A kinematic and electromyographic comparison of a Grammont-style reverse arthroplasty combined with a l'Episcopo transfer compared to a lateralized humeral component reverse for restoration of active external rotation.

Abstract

To assess kinematic and electromyographic findings of two designs of reverse total shoulder arthroplasty (RTSA). We tested two hypotheses: (H1) Grammont-style (RTSA) with l'Episcopo tendon transfer gains similar postoperative kinematic findings of a lateralized humeral component RTSA without a tendon transfer; (H2) RTSA with lateralized humeral component induces earlier shoulder muscle activation during shoulder motion compared with standard Grammont prosthesis with l'Episcopo transfer. Twenty-five patients with a cuff tear arthropathy, pseudoparalysis, and a positive dropping sign were sequentially included. A Grammont-style RTSA with a l'Episcopo tendon transfer was implanted in 13 patients (medialized humerus and transfer group) and an on-lay curved-stem RTSA in 12 (lateralized humerus group). Constant score; the disabilities of the arm, shoulder, and hand (DASH) score; and active shoulder range of motion (flexion, abduction, external rotation, and internal rotation) were measured pre- and postoperatively. Upper limb kinematic and surface electromyography (EMG) (anterior and posterior deltoid; upper, middle, and lower trapezius; and the upper and lower latissimus dorsi muscles) were recorded during active range of motion. At > 24months of follow-up, the Constant and DASH scores and active shoulder range of motion improved in both groups (p < 0.0001). Internal rotation was significantly higher in the lateralized humerus group than in the medialized humerus and transfer group (p = 0.004). The dropping sign was found in 0/12 patients in the lateralized humerus group and in 5/13 patients in the medialized humerus and transfer group (p < 0.001). Kinematic findings were similar between groups (p = 0.286). A pattern of earlier muscle activation in the lateralized humerus group was found as compared to the transfer group as follows: (i) posterior deltoid during external rotation (p = 0.004); (ii) upper latissimus dorsi and middle and lower trapezius during flexion (p = 0.004, p = 0.005, and p = 0.042, respectively); (iii) lower latissimus dorsi during abduction (p = 0.016). RTSA with a lateralized humeral component provided similar active external rotation restoration to a Grammont-style RTSA with a l'Episcopo procedure in patients with a dropping sign. The posterior deltoid muscle underwent earlier activation during active external rotation in the lateralized humerus RTSA group as compared to the tendon transfer group. Our findings demonstrated the H1 and only partially demonstrated the H2 of the study.