P2.E095: A Quality Improvement Project to Reduce Vancomycin Utilization in the Neonatal Intensive Care Unit

Purpose/Objectives: Prior to 2023, empiric management of late onset sepsis (LOS) in our Neonatal Intensive Care Unit (NICU) consisted of vancomycin and gentamicin. The American Academy of Pediatrics and the Pediatric Infectious Diseases Society recommend avoiding empiric vancomycin in neonates without risk factors for methicillin-resistant Staphylococcus aureus (MRSA) due to vancomycin associated nephrotoxicity, ototoxicity and emergence of resistant organisms. This project aimed to reduce unnecessary utilization of vancomycin by changing our LOS empiric antibiotic regimen. The aim was to reduce the mean vancomycin antibiotic utilization rate (AUR) in the NICU by 30%.

Design/Methods: Patients admitted to the NICU from March 15, 2023 through February 29, 2024 with suspected LOS were included. Patients receiving vancomycin for other indications, including necrotizing enterocolitis (NEC), were excluded. Data were captured by weekly chart review of patients initiated on LOS empiric antimicrobial therapy. A multidisciplinary team used the Model for Improvement to design and execute the project. Interventions tested using Plan, Do, Study, Act cycles included: education sessions, clinical practice guideline (CPG) implementation, and prospective audits with real-time feedback for patients receiving vancomycin for over 48 hours.The outcome measure was the vancomycin AUR measured in DOT per 1000 patients days. The AUR was plotted monthly on a U-chart and analyzed for special cause variation (Figure 1). The main process measure was CPG adherence. Balancing measures included: Gram-positive sepsis-related mortality, delayed onset of effective antimicrobial treatment, vancomycin-associated acute kidney injury (AKI) and oxacillin-associated hepatotoxicity events.

Results: During the intervention period, fifty patients underwent LOS evaluations and 41 (82%) received recommended empiric treatment with oxacillin and gentamicin. The vancomycin AUR decreased by 39.3%, from 28 to 17 DOT/1000 patient days. We met rules for special cause variation in March 2023, and shifted our center-line.
Our CPG adherence was 96%.There were no Gram-positive sepsis mortalities or AKI or hepatotoxicity events. Of the 41 patients treated empirically, three required changing from oxacillin to vancomycin due to oxacillin-resistant coagulase-negative Staphylococci bacteremia (two cases) and urinary tract infection (one case). None of these patients had prolonged bacteremia or dissemination of infection.

Conclusion/Discussion: This project achieved its aim to safely and effectively reduce empiric vancomycin use in patients with LOS. Provider education and the CPG were key, demonstrating the project’s importance and providing clinical decision-making support.