Plasma soluble fms-like tyrosine kinase 1 to placental growth factor ratio of 11.5 multiples of median predicts preeclampsia with severe features within 2 weeks of testing
Jimmy Espinoza,Vinicius F Calsavara,Sarah Kilpatrick,Sarosh Rana,Maged M Costantine,Kim Boggess,Blair J Wylie,Tiffany A Moore Simas,Judette M Louis,Stephanie L Gaw,Amy Murtha,Samantha Wiegand,Yvonne Gollin,Deepjot Singh,Robert M Silver,Danielle E Durie,Britta Panda,Errol R Norwitz,Irina Burd,Beth Plunkett,Rachel K Scott,Elizabeth Lemoine,Ravi Thadhani,S Ananth Karumanchi
DOI: https://doi.org/10.1016/j.ajog.2024.05.050
Abstract:Background: Angiogenic imbalances, characterized by an excess of antiangiogenic factors (soluble fms-like tyrosine kinase 1) and reduced angiogenic factors (vascular endothelial growth factor and placental growth factor), contribute to the mechanisms of disease in preeclampsia. The ratio of soluble fms-like tyrosine kinase 1 to placental growth factor has been used as a biomarker for preeclampsia, but the cutoff values may vary with gestational age and assay platform. Objective: This study aimed to compare multiples of the median of the maternal plasma soluble fms-like tyrosine kinase 1 to placental growth factor ratio, soluble fms-like tyrosine kinase 1, placental growth factor, and conventional clinical and laboratory values in their ability to predict preeclampsia with severe features. Study design: We conducted a cohort study across 18 United States centers involving hospitalized individuals with hypertension between 23 and 35 weeks' gestation. Receiver operating characteristic curve analyses of maternal plasma biomarkers, highest systolic or diastolic blood pressures, and laboratory values at enrollment were performed for the prediction of preeclampsia with severe features. The areas under the curve were compared, and quasi-Poisson regression models were fitted to estimate relative risks. The primary outcome was preeclampsia with severe features within 2 weeks of enrollment. Secondary outcomes were a composite of severe adverse maternal outcomes (elevated liver enzymes, low platelets count, placental abruption, eclampsia, disseminated intravascular coagulation, and pulmonary edema) and a composite of severe adverse perinatal outcomes (birth weight below the third percentile, very preterm birth [<32 weeks' gestation], and fetal or neonatal death). Results: Of the 543 individuals included in the study, preeclampsia with severe features within 2 weeks was observed in 33.1% (n=180) of them. A receiver operating characteristic curve-derived cutoff of 11.5 multiples of the median for the soluble fms-like tyrosine kinase 1 to placental growth factor plasma ratio provided good sensitivity (90.6%), specificity (76.9%), positive predictive value (66.0%), negative predictive value (94.3%), positive likelihood ratio (3.91), negative likelihood ratio (0.12), and accuracy (81.4%) for preeclampsia with severe features within 2 weeks. This cutoff was used to compare test positive cases (≥ cutoff) and test negative cases (< cutoff). Preeclampsia with severe features (66.0% vs 5.7%; P<.001) and composites of severe adverse maternal (8.11% vs 2.7%; P=.006) or perinatal (41.3% vs 10.14%; P=.001) outcomes within 2 weeks were more frequent in test positive cases than in test negative cases. A soluble fms-like tyrosine kinase 1 to placental growth factor plasma ratio ≥11.5 multiples of the median was independently associated with preeclampsia with severe features (adjusted incidence rate ratio, 9.08; 95% confidence interval, 6.11-14.06; P<.001) and a composite of severe adverse perinatal outcomes (adjusted incidence rate ratio, 9.42; 95% confidence interval, 6.36-14.53; P<.001) but not with a composite of severe adverse maternal outcomes (adjusted incidence rate ratio, 2.20; 95% confidence interval, 0.95-5.54; P=.08). The area under the curve for the soluble fms-like tyrosine kinase 1 to placental growth factor plasma ratio in multiples of the median (0.91; 95% confidence interval, 0.89-0.94) for preeclampsia with severe features within 2 weeks was significantly higher (P<.001 for all comparisons) than either plasma biomarker alone or any other parameter with the exception of absolute soluble fms-like tyrosine kinase 1 to placental growth factor plasma ratio values. Conclusion: A soluble fms-like tyrosine kinase 1 to placental growth factor plasma ratio ≥11.5 multiples of the mean among hospitalized patients with hypertension between 23 and 35 week's gestation predicts progression to preeclampsia with severe features and severe adverse perinatal outcomes within 2 weeks.