Local Antibiotics and the Risk of Antimicrobial Resistance in Extremity Fractures Complicated by Fracture-Related Infection

Background: We evaluated antimicrobial resistance (AMR) patterns following local antibiotic use in a large cohort of patients with fractures from the PREP-IT (A Program of Randomized Trials to Evaluate Preoperative Antiseptic Skin Solutions in Orthopaedic Trauma) study. We hypothesized that, among patients with extremity fractures who developed fracture-related infection (FRI), there would be no difference in AMR rates between those who had or had not received local antibiotic therapy with surgical fixation. Methods: This was a secondary analysis of all patients in the PREP-IT trial who developed FRI. Patient demographics, injury and fracture characteristics, and the primary outcome of the presence of an antimicrobial-resistant FRI were evaluated on the basis of whether the patient had or had not received local antibiotics in the operating room prior to, or at, definitive fixation. Results: A total of 555 FRIs in 546 patients (mean age, 50 years; 39% female; and 82% White) were included. A total of 268 fractures (264 patients) received local antibiotics. The Injury Severity Score and the proportion of open fractures were higher among patients and fractures that received local antibiotics, respectively. There were more Gustilo-Anderson type- IIIB or IIIC fractures in the local antibiotic group, but the rate did not differ significantly from that in the group with no local antibiotics (20% versus 14%; p = 0.14). Other baseline and fracture characteristics were similar between the groups, with the exception of age (lower in the group with local antibiotics). When examining FRIs with gram-positive organisms, we found that 3 (1.7%) of the FRIs in fractures that had been treated with local vancomycin had organisms resistant to vancomycin compared with 2 (0.9%) of the FRIs in fractures for which local vancomycin had not been used (p = 0.67). When examining FRIs with gram-negative organisms, the number of FRIs with aminoglycoside-resistant organisms was 8 (11.6%) among fractures that received local aminoglycosides and 10 (6.2%) among fractures that did not receive local aminoglycosides (p = 0.26). Conclusions: Among extremity fractures that developed FRI, we were unable to detect differences in the rates of AMR between fractures treated with or without local antibiotic prophylactic strategies in our analysis of a randomized trial of various skin preparation solutions for extremity trauma surgery. These findings provide cautious reassurance regarding the safety of local antibiotics but underscore the need for further prospective analysis.