To determine whether cesarean delivery is associated with a better outcome than vaginal delivery for infants weighing less than 1500 g (very low birth weight), we examined neonatal mortality and the incidence of intraventricular hemorrhage (IVH) in 1765 very low birth weight inborn infants admitted to seven neonatal intensive care centers. The cesarean rate was 32.5% for infants weighing 501-750 g and 52.4% for infants weighing 751-1000 g. The neonatal death rate was 53.1% for infants weighing 501-750 g delivered by cesarean, compared with 64.3% for vaginally born infants (P = .046). However, for infants weighing 1001-1250 g, the neonatal death rate for infants delivered by cesarean was 14.4%, compared with 7.8% for infants born vaginally (P = .02). The incidence of IVH was significantly lower in infants born by cesarean than in those born vaginally only in the 1251-1500-g birth weight interval (11.8 versus 18.9%; P = .03). Compared with women delivering vaginally, those delivered by cesarean were more likely to attain a higher mean gestation, to have preeclampsia, and to have a breech presentation, and less likely to be in labor. After adjusting by logistic regression for gestational age, preeclampsia, breech presentation, presence or absence of labor, and for center effects, the odds ratio for neonatal death was 1.00 (95% confidence interval 0.71-1.41); for IVH, the odds ratio was 0.85 (95% confidence interval 0.61-1.19). These data suggest that after accounting for certain maternal and fetal factors, cesarean delivery is not associated with a lower risk of either mortality or IVH.