Abstract
OBJECTIVE: We examined ampicillin dosing in pediatric patients across 3 conditions: (1) bacterial lower respiratory tract infections (LRTIs) in infants and children >3 months, (2) neonates with suspected or proven sepsis, and (3) neonates with suspected central nervous system (CNS) infections. We compared our findings to dosing guidance for these specific indications. DESIGN: Retrospective cohort study. SETTING: The study included data from 32 children's hospitals in the United States. METHODS: We reviewed prescriptions from the SHARPS study of antimicrobials, a survey of antibiotic prescribing from July 2016 to December 2017. Prescriptions were analyzed for indication, total daily dose per kilogram, and presence of antimicrobial stewardship program (ASP) review. LRTI prescriptions were compared to IDSA recommendations for community-acquired pneumonia. Neonatal prescriptions were compared to recommendations from the American Academy of Pediatrics (AAP). Prescriptions were categorized as "optimal" (80%-120% of recommended dosing), "suboptimal" (<80% of recommended dosing), or "excessive" (>120% of recommended dosing). RESULTS: Among 1,038 ampicillin prescriptions, we analyzed 88 prescriptions for LRTI, 499 prescriptions for neonatal sepsis, and 27 prescriptions for neonatal CNS infection. Of the LRTI prescriptions, 77.3%were optimal. Of prescriptions for neonatal sepsis, 81.6% were excessive compared to AAP bacteremia recommendations but 78.8% were suboptimal compared to AAP meningitis guidelines. Also, 48.1% of prescriptions for neonatal CNS infection were suboptimal, and 50.6% of prescriptions were not reviewed by the ASP. CONCLUSIONS: LRTI dosing is generally within the IDSA-recommended range. However, dosing for neonatal sepsis often exceeds the recommendation for bacteremia but is below the recommendation for meningitis. This variability points to an important opportunity for future antimicrobial stewardship efforts.