Reverse shoulder arthroplasty (RSA) is a procedure to reduce pain and restore function in patients with rotator cuff arthropathy1. In vitro studies suggest that modifications in prosthesis design and surgical technique can improve functional outcomes after RSA2, but little work has been done to quantify in vivo kinematics following RSA. The aim of this ongoing study is to determine the effects of surgical technique and prosthesis geometry on in vivo functional outcomes after RSA. We hypothesized that greater humeral retroversion would be related to a more posterior contact path and that greater lateralization would be related to greater strength and better patient-reported outcomes (PROs).
17 patients received RSA (10M, 7F, age 69.5±7.4 years) using a standard 145-degree humeral implant (Wright Med/Tornier) or 135-degree humeral implant (Arthrex) within 2.5±1.2 years of participating in this IRB-approved study. Surgical parameters of glenoid lateralization and humeral retroversion were recorded from surgical notes. ASES, DASH, and Constant-Murley surveys (CMS) were completed at testing. Participants performed 3 trials of scapular plane abduction while synchronized biplane radiographs of the shoulder were collected at 50 images/s for 2 seconds. Subject-specific models of the humerus and scapula with respective implants were created from CT scans and matched to the biplane radiographs to measure scapular and humerus motion with sub-millimeter accuracy3. The center of the contact region of the polyethylene and glenosphere was determined at 5-degree increments of glenohumeral (GH) abduction and averaged across trials. Isokinetic torque was recorded over the full ROM for flexion/extension, ab/adduction, and internal/external rotation at 30°/second using a Biodex. Both the peak torque and the total work done were used to quantify strength. Pearson’s correlations were used to identify associations between the anterior-posterior and superior-inferior location of the center of contact at every 5° of GH abduction and surgical parameters. Spearman’s correlations were used to identify associations between strength, surgical factors, and patient-reported outcomes (PROs) with significance set at p<0.05.
Ten patients had 20° of humeral retroversion, four had 30°, and two had 40°. Ten patients had 0-2.0 mm lateralization, three had 2.1-4.0 mm, and four had 4.0-7.0 mm. During abduction, the center of contact path was posterior and inferior to the center of the glenosphere in all subjects. Increased retroversion was positively correlated with a more superior location of the center of contact between 35° and 60° of GH abduction (all p<0.01; R>0.85). No other correlations were found for center of contact (all p>0.31). Total adduction work was correlated with lateralization (?=0.58, p=0.015), and peak torque in external rotation was correlated with lateralization (?=0.57, p=0.017). CMS was correlated with total work in abduction, external and internal rotation (all ?=0.48, all p<0.047).
Retroversion is associated with in vivo contact kinematics, and increased glenoid lateralization is associated with increased strength and better PROs after RSA, supporting our hypothesis. The results indicate surgical technique is associated with strength and PROs. Our results provide in vivo evidence confirming previous computational modeling and cadaver-based studies that demonstrated increased strength following RSA with lateralized designs.