Infection following mini-open rotator cuff repair: a single surgeon experience

Background: Surgical technique has been shown to influence risk of surgical site infection following rotator cuff repair (RCR). Few studies have reported the rate of infection associated with mini-open RCR. The goal of this study was to report the postoperative infection rate and risk factors for infection among patients undergoing RCR performed by a single surgeon using a modified mini-open technique. Our hypothesis was that the rate of infection after mini-open RCR would be lower than previously reported for this surgical approach. Methods: We retrospectively reviewed an institutional shoulder surgery database to identify patients who underwent mini-open RCR performed by 1 surgeon at an academic tertiary care institution between 2003 and 2020. Patient records were reviewed to determine which individuals returned within 3 months postoperatively with a superficial or deep surgical site infection requiring operative management. Patient demographics, preoperative clinical characteristics, intraoperative variables, microbiological findings, infection management, and clinical course after infection were recorded. Backward elimination multivariate regression was used to assess for significant risk factors for infection. Results: Of the 925 patients identified, 823 (89%) had at least 3 months of follow-up and were included for further analysis. A majority of the patients undergoing RCR were men (57%). The mean age was 58.4 +/- 9.9 years, and the mean body mass index was 29.3 +/- 5.9 kg/m2. Fourteen cases (1.7%) of postoperative surgical site infection were identified in 13 patients. Ten infections (1.2%) were superficial and 4 (0.49%) were deep. The most commonly identified organisms were Staphylococcus aureus and Cutibacterium acnes. Male sex (odds ratio [OR] 4.3, 95% CI 1.2-15.3) and diabetes mellitus (OR 3.9, 95% CI 1.2-12.6) were found to be associated with greater odds of infection. The RCR construct was found to be intact in all 10 patients with superficial infections and 2 of the 4 patients with deep infections. All infections were successfully treated with 1 round of surgical debridement and wound irrigation, and with 6 or fewer weeks of intravenous antibiotic therapy. All patients with postoperative infections recovered with no sequelae at a median final follow-up of 63.5 months (range, 3-215 months). Conclusions: This single-surgeon series of a large patient cohort undergoing mini-open RCR over an 18-year period demonstrated a low overall infection rate of 1.7%. Only 4 infections were deep, which suggests that deep infection after mini-open RCR is uncommon and approximates infection rates seen with arthroscopic techniques.