You are on page 1of 17


Ingrid G.L van de Port, MSc
Sharon Wood-Dauphinee, PhD, PT
Eline Lindeman, PhD, MD
Gert Kwakkel, PhD

From the Center of Excellence for
Rehabilitation Medicine Utrecht,
Rehabilitation Center De Hoogstraat,
Utrecht, The Netherlands (IGLvdP,
EL, GK); Rudolf Magnus Institute of
Neuroscience, Department of
Neurology and Neurosurgery,
University Medical Center, Utrecht,
The Netherlands (IGLvdP, EL, GK);
Vrije Universiteit Medical Center,
Department of Rehabilitation,
Amsterdam, The Netherlands (GK);
and School of Physical and
Occupational Therapy, Department
of Epidemiology and Biostatistics,
McGill University, Montreal, Canada

All correspondence and requests for
reprints should be addressed to I.G.L.
van de Port, MS, Rehabilitation
Center De Hoogstraat,
Rembrandtkade 10, NL-3583 TM
Utrecht, The Netherlands.

This study was undertaken as part
of the Long-Term Prognosis of
Functional Outcome in Neurological
Disorders program, supervised by the
Department of Rehabilitation
Medicine of the VU Medical Center,
Amsterdam and supported by the
Netherlands Organisation for Health
Research and Development (project
no. 1435.0020).
American Journal of Physical
Medicine & Rehabilitation
Copyright © 2007 by Lippincott
Williams & Wilkins
DOI: 10.1097/PHM.0b013e31802ee464



Effects of Exercise Training
Programs on Walking Competency
After Stroke
A Systematic Review

van de Port IGL, Wood-Dauphinee S, Lindeman E, Kwakkel G: Effects of
exercise training programs on walking competency after stroke: a systematic
review. Am J Phys Med Rehabil 2007;86:935–951.
To determine the effectiveness of training programs that focus on lower-limb
strengthening, cardiorespiratory fitness, or gait-oriented tasks in improving gait,
gait-related activities, and health-related quality of life after stroke. Randomized
controlled trials (RCTs) were searched for in the databases of Pubmed, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic
Reviews, DARE, Physiotherapy Evidence Database (PEDro), EMBASE, Database of the Dutch Institute of Allied Health Care, and CINAHL. Databases were
systematically searched by two independent researchers. The following inclusion
criteria were applied: (1) participants were people with stroke, older than 18 yrs;
(2) one of the outcomes focused on gait-related activities; (3) the studies
evaluated the effectiveness of therapy programs focusing on lower-limb strengthening, cardiorespiratory fitness, or gait-oriented training; and (4) the study was
published in English, German, or Dutch. Studies were collected up to November
2005, and their methodological quality was assessed using the PEDro scale.
Studies were pooled and summarized effect sizes were calculated. Best-evidence
synthesis was applied if pooling was impossible. Twenty-one RCTs were included, of which five focused on lower-limb strengthening, two on cardiorespiratory fitness training (e.g., cycling exercises), and 14 on gait-oriented training.
Median PEDro score was 7. Meta-analysis showed a significant medium effect of
gait-oriented training interventions on both gait speed and walking distance,
whereas a small, nonsignificant effect size was found on balance. Cardiorespiratory fitness programs had a nonsignificant medium effect size on gait speed. No
significant effects were found for programs targeting lower-limb strengthening. In
the best-evidence synthesis, strong evidence was found to support cardiorespiratory training for stair-climbing performance. Although functional mobility was
positively affected, no evidence was found that activities of daily living, instrumental activities of daily living, or health-related quality of life were significantly
affected by gait-oriented training. This review shows that gait-oriented training is
effective in improving walking competency after stroke.
Key Words:
Cerebrovascular Diseases, Systematic Review, Exercise Therapy,
Gait-Related Activities

November 2007

Exercise Training After Stroke


(3) the studies evaluated the effectiveness of therapy programs focusing on lowerlimb strengthening. achieving and maintaining the ability to walk in the home and in the community is an important aim of stroke rehabilitation. free weights.. and walking for specified distances. Gait-related activities were defined in the present study as activities involving mobility-related tasks. with the aim of improving strength and muscular endurance. We conducted a systematic review of the literature on the effects of lower-limb strength training. masses. No. stroke was defined according to the World Health Organization definition as an acute neurological dysfunction of vascular origin with sudden (within seconds) or at least rapid (within hours) occurrence of symptoms and signs corresponding to the involvement of focal areas in the brain. such as a random-numbers table or concealed envelopes (Pubmed 1990). Physiotherapy Evidence Database (PEDro). cardiorespiratory fitness training. Studies were collected up to November 2005. and the treatments to be administered are selected by a random process. 11 . or Dutch. 86. They also noted that there have been few studies including strength and mixed training. as established in the general population. (2) one of the study outcomes focused on gait-related activities. Crossover designs were treated as RCTs by taking only the outcomes after the first intervention phase. and abstracts published in proceedings of conferences were also examined. Randomized controlled trials (RCTs) were included if they met the following inclusion criteria: (1) participants were patients with stroke older than 18 yrs.S troke is a major cause of disability in the developed world. Recently. They suggested that programs concentrating on cardiorespiratory fitness resulted in improved scores for walking ability and maximum walking speed. German. can be extrapolated to persons who are disabled by stroke. in which concurrent enrollment and follow-up of the test.9 RCT was defined as a clinical trial involving at least one test treatment and one control treatment. virtual reality.3 Cardiorespiratory fitness training was defined as training intended to improve the cardiorespiratory component of fitness. Rehabil. or electrical stimulation) were excluded. EMBASE.1 Because independent gait is closely related to independence in activities of daily living (ADL). without aiming to improve gait perforAm. ● Vol. 14% walk with assistance. and mixed training programs on gait.”5 In addition.5. and health-related quality of life in those who had sustained a stroke. exercise therapy. or isokinetic devices. specialized machine weights. Database of the Dutch Institute of Allied Health Care. The following MeSH headings and key words were used for the electronic databases: cerebrovascular accident. Definitions In the present review. and CINAHL (1980 through November 2005)) were systemati- 936 van de Port et al. gait. J.7 Walking competency was defined as “the level of walking ability that allows individuals to navigate their community proficiently and safely. Studies evaluating specific neurological treatment approaches applying gait manipulations (e. Electronic databases (Pubmed. such as stair walking. rehabilitation. cardiorespiratory training. it is necessary to systematically evaluate the effects of the different training programs that aim to restore walking competency. cycling. WE). cally searched by two independent researchers (IvdP.. cardiorespiratory fitness. turning. 64% of survivors walk independently at the end of rehabilitation. Cochrane Database of Systematic Reviews. by using specific devices such as body weight– supported training.8 To optimize the treatment of those with stroke. DARE. walking quickly. making transfers.2 Saunders and colleagues3 evaluated the evidence for the effects of strength training. Cochrane Central register of Controlled Trials. Med. (4) the study was published in English.g. walking.and control-treated groups is ensured. METHODS Literature Search Potentially relevant studies were identified through computerized and manual searches. and 22% are unable to walk. typically performed for extended periods of time on ergometers (e. narrative reviews. there has been increasing interest in combinations of strength and cardiorespiratory training. gait-related activities. that are typically carried out by making repeated muscle contractions resisted by body weight. Lower-limb strength training was defined as prescribed exercises for the lower limbs.g. The full search strategy is available on request from the corresponding author. and randomized controlled trial. and gait-oriented training on gait. even in those patients in which the intervention was initiated beyond 6 mos after stroke. there is growing evidence that the link between physical training and improved cardiorespiratory fitness. or gait-oriented training. Bibliographies of review articles. often resulting in difficulties in walking.4 – 6 Salbach and colleagues5 suggested that high-intensity taskoriented practice may enhance walking competency in patients with stroke better than other methods. and these studies have been inconclusive. According to the Copenhagen Stroke Study. Phys. neurology. rowing). elastic devices. and (5) the design was an RCT. in which gait and gait-related tasks are practiced using a functional approach.

