556 - Neonatal intensive care unit antibiotic utilization from 2009 to 2019

Poster Number: 556
Publication Number: 556.428

Dustin D. Flannery, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Alvaro Zevallos Barboza, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Sagori Mukhopadhyay, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Jeffrey S. Gerber, CHOP, Philadelphia, PA, United States; Di Shu, University of Pennsylvania & Children’s Hospital of Philadelphia, Philadelphia, PA, United States; sean Hennessy, University of Pennsylvania Perelman School of Medicine, PHILADELPHIA, PA, United States; Kelly Wade, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States; Karen M. Puopolo, Childrens Hospital of Philadelphia, Philadelphia, PA, United States

Background: Antibiotics are the most common medications administered to infants in neonatal intensive care units (NICUs). The impacts of neonatal antibiotic stewardship efforts related to sepsis risk assessment and vancomycin reduction strategies, as well as cefotaxime shortage/discontinuation, on NICU antibiotic utilization over the last decade are not well understood.

Objective: To assess antibiotic utilization trends across a large cohort of infants admitted to academic and community NICUs across the US over an 11-year period.

Design/Methods: Repeated cross-sectional study of inborn infants admitted to the NICU in the Premier Database from 2009-2019. We analyzed by year the proportion of a) all NICU infants and b) extremely low birth weight (ELBW; < 1000 grams) infants receiving specific antibiotics for at least one dose. Absolute differences and relative differences between 2009 and 2019 proportions were calculated overall and for each medication. Generalized linear regression was used to estimate annual absolute difference and 95% CI, accounting for clustering by center.

Results: From 2009-2019, 1,144,023 infants were admitted to 657 NICUs, of which 31,200 (2.7%) were ELBW. Infants were 54.7% male sex, 56.1% White race, 14.7% Black race, and 11.4% Hispanic ethnicity; median length of stay was 5 days (IQR 3,13). Overall, 46.6% of NICU infants and 90.1% of ELBW infants received an antibiotic. There was a 37% relative reduction in all infants receiving antibiotics but only a 3% reduction specifically among ELBW infants (Table). For the overall cohort, there were decreases in relative proportions of infants who received ampicillin and gentamicin (each by 37%), vancomycin (by 49%), and cefotaxime (by 94%), while ceftazidime use increased (by 170%) (Table and Figure 1). For the ELBW cohort, there were decreases in relative proportions of infants who received gentamicin (by 8%), cefotaxime (by 87%), and vancomycin (by 31%), and increases for cefepime (by 184%), nafcillin/oxacillin (by 101%), and piperacillin/tazobactam (by 57%) (Table and Figure 2).Conclusion(s): This study of over 1 million NICU infants demonstrates significant changes in antibiotic utilization over 2009-2019. We speculate that implementation of antibiotic stewardship efforts targeting sepsis risk assessment and vancomycin reduction, and the evolving national shortage of cefotaxime, may have contributed to our findings. Continued antibiotic stewardship efforts, especially among ELBW infants, and utilization surveillance remain important to optimize antibiotic use in NICUs.
Table: Antibiotic utilization trends
Figure 1: Antibiotic utilization trends over time (overall NICU cohort)