Robot-Assisted Gait Training in Patients With Parkinson Disease: A Randomized Controlled Trial

Background. Gait impairment is a common cause of disability in Parkinson disease (PD). Electromechanical devices to assist stepping have been suggested as a potential intervention. Objective. To evaluate whether a rehabilitation program of robot-assisted gait training (RAGT) is more effective than conventional physiotherapy to improve walking. Methods. A total of 41 patients with PD were randomly assigned to 45-minute treatment sessions (12 in all), 3 days a week, for 4 consecutive weeks of either robotic stepper training (RST; n = 21) using the Gait Trainer or physiotherapy (PT; n = 20) with active joint mobilization and a modest amount of conventional gait training. Participants were evaluated before, immediately after, and 1 month after treatment. Primary outcomes were 10-m walking speed and distance walked in 6 minutes. Results. Baseline measures revealed no statistical differences between groups, but the PT group walked 0.12 m/s slower; 5 patients withdrew. A statistically significant improvement was found in favor of the RST group (walking speed 1.22 +/- 0.19 m/s [P = .035]; distance 366.06 +/- 78.54 m [P <.001]) compared with the PT group (0.98 +/- 0.32 m/s; 280.11 +/- 106.61 m). The RAGT mean speed increased by 0.13 m/s, which is probably not clinically important. Improvements were maintained 1 month later. Conclusions. RAGT may improve aspects of walking ability in patients with PD. Future trials should compare robotic assistive training with treadmill or equal amounts of overground walking practice.