Abstract
BACKGROUND: Acromion stress fractures (ASF) or scapular spine fractures (SSF) following reverse total shoulder arthroplasty (RTSA) are common complications with impaired clinical outcome. The underlying biomechanical factors remain unclear. The aim of this study was to evaluate basic demographic and radiographic parameters predicting occurrence of different types of ASF/SSF in a large single-center study cohort. METHODS: A total of 860 RTSA (805 patients) with available minimum follow-up of 2 years were implanted between 2005 and 2018 at a tertiary academic center. All RTSA with subsequent ASF/SSF (n = 45 in 43 shoulders [42 patients, 5%]) were identified and classified as Levy I to III. Predictive demographic, surgical, and radiographic factors were evaluated for each subtype and compared to the control group (817 RTSA, 763 patients). The radiographic analysis included critical shoulder angle, lateralization shoulder angle (LSA), distalization shoulder angle (DSA), acromio-humeral distance (ACHD), acromial thickness, deltoid tuberosity index, deltoid length, and center of rotation. RESULTS: Of the 45 ASF/SSF in 42 patients, 8 were classified as Levy I, 21 as Levy II, and 16 as Levy III. Demographic analysis revealed indication as risk factor for Levy I fractures, higher American Society of Anesthesiologists score as risk for Levy type II fractures and higher age as risk factor for Levy type III fractures. None of the measured radiographic parameters were predictive for occurrence of Levy type I and Levy type II ASF. However, analysis of Levy III SSF revealed a higher postoperative LSA (89° ± 10° vs. 83° ± 9°, P = .015), a lower postoperative DSA (45° ± 8° vs. 53° ± 12°, P = .002), less distalization (ACHD of 33 ± 8 mm vs. 38 ± 10 mm, P = .049), and a more medial center of rotation preoperatively (COR-LA 16 ± 8 mm vs. 12 ± 7 mm, P = .048) as predictive radiographic factors. CONCLUSION: The present analysis showed a significant association of higher postoperative LSA, lower DSA, a lower ACHD, and higher age as predictive factor only for Levy type III fractures. Some of these factors can be surgically influenced and this knowledge can be of value for preoperative planning and surgical execution to avoid these complications.