Prediction of adverse outcomes for early-onset fetal growth restriction: a prospective study

Early-onset growth restriction (eoFGR) has a significantly increased risk of adverse outcomes. This study aimed to apply ultrasonic Doppler measurement and Fetal heart quantification (Fetal HQ) to predict the risk of adverse outcomes in eoFGR fetuses. In this prospective study of singleton pregnancies with eoFGR, fetuses were stratified into a group with adverse outcomes and those without. The primary outcome was the composite adverse outcome,defined as a composite of one or more of the following outcomes: perinatal death, Apgar score < 7 at 5 min, umbilical venous blood pH < 7.0, etc. Exclusion criteria included confirmed fetal structural or chromosomal anomalies, poor-quality ultrasound images, etc. All participants were performed both Doppler ultrasound and Fetal HQ assessment. Independent predictors of adverse outcomes in eoFGR fetuses were identified using multivariable logistic regression analysis. We examined 137 normal fetuses and 85 fetuses with eoFGR between 22 and 38 weeks’ gestation. Of these eoFGR cases, 43 were identified with adverse outcomes, and 42 were without. Compared to those without the composite adverse outcome, eoFGR fetuses with it exhibited significantly lower umbilical vein blood flow (UVF) and global sphericity index (GSI), decreased fractional shortening (FS) of right ventricular segments 11–17, and higher rates of umbilical artery (UA) abnormalities, ductus venosus (DV) and coronary artery (CA) dilation. The addition of UVF percentile, CA dilation, or GSI percentile significantly enhanced the AUC for predicting the composite adverse outcome compared to UA abnormalities alone, with the UA abnormalities + UVF percentile combination achieving the highest value. Although the combined model of UA abnormalities, CA dilation, UVF percentile, and GSI percentile achieved the highest AUC of 0.92 for prediction of the composite adverse outcome, it did not yield a statistically significant improvement over the model incorporating only UA abnormalities and UVF percentile (AUC = 0.85). Reduced UVF, GSI, Doppler abnormalities in the UA, DV, and CA dilation were associated with the composite adverse outcome in eoFGR fetuses. A model including UVF percentile and UA abnormalities significantly improved the predictive performance for composite adverse outcome compared to models using UA abnormalities alone or supplemented by CA dilation or GSI percentile.