Abstract
BACKGROUND: Enterococcal infections represent 10% of intensive care unit (ICU)-acquired infections and are associated with adverse outcomes, particularly in the case of E. faecium infections. Some studies focusing on non-critically ill patients have cast doubt on the non-inferiority of daptomycin compared to other antibiotics for these infections. We aimed to describe antibiotic treatment practices for monomicrobial E. faecium bloodstream infections (BSI) in patients admitted to the ICU, and to identify factors associated with treatment failure. METHODS: This retrospective multicenter study included adult patients presenting with a monomicrobial E. faecium BSI during their stay in one of 11 ICUs of Paris University hospitals between 2017 and 2022. Patients receiving daptomycin as a definitive antibiotic regimen were compared to those receiving another molecule using competing-risks models and propensity score-based analyses. The primary outcome was a composite criterion of treatment failure including: (1) prolonged bacteremia (≥3days) under treatment and/or (2) relapse within 30 days after the end of treatment and/or (3) the need for salvage therapy within 30 days. RESULTS: Of the 166 included patients, 26 received daptomycin as a definitive antibiotic regimen, at a median dose of 10 [10-10] mg/kg/day and 140 patients received non-daptomycin-based regimens (vancomycin (69%), linezolid (19%) or a beta-lactam (11%)). Source of BSI was predominantly unknown (38%), digestive (37%) or an intravascular device (16%). All isolates were vancomycin-susceptible. Median daptomycin MIC was 3 [2-3.25] mg/L. Time to adequate antibiotic regimen (1 [0-2] vs 1 [0-2] days, p=0.22) and rates of source control (63% vs 43%, p=0.17) were not different between groups. Regarding primary outcome, daptomycin as a definitive antibiotic regimen was associated with a higher rate of treatment failure in the multivariate adjusted analysis using a Fine and Gray model (aSHR 2.53 [1.14-5.62], p=0.022). Sensitivity analyses using propensity score overlap weighting (HR 2.48 [1.05-5.86], p=0.038) or with only patients treated by vancomycin as comparators (aSHR 2.26 [1.00-5.10], p=0.051) yielded similar results. CONCLUSIONS: In this multicenter retrospective cohort of critically ill patients with monomicrobial E. faecium bloodstream infections, use of daptomycin as the definitive antibiotic regimen was associated with a higher rate of treatment failure. Further prospective studies are needed.