Clinical Rehabilitation http://cre.sagepub.com/
The effect of whole body vibration on balance, gait performance and mobility in people with stroke: A systematic review and meta-analysis Xiaotian Yang, Pu Wang, Chuan Liu, Chengqi He and Jan D Reinhardt Clin Rehabil published online 13 October 2014 DOI: 10.1177/0269215514552829 The online version of this article can be found at: http://cre.sagepub.com/content/early/2014/10/08/0269215514552829
Published by: http://www.sagepublications.com
Additional services and information for Clinical Rehabilitation can be found at: Email Alerts: http://cre.sagepub.com/cgi/alerts Subscriptions: http://cre.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Oct 13, 2014 What is This?
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
552829
research-article2014
CRE0010.1177/0269215514552829Clinical RehabilitationYang et al.
CLINICAL REHABILITATION
Article
The effect of whole body vibration on balance, gait performance and mobility in people with stroke: A systematic review and meta-analysis
Clinical Rehabilitation 1–13 © The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0269215514552829 cre.sagepub.com
Xiaotian Yang*1,2, Pu Wang*1,2, Chuan Liu1,2, Chengqi He1,2 and Jan D Reinhardt3
Abstract Objective: To examine the effect of whole body vibration on balance, gait performance and mobility among people with stroke. Method: A systematic review was conducted by two independent reviewers who completed the article search and selection. We included randomized controlled trials published in English examining effects of whole body vibration on balance, gait, mobility, muscle strength and muscle tone in adults with a clinical diagnosis of stroke. Articles were excluded if they were research studies on people with other primary diagnosis, abstracts published in the conferences or books. The Cochrane risk of bias tool was used to assess the methodological quality of the selected studies. Data source: Sources included Cochrane Central Register of Controlled Trials, Pubmed, MEDLINE, CINAHL, EMBASE, PEDro, PsycINFO, Science Citation Index, ClinicalTrials.gov, Current Controlled Trials, Stroke Trials Registry, and reference lists of all relevant articles. Result: Eight randomized controlled trials (nine articles) involving 271 participants were included in this meta-analysis. No significant improvement was found regarding Berg balance scale (SMD=-0.08, 95%CI=1.35 to 1.19, P=0.91), mobility (SMD=0.45, 95%CI=-0.46 to 1.37, P=0.33), maximal isometric contracion of knee extension strength (SMD=0.23, 95%CI=-0.27 to 0.74, P=0.36), and maximal isometric contracion of knee extension strength (SMD=0.09, 95%CI=-0.38 to 0.56, P=0.71). Conclusion: There was no evidence for effects of whole body vibration on balance in people with stroke. Effects of whole body vibration on mobility and gait performance remain inconclusive. More large and high-quality trials are required.
1Department
of Rehabilitation Medicine, West China Hospital, Sichuan University, Chengdu, Sichuan Province, People’s Republic of China 2Key Laboratory of Rehabilitation Medicine in Sichuan, Chengdu, Sichuan Province, People’s Republic of China 3Institute for Disaster Management and Reconstruction of Sichuan University and Hong Kong Polytechnic University, Chengdu, Sichuan Province, People’s Republic of China
*These
authors contributed equally to this study.
Corresponding author: Chengqi He, Department of Rehabilitation Medicine, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan Province, People’s Republic of China. Email: [email protected]
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
2
Clinical Rehabilitation
Keywords Stroke, mobility, vibration, meta-analysis Received: 8 March 2014; accepted: 27 August 2014
Introduction Limitations of mobility after stroke are very common, and current treatment is only partially successful at reducing this. Whole-body vibration (WBV) therapy has been introduced recently, but its benefits are not yet well established with good evidence. WBV therapy generates a vertical oscillation or a horizontal movement upon the individual placed on the platform. The contact surface of the platform transmits a vibration stimulus from the feet to the rest of the body.1 WBV is supposed to influence postural control through activating the Ia and II afferents of muscle groups2 and inducing sensory stimulation of foot-sole afferents.3 Vibration activates themyotatic reflex and produces rapid change in muscle length.4 Additionally, there is some preliminary evidence that WBV can improve proprioceptive function5 by detecting the stretching of muscles, which activate tonic vibration reflex.4 Thus, WBV training may have potential benefits on reducing disability through improving balance, gait performance, and mobility in people with stroke. Although there is a number of studies related to effects of WBV therapy on functional recovery in stroke patients, evidence is still inconclusive.6–14 The aim of this review is to systematically assess randomized controlled trials featuring WBV therapy in people with stroke regarding effects on function and activity, eapecially balance, gait performance and mobility.
