Early pregnancy placental biomarkers as predictors of preeclampsia with severe features

Abstract Introduction Multiple studies have evaluated late second and third trimester values for soluble Fms‐like tyrosine kinase 1 (sFlt‐1), placental‐like growth factor (PlGF), and sFlt‐1:PlGF ratio for preeclampsia risk stratification, but few studies have assessed sFlt‐1:PlGF ratio earlier in pregnancy. Objective To investigate the utility of early pregnancy sFlt‐1:PlGF ratio for predicting preeclampsia with severe features, and to compare its predictive ability with that of the United States Preventive Services Task Force (USPSTF) risk assessment algorithm. Methods This was a secondary analysis of the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers‐to‐be (nuMoM2b). Of 10,038 participants, 1977 (enriched for adverse outcomes) had blood, which had been drawn at the first (6w0d–13w6d) and second (16w0d–21w6d) visits, assayed for sFlt‐1 and PlGF, allowing for the calculation of sFlt‐1:PlGF ratios from both visits as well as the change between visits. The primary outcome for this analysis was preeclampsia with severe features. The predictive ability of the sFlt‐1:PlGF ratio was compared to that of the USPSTF risk assessment algorithm using area under the receiver operating characteristic curves (AUROC). Youden indexes were used to identify optimal sFlt‐1:PlGF ratio cut‐points. The predictive characteristics of these cut‐points were compared to those of a positive USPSTF risk assessment result. Results Preeclampsia with severe features was diagnosed in 240 (12.1%) eligible participants, among whom the mean gestational age at delivery was 36w0d. The optimal sFlt‐1:PlGF ratio for predicting this outcome was 19.1 at visit 1, 6.2 at visit 2, and a decrease of 15.0 between visits. The sFlt‐1:PlGF ratio was inferior to the USPSTF risk assessment algorithm in predicting this diagnosis (AUROC: 0.51–0.60 vs. 0.60, p = 0.0014–0.048). At both visits and across visits, the sFLt‐1:PlGF ratio and USPSTF risk assessment algorithm predicted preeclampsia with severe features with comparable sensitivity (visit 1: 63.0%, visit 2: 49.8%, between visits: 50.0% vs. USPSTF: 61.5%) and specificity (visit 1: 43.1%, visit 2: 68.6%, between visits: 53.3% vs. USPSTF: 58.0%). Conclusions Early pregnancy sFLt‐1:PlGF ratio is a poor risk assessment tool for preeclampsia with severe features.