Evaluation of Piperacillin-Tazobactam for Antibiotic Prophylaxis in Traumatic Grade III Open Fractures.

Abstract

Background: Broad-spectrum antibiotic agents are sometimes utilized for prophylaxis of Gustilo grade III open fractures. However, this practice is not recommended by current guidelines, and it is unknown how patient outcomes are impacted. This study aimed to determine if prophylaxis with piperacillin-tazobactam (PT) results in different rates of infection versus guideline-concordant therapy (GCT). Patients and Methods: This was a single-center, retrospective cohort study of adult trauma patients with Gustilo grade III open long bone fractures admitted between January 2008 and August 2018. The primary outcome of infection (superficial or deep) at six weeks and secondary outcomes of delayed union, nonunion, Clostridioides difficile, and development of resistant organisms were abstracted from medical records. Guideline-concordant therapy was defined as a first-generation cephalosporin with or without an aminoglycoside. Univariable and multivariable analyses controlling for injury severity score (ISS) were performed. Results: One hundred twenty patients were included; 97 (81%) received PT, 23 (19%) received GCT. Common injury mechanisms were motor vehicle/motorcycle accident (57%) and falls (17%), and a majority involved a lower extremity (65%). Baseline characteristics were similar except higher median ISS in PT (14; interquartile range [IQR], 9-22) versus GCT (9; IQR, 9-14). Guideline-concordant therapy was given for a median of four (range, 2-8) days and PT for six (range, 3-11) days (p\u2009=\u20090.078). On univariable analysis, PT patients had more infections at six weeks (23.7% vs. 4.3%; p\u2009=\u20090.042), but multivariable analysis demonstrated no difference (odds ratio [OR], 5.81; 95% confidence interval [CI], 0.73-46.25; p\u2009=\u20090.096). Patients receiving prophylaxis with PT had a longer median length of stay at 16 days (range, 10-22) versus nine days (range, 4-16). No statistically significant differences in delayed union, non-union, Clostridioides difficile, or development of resistant organisms were observed. Conclusions: Broad-spectrum antibiotic prophylaxis with PT did not improve infection rates compared to GCT, suggesting it may not be warranted.