Does isolated glenosphere lateralization affect outcomes in reverse shoulder arthroplasty?

Introduction: While lateralization of the glenohumeral center of rotation during reverse shoulder arthroplasty (RSA) has benefits of maintaining tension on the remaining rotator cuff and decreasing implant impingement on the glenoid, few clinical studies have evaluated the isolated effect of glenoid lateralization in RSA. The purpose of this study was to evaluate if clinical outcomes are affected by isolated glenosphere lateralization using a single implant design.

Methods: A retrospective review from a multicenter shoulder arthroplasty research database was performed between 2011 and 2018 using a single implant system to perform this case-controlled study. Inclusion criteria included primary RSAs with adequate preoperative and postoperative active and passive range of motion (ROM) measurements, outcome scores, and a minimum two-year follow-up. Revision shoulder arthroplasties and RSA for fractures were excluded from analysis. 102 RSAs (61 females, 41 males) using a +4mm lateralized glenosphere were compared to 102 sex, age, and glenosphere diameter matched control shoulders with standard glenospheres (whose center of rotation (CoR) is 2mm lateral to the glenoid fossa). The mean age at surgery was 70.4 years. Mean follow up was 43.6+18.9 months. All RSAs were performed with the same implant system (Equinoxe, Exactech, Gainesville, FL). Clinical outcome measures included ROM, ASES, Constant, UCLA, SST, SPADI scores, and VAS pain scores. We used the chi-squared test and Fisher exact test for bivariate analysis and the student t-test for continuous variables.

Results: Both groups were of similar average age and follow-up. They also had comparable rates of prior surgery and comorbidities. The lateralized glenosphere group had a slightly higher BMI (31.2 vs. 29.2, p=0.04). Both groups demonstrated significant improvements in all outcome scores that exceeded the MCID and the SCB. The groups demonstrated similar preoperative, postoperative and improvements in ROM as well as outcome scores. The overall complication rate was similar between groups (4% in lateralized and 5% in controls, p=0.73). Scapular or acromial fractures differences were not statistically significant between groups (1% in lateralized group vs. 3% in standard group, p=0.31). Scapular notching was more frequent in the standard group compared to the lateralized group (9% vs. 2%, p=0.03).

Conclusion: In a medialized glenosphere/lateralized humerus design, a +4mm lateralized glenosphere provided no significant advantage in postoperative pain, ROM, or outcome scores. However, lateralized glenospheres did demonstrate significantly lower scapular notching rates.

Level of evidence: III; retrospective cohort comparison; treatment study.