Role of first‐trimester umbilical vein blood flow in predicting large‐for‐gestational age at birth

Abstract

To describe umbilical vein (UV) hemodynamics at 11 + 0 to 13 + 6 weeks of gestation in pregnancies delivering a large-for-gestational-age (LGA) neonate, and to build a multiparametric model, including pregnancy and ultrasound characteristics in the first trimester, that is able to predict LGA at birth. This was a matched case-control study, of singleton pregnancies that underwent ultrasound examination at 11 + 0 to 13 + 6 weeks for aneuploidy screening, at a single center over a 4-year period. Cases were women who delivered a neonate with birth weight (BW) > 90th centile for gestational age and sex, according to local birth-weight standards, while controls were those who delivered a neonate with BW ranging between the 10th and 90th centiles, matched for maternal and gestational age, at a ratio of 1:3. Each included case underwent Doppler assessment of the uterine arteries and UV, including measurement of its diameter, time-averaged maximum velocity (TAMXV) and UV blood flow (UVBF). UVBF and its components were expressed as Z-scores. Fisher's exact test and Mann-Whitney U-test were used to compare differences in maternal biomarkers and ultrasound characteristics between pregnancies complicated by LGA and controls. Logistic regression and receiver-operating-characteristics (ROC) curve analyses were carried out to identify independent predictors of LGA and to build a multiparametric prediction model integrating different maternal, pregnancy and ultrasound characteristics. Subgroup analysis was also performed, considering women who delivered a neonate with BW > 4000 g. In total, 964 pregnancies (241 with LGA at birth and 723 without) were included in the study. In LGA pregnancies compared with controls, UV-TAMXV Z-score (0.8 (interquartile range (IQR), 0.4-1.5) vs 0.0 (IQR, -0.3 to 0.5); P ≤ 0.001) and UVBF Z-score (1.3 (IQR, 0.8-1.9) vs 0.1 (IQR, -0.4 to 0.4); P ≤ 0.001) were higher, while there was no difference in median UV diameter Z-score (P = 0.56). Median uterine artery pulsatility index multiples of the median (MoM; 0.94 (IQR, 0.78-1.12) vs 1.02 (IQR, 0.84-1.19); P = 0.04) was significantly lower in LGA pregnancies. On multivariate logistic regression analysis, maternal body mass index (BMI; adjusted odds ratio (aOR), 1.2 (95% CI, 1.1-1.7); P < 0.001), parity (aOR, 1.4 (95% CI, 1.2-1.6); P < 0.001), pregnancy-associated plasma protein-A (PAPP-A) MoM (aOR, 1.1 (95% CI, 1.0-1.6); P = 0.04) and UVBF Z-score (aOR, 1.6 (95% CI, 1.1-1.9); P < 0.001) were associated independently with LGA. A multiparametric model integrating parity, BMI and PAPP-A MoM provided an area under the ROC curve (AUC) of 0.72 (95% CI, 0.67-0.76) for the prediction of LGA. The addition of UVBF Z-score to this model significantly improved the prediction of LGA provided by maternal and biochemical factors, with an AUC of 0.79 (95% CI, 0.75-0.83; P = 0.03). Similarly, the model incorporating UVBF Z-score predicted BW > 4000 g with an AUC of 0.83 (95% CI, 0.75-0.93). UVBF measured at the time of the 11-14-week scan is associated independently with, and is predictive of, LGA and BW > 4000 g. Adding measurement of UVBF to a multiparametric model that includes maternal (parity and BMI) and biochemical (PAPP-A) parameters improves the diagnostic accuracy of prenatal screening for LGA at birth. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.