1997 and May 18, care period in a stroke The patients were
1998. All patients had unit. assessed and
a diagnosis of stroke followed up by
and needed continuous a PT and an OT;
rehab after discharge the assessments
from the acute ward. were made
TOPICS IN STROKE REHABILITATION/SPRING 2006

at admission/
discharge from
the acute ward,
admission/
discharge from the
rehab program,
and at 6 and 12
months post-
stroke.
Askim Total: 62 Diagnosis of an acute N/A Gr1: 76.9 Gr1: The extended service Gr2: The ordinary In principle, the N/A 6, 26, 52 2, 0, 1
2004426 Gr1: 31 stroke according to Gr2: 76.3 consists of stroke unit service consists same primary wk
Gr2: 31 the WHO definition treatment combined of treatment health care
of stroke; SSS score with a home-based in a combined system was
> 2 points and <58 program of follow-up acute and rehab available to both
points; living at home care coordinated by a stroke unit and the ordinary
before the stroke; mobile stroke team that further follow-up service and the
inclusion within 72 hr offers early supported organized by rehab extended service
after admission to the discharge and works in clinics and/or the patients, but
stroke unit and within close cooperation with primary health the cooperating
7 days after the onset the primary health care care system. mobile team was
of symptoms; able system during the first 4 only available
and willing to provide wk after discharge. The to the extended
informed consent. mobile team was based service patients.
within the stroke unit and
consisted of a nurse, a PT,
an OT, and the consulting
service of a physician.

Baskett Total: 100 A need for ongoing N/A Gr1: 71.7 Gr2: Home therapy: Gr1: Outpatient N/A Gr1: 5 hr/ Follow- 2, 0, 1
1999428 Gr1: 50 PT and/or OT Acute SD: 9.1 Therapists devised a group: Multidisci- session, 2–3x/ ups at 6
Gr2: 50 after discharge as program of exercises and plinary team as- wk wk and 3
determined by the Gr2: 67.8 activities in collaboration sessment followed months
treating clinicians; SD: 11.6 with the subject; the by goal setting. PT Gr2: 13 wk of (end of
subjects were able to subject could continue practice followed treatment treatment)
travel to the hospital this program throughout principles outlined
outpatient therapy the following wk. This by Bobath, Corr,
departments; subjects was a functional approach and Shepherd.
were residents in a that incorporated, as OT practice was
private home or rest far as possible, goals set centered mainly
home (residential toward restoration or on Bobath neu-
homes for frail elderly improvement of normal rodevelopmental
people) within the activities within the and Corr and
health service region of home and extending the Shepherd motor
the hospital. boundaries of limitation relearning treat-
Hemiplegia: that people had set for ments, together
Gr1: 23 R; 27 L themselves. with Kielhofner’s
Gr2: 16 R; 34 L model of human
occupation.
Appendix 3M
231

continues
APPENDIX 3M. Continued
232

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Bautz- Total: 82 All patients with N/A Gr1: 79.5 Gr1: Early supported Gr2: Conventional Initial care in N/M 3 and 6 2, 0, 1
Holter Gr1: 42 acute stroke (onset Acute SD: 7.5 discharge: Patients were rehab: Patients acute stroke months
2002427 Gr2: 40 less than 6 days prior assessed by a hospital- received the unit. follow-up
to hospitalization); Gr2: 78.0 based multidisciplinary conventional
home-dwelling and not SD: 4.0 project team consisting procedures for
severely disabled prior of a nurse, an OT, and discharge and
to the stroke; no other a PT. One of these 3 continued rehab.
medical condition (primary contact) started
likely to preclude immediate preparations
rehab and medically for discharge and
stable (Barthel score continued rehab provided
5–19; 72 hr post- by the general community
stroke). services organized in
Hemiplegia: 11 different local areas.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr1: 16 R; 23 L Caring for the group was

Gr2: 23 R; 14 L encouraged to establish a
multidisciplinary team for
each patient and support
offered.

Claesson Total: 249 Patients aged 70 yr Less than 7 Gr1: 80.1 Gr1: Stroke Unit (SU): Gr2: General N/A Patients Follow-up 0, 0, 1
2000429 Gr1: 166 or over, acute stroke days Gr2: 79.7 The treatment program Ward: Received in need of after 12
Gr2: 83 within 7 days before was built on the principal conventional acute prolonged months
admission, living in of SU care with a team medical care, PT, rehab were
their own homes in work concept for nursing and OT, although transferred
Göteborg, Sweden, and rehab. Support and not within the to geriatric
without recognized information for relatives framework of stroke units.
need of care. was an important part a structured
Hemiplegia: of the program, as was stroke unit care
Gr1: 85 R; 74 L the focus on the patients’ approach.
Gr2: 37 R; 35 L needs and participation
Speech disorder: in the treatment. Patients
Gr1: 79 in need of prolonged
Gr2: 41 rehab were transferred
to geriatric SU. Patients
received significantly
more occupational and
physical therapy than the
General Ward group.
Corr Total: 110 Patient discharged alive Days from Gr1: 75.1 Gr1: Intervention Gr2: Control N/A Gr1: Patients Follow-up 2, 0, 1
1995430 Gr1: 55 between April 1991 stroke onset SD: 13.75 group: Rehab at home group: No special were assessed at 1 year
Gr2: 55 and January 1992 to stroke unit by an OT. Interventions intervention by an OT at post-stroke
from 2 stroke units admission: Gr2: 75.8 based on the model or follow-up, 2, 8, 16, and
in South Glamorgan, Gr1: 11 days SD: 10 of human occupation. although they 24 wk after
irrespectively of SD: 21.5 days Included: teaching could receive any discharge.
discharge destination. Gr2: 10 days new skills, facilitating available services
SD: 12.75 days more independence as required. No
in ADL, facilitating contact with
return of function, researcher.
enabling patients to use
equipment supplied by
other agencies, giving
information to the patient
and caregiver, referring
to or liaising with other
agencies. This model was
in addition to any other
follow-up services and
community PT.

Drummond Total: 128 Hemiplegia: Gr1: 29.52 days Gr1: 58.95 Gr1: Leisure rehab: Gr3: Control Hospital and Gr1&2: Once Follow-ups 2, 0, 1
1995432 Gr1: 21 Gr1: 8 R; 12 L SD: 21.75 days SD: 13.11 The treatment program No additional social services a wk at 3 and 6
Gr2: 21 1 bilateral reflected personal input over months
Gr3: 23 Gr2: 7 R; 14 L Gr2: 28.38 days Gr2: 70.10 preferences and abilities. which they were
Gr3: 8 R; 15 L SD: 10.43 days SD: 6.69 30 min of treatment per receiving from
wk for first 3 months, hospital and social
Gr3: 25.44 days Gr3: 68.65 then 30 min of treatment services.
SD: 17.80 days SD: 9.95 a fortnight for 3 months.
Gr2: Standard customary
OT: OT activities such as
transfers, washing and
dressing practice, and
perceptual treatment.
In situations where
subjects were virtually
independent, the visits
were checkup visits and
subjects were questioned
about their progress and
any existing problems.
(Same time as time of Gr1
Appendix 3M

with therapist.)
233

continues
APPENDIX 3M. Continued
234

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Drummond Total: 315 Patients admitted with Within 5 wk of N/A Gr1: Stroke unit (no Gr2: Standard N/A N/A Follow-ups 1, 0, 1
1996a431 Gr1: 176 diagnosis of stroke acute stroke description of specific customary ward at 3 and 6
Gr2: 139 between January 1, treatment received). (no description of months
1991 and June 30, specific treatment
1993. received).

