Multiple Marker Second Trimester Serum Screening for Pre-eclampsia

Abstract

There are indications that circulating levels of both free β-human chorionic gonadotropin (β-hCG) and inhibin A are increased in pregnant women who later develop pre-eclampsia. Results with α-fetoprotein (AFP) are inconsistent, but one study suggested that women who later develop preeclampsia have lower levels of unconjugated estriol (uE 3 ). The goal of this study was to learn whether an appropriate combination of these markers might be an effective screening tool for preeclampsia in the second trimester. Nineteen women with preeclampsia (increased systolic and diastolic blood pressures [30 and 20 mm Hg, respectively]; proteinuria exceeding 100 mg/liter; and plasma uric acid of at least 357 μmol/liter) had estimates of the four putative markers in 32 serum samples. For each study sample, three control samples were collected from women with uneventful pregnancies, matched with the study patients for maternal age and gestational age. Levels of free β-hCG and inhibin A were significantly increased in the preeclamptic pregnancies, whereas concentrations of uE 3 were reduced. The differences between preeclamptic and unaffected pregnancies were most marked after 19 weeks' gestation. The median inhibin A value was 1.7 multiples of the median for unaffected pregnancies, and the median free β-hCG level was 2.1 multiples of the control median. The median uE 3 in preeclamptic pregnancies was 0.8 multiples of the median 2 weeks or more before the onset of proteinuria. There were no substantial differences in AFP. Combining inhibin A, free β-hCG, and uE 3 as a screening measure would detect an estimated 55% of affected pregnancies with 5% false-positive results. The screening performance of these markers can be estimated from the means, standard deviations, and correlations between the markers. This small-scale study suggests that inhibin A, free β-hCG, and uE 3 all are associated with the subsequent development of preeclampsia. Because levels of the markers are not highly correlated in unaffected pregnancies, each of them provides independent predictive information. It would be of interest to utilize screening in controlled trials of preventive measures in women at high risk of preeclampsia.