University of Tennessee Health Science Center Memphis, Tennessee, USA
Introduction: Following burn injury, patients are at increased risk of infection due to the loss of their primary barrier to infection, often subject to protracted hospital stays, repeated surgical procedures, and multiple exposures to topical and systemic antimicrobials. Severe burn injury is often cited as having an increased incidence of difficult to treat pathogens (DTp), compared to general hospitalized patients. The purpose of this study is to determine incidence of DTp after burn injury, which factors are associated with their development, and subsequent outcomes.
Methods: This single center, retrospective study assessed patients with thermal or inhalation injury who had a positive culture resulting in initiation of treatment (i.e. excision, topical, or systemic anti-microbials). Demographic data, pathogen and resistance profiles, and prior exposure to topical and systemic antimicrobials were collected. Pathogens were considered DTp if they were MDR, XDR, MRSA, ESBL-producing, AmpC-producing, carbapenem resistant, DTR Pseudomonas sp., CRAB, or Stenotrophomonas sp.
Results: Sixty-five patients who grew 376 pathogens were included in the final analysis. Two-hundred thirteen (56.7%) pathogens were considered DTp. Prior exposure to 7 of the 11 collected topical antimicrobials and 9 of 11 systemic antimicrobial classes were significantly associated with future development of a DTp. This remained true for 6 and 8, respectively, after controlling for percent burn and presence of flame injury via logistic regression. The most predictive model for developing DTp identified flame injury [OR 5.4 (95% CI 2.0, 14.2)] and exposure to bacitracin [2.9 (1.7, 4.9)], silver nitrate [2.0 (1.2, 3.4)], and anti-pseudomonal beta-lactams [2.7 (1.6, 4.8)] as the strongest predictors for developing DTp. These four variables remained significant predictors after controlling for culture evolution over time within each patient (e.g. repeated measures). As there were only 4 deaths, a cox-proportional hazard analysis was not feasible. Though the Kaplan-Meier plot according to DTp revealed a clear divergence in mortality (Logrank p = 0.0583).
Conclusions: In this analysis, exposure to topical and systemic antibiotics were associated with development of DTp.
