Effects of virtual reality training on gait biomechanics of individuals post-stroke

Objective: To evaluate gait biomechanics after training with a virtual reality (VR) system and to elucidate underlying mechanisms that contributed to the observed functional improvement in gait speed and distance. Design: A single blind randomized control study. Setting: Gait analysis laboratory in a rehabilitation hospital and the community. Participants: Fifteen men and three women with hemiparesis caused by stroke. Interventions: Subjects trained on a six-degree of freedom force-feedback robot interfaced with a VR simulation. Subjects were randomized to either a VR group (n = 9) or non-VR group (NVR, n = 9). Training was performed three times a week for 4 weeks for approximately 1 h each visit. Main outcome measures: Kinematic and kinetic gait parameters. Results: Subjects in the VR group demonstrated a significantly larger increase in ankle power generation at push-off as a result of training (p = 0.036). The VR group had greater change in ankle ROM post-training (19.5%) as compared to the NVR group (3.3%). Significant differences were found in knee ROM on the affected side during stance and swing, with greater change in the VR group. No significant changes were observed in kinematics or kinetics of the hip post-training. Conclusions: These findings are encouraging because they support the potential for recovery of force and power of the lower extremity for individuals with chronic hemiparesis. It is likely that the effects of training included improved motor control at the ankle, which enabled the cascade of changes that produced the functional improvements seen after training. (C) 2010 Elsevier B.V. All rights reserved.