Prediction of large‐for‐gestational‐age neonate by routine third‐trimester ultrasound

ABSTRACT Objectives First, to evaluate and compare the performance of routine ultrasonographic estimated fetal weight (EFW) and fetal abdominal circumference (AC) at 31 + 0 to 33 + 6 and 35 + 0 to 36 + 6 weeks' gestation in the prediction of a large‐for‐gestational‐age (LGA) neonate born at ≥ 37 weeks' gestation. Second, to assess the additive value of fetal growth velocity between 32 and 36 weeks' gestation to the performance of EFW at 35 + 0 to 36 + 6 weeks' gestation for prediction of a LGA neonate. Third, to define the predictive performance for a LGA neonate of different EFW cut‐offs on routine ultrasound examination at 35 + 0 to 36 + 6 weeks' gestation. Fourth, to propose a two‐stage strategy for identifying pregnancies with a LGA fetus that may benefit from iatrogenic delivery during the 38 th gestational week. Methods This was a retrospective study. First, data from 21 989 singleton pregnancies that had undergone routine ultrasound examination at 31 + 0 to 33 + 6 weeks' gestation and 45 847 that had undergone routine ultrasound examination at 35 + 0 to 36 + 6 weeks were used to compare the predictive performance of EFW and AC for a LGA neonate with birth weight > 90 th and > 97 th percentiles born at ≥ 37 weeks' gestation. Second, data from 14 497 singleton pregnancies that had undergone routine ultrasound examination at 35 + 0 to 36 + 6 weeks' gestation and had a previous scan at 30 + 0 to 34 + 6 weeks were used to determine, through multivariable logistic regression analysis, whether addition of growth velocity, defined as the difference in EFW Z ‐score or AC Z ‐score between the early and late third‐trimester scans divided by the time interval between the scans, improved the performance of EFW at 35 + 0 to 36 + 6 weeks in the prediction of delivery of a LGA neonate at ≥ 37 weeks' gestation. Third, in the database of the 45 847 pregnancies that had undergone routine ultrasound examination at 35 + 0 to 36 + 6 weeks' gestation, the screen‐positive and detection rates for a LGA neonate born at ≥ 37 weeks' gestation and ≤ 10 days after the initial scan were calculated for different EFW percentile cut‐offs between the 50 th and 90 th percentiles. Results First, the areas under the receiver–operating characteristics curves (AUC) of screening for a LGA neonate were significantly higher using EFW Z ‐score than AC Z ‐score and at 35 + 0 to 36 + 6 than at 31 + 0 to 33 + 6 weeks' gestation ( P < 0.001 for all). Second, the performance of screening for a LGA neonate achieved by EFW Z ‐score at 35 + 0 to 36 + 6 weeks was not significantly improved by addition of EFW growth velocity or AC growth velocity. Third, in screening by EFW > 90 th percentile at 35 + 0 to 36 + 6 weeks' gestation, the predictive performance for a LGA neonate born at ≥ 37 weeks' gestation was modest (65% and 46% for neonates with birth weight > 97 th and > 90 th percentiles, respectively, at a screen‐positive rate of 10%), but the performance was better for prediction of a LGA neonate born ≤ 10 days after the scan (84% and 71% for neonates with birth weight > 97 th and > 90 th percentiles, respectively, at a screen‐positive rate of 11%). Fourth, screening by EFW > 70 th percentile at 35 + 0 to 36 + 6 weeks' gestation predicted 91% and 82% of LGA neonates with birth weight > 97 th and > 90 th percentiles, respectively, born at ≥ 37 weeks' gestation, at a screen‐positive rate of 32%, and the respective values of screening by EFW > 85 th percentile for prediction of a LGA neonate born ≤ 10 days after the scan were 88%, 81% and 15%. On the basis of these results, it was proposed that routine fetal biometry at 36 weeks' gestation is a screening rather than diagnostic test for fetal macrosomia and that EFW > 70 th percentile should be used to identify pregnancies in need of another scan at 38 weeks, at which those with EFW > 85 th percentile should be considered for iatrogenic delivery during the 38 th week. Conclusions First, the predictive performance for a LGA neonate by routine ultrasonographic examination during the third trimester is higher if the scan is carried out at 36 than at 32 weeks, the method of screening is EFW than fetal AC, the outcome measure is birth weight > 97 th than > 90 th percentile and if delivery occurs within 10 days than at any stage after assessment. Second, prediction of a LGA neonate by EFW > 90 th percentile is modest and this study presents a two‐stage strategy for maximizing the prenatal prediction of a LGA neonate. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.