Earticle

초록

영어

This study aims to synthesize randomized evidence regarding the effects of virtual reality (VR)–based rehabilitation following stroke on UL performance, gait and mobility, and functional independence, while also examining potential moderators such as immersion level, timing post-stroke, dosage, and comparator intensity. A comprehensive search was conducted across MEDLINE, Embase, Cochrane CENTRAL, Scopus, Web of Science, CINAHL, PEDro, ClinicalTrials.gov, and WHO ICTRP (January 2000–August 2025) without language restrictions. Eligible studies included randomized or quasi-randomized trials comparing VR interventions (non-, semi-, or fully immersive), either as standalone treatments or adjuncts, against usual care or active controls in adult stroke populations. Two independent reviewers performed study screening, data extraction, and risk of bias assessment using the RoB 2 tool; the certainty of evidence was evaluated with GRADE. When outcome measures were sufficiently homogeneous, random-effects meta-analyses were conducted employing Hedges’ g (standardized mean difference) or mean difference for common scales, with heterogeneity assessed via I² and τ² statistics; sensitivity and subgroup analyses were prespecified. Ten trials met inclusion criteria, encompassing inpatient, outpatient, and home/telerehabilitation settings and diverse VR platforms (including screen-based exergames, treadmill/robot-assisted VR, real-instrument VR, and immersive systems). The overall risk of bias was low in 3 trials, high in 2, and some concerns in 5. Where data pooling or consistent directional effects were feasible, results generally favored VR interventions, demonstrating small-to-moderate improvements in UL motor performance (e.g., FMA-UE, dexterity tests) and gait/mobility (e.g., TUG, 10-/6-Minute Walk Test, balance scales). Between-group differences in functional independence and participation measures (e.g., FIM, Barthel, USER-P) were smaller and more variable, particularly when control groups received active, dose-matched rehabilitation. Adverse events were infrequent and mild, including transient dizziness and headache; adherence and acceptability were generally high. VR represents a safe and practical adjunctive tool that can increase practice dosage and engagement, potentially leading to improved functional outcomes following stroke (most notably in UL function and gait/mobility) when combined with conventional, goal-directed therapy. Recommended pragmatic dosing (≥3 sessions/week for ≥4–6 weeks, 30–60 minutes/session), task-oriented content and careful patient selection. Future adequately powered, preregistered trials employing core outcome sets, dose-matched active comparators, and extended follow-up periods are warranted to elucidate effects on functional independence, participation, and cost-effectiveness.

목차

Abstract
Ⅰ. Introduction
Ⅱ. Methods
1. Protocol and reporting standards
2. Search strategy and study selection
3. Data items and outcomes
4. Risk of bias and certainty assessment
5. Synthesis and statistical analysis
Ⅲ. Results
1. UL performance
2. Gait outcomes
3. Functional independence
Ⅳ. Discussion
Ⅴ. Conclusion
References

키워드

저자

• Jae-Hun Shim [ Department of Physical Therapy, College of Health Science, Baekseok University ] Corresponding Author

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  한국신경근육재활학회지 [Korean Journal of Neuromuscular Rehabilitation]

• 간기

  계간

• pISSN

  2508-6456

• 수록기간

  2011~2026

• 등재여부

  KCI 등재

• 십진분류

  KDC 514 DDC 617