Mirror therapy for upper limb motor recovery after stroke: a meta-analysis

Objective To systematically evaluate the rehabilitation effect of mirror therapy (MT) on upper limb motor function in stroke patients, thereby providing evidence-based support for clinical rehabilitation decision-making. Methods Randomized controlled trials (RCTs) investigating MT for upper limb dysfunction after stroke were retrieved from the PubMed, Web of Science, Cochrane Library, and EMBASE databases. The risk of bias in the included studies was assessed using the Cochrane Risk of Bias tool 2.0 (RoB 2.0). A meta-analysis was performed, and the grade of evidence was determined using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. Results A total of 21 RCTs were ultimately included. The meta-analysis showed that MT significantly improved the total score of the Fugl-Meyer Assessment Upper Extremity (FMA-UE) in stroke patients (Mean Difference [MD] = 5.45, 95% Confidence Interval [95% CI]: 3.50–7.41, p < 0.001). Hand function showed a statistically significant but clinically modest improvement (MD = 1.16, 95% CI: 0.03–2.29, p = 0.04); however, sensitivity analysis indicated that this statistical significance was not robust and depended on a small number of high-weight studies. A positive trend toward improvement was observed for shoulder-elbow-forearm function (MD = 1.79, 95% CI: −0.02 to 3.60, p = 0.05). However, no statistically significant improvement was found for wrist joint function (MD = 0.65, 95% CI: −0.17 to 1.47, p = 0.12). Heterogeneity was low across outcomes. No clear evidence of publication bias was detected; however, the number of studies was small for some outcomes, limiting the power of funnel plots and Egger’s test. The overall quality of evidence was moderate to low. Conclusion MT may improve global upper limb motor function in stroke patients, with a more modest and less robust effect on hand function. The statistical significance of the hand-function finding depended on a small number of high-weight studies, and the evidence for hand function is of low quality; the overall evidence is low-to-moderate. Therefore, these findings, especially those for hand function, should be interpreted cautiously, and further high-quality studies are warranted. Systematic review registration https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD420261395137 , PROSPERO registration number: CRD420261395137.