Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Protocol - Intervention

Tai Chi for improving recovery after stroke

Authors' declarations of interest

Version published: 18 March 2018 Version history

https://doi.org/10.1002/14651858.CD010207.pub3
Collapse all Expand all

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To evaluate the effectiveness of Tai Chi on dependency and motor function and its safety for recovery of people after stroke.

Background

Description of the condition

Stroke is caused by the interruption of the blood supply to the brain. There are two main causes of stroke: hemorrhagic stroke caused by a ruptured blood vessel inside brain, and ischemic stroke caused by vascular insufficiency (such as cerebrovascular thromboembolism). About 80% of strokes are due to ischemic cerebral infarction and 20% are due to brain hemorrhage. It is characterized by rapidly developing clinical symptoms and signs of focal and, at times, global loss of cerebral function lasting over 24 hours or leading to death (Hatano 1976; WHO 2012). Stroke is a worldwide problem: in 2010 the absolute numbers of people with first stroke (16.9 million), stroke survivors (33 million), and stroke‐related deaths (5.9 million) had significantly increased since 1990 (68%, 84%, and 26% increase, respectively), with most of the disease burden in low‐ and middle‐income countries (Feigin 2014). Stroke can occur at any age, but half of all strokes occur in people aged over 70 years (Bamford 1988). With the worldwide population ageing, stroke has been ranked as the third leading cause of death in developed countries and the second leading cause of death in developing countries, and it is also a major cause of disability in adults (Doyle 2010; New Reference).

Of those who survive the acute stroke, about 50% may still experience some level of disability after six months (Wade 1987). It has been reported that among stroke survivors six months after discharge, 50% had one‐sided paralysis, 35% were depressed, 30% were unable to walk without assistance, 25% were dependent in activities of daily living, and 19% were aphasic (Kelly‐Hayes 2003). Family members have to learn rehabilitation techniques and adjust to changes in their relationships with the stroke survivor (Coombs 2007). Not only does it lower the quality of life for patients and their families, it also causes an immense economical burden on society. The estimated direct and indirect cost for stroke reached USD 59,800 to USD 230,000 per patient (Barker‐Collo 2010). In the USA, the costs of stroke reached USD 57.9 billion in 2006 (Doyle 2010), and stroke has become a severe public health problem worldwide.

Description of the intervention

The recovery from stroke is a complex and demanding process (Eilertsen 2010). The deficits created by stroke affect the survivor's wellbeing, and may involve cognition, function, self‐concept, health perception, role change, and relationships (Vanhook 2009). Recovery poststroke is dependent upon cortical reorganization and, therefore, upon the presence of intact cortex, especially in areas adjacent to the infarct (Teasell 2006). It has been proposed that exposure to stimulating and complex environments, and involvement in tasks or activities that are meaningful to the individual with stroke, may increase cortical reorganization and enhance functional recovery (Teasell 2006). Some interventions show promise for improving motor recovery, particularly those that focus on high‐intensity and repetitive task‐specific practice (French 2016; Langhorne 2009).

Tai Chi, also called Tai Chi Chuan or Tai Ji Quan, is a form of Chinese traditional martial art. The term Tai Chi translates as "supreme ultimate fist", "boundless fist", "great extremes boxing", or simply "the ultimate", which reflects the idea about the origin of the cosmos and the mechanism of the world. The theory and practice of Tai Chi evolved in agreement with many Chinese philosophical principles, including those of Taoism and Confucianism. The current practice of Tai Chi has developed into slow, circular, precise, and rhythmic movements carried out with a lowered centre of gravity (i.e. knees and hips held in flexion). It is based on the following three principles.

  1. The body is extended and relaxed; awareness of trunk alignment and deep breathing are prerequisites for achieving appropriate posture before movements can be learned and practiced.

  2. The mind must be alert but calm as one becomes more aware of the presence and movement of the body within its own space.