Drummond Total: 65 Patients who spoke N/A Gr1: 58.95 Gr1: Leisure rehab: Gr2: Received Hospital and 30 min a Follow-ups 1, 0, 1
1996b433 Gr1: 21 English, had no SD: 13.11 Advice and help falling standard social services session at at 3 and 6
Gr2: 21 severe comprehension into the following broad customary OT least once months
Gr3: 23 problems, no history Gr2: 70.1 categories: treatment, activities, such a wk for 3
of dementia, did not SD: 6.69 such as practice of as transfers and months after
need to be transferred transfers needed dressing practice. discharge,
for further medical Gr3: 68.65 for leisure pursuits; Control: then 30 min a
treatment, and had SD: 9.95 positioning; provision Gr3: No fortnight for 3
TOPICS IN STROKE REHABILITATION/SPRING 2006

an address in the of equipment; advice intervention. months

Nottingham District on obtaining financial
Health Authority. assistance and transport;
Hemiplegia: N/A and liaison with specialist
org. and providing
physical assistance, such
as referral to volunteer
agencies.
Duncan Total: 20 30–90 days after onset, N/A Gr1: 67.3 Gr1: Home-based Gr2: Usual care N/A Gr1: 3 N/A 2, 0, 1
1998103 Gr1: 10 minimal or moderately SD: 9.6 yr exercises program as prescribed by sessions/wk
Gr2: 10 impaired sensorimotor provided by PT. 10-min their physicians for 8 wk; 1.5
function, ambulatory Gr2: 67.8 warm-up (stretching and (6 patients had hr a session.
with supervision SD: 7.2 flexibility); 4 blocks: 1) home health visits, Patients
and/or assistive assistive and resistive 4 patients had instructed to
device, and living exercise using PNF outpatient PT). continue on
at home within 50 patterns to upper and Visit included their own for
miles of the University lower extremities or balance training, additional 4
of Kansas Medical theraband exercise to progressive wk.
Center. Exclusion: major muscle groups resistive exercises, Gr2:
Medical condition of upper and lower bimanual Patients had
that interfered with extremities; progression activities, approximately
outcome assessment when patient able to facilitative 39 visits,
or that limited complete 2 x 20 reps, exercises. approximately
participation in increased theraband 44 min a visit.
submaximal exercise resistance or increased
program; Mini-Mental PNF manual resistance;
State score <18; 2) 15 min of balance
receptive aphasia that exercise; 3) patients
interfered with the encouraged to use affected
ability to follow a 3- upper extremity.
step command.
Hemiplegia:
Gr1: 4 L; 6 R
Gr2: 4 L; 5 R
1 brain stem
Treated area: Upper/
lower extremity
Appendix 3M
235

continues
APPENDIX 3M. Continued
236

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Duncan Total: 92 Stroke within 30 Gr1: 77.5 Gr1: 68.5 Gr1: Exercise program: Gr2: Standard N/A 12–14 wk N/A 2, 1, 1
2003104 Gr1: 44 to 150 days; ability SD: 28.7 days SD: 9.0 ROM and flexibility for customary care:
Gr2: 48 to ambulate 25 feet shoulder, elbow, wrist, Home visits by
independently; mild- Gr2: 73.5 Gr2: 70.2 fingers, hip, ankle, and research staff every
to-moderate stroke SD: 27.1 days SD: 11.4 trunk; strengthening 2 wk for health
deficits defined by a resistive exercises: 2 sets education, vital
Fugl-Meyer score of of 10 repetitions; balance signs, and test of
27 to 90 for upper and training; upper extremity O2 saturation.
lower extremity; an functional use; and
Orpington Prognostic endurance training (up
Scale score of 2.0 to 30 min on bike): 36
to 5.2 and palpable sessions of 90 min.
wrist extension on
the involved side; and
TOPICS IN STROKE REHABILITATION/SPRING 2006

Folstein Mini-Mental
Status examination
score >16.
Hemiplegia:
Gr1: 18 R; 22 L
Gr2: 22 R; 22 L
Evans Total: 267 Moderately severe N/A Gr1: 74.6 Gr1&3: Gr2&4: In the same N/A Follow-ups 1, 0, 1
2002434 Gr1: 80 stroke was defined SD: 8.8 Care provided by a Care provided hospital, teams at 2 and 11
Gr2: 84 as persistent stroke physician and by general wards were comparable months
Gr3: 49 neurological deficit Gr2: 76.6 multidisciplinary team from admitting in seniority
Gr4: 54 affecting continence, SD: 7.2 with specialist experience physicians. The and specialist
mobility, or self-care in stroke management. generic nursing experience and
abilities and requiring Gr3: 75.4 Individualized rehab and therapy had access to
multidisciplinary SD: 12.6 plan with clearly defined staff on the similar resources
treatment. Confirmed goals based on joint ward provided in the hospital
diagnosis of stroke. Gr4: 73.4 assessments. rehab, but the and community.
Hemiplegia: SD: 13.5 specialist team Treatment after
N/A provided input for discharge was
assessments, goal provided by
setting, planning community
of treatment, services.
discharge
arrangement,
and liaison with
patients and
relatives.

Falconer Total: 121 Adults admitted to Gr1: 23.5 days Gr1: 68.6 Gr1: Critical Path Method Gr2: Control: N/A N/A N/A 1, 0, 1
1993435 Gr1: 53 Rehabilitation Institute SD: 22.3 days SD: 11.9 (CPM) (method for Functioned
Gr2: 68 of Chicago with visually charting the steps more like a
diagnosis of recent Gr2: 21.7 days Gr2: 67.6 that must be completed multidisciplinary
(within 120 days) SD: 15.5 days SD: 14.9 to carry out the project): team (discipline
stroke. Functioned more like an oriented).
Hemiplegia: N/A interdisciplinary team Usual care team
(team oriented). CPM members from
used to plan, monitor, each discipline
and control the patient’s individually
program and progress. evaluated
CPM designed to the patient.
facilitate interdisciplinary Semimonthly
communication and meeting to update
coordination. patient status.
Appendix 3M
237

continues
APPENDIX 3M. Continued
238

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Fang Total: 128 Stroke was defined 1 wk post Gr1: 65.49 Gr1: Gr2: N/A Gr1: 45 min/ Follow- 2, 0, 1
2003436 Gr1: 50 as an acute onset of stroke SD: 10.94 Therapy included Bobath Did not receive day, 5 days/ up at 6
Gr2: 78 neurological deficits techniques and passive professional wk for 4 wk months
lasting more than Gr2: 61.8 movement training of the or regular
24 hr or leading SD: 10.94 affected limb. Initiated physiotherapy
to death with no within the first wk after during the whole
apparent cause other stroke onset. hospitalization
than cerebrovascular period. No
disease. Confirmed specific cognitive
CVA diagnostic. or acupuncture
therapy was
administered.
Stroke-related
symptoms and
TOPICS IN STROKE REHABILITATION/SPRING 2006

complications
were treated with
the help of a
multidisciplinary
team (the same for
Gr1).

Feigenson Total: 667 Patients with stroke Gr1: 64.61 days Gr1: 67.14 All groups: Similar Gr2: Therapy N/A N/R Follow-up 0, 0, 1
1979437 Gr1: 589 admitted for rehab. SD: 77.65 days SD: 11.73 therapeutic programs. provided off the at 6 wk
Gr2: 78 Hemiplegia: Same standardized units in separate (end of
Gr1: 266 R; 262 L; 150 Gr2: 48.75 days Gr2: 69.03 protocols utilized therapeutic treatment)
both; 460 none SD: 47.81 days SD: 10.74 throughout the hospital facilities.
Gr2: 35R; 33L; 10 Acute for nursing, PT, OT, and
both; 9 none speech therapy. Most
of the staff had rotated
through the stroke unit.
Family teaching was
emphasized on all units.
Gr1: Stroke unit; therapy
on the unit.
Feldman Gr1: 42 Have a diagnosis of N/A Gr1: Range, Gr1: Rehab Gr2: Control: N/A N/A N/A 2, 0, 0
1961438 Gr2: 40 hemiparesis secondary 37–82 program prescribed Program of
to cardiovascular Gr2: Range, by a physician on the functionally
disease; be admitted 36–80 staff in the Department oriented medical
to hospital within of Physical Medicine care supervised
2 months of the and Rehabilitation. The by the personnel
onset of hemiparesis; programs were prescribed of the medical
demonstrate a individually and were and neurological
nontransient varied according to wards. Training in
neuromuscular deficit. patients’ needs and sitting, standing,
progress. The prime and balancing
limitation was the patient’s was usually
ability to participate. instituted early.
Training was given
in ambulation
and self-care
activities. Social
and vocational
services were
provided when
indicated through
the hospital
social service
or community
agencies.

