Prediction of perinatal mortality in early-onset fetal growth restriction: A post hoc analysis of the Dutch STRIDER trial to predict perinatal mortality in early-onset fetal growth restriction

Abstract

Objective: Early-onset fetal growth restriction affects about 0.3% of pregnancies, posing high perinatal risks due to placental insufficiency. Early-onset fetal growth restriction often coincides with early-onset pre-eclampsia, associated with significant mortality and morbidity. Clinical management varies among clinicians, with emphasis on intensive monitoring and timely delivery. Our objective was to improve clinical prediction of perinatal mortality in early-onset fetal growth restriction for parental counseling.

Study design: This was a secondary analysis of prospective cohort data from the Dutch STRIDER trial. The study included 215 pregnant women diagnosed with severe early-onset fetal growth restriction between 20 + 0 and 29 + 6 weeks of gestation, from tertiary and secondary antenatal care centers in The Netherlands. Maternal and fetal characteristics were collected at inclusion, including sonographic and laboratory measurements. Analysis was performed using univariable and multivariable binary logistic regression to create a prediction model for perinatal mortality. The main outcome measures were fetal demise and neonatal mortality up to discharge.

Results: 215 Participants were included for this analysis. Perinatal mortality occurred in 84 (39 %) cases; 51 (24 %) were fetal and 33 (15 %) neonatal. Fetal abdominal circumference, gestational age at diagnosis, estimated fetal weight Multiple of Median, absent or reversed end-diastolic flow of the umbilical artery, umbilical artery pulsatility index Multiple of Median, non-Caucasian ethnicity, male sex, placental growth factor level and uterine artery pulsatility index were independent predictors of perinatal mortality. Randomization allocation (sildenafil or placebo) had no predictive value for mortality. The prediction model including gestational age at diagnosis, estimated fetal weight Multiple of Median and umbilical artery pulsatility index Multiple of Median showed an area under the receiver operating characteristic curve of 0.840 (P < 0.01). Placental growth factor was measured in a subset of patients and was an independent prognostic factor and performed significantly better within the predictive model, however it did not improve the predictive value of the model.

Conclusions: Prediction of perinatal mortality in early-onset fetal growth restriction is feasible with commonly available tests and measurements and could support decision making in management of pregnancy. However, implementation in practice requires further studies.

Keywords: Early-onset fetal growth restriction; Mortality; Perinatal mortality; Prediction; Predictive model.