Methods Two authors independently (XTY and PW) searched for relevant literature in the following electronic databases: Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, 2013 Issue 4), MEDLINE (1950 through November 2013; via Ovid), Pubmed (1966 through
November 2013), CINAHL (1982 through November 2013; via Ovid), EMBASE (1988 through November 2013; via Ovid), PEDro (1929 to November 2013; via website) and PsycINFO (1806 to November 2013; via Ovid).We applied language restrictions in English. We developed the search strategy for MEDLINE (via Ovid) (online supplementary material Appendix 1) and adapted this strategy for the other databases. Keywords included: vibration, whole body vibration, WBV, biomechanical stimulation, hemiplegia, haemorrhage, hemorrhage, cerebrovascular disorders, brain ischemia, stroke, poststroke, brain infarction, paresis, basal ganglia cerebrovascular disease, carotid artery diseases, intracranial arterial diseases, intracranial embolism and thrombosis, intracranial hemorrhages, vertebral artery dissection, randomized controlled trials, clinical trial, controlled clinical trial, trial, randomized, randomly, placebo. We also handsearched reference lists of all relevant articles and the following international trial registers: ClinicalTrials.gov (http://www.clinicaltrials.gov/); Current Controlled Trials (http://www. Controlled–trials.com); Stroke Trials Registry (http://www.strokecenter.org/trials/) and WHO International ClinicalTrials Registry Platform (http://www.who.int/ictrp/en/). The latest search was performed on 8th November 2013. Inclusion criteria were as follows: randomized controlled trials (RCT); published in English; adults with a clinical diagnosis of stroke; WBV was given aimed at improving function; control interventions included sham treatment, other exercise programmes or conventional treatment. Articles were excluded if they were: studies on patients with other primary diagnosis (e.g. Alzheimer’s and Parkinson’s diseases); conference abstracts or commentaries.
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
3
Yang et al. From the perspective of rehabilitation, effective interventions should not only reduce impairments, but also improve activity and participation. We focused on used commonly outcome measures reflecting function and activity in people with stroke according to the International Classification of Functioning, Disability and Health.15,16 We defined primary outcomes as measures of: (1) Balance (Berg balance scale). (2) Gait performance (e.g. Timed up and go test, comfortable gait speed, fast gait speed, tinetti gait test). (3) Mobility (e.g. Six-minute walk test, 10 metre walk test, Rivermead Mobility Index). We defined secondary outcomes as measures of: (1) Spasticity (Modified Ashworth Scale). (2) Muscle strength (e.g. Maximum isometric and eccentric torque, rate of torquedevelopment, root-mean-squared electromyography, median frequency of vastuslateralis, co-activation of knee flexors, the Motricity Index). (3) Adverse events. Two authors (XTT and CL) independently assessed each study’s risk of bias according to the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions.17 We evaluated the following domains: random sequence generation (selection bias); allocation concealment (selection bias); blinding of participants, personnel and outcome assessors (performance bias and detection bias); selective reporting (reporting bias); incomplete outcome data(reporting bias) and other bias. We assigned a judgement of low, unclear or high risk of bias based on the information published in each study. Reviewers were not blinded to authors, institutions, or journal of included studies. When disagreement existed between two reviewers, a third reviewer made the final decision. If necessary, we contacted each study author to obtain complete data for consummating ‘risk of bias’ assessment. Meta-analysis regarding effects of WBV was performed if two or more studies used the same outcome measure. We extracted means and standard
deviations for each outcome from each treatment group. For the combined data weighted mean difference (WMD), standardized mean difference (SMD) and 95% confidence interval (CI) were computed. Review Manager 5.2 software (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) was used for all analyses. The results of meta-analysis are presented using forest plots. We quantified the heterogeneity of the included studies using the I2 statistic, which describes the percentage of variation across studies due to heterogeneity. Fixed effects model should be used to combine studies when I2 value was lower than 50%.17 Otherwise, random effects model would be used. Publication bias was not investigated with funnel plots as there were fewer than 10 studies in the meta-analysis because test power is usually too low to distinguish change from real asymmetry.18