Fjaertoft Gr1: 160 Patients from the Within 72 hr of N/A Gr1: Further follow-up Gr2: Both groups N/A Follow-up 2, 0, 0
2003439 Gr2: 160 city of Trondheim, admission and was organized by the Comprehensive received similar at 52 wk
Norway, admitted to less than 7 days primary health care follow-up stroke stroke unit care
the stroke unit with after the onset system. service organized during the acute
signs and symptoms of symptoms by a mobile team. phase with
of an acute stroke; focus on early
inclusion within 72 hr mobilization/
of admission and less rehab combined
than 7 days after the with a
onset of symptoms; standardized
SSS score between 2 acute medical
and 57 points; ability treatment
to live independently program.
before the onset
of stroke; and no
participation in other
Appendix 3M

trials.
239

continues
APPENDIX 3M. Continued
240

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Gibson Total: 50 Patients hospitalized N/A Gr1: 74.7 Gr1: Occupational Gr2: Control: N/A N/A Follow-up 0, 0, 1
1997440 Gr1: 25 in acute care after Acute SD: 7.25 adaptation group: ROM exercises at 3 wk
Gr2: 25 onset of CVA, were at Received combination of and facilitation (end of
least age 60 yr, were Gr2: 73.5 occupational readiness for the affected treatment)
seen for OT treatment SD: 5.75 (ROM, bedside ADL) and upper extremity,
starting in acute care occupational activities that cognitive activities,
and followed through stimulate occupational and bedside ADL.
rehab at the same environment/role (OE/R) When transferred
facility, and received tasks (gardening, domino to rehab unit,
the customary amount playing, cooking). subjects received
of OT intervention in kitchen activity,
acute care; ability to a community
tolerate minimum of outing, and ADL
3 hr of therapy per focused treatment.
TOPICS IN STROKE REHABILITATION/SPRING 2006

day, follow directions,

and potential for
improvement in
function.
Hemiplegia:
24/26
Gr1: 11 R; 14 L
Gr2: 15 R; 10 L
Gilbertson Total: 138 Clinical diagnosis Gr1: 31 days Gr1: 71 Gr1: Intervention service: Gr2: Control N/A Gr1: 10 visits Follow- 2, 0, 1
2000441 Gr1: 67 of stroke; admitted SD: 10 days SD: 30.5 Designed to be client- group: Routine within 6 wk ups at 8
Gr2: 60 to a Glasgow royal centered; developed service (routine Gr2: N/A wk and 6
infirmary NHS trust; Gr2: 23 days Gr2: 71 through focus group follow-up). months
had been to the OT SD: 13.25 days SD: 58 sessions with patient, Inpatient from
department; and caregivers and local OT multidisciplinary baseline
discharge date had staff; 6 wk domiciliary rehab pre-
been set. program (approximately discharge home
Hemiplegia: 10 visits lasting 30–45 visit for selected
Gr1: 34 R; 33 L min); tailored to recovery patients; provision
Gr2: 33 R; 38 L goals identified by the of support services
patient such as regaining and equipment,
self-care or domestic regular
or leisure activities; OT multidisciplinary
worked with patient to review at a stroke
achieve these goals and clinic, and selected
also liaised with other patients referred to
agencies for advice, a medical hospital.
services, and equipment.

Gürsel Total: 43 Patients with no N/A M/F: Gr1: Rehab program for a Gr2: Patients N/A Gr1: 14 wk Follow- 0, 0, 0
1998442 Gr1: 21 previous history of Gr1: 12/9 mean time of 14 wk. The were informed in rehab up at 6
Gr2: 22 completed stroke and Gr2: 14/8 same standard customary about the disease program months
with relatives who Age: therapy techniques were and instructed to (end of
agreed to take the Gr1: 63 applied to all patients. exercise according treatment)
patient to the clinic (11) to program at
for a 2nd evaluation 6 Gr2: 59 home.
months later. Gr1 was (13)
formed by patients
admitted to the rehab
unit in the early phase.
The other patients who
for socioeconomic
reasons or a lack of
rehab beds received no
rehab were recruited in
the 2nd group.
Side of hemiplegia:
Gr1: 13 R; 8 L
Gr2: 11 R; 11 L
Appendix 3M
241

continues
APPENDIX 3M. Continued
242

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Hamrin Total: 59 N/A N/A Gr1: 73 Gr1: Activation program Gr2: Control: N/A Gr1: For 4 wk Follow- 2, 0, 1
1982443 Gr1: 33 SD: 10 in nursing. The following General medical up at 11
Gr2: 26 were emphasized: ward without months
Gr2: 72 psychological activation, activation program
SD: 11 verbal communication, in nursing.
prophylaxis against
contracture, prophylaxis
against pressure sores,
movement therapy,
ambulation, personal
hygiene, dressing, feeding,
bladder elimination, and
bowel elimination.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Hayes Total: 30 All participants had N/A Gr1: 70 Gr1: Received an early Gr2: Received All evaluated N/A Follow-up 1, 0, 0
1986444 Gr1: 15 diagnoses of acute Acute SD: 12 intervention of rehab care standard and treated by at end of
Gr2: 15 CVA upon admission within 72 hr of hospital rehabilitative same physical treatment
to a large metropolitan Gr2: 70 admission. treatment between therapist.
teaching hospital. SD: 9 4–15 days after
Hemiplegia: admission.
Gr1: 5 R; 10 L
Gr2: 7 R; 6 L
Holmqvist Total: 81 Acute stoke, 5 to 7 days Gr1: 70.8 Gr1: Rehab at home. Gr2: Routine Managed in the N/A Follow- 2, 0, 1
1998445 Gr1: 41 independence SD: 7.6 Team: 2 PTs, 2 OTs, one rehab: All patients wards before up at 3
Gr2: 40 in feeding and speech therapist, and in this group were randomization months
continence, Mini- Gr2: 72.6 a social worker on a also admitted to Total duration: 3 (end of
Mental State SD: 8.9 consulting basis. In each the Department to 4 months treatment)
Examination score of case, the case manager (a of Neurology.
>23, impaired motor therapist) was responsible Heterogeneous set
capacity and/or, for coordination of of interventions
dysphasia. discharge procedure, ranging from the
Hemiplegia: at-home therapy, liaison best established
Gr1: 15 R; 24 L between therapists, and in the hospital,
Gr2: 23 R; 14 L contact with neurologist day care, and/or
responsible. The program outpatient care to
emphasized a task- others introduced
and context-oriented during the study
approach, which implies period; for
that the patient performs example, daily
guided, supervised, or afferent sensory
self-directed activities. stimulation by
low-frequency
TENS and home-
based rehab.
Appendix 3M
243

continues
APPENDIX 3M. Continued
244

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Indredavik Total: 320 Signs and symptoms Acute onset Gr1: 74.0 Gr1: Extended Stroke Gr2: Ordinary N/A N/A Follow-ups 1, 1, 1
2000446 Gr1: 160 of acute stroke Gr2: 73.8 Unit Service (ESUS). Stroke Unit at 6 and 26
Gr2: 160 according to the WHO Stroke unit treatment Service (OSUS). wk
definition of stroke; similar to OSUS Stroke unit
SSS score between 2 combined with service treatment
and 57 points; living from a mobile team that – according to
at home before stroke; offers early supported evidence-based
included within 72 hr discharge and coordinates recommendations
after admission to the further rehab and follow- – combined with
stroke unit and within up in close cooperation further inpatient
7 days after the onset with the primary health rehab when more
of symptoms; lack of care system. Team: long-term rehab
participation in other nurse, OT, PT, and a is necessary, and a
trials; provision of physician offering part- follow-up program
TOPICS IN STROKE REHABILITATION/SPRING 2006

informed consent. time services. The team organized by the

tried to establish a service primary health
and support system that care system after
allowed the patient to discharge. The
live at home as soon as service includes:
possible after the stroke systematic
and to continue necessary diagnostic
training and rehab at evaluation;
home, in a day clinic, standardized
or by a combination of observation of
these 2 alternatives. For vital signs and
some patients with more neurological
extensive needs, the deficits; an acute
team also offered training medical treatment
and support at home in program; very
addition to service from early mobilization
other agencies. and rehab in a
stroke unit.
Jongbloed Total: 40 Patients had incurred N/A 69.6 Gr1: Experimental Gr2: Control: N/A Five 1-hr Follow-ups 1, 0, 0
1991447 Gr1: 20 a stroke within the Acute SD: 11 leisure activities including Subjects were visits at home at 5 wk
Gr2: 20 past 15 months, had individual or social visited by an over a 5-wk (end of
completed a rehab Gr1: 68.2 activities carried out in OT and asked consecutive treatment)
program, were not the home or community questions about period and 13 wk
on antidepressant Gr2: 69.6 environment. leisure activity
medications, were not involvement
severely aphasic, and throughout their
had a close relative or life span. The
friend who was willing therapist also
to participate in the asked each person
study. about the effects of
the stroke on his
Hemiplegia: or her life.
Gr1: 12 R; 8 L
Gr2: 12 R; 8 L