gu values of individual studies were averaged to obtain a weighted summarized effect size (SES). The fixed-effects model was used to decide whether the SES was statistically significant. Eligibility criteria specified Random allocation Concealed allocation Baseline prognostic similarity Participant blinding Therapist blinding Outcome assessor blinding More than 85% follow-up for at least one primary outcome Intention-to-treat analysis Between. which resulted in smaller variances. lower-limb strengthening. and/or numbers of studies found.14 For all outcome variables.3 We defined gait-oriented training as training intended to improve gait performance and walking competency in terms of different parameters of gait (e. Outcomes were pooled if the studies were comparable in terms of the type of intervention (i. stride and step length). point estimates were derived from graphs presented in the article. 3.11 (Table 1).05 (two tailed). divided by the pooled SDi of the experimental and control groups at baseline.14. PEDro scores were used as a basis for best-evidence syntheses and to discuss the methodological strengths and weaknesses of the studies. and those above 0. 5. intervention category. In the case of persistent disagreement. 2. those from 0.15 Best-Evidence Synthesis A best-evidence synthesis was conducted if pooling was impossible because of differences in outcomes. 6. The impact of sample size was addressed by estimating a weighting factor wi for each study and applying greater weight to effect sizes from studies with larger samples. 7. and the weights of each study were combined to estimate the variance of the SES. The homogeneity (or heterogeneity) test statistic (Q statistic) of each set of effect sizes was examined to determine whether studies shared a common effect size from which the variance could be explained by sampling error alone.8 as medium. On the basis of the classification by Cohen. Methodological Quality Two independent reviewers (IvdP and WE) assessed the methodological quality of each study using the PEDro scale10. November 2007 Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Yes/No Because gi tends to overestimate the population effect size in studies with a small number of patients. and standard deviation [SD] of the outcome scores in the experimental and control groups at baseline) were entered in Excel for Windows. a third reviewer made the final decision after discussions with the primary reviewers. Pooled SDi was estimated using the baseline SDs of the control and experimental groups.12 If additional information was required. a correction was applied to obtain an unbiased estimate: gu (unbiased Hedges’ g). or gait-oriented training) and if they assessed the same construct. or postintervention distributions. Using criteria based on the methodological quality score of the PEDro scale. 11.2 to 0. P values. stride and stepping frequency. 8. we contacted the authors or derived SDs from t or F statistics. The effect size gi (Hedges’ g) for individual studies was assessed by calculating the difference in mean changes between the experimental and control groups. 10. the percentage of total variation across the studies was calculated as I2. Quantitative Analysis Data contained in the abstract (numbers of patients in the experimental and control groups. which gives a better indication of the consistency between trials.g.e. the critical value for rejecting H0 was set at a level of 0. we classified the studies as high quality (four points or more) or low quality (three points or less).2 were classified as small.16 a random-effects model was applied.TABLE 1 The 11 items of the Physiotherapy Evidence Database (PEDro) scale for methodological quality 1. mean difference in change score.. Subsequently. cardiorespiratory fitness. gait speed.13 SES was expressed as the number of standard deviation units (SDUs) and a confidence interval (CI).16 When significant heterogeneity was found (I2 values ⬎50%). 4..7 SubExercise Training After Stroke 937 . 9. effect sizes below 0.15 Because the Q statistic underestimates the heterogeneity in a meta-analysis. and/or walking endurance.8 as large. If necessary.or within-group statistical analysis for at least one primary outcome Point estimates of variability given for at least one primary outcome mance as such.

49.53. except the work by Glasser.47 Studies focusing on cardiorespiratory fitness training included 104 participants.51 Time between stroke onset and the start of the intervention ranged from 3 mos46 to a 938 van de Port et al. we excluded 440 studies. mean of 4 yrs.4. All studies applied statistical analysis to group differences and reported point estimates and measures of variability. scored a minimum of six points. of whom 53 were assigned to the intervention group.50 –52 three that concentrated on cardiorespiratory fitness.46 – 48.47. In no study was the therapist blind to group status.65 Am.45 Screening of references of the articles led to another four studies46 – 49 being included.18 RESULTS The initial search strategy identified 486 relevant citations.4.50. The two RCTs focusing on cardiorespiratory fitness both scored six points. except for one. 11 .53–55 and 14 that targeted gait-oriented training. of whom 332 were assigned to the intervention group. All studies. this was regarded as no evidence.49-53. Time between stroke onset and the start of the intervention ranged from a mean of 16 days53 to more than 1 yr.56 – 60.54 One of these53 was used in our metaanalysis. and therefore they could be blinded. Sample sizes ranged from 2046. The studies centering on lower-limb strength training included 240 participants. This was as expected. with a median score of seven points (Table 3).59 gait speed (17 RCTs. 23 studies were included in the present systematic review (Fig. of whom 121 were assigned to the intervention group.49. 86.5.46. reasons for exclusion included studies not being randomized. 1). Forty-six full-text articles were selected. Individual study sample sizes were 1255 and 92 participants. the study by Lindsley et al.47 to a maximum of 133 participants. On the basis of title and abstract.5. J.5. All studies. One study on cardiorespiratory training53 failed to report baseline SDs.4. based on van Tulder et al. Only Katz and colleagues53 asked the patients to walk as far as they could. The selection included six RCTs that focused on strength training of the lower limb.58 and Macko and colleagues63.4.56. studies were categorized into four levels of evidence.46 TeixeiraSalmela.59. or the results are conflicting If the number of studies showing evidence was less than 50% of the total number of studies found within the same methodological quality category.22–24 Another 20 studies were excluded because the intervention did not meet the criteria. Despite being an RCT.66 Methodological Quality PEDro scores ranged from four to eight points.53. Quantitative Analysis Pooling was possible for balance (four RCTs.56 – 60.5. n ⫽ 743). or being conducted in a different patient population.56 – 66 Two RCTs concentrating on the effects of cardiorespiratory fitness employed the same population.61– 66 and walking distance (13 RCTs.53 A median of seven points (range four to eight) was scored by RCTs targeting gait-oriented training. We decided to include the early control group in our review because the number of patients who completed this trial was larger than that in the other control group. because the therapists had to conduct the therapy. using an intervention that did not fit within our definition. 2 and 3).53 respectively.55. three more were excluded because the studies were not RCTs19 –21 and three were excluded because the outcome measures did not reflect gaitrelated activities.25– 44 and one study was excluded because it focused on a subgroup of a larger RCT. Individual sample sizes ranged from 958 to a maximum of 100 participants. so SDs were derived from P values.68 walk test. Rehabil.17 1) Strong evidence: provided by generally consistent findings in multiple relevant high-quality RCTs 2) Moderate evidence: provided by generally consistent findings in one relevant high-quality RCT and one or more relevant low-quality RCTs 3) Limited evidence: provided by generally consistent findings in one relevant high-quality RCT or in one or more relevant low-quality RCTs 4) No or conflicting evidence: no RCTs are available. RCTs centering on lower-limb strengthening scored a median of seven points (range four to eight).46 specified the eligibility criteria. n ⫽ 692). Gait speed was measured over distances ranging from 5 to 30 m.52.52. so we used the SD of the postintervention measurement to calculate gi (Figs. Of these. No. and the second study was used to obtain additional information. Another study59 on gait-oriented training did not provide baseline SDs either.66 Dean.63 Balance was determined by the Berg Balance Scale (BBS)67 in all studies.48 was excluded because of a lack of information.4 Time between stroke onset and the start of the intervention varied between 8 days65 to a mean of 8 yrs.58. In total. Med.49.55 The studies focusing on gait-related training included 574 participants. Phys.sequently. The study by Richards and colleagues65 included two control groups.5.65 Walking distance was assessed by the 2-min51 or 6-min4. Table 2 shows the main characteristics of the 21 studies included in the present meta-analysis. ● Vol. n ⫽ 274).

