Co-Director of Maternal and Foetal Medicine, LAC-USC Women's and Children's Hospital, Director of Maternal Foetal Medicine at Good Samaritan Hospital, Medical Director, Pregnancy Help Center of San Gabriel Valley, Associate Professor, Division of Maternal Foetal Medicine, Department of Obstetrics and Gynaecology, University of Southern California, U.S.A.
There are a number of reasons for considering the association of human chorionic gonadotropin (hCG) with hyperemesis gravidarum. The temporal relationship between peak hCG levels and the most common time of nausea and vomiting has long been noted. More recently, the relationship of hCG to transient hyperthyroidism of hyperemesis gravidarum (THHG) has been described. Several experiments of nature have suggested that hCG plays a role. Let us consider these interrelated lines of evidence, beginning with the latter.
Nausea and vomiting of pregnancy (NVP) is more common in women with molar gestation, in which hCG is quite often markedly elevated. The consistent finding that NVP is more common with multiple gestation in which hCG is significantly elevated compared to singletons supports this observation as well. More recently, it has been noted that mothers carrying foetuses with Down's syndrome, a condition associated with elevated hCG, are more likely to have NVP. By way of contrast, in my experience, women with trisomy 18, in which hCG is very low, report much less nausea in pregnancy.
It has long been noted that peak concentrations of hCG and the most rapid rate of rise correspond to the time of maximal NVP (8-12 weeks' gestation). Several investigators have studied hCG concentrations in women with hyperemesis but the results are not consistent. Our group found that hCG was elevated in women with vomiting compared to gestational age-matched controls, but there was significant overlap (Figure 1). If hCG is the cause of hyperemesis, why do many women with high levels have little vomiting and vice versa?
Two other related observations suggest an answer to this. The rise in hCG corresponds temporally to a consistent thyroid stimulation in the mother (Figure 2). Since hCG closely resembles its sister glycoprotein hormone, thyroid stimulating hormone (TSH), it has been conjectured that hCG is the thyroid stimulator of pregnancy. Of greater interest to the question of the cause of NVP, however, is the observation that the severity of NVP correlates closely with the degree of biochemical hyperthyroidism.
During an 11-month period we studied 57 controls (30 with no vomiting and 27 with mild nausea and occasional vomiting) and 57 patients with hyperemesis gravidarum (38 with hyperemesis and 19 with severe hyperemesis).1 Hyperemesis gravidarum was defined as persistent vomiting that resulted in greater than 5% weight loss and large ketonuria on dip urine examination. All patients were seen at less than 16 weeks' gestation and had onset of vomiting in the first trimester. Exclusions were history of thyroid disease, goiter or presence of thyroid antibodies, fever or evidence of other conditions responsible for vomiting, and multiple gestation. Study and control subjects were enrolled contemporaneously and did not differ with respect to age, parity, or gestational age.
Human chorionic gonadotropin was higher among hyperemesis subjects compared to controls (97 ± 8 IU/mL, 29 ± 2 IU/mL, p < 0.001), but there was a weak correlation between hCG and Free T4 and TSH. Most interestingly, there was a strong association between the severity of NVP and the degree of hyperthyroidism (Figure 3). This finding has been shown by others as well.2 This strongly suggests that whatever the cause of the thyroid stimulation in pregnancy, it is very closely linked to the cause of hyperemesis. It would not appear to be the hyperthyroidism per se however, since hyperthyroidism itself rarely causes nausea and vomiting.