Kalra Total: 71 Inclusion: Stroke N/A Gr1: 77.2 Gr1: Individualized rehab Gr2: Treatment None N/A Follow-up 1, 1, 1
1995448 Gr1: 34 patients with SD: 8 yr program reviewed on according to at 3 yr
Gr2: 37 Orpington Prognostic a daily basis; focused existing general
Score >5 and referred Gr2: 80.4 on problems affecting ward practices (no
for rehab on a stroke SD: 12.1 yr ADLs and mobility in details available).
unit. the context of home
Hemiplegia: environment and support.
Gr1: 18 R; 14 L; 2
bilateral
Gr2: 19 R; 14 L; 4
bilateral
Appendix 3M
245

continues
APPENDIX 3M. Continued
246

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Kalra Total: 447 Inclusion: Patients <72 hr Gr1: 75 Gr1: Stroke unit Gr3: Domiciliary None Tx for 9 months 2, 0, 1
2000449 Gr1: 148 with moderately severe SD: 6 care provided by a stroke care 3months
Gr2: 150 stroke (persistent physician supported by patients were
Gr3: 149 neurological deficit Gr2: 77.3 a multidisciplinary team managed in their
affecting continence, SD: 6 with specialist experience own home by a
mobility, and in stroke management. specialist team
ability to look after Gr3: 77.7 Routine management consisting of a
themselves, requiring SD: 8 involved joint assessments doctor, nurse, PT,
multidisciplinary and goal setting, OT, and speech
treatment) who could coordinated treatment, and language
be supported at home and planned discharges. therapists, with
with nursing, therapy, Gr2: Stroke team support from a
and social services. patients were managed district nursing
Hemiplegia: on general wards and and social services
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr1: 69 R; 76 L remained under the care for nursing and

Gr2: 76 R; 67 L of admitting physicians. personal care
Gr3: 70 R; 76 L All patients were seen needs. Patients
by a specialist team were under the
with experience in joint care of the
stroke management. stroke team. Each
The team undertook patient had an
stroke assessments and individualized
collaborated with ward- care plan outlining
based nursing and therapy activities and
staff in goal setting, the objectives of
planning of treatment, treatment, which
discharge, arrangements, was reviewed
and liaison with patients at weekly
and relatives. Day-to-day multidisciplinary
treatment was provided meetings.
by staff on the ward.
Krespi Total: 704 Inclusion: Patients N/A Gr1: 62.9 Gr1: All hospitalized Gr2: Nurse/patient N/A N/A N/A 0, 0, 0
2003450 Gr1: 352 admitted between SD: 14.4 acute stroke patients were ratio 1:12 (day),
Gr2: 352 January 1997 and followed for at least 48 hr 1:24 (night). The
August 2000 to the Gr2: 61.23 in the Stroke Intermediate main aspects of
Istanbul Medical SD: 15.47 Care Unit: nurse/patient care (nursing,
Faculty Neurology ratio 1:4 (day). Stabilized physio) were not
Department. patients were transferred different from
in subacute care-bed: the rest of the
Hemiplegia: nurse/patient ratio 1: patients having
N/A 4 (day), 1:8 (summer other neurological
vacation), 1:12 (night). diseases. 2
One stroke fellow residents took
patient and 2 residents part in the
took part in the daily management.
management. Nurses Nurses were not
were dedicated to stroke dedicated to stroke
patients. 3 PTs evaluated patients. PTs were
patients on admission consulted when
and followed up on them needed.
until discharge. Follow-
up on patient weekly (by
multidisciplinary staff).

Lin 2004451 Total:19 Stroke onset greater Gr1: Gr1: 61.4 Home-based physical Gr2: Same as Gr1, N/A Once a wk for Follow-ups 1, 0, 1
Gr1: 9 than 1 yr; severe to 44.0 months SD: 11.2 therapy administered but 10 wk later 10 wk at 11 wk
Gr2: 10 moderate residual SD: by PTs, consisting of: (control group for and 22 wk
disability with Barthel 29.6 months Gr2: 62.8 50–60 min/session, the first 10 wk).
Index score 5–14; not SD: 9.4 motor facilitation,
involved in any kind Gr2: postural control training,
of rehab program in 62.8 months functional ambulation
the past 6 months; SD: 9.4 months training with gait
ability to follow verbal correction, ADL training,
instructions; and and daily exercise
living in the Nan-Tou programs tailor-made for
County area during the patients’ needs.
research.

Hemiplegia:
Gr1: 3 L; 6 R
Gr2: 2 L; 8 R
Appendix 3M
247

continues
APPENDIX 3M. Continued
248

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Lincoln Total: 76 No other medical N/A Gr1: 73.5 Standard customary Standard N/A N/A 6–16 1, 0, 1
1996452 Gr1: 39 problems requiring Acute SD: 7.5 stroke unit treatment. customary ward wk; no
Gr2: 37 continued treatment treatment. subsequent
on an acute medical Gr2: 65.5 follow-up
or geriatric ward; no SD: 10 done.
planned discharge date
within 2 wk; able to
tolerate rehab (in .5
hr sessions, 2 hr/day);
able to drink, eat, and
wash own face; able
to transfer from bed to
chair with help from
no more than 2 nurses;
TOPICS IN STROKE REHABILITATION/SPRING 2006

and independent for

toileting before stroke.
Lincoln Total: 282 Patients had no NA Median Gr1: Qualified- Gr2: Assistant- Gr1& 2: Gr1: Senior Follow-ups 2, 0, 1
1999453 Gr1: 95 planned discharge age: physiotherapist (QPT) physiotherapist Patients received research PT at 8 wk
Gr2: 94 date from the hospital Gr1: 73 group: The patients (APT) group: standard PT gave specific and 20 wk
Gr3: 93 within the next 7 (IQR: received standard PT in Patients were under the sessions for
days; were able to 65–81) addition to being treated treated by a supervision of a approximately
feed themselves and Gr2: 73 by a senior research PT. trained PT senior research 2 hr/wk.
wash their own faces (IQR: This additional treatment assistant. assistant. Gr2: Therapy
before the stroke; did 6–80) consisted of facilitation, Gr3: Routine sessions for
not have premorbid Gr3: 73 specific neuromuscular physiotherapy approximately
dementia; were (IQR: techniques, and functional (RTP) group 2 hr/wk by a
able to speak or to 64–80) rehab, all broadly based (control): Patients PT assistant.
understand English on the Bobath approach. received standard Gr3:
before the stroke; Motor and functional PT as is given PTsessions
had no premorbid tasks had to be practiced at the City each
severe impairment between therapy sessions. hospital. This weekday for
of the affected upper Ongoing assessments and therapy follows a approximately
limb, but had some specialized advice were Bobath approach 30–45 min.
impairment of arm given at each treatment. predominantly,
function; and were and most patients
reported by the received treatment.
physiotherapists as No additional
being able to tolerate treatment by the
a half-hour session of research PT.
PT daily.

Logan Total: 111 First-time stroke N/A Gr1: 71 Gr1: Enhanced OT Gr2: Standard N/A Varied Follow- 2, 0, 1
1997454 Gr1: 53 patients discharged Acute SD: 10.2 service: Patients were customary OT according to ups at 3
Gr2: 58 from hospital and seen and treated by a service: All each patient months
referred to the city Gr2: 74 single research OT. Access patients were (end of
of Nottingham’s and SD: 11.5 to aids and budgets for prioritized by the treatment)
the Boroughs of adaptations; provided senior OT. All and 3
Gelding and Beeston’s more frequent visits to urgent cases were months
social services OT patients than control. seen immediately;
departments. Diagnosis others were placed
checked with on a waiting list.
Nottingham stroke
register.
Hemiplegia:
Gr1: 29 R; 24 L
Gr2: 32 R; 26 L
Appendix 3M
249

continues
APPENDIX 3M. Continued
250

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Mayo Total: 114 Persons who had N/A Gr1: 70.3 Gr1: Prompt discharge Gr2: Regular N/A Variable per Follow-ups 2, 1, 1
2000455 Gr1: 58 persistent motor SD: 12.7 from hospital with discharge service individual at 1 month
Gr2: 56 deficits after stroke and immediate provision of (PT,OT,ST) as (end of
who had caregivers Gr2: 69.6 follow-up services by requested by treatment)
willing and able SD: 12.7 multidisciplinary team patient’s care and 3
to provide live-in (nursing, OT, PT, ST, provider and months
care for the subjects dietician). Individualized offered through
over a 4-wk period treatment. extended acute-
after discharge from care hospital
hospital. stay; inpatient or
Hemiplegia: outpatient rehab;
N/A and home care via
local community
health clinics, but
TOPICS IN STROKE REHABILITATION/SPRING 2006

private care was

also possible.

