Virtual reality balance training to improve balance and mobility in Parkinson's disease: a systematic review and meta-analysis.

In the last few years, virtual reality (VR) has been increasingly used to strengthen the effect of balance training (BT) in Parkinson’s disease (PD). We performed a systematic review and meta-analysis of randomized controlled trials (RCTs) to compare the effects of VR-BT relative to BT alone for improving balance and mobility PD subjects with balance/mobility difficulties. Four electronic databases were searched: two reviewers independently selected RCTs, extracted data, and applied the Cochrane risk-of-bias tool for randomized trials (version 2) and the GRADE framework for assessing the certainty of evidence. Primary outcomes were balanced (Berg Balance Scale-BBS), mobility (Timed Up and Go-TUG) and walking speed. Secondary outcomes were falls, walking distance and stability, spatial gait parameters, balance confidence, sensory integration ability, motor signs and quality of life. We included 22 studies (901 patients). Meta-analysis on fourteen trials (430 patients) showed a mean difference (MD) of 2.09 points (95% confidence interval [CI] 0.86–3.33) on BBS favoring VR-BT compared to BT (low certainty evidence). Subgroup analyses showed higher balance improvement in most affected subjects (moderate certainty evidence) and using VR rehabilitation-specific systems vs. VR non-specific systems. Eight trials (236 patients) assessing mobility showed a MD of 1.55 s (95% CI 0.04–3.06) on TUG favoring VR-BT (very low certainty evidence). No differences were observed in walking speed. Estimated effects were not maintained for any outcome at follow-up. This review suggests that VR-BT is more effective than BT to improve balance in PD subjects immediately after training, particularly in individuals with higher postural instability at baseline.