Effects of Exercise Therapy on Balance Capacity in Chronic Stroke: Systematic Review and Meta-Analysis

Item does not contain fulltextBACKGROUND AND PURPOSE: The purpose of this systematic review and meta-analysis was to investigate the effects of exercise training on balance capacity in people in the chronic phase after stroke. Furthermore, we aimed to identify which training regimen was most effective. METHODS: Electronic databases were searched for randomized controlled trials evaluating the effects of exercise therapy on balance capacity in the chronic phase after stroke. Studies were included if they were of moderate or high methodological quality (PEDro score >/=4). Data were pooled if a specific outcome measure was reported in at least 3 randomized controlled trials. A sensitivity analysis and consequent subgroup analyses were performed for the different types of experimental training (balance and/or weight-shifting training, gait training, multisensory training, high-intensity aerobic exercise training, and other training programs). RESULTS: Forty-three randomized controlled trials out of 369 unique hits were included. A meta-analysis could be conducted for the Berg Balance Scale (28 studies, n=985), Functional Reach Test (5 studies, n=153), Sensory Organization Test (4 studies, n=173), and mean postural sway velocity (3 studies, n=89). A significant overall difference in favor of the intervention group was found for the Berg Balance Scale (mean difference 2.22 points (+3.9%); 95% confidence interval [CI], 1.26-3.17; P<0.01; I(2)=52%), Functional Reach Test (mean difference=3.12 cm; 95% CI, 0.90-5.35; P<0.01; I(2)=74%), and Sensory Organization Test (mean difference=6.77 (+7%) points; 95% CI, 0.83-12.7; P=0.03; I(2)=0%). Subgroup analyses of the studies that included Berg Balance Scale outcomes demonstrated a significant improvement after balance and/or weight-shifting training of 3.75 points (+6.7%; 95% CI, 1.71-5.78; P<0.01; I(2)=52%) and after gait training of 2.26 points (+4.0%; 95% CI, 0.94-3.58; P