Nir Total: 155 Stroke survivors after 13 days Gr1: 72.3 Gr1: Concurrent Gr Control: Concurrent 1x/wk, 3 months 2, 0, 1
2004456 Gr1: 73 an average of 13 days SD: 6.8 treatment and a Concurrent treatment: 12 wk
Gr2: 82 in an acute care setting structured written nursing treatment only. Standard
Hemiplegia: Gr2: 73.8 intervention program. inpatient
Gr1: 35 R; 37 L SD: 7.6 Nursing students met rehab program
Gr2: 37 R; 45 L with the patients 1x/wk consisting of
for 12 consecutively half an hour of
weekly session of 1–2 hr PT and OT 5
each. The intervention days/wk.
was based on a written
guidebook based on
Orem’s model of self-
care. It emphasized the
broad care approach
that believes in the
ability of a person to be
active, responsible for
his or her condition, and
responsible for restoring
his own well-being.
Parry Total: 282 Patients had to be able 12 days N/A Gr1: Qualified Gr3: Routine Gr1&2: Patients Gr1&2: 10 Follow- 1, 0, 1
1999457 to participate in at least physiotherapy (QPT) physiotherapy received hr additional up at 5
Gr1: 94 30 min of routine PT group: Patients received (RPT) group: No additional PT. treatment wk post
Gr2: 93 per day and to have additional treatment additional therapy over a 5-wk treatment
Gr3: 95 motor impairment of a from a qualified PT. The was given. period (6 months)
previously functional additional PT aimed to
arm. reflect current British
Exclusion: N/A practice. Principles of
Hemiplegic: N/A treatment followed those
of the Bobath approach,
with some influence of
movement science-based
approaches.
Gr2: Assistant
physiotherapy (APG)
group: Patients received
additional treatment
from a trained supervised
assistant trained with the
help of a manual.
Appendix 3M
251

continues
APPENDIX 3M. Continued
252

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Partridge Total: 114 All patients with a N/A 76.5 Gr1: Standard amount Gr2: Standard Aims of Gr1: 60 min/ Follow- 2, 0, 0
2000458 Gr1:54 diagnosis of stroke, SD: 8.5 of therapy, plus higher amount of treatment with day ups at 6
Gr2: 60 according to WHO intensity of rehab; therapy; the Bobath Gr2: 30 min/ wk and 6
criteria, admitted over Duration: 60 min/day. Duration: 30 concept: mid- day months
a 2-year period to the min/day. line alignment, (end of
Canterbury Stroke postural treatment)
Unit. adjustment,
Hemiplegia: and symmetry
53 R; 60 L and control.
Maintenance
of full range
of movement.
Working to
improve base
TOPICS IN STROKE REHABILITATION/SPRING 2006

of support;
facilitating
all aspects
of transfers;
modifying
muscle tone;
sequencing of
movements;
balance re-
education;
and stability
and transfer
of weight.
Inhibitions;
of positive
support reaction;
Working on
shoulder
girdle and
scapulohumeral
rhythm.
Facilitating
functional hand/
arm movement.
All aspects of
reeducation.
Rudd N: 231 If patients lived alone, N/A
1997461 Gr1: 167 they needed to be able SD: 11
Gr2: 164 to perform functional
independent transfer, Gr2: 72
and if they lived with a SD: 12
willing caregiver they
needed to be able to
perform transfer with
assistance.
Gr1: 70 Gr1: Community therapy Gr2: Standard N/A Gr1 : Follow- 2, 0, 1
team. The patients customary care. Maximum ups at 3
were assessed for rehab Patients allocated duration of 3 months
needs before discharge to continue with months. (end of
in conjunction with the their treatment, Gr2: treatment)
hospital-based therapists discharge Maximum and 9
to set initial objectives planning, and duration of months
and to ensure continuity outpatient care in three 1-hr
of care. After discharge, the normal way. visits daily.
patients were given About half of
a planned course of the patients who
domiciliary PT, OT, and were admitted
speech therapy, with received treatment
visits as frequently as in general medical
considered appropriate or elderly care
(maximum 1 daily visit wards. Outpatient
from each therapist). Each resources available
patient had an individual to them included
care plan, which was a hospital-based
reviewed at a weekly team stoke clinic,
meeting. On discharge geriatric day
patients were referred hospital, generic
to standard customary domiciliary PT
services resources. The and speech and
community therapy team language therapy,
comprised a senior PT hospital outpatient
grade 1 with neurological PT and the usual
training, a senior OT community
grade 1, a halftime ST resources, a home
with adult neurological help for personal
training, and a full-time care, meals on
therapy aide. A consultant wheels, and
physician coordinated community nurse
the team and chaired the visits for specific
weekly clinical meeting. tasks.
Appendix 3M
253
continues
APPENDIX 3M. Continued
254

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Ruff Total: 113 N/A N/A Gr1: 78.98 Gr2: 7 days/wk Gr1: Standard N/A Gr1: 6 x/wk None 1, 0, 0
1999462 Gr1: 57 Gr2: 70.96 treatment program. customary 6- Gr2: 7 x/wk
Gr2: 56 Received an additional day treatment
7th day of rehab regiment. The
treatments, comprised of 7th day was
occupational, physical, open for rest and
and speech therapy. visitation, with the
patients receiving
necessary nursing
and medical
attention.

Rodgers Total: 92 Patient had to have Within 48 hr of Median age Gr1: Early supported Gr2: Standard N/A Gr1: The Follow- 2, 0, 1
1997460 a home address in admission Gr1: 73 discharge group: Stroke care group: discharge up at 3
TOPICS IN STROKE REHABILITATION/SPRING 2006

Gr1: 46 Newcastle; could not Range, discharge team consisted Inpatient and rehab service months
Gr2: 46 be living in residential 47–93 of OT, PT, SW, ST, and OT outpatient care was available post stroke
or nursing home care technician. The team was was provided by 5 days/wk,
prior to the incident Gr2: 73 based in the community standard hospital but home care
stroke; could not be Range, to establish a relationship and community component
severely handicapped 44–91 with referred stroke services. Discharge of the service
prior to the incident patients and their families planning and was available
stroke (Oxford early in hospital stay. services post all day,
Handicap Scale 0–3); Before each discharge, discharge arranged every day if
had no other condition a member of the team and provided required. No
likely to preclude would do a home visit according to the time limit
rehab; had to be without the patient and usual practice of involved for
medically stable with the day of the discharge each participating the duration
a Barthel Activities a team member would ward or unit. of the home
of Daily Living Index bring the patient home. Community care program.
between 5–19 at 72 hr support was
post stroke. provided by the
Hemiplegic: primary care team,
N/A and community
rehab services,
outpatient
services, and
social services as
were provided as
appropriate.
Rodgers 626 Stroke in the previous Gr1: 5 days Median age Gr1: Enhanced upper Gr2: Control Gr1&2: Gr1: Follow-ups 2, 0, 1
2003459 10 day resulting (2.5) median Gr1: 74 limb therapy time: 30 min Interdisciplinary Enhanced at 6 wk
Gr1: 62 in upper extremity Gr2: 75 of rehab from PT and OT treatment upper and 18 wk
Gr2: 61 impairment due to Gr2: 5 days 5 days/wk for 6 wk. program: extremity
one or more of the (2) median Current stroke therapy for 30
following: weakness, unit PT (Bobath min, 5 days/
sensory loss, ataxia, based) and wk for 6 wk
or visual impairment. OT practice Gr2: N/A
Exclusion: not combining
medically stable a normal
within 10 days; movement
persistent impaired approach within
conscious level in meaningful
days 0–10 (Glasgow activity and task
coma scale); significant analysis.
communication
difficulties; cognitive
problems; and severe
handicap prior to
stroke.
Hemiplegia:
Gr1: 28 R; 34 L
Gr2: 26 R; 35 L

