Is It Time to Rethink the Notion of Carbapenem-Sparing Therapy Against Extended-Spectrum β-Lactamase–Producing Enterobacteriaceae Bloodstream Infections? A Critical Review

Abstract

Objective:To present systematic recommendations for carbapenem-sparing therapy against extended-spectrum β-lactamases (ESBLs) Enterobacteriaceae bloodstream infections (BSIs) derived from a critical review of clinical data. Data Sources: A systematic literature search using PubMed and MEDLINE databases (January 1, 2012, to June 30, 2017) was performed using key MESH terms: ESBL or extended-spectrum β-lactamases and bacteremia or bloodstream infection with piperacillin/tazobactam, ciprofloxacin, levofloxacin, cefepime, cephamycins, carbapenem, doripenem, meropenem, and ertapenem. References within articles of interest were also evaluated. Study Selection and Data Extraction: All English language trials were considered, and results were limited to clinical efficacy trials. Articles were screened by title and abstract for inclusion. Data Synthesis: Studies comparing noncarbapenem versus carbapenem therapy for ESBL BSIs were critically analyzed to identify heterogeneity among studies. Data abstracted included empirical or definitive therapy, patient population, dosing, source of infection and severity, infectious etiology, and outcome. Conclusions: Completely sparing carbapenem therapy cannot be justified among patients with ESBL BSIs. Determining the source of infection is critical to identify patients for whom carbapenem-sparing therapy is appropriate.