Assessment of a Modified Antibiotic Prophylaxis Open Fracture Protocol

BACKGROUND: National guidelines support antibiotic prophylaxis for open fracture with cefazolin +/− aminoglycoside based on fracture grade and contamination. The purpose of this study was to assess a modified adult antibiotic prophylaxis open fracture protocol (AOFP) which recommended weight-based cefazolin for low-grade fractures or ciprofloxacin plus vancomycin for high-grade fractures. METHODS: Adult patients with open fractures admitted to Palmetto Health Richland between January 2012 and December 2016 were screened for study inclusion. Exclusion criteria were receipt of antibiotics for reasons other than open fracture, death prior to wound closure, and local admission time >48 hours after time of injury. Compliance to all elements of AOFP was assessed. Clinical endpoints including open fracture infection rates, epidemiology, and drug-related adverse events were compared between pre-implementation (January 2012 – December 2012) and post-implementation period (November 2015 – December 2016). χ (2) and t-tests as appropriate were used to compare outcomes between groups. RESULTS: Following exclusions 189 patients were included in the analysis (90 pre- vs. 99 post-AOFP, respectively). Post-AOFP, a 17% (16/93) adherence rate to all AOFP elements was found. Appropriate agents were selected in 82.8% (77/93). The most common reasons for non-adherence were incorrect dosing and prolonged antibiotic duration. Fracture site infection rates were 23.3% (21/90) and 7.1% (7/99) in pre- and post-AOFP groups, respectively (P = 0.001). Infections primary caused by Gram-negative pathogens in pre-AOFP and Gram-negative organisms comprised 62 and 40% of open fracture site infections in pre- and post-AOFP groups, respectively. Incidence of acute kidney injury, Clostridium difficile-associated diarrhea, and other antibiotic-associated AEs were rare and comparable between groups. Change in median days to infection (55.6 days vs. 56.55 days, P = 0.71) and median duration of antibiotics in hours (48.0 vs. 54.7, P = 0.59) was not significantly different post implementation. CONCLUSION: Local adherence to all elements of the modified AOFP was low, yet the appropriate agent(s) was used in majority of cases. The modified AOFP was associated with a numerical decrease in infection rates post-open fracture and comparable AEs. DISCLOSURES: P. B. Bookstaver, Rock Pointe: Content Developer, Consulting fee