Effect of combined version on impingement-free rotational range-of-motion in three different implant types in reverse...

The influence of combined humeral and glenoid component version on impingement-free range-of-motion (ROM) has not previously been evaluated in reverse total shoulder arthroplasty (rTSA). The purpose of this study was to determine the effects of combinations of humeral and glenoid component version on ROM in 3 different rTSA designs. We hypothesized that increased combined retroversion would decrease ROM.

Nine patients with posterior glenoid bone loss indicated for rTSA underwent preoperative computed tomography of the shoulder. Virtual templating was performed using 3 different rTSA designs: medial glenoid-medial humerus, medial glenoid-lateral humerus (MG-LH), and lateral glenoid-medial humerus (LG-MH). Each implant was placed in different combinations of glenoid (10°, 0°, -10°, -20°) and humeral component version (0°, -10°, -20°, -30°, -40°). Virtual simulations of glenohumeral ROM were performed at 0°, 30°, and 90° abduction, with impingement-free internal rotation (IR), external rotation (ER), and combined IR and ER measured. A scapular contact mesh was created to determine impingement locations. A linear mixed effects model was used to assess the association between implant type and ROM. Chi-square tests were used to evaluate differences in impingement locations between implant types.

Total (IR and ER) ROM decreased with increasing combined component retroversion. ER ROM increased with more combined retroversion version until 30°, while the contrary was found for IR ROM. In the linear mixed effects model, implant type, glenoid version, and humeral version were significantly associated with total ROM. Compared to the LG-MH implant, the MG-LH implant was associated with increased ROM, and the medial glenoid-medial humerus implant was associated with decreased ROM. Changes in glenoid version were significantly associated with IR ROM but not ER ROM, while changes in humeral version significantly impacted both IR and ER ROM. The most common locations of bony impingement were the posterior scapular neck and anterior/posterior glenoid at 0° abduction; the anterior/posterior glenoid at 30° abduction; and the superior glenoid, coracoid, and acromion at 90° abduction. There were significant differences in impingement locations between implant types at all abduction positions (P < .001).

This study demonstrates that total impingement-free ROM in rTSA decreases with combined component version greater than 30° of retroversion. The study also demonstrates increased total ROM with implant designs that create a more lateralized center of rotation (MG-LH, LG-MH). Impingement location changes with implant design, with less scapular neck impingement found with the LG-MH design. The study is limited by not being able to account for scapulothoracic motion or soft tissue considerations.