If hCG is the thyroid stimulator of pregnancy, how can there be such a weak association between hCG and the degree of hyperthyroidism? The answer to this question may lie in the action of variants of hCG of differing thyrotropic potential. The principle difference between hCG and other glycoprotein hormones (luteinizing hormone [LH], follicle stimulating hormone [FSH], TSH) is the large 31 amino acid tail, the ß-carboxyterminal portion (ß-CTP) (Figure 4). The ß-CTP has four sites for glycosylation. Variants of hCG that lack the ß-CTP do not bind sialic acid moieties as readily as intact hCG. Whereas LH is ten times more potent than intact hCG at stimulating the TSH receptor, mutant hCG lacking the ß-CTP is as potent as LH. This variant form of hCG, having greater thyrotropic potential, may also be found in otherwise normal pregnancies characterized by hyperemesis.3
What is commonly referred to as immuno-reactive hCG is a mixture of free ß-hCG, free a-hCG, deglycosylated hCG, asialo-hCG, and hCG completely lacking the CTP. Molar tissue, in which only 3% of the measured hCG is fully glycosylated and a majority of the hCG is not intact, is an ideal source for studying variants of hCG. We collaborated with Dr. Jerome Hershman who extracted large amounts of hCG from these tissues and separated them by pH. Basic fractions of hCG were much more potent than intact hCG in stimulating the release of cyclic adenosine monophosphate (cAMP) when they combined with human TSH and LH receptors. These basic fractions correspond to the deglycosylated hCG lacking the ß-CTP.4
Other mutations of hCG cause it to be unable to dimerize, i.e., for a and ß subunits to join, which they must for normal activity. Such mutations may manifest as elevated free ß-hCG. This is seen in molar gestations. We found that, compared to 23 gestational age-matched controls, 39 patients with hyperemesis had elevated free ß-hCG (101 ± 70 ng/mL, 31 ± 31 ng/mL, p < 0.001) (Figure 5).5 There was also a significant difference between groups for total hCG (9,237 ± 3,613 ng/mL, 5,543 ± 2,290 ng/mL, p < 0.01), but not for free a-hCG (399 ± 231 ng/mL, 377 ± 214 ng/mL). Thus, part of the puzzle of the relationship between hCG and NVP may lie in the fact that measurements of intact hCG do not reflect various fractions of differing potency.
Another important clue to the hormonal etiology of NVP may lie in variation within the hCG receptor. Mutations have been described in the glycoprotein hormone receptors just as for the hormones themselves. Some of these mutations, so-called gain of function mutations, offer another explanation for the variation between hCG and thyroid stimulation, and hCG and hyperemesis itself. A recently published case establishes just such an association.6 A 27-year-old woman experienced hyperemesis throughout pregnancy associated with hyperthyroidism. Both resolved after delivery. This patient was found to have a familial gain-of-function mutation in the TSH receptor making her extremely sensitive to the circulating hCG in pregnancy. This case may be related to those previously described in which there is an association of hyperemesis, hyperthyroidism and virilization. Such cases could be explained by a mutation in the glycoprotein hormone receptor in an area common to the TSH and LH/hCG receptor or by variants of hCG that have greater potency in stimulating not only the TSH receptor but also the LH/hCG receptor. They may also provide a further clue to the puzzle that still remains. How does hCG cause hyperemesis? Despite the evidence that hCG is very closely related to the cause of hyperemesis, there still lacks a good explanation for how hCG can cause hyperemesis. The effect of hCG on steroid hormone synthesis is well-known, as is the fact that steroid hormones modulate the response to nausea. It is likely that the hyperthyroidism seen with hyperemesis is only a marker for increased hCG interaction with related glycoprotein hormone receptors with the most significant interaction occurring within the ovary and affecting steroid metabolism.
Along these lines, increased levels of oestradiol have been found by some investigators including our own group, but not by all. It is possible that oestradiol or testosterone, or the rate of change in these two hormones, is responsible for the nausea and vomiting of pregnancy and some of the related mood changes that have been described. More work is needed in this area.
RELATED PUBLICATIONS BY THE SPEAKER:
Goodwin TM. Hyperemesis gravidarum. Clin Obstet Gynecol 1998;41:597-605.
Goodwin TM, Hershman JM. Hyperthyroidism due to inappropriate production of human chorionic gonadotropin. Clin Obstet Gynecol 1997;40:32-44.
Pekary AE, Jackson IM, Goodwin TM, Pang XP, Hein MD, Hershman JM. Increased in vitro thyrotropic activity of partially sialated human chorionic gonadotropin extracted from hydatidiform moles of patients with hyperthyroidism. J Clin Endocrinol Metab 1993;76:70-4.
Goodwin TM, Montoro M, Mestman JH. Transient hyperthyroidism and hyperemesis gravidarum: clinical aspects. Am J Obstet Gynecol 1992;167:648-52.
Back to home