Infections following arthroscopic rotator cuff repair: incidence, risk factors, and prophylaxis

The primary aim of the present study was to determine the incidence of infections following arthroscopic rotator cuff repair in a single department over a 10-year period. Secondary goals included the evaluation of potential risk factors for infections and to investigate the efficacy of perioperative antibiotic prophylaxis in the reduction of infectious complications. A retrospective evaluation of 3294 all-arthroscopic rotator cuff repairs performed between 2004 and 2014 at a single institution was conducted to determine the rate of infection in all-arthroscopic rotator cuff repairs. Detailed data including demographics, co-morbidities, initial surgical procedure, time from index surgery to infection, clinical presentation, isolated pathogens, and subsequent treatment were recorded of patients with infections. Univariate logistic regression was performed, and a multivariate model was developed to identify variables that were predictive of infections following arthroscopic rotator cuff repair. The rate of infection was 8.5/1000 in whom rotator cuff repairs were performed (95 % CI 0.58-1.23 %) during the study period. The most commonly identified pathogen was Staphylococcus epidermidis (n = 11, 39.3 %), followed by Propionibacterium acnes (n = 8, 28.6 %) and Staphylococcus aureus (n = 2, 7.1 %). Patients presented at our institution with signs of infection an average of 28.9 +/- 14.7 days after the index surgery. The leading symptom that was present in all patients was diffuse or localized shoulder pain, followed by local signs of infection in 19 (67.9 %), secretion in 14 (50 %), and fever in 9 (32.1 %) patients, respectively. Univariate and multivariate analyses identified the male gender, age over 60, and duration of surgery over 90 min as predictive factors for infection. Administration of perioperative antibiotic prophylaxis was an independent mitigating factor for postoperative infection and reduced the infection rate from 1.54 % (95 % CI 0.98-2.30 %) to 0.28 % (95 % CI 0.10-0.67 %). The overall incidence of infection was 8.5/1000 arthroscopic rotator cuff repairs over a 10-year period. Gender, age, and length of surgery were identified as predictive factors for infection. The administration of perioperative antibiotic prophylaxis significantly reduced the overall risk of infection and is therefore recommended in all-arthroscopic rotator cuff repairs. However, current prophylactic treatment strategies were not equally efficient for all types of pathogens. Retrospective comparative study, Level III.