Kinesiotape effects on ankle functional performance and perceived stability in chronic ankle instability: a systematic review and meta-analysis

Protocol and guidance

This study was conducted according to the PRISMA (preferred reporting items for systematic reviews and meta-analysis) guidelines¹⁶. The protocol for this review was registered with PROSPERO (CRD42024497654).

Search strategy

PubMed, Web of Science, Scopus, Physiotherapy Evidence Database (PEDro) and SPORTDiscus were searched for studies published in English from 2014 to September 2025. The starting year was selected because the International Ankle Consortium published standardised selection criteria for individuals with CAI in 2014. The final search was conducted on 6 September 2025. Terms related to ‘kinesiotape’ and ‘chronic ankle instability’ were used to prepare the search strategy. The keywords were ‘kinesio tap*’ OR ‘kinesiotap*’ OR ‘elastic tap*’ OR ‘athletic tap*’ OR ‘orthotic tap*’ AND ‘chronic ankle instability’ OR ‘functional ankle instability’, which were found in the title, abstract or keywords. The search strategy for each database is presented in Supplementary Table 1. Reference lists of included studies and relevant systematic reviews were also manually searched.

Eligibility criteria

In this systematic review, the research question is developed using the PICO strategy:

P: Population: Individuals with CAI according to the inclusion criteria of the International Ankle Consortium¹⁷.

I: Intervention: KT.

C: Comparison: Compare with other taping conditions (rigid tape [RT], mulligan tape [MT], athletic tape [AT], dynamic tape [DT], sham tape [ST] or NT).

O: Outcome: Objective ankle functional performance (dynamic balance, static balance, proprioception, muscle activation, muscle strength and sports performance) and subjective perceived ankle stability, assessed using validated instruments at baseline and post-intervention.

Studies included in the systematic review should meet the following criteria:

1. Individuals with CAI needed to follow the International Ankle Consortium criteria:

   • History of at least one ankle sprain.

   • History of giving way, recurrent sprain or instability.

   • Self-reported ankle instability or function confirmed with a validated questionnaire: Ankle Instability Instrument: answer yes to at least 5 yes/no questions (this should include question 1 plus 4 others); Cumberland Ankle Instability Tool: score of ≤24; Identification of Functional Ankle Instability: score of ≥11; Foot and Ankle Ability Measure activities of daily living subscale <90%, sport subscale < 80%; Foot and Ankle Outcome Score: score of <75% in three or more categories^18-22.

2. Comparing KT with other taping conditions (RT, AT, MT, DT, ST, or NT).

3. KT is applied to the ankle joint, and at least one ankle functional performance or perceived stability is assessed with validated instruments.

4. Research designs only including randomised controlled trials (RCTs) (including cross-over designs).

The following exclusion criteria were followed:

1. Review articles, meta-analyses, clinical practice guideline, conference proceeding, case reports or protocol.

2. Combination KT with other exercise therapies (e.g. balance exercise, weight bearing exercise) or comparison of KT with other ankle orthoses.

3. A history of previous surgeries to the musculoskeletal structures (i.e. bones, joint structures, nerves) in either lower extremity.

4. A history of a fracture in either lower extremity requiring realignment.

5. Acute injury to musculoskeletal structures of other joints of the lower extremity in the previous 3 months that impacted joint integrity and function (i.e. sprains, fractures), resulting in at least 1 interrupted day of desired physical activity.

Study selection

After excluding duplicates in the literature, two reviewers (F.S. and Y.R.) reviewed the titles and abstracts of the studies. Based on the established selection criteria, the reviewers individually reviewed the full text of the remaining studies. Any disagreements in screening were resolved by discussion, and if consensus could not be reached, a third reviewer (K.T.) was consulted.

