198 Morbidity and mortality among very-low-birth-weight neonates with intrauterine growth restriction Ira M. Bernstein, MD, a Jeffrey D. Horbar, MD, b, e Gary J. Badger MS, c Arne Ohlsson, MD, d and Agneta Golan, MD, d for the Vermont Oxford Network Burlington, Vermont, and Toronto, Ontario, Canada OBJECTIVE: We sought to determine the associations between intrauterine growth restriction and neonatal morbidity and mortality, as well as the impact of prenatal glucocorticoid administration on the frequency of specific complications of prematurity among neonates with intrauterine growth restriction. STUDY DESIGN: We examined the association between intrauterine growth restriction and adverse neonatal outcomes in a population of 19,759 singleton very-low-birth-weight neonates without major birth defects. We included neonates from 25 to 30 weeks’ gestation entered in the Vermont Oxford Network database between 1991 and 1996 by 196 institutions. Intrauterine growth restriction was defined as the 10th percentile for birth weight according to the 1993 US national statistics. Odds ratios were estimated according to stepwise logis- tic regression for each neonatal outcome. Potential explanatory variables included gestational age, intrauter- ine growth restriction, race, prenatal care, prenatal glucocorticoid administration, route of delivery, fetal sex, and birth within versus postnatal transfer to a network institution. RESULTS: There was a statistically significant association of intrauterine growth restriction with neonatal death (odds ratio, 2.77; 95% confidence interval, 2.31-3.33), necrotizing enterocolitis (odds ratio, 1.27; 95% confidence interval, 1.05-1.53), and respiratory distress syndrome (odds ratio, 1.19; 95% confidence interval, 1.03-1.36). There was a trend (P < .10) toward association of intrauterine growth restriction with increased risks of intraventricular hemorrhage (odds ratio, 1.13; 95% confidence interval, 0.99-1.29) and severe intra- ventricular hemorrhage (grades III and IV; odds ratio, 1.25; 95% confidence interval, 0.98-1.59). Maternal prenatal glucocorticoid administration was associated with significantly lower risks of respiratory distress syndrome (odds ratio, 0.51; 95% confidence interval, 0.44-0.58), intraventricular hemorrhage (odds ratio, 0.67; 95% confidence interval, 0.61-0.73), severe intraventricular hemorrhage (odds ratio, 0.50; 95% confi- dence interval, 0.43-0.57), and death (odds ratio, 0.54; 95% confidence interval, 0.48-0.62). The benefits of prenatal glucocorticoid therapy for growth-restricted newborns were similar to those among normally grown infants. CONCLUSIONS: Intrauterine growth restriction within the range of 501 to 1500 g birth weight is associated with increased risks of neonatal death, necrotizing enterocolitis, and respiratory distress syndrome. Prenatal corticosteroid use was associated with decreased risks of all outcomes studied except necrotizing enterocoli- tis.We found no evidence that this benefit was dependent on fetal size. (Am J Obstet Gynecol 2000;182:198- 206.) Key words: Death, intrauterine growth restriction, intraventricular hemorrhage, necrotizing enterocolitis, prenatal glucocorticoids, respiratory distress syndrome, small for gestational age The influence of intrauterine growth restriction (IUGR) on the neonatal morbidity associated with pre- maturity has been disputed in the literature. Initial evi- dence suggested that IUGR was associated with reduc- tions in the incidences of respiratory distress syndrome (RDS) 1, 2 and intraventricular hemorrhage. 2, 3 These ob- servations supported a view of IUGR as an adaptive reac- tion to adverse intrauterine conditions that, through the initiation of a fetal stress response, led to improvements in gestational age–specific neonatal morbidity. 4, 5 Recent data have suggested that there is either no change 6-10 or a significant increase 11, 12 in the risk of RDS associated with IUGR and no difference in the incidence of intra- ventricular hemorrhage. 7 However, these studies failed to adjust for potential confounding variables that can alter the interpretation of the risk specific to IUGR. This study therefore examined the association of IUGR with neonatal morbidity and mortality while adjusting for From the Departments of Obstetrics and Gynecology, a Pediatrics, b and Medical Biostatistics, c University of Vermont College of Medicine, and the Department of Pediatrics, University of Toronto. d A complete list of participating investigators in the Vermont Oxford Network and their in- stitutional affiliations appears at the end of the article Received for publication April 23, 1999; revised June 10, 1999; ac- cepted August 3, 1999. Reprint requests: A. Lynn Stillman, Network Administrator, Vermont Oxford Network, 444 S Union St, Burlington, VT 05401. Copyright © 2000 by Mosby, Inc. 0002-9378/2000 $12.00 + 0 6/ 1/ 101977