Comparison of robot-assisted gait training, body weight-supported treadmill training, and conventional gait training in patients with chronic stroke using gait analysis and surface electromyography measurements

This study aimed to compare the effects of robot-assisted walking, body weight-supported treadmill training, and conventional gait training on walking function, gait patterns, and muscle activity in patients with unilateral ischemic stroke, and to determine whether any of these approaches demonstrated superiority over the others. A total of 48 patients who met the eligibility criteria and applied to the Department of Physical Medicine and Rehabilitation at Pamukkale University Faculty of Medicine were included in the study. Participants were randomized into three groups: the first group received robot-assisted gait training using RoboGait, the second group underwent body weight-supported treadmill training with Zebris RehaWalk®, and the third group received conventional gait training. Pre- and post-treatment assessments included clinical evaluations as well as kinematic analysis with a two-dimensional camera system, surface electromyography (sEMG), plantar pressure analysis, and gait analysis. At the end of the intervention, all groups demonstrated significant improvements in the 6-Minute Walk Test (6MWT), Functional Ambulation Scale (FAS), and Fugl-Meyer Assessment (FMA) scores (p<0.05); however, no significant differences were found between the groups in terms of clinical measures (p>0.05). In the kinematic analysis, significant increases in joint range of motion were observed in the RoboGait group for knee flexion and ankle dorsiflexion on both sides (p<0.05), with no change in hip flexion. In the RehaWalk group, all joints on the affected side showed significant improvements, and on the unaffected side, both hip flexion and ankle dorsiflexion increased significantly (p=0.036; 0.035). In the conventional group, only hip flexion and ankle dorsiflexion on the affected side showed significant improvement (p=0.023; p=0.002). There was no significant intergroup difference in joint range of motion after treatment (p>0.05). sEMG analyses showed significant increases in muscle activation amplitudes of the biceps femoris, tibialis anterior, and gastrocnemius in all groups, with no significant difference between them (p>0.05). However, only the RehaWalk group showed significant reductions in co-contraction indices for knee and ankle muscles (p<0.05). There was no significant intergroup difference in sEMG analyses after treatment (p>0.05). In gait analysis, the RehaWalk group showed statistically significant changes in multiple parameters: double support phase, gait speed, double step length, double step duration, stance phase, swing phase, single support phase, and step length on both sides, as well as reduced step duration on the affected side. The RoboGait group showed significant improvements in double step length, gait speed, and step length on both sides. The conventional group showed significant improvements in gait speed, double step length, double support phase, step length on the unaffected side, and single support phase on the affected side. In pressure analysis, gait line length significantly increased in all groups. However, single support trajectory length significantly increased on both sides in the RehaWalk group, and only on the unaffected side in the RoboGait group. Additionally, the RehaWalk group demonstrated significant improvements in anterior-posterior positioning and lateral symmetry. In between-group comparisons, the RehaWalk group demonstrated significantly higher values in double step length, bilateral step length, and single support trajectory on the affected side compared to the conventional group (p<0.05). In conclusion, all three gait rehabilitation approaches applied to patients with chronic stroke showed positive effects on both clinical outcomes and objective parameters. Among these, the Zebris RehaWalk® system provided statistically significant and distinct improvements, particularly in spatiotemporal and pressure-related variables such as step length, double step length, and single support trajectory, when compared to conventional approaches. With its technology that supports visual feedback, pressure-guided control, and active patient participation, the Zebris RehaWalk® system can be considered a more effective alternative than traditional methods in improving gait quality and symmetry in chronic stroke patients. Keywords: Stroke rehabilitation, robotic rehabilitation, body weight-supported treadmill training, gait analysis, surface electromyography