Patient characteristics are more important than preoperative radiographic factors in predicting the risk of acromial stress fracture following reverse total shoulder arthroplasty.

Abstract

Background: Acromial stress fractures (ASFs) after reverse total shoulder arthroplasty (rTSA) can have a devastating effect on a patient's overall function, causing decreased range of motion and patient satisfaction. The purpose of this study was to evaluate preoperative patient demographics and radiographic findings associated with risk for ASF after rTSA.

Methods: A retrospective review was performed on patients who underwent primary rTSA by a single surgeon between December 2011 and December 2022. Inclusion criteria were primary rTSA with minimum 6-month postoperative radiographs. Preoperative demographics, comorbidities, and radiographic factors (acromial thickness, critical shoulder angle, humeral cortical thickness, and acromiohumeral distance) were reviewed. Postoperative radiographs were all evaluated for the presence of ASF and classified according to the Levy classification. Univariate analyses followed by logistic regression analyses were performed to evaluate for significant risk factors. A Bonferroni correction was then performed.

Results: A total of 757 rTSAs were included and 58 (7.7%) ASFs identified. ASFs were identified at a median of 14 weeks postoperatively. Across the entire cohort, including both male and female patients, multiple factors including lower body mass index, inflammatory arthritis, and decreased acromial thickness were identified as initially being correlated with ASF. Among females, the presence of inflammatory arthritis, the use of preoperative osteoporosis medications, humeral cortical thickness, and acromial thickness were initially identified as increasing the risk of ASF. For males, surgery on the dominant arm and lower body mass index were initially identified as increasing the risk of ASF. After a Bonferroni correction was completed and the P value representing significance was set at P < .0025, only the presence of inflammatory arthritis in females was found to be statistically significant (odds ratio: 4.87, 95% confidence interval: 2.16-10.96, P < .001).

Conclusion: This study points to multiple preoperative patient characteristics that showed initial correlation with an increased risk of ASF and demonstrates the importance of future study. After the statistical significance was revised and set at P < .0025, females with inflammatory arthritis are associated with increased risk of ASF after rTSA. These findings suggest that overall patient health and fragility may be more predictive of ASF rather than preoperative radiographic characteristics that were studied. We propose that further study of other factors, such as implant position, patient nutrition, and postoperative rehabilitation parameters, be performed to determine their impact on ASF risk.