Sivenius Total: 95 N/A 1 wk Gr1: 71.5 Gr1: Intensive treatment Gr2: Normal The patients Gr1: 30 min/ Follow-up 2, 0, 0
1985463 Gr1: 50 SD:10.5 (IT): One department treatment (NT): received the session, twice at 1 wk
Gr2: 45 was redesigned as a rehab The patients were normal physical a day when in (end of
Gr2: 70.1 unit with the especial discharged from therapy in the medical treatment)
SD: 9.1 purpose of treating stroke these departments the standard ward of the
patients. The remaining to their homes customary University
patients were treated in or, if it was not medical wards, Hospital, 4
neurological wards of the possible, to old the duration and sessions/wk,
University Hospital. The age homes or amount of which 15 days/
principle of the treatment chronic care was determined month.
was that PT should be departments by the internists. Gr2: 3
given as long as functional of community sessions/wk,
recovery was taking hospitals where 15 days/
place or the patient could some of them were month.
perform independently able to obtain
at home. PT. The guiding
principle was,
however, that no
Appendix 3M

patient’s condition
worsened as a
result of study.
255

continues
APPENDIX 3M. Continued
256

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Smith Total: 133 Inclusion: Confirmed Gr1: 35 days Gr1: 63 Gr1&2: PT and OT Gr3: Control: No None Gr1: 4 full Follow- 1, 0, 1
1981464 Gr1: 46 diagnosis of stroke. Gr2: 41 days Gr2: 66 treatment in groups and routine rehab; days/wk up at 3
Gr2: 43 Patients should be Gr3: 37 days Gr3: 65 individually. regularly visited at for up to 6 months
Gr3: 44 able to manage the home by a health months. (end of
most intensive of the 3 visitor; referred Gr2: treatment)
treatment regimens. back to hospital 3 times per
Exclusion: Patients or other services half day
making a full recovery if necessary; and for up to 6
while in hospital. encouraged to months.
Patients too old or continue with Gr3: N/A
too frail for intensive exercises taught
rehab. Patients living while in hospital.
outside the district.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Smith Total: 307 Patients who were < 60 days 72.5 Gr1: Stroke unit, Gr2: General N/A N/A N/A 1, 0, 0
1982465 Gr1: 78 conscious and had independent patients; ward, independent
Gr2: 47 an established or duration: 34.8 hr of PT, patients;
Gr3: 49 developing hemiplegia 56.8 hr of OT; duration: 41.1
Gr4: 60 at the time of total number of treatment hr of PT, 71.2 hr
assessment. Patients sessions: 37.6 days of PT, of OT;
were classified as 36.4 days of OT. total number of
independent if they Gr3: Stroke unit, treatment sessions:
could carry out the dependent patients; 35.2 days of PT,
following 7 activities duration: 67.2 hr of PT, 39.3 days of OT.
without human 93.9 hr of OT; Gr4: General
assistance: getting total number of treatment ward, dependent
in and out of bed; sessions: 82.7 days of PT, patients;
dressing; indoor 70.7 days of OT. duration: 127.6
mobility; toileting hr of PT, 125.8 hr
and personal hygiene; of OT;
cooking a simple hot total number of
meal; feeding oneself; treatment sessions:
and controlling 128.1 days of PT,
heat, light and 98.1 days of OT.
communication.
Sonada Total: 106 Stroke patients; Gr1: 49.8 days Gr1: 62.8 FIT program: 40 min of Standard N/A Gr1: N/A 0, 0, 0
2004466 Gr1: 58 days since onset to Gr2: 54.1 days Gr2: 64.7 PT and 40 min of OT customary rehab: 7 days/wk
Gr2: 48 admission ranged from similar to conventional 40 min of PT, 40 Gr2:
30–80; had a motor rehab, 7 days/wk; and min of OT, 5 days/ 5 days/wk
subscore of the FIM speech therapy 5 days/wk. wk; gait exercises
less than 80; and had In addition, patients were related to ADLs;
a cognitive subscore encouraged to stay out orthoses used if
of the FIM greater of sleeping area during necessary; passive
than 25. daytime. Patients freely ROM exercises
ambulated in the corridor; for affected side
spoke and interacted and muscle
with each other; and self- strengthening
initiated exercises and exercises for
self-care activities. unaffected side;
and speech
therapy 5 days/wk.

Sulch Total: 152 Patients transferred < 2 wk Gr1: 75 Gr1: Integrated care Gr2: N/A N/A Follow- 1, 0, 0
2002467 Gr1: 76 to a stroke rehab unit SD: 11 pathways: Therapeutic Multidisciplinary up at 6
Gr2: 76 within 2 wk of the activities were grouped team care: Patients months
acute event. Gr2: 74 according to stage and were assessed
SD: 10 predicted patients’ for individual
needs. Key goal for each needs and a
therapeutic intervention customized rehab
and the time estimated program was
to achieve these were designed under
defined in advance. the supervision
The integrated care of a consultant.
pathway covered all Therapeutic
aspects of inpatient activities, goals,
rehab from admission and the time
to discharge. These taken to achieve
goals were discussed in these goals
weekly multidisciplinary were discussed
meetings and changed in weekly
on the basis of patients’ multidisciplinary
progress. meetings and
changed on the
basis of patients’
progress.
Appendix 3M
257

continues
APPENDIX 3M. Continued
258

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Sunderland Total: 137 Clinical diagnosis of Gr1: 8 days Gr1: 65 yr Gr1&3: Enhanced Gr2&4: Standard N/A Follow-up 2, 0, 1
1992468 Gr1: 36 unilateral acute stroke, SD: 8.25 days SD: 14 yr therapy. More intensive Conventional customary OT. at end of 6
Gr2: 35 supported by CT treatment for the arm. therapy. months
Gr3: 29 scan when available. Gr2: 10 days Gr2: 68 yr Used behavioral methods Based on the
Gr4: 32 Inability to complete SD: 7.25 days SD: 8 yr to encourage the patient neurophysiological
the Nine Hole Peg and family to be active techniques
test (NHPT) within Gr3: 9 days Gr3: 67 yr participants in arm described in the
18 s when using the SD: 7.5 days SD: 11.5 yr rehab and to help the text by Bobath
affected hand despite patient avoid becoming a and Johnstone.
visual and verbal Gr4: 8 days Gr4: 70 yr passive recipient of expert This text describes
prompts to try to SD: 7.25 days SD: therapy. Specific aims the major
overcome any visual 12.25 yr were to promote greater techniques used.
neglect. Being well and adherence to self-directed The emphasis is
alert enough to receive exercise programs, to on expert hands-
TOPICS IN STROKE REHABILITATION/SPRING 2006

active therapy. Patients combat overprotectiveness on treatment

who had suffered from spouses, to prevent by the therapist
previous strokes were learned non-use of the and patients are
only included if it was affected UE, and to not routinely
clear that they had full facilitate learning of new instructed to
use of the arm before motor skills. An eclectic exercise between
the present episode. approach was taken in therapy sessions.
Hemiplegia: selection of treatment Active movement
Gr1: 22 R; 14 L techniques, which may not be
Gr2: 19 R; 16 L included Bobath exercises, encouraged until
Gr3: 15 R; 14 L EMG biofeedback, abnormal muscle
Gr4: 17 R; 15 L microcomputer games, tone is well
and goal-setting. controlled.
Emphasis was placed
on giving the patient
tasks of graded difficulty
and providing objective
feedback on performance.
Patients received more
than double the amount
of therapy for the UE.
They also received more
encouragement to practice
between sessions.
Teng Total: 114 Persons with persistent N/A N/A Gr1: Home group: 4- Gr2: Standard N/A N/A Follow- 1, 0, 1
2003469 Gr1: 58 motor deficits after wk tailor-made home customary care: up at 3
Gr2: 56 stroke who had program of rehab and Current practice months
caregivers willing and nursing services. for discharge
able to provide live-in planning and
care for the subject referral for follow-
over 4-wk period up services (such
after discharge from as extended
hospital. acute-care hospital
Hemiplegia: N/A stay, inpatient
rehab, outpatient
care, private care,
and home care
provided by local
community health
service center).

Turton Total: 22 Participants who Gr1: 24 days Gr1: 59 Gr1: Home therapy Gr2: Control: Both groups N/A N/A 0, 0, 1
1990470 Gr1: 10 achieved less than SD: 25.8 days SD: 11.97 2–3 x/day. The patients Patients were followed regimes
Gr2: 12 95% of normal were provided with a visited at home for recommended
performance on a task Gr2: 16 days Gr2: 58 booklet and a program assessment only. by the hospital
requiring a repeated SD: 6.1 days SD: 6.86 of exercises, which Both groups rehab staff.
reaching action (a were considered to followed regimes
peg transfer test). be appropriate to recommended by
Those with sensory their individual stage the hospital rehab
or proprioceptive of recovery and to staff.
problems in addition the problems they
to motor deficits, experienced. The patients
apraxia, and were asked to practice
perceptual or cognitive their exercises at the
impairments had to agreed times until the
be able to understand next visit, when the home
instructions. therapy program could
Hemiplegia: be reviewed following
Gr1: 7R; 3L reassessment.
Gr2: 6R; 6L
Appendix 3M
259

continues
APPENDIX 3M. Continued
260

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Walker Total: 185 Hemiplegia: < 1 month Gr1: Gr1: Standard OT: The Gr2: No OT: N/A N/A Follow-up 2, 0, 1
1999471 Gr1: 84 Gr1: 39 R; 45 L 73.6 yr aim of therapy was Patients received at 1 month
Gr2: 79 Gr2: 41 R; 40 L SD: 8.1 yr independence in personal no additional
and instrumental ADLs. input from the
Gr2: Personal care included research therapist,
75.1 yr activities such as bathing, but may have
SD: 8.6 yr feeding, dressing, received input
and stair mobility. from existing
Instrumental ADLs services, as would
included activities such as occur in routine
outdoor mobility, driving practice.
a car, traveling by public
transport, and household
chores. The focus of
TOPICS IN STROKE REHABILITATION/SPRING 2006

therapy was active

intervention rather than
assessment or liaison.

