Comparison of anterior shoulder pain and internal rotation dysfunction after anatomic and reverse shoulder arthroplasty for osteoarthritis.

Abstract

There is controversy regarding whether reverse shoulder arthroplasty (rTSA) and anatomic shoulder arthroplasty (aTSA). rTSA changes the shoulder's anatomy and biomechanics, altering soft tissue tension and arm length. The effects of these changes to the anatomy are incompletely understood and may lead to anterior shoulder pain and dysfunction. In addition, decreased internal rotation is a known limitation of rTSA that is less frequently seen in aTSA. Differences in anterior shoulder pain and internal rotation dysfunction may not be apparent on standardized instruments of outcomes assessment. This study aims to develop a novel anterior shoulder pain and dysfunction score (ASPDS) to classify anterior shoulder dysfunction and use an existing questionnaire, the functional internal rotation (FIR) score, to quantify internal rotation (IR) deficiency following rTSA compared with aTSA when performed for osteoarthritis. A retrospective review of a single-institution database was conducted of patients 2 years postoperative from aTSA or rTSA for glenohumeral osteoarthritis. Exclusion criteria included shoulder arthroplasty for indication other than osteoarthritis, subscapularis not repaired, or signs of rotator cuff tearing preoperatively. Walch glenoid morphology was classified based on computed tomography scan or radiograph. A novel 7-question ASPDS questionnaire was developed to target actions that require anterior deltoid function. The previously published 10-question FIR score was used to assess actions that necessitate IR function. Outcomes included scores for the ASPDS, FIR score, American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form score (ASES), visual analog scale (VAS) for pain, and Single Assessment Numeric Evaluation (SANE). Internal reliability of the questionnaires was assessed using Cronbach alpha. Twenty-six patients were included in each aTSA and rTSA group per the a priori power analysis. Demographic factors were not significantly different between groups (P > .05). Mean ASPDS scores were lower in the rTSA group (29.1 ± 5.1) compared with the aTSA group (32.7 ± 4.5) (P = .001). Mean FIR score was also worse in the rTSA group (43.1 ± 6.5) compared with the aTSA group (46.8 ± 7.3) (P = .004). No difference was observed between the rTSA and aTSA groups for ASES, SANE, and VAS scores (P > .05). Cronbach alpha was 0.889 for the ASPDS questionnaire and 0.935 for the FIR score. The ASPDS is a reliable questionnaire for characterizing anterior shoulder pain and dysfunction following shoulder arthroplasty. Initial findings suggest that anterior shoulder pain and dysfunction and decreased internal rotation are more common in rTSA compared with aTSA when performed for osteoarthritis, with differences observed at 2 years postoperatively. These differences in outcomes were not captured by traditional shoulder outcomes questionnaires.