There are no reported randomized trials to determine the ideal timing of antibiotic treatment for intra-amniotic infection. We evaluated the effect of intrapartum versus immediate postpartum treatment of intra-amniotic infection on maternal and neonatal morbidity and mortality. Two hundred fifty-seven women with clinically diagnosed intra-amniotic infection who had amniotic fluid cultures were evaluated. Patients received treatment with penicillin and gentamicin, but the timing of the treatment was determined at the physician's discretion. Most patients (82%) received intrapartum treatment; the remaining women (18%), mainly those with an anticipated short interval before delivery, received the same antibiotics immediately postpartum. As expected, the postpartum treatment group had a significantly shorter diagnosis-to-delivery interval (1.9 +/- 2.1 versus 4.7 +/- 4.3 hours; P less than .001) and a lower maximum temperature during labor (100.8 +/- 0.7 versus 101.0 +/- 0.8F; P = .038). The two treatment groups did not differ in distribution of low birth weight infants, frequency of maternal bacteremia, mode of delivery, or organisms isolated from the amniotic fluid. There were no differences in maternal outcome, but the incidence of neonatal sepsis was significantly lower in the intrapartum treatment group (2.8 versus 19.6%; P less than .001). Neonatal mortality from sepsis was also lower in the intrapartum treatment group (0.9 versus 4.3%), but this difference was not statistically significant. The reduced frequency of neonatal septicemia observed in the intrapartum-treated group might reflect early intrauterine therapy for the infected fetus.