Kinesio Taping for Balance Function after Stroke: A Systematic Review and Meta-Analysis

Objective With the increasing social and economic burdens of balance impairment after stroke, the treatment for balance impairment after stroke becomes a major public health problem worldwide. Kinesio taping (KT) as a part of clinical practice has been used widely in the treatment of balance impairment after stroke. However, the clinical effects of KT for balance function have not been confirmed. The objective of this study is to investigate the effects and safety of KT for balance impairment after stroke. Methods We conducted a systematic review (SR) and meta-analysis of randomized controlled trials (RCTs) on the effects of KT for balance impairment after stroke. We searched the following databases: (1) English databases: EMBASE (via Ovid), MEDLINE (via Ovid), the Cochrane library, PubMed, and PEDro; (2) Chinese databases: China Biology Medicine (CBM), Wan Fang database, China National Knowledge Infrastructure (CNKI), and VIP. Besides, hand searches of relevant references were also conducted. We systematically searched from the inception to December 2018, using the keywords (Kinesio, Kinesio Tape, tape, or Orthotic Tape) and (stroke, hemiplegia, or hemiplegic paralysis) and (balance or stability). The search strategies were adjusted for each database. The reference lists of included articles were reviewed for relevant trials. For missing data, we contacted the authors to get additional information. Results 22 RCTs involved 1331 patients, among which 667 patients in the experimental group and 664 patients in the control group were included. Results of meta-analysis showed that, compared with conventional rehabilitation (CR), there was significant difference in Berg Balance Scale (BBS) (MD=4.46, 95%CI 1.72 to 7.19, P=0.001), Time Up and Go Test (TUGT) (MD=-4.62, 95%CI -5.48 to -3.79, P < 0.00001), functional ambulation category scale (FAC) (MD=0.53, 95%CI 0.38 to 0.68, P < 0.00001), Fugl-Meyer assessment (FMA-L) (MD=4.20, 95%CI 3.17 to 5.24, P < 0.00001), and Modified Ashworth Scale (MAS) (MD=-0.38, 95%CI -0.49 to -0.27, P < 0.00001). The results of subgroup analysis showed that there was no significant difference between KT and CR with ≤4 weeks treatment duration (< 4 weeks: MD=5.03, 95%CI -1.80 to 11.85, P=0.15; =4 weeks: MD=4.33, 95%CI -1.50 to 10.15, P=0.15), while there was significant difference with more than 4-week treatment duration (MD=4.77, 95%CI 2.58 to 6.97, P < 0.0001). Conclusions Based on current evidence, KT was more effective than CR for balance function, lower limb function, and walking function in poststroke patients. Longer treatment duration may be associated with better effects. However, more well-conducted RCTs are required in the future.


Introduction
Stroke is a common clinical cerebrovascular disease, including cerebral infarction, intracerebral hemorrhage (ICH), and subarachnoid hemorrhage (SAH), with a high morbidity, which is a major cause of disability and death among people around the world [1]. As the world's population ages, the burden of stroke increases rapidly [2]. WHO Global status report on noncommunicable disease (NCD) showed an estimated 6.7 million NCD deaths were caused by stroke; 33 million of stroke survivors needed long-term follow-up and secondary preventive measures [3]. A study on 84,184 acute stroke patients in England, Wales, and Northern Ireland showed that the total cost of health and social care each year was m3.60 billion in the first five years after admission [4]. A study on economic burden of informal care attributable to 2 Evidence-Based Complementary and Alternative Medicine stroke among those aged 65 years or older in China showed that the average annual cost of informal care associated with stroke was 10,612 RMB per stroke survivor [5].
Balance impairment is one of the common daily functional problems in stroke patients, which seriously affects the patient's daily life and work. According to the severity of stroke, the manifestations of balance impairment can be classified into stepping, standing, and sitting balance impairment [6]. Poor balance ability is often associated with increased risk of falls, disability, and even mortality. A study based on 41 community-dwelling people with stroke showed 50% (21/41) participants were classed as fallers, of whom 10 had fallen repeatedly [7]. Another study based on 522 adults showed that, by 2 years after stroke, 23.5% (124/522) adults had fallen at least once, 14.2% (74/522) had 2 or more falls and 5.4% (28/522) had a fracture [8]. Especially for old people, owing to poor balance ability, they have to face the risk of body injuries caused by falls for a long time [9], some severe falls can even lead to paralysis or death. This damage is undoubtedly serious and complicated. Hence, it is urgent for stroke patients to improve their balance ability [10][11][12][13][14][15].
Kinesio taping (KT) was introduced by Kenzo Kase in 1996, which normally involves a combination of applying appropriate tension along the elastic therapeutic tape and placing the target muscle in a stretched position and is widely used as an interesting and relatively novel method for various clinical treatments. KT has gained popularity in clinical practice and been used in clinical practice worldwide. By supporting weak muscle, relaxing overstretched muscle, and reducing pain, KT can promote functional use of the upper and lower extremity, in further to improve balance ability [16]. Besides, KT was also a good option for stroke patients who had asymmetrical and imbalanced body posture [17]. Recent studies showed that KT has been used as an adjuvant rehabilitation therapy in the treatment of balance impairment after stroke [18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36], but the conclusions were inconsistent. The effect of KT on balance impairment of poststroke patients is still controversial. Therefore, a systematic review (SR) and meta-analysis of randomized controlled trials (RCTs) was performed to investigate the efficacy and safety of KT on balance impairment after stroke.
Hand searches of relevant references were conducted. Clinical trials registration websites, including https://clinicaltrials.gov/ and http://www.chictr.org.cn, were also searched for possible RCTs.
All the databases were conducted from their inception to December 2018. The search terms (Kinesio, Kinesio Tape, tape, or Orthotic Tape) and (stroke, hemiplegia, or hemiplegic paralysis) and (balance or stability) were used and the search strategies were adjusted for each database. Details of search strategies were given in Appendix A.