0. Exercise Training After Stroke 939 . Z ⫽ 1.50. ⫺0.83. A homogeneous nonsignificant SES was found compared with control groups (SES [fixed].1%). P ⫽ 0. 0.47. A heterogeneous nonsignificant SES was found compared with the control groups (SES [random].07. I2 ⫽ 0%).28. Lower-Limb Strengthening 46. CI.47. ⫺0. ⫺0. Cardiorespiratory Fitness Training Two studies involving cardiorespiratory training53. P ⫽ 0. I2 ⫽ 21%).98.FIGURE 1 Inclusion and exclusion criteria for the present review.667.13 SDU. Z ⫽ ⫺0.00 SDU. I2 ⫽ 57. Three studies (n ⫽ 200)50 –52 determined walking distance and found a homogenous nonsignificant SES compared with control November 2007 groups (SES [fixed]. CI. Z ⫽ 0.03 to 0.36 SDU. ⫺0.73 to Four studies targeting lower-limb strengthening (n ⫽ 107) measured gait speed.55 (n ⫽ 104) assessed gait speed. CI. P ⫽ 0.28 to 0.75.

unilateral paretic and nonparetic knee extension. based on the use of a static dynamometer. timed up and go. Intervention aimed at increasing strength did not result in differences in walking between groups. ambulation time Outcome 6 wks. Differences in ambulation times and FAP scores were nonsignificant. C: Passive range of motion. C: Conventional therapy. Med. Mean Days at Inclusion Intervention I: Motor reeducation program for the paretic lower limb. J.50 Ouellette et al. I: Conventional therapy plus progressive resistance exercises performed with weights at the waist or on the lower extremities. Phys. functional limitation and disability (LLFDI). I: Maximal concentric isokinetic strength training. finger-to-nose movements. TABLE 2 Characteristics of the studies included in the review 12 wks. three times a week. gait speed. C: Motor reeducation program for the paretic upper limb. Progressive resistance training was not effective compared with the same exercises without resistance.51 42 (21/21) 25 (12/13) Bourbonnais et al. I: Therapeutic exercise program based on neurophysiological and development theories and gait training plus isokinetic training. Treatment of the lower limb produces an improvement in gait velocity and walking speed. ● Vol. Rehabil.47 Lower-limb strength training Glasser46 20 (10/10) Study 6 mos to 6 yrs after stroke (874) ⬍6 mos after stroke (38) Chronic (1096) ⬎6 mos (1460) 3–6 mos (137) Time Since Stroke. C: Therapeutic exercise program based on neurophysiological and development theories and gait training. C: Bilateral range of motion and upper-body flexibility exercises. 5 days/wk. gait speed Motor function (FM). 2 hrs/day Intensity Progressive resistance training safely improves lower-limb strength in the paretic and nonparetic limb and results in reductions in functional limitations and disabilities. quality of life (SF36) Functional Ambulation Profile (FAP). 86. ankle dorsiflexion. quality of life (SIP) Disability (CMSA Disability Inventory). gait speed. peak muscle power. and plantarflexion. walking distance 6 wks. stair climbing. I: High-intensity resistance training program consisting of bilateral leg press. Author’s Conclusion .940 van de Port et al. Am. based on the use of a static dynamometer. 11 n(E/C) 106 (54/52) Moreland et al. three times a week. walking distance. No. depression (GDS). three times a week During rehabilitation (mean 8 wks). three times a week Lower-extremity muscle strength. 30 mins Lower-limb strength. stairclimbing speed.52 20 (10/10) Kim et al. 45 mins 5 wks. gait speed. chair rising.

30 mins. three times a week. Lower-extremity scores and gait velocity were significantly different. last 6 wks: three times a week. This one started later and was not as intensive (CON). balance (BBS). gait speed Gait speed. balance (BBS) Walking distance. ambulation (BI).November 2007 Exercise Training After Stroke 941 n(E/C) 12 (7/5) Duncan et al. gait speed. resistance exercise with a Kinetron isokinetic device and a treadmill. and endurance. I: Therapist-supervised homebased exercise program to improve strength. after submaximal aerobic training. instrumental ADL. balance (BBS). Mean Days at Inclusion 12 wks. motor function (FM). C: Conventional therapy. 10 times a week. first 2 wks: five times a week. gait speed. C1: Started early and was as intensive as for the experimental group but included more traditional approaches to care (ECON). Intervention Subacute (16) Time Since Stroke. ⬎1 yr after stroke (1315) Acute (about 10 days) I: Regular therapy and leg cycle ergometer training. 10 times a week. ADL. 40 mins 8 wks. five times a week. 90 mins Exp: 5 wks. balance. Author’s Conclusion . quality of life Balance (FM-B). 50 mins ECON: 5 wks.59 20 (10/10) Gait-oriented training Richards 27 (10/8/9) et al. 30 mins Intensity Motor function (FM). Group results demonstrated that gait velocity was similar in the three groups. 50 mins CON: 5 wks. C: Arm and hand exercise while sitting. walking distance. C: Usual care. 60 mins 8 wks. The experimental group attained significant improvement compared with the control group in cardiovascular fitness and gait speed. C2: Therapy was composed of similar techniques as provided to the other control group. three times a week.55 Cardiorespiratory training Katz-Leurer 90 (46/44) et al. workload. exercise time Outcome The experimental group showed greater improvement of neurological impairment and lower-extremity function. I: Intervention group participating in a water-based exercise program that focused on leg exercise to improve inclusive cardiovascular fitness and gait speed. including walking distance. Stroke patients in the subacute stage improved some of their aerobic and functional abilities.53 Study TABLE 2 Continued Subacute (61) I: Intensive and focused approach incorporating the use of tilt table and limb-load monitor.65 Chu et al.

