Prediction of large-for-gestational-age neonates: screening by maternal factors and biomarkers in the three trimesters of pregnancy

Objective To develop a model based on maternal characteristics and medical history (maternal factors) for the prediction of delivery of large-for-gestational-age (LGA) neonates, and to examine the potential value of first-, second-and third-trimester fetal biometry and biomarkers in improving such a model. Methods This was a screening study in 76 300, 54 999, 25 727 and 6181 singleton pregnancies at 11-13, 19-24, 30-34 and 35-37weeks' gestation, respectively. The a-priori risk for LGA with birth weight >95th percentile (LGA >95th) was calculated using multivariable logistic regression analysis to determine which of the maternal factors had a significant contribution. Regression analysis was then used to determine whether screening by a combination of maternal factors, fetal biometry and various biophysical and biochemical markers had significant contribution in predicting delivery of LGA neonates. Results The likelihood of LGA >95th increased with increasing maternal weight and height and was lower in women of Afro-Caribbean and South Asian racial origins, in cigarette smokers and in nulliparous women. The risk was higher in women with pre-existing diabetes mellitus Type I and lower in those with chronic hypertension. In parous women, the risk increased with birth-weight Z-score in previous pregnancy and prior history of gestational diabetes and decreased with interpregnancy interval. Screening by maternal factors at 11-13weeks predicted 32%, 44% and 60% of LGA >95th at false-positive rates (FPRs) of 5%, 10% and 20%, respectively. With the addition of fetal biometry, the detection rates improved to 37%, 51% and 68% at 19-24weeks, 50%, 65% and 81% at 30-34weeks and 60%, 73% and 85% at 35-37 weeks at FPRs of 5%, 10% and 20%, respectively. The addition of biomarkers did not improve the detection rates achieved when screening by a combination of maternal factors and fetal biometry. Conclusion Combined screening by maternal factors and fetal biometry can predict a high proportion of pregnancies that will deliver LGA neonates. Copyright (C) 2015 ISUOG. Published by John Wiley & Sons Ltd.