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

Objectives: First, to evaluate and compare the performance of routine ultrasonographic estimated fetal weight (EFW) and fetal abdominal circumference (AC) at 31+0 - 33+6 and 35+0 - 36+6 weeks’ gestation in the prediction of large for gestational age (LGA) neonates born at ≥37 weeks’ gestation. Second, to assess the additive value of fetal growth velocity between 32 and 36 weeks’ gestation on the performance of EFW at 35+0 - 36+6 weeks’ gestation for prediction of LGA neonates. Third, to define the predictive performance for LGA neonates of different EFW cut-offs at routine ultrasound examination at 35+0 - 36+6 weeks’ gestation. Fourth, to propose a two-stage strategy for identifying pregnancies with LGA fetuses that may benefit from iatrogenic delivery during the 38th gestational week.

Methods: First, data from 21,989 singleton pregnancies that had undergone routine ultrasound examination at 31+0 - 33+6 weeks’ gestation and 45,847 that had undergone routine ultrasound examination at 35+0 - 36+6 weeks were used to compare the predictive performance of EFW and AC for LGA neonates with birthweight >90th and >97th percentiles born at ≥37 weeks’ gestation. Second, data from 14,497 singleton pregnancies that had undergone routine ultrasound examination at 35+0 - 36+6 weeks’ gestation and had a previous scan at 30+0 – 34+6 weeks were used to determine, through multivariable logistic regression analysis, whether addition of growth velocity, defined by a difference in EFW and AC Z-scores between the early and late third trimester scans divided by the time interval between them, improved the performance of EFW at 35+0 - 36+6 weeks in the prediction of delivery of LGA neonates born at ≥37 weeks’ gestation. Third, in the database of the 45,847 pregnancies that had undergone routine ultrasound examination at 35+0 - 36+6 weeks’ gestation the screen positive and detection rate of LGA neonates born at ≥37 weeks’ gestation and at ≤10 days from the initial scan were calculated for different EFW percentile cut-offs between the 50th and 90th percentile.

Results: First, the areas under the receiver operating characteristic curves (AUROC) of screening for LGA neonates were significantly higher with EFW Z-score than AC Z-score and at 35+0 - 36+6 than at 31+0 - 33+6 weeks’ gestation (p<0.001). Second, the performance of screening for LGA neonates achieved by EFW Z-score at 35+0 - 36+6 weeks was not significantly improved by addition of EFW or AC growth velocity. Third, in screening by EFW >90th percentile at 35+0 - 36+6 weeks’ gestation the predictive performance for LGA neonates born at ≥37 weeks’ gestation was modest (65% and 46% for neonates with birthweight >97th and >90th percentiles, respectively, at screen positive rate of 10%), but the performance was better for prediction of LGA neonates born at ≤10 days from the scan (84% and 71% for neonates with birthweight >97th and >90th percentiles, respectively, at screen positive rate of 11%). Fourth, screening by EFW >70th percentile at 35+0 - 36+6 weeks’ gestation predicted 91% and 82% of LGA neonates with birthweight >97th and >90th percentiles born at ≥37 weeks’ gestation, at screen positive rate of 32%, and the respective values of screening by EFW >85th percentile for prediction of LGA neonates born at ≤10 days from 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 >70th percentile should be used to identify pregnancies in need for another scan at 38 weeks and in the latter those with EFW >85th percentile should be considered for iatrogenic delivery during the 38th week.

Conclusions: First, the predictive performance for LGA neonates 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 birthweight >97th than >90th percentile and if delivery occurs within 10 days than at any stage after assessment. Second, prediction of LGA neonates by EFW >90th percentile is modest and the study presents a two-stage strategy for maximizing the prenatal prediction of LGA neonates.