Clinical Relevance for the Veteran

Purpose of the Work. A robotic exercise device has been developed (MIT-Manus) and shown in a research setting to be capable of providing therapy to the arm of patients with weakness due to stroke, which simulates a conventional therapy known as skateboard therapy. We evaluated whether this robot-assisted task-specific therapy could also improve motor function in stroke patients with continuous, stable arm impairment. Subjects and Procedures. In a pilot intervention study conducted at the Baltimore Department of Veterans Affairs (VA) Medical Center without controls, we enrolled subjects with stable shoulder and elbow deficits due to stroke, stroke onset at least 6 months before enrollment, and a Motor Power Assessment (MPA) grade of 3 or less. We delivered 18 sessions of robot-assisted taskoriented therapy over 3 weeks. Primary outcome measures included the MPA, the upper-limb Fugl-Meyer (FM) Assessment Test, the Wolf Motor Function Arm Test, and the Motor Status Score for shoulder and elbow. We used the Wilcoxon Signed Test for paired data to assess preand posttreatment outcomes. Results. Statistically significant improvements were observed for the severely impaired group when we compared baseline and postintervention outcomes measured with the FM Assessment Test (mean increase of 1.5, p = 0.003), and the MPA (mean increase of 4.3, p = 0.006). Trends of improvement were observed but not significant among subjects who were moderately impaired. Relevance to the Veteran Population. Upperlimb dysfunction due to weakness and spasticity is a leading cause of chronic disability in veterans with stroke. This is a particular problem in more severely disabled, nonambulatory patients where loss of fine motor coordination, proximal limb weakness, and spasticity can interfere with activities of daily living. These more disabled patients are overrepresented in the VA population and therefore are particularly important to the VA. Our findings indicate that improvement in upper-limb motor function is not only for those with moderate impairments, as previously held, but is also for chronic stroke patients with severe impairments as well. Upper-limb neurorehabilitation has the potential to improve the VA systems delivery of care in several ways. First, if evidence from a controlled trial indicates that the robot-assisted therapy is equal in efficacy to traditional exercises for stroke patients with chronic upper-limb impairment, then having one therapist oversee several robotic workstations at one time would be possible, thereby potentially lowering costs. In addition, as the per-unit cost of the robot decreases, placement of remotely monitored units in the field for home therapy may be possible. If the robot were significantly more effective in improving function than the current standard of care, then adoption of robotic therapy would improve care. Finally, robotic therapy may reduce long-term costs of care, such as nursing home and hospital readmissions. These opportunities for improved VA services exist, but underlying evidence of efficacy begins with this study. Leah R. MacClellan, MSPH