Stroke has an extremely high mortality and disability rate, making it a critical public health issue. Patients with stroke are often accompanied by lower limb motor dysfunction, which severely impairs their quality of life. In the past, one-on-one manual rehabilitation training was commonly administered to such patients, which consumes a large amount of human resources and makes it difficult to guarantee the training quality. In recent years, the advent of exoskeleton robots is expected to address this pain point. Exoskeleton robots can be classified into two major categories, fixed and mobile based on their structural forms and operational postures. However, it remains unclear whether these two distinctly designed robots are effective in improving lower limb motor function and quality of life in stroke patients, and whether there are differences in their therapeutic effects.

To investigate the effects and therapeutic benefits of stationary versus mobile exoskeleton robots in enhancing lower limb function and quality of life among stroke patients.

Forty stroke inpatients admitted to the Department of Rehabilitation Medicine at the First Medical Center of Chinese PLA General Hospital between 2021 and 2024 were enrolled as study participants. Using a random number table method, they were randomly assigned to either the fixed robot group (n=20) or the mobile robot group (n=20). In addition to conventional rehabilitation training, the fixed robot group underwent Lokomat robotic-assisted gait training for 20 minutes per session, five times weekly, over two consecutive weeks. The mobile robot group received Aikang robotic-assisted training under the same regimen: 20 minutes per day, five times per week, for two consecutive weeks, alongside standard rehabilitation. Outcome measures were collected before and after the intervention period and included the Fugl-Meyer Assessment of Lower Extremity (FMA-LE), peak torque (PT) of hip and knee flexion and extension, absolute error angle (AE) angular error of the affected knee joint, Functional Ambulation Category (FAC) scale, gait analysis parameters, and the Modified Barthel Index (MBI).

All 40 patients completed the entire trial protocol with no dropouts reported during the study. Intergroup comparisons: prior to training, there were no statistically significant differences between the two groups in FMA-LE scores, AE angle of the affected knee joint, MBI scores, PT of hip and knee flexion/extension, and FAC grades (P>0.05). After training, no statistically significant differences were found between the two groups in FMA-LE scores, MBI scores, PT of hip and knee flexion/extension, and FAC grades (P>0.05), whereas the AE angle of the affected knee joint in the fixed robot group was significantly lower than that in the mobile robot group (P<0.05). Intragroup comparisons: after training, both groups exhibited a significant increase in the lower extremity FMA-LE scores and MBI scores, with a significant improvement in the AE angle of the affected knee joint and the PT of hip and knee flexion/extension compared with the baseline (P<0.05). The FAC grades of both groups were significantly better than those at baseline (P<0.05). Gait analysis was conducted in only 7 patients in the fixed robot group and 8 patients in the mobile robot group. For gait parameters, intergroup comparisons showed no statistically significant differences in walking speed, step length and gait cycle between the two groups both before and after training (P>0.05). Intragroup comparisons indicated that walking speed was significantly increased after training in both groups (P<0.05); the fixed robot group had a significant increase in step length (P<0.05) and a significant reduction in gait cycle (P<0.05) compared with the baseline.

Both fixed exoskeleton robots and mobile exoskeleton robots can effectively improve the lower limb function and quality of life of stroke patients, and fixed exoskeleton robots have more advantages in improving patients' knee joint proprioception.

脑卒中致死率和致残率极高，是重要的公共卫生问题。脑卒中患者常伴有下肢功能障碍，严重影响了其生活质量。既往针对此类患者常进行人工一对一训练，大量消耗人力，训练质量难以保障。近年来外骨骼机器人的问世有望解决这一痛点。外骨骼机器人可依据结构形式与作业姿态分为固定式与移动式两大类型。两种风格迥异的机器人在改善脑卒中患者下肢功能和生活质量方面是否有效，并且是否存在疗效差异尚不明确。

探究固定式外骨骼机器人和移动式外骨骼机器人对脑卒中患者下肢功能和生活质量的改善效果和疗效优势。

选取2021—2024年中国人民解放军总医院第一医学中心康复医学科的40例脑卒中住院患者作为研究对象，使用随机数字表法将其分为固定机器人组（20例）和移动机器人组（20例）。固定机器人组在常规康复训练以外增加Lokomat机器人训练，20 min/d，5次/周，持续2周。移动机器人组在常规康复训练以外增加艾康机器人训练，20 min/d，5次/周，持续2周。在干预前和干预结束后分别采集各项评价指标，包括Fugl-Meyer量表下肢部分（FMA-LE）评分、髋关节屈/伸峰力矩、膝关节屈/伸峰力矩、患侧膝关节绝对误差（AE）角度、功能性步行分级量表（FAC）分级、步态分析、改良Barthel指数（MBI）。

40例患者完成全部试验流程，过程中无患者脱落。组间比较：训练前，两组患者FMA-LE评分、患侧膝关节AE角度、MBI评分，髋、膝关节屈/伸峰力矩，FAC分级比较，差异无统计学意义（P>0.05）。训练后，两组患者FMA-LE评分、MBI评分，髋、膝关节屈/伸峰力矩，FAC分级比较，差异无统计学意义（P>0.05）；固定机器人组患者患侧膝关节AE角度小于移动机器人组（P<0.05）。组内比较：两组患者训练后FMA-LE评分、MBI评分均较训练前升高，患侧膝关节AE角度及髋、膝关节屈/伸峰力矩较训练前改善（P<0.05）。两组患者训练后FAC分级优于训练前（P<0.05）。固定机器人组、移动机器人组步态分析测试仅进行了7例和8例。组间比较：训练前后，两组患者步行速度、步长、步行周期比较，差异无统计学意义（P>0.05）。组内比较：两组患者训练后步行速度均较训练前增加（P<0.05），固定机器人组患者的步长较训练前提高（P<0.05），固定机器人组的步行周期较训练前缩短（P<0.05）。

固定式外骨骼机器人和移动式外骨骼机器人均可有效改善脑卒中患者的下肢功能和生活质量，且固定式外骨骼机器人在改善患侧膝关节本体感觉方面更具优势。