Development of a fetal risk assessment score for the prediction of neonatal outcome in the growth-restricted fetus

Abstract

Objective: To develop and analyze a fetal risk assessment score (FRAS) that incorporates fetal arterial and venous blood flow studies (BFS), amniotic fluid volume, the non-stress test (NST) and an estimated fetal weight to improve the ability of antenatal testing to identify fetuses at risk for poor perinatal outcome and compare it to the Biophysical Profile (BPP). Study design: The Perinatal data base of the author’s institution was searched for all patients with singleton gestation with the diagnosis of intrauterine growth restriction, and who had both a biophysical profile (BPP) and fetal BFS (umbilical and middle cerebral artery, ductus venosus) within 4 days of delivery. Fetuses with major congenital abnormalities, chromosomal anomalies, or who delivered less than 25 weeks gestation were excluded. A FRAS score was developed by assigning numerical points for increasing abnormal arterial and venous BFS, and one point each for a non-reactive NST, oligohydramnios or if the fetus was small for gestational age. Recommendations for delivery were based on the clinical situation and the results of the Biophysical Profile (BPP); the FRAS score was not available to the attending physician. The FRAS was then compared to the BPP for the prediction of poor neonatal outcome (significant neonatal complications or prolonged hospital stay) using receiver operating characteristic (ROC) curve analysis and χ2 analysis. Results: Two hundred twenty-nine patients were included in the study. The results of the ROC analysis showed that the designed FRAS (area: 0.802) was slightly better than the BPP (area: 0.659) at predicting poor perinatal outcome in a group of growth-restricted fetuses. Conclusion: The study gives support to the hypothesis that combining biophysical tests with BFS will improve the identification of potential high-risk patients at increased risk for poor neonatal outcome, but prospective, randomized studies are needed to confirm this hypothesis.