Data extraction

Data extraction was completed by an independent reviewer (F.S.). For all includable studies, the following data were extracted: author, published date, study design, participant characteristics (numbers, sex, age), intervention and comparison, KT application method, applying time, outcome measures related to ankle functional performance and perceived stability. The mean and standard deviation of the relevant outcomes were retrieved, and a meta-analysis was carried out when commonalities in population characteristics, control intervention and outcomes were found. If outcomes were reported for the CAI group and the non-CAI group, we extracted the outcome of the CAI group. During the different followup time points, the outcomes that were closest to the end of the intervention period were extracted.

Quality and risk of bias assessment

Quality and risk of bias were assessed using the original Cochrane Collaboration Risk of Bias tool (RoB 1)²³. Using this tool, it can determine whether biases in reporting, selection, performance, detection, attrition and other areas are ‘high risk’, ‘unclear’ or ‘low risk’. The two reviewers reached a consensus to resolve any disagreement on the risk of bias assessment.

Data synthesis

Review Manager (RevMan)²⁴ version 5.4 (The Cochrane Collaboration, 2023) was used for all data synthesis. We used standardised mean differences (SMDs) with 95% confidence intervals (CIs) to assess the outcomes and considered p < 0.05 as statistically significant. We assessed heterogeneity using the I² test²⁵. Fixed effects models were used to pool outcomes when significant heterogeneity was absent (I² < 50%), and random effects models were used when significant heterogeneity was present (I² ≥ 50%). The pooled SMD was interpreted as follows: small effect (<0.40), moderate effect (0.40–0.70) and large effect (>0.70)²⁶.

Meta-analyses were organised by comparison intervention type (NT, RT or ST) and by outcome measures (dynamic balance, static balance, proprioception and perceived stability). WebPlotDigitizer (https://automeris.io/WebPlotDigitizer) was used to extract data from studies that included graphs²⁷. Data that could not be pooled due to differences in outcome measures and the lack of studies were summarised with p-values and narrative synthesis.

The quality of the evidence for each meta-analysis was assessed by two reviewers using the GRADE (grading of recommendations assessment, development and evaluation methodology) approach28. We downgraded the quality of evidence when the following issues were discovered29: (1) Risk of bias (>75% of studies do not have low risk of bias); (2) Inconsistency (large heterogeneity based on the similarity of point estimates, statistical heterogeneity and I² > 50%); (3) Imprecision (included population inadequately reflects selection criteria of the review, estimate of the effect is sufficiently imprecise if the upper or lower confidence limit crosses the effect size of 0.5 in either direction); (4) Indirectness (studies with indirect comparison); and (5) Publication bias (asymmetry in funnel plots where a sufficient number of studies was available), which was not possible to assess in the present review due to low number of studies in each analysis (<10 studies). The evidence quality was classified as high, moderate (downgraded one level), low (downgraded two levels) or very low (downgraded ≥three levels).

Study selection and characteristics Search results

We identified 294 studies from the 5 databases and finally included 17 of them in this systematic review^30-46. The screen process and results are shown in Figure 1.

Figure 1.

Study characteristics

A total of 451 participants were included across the 17 studies, of whom 375 had CAI and 76 were healthy controls. Healthy participants were included in studies that used between-group comparisons to evaluate the effects of KT. All participants were young adults aged 18–36 years. The average sample size was 27 (range 15–35). Athletic participants were included in six studies^31,33-36,40. To investigate the effects of KT, 15 studies compared KT with NT^{30-38,40-41,43-46} and 6 studies compared KT with ST, which used the same type of KT but with no tension^{30,38,39,43,44,46}. In addition, KT was compared with other tape conditions in seven studies (RT, AT, DT and MT)^{32,36,37,40,42-44}. The effects of KT on ankle functional performance (dynamic balance, static balance, proprioception, muscle activation, muscle strength and sports performance) were investigated using 25 outcome measures in 17 studies. Furthermore, two studies assessed KT’s impacts on perceived stability^32,43. In 11 studies, outcomes were evaluated immediately following application of KT^{30,32,36-38,41-46}, while in 5 studies outcomes were evaluated at about 20–30 minutes after KT application^31,33-35,39, since prior research confirmed that KT is most effective at 20 minutes after treatment⁴⁷. In addition, outcomes were evaluated at 24 hours after KT application in two studies^39,40. The details of the author, published date, study design, participant characteristics (numbers, sex, age), intervention and comparison, KT application method, applying time and outcome measures of the studies are presented in Supplementary Table 2.

