Enhancing motor function after stroke: a systematic review and meta-analysis of bioelectrical feedback interventions

Stroke is a leading cause of long-term disability, with hemiplegia affecting approximately 80% of survivors. Conventional rehabilitation shows limited effectiveness, and electromyographic (EMG)-based bioelectrical feedback has emerged as a promising adjunctive therapy.

Following PRISMA 2020 guidelines, the Cochrane Library, PubMed, EMbase, Web of Science, and CNKI were searched from 1 January 2010 to 31 March 2025. Two researchers independently screened controlled clinical studies (randomized or non-randomized) evaluating EMG-based bioelectrical feedback for post-stroke hemiplegia. Risk of bias was assessed using the Cochrane Risk of Bias 2 (RoB 2) tool for randomized trials and ROBINS-I for non-randomized studies, with analyses performed in RevMan 5.3. For random-effects analyses with at least four contributing studies, 95% prediction intervals (PIs) were reported alongside 95% confidence intervals.

Eight studies comprising 549 patients (275 intervention, 274 control) were included. Bioelectrical feedback significantly improved total Fugl-Meyer Assessment scores [MD = 9.50, 95% CI (3.41, 15.60), p = 0.002], Activities of Daily Living scores [MD = 8.80, 95% CI (3.67, 13.94), p = 0.0008], and Fugl-Meyer motor subscale scores [MD = 6.83, 95% CI (1.52, 12.14), p = 0.01] versus conventional therapy. Neurophysiological outcomes also improved, with greater EMG amplitude [MD = 0.03 mV, 95% CI (0.00, 0.06), p = 0.03] and larger active range of motion [MD = 4.74°, 95% CI (1.99, 7.50), p = 0.0007]. Complications were significantly reduced [OR = 0.26, 95% CI (0.14, 0.49), p < 0.0001]; because control-group event rates were not rare (approximately 20-40%), this corresponds to an estimated 54-67% relative risk reduction rather than 74%. Heterogeneity was high for most continuous outcomes (I² = 69-98%), and 95% PIs for EMG amplitude, AROM, and FMA motor subscale crossed zero, indicating that a future comparable study could plausibly report null or unfavorable effects despite positive pooled estimates. The small number of studies per outcome (3-6) precluded subgroup analyses and formal assessment of publication bias.

EMG-based bioelectrical feedback may improve motor function and activities of daily living and may reduce complications in hemiplegic stroke patients. However, high heterogeneity, few pooled studies per outcome, geographic concentration, and inability to assess publication bias warrant cautious interpretation. Adequately powered randomized controlled trials with blinded outcome assessment are needed before firm clinical recommendations can be made.