2302. Bloodstream Infections Due to Carbapenem-Resistant Gram-Negative Bacteria in Pediatric Intensive Care Unit (PICU): Risk Factors and Outcomes

BACKGROUND: Bloodstream infections (BSI) caused by multidrug-resistant bacteria are associated with poor outcome and increased cost. We investigated risk factors for carbapenem resistance (CR) and outcome associated with the development of BSI due to Gram-negative (GN) bacteria in PICU patients, a very vulnerable population. METHODS: We reviewed the records of 1 month–15 year old patients with documented GN BSI hospitalized in a PICU from 2005 to 2017. Isolates with meropenem MIC ≥16 mg/L were considered as resistant. Demographics, clinical characteristics, potential risk factors for acquisition of resistant strains, treatment, potential source control and outcome were recorded. Outcome was determined as microbiological response (negative blood cultures) within 5 days and mortality within 30 days. Both univariate and multivariable logistic regression analysis was performed and odds ratios (OR) with 95% confidence intervals (CI) were presented. RESULTS: 81 patients with GN BSI were studied (34.6% Pseudomonas aeruginosa, 34.6% Acinetobacter baumannii and 30.9% Enterobacteriaceae), 21 with CR isolates. Risk factors for CR BSI were: prior carbapenem use (OR: 3.86, 95% CI: 1.10, 13.82) and renal replacement therapy (OR: 3.86, 95% CI: 1.10, 13.82). In multivariable outcome analysis, high levels of CRP (OR: 0.99, 95% CI: 0.99, 0.999), renal replacement therapy (OR: 0.11, 95% CI: 0.01, 0.71) and inotrope administration (OR: 0.30, 95% CI: 0.09, 0.91) were associated with poor microbiological response, whereas source control (OR: 2.99, 95% CI: 1.01, 9.43) with better microbiological response. High PRISM score III (OR: 1.15, 95% CI: 1.04, 1.29) and CR (OR: 5.07, 95% CI: 1.47, 19.36) were both independently associated with worse outcome, whereas source control was the only independent factor preventing death (OR: 0.24, 95% CI: 0.06, 0.78). In patients with CR BSI, administration of at least two active antimicrobials was associated with better outcome (OR: 10.80, 95% CI: 1.33, 237.05). CONCLUSION: Prior carbapenem use is associated with carbapenem-resistant BSI development in PICU, which in turn is an independent risk factor for mortality. Source control is associated with better microbiological response within 5 days, as well as with decreased mortality. DISCLOSURES: All authors: No reported disclosures.