Quality and risk of bias assessment

The results of risk of bias are shown in Supplementary Figures 1 and 2. Overall, the included studies were judged to have a high risk of bias, with 15 studies rated as high risk and 2 as unclear risk; no study was considered as low risk. At the domain level, allocation concealment was frequently rated as high risk. Blinding of participants and personnel and blinding of outcome assessment were often judged as unclear risk because these procedures were insufficiently reported. Incomplete outcome data was generally judged as low risk, whereas selective reporting was mostly rated as low or unclear risk. In crossover studies, other bias was frequently judged as unclear because washout periods were not consistently reported, making it difficult to determine the potential for carry-over effects.

Findings Dynamic balance

Nine studies examined the effects of KT on dynamic balance^{30,32,34,36-39,41,43}. Dynamic balance was evaluated using various outcome measures, including the star excursion balance test (SEBT)/Y Balance, Limits of stability, Motor control test, and Adaptation test. Only SEBT and YBT outcomes were sufficiently comparable across studies to permit meta-analysis.

For the pooled analyses, separate meta-analyses were conducted according to the direction of reach and the comparator. Compared with NT, KT was associated with statistically significant improvements in the posterolateral direction based on data pooled from seven studies^{30,32,34,36,38,39,41} (SMD, 0.45; 95% CI, 0.23–0.66; low-quality evidence) (Figure 2 and Table 1). In the posteromedial direction, eight studies were pooled^{30,32,36-39,41} and KT was associated with a statistically significant improvement compared with NT (SMD, 0.46; 95% CI, 0.26–0.67; low-quality evidence) (Figure 2 and Table 1).

Figure 2.

Compared with RT, KT was associated with a statistically significant improvement in the anterior direction based on data from two studies^32,36 (SMD, 0.58; 95% CI, 0.16–1.01; low-quality evidence) (Supplementary Figure 3 and Table 1).

In contrast, compared with ST, pooled analyses from two studies showed no statistically significant differences in any SEBT direction^30,38 (Supplementary Figure 4 and Table 1).

One study was not included in the meta-analysis because dynamic balance was assessed using different outcome measures⁴³. This study reported 32 between-group comparisons, of which only two showed statistically significant differences (Supplementary Table 3), suggesting limited consistency across the non-pooled results.

Static balance

Five studies evaluated the effect of KT on static balance^{30,32,37,38,43} using different outcome measures, including centre of pressure (COP) data variables during the one-leg standing test, Balance Error Scoring System, time-to-boundary, sensory organisation test and sway velocity of centre of gravity in unilateral stance. Only COP outcomes were sufficiently comparable across studies to permit meta-analysis.

For the pooled analyses, KT was associated with a statistically significant decrease in mediolateral COP range compared with no tape (NT) based on data from two studies^32,38 (SMD, −0.38; 95% CI, −0.69 to −0.06; moderate-quality evidence) (Figure 3 and Table 2). KT was also associated with a statistically significant decrease in sway area compared with NT based on data from two studies^37,38 (SMD, −0.36; 95% CI, −0.69 to −0.03; low-quality evidence) (Figure 3 and Table 2). No statistically significant differences were observed for other pooled static balance outcomes.

Figure 3

Two studies were not included in the meta-analysis because static balance was assessed using different outcome measures^30,43. These studies reported 44 between-group comparisons, of which only one showed a statistically significant difference (Supplementary Table 4), indicating limited consistency across the non-pooled results.

