147. Non-Fermenting Gram-Negative Bloodstream Infection: A Multicenter Retrospective Cohort Study

Abstract

BackgroundFew data exist on clinical characteristics, therapeutic management and outcome of patients with Non-Fermenting Gram-negative bloodstream infection (NFGN-BSI). Our aim is to describe a large cohort of patients with NFGN-BSI and to investigate risk factors for 30-day mortality. Further, the impact of the new difficult to treat resistance (DTR) definition will be investigated.MethodsRetrospective multicenter study of patients diagnosed with NFGN-BSI at three large Italian hospitals (Bologna, Genova, Torino), over a 4-year period (2013–2016). Exclusion criteria: age <18 years, clinical data not available, polymicrobial BSI, death within 72 hours from drawing index blood cultures (BCs). Carbapenem resistance (CR) was defined according to 2015 CDC definitions, and DTR as resistance to all β-lactams and fluoroquinolones. Active empiric therapy (AET) was defined as at least one in vitro active drug administered within 24 hours from drawing index BCs. Endpoint was all-cause 30-day mortality.Results521 patients with NFGN-BSI were analyzed: 63.3% male, median age 67 (IQR 55–78) years, median Charlson index 6 (IQR: 3–7). Most episodes were hospital acquired 72.9%. Etiology distribution: Pseudomonas aeruginosa 69.9%, Acinetobacter baumannii 19.6%, and Stenotrophomonas maltophilia 10.6%. CR and DTR rates were 38.6% and 26.9%. Main infection sources were deemed as primary 50.7%, CVC-related 26.5%, and lower respiratory tract 16.3%. Source control and ID consultation were performed in 33.4% and 47.6% of cases. AET rate was 38.2%. Empiric and definitive antibiotic treatment cohorts consisted of 377 and 472 patients, respectively. There was high heterogeneity in antibiotic choice with 30 and 48 different regimens in empiric and definitive cohort, respectively. Combination therapy was administered in 22.3% of empiric cohort and in 37.3% of definitive cohort patients. Independent risk factors for 30-day mortality were age (HR 1.03, 95% CI 1.01–1.05, P = 0.001), SOFA (HR 1.25, 1.15–1.36, <0.001), DTR (HR 2.73, 1.60–4.65, <0.001), and AET (HR 0.50, 0.25–0.99, 0.05).ConclusionHigh heterogeneity in therapeutic management of patients with NF-GNBSI was observed. DTR was a strong predictor of mortality, AET was associated with improved outcome.Disclosures All authors: No reported disclosures.