Appropriate non-carbapenems are not inferior to carbapenems as initial empirical therapy for bacteremia caused by extended-spectrum beta-lactamase-producing Enterobacteriaceae: a propensity score weighted multicenter cohort study.

The efficacy of empirical non-carbapenem antibiotics for extended-spectrum beta-lactamase-producing Enterobacteriaceae bacteremia (ESBL-B) is still inconclusive. We conducted a multicenter retrospective cohort study to evaluate the efficacy of empirical non-carbapenem antibiotics for treating ESBL-B. Electronic medical records of individuals who were diagnosed with ESBL-B were reviewed between January 2010 and December 2014 at four university hospitals in Korea. Patients were classified into non-carbapenem and carbapenem groups according to the empirical antibiotic regimen. Patients treated with appropriate empirical antibiotics and who subsequently received carbapenems as definitive therapy were included in the analysis. The inverse probability of treatment weights, a statistical method that adjusts baseline statistics by giving weights based on propensity score, was used. During the study period, 232 adequately treated patients with ESBL-B were included in the analysis: 49 patients in the non-carbapenem group and 183 in the carbapenem group. The baseline characteristics and severity of infection were similar after propensity score weighting. The 30-day mortality rates for the two groups were not statistically significantly different (non-carbapenems 6.3% and carbapenems 11.4%; P = 0.42). In a multivariate analysis, empirical treatment with non-carbapenem antibiotics was not associated with 30-day all-cause mortality (HR 1.02, 95% CI 0.99-1.06, P = 0.14). In a subgroup analysis, empirical treatment with piperacillin-tazobactam was also not associated with 30-day all-cause mortality (HR 1.21, 95% CI 0.37-4.00, P = 0.75). Appropriate non-carbapenems were not inferior to carbapenems as initial empirical therapy for ESBL-B.