Proprioception

Four studies investigated the effect of KT on proprioception^30,38,44,45, using various outcome measures, including joint position sense, threshold to detect passive motion, force sense and the Ankle Inversion Discrimination Apparatus for Landing. Only inversion joint position sense outcomes were sufficiently comparable across studies to permit meta-analysis.

Pooled analysis of two studies showed no statistically significant difference between KT and NT for inversion joint position sense^30,45 (SMD, −0.49; 95% CI, −1.37–0.39), based on very low-quality evidence (Supplementary Figure 5 and Table 3). Two studies were not included in any meta-analysiss^38,44 due to differences in outcome measures. These two studies reported 44 between-group comparisons, none of which showed statistically significant differences (Supplementary Table 5).

Muscle activation

Three studies examined the effect of KT on muscle activation^31,40,44. Reported outcomes included peak muscle activity, onset time and latency responses, antagonist co-activation and root mean square (RMS) values of muscle activity.

Compared with NT, one study found that KT was associated with higher lateral and medial gastrocnemius activity and lower peroneus longus activity during a countermovement jump task³¹. Another study reported that both KT and MT were associated with changes in onset time, short- and medium-latency response magnitude and antagonist co-activation during sudden inversion perturbation compared with NT⁴⁰. A third study showed that, during computerised dynamic posturography, lateral gastrocnemius RMS was lower with AT than with KT under one sensory organisation condition, whereas peroneus longus RMS was lower with KT than with ST during a forward-small motor control test perturbation⁴⁴.

And it was shown that both the KT and MT can affect the onset time, magnitude and antagonist co-activation of muscles during short and medium latency response compared with NT⁴⁰ (Supplementary Table 6).

Muscle strength

Only one study examined the effect of KT on muscle strength using ankle isokinetic muscle strength³⁴. Compared with NT, KT was associated with increased average ankle evertor muscle strength. (Supplementary Table 7).

Sports performance

Seven studies evaluated the effects of KT on sports performance^{31,34-36,39,42,46}, using a variety of outcome measures, including jumping tasks, functional hopping or agility tests and gait-related tasks.

For jumping-related outcomes, one study reported that, during countermovement jump performance, KT was associated with reduced ankle inversion/eversion range of motion and increased ankle dorsiflexion–plantarflexion range of motion, jump height, ankle plantarflexion angular velocity, average peak force, rate of force development, peak power and ankle joint power compared with NT³¹. In another study of unexpected jumping, KT was associated with reduced ankle plantarflexion at initial contact, increased peak ankle dorsiflexion and eversion angles and moments during early landing, and reduced peak vertical ground reaction force compared with NT and ST⁴⁶.

For functional hopping and agility tasks, the findings were inconsistent. Although one study reported a statistically significant difference in Figure-of-8 test performance between KT and NT³⁶, another study found no statistically significant difference for the same test³⁴. In addition, one study found that, in the single hop test, the KT group achieved greater hop distance than the sham group after 20 minutes and 24 hours, while in the side hop test, a statistically significant difference was observed only in average time after 20 minutes³⁹.

For gait-related outcomes, one study reported that KT was associated with increased ankle eversion during early stance in treadmill walking compared with NT, but no difference was observed during late stance⁴². By contrast, two studies found no statistically significant effects of KT on gait initiation profile or on the temporal and spatial parameters of gait initiation^34,35.

Overall, evidence for sports performance outcomes was heterogeneous and inconsistent across studies (Supplementary Table 8).

Perceived stability

Two studies investigated the perceived stability effects of KT^32,43. Visual analogue scale scores were sufficiently comparable across studies to permit meta-analysis, with higher scores indicating greater perceived stability. Pooled analysis showed that KT was associated with higher perceived stability than NT^32,43 (SMD, 0.89; 95% CI, 0.51–1.28), based on low-quality evidence from two studies (Figure 4 and Table 4).

Figure 4