functional mobility (FAC). stride length. gait speed. The combined program of muscle strengthening and physical conditioning resulted in gains in all measures of impairment and disability. quality of life (NHP). 86. 4 wks. Mean Days at Inclusion Intensity I: Physiotherapy treatment plus ambulation on a motor-driven treadmill at a comfortable walking speed. gait cycle Sit to stand. 60–90 mins aerobic exercises (graded walking plus stepping or cycling). ninehole peg test Gait speed. lower-extremity muscle strengthening. Walking distance. I: Treadmill retraining with the 4 wks. 11 n(E/C) 13 (6/7) 12 (6/6) 18 (10/8) 25 (13/12) Study TeixeiraSalmela et al. sit to stand. C: Similar organization and delivery as the experimental group. C: No intervention. level of physical activity (HAP).66 Dean et al. balance. three times a instruction to walk for as long week. assistance.61 TABLE 2 Continued ⬍90 days (34. and gastrocnemius muscular activity.2) ⬎ 3 mos (658) ⬎9 mos (2799) Time Since Stroke. 10 wks. Author’s Conclusion . three times a week. C: Physiotherapy treatment plus ambulation on a floor surface at a comfortable speed using walking aids. Phys.62 Laufer et al. Treadmill training may be more effective than conventional gait training in improving gait parameters such as functional ambulation. 60 mins workstations designed to strengthen the muscles in the affected leg in a functional way and practicing locomotionrelated tasks. 8–20 mins I: Program consisting of warm-up. C: Conventional physiotherapy. walking distance. except that it was designed to improve the function of the affected upper limb. but there were no statistically significant differences in gait between the conventional and treadmill retraining groups. gait speed. J. gait speed Outcome Improvements were seen. percentage of paretic single-stance period. Med. and the step test showed significant improvements between groups. 60 mins as patients felt comfortable. ADL.58 Liston et al. gait speed. No. timed up and go. 3 wks. Am. Intervention Standing balance. ● Vol.942 van de Port et al. and resting periods as needed. step test Muscle strength and tone. and cool-down. three times I: Circuit program including a week. five times a week. Rehabil. This task-related circuit training improved locomotor function in chronic stroke.

C: Physiotherapeutic gait therapy based on the latest principles of proprioceptive neuromuscular facilitation and Bobath concepts.57 n(E/C) Study TABLE 2 Continued Subacute (43) 30–150 days (76) 6 mos to 5 yrs (822) ⬎4 wks (114. 30 mins 4 wks. C: Upper-limb group activities including functional tasks. 90 mins 4 wks. Mean Days at Inclusion I: Mobility-related group activities including endurance tasks and functional tasks. home exercise program consisting of exercises to lengthen and strengthen lower-limb muscles and to train balance and coordination. The mobility group showed significantly better locomotor ability than the upper-limb group. endurance. I1: Conventional physiotherapy plus limited progressive treadmill training (LTT). quality of life (SA-SIP30) Gait speed. The intervention program significantly increased walking speed and walking capacity compared with the control group. balance. five times a week. gait speed. Author’s Conclusion . C: Low-intensity. I: Both treadmill and overground walking. step length and width. upper-extremity function. timed up and go.56 Duncan et al. 30 mins Intensity Lower-extremity muscle and grip strength. cadence. step test. cadence. walking distance Upper-limb function (MAS. stride length. progressive exercise program produced gains in endurance. C: Usual care. with the proportion of treadmill walking decreasing by 10% each week. JTHFT). balance (BBS).64 Blennerhassett et al. Intervention 4 wks. three times at week. 12 sessions. Findings support the use of additional task-related practice during inpatient stroke rehabilitation. 60 mins 12 wks.November 2007 Exercise Training After Stroke 943 27 (13/14) 92 (44/48) Ada et al. three times a week.4 30 (15/15) 60 (20/20/20) Pohl et al. I: Exercise program designed to improve strength and balance and to encourage more use of the affected extremity. Structured STT in poststroke patients resulted in better walking abilities than LTT or conventional physiotherapy. motor function (FM). and mobility beyond those attributable to spontaneous recovery and usual care. walking distance Gait speed. functional mobility (FAC) Outcome This structured.6) Time Since Stroke. I2: Conventional physiotherapy plus structured speeddependent treadmill training (STT).

three times a week 6 months. I: Progressive task-oriented modality to optimize locomotor relearning. Phys. experimental vs. walking distance. Bobath-oriented. modified Ashworth scale. gross motor function (RGMF).60 Salbach et al. functional mobility (RMI). Walking Impairment Questionnaire. FM. 40 mins Gait speed. physical activity (PAS) Timed up and go. speed. mobility. C: Upper-extremity activities. Chedoke–McMaster stroke assessment. leg muscle strength. C. Med. walking distance. endurance. Rivermead mobility index. Fugl Meyer. five times a week. JTHFT. walking distance 6 wks. SA-SIP30. Barthel index. RGMF. Author’s Conclusion E/C. C: Seated upper-extremity program. I: Ten functional tasks designed to strengthen the lower extremities and enhance walking balance. endurance. MAS. Intervention 19 wks. RMI. Human Activity Profile. Sickness Impact Profile. C: Conventional therapy.63 Pang et al. Stroke Adapted-Sickness Impact Profile 30. Both functional mobility and cardiovascular fitness improved more after the intervention than after conventional care. GDS. three times a week. Mean Days at Inclusion I: Progressive fitness and mobility exercise program designed to improve cardiorespiratory fitness. LLFDI. Social Functioning 36. walking distance. WIQ. 86. three times a week. The task-oriented intervention significantly improved gait speed and walking distance. Functional Ambulation Profile. Physical Activity Scale.49 ⬎1 yr (1881) ⬎6 mos after stroke (1125) Chronic (228) ⬍6 wks (44) Time Since Stroke. CMSA.5 Macko et al. HAP. gait speed. and distance. 11 50 (25/25) 91 (44/47) 61 (32/29) 63 (32/31) Eich et al. balance (BBS). NHP. J. ECON. Late Life Function and Disability Instrument. walking quality Outcome 6 wks. BBS. intervention group. No. Rehabil. SIP. FAP. Jebsen Taylor hand function test. 60 mins Gait speed. Bobath-oriented plus treadmill training. Walking Impairment Questionnaire (WIQ) Muscle strength. 60 mins Intensity The intervention group had significantly greater gains in cardiorespiratory fitness. FM-B. Addition of aerobic treadmill training to Bobath-oriented physiotherapy resulted in significant improvement in gait speed and walking distance. control group. and paretic leg strength. PAS. n(E/C) Study TABLE 2 Continued . balance. I: Individual physiotherapy. Rivermead gross motor function. Geriatric Depression Scale. Fugl Meyer balance. C: Individual physiotherapy. ● Vol. I. Berg balance scale. Am. balance (BBS).944 van de Port et al. early control group. providing cardiovascular conditioning. SF36. control group. BI. and mobility. Notthingham health profile.

