Effects of Balance Exercise Interventions on Balance-Related Performance in People With Multiple Sclerosis: A Systematic Review and a Meta-Analysis of Randomized Controlled Trials.

Background: Balance training covers a range of different modalities and complexity levels for people with multiple sclerosis (MS). When evaluating the effects of balance training across different kinds of interventions, determination of the specific intervention content that predict effects are needed.

Objective: To investigate the effects of balance training on gait and dynamic balance outcomes.

Methods: Four databases were systematically searched. Randomized controlled trials involving people with MS (Expanded Disability Status Scale [EDSS] score ≤7.5) where at least 50% of the intervention targeted balance control were included. Interventions were categorized based on training types. Risk-of-bias was assessed using the Tool for the Assessment of Study Quality and Reporting in Exercise (TESTEX).

Results: A total of 18 included studies involved 902 people with MS (EDSS range from 0 to 7.5). Interventions evaluated with a balance composite score or a mobility test showed a moderate effect size (ES = 0.46 [95% confidence interval (CI) = 0.18 to 0.74]; p < .01) and a small overall ES (ES = 0.19 [95% CI = 0.01-0.36]; p = .04), respectively, across different training types. Stepping and gait speed outcomes showed no effect. Cognitive dual-task training showed a significant effect (ES = 0.81 [95% CI = 0.24 to 1.37]) on subgroup level, when evaluated with a mobility outcome measure. The median TESTEX score on study quality and reporting was 11 (maximum score = 15).

Conclusions: Improvements of balance were found across interventions when measured by balance composite scores and mobility tests, but not when measured by stepping or gait speed outcomes. Large training volume was positively associated with effect on balance. A definition of intensity in balance training is needed for evaluation of its impact on the effect of balance interventions.