Prolonged use of antibiotics after birth is associated with increased morbidity in preterm infants with negative cultures

Abstract

Background: Most preterm infants are exposed to a variable duration of antibiotic therapy after birth despite negative cultures. Data is emerging about the risks of prolonged antibiotics. We sought to assess the association between length of initial antibiotic course and neonatal outcomes in a cohort from a single large perinatal center. Methods: Retrospective cohort study of prospectively collected data on all infants with a birth weight of less than 1250 g hospitalized in our NICU in a 4 year window and who had negative blood and CSF cultures in the first 2 days of life. The primary outcome is a composite of necrotizing enterocolitis (NEC), late onset sepsis (LOS) and death evaluated using multivariable regression analysis. Results: A total of 620 infants less than 1250 g with negative cultures were eligible for study over a 4 year period. The 238 infants with more than 5 days initial antibiotic use were significantly smaller and of lower gestational age than the 382 infants who received up to 5 days of antibiotics. Their mothers had more clinical chorioamnionitis, less maternal hypertension and greater perinatal use of antibiotics. On multivariate analysis, infants who received empiric antibiotics for longer than 5 days had higher rates of neonatal morbidities after adjusting for gestational age, SNAP II, small-for-gestational age status, gender, maternal hypertension, prenatal steroid treatment, clinical chorioamnionitis, intrapartum antibiotic treatment, and multiple births. Composite outcome OR: 1.83 (1.15 to 2.92), LOS OR: 2.02 (1.20 to 3.39), bronchopulmonary dysplasia OR: 1.58 (1.04 to 2.29). Mortality and NEC were not significantly different. Conclusion: More than 5 days of antibiotic treatment in very preterm infants with negative cultures was associated with increased morbidity in our population, and that included BPD. It is of note that patterns of increased morbidity and/or mortality differ between studies. Prospective trials of clinical protocols for starting and stopping antibiotics in the very preterm infants are required.