Electromyographic biofeedback therapy for improving limb function after stroke: A systematic review and meta-analysis

Upper and lower limb impairment is common after stroke. Electromyographic biofeedback therapy is a non-invasive treatment, and its effectiveness in functional rehabilitation of the limb after stroke still remains uncertain.

The objective of this study was to evaluate whether electromyographic biofeedback can improve upper and lower limb dysfunction in stroke patients.

PubMed, Embase, Cochrane Library, and Physiotherapy Evidence Database (PEDro) were searched from inception to 1st May 2022. Inclusion criteria were randomized controlled clinical trials of electromyographic biofeedback therapy interventions reporting changes in upper and lower limb function in post-stroke patients. Data were extracted by two independent reviewers and pooled in random-effects models using Review manager (RevMan) software.

Our analyses included 10 studies enrolling a total of 303 participants. Electromyographic biofeedback therapy can effectively improve limb function after stroke (standardized mean difference [SMD], 0.44; 95% confidence interval [CI], 0.12-0.77; P = 0.008) and in subgroup analyses, the effect sizes of short-term effect (SMD, 0.33; 95% CI, 0.02-0.64; P = 0.04) was significant, but the long-term was not (SMD, 0.61; 95% CI, -0.11-1.33; P = 0.10). In addition, Electromyographic biofeedback therapy can improve the active range of motion of shoulder (SMD, 1.49; 95% CI, 2.22; P<0.0001) and wrist joints (SMD, 0.77; 95% CI, 0.13-1.42; P = 0.02) after stroke.

In this meta-analysis, electromyographic biofeedback therapy intervention can improve upper and lower limb function in patients with stroke. Short-term (less than one month) improvement after electromyographic biofeedback therapy was supported, while evidence for long-term (more than one month) benefits was lacking. Range of motion in the glenohumeral and wrist joints were improved. Stronger evidence for individualized parameters, such as optimal treatment parameters and intervention period, is needed in the future.

[https://www.crd.york.ac.uk/prospero/display_record.php?recordID=267596], identifier [CRD42022354363].