Serum Concentrations of Human Chorionic Gonadotrophs and Immunoreactive Inhibin in Early Pregnancy and Recurrent Miscarriage: A Longitudinal Study

EDITORIAL COMMENT: We accepted this paper for publication because we thought readers would be interested to read the results of the careful investigations of this series of patients with previous miscarriages. However the reviewers thought it pertinent to remind readers that it is not generally accepted that chorionic gonadotrophin hormone is a useful test for fetal welfare in early or late pregnancy. Moreover, there is a distressing tendency in recent publications to ignore classical papers of the past or indeed any paper that was published before the mid 1980s and hence not readily available on a quick computer search. It seems appropriate to state that previous papers on such topics as hormone levels in women with threatened abortion have shown that measurements of oestrogen and progesterone levels in blood and urine provided important prognostic information. In one study of 55 such patients oestrogen excretion was below the first percentile for the period of gestation in 22 patients all of whom aborted, usually within a week of the measurement. Twenty two of the 33 patients with oestrogen values in the normal pregnancy range had pregnancies which continued and resulted in surviving infants. Eleven patients with normal oestrogen values subsequently aborted. These abortions occurred more than 1 week after the original assessment and usually in the second trimester. Pregnanediol measurements provided comparable information (A). In many centres urinary or serum oestrogen and pregnanediol levels continue to be assessed in patients with a bad past reproductive history or in cases with threatened miscarriage although ultrasonography has largely supplanted the use of these biochemical markers in the latter case. Moreover the reviewer has reread the articles referred to in this paper and the results presented do not, in our opinion, disagree with the above statements. (A) Brown JB et al. Hormone levels in threatened abortion. J Obstet Gynaecol Br Commonw 1970; 77: 690–700. Authors' Response to Editorial Comment We have reported a longitudinal study of hormonal changes in early pregnancy and miscarriage in which measurements were commenced before there was any clinical evidence of potential outcome of the pregnancies. In contrast, in the study of Professor Brown referred to above and in most other early similar studies, assays of hormones began at the time the mother presented with clinical symptoms of threatened miscarriage. While ultrasound remains the main diagnostic means of monitoring the progress of early pregnancy, in our experience the serum level of HCG is a useful complementary biochemical test. We have found serum progesterone to be of minimum prognostic value but have not evaluated urinary pregnanediol or urinary oestrogen. Summary: Serum concentrations of immunoreactive inhibin (ir-inhibin) and human chorionic gonadotrophin (HCG) have been measured during the first trimester in a longitudinal study of pregnant women attending a recurrent miscarriage clinic. In 30 singleton pregnancies (Group 1) that continued successfully to term, the median concentration of ir-inhibin initially declined from 1,140 pg/mLat week 4–5 then rose back to comparable values between weeks 7 and 10 but to decline again to reach the significantly lower level of 840 pg/mL (p<0.01) at week 15–16. Serum levels of HCG showed the classical profile of normal pregnancy reaching a median peak value of 65,600 IU/L(1st IRP) at week 8–9. In 7 pregnancies that miscarried but earlier had evidence on ultrasound of an active fetal heart, HCG levels in the first 9 weeks were consistently below the 10th percentile for Group 1 pregnancies (p<0.001). Levels of irinhibin were also suppressed but to a lesser extent. In 6 of 7 afetal pregnancies, HCG levels during the first 9 weeks were again markedly subnormal. The levels of irinhibin varied between high normal and subnormal. In none of the pregnancy groups was a correlation found between irinhibin and HCG concentrations. In a single pregnancy with an anencephalic fetus, while levels of irinhibin and HCG were not depressed, peak values were not reached until week 12. The study shows that the level of irinhibin in the maternal serum in early pregnancy is of little value as a prognostic indicator of pregnancy outcome. It confirms that a subnormal HCG level is a useful predictor of early pregnancy failure.