Virtual Reality-Enhanced Partial Body Weight-Supported Treadmill Training Poststroke: Feasibility and Effectiveness in 6 Participants.

Arch Phys Med Rehabil, Jan 2010; 91:115-122.
Walker, Martha L; Ringleb, Stacie I; Maihafer, George C; Walker, Robert; Crouch, Jessica; VanLunen, Bonnie; Morrison, Steven

Summary by John Kelly, PTA, BS

The purpose of this study was to determine the effectiveness of a low-cost virtual reality system paired with partial body weight supported treadmill training (BWSTT) on balance and walking ability in participants poststroke. The investigators used a before-after comparison of a group of participants less than one year post right ischemic CVA. The authors concluded that the combination of BWSTT and a simple virtual reality system improves performance of walking speed and duration. Further, it was concluded that the intervention was effective when delivered at typical outpatient intensity, and that a low cost virtual reality system is feasible for clinic use.

The participants participating in this study were between 49 and 70 years of age, and between 3 and 38 weeks post stroke. All participants were able to walk 20 feet independently with or without an assistive device. All participants had been discharged from inpatient and outpatient physical therapy prior to participating in the study, and had not participated in treadmill training. Measures used to determine balance and walking ability were the Functional Gait Assessment, overground walking speed, and the Berg Balance Scale. Mean treadmill walking speed, treadmill walking duration, and percentage of body weight supported were also reported.

The study was conducted over 12 visits, completed 2-3 times per week, similar to the intensity common in outpatient physical therapy clinics. The total training time of the participants was less than 4 hours, meaning sessions averaged approximately 20 minutes. This dosage was selected by the researchers to mimic the typical schedule of an outpatient physical therapy clinic.

The intervention was provided using body weight support over a treadmill, paired with a 51 inch video display screen placed directly in front of the participants. Participants wore an inertial orientation tracking device attached to a baseball cap. The virtual reality system was based on custom designed software that is not yet commercially available, but was designed to be a low-cost product that may be feasible for use in a physical therapy clinic. The feedback display was presented at a constant rate of speed (4mph), despite the rate of participant walking, and consisted of a video representation of a city street. The program included a virtual walking companion which provided auditory feedback when initiated by the therapist and inertial tracking device, including encouragement, distances walked, and data relative to posture.

Body weight support was initially provided at 20-40% of the participant’s body weight, and initial training duration was 10 minutes. During subsequent training sessions, participants were subjected to 5 minutes at a “slow walking speed” based on the participant’s overground walking speed. The treadmill then was programmed to increase speed “for a few minutes” at a faster, challenging speed, after which a moderate pace was assumed for the remainder of the intervention, which was .2mph faster than the initial speed. The starting speed for the next session was determined by the physical therapist based on patient tolerance and vital signs at the end of the previous session. When deciding which parameters to progress, more emphasis was placed on increasing duration rather than the speed of walking. Body weight support was also decreased throughout the intervention, from an average of 31.6% initially, to an average of 18.33% of the subject’s body weight supported at the end of the study. The researchers did not report provision of any manual assistance or facilitation during the intervention.

The study started with 7 participants, with 1 withdrawing after 11 treatment sessions. The reason for the subject’s withdrawal was not given, but there were no adverse affects reported. Participants increased overground walking speed by an average of 38%, improved Berg Balance Test score by an average of 10%, and increased Functional Gait Assessment scores by an average of 30%. Two participants discontinued the use of their cane, and one progressed from a quad cane to a straight cane. In addition to very positive improvements in functional objective measures, the researchers concluded that the equipment used, patient schedule, and therapist intervention strategies were feasible for implementation within a typical outpatient physical therapy clinic.