BACKGROUND AND OBJECTIVE: In 2014 at Nationwide Children’s Hospital, the Neonatal Antimicrobial Stewardship Program recommended nafcillin over vancomycin for empirical therapy of possible late-onset sepsis (LOS) in infants without a history of methicillin-resistant Staphylococcus aureus colonization or infection. We report our experience with this guideline and assess its safety. METHODS: We retrospectively reviewed all infants who received nafcillin or vancomycin for empirical treatment of possible LOS at 3 NICUs before (January 2013–May 2014) and after (January 2017–March 2019) implementation of a vancomycin reduction guideline. Safety measures included duration of blood culture positivity, recurrence of infection with the same previously identified pathogen in the 14 days after discontinuation of antibiotic therapy, and mortality. RESULTS: Among 366 infants who received a first antibiotic course for possible LOS, 84% (95 of 113) and 25% (62 of 253) received empirical therapy with vancomycin before and after the guideline implementation, respectively, representing a 70% reduction. Nafcillin use increased by 368%. Duration of blood culture positivity did not differ before and after the guidance. In 2 infants, antibiotic therapy was restarted within 14 days of discontinuation of the initial therapy for recurrence of the same infection; both had received empirical vancomycin. Overall in-hospital mortality was 10%, and there was no difference before (9%) and after (10%) implementation of the vancomycin reduction guidance (odds ratio, 0.97). CONCLUSIONS: Nafcillin can be a safe alternative to vancomycin for empirical therapy of LOS among NICU infants who do not have a history of methicillin-resistant S aureus infection or colonization.

No abstract