Our objective was to assess the perinatal management and neonatal outcomes of premature, severely intrauterine growth-restricted (IUGR) neonates. A cohort of neonates <1000 grams, < or = first percentile for weight, and <37 weeks' gestation was identified and matched 2:1 to two control sets of premature, appropriate-for-gestational age (AGA) infants-one with similar gestational age (AGA-GA group) and the other with similar birth weight (AGA-BW group) to determine the effect of IUGR on the outcome of the premature infant. The IUGR group was then examined in detail for descriptive statistics. Data were analyzed by t-tests and Chi-square analyses where appropriate. The IUGR infants had worse outcomes than AGA-GA controls but had somewhat better results than the AGA-BW controls. In the IUGR group, a birth weight less than 550 grams was significantly associated with neonatal death (p < 0.001). However, increasing gestational age was not associated with neonatal survival (p = 0.661) if birthweight remained below 550 grams. Classical cesarean delivery was associated with neonatal death (p = 0.003). Neonatal variables associated with poor outcome included patent ductus arteriosus (p = 0.034), feeding intolerance (p = 0.046), and failure to thrive (p = 0.05). Overall, neonatal survival was 73%. Of the surviving neonates, 69% had evidence of neurodevelopmental delay when tested at 6 and 12 months. Premature, growth-restricted neonates with birth weights of <550 grams versus those of >550 grams have dismal outcomes despite a gestational age that is compatible with survival.