Wall Total: 20 Inclusion: Subjects 18 months to 45–70 yr A series of 10 exercises Gr3: PT and N/A Each exercise Before 1, 0, 1
1987472 Gr1: 5 demonstrated a 10 yr to promote an efficient home-supervised: lasted 5 min: treatment,
Gr2: 5 reduced support phase gait pattern and weight Patients attended 1.75 min at 1-month
Gr3: 5 time on the affected transfer was designed PT for 1 hr performing intervals
Gr4: 5 lower extremity; hierarchically in terms of once a wk to 10 repetitions during
all were capable complexity. Gr1: Patients undertake the and a treatment,
of walking with or had PT for 1 hr twice a exercises under 45-s rest. A and for 3
without a cane. wk. The exercises were the supervision of total of 10 months
Treated area: Lower taught by the PT who a PT. The patients exercises were after
extremity provided feedback and also performed performed treatment
corrected the patient. the exercises once in 1 hr with
Gr2: Patients performed a wk in their a 5-min rest
the same exercises as Gr1, home under the interval after
at home with supervision supervision of 15 min of
from their spouse. PT their spouse. exercises.
monitored the home Gr4: Control: No
exercise program to exercise treatment
provide the spouse with was given.
clarification on matters
arising.
Werner Total: 54 Between 6 months Gr1: 2.9 wk Gr1: 59 Gr1: Interventions Gr2: No outpatient N/A Session/wk: Follow- 2, 0, 1
1996279 Gr1: 33 and 5 yr after hospital SD: 1.8 wk SD: 19 oriented toward therapy. 2 hr of OT/PT ups at 3
Gr2: 21 discharge following an functional tasks such 4 x/wk a for months
acute middle cerebral Gr2: 3.3 wk Gr2: 66 as transfers, walking, 12 wk (end of
artery vascular event SD: 1.9 wk SD: 13 self-care, and feeding by treatment)
(CT scan/clinical Acute using modalities such as: and 3
presentation); strengthening, stretching, months
functional mobilization, muscle post
comprehension; no retraining/facilitation; treatment
medically significant given by trained OT and
comorbidities that PT.
would compromise
functional status
beyond the effects
of the stroke; no
generalized sensory
deficit; no history of
recurrent stroke; no
ongoing therapeutic
rehab intensity in past
4 months; or ADL
limitation.
Hemiplegia: N/A
Appendix 3M
261

continues
APPENDIX 3M. Continued
262

Age, yr Session
Author Sample Time since (mean, SD Comparison Concurrent frequency Follow-up Quality
and year size Population details onset for control) Treatment group therapy and duration duration (R, B, W)

Wikander Total: 34 Incontinent stroke Gr1: 12.7 days Gr1: 74.0 Gr2: Rehab governed by Gr1: Standard N/A N/A N/A 1, 0, 1
1998473 Gr1: 13 patients SD: 1.1 days SD: 1.5 FIM. customary
Gr2: 21 Hemiplegia: Repeated assessments methods of rehab.
Gr1: 9 R; 12 L Gr2: 17.5 days Gr2: 74.6 using an FIM instrument Standard
Gr2: 4 R; 9 L SD: 1.3 days SD: 91.8 were done by the staff customary method
to determine the rehab of assessment
planning and fix goal following
making on an individual the Bobath
patient-by-patient basis. concept. The
Bobath concept
underlines the
need for inhibiting
the stroke patient
in specific ways
TOPICS IN STROKE REHABILITATION/SPRING 2006

so that he or
she will have to
relearn normal
movements, which
makes the patient
dependent on
ward staff for a
longer period of
time.

Young Total: 95 (Inclusion criteria from N/A Median Gr1: Home PT. The Gr2: Day hospital. N/A Gr1: 11 N/A 1, 0, 0
1993474 Gr1: 52 a previous study): age: patients received active The patients sessions in 8
Gr2: 43 Over 60 yr of age; and Gr1: 69 hands-on standard received standard wk.
about to be discharged Range, treatments by PT for a PT. Gr2: 2 x/wk
home after a new 60–89 total of 11 sessions in for a total of
stroke that had caused 8 wk. 15 sessions in
persisting disability. Gr2: 70 8 wk.
Exclusion: N/A Range,
60–88

Note: ADL = activities of daily living; CVA = cerebrovascular accident; EMG-BFB = electromyographic biofeedback; F = female; GP = general practitioner; Gr = group; IQR = interquartile range; L = left; M = male; N/A =
not available/not applicable; N/M = not measured; N/R = not reported; OT = occupational therapy; PT = physiotherapy; Quality R, B, W = randomization, blinding, withdrawals; R = right; SSS: Scandinavia Stroke Scale; ST =
speech and language therapist; SW = social worker; PNF = proprioceptive neuromuscular facilitation; ROM = range of motion; TENS = transcutaneous electrical nerve stimulation; UE = upper extremity; WHO = World Health
Organization; x/wk = times per week.
APPENDIX 4. Existing Guidelines on Stroke Rehabilitation
Department of Veterans
Evidence-Based Affairs / Department of
Rehabilitation Clinical Practice Defense US Army (VA/
intervention Guidelines DoD) (2003)75
Therapeutic Quality of Levels I, II, and III evidence
exercises published (see Table 2)
evidence
Clinical Existing good evidence
recommendations that the use of overhead
pulleys should be avoided.
Fair evidence to support
that stroke patients should
participate in a regular
strengthening and aerobic
exercise program. Poor
evidence to support that
strengthening should be
included in the acute phase.
Biofeedback Quality of Level I
published
evidence
Clinical There is insufficient
recommendations evidence as to whether
routine use of biofeedback
should or should not be
used as an intervention for
patients undergoing stroke
rehabilitation.
Scottish
Heart and Stroke Intercollegiate
Foundation of Ontario Royal College of Guidelines Network
(2002) 71 St. Joseph’s Health Care London (2004)73 Physicians (2004) 72 (SIGN) (2002) 74
Levels I and II Levels I and II N/A N/A
Good evidence existing Limited evidence that the restorative N/A N/A
on enhancing sensory- approach increases the number of patients
motor recovery of who are institutionalized, and that the
the upper limb as Bobath treatment improves the quality of
an intervention. Fair gait.
evidence about avoiding Strong evidence that strength training
the use of overhead improves gait, that balance training
pulleys for patients with improves outcomes, and that sensory-
motor impairments and motor training with robotic devices
moderate shoulder pain. improves functional and motor outcomes
of the shoulder and elbow. Moderate
evidence that aggressive range of motion
therapy, commonly using overhead pulleys,
results in markedly increased shoulder
pain, and that the gentler range of motion
program is preferred.
Levels I and II Level I Levels I and II Level I
Electromyography Strong evidence that biofeedback training Existing evidence that Existing fair
(EMG) biofeedback improves gait and standing. biofeedback systems evidence that EMG
should be included should be routinely biofeedback should
as an intervention for excluded as an not be used routinely
patients who have a intervention for post-stroke in the rehabilitation
high level of motor rehabilitation. of function and
return for reducing movement.
motor impairment and
improving functional
motor recovery.
Appendix 4
263
continues
APPENDIX 4. Continued
264

Department of Veterans Scottish

Evidence-Based Affairs / Department of Heart and Stroke Intercollegiate
Rehabilitation Clinical Practice Defense US Army (VA/ Foundation of Ontario Royal College of Guidelines Network
intervention Guidelines DoD) (2003)75 (2002) 71 St. Joseph’s Health Care London (2004)73 Physicians (2004) 72 (SIGN) (2002) 74

Acupuncture Quality of N/A N/A Level I Level I N/A

published
evidence

Clinical N/A N/A Conflicting evidence that acupuncture is Insufficient evidence to N/A
recommendations effective for improving stroke outcomes. make definitive statements
about the role of
acupuncture. Acupuncture
should only be used as an
intervention for patients
undergoing post-stroke
rehabilitation in the
context of ongoing trials.
TOPICS IN STROKE REHABILITATION/SPRING 2006

Balance Quality of N/A N/A Level I N/A N/A

training published
evidence

Clinical N/A N/A Existing strong evidence that balance N/A N/A
recommendations training post-stroke improves outcomes.

