Catheter-related bloodstream infections: predictive factors for Gram-negative bacteria aetiology and 30 day mortality in a multicentre prospective cohort.

BACKGROUND: Catheter-related bloodstream infections (CRBSIs) increase morbidity and mortality, prolong hospitalization and generate considerable medical costs. Recent guidelines for CRBSI recommend empirical therapy against Gram-positive bacteria (GPB) and restrict coverage for Gram-negative bacteria (GNB) only to specific circumstances. OBJECTIVES: To investigate predictors of GNB aetiology in CRBSI and to assess the predictors of outcome in patients with CRBSI. METHODS: Patients with CRBSI were selected from the PROBAC cohort, a prospective, observational, multicentre national cohort study including patients with bloodstream infections consecutively admitted to 26 Spanish hospitals in a 6 month period (October 2016-March 2017). Outcome variables were GNB aetiology and 30 day mortality. Adjusted analyses were performed by logistic regression. RESULTS: Six hundred and thirty-one episodes of CRBSI were included in the study. Risk factors independently related to GNB aetiology were central venous catheter (CVC) [OR 1.60 (95% CI: 1.05-2.44), P = 0.028], sepsis/septic shock [OR: 1.76 (95% CI: 1.11-2.80), P = 0.016], antibiotic therapy in the previous 30 days [OR: 1.56 (95% CI: 1.02-2.36), P = 0.037], neutropenia <500/µL [OR: 2.01 (95% CI: 1.04-3.87), P = 0.037] and peripheral vascular disease [OR: 2.04 (95% CI: 1.13-3.68), P = 0.018]. GNB were not associated with increased mortality in adjusted analysis, while removal of catheter [OR: 0.24 (95% CI: 0.09-0.61), P = 0.002] and adequate empirical treatment [OR: 0.37 (95% CI: 0.18-0.77), P = 0.008] were strong protective factors. CONCLUSIONS: Our study reinforces the recommendation that empirical coverage should cover GNB in patients presenting with sepsis/septic shock and in neutropenic patients. Catheter removal and adequate empirical treatment were both protective factors against mortality in patients with CRBSI.