Feasibility of a Virtual Reality Interface for Shaping Compensation During Motor Rehabilitation

PURPOSE: The objective of this study was to determine the feasibility and preliminary efficacy of a virtual reality (VR)–based intervention for improving upper-extremity range of motion (ROM) and decreasing compensatory trunk movement in persons with chronic hemiparesis due to stroke. Repetitive, task-based practice is the prevailing strategy for inducing neuroplastic change and improving upper-extremity function, activity performance, and participation in persons with stroke; however, the required repetition dose is often not achieved in clinical settings, clients can become bored, and maladaptive compensatory movements are often not addressed. A customizable VR interface, namely VRShape, was developed to engage clients to perform high doses of reaching movements while shaping the amount of trunk compensation used through objective monitoring and real-time feedback.

DESIGN: A small sample of persons with chronic stroke (N = 5) were recruited from the St. Louis area to test the feasibility and preliminary efficacy of using VRShape as an intervention. Participants took part in 18 sessions over the course of 6 wk during which they performed reaching movements for 1 hr while playing video games of their choice. Participants were ages 44–78 yr, were at least 6 mo poststroke, had mild to moderate impairment (National Institutes of Health Stroke Scale arm/motor scores of 1–3), and exhibited at least 20° of compensatory trunk flexion.

METHOD: Outcomes related to feasibility were collected at each session using the System Usability Scale (SUS), Intrinsic Motivation Inventory (IMI), and ITC–Sense of Presence Inventory (ITC–SOPI). Outcomes related to preliminary efficacy were collected pre- and postintervention and included a functional reach test recorded and analyzed with an eight-camera video motion capture system and the Action Research Arm Test (ARAT). Repetitions and movement kinematics were also recorded at each session using VRShape.

RESULTS: Participants completed 90 total sessions without major technical issues and completed an average of 512.12 (SD = 95.17) reaching repetitions within each 1-hr session. The system was found to be usable, motivating, engaging, and safe as represented by excellent scores on the SUS (M = 85.19, SD =18.75), high IMI interest/enjoyment scores relative to other systems (M = 43.16, SD = 8.80), high ITC–SOPI engagement scores (M = 3.94, SD = 1.02), and low ITC–SOPI negative effects scores (M = 1.29, SD = 0.66). Preliminary analysis of efficacy outcomes related to reaching ROM, shoulder ROM, and trunk ROM showed no statistically significant improvement (p > .05); however, further analysis must be performed. While some participants improved their individual ARAT scores, at the group level there was no statistically significant improvement from pre- to postintervention (p > .05).

CONCLUSION: VRShape is a usable, motivating, engaging, and safe intervention that elicits a very large dose of movement repetitions in a small amount of time. Participants were able to take part in an intense intervention using this system with no major technical issues. Preliminary analysis shows no statistical improvement related to efficacy; however, further analysis must be performed. These results show that VRShape may provide the basis for a useful VR-based motor intervention.