Inclusion Criteria.
Trials were screened based on the following inclusion criteria: (1) RCTs of KT for stroke patients with balance impairment; (2) patients with balance impairment after stroke diagnosed according to stroke diagnostic criteria with clear history and manifestations and confirmed by the brain Computed Tomography (CT) or Magnetic Resonance Imaging (MRI) and the age and sex are not limited; (3) the experimental group was treated with KT along with or without conventional rehabilitation (CR). The control group can be CR, sham KT, or placebo; (4) primary outcome is Berg Balance Scale (BBS). Secondary outcomes include related outcomes to lower limb function, spasticity, walking function.

Exclusion Criteria.
Trials met the following criteria would be excluded: (1) Trials were non-RCTs, such as literature review, case report, and expert treatment experience summary; (2) balance impairment was not caused by stroke, for example, caused by Parkinson's disease, pediatric cerebral palsy, knee surgery, or other diseases; (3) duplicate or the data cannot be extracted.

Studies Selection.
All the retrieved studies were imported into Endnote (X8), and the filter tool was used to delete the duplicated studies. Two reviewers (YJH and DLZ) independently screened the titles and abstracts according to the inclusion and exclusion criteria. After screening, two reviewers (YJH and DLZ) cross checked and downloaded the full text of all possibly relevant studies for further assessment. The disagreements were resolved through team discussion.

Data Extraction.
A standardized data extraction form was designed in advance. The following information of the included studies, first author; publication year; country of origin; participant characteristics; evaluation time and outcomes, were independently extracted by two reviewers (QWX and YJH). If the trials had more than 2 groups or factorial designs and multiple comparisons, we extracted only the information and data of interest reported in the original articles. In case of disagreements, a third reviewer (RJJ) was involved.
2.6. Risk-of-Bias Assessments. The methodological quality for the included RCTs was independently assessed by 2 reviewers (YJH and QC) based on Cochrane risk-of-bias criteria [19]. The Cochrane risk-of-bias criteria included the randomization sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias. Each quality item was graded as low, high, or unclear. The disagreements of assessment were resolved through team discussion. 18 of studies included in quantitative synthesis (meta-analysis) Figure 1: PRISMA flow chart for study selection.

Statistical Analysis.
We performed meta-analysis to calculate mean difference (MD) and 95% CIs using the Mantel-Haenszel statistical method. Heterogeneity test of each outcome was conducted by using the Chi-square test with no significance (P>0.05), and 2 statistic evaluated the degree of heterogeneity. If 2 < 50%, many similar studies could be considered to have no homogeneity, the fixed effect model adopted for a meta-analysis; otherwise random effects model was performed. If substantial heterogeneity was detected, subgroup analysis or sensitivity analysis could be applied to explore the causes of heterogeneity. If the sources of heterogeneity could not be determined, descriptive analysis was adopted.

Study Selection.
A total of 414 potentially relevant articles were retrieved. After removing duplicates, titles and abstracts of remaining articles were screened for inclusion. After reading full texts, 22 RCTs were included in SR and 18 RCTs for meta-analysis. Excluded articles with explanations were listed in Appendix B. Flow diagram for selection of the included studies was showed in Figure 1.  [18][19][20], and 1 was in Iran [21]. Disease duration varied from days to years. The control group in 20 RCTs utilized CR, while 2 RCTs used sham KT [20,22]. The locations of KT were on lower limbs (quadriceps, triceps surae, tibialis anterior, ankle, and so on).

