The Combined Impact of Virtual Reality Neurorehabilitation and Its Interfaces on Upper Extremity Functional Recovery in Patients With Chronic Stroke

BACKGROUND AND PURPOSE: Although there is strong evidence on the beneficial effects of virtual reality (VR)-based rehabilitation, it is not yet well understood how the different aspects of these systems affect recovery. Consequently, we do not exactly know what features of VR neurorehabilitation systems are decisive in conveying their beneficial effects. METHODS: To specifically address this issue, we developed 3 different configurations of the same VR-based rehabilitation system, the Rehabilitation Gaming System, using 3 different interface technologies: vision-based tracking, haptics, and a passive exoskeleton. Forty-four patients with chronic stroke were randomly allocated to one of the configurations and used the system for 35 minutes a day for 5 days a week during 4 weeks. RESULTS: Our results revealed significant within-subject improvements at most of the standard clinical evaluation scales for all groups. Specifically we observe that the beneficial effects of VR-based training are modulated by the use/nonuse of compensatory movement strategies and the specific sensorimotor contingencies presented to the user, that is, visual feedback versus combined visual haptic feedback. CONCLUSIONS: Our findings suggest that the beneficial effects of VR-based neurorehabilitation systems such as the Rehabilitation Gaming System for the treatment of chronic stroke depend on the specific interface systems used. These results have strong implications for the design of future VR rehabilitation strategies that aim at maximizing functional outcomes and their retention. Clinical Trial Registration- This trial was not registered because it is a small clinical study that evaluates the feasibility of prototype devices.