  3. All body movements require well‐coordinated sequencing of body segments (Wolf 1997).

How the intervention might work

Tai Chi training has been extensively adopted for strengthening body and improving health by flowing circular movements, balance, and weight shifting, breathing techniques, and cognitive tools such as visualization and focused internal awareness (Yeh 2010). The results of previous studies suggest that Tai Chi is a moderate intensity exercise that is aerobic in nature (Field 2011). Increased balance and decreased falls, improved gait, and increased muscle strength and flexibility are the most frequently reported benefits of Tai Chi (Chang 2010; Field 2011). The psychological benefits reported for Tai Chi include improved attentiveness and sleep, and reduced stress and anxiety. Cardiovascular functions that have been reported to change following Tai Chi include reduced heart rate and blood pressure, increased vagal activity, and reduced cholesterol. Tai Chi has also been reported to reduce pain in people with fibromyalgia, osteoarthritis, and rheumatoid arthritis (Chang 2010; Field 2011; Lee 2010; Nomura 2010). Clinical studies have reported that Tai Chi is an intervention that may benefit many chronic diseases, including Parkinson's disease (Li 2012), depression in the elderly (Lavretsky 2011), chronic heart failure (Yeh 2008; Yeh 2011), chronic obstructive pulmonary disease (Chan 2010; Yeh 2010), and rheumatoid arthritis (Han 2009; Wang 2011). Tai Chi exercise has been reported to help community dwelling older adults to enhance mobility, manage the fear of falling, improve physical and cognitive functioning, and increase quality of life (Au‐Yeung 2009; Huang 2011; Liu 2010; Taylor‐Piliae 2010; Wang 2010). Typical disabilities following stroke include poor neuromuscular control, high blood pressure, and depression (Taylor‐Piliae 2007). Therefore, the physical and psychological benefits from practicing Tai Chi, especially increased balance, improved gait, decreased stress and anxiety, and improved cardiovascular and cognitive functions, may also help stroke survivors.

Why it is important to do this review

Tai Chi can be practiced at any time and in any place with no need for special equipment. It has been reported to be safe, feasible, and potentially beneficial for patients after stroke in clinical studies (Au‐Yeung 2009; Hart 2004; Taylor‐Piliae 2012). Tai Chi may potentially enhance recovery of people with stroke. To systematically summarize the current available clinical evidence on the effect of Tai Chi on patients after stroke, we propose to conduct this review in order to provide clear evidence on the effectiveness and safety of Tai Chi for improving recovery in stroke patients.

Objectives

To evaluate the effectiveness of Tai Chi on dependency and motor function and its safety for recovery of people after stroke.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomized controlled trials (RCTs) irrespective of masking, publication status, or language. We will also include randomized crossover studies.

Types of participants

People of any age with a diagnosis of stroke that fulfils the clinical criteria of the World Health Organization (WHO) of "Rapidly developed clinical signs of focal disturbances of cerebral function, lasting more than 24 hours or leading to death, with no other apparent cause than of vascular origin" (Hatano 1976). There will be no limitation in relation to time since stroke.

Types of interventions

The treatment intervention will be any exercise activity in the name of Tai Chi. The control intervention will include no treatment, another exercise program, or conventional treatment. The exercise program could be any physical activities organized to improve health. In addition to the studied treatment and control intervention, some co‐interventions such as diet, education programs, life modification, or medicine may be administered simultaneously to the treatment and control group. We will include studies with co‐interventions if all arms of a study receive the same co‐interventions, except systematically organized exercise program which may dilute the effect of Tai Chi.

Types of outcome measures

We will collect the following primary and secondary outcome measures after the Tai Chi exercise program and at the end of follow‐up, if available.

Primary outcomes

  1. Dependency, assessed by any scales with reported validated studies, such as the modified Rankin Scale (mRS), Care Dependency Scale (CDS), Functional Independence Measure (FIM), Barthel index (BI), or Modified Barthel Index (MBI).

  2. Motor function, measured by any scales with reported validated studies, such as Motor Assessment Scale (MAS) or Clinical Outcome Variables (COVS).

If more than one dependency or motor function outcome are reported, we will compute the standardized mean difference (SMD) in meta‐analysis for combining the scale results. We will make different combinations of global motor function assessments or specific ones to upper limb and lower limb.

Secondary outcomes

  1. Measures of specific upper and lower limb function, such as Wolf Motor Function Test (WMFT) or Fugl‐Meyer Assessment (FMA).

  2. Validated scales of neurological function, such as Canadian Neurological Scale (CNS) or National Institutes of Health Stroke Scale (NIHSS).

  3. Measures of balance, such as Berg Balance Scale (BBS).

  4. Measures of gait, such as Tinetti Gait Scale (TGS) or Gait Assessment and Intervention Tool (GAIT).

  5. Measures of mood, such as Beck Depression Inventory (BDI), Geriatric Depression Scale (GDS), or General Health Questionnaire (GHQ).

  6. Measures of cognitive functions, such as Mini‐Mental State Examination (MMSE) or Montreal Cognitive Assessment (MoCA).

