What Are Risk Factors for Intraoperative Humerus Fractures During Revision Reverse Shoulder Arthroplasty and Do They Influence Outcomes?

Abstract

Background: With the increase in shoulder arthroplasty rates, the number of perioperative complications, such as periprosthetic fractures, continues to be a rise; however, the risk factors and incidence of intraoperative complications, such as fractures, during revision reverse shoulder arthroplasty are not well established. Questions/purposes: We evaluated patients receiving a reverse shoulder arthroplasty to determine (1) the frequency and characteristics of intraoperative humerus fractures, (2) the risk factors for fracture, (3) the complications associated with treatment of fractures, and (4) clinical and functional outcomes after treatment. Methods: Using one institution’s total joint registry, we performed a retrospective analysis of medical records of 224 patients (230 shoulders) who underwent revision surgery to reverse shoulder arthroplasty, from 2005 to 2012, for failed total shoulder arthroplasty. Reverse shoulder arthroplasty was used when there was a deficient rotator cuff, prior instability, or deficient glenoid bone stock. Intraoperative periprosthetic humerus fractures occurred in 36 shoulders (16%) (36 patients). The clinical outcome analysis included 29 patients with a minimum 2-year clinical followup (mean followup, 3.1 years; range, 2.0−6.3 years). The control group consisted of 188 patients (194 shoulders), and it was used for risk-factor calculation; whereas only 150 patients (154 shoulders) in the control group had a minimum 2-year followup, and thus only 150 patients (154 shoulders) made up the comparators (controls) for outcome-analysis comparisons. Risk factors were assessed using univariate analysis with odds ratios (OR), whereas implant survival and complications were assessed using the Kaplan-Meier method. Results: Three displaced and 33 nondisplaced fractures occurred during revision reverse total shoulder arthroplasties. Most of the fractures (81%) occurred during component removal of cemented (n = 11) and cementless (n = 25) components. Intraoperative fractures only were treated with stabilization of the prosthetic stem in 28 patients, while adjunctive internal fixation was used in eight patients. Risk of intraoperative periprosthetic fractures was increased by factors including female sex (n =18 women; OR, 2.41; range, 1.11−5.68; p = 0.03); history of instability (n = 27; OR, 2.65; range, 1.18−5.93; p = 0.02); and prior hemiarthroplasty (n = 22; OR, 2.34; range, 1.13−4.84; p = 0.03). There were two postoperative fractures in patients who had an intraoperative fracture and both were treated nonoperatively. Overall, three (8%) revision procedures were performed in patients with intraoperative fractures, with 2- and 5-year survivorship estimates of 94% and 85%, respectively, compared with 89% and 84%, respectively for patients without an intraoperative fracture (p = 0.45). At latest followup, patients experienced good postoperative pain relief, improved shoulder abduction, and good American Shoulder and Elbow Surgeon and Simple Shoulder Test scores. Conclusions: Intraoperative humeral fractures occur in approximately 16% of shoulders undergoing revision surgery. Fractures during revision reverse TSA are not uncommon secondary to the risks of component removal in revision surgery and poor remaining bone stock. The risk seems to be greatest for female patients, patients with instability, and patients who have undergone previous hemiarthroplasties. Intraoperative humeral fractures should be approached in a systematic way to achieve anatomic reduction and stable fixation. When properly stabilized, these fractures appear not to substantially influence overall final outcome. This study provides a foundation for future investigation of methods to reduce the risk for intraoperative humeral fractures attributable to reverse revision TSA. Level of Evidence: Level III, therapeutic study.