Role of placental, fetal and cardiovascular markers in predicting adverse outcomes in women with suspected or confirmed pre-eclampsia

Abstract

STUDY OBJECTIVES: To assess the performance of placental, fetal and cardiovascular markers in the prediction of adverse outcomes in women with suspected or confirmed preeclampsia. METHODS: In this prospective prognostic accuracy study, women with suspected or confirmed preeclampsia underwent a series of investigations to measure maternal hemodynamic indices, mean arterial pressure (MAP), augmentation index (AIx), ophthalmic artery peak ratio (PR), uterine artery pulsatility index (UTA-PI), fetal biometry and Doppler studies, soluble fms-like tyrosine kinase-1 (sFLT-1), and placental growth factor (PlGF). The performance of these markers in isolation or in combination, to predict adverse perinatal outcomes and adverse maternal outcomes was then assessed using receiver-operating characteristics (AUROC) analysis. RESULTS: We included 126 women with suspected (n=31) or confirmed preeclampsia (n=95) with a median gestational age of recruitment of 33.9 weeks (interquartile range 30.9-36.3). Adverse perinatal outcomes were associated with a higher median UTA-PI (1.3 vs 0.8, p<0.001), ophthalmic artery PR (0.8 vs 0.7, p=0.01), and umbilical artery PI (82.0 vs 68.5 percentile, p<0.01), and lower median estimated fetal weight (4.0 vs 43.0 percentile, p<0.001), abdominal circumference (4.0 vs 63.0 percentile, p<0.001), middle cerebral artery PI (28.0 vs 58.5 percentile, p<0.001), and cerebroplacental ratio (18.0 vs 46.5 percentile, p<0.001). Pregnancies with adverse perinatal outcomes also had a higher median sFLT-1 (8208.0pg/mL vs 4508.0pg/mL, p<0.001), lower PlGF (27.2pg/mL vs 76.3pg/mL, p<0.001) and a higher sFLT-1/PlGF ratio (445.4 vs 74.4, p<0.001). The best performing individual marker for predicting adverse perinatal outcomes was the sFLT-1/PlGF ratio (AUROC 0.87, 95% CI 0.81-0.93) followed by estimated fetal weight (AUROC 0.81, 95% CI 0.73-0.89). Adverse maternal outcomes were associated with a higher median sFLT-1 (7471.0pg/mL vs 5131.0pg/mL, p<0.001), sFLT-1/PlGF ratio (204.3 vs 99.3, p<0.001) and a lower PlGF (37.0pg/mL vs 66.1pg/mL, p=0.01) and estimated fetal weight (16.5 vs 37.0 percentile, p=0.04). All markers performed poorly in predicting adverse maternal outcomes with sFLT-1 (AUROC 0.69, 95% CI 0.59-0.78) and sFLT-1/PlGF (AUROC 0.69, 95% CI 0.59-0.78) demonstrating best individual performance. The addition of cardiac, fetal, or other placental indices to sFLT-1/PlGF did not improve the prediction of adverse maternal or perinatal outcomes. CONCLUSION: The sFLT-1/PlGF ratio performs well in predicting adverse perinatal outcomes and poorly in predicting adverse maternal outcomes in women with suspected or diagnosed preeclampsia. The addition of cardiac, fetal and vascular indices to these markers is unlikely to improve prognostic performance of the sFLT-1/PlGF ratio. This article is protected by copyright. All rights reserved.