58 Liston et al.5. 0.5.30 – 0.49 1 2 3 4 5 6 7 8 9 10 11 Total Score No Yes Yes Yes Yes 1 1 1 1 1 0 0 1 1 0 0 1 1 1 1 0 1 0 0 0 0 0 0 0 0 0 1 0 1 1 1 1 1 1 1 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 4 8 6 8 7 Yes Yes 1 1 0 0 1 1 0 0 0 0 1 1 1 1 0 0 1 1 1 1 6 6 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 0 0 0 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 1 1 1 1 1 1 1 0 1 1 1 1 0 1 1 1 1 1 1 1 1 0 1 0 1 0 0 1 0 0 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 6 7 4 4 7 6 6 8 8 8 8 8 5 8 Because only one study analyzed the effect of cardiorespiratory training on balance55 and one on walking distance.05 to 0.50 Moreland et al. Z ⫽ 3.56 – 60. In addition. I2 ⫽ 31. CI. P ⫽ 0.43. I2 ⫽ 61. Cardiorespiratory Fitness Training There is limited evidence that cardiorespiratory training negatively affects balance. P ⬍ 0.63 assessed the effect of gait-oriented training on walking distance. Gait-Oriented Training Standing balance showed no statistically significant differences between control and experimental groups in two high-quality studies focusing on gait-oriented training. Gait-Oriented Training Four studies assessed balance after gait-oriented training4. Although they used different measures to determine stair-climbing performance.65 Duncan et al.53 these results are described in the best-evidence syntheses.61 Pohl et al.62 Laufer et al. CI.56. however.47 These findings provide strong evidence that programs focusing on lower-limb strengthening do not produce greater improvement in stair-climbing ability than conventional care.27– 0.53 One high-quality study53 on cardiorespiratory fitness training also assessed stair climbing by asking the patients to climb as many stairs as possible at comfortable speed. both studies concluded that changes in stair climbing did not significantly differ between the experimental and control groups.64 whereas another high-quality study failed to find significant results in favor of Exercise Training After Stroke 945 .52 Cardiorespiratory fitness training Katz-Leurer et al.57 Eich et al. 0. suggesting limited evidence in favor of cardiorespiratory training for improving stair climbing. ⫺0.59 – 66 evaluated gait speed and found a homogenous significant SES (SES [fixed]. Z ⫽ 1. One study also evaluated health-related quality of life by means of the Short Form 36 and concluded that there was no significant differNovember 2007 ence between groups.52 selected stair climbing as a secondary outcome measure. SDU.62 Two high-quality studies.64 Ada et al. A heterogeneous significant SES was found compared with the control groups (SES [random].TABLE 3 Physiotherapy Evidence Database (PEDro) scores for each RCT Study Lower-limb strength training Glasser46 Kim et al.2%). presented statistically significant differences between groups on the functional ambulation category.51 Ouellette et al.62 SDU. The experimental group performed significantly better than the control group.59 Teixeira-Salmela et al.55 and there is limited evidence for a positive impact of cardiorespiratory training on walking distance.01.63.5. Best-Evidence Syntheses Lower-Limb Strengthening Two high-quality studies on lower-limb strengthening47. I2 ⫽ 0%).58. Moreover.56 Duncan et al. nine studies (n ⫽ 451)4.3%).5 Macko et al. Twelve studies centered on gaitoriented training (n ⫽ 501)4.61.55 Gait-oriented training Richards et al.59.49.66 Dean et al.63 Pang et al.73.53 Chu et al. P ⬍ 0.45 SDU.95. CI. there was limited evidence that programs of lower-limb strengthening are not superior to conventional care in improving health-related quality of life. 0.60 Salbach et al. 0. 0.59 and found a homogenous nonsignificant SES (SES [fixed].4 Blennerhassett et al.01.47 Bourbonnais et al. Z ⫽ 4.61.

whereas limited evidence was found that there is no effect of gaitoriented training on walking quality. In agreement with the above findings. also found positive effects on gait speed. gait-oriented training on the Rivermead Mobility Index.60 The above findings provide strong evidence that standing balance.56. Rehabil.62 or health-related quality of life.59. by means of leg cycle ergometers and water-based exercises.59 although one lowquality study did find significant differences in quality of life between groups. walking distance.FIGURE 2 Summarized effect of gait speed (mean and 95% CI). IADL. Med. Strong evidence was found for improved functional mobility after gait-oriented training. No. programs focusing on lower-limb strengthening alone failed to show significant effects on gait speed and walking distance. Phys. In addition. 86. The results showed positive. or quality of life are not significantly more improved by gait-oriented training than by conventional care. DISCUSSION This systematic review included 21 high-quality RCTs. whereas no significant effects were found on balance control as measured by the BBS. a bestevidence synthesis showed that lower-limb strength training did not affect outcomes such as stair climbing or health-related quality of life.69 there is some discussion about the clinical implication of the changes assessed by the BBS.62.66 Finally.63 The high-quality studies also found no significant effects of gait-oriented training on outcomes such as ADL.65 instrumental ADL. ● Vol. In contrast.14 m/sec for gait speed and 41. J. The small number of studies that evaluated cardiorespiratory fitness training using nonfunctional approaches. Although there is evidence that the BBS is a responsive tool. 11 . one high-quality study concluded that there were no significant differences in walking quality between the control and experimental groups. there is some evidence that cardiorespiraAm. whereas strong evidence was found for a favorable effect of cardiorespiratory training on stair-climbing performance.70 The significant SES for gait-oriented training programs corresponds to a mean improvement of 0.4. significant effects of gait-oriented training on gait speed and 946 van de Port et al. ADL.2 m on the 6-min walk test.

45 Because it is Exercise Training After Stroke 947 . strong evidence was found that balance. In other words. and they treated these as continuous scales. Recent electroneurophysiological studies in which the EMG activity of the paretic muscles was serially recorded45 and studies recording improvements in standing balance74.71 Because gait speed over a short distance overestimates walking distance in a 6-min walk test. ADL.53 Finally. suggesting that gait and gait-related activities should be directly targeted. IADL. although functional mobility was positively impacted.64 also suggests that.72 one should realize that improving gait speed does not automatically result in improvements in walking distance. However. such as increased anticipatory activation of muscles of the nonparetic leg. This finding supports the general view of motor learning that exercise regimens mainly induce specific treatment effects.7. therapy-induced improvements do not automatically generalize to significant gains in gait performance. because the authors used ordinal scales to assess balance and ADL. or health-related quality of life were not significantly affected by gait-oriented training. these conclusions need to be interpreted with some caution. the training programs need to focus primarily on the relearning of functional gait-related skills that are relevant to the individual patient’s needs.74 or stride lengthening of the nonparetic leg32 during walking.52.47. despite the significant improvement in strength. This November 2007 underscores the fact that training should be task specific. The lack of evidence to support the relationship between strength gains and improvements in walking ability47. tory training negatively affects balance55 and has a positive impact on walking distance. Closer associations have been found with compensatory adaptive changes on the nonparetic side.75 have shown that task-related improvements were poorly related to physiologic gains on the paretic side.FIGURE 3 Summarized effect size of walking distance (mean and 95% CI). reporting means and CIs. there is growing evidence that functional improvements are closely related to the use of compensatory movement strategies in which patients learn to adapt to existing impairments.75 strategies using increased weight bearing above the nonparetic leg during standing. The main finding of the present review is that programs focusing on cardiorespiratory and gaitoriented training are more beneficial in improving walking competency than programs centered on strengthening.73 The mechanisms underlying therapy-induced improvements in gait performance are not yet well understood. In other words.

