Virtual reality-based rehabilitation for upper extremity recovery after stroke: a systematic review and meta-analysis of randomized controlled trials

Background Stroke often results in upper extremity motor impairments, limiting functional independence. Conventional rehabilitation, while effective, faces challenges such as low engagement and high resource demand. Virtual reality (VR)-based rehabilitation may promote neuroplasticity through immersive and repetitive training.Objectives To systematically evaluate randomized controlled trials (RCTs) assessing the effectiveness of VR-based rehabilitation on upper extremity motor recovery in stroke patients, and to explore the impact of system type, stroke chronicity, and intervention dosage.Methods Following PRISMA 2020 guidelines (PROSPERO: CRD420251065021), a systematic search was conducted in PubMed, Scopus, Web of Science, and CENTRAL for RCTs published between 2010 and March 2025. Studies included adult stroke survivors receiving VR interventions targeting upper limb recovery. A random-effects meta-analysis calculated standardized mean differences (SMDs), and risk of bias was assessed using RoB 2.0.Results Twenty-four RCTs (n = 1,627) were included. Meta-analysis of 21 studies showed a moderate effect favoring VR (SMD = 0.58; 95% CI: 0.38–0.78; p < 0.001; I2 = 62%). Immersive VR, chronic stroke, and higher intervention doses (>600 min) showed stronger effects.Conclusions VR-based rehabilitation significantly enhances upper extremity motor function post-stroke. Immersive systems and sufficient dosage yield optimal outcomes, supporting VR as a valuable adjunct to conventional therapy.