Effects of a robot-assisted sensorimotor training program with vibro-tactile feedback on proprioception and motor function in people with chronic stroke

Abstract

Proprioceptive deficits are exhibited in nearly 50% of all stroke survivors and is associated with poor upper limb motor function and impaired activities of daily living function. Improving proprioception enhances motor learning and improves motor performance in non-stroke populations. Thus, improving proprioception could serve as an additional route to enhance motor recovery after stroke. The aims of this study were to examine whether a robot-aided sensorimotor training regimen requiring active wrist movements administered without vision would improve the proprioceptive acuity, proprioception-related somatosensory evoked potential (SEP) measures and motor performance in adults with chronic stroke. METHODS: Twelve adults with chronic stroke were recruited (Median age: 63 years, 42 – 74 years; median time after stroke: 12 months; median Fugl-Meyer Assessment – UE: 65 points). Participants completed two training sessions in two consecutive days (total training time: 1 hour). Users grasped the robot handle and performed wrist adduction/abduction movements to tilt a virtual board on which a virtual ball rolled. Users were tasked to roll the ball to a target area on the board. Real-time, vibro-tactile feedback about ball position and speed was provided to the forearm. The task difficulty increased as the user continued training. Assessments were conducted before, immediately after, and two days after the intervention. Outcome measures were wrist proprioceptive acuity indicated by the just-noticeable-difference (JND) threshold, spatial errors of wrist tracing tasks, movement time and endpoint error of a wrist pointing task and short-latency SEPs induced by median nerve stimulation. RESULTS: The stroke group significantly reduced JND thresholds at posttest and retention in comparison to the pretest (Medians: pretest: 1.8°, posttest: 1.4°, retention: 1.3°; W = 10, p = 0.0042 for both comparisons). Higher reduction in the JND threshold was associated with a higher increase in the P27-N30 peak-to-peak SEP amplitude at retention compared to pretest. Changes in SEP measures and motor measures across visits did not reach statistical significance. DISCUSSION: This exploratory, proof-of-concept study documented that proprioceptive function of adults with chronic stroke improved after a brief active proprioceptive-motor training. If proven effective, such interventions or elements could be employed in clinical practice in addition to existing rehabilitation approaches.