86. resulting from the use of compensatory or adaptive movement strategies to perform functional tasks such as walking.78 Energy expenditure required to perform routine ambulation is increased approximately 1. 11 . Rehabil. however. these variables can be used to identify those patients who are at risk for mobility decline. No. increased intensity)86—aspects that need further investigation. because function-oriented training is effective in improving walking competency. this effect was not statistically significant.53. Med. This is in accordance with the Cochrane review of Saunders and colleagues. Lai and coworkers84 concluded that depressive symptoms do not restrict gains in functional outcome as a result of physical exercise. because about 20% of all chronic stroke patients show a significant decline in mobility status in the long run. Am. and gait-oriented training we used were arbitrary. This is an important finding for clinical practice.85 Future studies should elucidate whether a functional training program can improve walking competency in patients who are susceptible to a decline in mobility such as the very old. First. the long-term effects of these training interventions need to be investigated. we found that the presence of depressive symptoms.53 These results are in agreement with the findings in the recently conducted review of Pang et al. because balance is highly related to independent gait.85 In other words. CONCLUSION This review shows that gait-oriented training.79 The lower walking speeds observed in patients with hemiparesis (30 m/min) consume approximately the same amount of oxygen (10 ml/kg per minute)80 as healthy people require when walking approximately twice as fast (i. the definitions of strengthening. more RCTs are needed to allow conclusions on the effects of nonspecific cardiorespiratory training on walking competence. Recently. fatigue. longitudinal kinematic and neurophysiologic studies are needed for a better understanding of the underlying mechanisms of functional improvement. those severely compromised. the number of studies investigating energy expenditure after stroke is limited. The moment at which these gait-oriented treatments are introduced seems not to be restricted to a particular phase after stroke or to a particular type of stroke. Although only two studies focusing on the effect of cardiorespiratory fitness interventions (without walking) on gait speed could be included.82 However.77.3 The only study that assessed the effect of cardiorespiratory training on walking distance showed that cardiorespiratory training was beneficial in improving distance walked. targeting improved strength and cardiorespiratory fitness.76 Obviously. Future studies should establish whether the improvements in gait speed and walking distance that have been described are of clinical relevance for independent community ambulation. ● Vol. Another study suggested that persons with severe depressive symptoms may be particularly responsive to therapeutic intervention.still unclear which compensatory characteristics are most closely related to gains in walking competency.77 These high energy demands are frequently associated with less efficient motor control in hemiplegic compared with healthy subjects. In addition. In addition.5 indicated that 948 van de Port et al.0-fold in hemiparetic stroke patients compared with normal control subjects. Further improvement of stroke rehabilitation could be achieved by identifying which patients benefit most from supervised83 physical fitness training programs. the study was subject to certain limitations. current debate has concentrated on whether the critical variable for therapeutic efficacy is task specificity or the intensity of the effort involved in therapeutic activities (increased volume. a positive effect on walking speed was found. the review did not include papers written in languages other than English. and those who are depressed. because progressive bicycling programs resulted in significant gains in walking endurance. The present review also suggests that enhancing walking endurance by improving physical condition seems to be less specific. is the most successful method to improve gait speed and endurance. cardiorespiratory fitness. Although this systematic review aimed at identifying all relevant trials. J.81 However. German. because it has been suggested that the energy costs of walking are substantially higher in people with stroke than in normal individuals. Salbach et al. They also suggested that exercise may help reduce poststroke depressive symptoms.22 Recently. increased level of participation. In addition.45 Progression in training programs seems to be an important aspect of improving walking endurance.5.5 The fact that balance is also improved by cardiorespiratory training might also suggest that it would be beneficial in improving gait speed and walking distance. 60 m/min). reduced cognitive status. most effects were gained in the group of patients with a moderate walking deficit. Phys. and an inactive lifestyle are important factors related to a gradual decline in mobility over time. improving aerobic capacity as a reflection of physical condition is an important factor in restoring walking competency. or Dutch or studies focusing on body weight–support treadmill-training 2..e.

