The risks of spontaneous preterm delivery and perinatal mortality in relation to size at birth according to fetal versus neonatal growth standards

Abstract

Objective: The aim of this study was to test the null hypothesis that size at birth relative to fetal or neonatal growth standards is not a significant variable related to the risk of spontaneous preterm delivery.

Study design: This was a hospital-based cohort study of consecutive births at a tertiary care perinatal center from January 1, 1985, to December 31, 1996. A total of 37,377 pregnancies met the following inclusion criteria: (1) singleton gestation, (2) 25 to 40 weeks' gestation, and (3) no anomalies. Neonates were divided into 5 birth weight categories according to either fetal (uncorrected for sex) or neonatal (corrected for sex) growth standards, as follows: (1) intrauterine growth restriction, birth weight <3rd percentile; (2) borderline intrauterine growth restriction, birth weight > or = 3rd percentile and <10th percentile; (3) appropriate for gestational age, birth weight from 10th percentile through 90th percentile; (4) borderline large for gestational age, birth weight >90th percentile but < or = 97th percentile, and (5) large for gestational age, birth weight >97th percentile. Logistic regression analysis was used to estimate the independent effect of birth weight category on the risk of preterm delivery after spontaneous onset of labor, with the appropriate-for-gestational-age group serving as a reference.

Results: When fetal growth standards were applied, there was a significant increase in the risk of spontaneous preterm delivery when birth weight was outside the appropriate-for-gestational-age range (odds ratios of 2.5, 1.4, 1.2, and 1.9 for intrauterine growth restriction, borderline intrauterine growth restriction, borderline large-for-gestational age, and large-for-gestational-age groups, respectively). In contrast, when neonatal growth standards were applied, the risks of spontaneous preterm delivery in intrauterine growth restriction, borderline intrauterine growth restriction, and large-for-gestational-age groups were significantly lower (odds ratios of 0.5, 0.7, and 0.7 for intrauterine growth restriction, borderline intrauterine growth restriction, and large-for-gestational-age groups, respectively) because of an underestimation in the number of fetuses with abnormal size at birth delivered prematurely. With both fetal and neonatal growth standards there was a 5-to 6-fold greater risk of perinatal death for both preterm and term fetuses with intrauterine growth restriction.

Conclusion: Fetal growth standards are more appropriate in predicting the impact of birth weight category on the risk of spontaneous preterm delivery than are neonatal growth standards. When fetal standards are applied, the risks of preterm birth in both extreme abnormal birth weight categories (intrauterine growth restriction and large for gestational age) are 2- to 3-fold greater than the risk among appropriate-for-gestational-age infants.