Results Our intial search identified a total of 191 articlesbut 56 records had to be removed due to duplication (Figure 1). After screening titles and abstracts, we obtained 12 full text articles. Of those, two studies did not allocated subjects based on randomization.19, 20 Reports written by Lau et al.8 and Pang et al.10 were based on the identical data. Only nine articles (eight trials) met all eligibility criteria and were included in this meta-analysis. These eight trials comprised a total of 271 participants. The start of rehabilitation ranged from several days after stroke to more than six months post stroke. Two studies investigated the effects of WBV on fuction in the subacute phase of stroke.13,14 Control intervention types included no treatment, sham vibration, routine physiotherapy and exercise therapy on music (ETM). The stroke type of all trials included both ischaemic stroke and intracranial haemorrhage. Two trials7,13 studied the short-term effects of WBV after a single session, whereas six trials studied the long-term effects of WBV (up to nine weeks training programme) (Table 1). The assessments of the methodological quality are provided in Figure 2 and Figure 3 (online supplementary material). Only one study6 reported an
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
4
Clinical Rehabilitation
189 records identified through database searching
2 records identified through other sources
56 records removed due to duplication 135 records screened 123 articles excluded after reading the titles and abstracts 12 full articles assessed for eligibility 3 articles excluded. The study was not RCT (n=2) WBV was not inlcuded as treatment (n=1) 9 articles (8 trials) included in this metaanalysis
Figure 1. Flow diagram. Nine articles (eight trials) were included in this meta-analysis.
adequate method for blinding of participants and personnel. Four studies6, 8, 9, 14 including a total of 186 participants assessed balance with the Berg balance scale. We used the raw data directly under the condition of unbalanced baseline.6 The pooled mean difference for balance was -0.08 (95%CI=-1.35 to 1.19, Z=0.12, P=0.91) without evidence for statistical heterogeneity (I2=0%). No differences were observed between WBV as compared to control (Figure 4A). When we excluded the study6 with unbalanced baseline data, the result did not change (Figure 4B). In addition to post-intervention data, two trials8,14 also reported follow-up data. This allowed us to perform a sensitivity analysis based on the different time points (Figure 4C). However, the sensitivity analysis still keep the same result with the primary analysis. Figure 5 summarises the results for WBV effects on gait performance measured with the Timed Up
and Go test (TUG).6,7 Significant differences in one outcome measure: TUG (P
The effect of whole body vibration on balance, gait performance and mobility in people with stroke: A systematic review and meta-analysis Xiaotian Yang, Pu Wang, Chuan Liu, Chengqi He and Jan D Reinhardt Clin Rehabil published online 13 October 2014 DOI: 10.1177/0269215514552829 The online version of this article can be found at: http://cre.sagepub.com/content/early/2014/10/08/0269215514552829
Published by: http://www.sagepublications.com
Additional services and information for Clinical Rehabilitation can be found at: Email Alerts: http://cre.sagepub.com/cgi/alerts Subscriptions: http://cre.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Oct 13, 2014 What is This?
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
552829
research-article2014
CRE0010.1177/0269215514552829Clinical RehabilitationYang et al.
CLINICAL REHABILITATION
Article
The effect of whole body vibration on balance, gait performance and mobility in people with stroke: A systematic review and meta-analysis
Clinical Rehabilitation 1–13 © The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0269215514552829 cre.sagepub.com
Xiaotian Yang*1,2, Pu Wang*1,2, Chuan Liu1,2, Chengqi He1,2 and Jan D Reinhardt3
Abstract Objective: To examine the effect of whole body vibration on balance, gait performance and mobility among people with stroke. Method: A systematic review was conducted by two independent reviewers who completed the article search and selection. We included randomized controlled trials published in English examining effects of whole body vibration on balance, gait, mobility, muscle strength and muscle tone in adults with a clinical diagnosis of stroke. Articles were excluded if they were research studies on people with other primary diagnosis, abstracts published in the conferences or books. The Cochrane risk of bias tool was used to assess the methodological quality of the selected studies. Data source: Sources included Cochrane Central Register of Controlled Trials, Pubmed, MEDLINE, CINAHL, EMBASE, PEDro, PsycINFO, Science Citation Index, ClinicalTrials.gov, Current Controlled Trials, Stroke Trials Registry, and reference lists of all relevant articles. Result: Eight randomized controlled trials (nine articles) involving 271 participants were included in this meta-analysis. No significant improvement was found regarding Berg balance scale (SMD=-0.08, 95%CI=1.35 to 1.19, P=0.91), mobility (SMD=0.45, 95%CI=-0.46 to 1.37, P=0.33), maximal isometric contracion of knee extension strength (SMD=0.23, 95%CI=-0.27 to 0.74, P=0.36), and maximal isometric contracion of knee extension strength (SMD=0.09, 95%CI=-0.38 to 0.56, P=0.71). Conclusion: There was no evidence for effects of whole body vibration on balance in people with stroke. Effects of whole body vibration on mobility and gait performance remain inconclusive. More large and high-quality trials are required.