5:223–6 12. New York.18:249–52 22.81:995–1005 31. Haddock CK: Combining estimates of effect size. Mu¨ller F. Richards L. Weinandy-Trapp M. Pedro-Cuesta J. 1994. and Prevention. Collen FM. Macko RF. Cochrane Database Syst Rev 2004. Recommendations on stroke prevention. J Rehabil Res Dev 2004. controlled trial. Giakas G: An exploration of the effects of weighted garments on balance and gait of stroke patients with residual disability. Green J.17:765–76 32. J Am Geriatr Soc 2005. Richards CL. Aruin AS. Wade DT. N Engl J Med 2005. Rowe PJ. Raudenbush S: Random effects model. 1994. Stroke 2003. BMJ 1992. Nakayama H. Mayo NE. Sherrington C. Moseley AM.18:30–6 20. Wagenaar RC: Effect of duration of upper. Cohen J: Statistical Power Analysis for the Behavioral Sciences.359:199– 203 10. and Metabolism. Russell Sage Foundation. Young J: Physiotherapy for patients with mobility problems more than 1 year after stroke: a randomised controlled trial. Dekker J: The impact of physical therapy on functional outcomes after stroke: what’s the evidence? Clin Rehabil 2004. Herbert RD. Buckley EL. van den Ende CH: Occupational therapy for stroke patients: a systematic review.72:781–9 21. Smith GV.83:713–21 30. Hill E. randomised trial. REFERENCES 1.53:576–82 23. Int J Rehabil Res 2003. diagnosis. Clin Rehabil 1999. and therapy. von Koch L. Arch Phys Med Rehabil 1995. Warlow CP: Physiotherapy intervention late after stroke and mobility. Dekker J.16: 473–80 35. Duncan PW. Aust J Physiother 2004. Almazan J. 1977 16. Elkins M: Reliability of the PEDro scale for rating quality of randomized controlled trials. Forster A. Gulanick M. Maletsky R. Green G: Shared responsibility for ongoing rehabilitation: a new approach to home-based therapy after stroke.CD003316 4.15:390–7 34. Robb GF. Russell Sage Foundation.76:27–32 2. Pierson-Carey CD. Clin Rehabil 2004. and the Stroke Council. Brown DA. Goldberg AP.12:131–8 18. Phys Ther 1992. Forrester LW: Task-oriented aerobic exercise in chronic hemiparetic stroke: training protocols and treatment effects.5:198–202 24. Salbach NM. Scheidtmann K. Clin Rehabil 2002. McClellan R. Broad JB. van de Nes JC. Kostulas V. Wood-Dauphinee S. Greig CA. Pollock AS. Widen HL: Randomized controlled trial of rehabilitation at home after stroke: one-year follow-up of patient outcome. Malouin F. Allen JB.13: 23–33 9. in Cooper HM. Green J. the Council on Cardiovascular Nursing.18:833–62 19. Baskett JJ. New York. Subcommittee on Exercise. Wood-Dauphinee S: The effect of a task-oriented walking intervention on improving balance self-efficacy poststroke: a randomized. Kwakkel G. Jaffe DL. Forster A. Jorgensen HS. Silver KH. Hedges LV (eds): The Handbook of Research Synthesis. Clin Rehabil 2004. A systematic review within the framework of the Cochrane Collaboration Back Review Group. Maher CG. Hendriks HJ. Kwakkel G. Richards CL.82:432–48 33. Rehabilitation after stroke. Perera S. Cote R: A task-orientated intervention enhances walking distance and speed in the first year post stroke: a randomized controlled trial. Jones D. Higgins JP. Hicks RR: Balance and mobility following stroke: effects of physical therapy interventions with and without biofeedback/forceplate training. Stroke 2003. Koes BW: The effectiveness of acupuncture in the management of acute and chronic low back pain. Lancet 2002. Young A. 41:283–92 11.12:45–57 7. Phys Ther 2002. Altman DG: Measuring inconsistency in meta-analyses. Bouter LM.34:2173–80 5. BMJ 2003. Duncan P. Salbach NM. Deeks JJ.304: 609–13 November 2007 Exercise Training After Stroke 949 . Mead GE: Physical fitness training for stroke patients. van Tulder MW. Stroke—1989. Stroke 2004. Cherkin DC. Cardiac Rehabilitation. Raaschou HO. Top Stroke Rehabil 2005. Pomeroy VM. Man Ther 2000. et al: Randomized clinical trial of therapeutic exercise in subacute stroke. Lao L. Studenski S: Gains in distance walking in a 3-month follow-up poststroke: what changes? Neurorehabil Neural Repair 2004. Clin Rehabil 2004. Koenig E: Treadmill training in early poststroke patients-do timing and walking ability matter? Neurol Rehabil 1999. Durward BR. in Cooper HM. Jongbloed L: Prediction of function after stroke: a critical review. Stroke 1986. Phys Ther 2003. Olsen TS: Recovery of walking function in stroke patients: the Copenhagen Stroke Study. Hedges LV (eds): The Handbook of Research Synthesis.34:676–87 37. Cup EH. Can augmented physiotherapy input enhance recovery of mobility after stroke? A randomized controlled trial.24:1113–23 36. Bogle S. Spine 1999. Lew HL: Stepping over obstacles to improve walking in individuals with poststroke hemiplegia. Report of the WHO Task Force on Stroke and other Cerebrovascular Disorders.26:309–12 8. resource-efficient mobility program after discharge from rehabilitation improves standing in people affected by stroke: placebo-controlled. Collen F. resource use and cost. Young J.327:557–60 28. Maher CG. Wade D: Combined analysis of two randomized trials of community physiotherapy for patients more than one year post stroke. Hocking C. pp 261–82 14. Potvin M. Phys Ther 2001. Wood-Dauphinee S. Steultjens EM. Reekie G. Ada L: A six-week. Hanke TA.30:2112–8 17. Robichaud-Ekstrand S. 1994. Studenski S. Sharma A: Base of support feedback in gait rehabilitation. 301–23 15.352:1677–84 3. Hanley JA. Saunders DH. Russell Sage Foundation. Evans B. Whitman R. Hedges LV: Fixed effects model. Wood-Dauphinee S: Use of an intensive task-oriented gait training program in a series of patients with acute cerebrovascular accidents. Stroke 1989. Costa F. Hedges LV (eds): The Handbook of Research Synthesis. Geiger RA. Van der Wees PJ.18:529–37 26. Van Peppen RP. Falconer M. O’Keefe J. Mayo NE. Ivey FM. Moseley AM: PEDro. Richards CL. 20:1407–31 29. Sherrington C. Cerebrovasc Dis 2001. pp 285–300 13.35: 1230–40 27. Macko RF: “Task-oriented” exercise improves hamstring strength and spastic reflexes in chronic stroke patients. Clin Rehabil 2001.ACKNOWLEDGMENTS We wish to thank Wieteke Ermers (WE) from the University of Maastricht and Hans Ket from the VU Medical Library for the literature search. Academic Press. Dobkin BH: Clinical practice.18: 509–19 6. Hanley JA. Pohl PS. New York. Physical Activity. in Cooper HM. New York.and lower-extremity rehabilitation sessions and walking speed on recovery of interlimb coordination in hemiplegic gait. A database of randomized trials and systematic reviews in physiotherapy. Thompson SG. the Council on Nutrition. Wulf D.50:163–7 25. et al: Physical activity and exercise recommendations for stroke survivors: an American Heart Association scientific statement from the Council on Clinical Cardiology. Paul JP: The effect of independent practice of motor tasks by stroke patients: a pilot randomized controlled trial. Shadish WR. Brunner H. Prevost J. Herbert RD. Gordon NF. Berman B. Stroke 1999.

