Time Dependent Clinical Predictive Model of Maternal Complications, Based on Placental Angiogenic Biomarkers: A Prospective Multicenter Study

Le résumé fourni par la source

ABSTRACT Objective To develop a prediction model for maternal adverse outcomes in high‐risk pregnant populations based on clinical history, biophysical parameters, and placental angiogenic biomarkers. Design Prospective, observational, and multicenter study. Setting Two university hospitals in Northern Italy. Population Women affected by hypertensive disorder of pregnangy (HDP) and/or foetal growth restriction (FGR). Methods Fms‐like tyrosine kinase‐1 (sFlt‐1) and placental growth factor (PlGF) were measured at recruitment. First, the whole cohort was stratified according to sFlt‐1/PlGF ratio risk categories. Second, the Cox Model was adapted with sFlt‐1/PlGF ratio integrated as a continuous variable together with demographic, clinical and biophysical risk factors to develop a prediction model for the maternal adverse outcomes (severe pre‐eclampsia, HELLP syndrome, abruptio placenta and liver haematoma). Results We recruited 335 consecutive singleton high‐risk pregnancies (109 with HDP, 95 with HDP‐FGR, and 131 with FGR). Seventy‐one of these developed severe‐pre‐eclampsia, 11 suffered from placental abruption and seven developed HELLP syndrome. At survival analysis, the probability to develop adverse outcomes was significantly different between patients with medium (38–85, 38–110) to very high (> 85/110) sFlt‐1/PlGF ratio compared to those with low sFlt‐1/PlGF ratio (< 38). The Cox multivariable clinical model was significantly associated with maternal complications: the area under the curve was 89.5% after 1 week, 93.8% after 2 weeks, 93.7% after 4 weeks and 96.1% after 8 weeks. Conclusions In a cohort of women with HDP and/or FGR, a prediction model, based on demographic, clinical, biophysical risk factors and sFlt‐1/PlGF ratio, can predict the risk of developing maternal complications from the time at diagnosis and up to 1, 2, 4, and 8 weeks. The inclusion of sFlt‐1/PlGF ratio, as a continuous variable, significantly improved the risk prediction.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.