Constraint- Quality of Level I N/A Level I N/A N/A

induced published
movement evidence
therapy

Clinical This therapy should be N/A This therapy should be included as an N/A N/A
recommendations considered as a treatment intervention for patients undergoing stroke
for a select group of patients rehabilitation.
with 20º of wrist extension
and 10º of finger extension,
who have no sensory and
cognitive deficits.
Electrical Quality of Levels I and II Level I Level I Level I N/A
stimulation published
evidence

Clinical Recommendation for the
recommendations use of functional electrical
stimulation (FES) for gait
training.
There is insufficient evidence
as to whether multi-channel
FES should or should not be
included as an intervention
for severe hemiplegic patients
with gait impairment. FES
should be included as an
intervention for the treatment
of the shoulder in sub- and
post-acute stroke patients.
FES should be included Recommendation for the use of FES for FES should not be N/A
as an intervention for the gait training. FES should be included as routinely considered as
treatment of the shoulder an intervention for the treatment of the an intervention for stroke
in sub- and post- shoulder in sub- and post-acute stroke patients.
acute stroke patients. patients.
Electromyography
(EMG) biofeedback and
electrical stimulation of
the wrist and the forearm
should be included as an
intervention to reduce
motor impairment and
improve functional
motor recovery.

Therapy for Quality of N/A N/R Levels I and II N/A N/A

shoulder published
subluxation evidence

Clinical N/A There is existing good Existing good evidence that FES should N/A N/A
recommendations evidence that FES be included as an intervention for the
should be included as reduction of shoulder subluxation in sub-
an intervention for the acute stroke patients. There is limited
reduction of shoulder evidence that shoulder slings influence
subluxation in sub-acute clinical outcomes. Shoulder strapping
stroke patients. and prolonged positioning do not result
There is limited evidence in significant reductions in pain or
that shoulder slings improvements in upper limb function
influence clinical or range of motion. Proper positioning
outcomes. of the hemiplegic shoulder helps in the
prevention of subluxation.
Appendix 4
265

continues
APPENDIX 4. Continued
266

Department of Veterans Scottish

Evidence-Based Affairs / Department of Heart and Stroke Intercollegiate
Rehabilitation Clinical Practice Defense US Army (VA/ Foundation of Ontario Royal College of Guidelines Network
intervention Guidelines DoD) (2003)75 (2002) 71 St. Joseph’s Health Care London (2004)73 Physicians (2004) 72 (SIGN) (2002) 74

Transcutaneous Quality of N/A N/A Level I Level I N/A

electrical nerve published
stimulation evidence
(TENS)

Clinical N/A N/A There is conflicting evidence that TENS Recommendation that N/A
recommendations improves a variety of outcomes, including high-intensity TENS
motor recovery, spasticity and ADL. should be included as an
intervention for patients
undergoing stroke
rehabilitation, if non
steroidal anti-inflammatory
analgesia has no significant
TOPICS IN STROKE REHABILITATION/SPRING 2006

effect. TENS should

only be used as a routine
intervention for improving
muscle control in the
context of ongoing trials.

Therapeutic Quality of N/A N/A Level I N/A N/A

ultrasound published
evidence

Clinical N/A N/A Existing evidence that adding ultrasound N/A N/A
recommendations therapy to range of motion exercises does
not change outcomes.
Gait training Quality of Levels I and II
published
evidence
Clinical Treadmill training with partial
recommendations body weight support should
be included as an adjunct
to conventional therapy for
patients undergoing stroke
rehabilitation. Conflicting
evidence for FES for gait
training following stroke,
and insufficient evidence
to recommend multi-
channel FES for severe
hemiplegic patients with gait
impairment.
Task-oriented Quality of Level I
training published
evidence
Clinical There is insufficient evidence
recommendations to recommend for or
against neurodevelopmental
training compared to other
approaches for motor
retraining post- stroke.
Stroke Prevention and
Educational Awareness
Diffusion (SPREAD)
Collaboration (2000)76
Quality of Level I
published
evidence
Clinical Existing evidence to
recommendations encourage functional task
training for stroke patients
with functional deficits in
affected limbs.
N/A
N/A
Level I
Existing evidence to
encourage functional
task training for stroke
patients with functional
deficits in affected limbs.
Agency for Health Care
Policy and Research
(AHCPR) (1995)70
Level II
Evidence to encourage
functional task training
for stroke patients with
functional deficits in
affected limbs.
Level I Level I N/A
Conflicting evidence on the use of treadmill Treadmill training with N/A
training, and body-weight support and partial body weight
treadmill training. support should be
Strong evidence was found to include included as an adjunct to
FES for patients undergoing stroke conventional therapy for
rehabilitation, and for the use of rhythmic patients undergoing stroke
auditory stimulation in patients post-stroke. rehabilitation.
Level I N/A Level I
Conclusion that the benefit of leisure N/A Existing evidence to
therapy post-stroke is uncertain. support the use of
task-specific training
to improve task
performance.
Appendix 4
267
continues
APPENDIX 4. Continued
268

Department of Veterans Scottish

Evidence-Based Affairs / Department of Heart and Stroke Intercollegiate
Rehabilitation Clinical Practice Defense US Army (VA/ Foundation of Ontario Royal College of Guidelines Network
intervention Guidelines DoD) (2003)75 (2002) 71 St. Joseph’s Health Care London (2004)73 Physicians (2004) 72 (SIGN) (2002) 74

Sensory Quality of N/A N/A Level I N/A N/A

intervention published
evidence

Clinical N/A N/A For sensory function training of the hand, N/A N/A
recommendations strong evidence was found that perceptual
training interventions improve perceptual
functioning.
For visual attention retraining, there
is strong evidence that visual scanning
techniques improve visual neglect with
associated improvements in function.
Existing conflicting evidence that external
TOPICS IN STROKE REHABILITATION/SPRING 2006

sensory interventions such as eye

movement feedback are beneficial in the
treatment of visual neglect.
Intensity and Quality of Level I N/A Levels I and II Level I Levels I and III
organization of published
rehabilitation evidence

Clinical There is insufficient evidence
recommendations as to whether inpatient
stroke rehabilitation care
demonstrates superiority in
outcomes when compared
to outpatient stroke
rehabilitation.
There is also good evidence
that early rehabilitation
therapy should be provided
as soon as the patient’s
medical status is stable.
N/A Existing good evidence that: a faster There is good evidence There is good
recovery and an earlier discharge from the that acute care for patients evidence that
hospital can result from intense therapies undergoing post-stroke stroke patients in
delivered in short periods of time within rehabilitation in a hospital the acute phase
the post-stroke rehabilitation process; should be delivered in a should be treated in
care pathways do not improve post-stroke ward or ward area with a multidisciplinary
rehabilitation outcomes; care pathways specialists with expertise in stroke unit.
do not reduce hospital costs or decrease stroke management (i.e., a There is poor
the length of hospital stays for patients stroke unit). evidence to support
undergoing post-stroke rehabilitation; There is also good evidence that occupational
post-stroke rehabilitation provided by that stroke patients should therapy treatment
specialized stroke units enables patients only be managed at home should be based on
to maintain both short- and long-term if the services delivered an assessment of the
functional improvements, when compared at home are part of a individual’s unique
with general medical units; stroke service provided problems.
patients undergoing post-stroke by specialists and that
rehabilitation can maintain and even otherwise, these patients
improve functional outcomes for up to 1 should be admitted to the
year; additional home-based therapy does hospital for initial care and
not improve the overall functional outcome assessment.
scores for patients undergoing post-stroke
rehabilitation.
The evidence is fair as to whether the
improvement of patients’ functional
outcomes is a direct result of an early
admission to post-stroke rehabilitation.

European Stroke Initiative

(2003)78

Intensity and Quality of Level not reported

organization of published
rehabilitation evidence
(continued)

Clinical The evidence is poor as to

recommendations whether stroke patients for
whom care services are able
to provide adequate and
Appendix 4

flexible support within 24 hrs

should be considered for a
home management program.
269

Note: This previously existing clinical practice guideline was looked at though it was deemed not applicable: the European Stroke Initiative (2003).78 N/A = not applicable; N/R = not reported (N/R); N/C = not considered.