Pellegrino L. Richards L.74:612–20 66. Perera S. McKay HA.18:199–211 42. Carmeli E. Dumas F. self-reported function.85:886–95 75. 86. Mehrholz J.27:27–36 68. controlled pilot trial. Macko RF. Lindsley HG. Stroke 2005. Williams JI. Eng JJ. Lieu FK: Gait outcomes after additional backward walking training in patients with stroke: a randomized controlled trial. Huidekoper SC. Savastano G. Morris SL.18:27–39 74. Yang YR. Arch Phys Med Rehabil 2003. Bampton J. Gylfadottir S: Exercise leads to faster postural reflexes. Stroke 1998. et al: A randomized. controlled trial. A preliminary report. Yang YR: Efficacy of Bobath versus orthopaedic approach on impairment and function at different motor recovery stages after stroke: a randomized controlled study. Stroke 2002. Aust J Physiother 2001.10:35–50 70. Eng JJ. Yen LL. Bean JF. Richards CL. Brouwer BJ. Crompton S: A treadmill and overground walking program improves walking in persons residing in the community after stroke: a 950 placebo-controlled. Wood-Dauphinee S.81:409–17 59. Stanghelle JK: Bobath or motor relearning programme? A comparison of two different approaches of physiotherapy in stroke rehabilitation: a randomized controlled study.38:69–78 44. Chen HI. Clin Rehabil 2005. Studenski SA: Coordination of hemiparetic locomotion after stroke rehabilitation. Beaudoin N. Harris JE. J. Moreland JD. Glasser L: Effects of isokinetic training on the rate of movement during ambulation in hemiparetic patients. Ouellette MM. Wang RY. Wood-Dauphinee S: The role of technology in task-oriented training in persons with subacute stroke: a randomized controlled trial. Marigold DS. Ivey FM. Fasotti L. Dawson AS. Berg K.250:1206–8 61. Clin Rehabil 2005.84:1609–14 54. Mach H. Malouin F. Clin Rehabil 2004. Garland SJ. Scand J Rehabil Med 1995. Phys Ther 1986. Musser L. Berg K. Walker C. Williams JI: The Balance Scale: responsiveness to clinically meaningful changes. Di Lauro A. J Am Geriatr Soc 2005.84:1433–40 52. Am J Phys Med Rehabil 2002. Wood-Dauphinee S. Stevenson TJ: Detecting change in patients with stroke using the Berg Balance Scale.38. Med. Richards CL. DeSouza CA. Willems DA. Dite W: Additional task-related practice improves mobility and upper limb function early after stroke: a randomised controlled trial.10:265–73 48. No. Hann AW. et al: Progressive resistance strengthening exercises after stroke: a single-blind randomized controlled trial. Dickstein R.28:326–30 69. Arch Phys Med Rehabil 1999. Ada L. et al: High-intensity resistance training improves muscle strength. Malouin F. 11 . Marcovitz E: The effect of treadmill training on the ambulation of stroke survivors in the early stages of rehabilitation: a randomized study. Gylfadottir S: Water-based exercise for cardiovascular fitness in people with chronic stroke: a randomized controlled trial. Age Ageing 2000. Phys Ther 2000. Duncan PW. Liston R. Nadeau S. Wood-Dauphinee S. et al: Treadmill aerobic exercise training reduces the energy expenditure and cardiovascular demands of hemiparetic gait in chronic stroke patients. and fewer falls in older persons with chronic stroke. Arch Phys Med Rehabil 2004. Chefez Y. randomized trial. Restor Neurol Neurosci 2004. Pohl M. and disability in long-term stroke survivors. Dodd KJ. Brouwer B: Muscle strengthening and physical conditioning to reduce impairment and disability in chronic stroke survivors.36:2206–11 64. Ritschel C. Goldsmith CH. Shochina M: The effect of early aerobic training on independence six months post stroke.29: 2055–60 60. Pang MY. Teixeira-Salmela LF. Katz-Leurer M.19:155–64 40.17:735–41 55. Wallace D.19:250–8 46. Richards CL. Chu KS. Macko RF. Neurorehabil Neural Repair 2004. Can J Rehabil 1997. Dean CM. Bourbonnais D. Eng JJ.53: 416–23 62. Williams JI: The Balance Scale: reliability assessment with elderly residents and patients with an acute stroke. Tretter LD.80:1211–8 67. Richards CL. Harris B. Kim CM. Arch Phys Med Rehabil 2003. Rehabil.22:281–99 72. Dean CM.50:219–24 58. Kollen B. Hesse S: Aerobic treadmill plus Bobath walking training improves walking in subacute stroke: a randomized controlled trial.33: 553–8 65.80:886–95 39. Eich HJ. Chen CY. Tallis RC: Conventional physiotherapy and treadmill re-training for higher-level gait disorders in cerebrovascular disease. Blennerhassett J. Aust J Physiother 2004. Hall JM.14:361–9 57. controlled trial. Clin Rehabil 2001. Shochina M. improved balance and mobility. Harris JE. Mickelborough J. Arch Phys Med Rehabil 2000. Katz-Leurer M. Geurts AC. Malouin F: Task-related circuit training improves performance of locomotor tasks in chronic stroke: a randomized. van Limbeek J: Recovery of standing balance in postacute stroke patients: a rehabilitation cohort study.15:415–21 73. Forrester LW. Bilodeau S. Neurol Rep 1994. Huijbregts MP. Eng JJ. Clin Rehabil 2004. Bravo G. et al : A randomized trial on the efficacy of intensive rehabilitation in the acute phase of ischemic stroke. Stroke 2004. Langhammer B. Clin Rehabil 2000.81:890–7 51.19: 29–34 49.29:311–8 45. Bouchard JP. et al: Treadmill exercise rehabilitation improves ambulatory function and cardiovascular fitness in patients with chronic stroke: a randomized. Bravo G. Inglis JT. Laufer Y.53:1667–74 50.47: 29–38 71. ● Vol. Kautz SA. Neptune RR. Dawson AS: Effects of isokinetic strength training on walking in persons with stroke: A double-blind controlled pilot study. Phys.66:673–6 47. Ruckriem S: Speed-dependent treadmill training in ambulatory hemiparetic stroke patients: a randomized controlled trial. Friedlander Y: The influence of early aerobic training on the functional capacity in patients with cerebrovascular accident at the subacute stage. controlled pilot study of a home-based exercise program for individuals with mild and moderate stroke. 63. Ivanova TD. Macintyre DL. Gravel D. Neurorehabil Neural Repair 2005. Clin Rehabil 2003. J Rehabil Res Dev 2001.35:1404–9 53. Olney SJ. Arch Phys Med Rehabil 1993. Werner C. Carmeli E. Dawson AS. Lindeman E: Understanding the pattern of functional recovery after stroke: facts and theories. de Haart M. Stroke 1997. Kwakkel G.19: 264–73 41. Miller KJ: Recovery of Am. Wang RY. J Stroke Cerebrovasc Dis 2001. J Am Geriatr Soc 2005. Morris ME: Outcomes of progressive resistance strength training following stroke: a systematic review. Ozkaplan A. Malouin F: Walking speed over 10 metres overestimates locomotor capacity after stroke. Lepage Y. LeBrasseur NK. Arch Phys Med Rehabil 2003. Harris JE: A community-based fitness and mobility exercise program for older adults with chronic stroke: a randomized. Dean CM. Brunet D: Task-specific physical therapy for optimization of gait recovery in acute stroke patients.18: 640–51 43. Culham EG: Use of visual feedback in retraining balance following acute stroke.85:870–4 56. Yen JG. Duncan P. Forget R: Effect of force-feedback treatments in patients with chronic motor deficits after a stroke. Steward MR: The effects of Kinetron training on gait patterns with strokes.84:1486–91 van de Port et al. Dawson AS. J Neurol 2003. Arch Phys Med Rehabil 2004.

3:85–8 Hash D: Energetics of wheelchair propulsion and walking in stroke patients. 78. Monga TN. Monga T. Gait Posture 1999.36:2676–80 83. Orr W: External work of walking in hemiparetic patients. Lindeman E: Susceptibility to deterioration of mobility long-term after stroke: a prospective cohort study. Eng JJ. Griffin MP. Qureshy H. method.37: 476–81 84. 77. Richards L.41:293–312 Exercise Training After Stroke 951 . 79. van Wijk I. Lai SM. Protas EJ: Gait outcomes after acute stroke rehabilitation with supported treadmill ambulation training: a randomized controlled pilot study. 83:1258–65 Olney SJ. Kollen B. Van de Port I. et al: A randomized controlled trial of supervised versus unsupervised exercise programs for ambulatory stroke survivors. Kwakkel G: Predicting improvement in gait after stroke: a longitudinal prospective study.54:240–7 85. van de Port I. Arch Phys Med Rehabil 1991. Orthop Clin North Am 1978. Brouwer B. Lim PA. Studenski S. Lindeman E. Kwakkel G. Olney SJ.37:167–71 86. Twisk J. Gylfadottir S: The use of aerobic exercise training in improving aeobic capacity in individuals with stroke: a meta-analysis. Arch Phys Med Rehabil 2003. J Rehabil Res Dev 2004. Waters RL. Scand J Rehabil Med 1971. Stroke 2006. 20:97–111 da Cunha IT Jr. Mulroy S: The energy expenditure of normal and pathologic gait. et al: Therapeutic exercise and depressive symptoms after stroke. J Am Geriatr Soc 2006. McBride ID: Work and power in gait of stroke patients. Dawson AS.84:1753–9 Pang MYC. Stroke 2006. Patten C. Arch Phys Med Rehabil 2002. Brown HE: Weakness and strength training in persons with poststroke hemiplegia: rationale.9:207–31 82. Nymark J. 80. Lexell J. standing balance and functional mobility after stroke. and efficacy.76. Henson H. Clin Rehabil 2006.9:372–4 November 2007 81.72:309–14 Gerson J. Stroke 2005.