Rate of and Risk Factors for Reoperations After Open Reduction and Internal Fixation of Midshaft Clavicle Fractures

Background: Reoperation rates following open reduction and internal fixation (ORIF) of midshaft clavicle fractures have been described, but reported rates of nonunion, malunion, infection, and implant removal have varied. We sought to establish baseline rates of, and risk factors for, reoperations following clavicle ORIF in a large population cohort. Methods: Administrative databases were used to identify patients sixteen to sixty years of age who had undergone an ORIF of a closed, midshaft clavicle fracture from April 2002 to April 2010. The primary outcome was a reoperation within two years (isolated implant removal, irrigation and debridement [deep infection], pseudarthrosis reconstruction [nonunion], or clavicle osteotomy [malunion]). The secondary outcome was rare perioperative complications, including pneumothorax, subclavian vasculature injury, and brachial plexus injury. A multivariable logistic regression analysis was performed to determine the influence of patient and provider factors on these outcomes. Results: We identified 1350 patients who underwent midshaft clavicle ORIF (median age, thirty-two years [interquartile range, twenty-one to forty-four years]; 81.3% male). One in four patients (24.6%) underwent at least one clavicle reoperation. The most common procedure was isolated implant removal (18.8%), and females were at highest risk (odds ratio [OR], 1.7; p = 0.002). The median time to implant removal was twelve months. A reoperation secondary to nonunion, deep infection, and malunion occurred in 2.6%, 2.6%, and 1.1% of the patients after a median of six, five, and fourteen months, respectively. Risk factors for clavicle nonunion included female sex (OR, 2.2; p = 0.04) and a high comorbidity score (OR, 2.8; p = 0.009). For surgeons, fewer years in practice was associated with a small risk of the patient developing an infection (OR, 1.1; p < 0.001). Sixteen pneumothoraces (1.2%) were identified; however, brachial plexus and subclavian vessel injuries were each found in five or fewer patients. Conclusions: Following clavicle ORIF, one in four patients underwent a reoperation. The most common procedure was implant removal, and although the rates of reoperations secondary to nonunion, malunion, and infection were low they were higher than previously reported. Pneumothoraces and neurovascular injuries were infrequent and should continue to be considered rare complications of clavicle ORIF. Level of Evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.