1Department
of Rehabilitation Medicine, West China Hospital, Sichuan University, Chengdu, Sichuan Province, People’s Republic of China 2Key Laboratory of Rehabilitation Medicine in Sichuan, Chengdu, Sichuan Province, People’s Republic of China 3Institute for Disaster Management and Reconstruction of Sichuan University and Hong Kong Polytechnic University, Chengdu, Sichuan Province, People’s Republic of China
*These
authors contributed equally to this study.
Corresponding author: Chengqi He, Department of Rehabilitation Medicine, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan Province, People’s Republic of China. Email: [email protected]
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
2
Clinical Rehabilitation
Keywords Stroke, mobility, vibration, meta-analysis Received: 8 March 2014; accepted: 27 August 2014
Introduction Limitations of mobility after stroke are very common, and current treatment is only partially successful at reducing this. Whole-body vibration (WBV) therapy has been introduced recently, but its benefits are not yet well established with good evidence. WBV therapy generates a vertical oscillation or a horizontal movement upon the individual placed on the platform. The contact surface of the platform transmits a vibration stimulus from the feet to the rest of the body.1 WBV is supposed to influence postural control through activating the Ia and II afferents of muscle groups2 and inducing sensory stimulation of foot-sole afferents.3 Vibration activates themyotatic reflex and produces rapid change in muscle length.4 Additionally, there is some preliminary evidence that WBV can improve proprioceptive function5 by detecting the stretching of muscles, which activate tonic vibration reflex.4 Thus, WBV training may have potential benefits on reducing disability through improving balance, gait performance, and mobility in people with stroke. Although there is a number of studies related to effects of WBV therapy on functional recovery in stroke patients, evidence is still inconclusive.6–14 The aim of this review is to systematically assess randomized controlled trials featuring WBV therapy in people with stroke regarding effects on function and activity, eapecially balance, gait performance and mobility.
Methods Two authors independently (XTY and PW) searched for relevant literature in the following electronic databases: Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, 2013 Issue 4), MEDLINE (1950 through November 2013; via Ovid), Pubmed (1966 through
November 2013), CINAHL (1982 through November 2013; via Ovid), EMBASE (1988 through November 2013; via Ovid), PEDro (1929 to November 2013; via website) and PsycINFO (1806 to November 2013; via Ovid).We applied language restrictions in English. We developed the search strategy for MEDLINE (via Ovid) (online supplementary material Appendix 1) and adapted this strategy for the other databases. Keywords included: vibration, whole body vibration, WBV, biomechanical stimulation, hemiplegia, haemorrhage, hemorrhage, cerebrovascular disorders, brain ischemia, stroke, poststroke, brain infarction, paresis, basal ganglia cerebrovascular disease, carotid artery diseases, intracranial arterial diseases, intracranial embolism and thrombosis, intracranial hemorrhages, vertebral artery dissection, randomized controlled trials, clinical trial, controlled clinical trial, trial, randomized, randomly, placebo. We also handsearched reference lists of all relevant articles and the following international trial registers: ClinicalTrials.gov (http://www.clinicaltrials.gov/); Current Controlled Trials (http://www. Controlled–trials.com); Stroke Trials Registry (http://www.strokecenter.org/trials/) and WHO International ClinicalTrials Registry Platform (http://www.who.int/ictrp/en/). The latest search was performed on 8th November 2013. Inclusion criteria were as follows: randomized controlled trials (RCT); published in English; adults with a clinical diagnosis of stroke; WBV was given aimed at improving function; control interventions included sham treatment, other exercise programmes or conventional treatment. Articles were excluded if they were: studies on patients with other primary diagnosis (e.g. Alzheimer’s and Parkinson’s diseases); conference abstracts or commentaries.
