1144. Evaluation of Nafcillin vs. Vancomycin as Empiric Therapy for Late-Onset Sepsis in the Neonatal Intensive Care Unit

BACKGROUND: Empiric therapy for possible late-onset sepsis (LOS) due to Gram-positive bacteria in the neonatal intensive care unit (NICU) has included vancomycin to cover, among other pathogens, coagulase-negative staphylococci (CoNS) which are the most frequent cause of bloodstream infections (BSI). In 2015, Nationwide Children’s Hospital (NCH) neonatal antimicrobial stewardship (nASP) team recommended nafcillin rather than vancomycin, in combination with gentamicin, as the preferred agent for empiric therapy of LOS in infants not colonized with methicillin-resistant Staphylococcus aureus, irrespective of presence of a central venous catheter. The NCH nASP team provides oversight for 6 Columbus NICUs, and Our objective was to evaluate the impact of the vancomycin reduction program, with secondary objectives including duration of therapy, recurrence of BSI within 14 days of completion of therapy, and mortality. METHODS: The pharmacy database at NCH was queried with respect to all nafcillin and vancomycin use from 2013–2018. Pertinent clinical and laboratory data were obtained from the electronic health record (EHR) on all infants who had nafcillin or vancomycin therapy initiated, with each initiation defined as an antibiotic “course.” RESULTS: From 1/2013 to December 2014 (pre-vancomycin reduction), there was an average of 112 vancomycin and 42 nafcillin courses provided to infants each year. From 1/2015 to December 2018, the use of nafcillin increased to an average of 90 courses while vancomycin decreased to 55 courses per year (P < 0.01). Since the institution of the vancomycin reduction program, preliminary EHR review of 50 infants has shown that 9 had a positive blood culture (7 CoNS; 2 methicillin-susceptible S. aureus; 1 Escherichia coli). All CoNS isolates were resistant to nafcillin, and all infants sterilized the blood culture within 24 hours of vancomycin. The overall median length of therapy was 3 days with nafcillin or vancomycin. However, when excluding rule outs, the median duration of therapy was 9.5 days. There was no BSI recurrence or infection-related death. CONCLUSION: An empiric antibiotic regimen that includes nafcillin rather than vancomycin for possible LOS in high-risk infants in the NICU effectively and safely reduced overall vancomycin use. DISCLOSURES: All authors: No reported disclosures.