  7. Quality of life, measured by any scales with reported validated studies, such as the Medical Outcomes Study Short Form 36 (SF‐36), Stroke Specific Quality of Life Scale (SS‐QOL). London Handicap Scale (LHS), Nottingham Health Profile (NHP), Stroke Impact Scale (SIS) or EuroQoL‐5 Dimensions (EQ‐5D).

  8. All‐cause death.

  9. Possible adverse events caused by practicing Tai Chi, such as sore muscles or sprains.

Search methods for identification of studies

See the 'Specialized register' section in the Cochrane Stroke Group module. We will not apply any date or language restrictions in the search for trials and we will arrange translation of trials published in languages other than English. Two review authors (HWZ and RZ) will be responsible for searching the literature.

Electronic searches

We will search the Cochrane Stroke Group Trials Register and the following electronic bibliographic databases.

  1. The Cochrane Central Register of Controlled Trials (CENTRAL) (Cochrane Library, latest issue).

  2. MEDLINE Ovid (from 1946).

  3. Embase Ovid (from 1980).

  4. CINAHL EBSCO (Cumulative Index to Nursing and Allied Health Literature; from 1982).

  5. AMED Ovid (Allied and Complementary Medicine Database) (from 1985).

  6. PEDro (Physiotherapy Evidence Database) (www.pedro.org.au/).

  7. REHABDATA (www.naric.com/research/rehab/).

We will search the following Chinese databases.

  1. CBM (Chinese BioMedical Literature Database, from 1978).

  2. TCMonline (from 1949).

  3. Index to Taiwan Periodical Literature System (from 1970).

  4. Wanfang Data (www.wanfangdata.com/).

We developed the MEDLINE search strategy with the help of the Cochrane Stroke Group Information Specialist and will adapt it for the other databases (Appendix 1).

Searching other resources

We will also search the following trials and research registers.

  1. US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov/).

  2. WHO International Clinical Trials Registry Platform (apps.who.int/trialsearch/).

  3. EU Clinical Trials Register (www.clinicaltrialsregister.eu).

  4. Stroke Trials Registry (www.strokecenter.org/trials/).

  5. ISRCTN (www.ISRCTN.com).

  6. Centre Watch Clinical Trials listing service (www.CenterWatch.com).

In an effort to identify further published, unpublished and ongoing trials we will search the following.

  1. Open Grey (System for Information on Grey Literature in Europe, formerly SIGLE: www.opengrey.eu/).

  2. The Canadian Agency for Drugs and Technologies in Health (CADTH) Grey Matters gray literature checklist (www.cadth.ca/resources/finding‐evidence/grey‐matters).

We will also search the reference lists of all included studies and any relevant systematic reviews identified for further references to relevant trials. We will use the Web of Science Cited Reference Search facility to forward search citation references of included studies.

We will contact experts and organizations in the field to obtain additional information on relevant trials. Additionally, we will contact the original study authors for clarification and further data if trial reports are unclear.

Data collection and analysis

Selection of studies

Two review authors (HWZ and RZ) will independently review the titles and abstracts of the literature search results. They will discard studies that are not eligible for inclusion in the review and retain those with relevant data or information. We will retrieve the full texts of potentially eligible articles for further assessment. We will categorize every record as either include, exclude, or unclear. We will resolve any disagreements by discussion and consensus. When an article is categorized as unclear because there is unclear information or missing data, we will contact the trial authors for clarification. We will record any communication with trial authors. We will list all studies excluded after full‐text assessment in a 'Characteristics of excluded studies' table. The study selection process will be illustrated in a PRISMA diagram.

Data extraction and management

Two review authors (HWZ and RZ) will independently perform data extraction using a pretested data extraction form (Appendix 2). Where a study has more than one publication, we will only use the publication with the most complete and recent data. Where relevant outcomes are only published in an earlier publication, we will use these data. We will highlight any discrepancies between published versions. We will request any further information required from the original study author by written correspondence and will include any relevant information obtained in this manner in the review. The two review authors will resolve any disagreements in consultation with a third review author (ZXL).

Assessment of risk of bias in included studies

Two review authors (HWZ and RZ) will independently assess the risk of bias in the included studies using Cochrane's 'Risk of bias' assessment tool (Higgins 2011). To detect potential biases, we will assess the following domains.

