Detection of changes of hCG, progesterone and estradiol serum levels in threatened abortion in the first three months of gestation

The purpose of this study is finding the importance of hCG, estradiol, and progesterone serum level changes in prediction of spontaneous abortion. In a prospective study of 34 women with threatened abortion between 5 and 12 weeks gestation, serum levels of human chorionic gonadotropin (hCG), progesterone and estradiol (E2) were measured by radioimmunoassay. The same hormones were also determined in a control group of 44 women with normal intrauterine pregnancy of corresponding gestational age. HCG was determined consecutively in two serum samples obtained at 2- to 4- day intervals progesterone and E2 measurements were performed in the first serum sample. Pregnancy continued in the normal controls and in 18 patients with threatened abortion beyond 20 weeks gestation (viable pregnancies) and 16 patients aborted spontaneously before this time. In this investigation we have used forion radioimmunoassy kits of Amersham company (Kavoshyar agency in Iran) with an acceptable specificity and sensitivity in Reference Laboratory in Iran. There was no significant difference in the mean maternal ages or gravidity between these groups (p>0.05). The doubling time of hCG in abortions were lower than those in normal pregnancies and viable threatened abortions, although statistical analysis was not possible due to small numbers of abortions with rising hCG titers. The levels of all three hormones were significantly lower (p<0.05 –p<0.0001) in abortions than those in normal pregnancies and viable threatened abortions of the same gestational age except the hCG in the fifth and sixth gestational weeks. Regardless of the levels of two other hormones, the doubling time of hCG concentration for the second time on fifth and sixth weeks of gestation can differentiate between abortion and viability. However, after the 7th gestational week serial determinations of hCG generally did not add much information to that obtained from its single measurement. Statistical analysis revealed no significant difference between normal pregnancies and viable threatened abortions in the means of any of the biochemical markers (P> 0.05). The best discriminatory progesterone concentration that predicted abnormal pregnancies was 10 ng/mL. No pregnancies continued normally when progesterone was 10 ng/mL. An E2 level above 200 pg/mL indicated a normal intrauterine pregnancy and therefore ruled out an abnormal pregnancy with a specificity of 95% and a sensitivity of 87.5%. Progesterone with a positive predictive value of 100% was by far the best predictor of abortion and hCG with a negative predictive value of 98.4% was the best estimator of viable pregnancy, although their efficiency was the same (97.4%). Like hCG, estradiol was also a better predictor of viable pregnancy than abortion (a negative predictive value of 97.6% versus a positive predictive value of 82%) but with less efficiency (93.6%). If paired values of hCG+estradioal, hCG+ progesterone, or progesterone + estradiol were low, abortion followed in 94,100, and 100% of cases, respectively. The combination that comprised low levels of all three hormones was unavoidably followed by abortion.