Fifty-eight women with severe preeclampsia between 28-34 weeks' gestation qualified for a randomized controlled trial to establish whether elective delivery 48 hours after administration of betamethasone (aggressive-management group) or delivery later as indicated by maternal or fetal condition (expectant-management group) was more beneficial to maternal and fetal outcome. Twenty women who qualified were not randomized because they developed maternal or fetal indications necessitating delivery within 48 hours; these newborns developed most of the complications. Expectant management was not associated with an increase in maternal complications, but it significantly prolonged the gestational age (mean 7.1 days; P less than .05), reduced the number of neonates requiring ventilation (P less than .05), and reduced the number of neonatal complications (P less than .05).