519. Multidrug-resistant Gram-Negative Bacteremia in Pediatric Patients: Is It Time to Change to Empiric Meropenem?

BACKGROUND: The rise of antimicrobial resistance among gram-negative (GN) pathogens has been dramatic nationally. Delayed initiation of active antimicrobial agents has been associated with poor outcomes. We aimed at evaluating the prevalence and treatment of multi-drug-resistant gram-negative (MDR-GN) bacteremia in our pediatric patients. METHODS: All episodes of GN bacteremia from 2017–2018 at our institution were retrospectively reviewed. GN defined as MDR in our study were carbapenem-resistant organisms (CRO), extended-spectrum β-lactamase (ESBL) producers, and GN that were resistant to cefepime and ≥2 classes of non-cephalosporin antimicrobial agents. Stenotrophomonas maltophilia was excluded. Ineffective empirical treatment (IET) is defined as an initial antibiotic regimen that is not active against the identified pathogen[s] based on in vitro susceptibility testing results. RESULTS: A total of 292 episodes of GN bacteremia were identified and 6 S. maltophilia were excluded. Of these, 29 bacteremic episodes in 26 patients were caused by MDR-GN organisms including 18 ESBL, 7 CRO, 1 ESBL and CRO, 3 non-ESBL/non-CRO cefepime-resistant MDR-GN. None of the CRO had carbapenemase genes detected. However, there was a patient with multiple sites of infection simultaneously with non-NDM CR Acinetobacter bacteremia and NDM-mediated CR-Klebsiella ventriculitis. The annual rate of MDR-GN bacteremia increased from 8% in 2017 to 12% in 2018. Almost half (48%) of episodes were community onset. Among these, all but one had underlying medical conditions with hospital exposure and most patients had central venous devices at the time of infection. 52% (15/29) episodes of MDR-GN bacteremia had IET. Despite IET, 47% (7/15) had negative blood cultures prior to initiation of effective therapy (6 ESBL and 1 P. aeruginosa). Various antibiotic regimens were used for CRO therapy as shown in Table 1. CONCLUSION: In our institution, MDR-GN infection is increasing. As such, empiric meropenem is currently recommended in BMT or neutropenic patients with suspected sepsis. However, empiric meropenem must be used judiciously as its widely use will lead to more selection of MDR pathogens. It is essential to continue monitoring of these MDR-GN to guide appropriate empiric regimens. [Image: see text] DISCLOSURES: All authors: No reported disclosures.