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
3
Yang et al. From the perspective of rehabilitation, effective interventions should not only reduce impairments, but also improve activity and participation. We focused on used commonly outcome measures reflecting function and activity in people with stroke according to the International Classification of Functioning, Disability and Health.15,16 We defined primary outcomes as measures of: (1) Balance (Berg balance scale). (2) Gait performance (e.g. Timed up and go test, comfortable gait speed, fast gait speed, tinetti gait test). (3) Mobility (e.g. Six-minute walk test, 10 metre walk test, Rivermead Mobility Index). We defined secondary outcomes as measures of: (1) Spasticity (Modified Ashworth Scale). (2) Muscle strength (e.g. Maximum isometric and eccentric torque, rate of torquedevelopment, root-mean-squared electromyography, median frequency of vastuslateralis, co-activation of knee flexors, the Motricity Index). (3) Adverse events. Two authors (XTT and CL) independently assessed each study’s risk of bias according to the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions.17 We evaluated the following domains: random sequence generation (selection bias); allocation concealment (selection bias); blinding of participants, personnel and outcome assessors (performance bias and detection bias); selective reporting (reporting bias); incomplete outcome data(reporting bias) and other bias. We assigned a judgement of low, unclear or high risk of bias based on the information published in each study. Reviewers were not blinded to authors, institutions, or journal of included studies. When disagreement existed between two reviewers, a third reviewer made the final decision. If necessary, we contacted each study author to obtain complete data for consummating ‘risk of bias’ assessment. Meta-analysis regarding effects of WBV was performed if two or more studies used the same outcome measure. We extracted means and standard
deviations for each outcome from each treatment group. For the combined data weighted mean difference (WMD), standardized mean difference (SMD) and 95% confidence interval (CI) were computed. Review Manager 5.2 software (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) was used for all analyses. The results of meta-analysis are presented using forest plots. We quantified the heterogeneity of the included studies using the I2 statistic, which describes the percentage of variation across studies due to heterogeneity. Fixed effects model should be used to combine studies when I2 value was lower than 50%.17 Otherwise, random effects model would be used. Publication bias was not investigated with funnel plots as there were fewer than 10 studies in the meta-analysis because test power is usually too low to distinguish change from real asymmetry.18
Results Our intial search identified a total of 191 articlesbut 56 records had to be removed due to duplication (Figure 1). After screening titles and abstracts, we obtained 12 full text articles. Of those, two studies did not allocated subjects based on randomization.19, 20 Reports written by Lau et al.8 and Pang et al.10 were based on the identical data. Only nine articles (eight trials) met all eligibility criteria and were included in this meta-analysis. These eight trials comprised a total of 271 participants. The start of rehabilitation ranged from several days after stroke to more than six months post stroke. Two studies investigated the effects of WBV on fuction in the subacute phase of stroke.13,14 Control intervention types included no treatment, sham vibration, routine physiotherapy and exercise therapy on music (ETM). The stroke type of all trials included both ischaemic stroke and intracranial haemorrhage. Two trials7,13 studied the short-term effects of WBV after a single session, whereas six trials studied the long-term effects of WBV (up to nine weeks training programme) (Table 1). The assessments of the methodological quality are provided in Figure 2 and Figure 3 (online supplementary material). Only one study6 reported an
Downloaded from cre.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 15, 2014
4
Clinical Rehabilitation
189 records identified through database searching
2 records identified through other sources
56 records removed due to duplication 135 records screened 123 articles excluded after reading the titles and abstracts 12 full articles assessed for eligibility 3 articles excluded. The study was not RCT (n=2) WBV was not inlcuded as treatment (n=1) 9 articles (8 trials) included in this metaanalysis
Figure 1. Flow diagram. Nine articles (eight trials) were included in this meta-analysis.
adequate method for blinding of participants and personnel. Four studies6, 8, 9, 14 including a total of 186 participants assessed balance with the Berg balance scale. We used the raw data directly under the condition of unbalanced baseline.6 The pooled mean difference for balance was -0.08 (95%CI=-1.35 to 1.19, Z=0.12, P=0.91) without evidence for statistical heterogeneity (I2=0%). No differences were observed between WBV as compared to control (Figure 4A). When we excluded the study6 with unbalanced baseline data, the result did not change (Figure 4B). In addition to post-intervention data, two trials8,14 also reported follow-up data. This allowed us to perform a sensitivity analysis based on the different time points (Figure 4C). However, the sensitivity analysis still keep the same result with the primary analysis. Figure 5 summarises the results for WBV effects on gait performance measured with the Timed Up
and Go test (TUG).6,7 Significant differences in one outcome measure: TUG (P
