518. Comparing the Mortality of Carbapenemase-Producing and Non-Carbapenemase-Producing Carbapenem-Resistant Enterobacteriaceae Bacteremia

BACKGROUND: Carbapenem-resistant Enterobacteriaceae (CRE) infection is an emerging clinical issue. One of the mechanisms of carbapenem-resistance is carbapenemase production. This study aimed to identify whether clinical outcomes differ by CRE resistance mechanism and to evaluate risk factors for mortality in patients with CRE bacteremia. METHODS: We conducted a retrospective cohort study comparing 14-day mortality between patients with carbapenemase-producing (CP)-CRE and non-CP-CRE bacteremia during January 2011 to October 2018. Only monomicrobial Escherichia coli or Klebsiella pneumoniae bacteremia were included in the study. A modified carbapenem inactivation method was used for phenotypic detection of carbapenemase production. The presence of a variety of carbapenemase genes was evaluated by PCR with specific primers. RESULTS: Of 134 patients with monomicrobial CRE bacteremia, 48 (35.8%) were infected with CP-CRE, and 86 (64.1%) were infected with non-CP-CRE. The most common carbapenemase in CP-CRE isolates was KPC (66.7%), followed by NDM-1 (18.8%), OXA-48-like (10.4%), and VIM (4.1%). Baseline characteristics were similar between the two groups (Table 1). However, the CP-CRE group was significantly more likely to undergo removal of eradicable foci and to have meropenem MIC >8 µg/mL. A total of 33 (24.6%) patients died within 14 days, including 9 (18.8%) in the CP-CRE group and 24 (27.9%) in the non-CP-CRE group. Deceased patients were more likely to have a higher Pitt bacteremia score, nosocomial acquisition, ineradicable or not-eradicated foci, immunosuppressant use, inappropriate definitive treatment (Table 2). Combination therapy for definitive treatment was associated with decreased mortality. In a multivariate analysis including carbapenemase production, a higher Pitt bacteremia score (aOR, 5.15), ineradicable or not-eradicated foci (aOR, 4.05) and combination therapy for definitive treatment (aOR, 0.35) were independent risk factors for mortality. CONCLUSION: Our study suggests that carbapenemase production is not a mortality risk factor in CRE bacteremia and provides additional evidence for early source control and combination therapy. [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.