  1. Random sequence generation.

  2. Allocation concealment.

  3. Blinding of participants and personnel.

  4. Blinding of outcome assessment.

  5. Incomplete outcome data.

  6. Selective reporting.

  7. Other sources of bias. We will consider baseline comparability as one of the other sources of bias.

We will categorize the risk of bias for each domain within the included studies to one of three levels, i.e. low, unclear, or high risk of bias), and will further summarize the risk of bias for each included study. Based on this, we will then use the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system to evaluate the quality of evidence for each individual outcome (GRADEpro GDT; Higgins 2011; Schünemann 2013). This approach involves consideration of risk of bias (methodological quality), and also directness of evidence, heterogeneity, precision of effect estimates, and risk of publication bias. Following the GRADE approach, we will assign the empirical evidence of each individual outcome to one of four levels: high, moderate, low, or very low quality.

We will resolve any discrepancies by discussion. If we cannot reach a consensus, we will consult a third review author (ZXL) to help make the final decision.

Measures of treatment effect

For dichotomous outcomes, we will express results as risk ratios (RRs) with 95% confidence intervals (CIs). To help interpret the results of meta‐analysis, we will calculate the number needed to treat for an additional harmful outcome (NNTH) and number needed to treat for an additional beneficial outcome (NNTB) across a range of assumed control risks.

Where continuous scales of measurement are used, we will use the mean difference (MD) or the standardized mean difference (SMD) if the study authors used different scales. We will pay special attention to the different directions of scales.

Unit of analysis issues

We will base the analysis of outcomes on the randomized individuals. We will give special attention to the designs of cluster RCTs, crossover studies or repeated measures on the same participants. If the analysis is correct to properly account for the cluster design in cluster RCTs, we will extract the effect estimates and their standard errors for meta‐analysis using the generic inverse‐variance method. If not, we will perform appropriate analyses of cluster RCTs. For crossover studies, we will extract only results from the first randomization phase. When repeated measures are reported, we will define several different outcomes based on different periods of follow‐up and will analyze these separately.

In the case of multiple intervention groups within a study, we will make pair‐wise comparisons relevant to the study objective. If necessary, we will combine relevant groups to make a single pair‐wise comparison.

Dealing with missing data

We will conduct both available‐case analysis and intention‐to‐treat (ITT) analysis. In available‐case analysis, because the general methods to impute missing data in individual studies are conservative, e.g. last observation carried forward, it is unlikely that this would lead to overestimation of effect sizes. We will consider the potential impact of missing data in the 'Risk of bias' tables and interpretation of the results. When 15% of data are missing, we will regard the study as having a high risk of bias. For dichotomous outcomes, we will investigate missing data by sensitivity analyses of worst‐best cases. For continuous outcomes, we will assume a fixed difference between the means of missing data and the measured outcome data in the sensitivity analysis.

Assessment of heterogeneity

We will quantify the heterogeneity using the I2 statistic, with values of 25%, 50%, and 75% corresponding to low, medium, and high levels of heterogeneity, respectively. We will use the Chi2 test with an alpha of 0.1 for statistical significance.

Assessment of reporting biases

We will investigate reporting biases visually using funnel plots and we will explore possible reasons for reporting biases other than publication bias, such as poor methodological quality and true heterogeneity.

Data synthesis

We will pool data using the random‐effects model. We will also analyze the fixed‐effect model to explore the influence of small‐study effects. We will use Cochrane's statistical software, Review Manager 5 (RevMan 5), for data synthesis (Review Manager 2014).

Subgroup analysis and investigation of heterogeneity

If sufficient information is available, we will conduct subgroup analyses to explore potential sources of heterogeneity based on the following.

  1. Baseline dependency or motor function score.

  2. Duration of practicing Tai Chi.

  3. Different programs of Tai Chi.

  4. Different control groups (different exercise program or conventional treatments).

We will tabulate adverse effects and assess them with descriptive techniques. We will also calculate the incidence rate of different type of adverse effects.

Sensitivity analysis

For dichotomous outcomes, we will perform worst‐best cases to explore the impact of incomplete or missing data. For continuous outcomes, we will assume a fixed difference between the mean values of missing data and the measured outcome data in the sensitivity analysis. We will use both fixed‐effect and random‐effects models to investigate the influence of small study effect.

We will perform sensitivity analysis to explore the effect of removing studies that are at high risk of bias from the analysis for primary outcomes.

'Summary of findings' table

The following outcomes are prespecified for presentation in the 'Summary of findings' table.

  1. Dependency.

  2. Motor function.

  3. Neurological function, mood, and cognitive functions.

  4. Quality of life.

  5. Possible adverse events caused by practicing Tai Chi.

We will consider the quality of evidence, potential benefits and harms, study context, and patients' value on the intervention benefit and possible harm when we interpret the results.