Intrapartum antibiotic prophylaxis for Group B Streptococcus and risks of unnecessary antibiotics

We much appreciated the Clinical Opinion by Turrentine1 regarding intrapartum antibiotic prophylaxis (IAP) for the prevention of neonatal group B Streptococcus (GBS) early-onset sepsis. The author raises concerns about cases of inadequate duration (<4 hours) of IAP before delivery. Indeed, as much as 25-40% of GBS-colonized women will not receive the 4 hours of IAP that are recommended by guidelines. To optimize fetal exposure, Turrentine recommends strategies such as (1) to postpone artificial rupture of membranes (or administration of oxytocin) until 4 hours IAP is completed and (2) to start IAP before the initiation of uterotonic agents in women who are admitted for induction of labor. Although the first step is feasible, a woman with a Bishop score <5 may be exposed to several antibiotic doses until delivery. The minimum duration of IAP for the prevention of early-onset GBS sepsis remains uncertain, because existing data are conflicting. There is evidence that bactericidal levels of β-lactam antibiotics in fetal blood are achieved as early as 3 minutes and that levels that exceed 10- to 179-fold of the minimal inhibitory concentration for GBS persist for up to 2 hours. Although neonatal colonization (NC) is only a proxy marker of infection, the study of NC gives useful information regarding the effects of IAP within the amniotic fluid. We demonstrated that intravenous ampicillin that is given <2 hours before delivery is very effective in reducing NC.2 More recently, we evaluated NC in a larger prospective cohort study. Almost 40% of 44 neonates without IAP exposure get colonized at 24-48 hours of birth, whereas rates of NC were low (between 9.3% and 0) in the 458 neonates who were exposed to IAP.3 Interestingly, rates of NC did not vary significantly in the range of <1–12 h before delivery (score test for trend of odds; P = .13). No cases of NC were found after 12 hours of IAP. Both studies provide strong evidence of a rapid action of ampicillin within the amniotic fluid, although they do not provide firm evidence regarding the time necessary to ensure the effectiveness. Although the labor process is capricious, delivery is largely a slow, rather than a fast, event (unfortunately for women and their caregivers!). Therefore, we raise concerns regarding intrapartum management strategies that perhaps allow some neonatal benefit, while exposing a greater number of women to unnecessary antibiotics, which is a probable harmful consequence to public health.4