Methodological Quality.
The risk of bias assessed by the Cochrane Collaboration's tool of the included RCTs was summarized in Figures 2(a) and 2(b). Among the 22 included RCTs, the randomization procedure was reported adequately in the 8 RCTs; 2 RCTs clearly reported the allocation concealment. Only 1 RCT described the blinding of participants and personnel. We considered RCTs using objective outcome indicator as "low quality"; there were 6 RCTs considered "low quality" in the blinding of outcome assessment. In incomplete outcome data section, only 1 RCT was considered "high quality" due to no reason for loss. All the 22 RCTs were rated "uncertain risks" in selective reporting for no protocol published in advance. In other bias section, we considered "uncertain risks" if the study did not report conflict of interest or source of funds.    We performed sensitivity analysis by removing RCTs one by one; the results remained unchanged.

MAS. 5 RCTs compared KT with CR
Other Outcomes. As shown in Table 2 [37] found that KT was more effective than CR on lower limb swelling in stroke patients. Chen Z et al. [39] reported that KT can effectively decrease the angle of strephenopodia; it was safe with no adverse effects. Besides, Tan TC et al. [31] found that KT had a positive effect in improving gait and walking ability of hemiplegia patients. Tang Y et al. [32] found KT with physical therapy may be favorable in improving functional outcome of sit-tostand transfer in stroke patients and use on both sides is more effective.

Summary of Findings.
To our knowledge, this is the first meta-analysis of KT for balance ability after stroke. Results of this meta-analysis showed that KT was more effective than CR for balance ability, lower limb function, and walking function in poststroke patients. But no significant difference was found between KT and sham KT in BBS. Possible causes may be only two studies included. However, the results of BBS were not consistent regarding of the treatment duration. In the subgroup analysis, we found that, compared with CR, KT was more effective with more than 4-week treatment duration. However, there was no significant difference between KT and CR with treatment duration ≤4 weeks, which suggested that long-term KT may be more effective than short-term KT. Except 10mMWT, there was significant difference between KT and CR in TUGT, FMA, FAC, MAS, and other outcomes.

Implications for Further Studies.
Balance impairment is one of the common daily functional problems in stroke patients, which is often associated with increasing risk of falls, poor walking function and low quality of life. KT is a safe and effective method, while the mechanism of KT is not clear at present, its therapeutic effect may be the stickiness and principles of kinematics and biomechanics, by sticking KT into different directions and using different tensions. Recent studies showed that KT had been used as an adjuvant rehabilitation therapy in the treatment of balance impairment after stroke. We retrieved 4 SRs of the clinical effects of KT [40][41][42][43], 1 SR was not only focus on stroke patients [40], and 3 of them focused on the motor function of lower extremity, which directly affected the balance function [41][42][43]. Compared with other outcomes, balance ability is the key for poststroke patients to return to family and society. 1 previous SR concluded that the current existing evidence was insufficient to support the use of KT over other modalities in clinical practice [40]. Due to these controversial conclusions, our SR focused on balance function and made a more comprehensive search in English and Chinese databases. 22 RCTs which met the inclusion criteria were included. Results of Wu YH et al. [43] showed that 4-week KT was more effective than 6 weeks KT, which was opposite to our results. The possible reasons for the opposite result may be that Wu YH et al. combined BBS and TUGT together, which certainly contributed to heterogeneity. Our SR took BBS as primary outcome and TUGT as secondary outcome. Results of our SR indicated that KT was superior to CR in both BBS and TUGT.

Strengths and Limitations.
This SR provides the latest evidence on KT for balance function after stroke based on the findings of relevant RCTs; the results of our meta-analysis showed that KT may be a beneficial complementary therapy in the balance rehabilitation process for stroke patients. The results of this study will provide evidence for the clinical application of KT. Different from current SRs, we conducted a comprehensive search and reporting in accordance with PRISMA (Appendix C). However, our study still has several limitations. First of all, the treatment duration of KT was inconsistent, from as short as an immediate treatment to 12 weeks, indicating a wide variation in designs. Second, the locations of KT were not standardized in lower limb, hip, thigh, and crus, which may also lead to discrepancy. Third, several outcomes included only 2 studies, which may lead to unreliable results.
Hence, more high methodological quality, larger sample size, and standard-designed trials are required to draw a definitive conclusion and provide a standard KT program.

Conclusion
Based on current evidence, this study demonstrated that KT was more effective than CR for balance ability, lower limb function and walking function in poststroke patients. Longer treatment duration may be associated with better effects. However, more well-conducted RCTs are required in the future. Table 3 No.   / Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale.

5
Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.

4-5
Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated.

4
Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).

5
Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.

6
Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.