Objectives: This study assessed the impact of discordant empirical antibiotic therapy on the outcome of bacteremia caused by extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli.
Methods: The clinical features and outcomes of a cohort of patients hospitalized with ESBL E. coli bacteremia between 2007 and 2008 were retrospectively reviewed. The effect of different antimicrobial regimens on patient outcomes was analyzed.
Results: ESBL E. coli accounted for 24.2% (207/857) of E. coli bacteremia cases. The urinary tract (43.6%) was the most common source of infection, followed by the hepatobiliary tract (23.0%). Discordant empirical antibiotic therapy was given to 52.0% patients. Admission to the intensive care unit was associated with the use of a carbapenem as empirical antibiotic therapy (p<0.001). Univariate analysis revealed no significant differences in 30-day mortality rates between patients receiving concordant and discordant empirical antibiotic therapy (23.5% vs. 19.8%, p=0.526), carbapenem and non-carbapenem empirical antibiotic therapy (29.8% vs. 19.1%, p=0.118), beta-lactam/beta-lactam inhibitor combinations (BLBLIs) and non-BLBLIs empirical antibiotic therapy (20.3% vs. 22.3%, p=0.734), cephalosporin and non-cephalosporin empirical antibiotic therapy (19.7% vs. 22.6%, p=0.639), and fluoroquinolone and non-fluoroquinolone empirical antibiotic therapy (8.3% vs. 22.4%, p=0.251). The findings were confirmed by multivariate analysis.
Conclusions: Despite a high proportion of discordant empirical antibiotic therapy, ESBL production had little effect on 30-day mortality. Whether the observation can be applied to different ESBL types is unknown and warrants further study.
Copyright © 2012 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.