PHYSIOLOGICAL RESPONSES OF INDIVIDUALS WITH SPINAL CORD INJURY DURING ROBOTIC-ASSISTED TREADMILL WALKING.

Abstract

Purpose/Hypothesis: Body weight supported treadmill training (BWSTT) has been shown to be effective in retraining walking following incomplete spinal cord injury (SCI), although its performance in the clinic is limited. Development of robotic devices which provide passive guidance during walking may increase performance of BWSTT, but may minimize voluntary effort by the subject. The purpose of this study was to investigate whether robotic-assisted treadmill walking provides a sufficient stimulus to generate appropriate muscle activity and metabolic responses compared to therapist-assisted walking. We hypothesized that therapist-assisted treadmill walking would generate increased Vo2 responses and more appropriate lower extremity muscle activity vs., robotic-assisted treadmill walking. Number of Subjects: Twelve individuals with motor incomplete SCI participated in the study. Materials/Methods: Cardiopulmonary, metabolic, and electromyographic (EMG) responses during therapist-assisted treadmill walking were compared to robotic-assisted ambulation with and without visual biofeedback. Robotic training was provided by the Lokomat and visual feedback of bilateral hip and knee torques during swing and stance phases were displayed on a computer screen. Subjects were asked to walk on the treadmill with <40% BWS for 10 minutes with either robotic- or therapist-assistance followed by 10 minutes of the other testing method. Standardized metabolic testing equipment was used to obtain cardiopul-monary measurements for 5 minutes in sitting and 2 minutes in standing prior to treadmill walking and throughout the testing period including 10 minutes of recovery. Electromyographic (EMG) activity was collected from key lower extremity muscles during each minute of treadmill walking. Results: Therapist-assisted treadmill walking demonstrated significantly higher (p < 0.01) Vo2, heart rate (HR), and minute ventilation (VE) during all 10 minutes compared with robotic-assisted walking without biofeedback. Providing visual biofeedback during the walking tasks minimized the differences between conditions, although Vo2 responses during therapist-assisted walking were greater (p < 0.05) during the last 3 minutes of walking. EMG activity was significantly higher only for the rectus femoris in pre-swing during therapist- vs. robotic-assisted walking without feedback. This difference was minimized with feedback. Conclusions: Therapist-assisted treadmill walking elicits increased aerobic responses and more appropriate muscle activity compared to robotic-assisted treadmill walking, although provision of biofeedback of joint torques may increase patient effort. Clinical Relevance: Understanding the physiological responses to robotic- or therapist-assisted BWSTT is important the relative contribution of passive guidance during walking tasks.