Patient and surgical risk factors for dislocation after reverse shoulder arthroplasty: a study by the American Shoulder and Elbow Surgeons multicenter complications of reverse shoulder arthroplasty research group

Background: The purpose of this study was to identify patient and surgical factors associated with dislocation after reverse shoulder arthroplasty (RSA) in a large multicenter cohort. Methods: A retrospective study using data from 15 institutions and 24 American Shoulder and Elbow Surgeons members across the United States was performed. Patients who underwent RSA between January 2013 and June 2019 with 3-month minimum follow-up were included. All components of methodology were determined by the Delphi method, an iterative survey process, requiring 75% consensus among all surgeons. Dislocations were defined as complete loss of articulation between the humeral component and the glenosphere with associated radiographic confirmation. Binary logistic regression was performed to determine predictors of dislocation following RSA. Results: Overall, 6621 patients with a mean follow-up of 19.2 ± 15.6 months were included. The incidence of dislocation was 2.1% (n = 138), 1.6% (n = 99) in primary RSA, and 6.5% (n = 39) among revision RSAs (P < .001). Dislocations occurred at a median of 7.0 weeks (interquartile range 3.0-36.0) after surgery with 22.5% (n = 31) following a trauma. Factors independently predictive of dislocation, in order of decreasing effect, were a preoperative diagnosis of nonunion fracture sequelae (odds ratio [OR] 8.31; P < .001), revision arthroplasty (OR 4.82; P < .001), the presence of a humeral spacer (OR 3.24; P < .001), a preoperative diagnosis of rotator cuff arthropathy or massive rotator cuff tear (OR 2.91; P < .001), the presence of a constrained polyethylene liner (OR 2.18; P = .001), male sex (OR 1.95; P = .001), and the lack of subscapularis repair (OR 1.61; P = .032) indicating a modest improvement in model fit compared to the null model. Conclusion: Patient factors such as being male, having large rotator cuff tears, undergoing revision RSA, or undergoing RSA for the sequelae of fracture nonunion appear to significantly increase the risk of postoperative dislocation after RSA. Surgical factors predictive of dislocation included the presence of a humeral spacer, constrained polyethylene liner, and lack of subscapularis repair, indicating a surgeon intraoperative awareness of potential instability.