Indications for intra-partum administration of antibiotics (IPA) have progressively broadened over the past several years. We wondered what impact these recommendations have had on the incidence of early-onset sepsis in our own institution. Specifically, we addressed two questions: 1) For which conditions, if any, has the use of IPA increased in our maternal population? 2) Has the increased use of IPA, if noted, reduced the frequency of GBS infection?

Methods: We correlated the use of IPA with risk factors for early-onset sepsis (premature rupture of membranes (pre-ROM); prolonged ROM(pro-ROM); chorioamnionitis) from 1989-1994. We determined the yearly incidence of early-onset GBS sepsis (EOGBS; + blood culture for GBS < 2 days of life). We correlated risk factors and IPA with EOGBS for each septic infant. Finally, we determined whether the incidence of early-onset infection with non-GBS organisms had changed.

Results: Over the 6 year study period, there were no significant changes in the number of deliveries (≈3000/yr), or infants at risk for infection: pre-ROM (≈250/yr), pro-ROM (≈175/yr), chorio (≈200/yr). The use of IPA roughly doubled over the study period for infants with pre-ROM(44% to 74%), pro-ROM (43% to 75%), and no identified risk factors (10% to 20%). Despite the increased use of IPA, the yearly incidence of EOGBS was not significantly reduced (14, 9, 8, 7, 9, 12 for each study year). 44% of infants with EOGBS had no identified risk factors for sepsis. 27% of mothers of EOGBS infants had received IPA. We saw no change in the frequency of early-onset infection by other pathogens (2, 9, 3, 6, 2, 4 for each study year). Infants with early-onset non-GBS sepsis were significantly more likely to have risk factors than EOGBS infants (77% vs 32% premature; 85% vs 56% some maternal risk factor; both p<0.01).

Conclusions: 1) Despite a two-fold increase in use of IPA for infants at risk, the incidence of EOGBS was not reduced. 2) Established maternal risk factors failed to predict sepsis in roughly half of EOGBS infants. 3) Maternal antibiotic use did not prevent infection in one-fourth of affected neonates. 4) Based upon our experience, closer adherence to newly published risk-based guidelines will be less effective in preventing EOGBS than had previously been hoped.