Effect of Time-Dose-Matched Virtual Reality Therapy on Upper Limb Dysfunction in Patients Poststroke: A Meta-Analysis of Randomized Controlled Trials

To investigate the efficacy and acceptability of virtual reality (VR) with time-dose-matched conventional therapy (CT) in patients poststroke with upper limb dysfunction.Cochrane, PubMed, Web of Science, Embase, and ProQuest were systematically searched up to May 24, 2021.Randomized controlled trials comparing VR with time-dose-matched CT in patients poststroke with upper limb dysfunction were included.The extracted data included efficacy (mean change in structure/function, activity, and participation scores), acceptability (dropouts for all reasons), adverse events, and characteristics of the included studies. The Cochrane risk of bias assessment tool was used to assess the risk of bias.Thirty-one randomized controlled trails were included. VR was superior to time-dose-matched CT in terms of the World Health Organization's International Classification of Functioning, Disability and Health structure/function, with a standardized mean difference (SMD) of 0.35, but not activity and participation. Subgroup analyses demonstrated that virtual environment was superior to CT in structure/function (SMD=0.38) and activity (SMD=0.27), whereas there were no significant differences between commercial gaming and CT in any World Health Organization International Classification of Functioning, Disability and Health domain. VR mixed with CT was more effective than time-dose-matched CT in structure/function (SMD=0.56), whereas VR only was not significantly different from CT. There were no significant differences in the incidence of adverse events and dropout rates between VR and CT.The results suggest that VR is superior to time-dose-matched CT in terms of recovery of upper extremity motor function, especially when a virtual environment is used or VR is mixed with CT. However, VR (VR only or mixed with CT) does not improve patients' daily activity performance and participation compared with CT. Overall, VR appears to be safe and acceptable as CT. Large-scale definitive trials are needed to verify or refute these findings.