Prediction of imminent preeclampsia at 35-37 weeks gestation

BACKGROUND: In the weeks preceding the clinical onset of preeclampsia, the maternal serum level of the angiogenic placental growth factor is decreased and that of the antiangiogenic factor soluble fms-like tyrosine kinase-1 is increased. Women presenting at specialist clinics with signs or symptoms of hypertensive disorders have been stratified according to concentrations of placental growth factor or the ratio of concentrations of soluble fms-like tyrosine kinase-1 and placental growth factor to determine clinical management for the subsequent 1-4 weeks. An alternative approach for the prediction of preeclampsia is use of the competing risks model, a Bayes' theorem based method, to derive patientspecific risk for preeclampsia by various combinations of maternal characteristics and medical history with multiples of the median values of biomarkers. OBJECTIVE: The purpose of this study was to compare the performance of screening for delivery with preeclampsia at <= 2 and <= 4 weeks after assessment at 35(+0)-36(+6) weeks gestation between the use of percentile cut-offs in placental growth factor alone or the soluble fms-like tyrosine kinase-1/placental growth factor ratio and the competing risks model. STUDY DESIGN: This was a prospective observational study in women who attended a routine hospital visit at 35(+0)-36(+6) weeks gestation in 2 maternity hospitals in England. The visits included the recording of maternal demographic characteristics and medical history and the measurement of serum placental growth factor and soluble fms-like tyrosine kinase-1 and mean arterial pressure. The areas under the receiver operating characteristics curves were used to compare the predictive performance for preeclampsia with delivery at <= 2 and <= 4 weeks from assessment of screening by placental growth factor alone and the soluble fms-like tyrosine kinase-1/placental growth factor ratio with that of a previously developed competing risks model with a combination of maternal factors, placental growth factor, soluble fms-like tyrosine kinase-1, and mean arterial pressure (triple test). RESULTS: First, the study population of 15,247 pregnancies included 326 pregnancies (2.1%) that subsequently experienced preeclampsia. Second, in the screening for delivery with preeclampsia at <= 2 and <= 4 weeks from assessment, the performance of the triple test was superior to that of placental growth factor alone or the soluble fms-like tyrosine kinase-1/placental growth factor ratio. The area under the receiver operating characteristics curves for preeclampsia at <= 2 weeks in screening by the triple test (0.975; 95% confidence interval, 0.964-0.985) was higher than that of placental growth factor alone (0.900; 95% confidence interval, 0.866-0.935; P<.0001) and the soluble fms-like tyrosine kinase-1/placental growth factor ratio (0.932; 95% confidence interval, 0.904-0.960; P=.0001). Similarly, the areas under the receiver operating characteristics curves for preeclampsia at <= 4 weeks in screening by the triple test (0.907; 95% confidence interval, 0.886-0.928) was higher than that of placental growth factor alone (0.827; 95% confidence interval, 0.800-0.854; P<.0001) or the soluble fms-like tyrosine kinase-1/placental growth factor ratio (0.857; 95% confidence interval, 0.830-0.883; P<.0001). Third, at most, screen-positive rates of 2-30% the detection rate of delivery with preeclampsia at <= 2 and <= 4 weeks that was achieved by the triple test was approximately 10% higher than that of the soluble fms-like tyrosine kinase-1/placental growth factor ratio and 20% higher than that of placental growth factor alone; the negative predictive value was similar for the 3 tests. CONCLUSION: At 35(+0)-36(+6) weeks gestation, the performance of screening for imminent delivery with preeclampsia by the competing risks model is superior to that of placental growth factor alone or the soluble fms-like tyrosine kinase-1/placental growth factor ratio.