Shoulder Musculoskeletal Disorder Rehabilitation Using a Robotic Device Based on Electromyography (EMG) Biofeedback: A Retrospective Cohort Study.

Background and Objectives: While shoulder injuries represent the musculoskeletal disorders (MSDs) most encountered in physical therapy, there is no consensus on their management. In attempts to provide standardized and personalized treatment, a robotic-assisted device combined with EMG biofeedback specifically dedicated to shoulder MSDs was developed. This study aimed to determine the efficacy of an 8-week rehabilitation program (3 sessions a week) using a robotic-assisted device combined with EMG biofeedback (RA-EMG group) in comparison with a conventional program (CONV group) in patients presenting with shoulder MSDs. Materials and Methods: This study is a retrospective cohort study including data from 2010 to 2013 on patients initially involved in a physical rehabilitation program in a private clinic in Chicoutimi (Canada) for shoulder MSDs. Shoulder flexion strength and range of motion were collected before and after the rehabilitation program. Forty-four patients participated in a conventional program using dumbbells (CONV group), while 73 completed a program on a robot-assisted device with EMG and visual biofeedback (RA-EMG group); both programs consisted of two sets of 20 repetitions at 60% of maximal capacity. Results: We showed that the RA-EMG had significantly greater benefits than the CONV group for shoulder flexion strength (4.45 [2.6;6.15] kg vs. 2.3 [0.90;4.775] kg, U = 761, p = 0.013) and for normalized strength (77.5 [51.3;119.1] % vs. 39.1 [16.6;89.2] %, U = 755, p = 0.016). In addition, the RA-EMG group showed a trend to greater absolute gain of ROM than the CONV group (10.0 [0;24.3] degrees vs. 5.5 [0;12.0] degrees, U = 1931, p = 0.067), and a greater benefit in normalized ROM was observed for the RA-EMG (7.4. [0;17.7] %) than the CONV group (4.6 [0;10.8], U = 1907, p = 0.046). Conclusions: The current retrospective cohort study showed that a specific and tailored 8-week rehabilitation program with constant effort by automatic adjustment of the level of resistance by EMG feedback induced greater benefits for shoulder flexion strength and a trend to improve range of motion compared to conventional rehabilitation in patients with shoulder MSDs. Future research should be pursued to determine the added potential of this approach for abduction and external rotation with a randomized controlled design.