The Prediction of Intrapartum Fetal Compromise According to the Expected Fetal Weight

Objectives: To assess the predictive accuracy of the expected fetal weight in the third trimester (ExFW3t), based on the estimated fetal weight (EFW) at mid-trimester ultrasound scan, for the prediction of intrapartum fetal compromise (IFC) (an abnormal intrapartum fetal heart rate or intrapartum fetal scalp pH requiring urgent cesarean section). Methods: This retrospective study included 777 singleton pregnancies that underwent a 20-week study and a 3t scan. The extrapolated EFW at 20 weeks to the 3t or ExFW3t was considered a proxy of the potential growth. The percentage difference with the actual 3t EFW (%ExFW3t) was compared with other ultrasonographic and clinical parameters-EFW centile (EFWc), middle cerebral artery pulsatility index (MCA PI) in multiples of the median (MoM), umbilical artery (UA) PI MoM, cerebroplacental ratio (CPR) MoM, and maternal height-for the prediction of IFC by means of the area under the curve (AUC) and Akaike Information Criteria (AIC). Results: Pregnancies with IFC presented higher values of UA PI MoM (1.19 vs. 1.09, p = 0.0460) and lower values of population and Intergrowth EFWc (45.9 vs. 28.9, p < 0.0001, 48.4 vs. 33.6, p = 0.0004), MCA PI MoM (0.97 vs. 0.81, p < 0.0001), CPR MoM (1.01 vs. 0.79, p < 0.0001), %ExFW3t (89.9% vs. 97.5%, p = 0.0003), and maternal height (160.2 vs. 162.9, p = 0.0083). Univariable analysis selected maternal height, EFWc, %ExFW3t, and UA PI MoM as significant parameters. However, %ExFW3t did not surpass the prediction ability of cerebral Doppler. Finally, multivariable analysis showed that the best models for the prediction of IFC resulted from the combination of cerebral Doppler (MCA PI MoM or CPR MoM), fetal weight (%ExFW3t or EFWc), and maternal height (AUC 0.75/0.76, AIC 345, p < 0.0001). Conclusions: Fetal weight-related parameters, including %ExFW3t, a proxy of the proportion of potential growth achieved in the 3t, were less effective than fetal cerebral Doppler for the prediction of IFC. The best performance was achieved by combining hemodynamic, ponderal, and clinical data.