Intrawound Tobramycin Plus Vancomycin to Prevent Surgical Site Infection in Tibial Fractures: The TOBRA Randomized Clinical Trial

JAMA. 2026 Apr 15:e264023. doi: 10.1001/jama.2026.4023. Online ahead of print.

Abstract

Importance: Previous research has suggested that intrawound vancomycin powder reduces deep surgical site infections among patients with periarticular tibial fractures at high risk of infection. It is unknown whether the addition of tobramycin powder further decreases infection rates.

Objective: To compare whether the combination of tobramycin plus vancomycin vs vancomycin alone delivered as intrawound powder at the time of definitive fixation reduces deep surgical site infections.

Design, setting, and participants: Open-label, assessor-masked, randomized clinical trial conducted at 39 US trauma centers. Eligible patients were adults with an operatively treated periarticular tibial fracture (either tibial plateau or pilon) who met 1 of 3 criteria for elevated infection risk. Enrollment occurred between June 18, 2021, and December 12, 2024 (final follow-up, July 15, 2025).

Interventions: Intrawound tobramycin (1.2 g) plus vancomycin (1.0 g) powder vs intrawound vancomycin (1.0 g) powder delivered at the time of definitive fixation.

Main outcomes and measures: The primary outcome was a deep surgical site infection requiring surgical management within 182 days of definitive fracture fixation. Secondary outcomes included deep surgical site infections with pathogens that were gram-negative only, deep surgical site infections with at least 1 pathogen that was gram-positive, deep surgical site infections with polymicrobial cultures, deep surgical site infections with negative culture results, and cellulitis or skin infections treated only with antibiotics.

Results: Among the 1660 participants randomized, 1528 (mean age, 47.0 [SD, 14.3] years; 603 female [39.5%]; 925 male [60.5%]) were included in the primary analysis. Deep surgical site infections occurred in 51 of 753 participants (182-day probability, 7.4%) in the tobramycin plus vancomycin group and 47 of 775 participants (182-day probability, 6.6%) in the vancomycin alone group (hazard ratio, 1.11; 95% bayesian credible interval, 0.75-1.66; posterior probability of superiority, 29.7%). The threshold required for superiority was not reached for any secondary outcome.

Conclusions and relevance: Among patients with operatively treated periarticular tibial fractures at high risk of infection, adding intrawound tobramycin powder to vancomycin powder at the time of definitive fixation did not reduce deep surgical site infections compared with vancomycin powder alone.

Trial registration: ClinicalTrials.gov Identifier: NCT02227446.

Associated data

  • ClinicalTrials.gov/NCT02227446