Abstract
BACKGROUND: Total shoulder arthroplasty (TSA) is the preferred treatment for glenohumeral arthritis refractory to nonoperative measures. However, some surgeons have argued for a role for hemiarthroplasty (HA) in the setting of a smooth glenoid that articulates appropriately with the humeral head. The purpose of this study is to evaluate long-term revision rates and short-term postoperative complications in patients undergoing either HA or TSA for glenohumeral arthritis. METHODS: A retrospective review of patients who underwent HA and TSA was conducted using a commercially available national database. Demographics, postoperative complications, risk factors, revision rates, and costs were analyzed using 2 sample t-tests, chi-squared tests, and multivariate logistic regressions. RESULTS: Patients were stratified by operation: (1) HA (n = 1615) or 2) TSA (n = 7845). Patients undergoing primary TSA had higher rates of prior ipsilateral rotator cuff repair and corticosteroid injections. At 2 years, patients who underwent HA, 3.0% of patients had revision surgery, compared to 1.6% of patients who underwent TSA (P = .002); at 5 years, 3.7% of the HA cohort (P < .0001) had revision surgery, compared to 1.9% of patients who underwent TSA. CONCLUSIONS: Patients undergoing TSA or RTSA for glenohumeral arthritis had higher preoperative co-morbidities but had no difference in short-term complication rates with a lower risk of revision surgery at both 2-year and 5-year follow-up when compared to HA. Increasing age, female sex, hyperlipidemia, postoperative infection, shoulder instability, and thromboembolism all independently increased odds for revision shoulder arthroplasty for glenohumeral arthritis. LEVEL OF EVIDENCE LEVEL: III.