Infection After Open Long Bone Fractures: Can We Improve on Prophylaxis?

Abstract

Current standards recommend antibiotic prophylaxis administered after open fracture injury. The purpose of this study was to assess culture results in patients with open fracture-associated infections, hypothesizing that cultures obtained do not vary by Gustilo-Anderson (GA) classification. We examined cultured bacterial species from patients with open long bone fractures that underwent irrigation and debridement at a Level 1 trauma center (2008-2016), evaluating our current and two hypothetical antibiotic protocols to assess whether they provided appropriate coverage. The antibiotic protocols included protocols 1 (cefazolin, with gentamicin added for type III fractures), 2 (vancomycin and cefepime) and 3 (ceftriaxone). GA classification was not associated with bacterial gram stain (P=0.161), nor was it predictive of mono- versus polymicrobial infection (P=0.094). Of 42 culture-positive infections, 31 were type III and 11 were type I or II fractures. 27% of the infections for type I or II fractures were caused by organisms targeted by protocol 1 (OR 0.18, 95% CI 0.04-0.82; P=0.027). There was no difference in coverage by fracture type among protocol 2 (P=0.771) or protocol 3 (P=0.891). For type III fractures, protocol 2 provided 94% appropriate coverage compared to 68% and 61% coverage by protocols 1 and 3, respectively. For open fractures complicated by infection, isolated bacterial organisms do not correlate with GA open fracture classification, suggesting that hypothetical protocol 2 should be used for all fracture types. Protocol 2's broad coverage, across all GA fracture types, may prevent infection by organisms not covered by current antibiotic prophylaxis.