Impact of Immersive Virtual Reality-Based Rehabilitation on Postural Balance, Functional Independence, Intervention Characteristics, and Clinical Outcomes Following Stroke: A Scoping Review
Keywords:
stroke; immersive virtual reality; postural balance; functional independence; neurological rehabilitation; scoping reviewAbstract
Background: Immersive virtual reality (IVR) is increasingly used in stroke rehabilitation, yet evidence on postural balance and functional independence is fragmented across mixed virtual reality (VR) modalities.
Objective: To map the evidence on VR-based rehabilitation after stroke, distinguishing immersive from non-immersive systems, with emphasis on balance, functional independence, intervention characteristics and outcomes.
Methods: A scoping review followed Joanna Briggs Institute guidance and PRISMA-ScR reporting. Seven databases, Google Scholar and reference lists were searched from January 2020 to November 2026. Two reviewers independently screened records and charted data.
Results: Of 2,230 records identified, 1,555 were screened after duplicate removal, 132 were assessed in full text and 12 systematic reviews and meta-analyses were included. Pooled estimates indicated small-to-moderate benefits for balance (Berg Balance Scale mean difference 1.35; standardised effect 0.31), gait and daily function, but most reviews combined immersive, semi-immersive, non-immersive or robot-assisted VR. The single review restricted to IVR (34 randomised trials) reported benefits for upper-limb impairment and selected gait parameters, but inconsistent effects on balance, functional performance and quality of life, with low to very low certainty. Evidence specific to middle cerebral artery stroke and dose–response was scarce, and follow-up and safety reporting were inconsistent.
Conclusions: IVR is a promising adjunct to conventional physiotherapy, but its independent contribution to balance and independence is unproven. Trials with defined immersion levels, standardised reporting, extended follow-up and stroke-subtype characterisation are needed.

