OC10.04: The role of angiogenic factors for first‐trimester screening of fetal growth restriction

Abstract

To assess if angiogenic factors could improve first-trimester screening for the prediction of fetal growth restriction (FGR). A nested case control study drawn from a prospective first trimester cohort (2007–2012). The outcome was the development of FGR (birthweight <10th centile with Doppler abnormalities). Logistic regression-based predictive models were developed for the prediction of early and late FGR, subdivided for abnormalities of umbilical artery. The model included the a priori risk (maternal characteristics), mean arterial pressure (MAP), uterine artery (UtA) Doppler (11–13 weeks), and the measurement of the angiogenic factors (8–11 weeks): placental growth factor (PlGF) and soluble Fms-like tyrosine kinase-1 (sFtl-1), normalized by logarithmic transformation. Of the 9,167 participants, 248 (2.7%) were diagnosed as FGR, from which 43 fetuses (0.5%) were early FGR, and 205 (2.3%) late FGR. Significant contributions for the prediction of early FGR were black ethnicity, chronic hypertension, previous FGR, smoking, together with MAP, UtA, PlGF and sFlt-1. A model achieved detection rates (DR) of 81% and 88% for 5% and 10% false positive rates (FPR), respectively (AUC: 0.96 [95%CI: 0.94–0.99]), and PlGF/sFlt-1 improved it by 26%. In those cases (49%) without pre-eclampsia (PE) DR was 81% at 10% of FPR. For late FGR, significant contributions were previous FGR, autoimmune disease, smoking together with MAP, UtA, PlGF and sFlt-1. The model achieved DR of 57% and 64% at 5% and 10% of FPR, respectively (AUC: 0.80 [95%CI: 0.75–0.84]), and PlGF/sFlt-1 improved it by 20%. In those cases (70%) without PE DR was 59% at 10% of FPR. Angiogenic factors are essential for the prediction of FGR. For early FGR, even if the association with PE, the prediction in the first trimester is attainable. For the prediction of late FGR, not influenced by PE and difficult to achieve so early, angiogenic factors improved it substantially.