OBJECTIVE To investigate the efficacy of virtual reality (VR)- and gaming-based interventions for improving upper extremity function post-stroke, and to examine demographic and treatment-related factors that may moderate treatment response. DATA SOURCES A comprehensive search was conducted within the PubMed, CINAHL/EBSCO, SCOPUS, Ovid MEDLINE and EMBASE databases for articles published between 2005 and 2019 (PROSPERO Registration number 95052). STUDY SELECTION Articles investigating gaming and VR methods of treatment for upper extremity weakness were collected with the following study inclusion criteria: 1) participants aged 18 or older with upper extremity deficits, 2) randomized controlled trials or prospective study design, 3) Downs-Black rating score of >= 18, and 4) outcome measure was the Wolf Motor Functioning Test (WMFT), the Fugl-Meyer (FM) or the Action Research Arm Test (ARAT). DATA EXTRACTION Thirty-eight articles met inclusion criteria. The primary outcome was proportional improvement on the WMFT, FM, or ARAT. The following individual or treatment factors were extracted: VR/gaming dose, total treatment dose, chronicity (> or < 6 months), severity of motor impairment, and presence of a gaming component. DATA ANALYSIS Random effects meta-analysis models were utilized to quantify 1) the proportional recovery that occurs following VR/gaming, 2) the comparative treatment effect of VR/gaming versus conventional physiotherapy, and 3) whether the benefit of virtual reality differed based on participant characteristics or elements of the treatment. RESULTS On average, VR/gaming interventions produced an improvement of 28.5% of the maximal possible improvement. Dose and severity of motor impairment did not significantly influence rehabilitation outcomes. Treatment gains were significantly larger overall (10.8%) when the computerized training involved a gaming component versus just visual feedback. VR/gaming interventions showed a significant treatment advantage (10.4%) over active control treatments. CONCLUSIONS Overall, VR/gaming-based upper extremity rehabilitation post-stroke appears to be more effective than conventional methods. Further in-depth study of variables impacting improvement, such as individual motor presentation, treatment dose, and the relationship between the two, are needed.

中文翻译：

---

基于虚拟现实和游戏的中风后上肢康复干预措施的有效性：元分析

目的 调查基于虚拟现实 (VR) 和游戏的干预措施在改善卒中后上肢功能方面的功效，并检查可能会缓和治疗反应的人口统计学和治疗相关因素。数据来源 在 PubMed、CINAHL/EBSCO、SCOPUS、Ovid MEDLINE 和 EMBASE 数据库中对 2005 年至 2019 年期间发表的文章（PROSPERO 注册号 95052）进行了全面搜索。研究选择 根据以下研究纳入标准收集调查游戏和 VR 治疗上肢无力的方法的文章：1) 18 岁或以上有上肢缺陷的参与者，2) 随机对照试验或前瞻性研究设计，3) Downs-Black评分 >= 18，和 4) 结果测量是 Wolf 运动功能测试 (WMFT)，Fugl-Meyer (FM) 或 Action Research Arm Test (ARAT)。数据提取 三十八篇文章符合纳入标准。主要结果是 WMFT、FM 或 ARAT 的比例改善。提取了以下个体或治疗因素：VR/游戏剂量、总治疗剂量、慢性（> 或 < 6 个月）、运动障碍的严重程度和游戏组件的存在。数据分析 随机效应荟萃分析模型被用来量化 1) 虚拟现实/游戏后发生的比例恢复，2) 虚拟现实/游戏与传统理疗的比较治疗效果，以及 3) 虚拟现实的益处是否因治疗的参与者特征或要素。结果 平均而言，VR/游戏干预产生了最大可能改进的 28.5%。运动障碍的剂量和严重程度对康复结果没有显着影响。当计算机化培训涉及游戏组件而不是仅视觉反馈时，治疗收益总体上显着更大 (10.8%)。VR/游戏干预比主动控制治疗显示出显着的治疗优势（10.4%）。结论 总体而言，基于 VR/游戏的中风后上肢康复似乎比传统方法更有效。需要进一步深入研究影响改善的变量，例如个体运动表现、治疗剂量以及两者之间的关系。8%) 当计算机化培训涉及游戏组件而不是视觉反馈时。VR/游戏干预比主动控制治疗显示出显着的治疗优势（10.4%）。结论 总体而言，基于 VR/游戏的中风后上肢康复似乎比传统方法更有效。需要进一步深入研究影响改善的变量，例如个体运动表现、治疗剂量以及两者之间的关系。8%) 当计算机化培训涉及游戏组件而不是视觉反馈时。VR/游戏干预比主动控制治疗显示出显着的治疗优势（10.4%）。结论 总体而言，基于 VR/游戏的中风后上肢康复似乎比传统方法更有效。需要进一步深入研究影响改善的变量，例如个体运动表现、治疗剂量以及两者之间的关系。