Late Antibiotic Use Among Preterm Infants Admitted to the Neonatal Intensive Care Unit.

Abstract

To describe antibiotic use patterns in preterm infants aged over 3 days. This was a retrospective cross-sectional cohort study using the Premier Health Database, including infants at or under 34 weeks' gestation who were admitted to a neonatal intensive care unit (NICU) from 2009 to 2023. Infants who were transferred, discharged, or who died before day 4 and antibiotic courses initiated on days 1 through 3 after birth were excluded. Late antibiotic regimens were defined by initiation on or after day 4 and categorized by the antibiotics prescribed on the first day of each course. Outcomes included the proportion of infants exposed to any late antibiotic, distribution of antibiotic regimens over the study period, and distribution of antibiotic regimens over the course of NICU hospitalization. Among 420 687 eligible infants admitted to 699 centers over 15 years, 65 398 infants (15.5%) were administered late antibiotics. Antibiotic administration was inversely correlated with gestational age, with 75.0% of infants born 22 to 24 weeks given late antibiotics. There were 1020 unique late antibiotic regimens identified: vancomycin and gentamicin (19.2%) and ampicillin and gentamicin (12.0%) were the most frequent late regimens. Between 2009 and 2023, use of nafcillin or oxacillin-containing regimens (7.3% to 17.8%) and use of piperacillin/tazobactam or cefepime-containing regimens (4.5% to 24.3%) increased, whereas the use of vancomycin-containing regimens decreased (58.8% to 36.0%). Antibiotic choice changed over the NICU care course, with cefazolin being the dominant exposure after 90 days of age. Significant heterogeneity in late antibiotic administration and content supports the need for late antibiotic stewardship guidance among preterm infants.