Ertapenem in the treatment of bacteremia caused by extended-spectrum beta-lactamase-producing Escherichia coli: a propensity score analysis

Objective: This study assessed the impact of ertapenem and other carbapenems on mortality in patients with monomicrobial extended-spectrum beta-lactamase-producing Escherichia coli (ESBL-EC) bacteremia. Methods: This non-concurrent prospective study included adult patients with ESBL-EC bacteremia during a 2.5-year period at a 2200-bed teaching hospital. We used a multivariate logistic regression model and Cox's proportional hazards model including propensity score analysis to assess variables associated with 30-day mortality. Results: Of 71 patients who met the study criteria, nine died within 3 days. Among the 62 remaining patients who received definitive antimicrobial therapy, 13 died within 30 days. Male gender, ICU stay, solid tumor, and primary bacteremia were independent predictors of 30-day mortality, whereas definitive antimicrobial therapy using either ertapenem or imipenem/meropenem was protective (p < 0.001 and p = 0.002, respectively). Adjustment by propensity score found that ertapenem appeared to exhibit more favorable outcomes, but the difference fell short of statistical significance (hazard ratio 0.02, p = 0.06). Inappropriate initial therapy was not a significant predictor of mortality. Conclusions: ICU stay, but not initial choice of empirical antimicrobial therapy, was a major predictor of mortality. Using a carbapenem as definitive therapy was a protective factor for 30-day mortality. The choice of ertapenem is reasonable for less severely-ill patients who are at risk of ESBL-EC bacteremia and unlikely to have infection due to Pseudomonas aeruginosa. (C) 2011 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.