Liu Shengye, Chen Fangyuan, Yin Jianqiao, Wang Guanqi, Yang Liyu
Department of Orthopedics, Shengjing Hospital of China Medical University, Sanhao Street 36, Shenyang, Liaoning, People's Republic of China.
Department of Rehabilitation, Shengjing Hospital of China Medical University, Puhe Road 16, Shenyang, Liaoning, People's Republic of China.
J Neuroeng Rehabil. 2025 May 29;22(1):121. doi: 10.1186/s12984-025-01649-1.
The purpose of this meta-analysis was to investigate the effects of Robotic exoskeleton gait training (REGT) on lower limb mobility, walking balance, functional scores and respiratory function in patients with spinal cord injury (SCI).
The PubMed, Embase, Cochrane Library databases were systematically searched from inception until December 24, 2024.
Eligible randomized controlled trials contained information on the population (SCI), intervention (REGT), and outcomes (walking speed and distance, walking balance, functional scores for SCI rehabilitation, respiratory function). Participants in the REGT intervention group were compared with those in conventional physical gait training (CPT) groups. Two independent researchers conducted the research,screened the articles, and assessed their eligibility.
Two independent researchers extracted key information from each eligible study. The authors' names, year of publication, setting, total sample size, REGT, CPT training schedule, baseline/mean difference (MD), and 95% confidence interval (CI) were extracted using a standardized form, and the methodological quality was assessed using the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) system.
Of 595 studies identified, 15 randomized controlled trials (n = 579) were included for meta-analysis. Compared with conventional physical gait training (CPT), REGT showed no significant efficacy in walking speed (10-Meter Walk Test, WMD (95%CI) = - 0.03 (- 0.06, 0.00) m/s, P = 0.08) and walking distance, (6-Minute Walk Test, WMD (95% CI) = -1.83 (- 14.48, 10.83) meters, P = 0.78). REGT showed statistically significant efficacy in walking stability (Timed Up and Go, WMD (95%CI) = 6.62 (0.35, 12.88) s, P = 0.04) and functional scores such as Walking Index for Spinal Cord Injury Version II (WMD (95%CI) = 2.17 (1.05, 3.29), P = 0.0001) and Lower Extremity Motor Score (WMD (95%CI) = 1.33 (0.58, 2.07), P = 0.0005). Additional Significant efficacy was also found in terms of respiratory function (forced expiratory volume in one second, WMD (95%CI) = 0.60 (0.05, 1.16) L, P = 0.03).
This meta-analysis discovered the evidence that robotic exoskeleton gait training can improve the walking balance, strength of lower limbs, functional scores and respiratory function in the patients with spinal cord injury (SCI) compared to conventional gait training (CPT). No obvious evidence showed that REGT has more advantages than CPT in improving walking speed and distance. REGT combined with CPT are more recommended in the discovery of walking speed and distance of patients above 6 months after SCI.
本荟萃分析旨在研究机器人外骨骼步态训练（REGT）对脊髓损伤（SCI）患者下肢活动度、行走平衡、功能评分及呼吸功能的影响。
对PubMed、Embase、Cochrane图书馆数据库进行系统检索，检索时间从建库至2024年12月24日。
符合条件的随机对照试验需包含有关研究对象（SCI）、干预措施（REGT）及结局指标（步行速度和距离、行走平衡、SCI康复功能评分、呼吸功能）的信息。将REGT干预组的参与者与传统物理步态训练（CPT）组的参与者进行比较。由两名独立研究人员开展研究、筛选文章并评估其合格性。
两名独立研究人员从每项合格研究中提取关键信息。使用标准化表格提取作者姓名、发表年份、研究地点、总样本量、REGT、CPT训练方案、基线/平均差值（MD）及95%置信区间（CI），并使用GRADE（推荐分级、评估、制定与评价）系统评估方法学质量。
在检索到的595项研究中，纳入15项随机对照试验（n = 579）进行荟萃分析。与传统物理步态训练（CPT）相比，REGT在步行速度（10米步行试验，加权均数差（WMD）（95%CI）=-0.03（-0.06，0.00）m/s，P = 0.08）和步行距离（6分钟步行试验，WMD（95%CI）=-1.83（-14.48，10.83）米，P = 0.78）方面未显示出显著疗效。REGT在行走稳定性（计时起立行走试验，WMD（95%CI）= 6.62（0.35，12.88）秒，P = 0.04）以及功能评分如脊髓损伤步行指数II版（WMD（95%CI）= 2.17（1.05，3.29），P = 0.0001）和下肢运动评分（WMD（95%CI）= 1.33（0.58，2.07），P = 0.0005）方面显示出统计学显著疗效。在呼吸功能方面（一秒用力呼气量，WMD（95%CI）= 0.60（0.05，1.16）L，P = 0.03）也发现了额外的显著疗效。
本荟萃分析发现，与传统步态训练（CPT）相比，机器人外骨骼步态训练可改善脊髓损伤（SCI）患者的行走平衡、下肢力量、功能评分及呼吸功能。没有明显证据表明REGT在提高步行速度和距离方面比CPT更具优势。对于SCI后6个月以上患者的步行速度和距离改善，更推荐REGT与CPT相结合。
