Evolving trends of systematic reviews on virtual reality for stroke rehabilitation

Objective: Virtual reality (VR) therapies have increasingly been adopted across medical fields, including neurorehabilitation for stroke recovery. Evidence from several systematic reviews (SRs) was explored, covering different aspects. We aim to explore overlaps, gaps, and trends of SRs focusing on VR stroke rehabilitation providing a foundation for improving the field and addressing current limitations. Materials and methods: We moved from a recent overview of reviews, searching multiple databases for all published SRs and the international database of prospectively registered SRs (PROSPERO) for ongoing SRs. Data extraction of study characteristics and methodological quality of SRs using AMSTAR 2 were obtained from a recent overview of reviews. Two independent reviewers conducted data analysis and visualization by the trend over time of published SRs with their included primary studies and ongoing SRs, methodological quality and other SR characteristics. Results: The data set consisted of 58 SRs, including 345 primary studies and 45 ongoing SRs, published between 2007 and 2022. The number of published and ongoing SRs significantly increased over time (R2 = 0.8654; R2 = 0.747, respectively). In the last three years, Asia accounts for the majority of publications (31%). Overall, the main outcome assessed over time was upper extremity function and activity in 67.2% of SRs. Most of the published SRs were judged "critically low" (77.6%). The number of included studies increased over time reaching a median of 17 studies with a median of 493 participants. Conclusions: In stroke rehabilitation, the published and ongoing SRs on VR have risen over time in terms of the number of publications, with some concerns about methodological quality and representation of countries around the world.