Purpose: Infections caused by ESBL-producing Enterobacteriaceae (ESBL-EB) are a major health problem, but data regarding elderly patients is lacking.
Methods: We performed a retrospective observational study quantifying the effects of antimicrobial treatment and primary infection site on clinical outcomes in an historical case series of 42 patients aged 80.7±10years admitted to an Internal Medicine ward in Italy for ESBL-EB bloodstream infections (BSI).
Results: At multivariate risk analysis, we found that urinary tract as primary infection site (RR=0.181 [0.037-0.886], p=0.035) and definitive antibiotic therapy (RR=0.517 [0.147-0.799], p=0.038) decreased the relative risk of a negative clinical response, while the respiratory tract origin increased the relative risk (RR=2.788 [1.407-9.228], p=0.025). Also regarding 30days mortality, multivariate risk analysis identified that urinary tract as primary infection site (RR=0.098 [0.011-0.743], p=0.025) and definitive antibiotic therapy (RR=0.236 [0.058-0.961], p=0.044) decreased the relative risk, while the respiratory origin increased the relative risk (RR=4.241 [1.040-17.295], p=0.014). We observed similar outcomes in patients definitively treated with carbapenems or with carbapenem-free treatments. Additionally, an initially inappropriate therapy did not correlate with worse outcomes if a switch to an effective definitive treatment was performed promptly.
Conclusions: Carbapenem-sparing regimens (e.g. piperacillin-tazobactam alone or with an aminoglycoside) could be empirically safely used in elderly patients at high risk of ESBL-EB BSI and for definitive treatment of ascertained cases if the primary site is the urinary tract, leaving early carbapenem use for cases at higher risk of death, such as those with pneumonia.
Keywords: Carbapenem; ESBL; Elderly; Enterobacteriaceae; Internal medicine; Outcome.
Copyright © 2017 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.