Clinical and economic outcomes for term infants associated with increasing administration of antibiotics to their mothers

Abstract

Implementation of national guidelines for the prevention of group B streptococcal (GBS) infections has led to an increase in intrapartum antibiotic use and reduction in early-onset GBS infections in newborns. Other outcomes, including the clinical diagnosis of sepsis in term infants, treatment with antibiotics, length of stay, and cost have not been described. To examine these outcomes, we performed an analysis of maternal and newborn data collected between 1998 and 2002 of 130 447 in-hospital births of newborns >or=37 weeks gestation and their mothers from a large vertically integrated healthcare organisation in Utah. The main outcome measures included: (i) the number of women delivering at term who received intravenous antibiotics; (ii) the number of newborns treated for 'clinical sepsis', which was defined as receiving antibiotics for >72 h and the number of newborns who received antibiotics for <or=48 h, i.e. a 'rule-out-sepsis' course. We also compared the lengths of stay and variable costs of infants whose mothers received antibiotics with those whose mothers did not. We found that the proportion of mothers who received intravenous antibiotics rose from 26.8% in 1998 to 40.6% in 2002 (P < 0.0001). The proportion of newborns treated for clinical sepsis ranged from 1.2% to 1.4% over the 5-year period. (P for trend = 0.04). After controlling for maternal chorioamnionitis, delivery by caesarean section and maternal GBS status, newborns of mothers who received antibiotics were significantly more likely to be treated for clinical sepsis than were newborns of mothers who had not received them [adjusted OR = 3.3; 95% CI 2.9, 3.8]. The average length of stay for newborns whose mothers were treated with antibiotics was 55.8 h compared with 41.6 h for those not treated (P < 0.0001). The cost of caring for newborns whose mothers received antibiotics was $740 compared with $638 for those whose mothers had not received them (P < 0.001).