The value of elevated second-trimester beta-human chorionic gonadotropin in predicting development of preeclampsia

Our purpose was to investigate the association of an elevated second-trimester serum beta-human chorionic gonadotropin concentration with the subsequent development of hypertension in pregnancy and to evaluate its utility as a screening test for preeclampsia.

We examined 6286 nondiabetic women with singleton pregnancies who, as part of triple-screen testing, had a serum beta human chorionic gonadotropin level drawn between 15 and 22 weeks' gestation between November 1, 1991, and November 30, 1994. Medical records of women with hypertension (n = 675) were reviewed, patients with chronic hypertension were excluded, and the remainder were classified as having gestational hypertension (n = 333), mild preeclampsia (n = 110), or severe preeclampsia (n = 84). The beta-human chorionic gonadotropin level expressed as multiples of the median adjusted for maternal weight and gestational age was compared between normotensive and hypertensive complicated pregnancies.

In the overall population beta-human chorionic gonadotropin levels > or = 2.0 multiples of the median during the second trimester were significantly associated with development of hypertension in pregnancy. The rate ratio for development of overall hypertension was 1.6 (95% confidence interval 1.3 to 2.0) and for preeclampsia 1.8 (95% confidence interval 1.3 to 2.6). When stratified by parity, a statistically significant association remained only among multiparous women, for overall hypertension (rate ratio 2.2, 95% confidence interval 1.6 to 3.2) and for preeclampsia (rate ratio 3.4, 95% confidence interval 2.1 to 5.6). Adjusting for confounding factors did not alter the results. In the overall population, with the use of 2.0 multiples of the median of beta-human chorionic gonadotropin as a cutoff value, the sensitivity of beta-human chorionic gonadotropin as a screen for development of hypertension was 15.6%, the specificity was 90.0%, and the positive predictive value was 12.8%.

Overall, second-trimester serum beta-human chorionic gonadotropin levels were elevated among women who had hypertension during pregnancy. In our population this association was statistically significant only among multiparous women. The utility of an elevated second-trimester beta-human chorionic gonadotropin level as a screening test for preeclampsia is limited.