外骨骼辅助仿人上肢运动训练治疗脑卒中后上肢运动障碍：EAMT 随机试验

Exoskeleton-Assisted Anthropomorphic Movement Training for the Upper Limb After Stroke: The EAMT Randomized Trial.

机构信息

Department of Rehabilitation Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China (Z.-J.C., J.X., N.X., X.-L.H.).

World Health Organization Cooperative Training and Research Center in Rehabilitation, Wuhan, China (Z.-J.C., J.X., N.X., X.-L.H.).

出版信息

Stroke. 2023 Jun;54(6):1464-1473. doi: 10.1161/STROKEAHA.122.041480. Epub 2023 May 8.

DOI:10.1161/STROKEAHA.122.041480

PMID:37154059

Abstract

BACKGROUND

Robot-assisted arm training is generally delivered in the robot-like manner of planar or mechanical 3-dimensional movements. It remains unclear whether integrating upper extremity (UE) natural coordinated patterns into a robotic exoskeleton can improve outcomes. The study aimed to compare conventional therapist-mediated training to the practice of human-like gross movements derived from 5 typical UE functional activities managed with exoskeletal assistance as needed for patients after stroke.

METHODS

In this randomized, single-blind, noninferiority trial, patients with moderate-to-severe UE motor impairment due to subacute stroke were randomly assigned (1:1) to receive 20 sessions of 45-minute exoskeleton-assisted anthropomorphic movement training or conventional therapy. Treatment allocation was masked from independent assessors, but not from patients or investigators. The primary outcome was the change in the Fugl-Meyer Assessment for Upper Extremity from baseline to 4 weeks against a prespecified noninferiority margin of 4 points. Superiority would be tested if noninferiority was demonstrated. Post hoc subgroup analyses of baseline characteristics were performed for the primary outcome.

RESULTS

Between June 2020 and August 2021, totally 80 inpatients (67 [83.8%] males; age, 51.9±9.9 years; days since stroke onset, 54.6±38.0) were enrolled, randomly assigned to the intervention, and included in the intention-to-treat analysis. The mean Fugl-Meyer Assessment for Upper Extremity change in exoskeleton-assisted anthropomorphic movement training (14.73 points; [95% CI, 11.43-18.02]) was higher than that of conventional therapy (9.90 points; [95% CI, 8.15-11.65]) at 4 weeks (adjusted difference, 4.51 points [95% CI, 1.13-7.90]). Moreover, post hoc analysis favored the patient subgroup (Fugl-Meyer Assessment for Upper Extremity score, 23-38 points) with moderately severe motor impairment.

CONCLUSIONS

Exoskeleton-assisted anthropomorphic movement training appears to be effective for patients with subacute stroke through repetitive practice of human-like movements. Although the results indicate a positive sign for exoskeleton-assisted anthropomorphic movement training, further investigations into the long-term effects and paradigm optimization are warranted.

REGISTRATION

URL: https://www.chictr.org.cn; Unique identifier: ChiCTR2100044078.

摘要

背景

机器人辅助手臂训练通常以平面或机械三维运动的机器人方式进行。目前尚不清楚将上肢（UE）自然协调模式整合到机器人外骨骼中是否可以改善结果。本研究旨在比较传统治疗师介导的训练与从 5 种典型 UE 功能活动中获得的类似人的粗大运动实践，这些活动根据需要在卒中后患者中使用外骨骼辅助进行管理。

方法

在这项随机、单盲、非劣效性试验中，由于亚急性卒中而导致 UE 运动障碍程度中度至重度的患者被随机分配（1:1）接受 20 次 45 分钟的外骨骼辅助拟人运动训练或常规治疗。治疗分配对独立评估者是盲的，但对患者或研究者不是盲的。主要结局是从基线到 4 周时 Fugl-Meyer 上肢评估的变化，与预先指定的 4 分非劣效性边界相比。如果证明非劣效性，则进行优越性检验。对主要结局进行了基线特征的事后亚组分析。

结果

2020 年 6 月至 2021 年 8 月期间，共纳入 80 名住院患者（67 名[83.8%]男性；年龄 51.9±9.9 岁；卒中发病后天数 54.6±38.0），随机分组并进行意向治疗分析。外骨骼辅助拟人运动训练（14.73 分；[95%CI，11.43-18.02]）的 Fugl-Meyer 上肢评估的平均变化高于常规治疗（9.90 分；[95%CI，8.15-11.65]）在 4 周时（调整后的差异，4.51 分[95%CI，1.13-7.90]）。此外，事后分析支持运动障碍程度中度至重度（Fugl-Meyer 上肢评估评分，23-38 分）的患者亚组。