OBJECTIVE: To evaluate whether intraventricular hemorrhage and periventricul ar leukomalacia are characterized by different risk factors. METHODS: In a cohort of 653 consecutive singleton neonates born after preterm membrane rupture, spont aneous preterm labor, or indicated preterm delivery at 24 to 33 weeks of gestati on from January 1, 1993, to December 31, 2002, we evaluated the obstetric and histopathologic pla cental variables in reference to the development of intraventricular hemorrhage (n = 44), periventricular leukomalacia (n = 19), or no ultrasonographic cerebral lesion (n = 589). Excluded were stillbirths and congenital anomalies. Statistic al analysis included Fisher exact test, Student t test, and stepwise logistic re gression analysis with a 2-tailed P < .05 consi-dered significant. RESULTS: Mu ltivariate analysis showed that occurrence of neonatal intraventricular hemorrha ge and periventricular leukomalacia were associated only with spontaneous premat urity (odds ratio = 1.9; 95%confidence interval 1.1-3.4) and gestational age a t delivery in weeks (odds ratio = 0.8; 95%confidence interval 0.7-0.9). Neonat es with intraventricular hemorrhage did not differ from those with periventricul ar leukomalacia in any obstetric or neonatal variable, but there was a higher ri sk of neurodevelopmental delay associated with periventricular leukomalacia. CON CLUSION: Among premature infants born at less than 34.0 weeks of gestation, intr aventricular hemorrhage and periventricular leukomalacia share common clinical c haracteristics, with spontaneous preterm delivery and gestational age at deliver y as the only independent antenatal predictors. METHODS: In a cohort of 653 consecutive singleton neonates born after preterm membrane rupture, spont aneous preterm labor, or indicated preterm delivery at 24 to 33 weeks of gestati on from January 1, 1993, to December 31, 2002, we evaluated the obstetric and histopathologic pla cental variables in reference to the development of intraventricular hemorrhage (n = 44), periventricular leukomalacia (n = 19), or no ultrasonographic cerebral lesion Statistic al analysis included Fisher exact test, Student t test, and stepwise logistic re gression analysis with a 2-tailed P <.05 consi-dered significant. RESULTS: Mu ltivariate analysis showed that occurred of neonatal intraventricular hemorrha ge and periventricular leukomalacia were associated only with spontaneous prematurity (odds ratio = 1.9; 95% confidence interval 1.1-3.4) and gestational age at delivery in weeks (odds ratio = 0.8; 95% confidence interval 0.7-0.9). Neonat es with intraventricular hemorrhage did not differ from those with periventricul ar leukomalacia in any obstetric or neonatal variable, but there was a higher ri sk of neurodevelopmental delay associated with periventricular leukomalacia. CON CLUSION: Among premature infants born at less than 34.0 weeks of gestation, intr aventricular hemorrhage and periventricular leukomalacia share common clinical c haracteristics, with spontaneous preterm delivery and gestational age at deliver y as the only independent antenatal predictors.