Clinical Endocrinology (1984) 20,695-700

MATERNAL HYPERTHYROIDISM AND CONGENITAL MALFORMATION IN THE OFFSPRING

NAOKO MOMOTANI, KUNIHIKO ITO, NOBORU HAMADA, YOSHIO BAN, YOSHIHIKO NISHIKAWA AND TAKASHI MIMURA

It0 Thyroid Clinic and Hospital, Tokyo, Japan

(Received 24 October 1983; revised 4 January 1984; accepted I January 1984)

SUMMARY

Six hundred and forty-three neonates from mothers with Graves’ disease were examined for major malformations of external organs to compare the influence of maternal hyperthyroidism vs. ingestion of methimazole (MMI) during the first trimester on the incidence of congenital malformations. The subjects were divided into four groups according to maternal therapy and thyroid status during the first trimester as follows: (1) infants whose mothers did not receive MMI and were hyperthyroid (Group I), (2) infants whose mothers did not receive MMI and were euthyroid (Group 2), (3) infants whose mothers received MMI and were hyperthyroid (Group 3) and (4) infants whose mothers received MMI and were euthyroid (Group 4). The prevalence of malformed infants in these four groups was 6.0% (three of 50), 0.3% (one of 350), 1.7% (two of 117) and 0.0% (none of 126), respectively. The incidence in Group 1 was significantly higher than that in Group 2 (P<O-OI). There was no discernible dose dependency of MMI on the occurrence of malformations. These findings suggest that maternal uncontrolled hyperthyroidism may cause congenital malformations and that the beneficial role of MMI treatment outweighs its teratogenic effect, if any.

Maternal hyperthyroidism has been suspected as a potential cause of malformations in the offspring. There have been several reports of malformed infants born to thyrotoxic mothers (Wallace, 1933, 1940; Whitelaw, 1947; Hepner, 1952; Holt, 19S9; Hawe & Francis, 1962; Orlova, 1966; Talbert et al., 1970). Whether the mothers ingested antithyroid drugs was not stated in these reports; however, methimazole (MMI) therapy has been reported to be associated with congenital scalp defects in five offspring of three treated mothers (Milham & Elledge, 1972; Mujtaba & Burrow, 1975). Because of the essentially anecdotal nature of these case reports, one cannot be certain of the true risk for malformation involved in uncontrolled thyrotoxicosis and antithyroid drug therapy.

Having a unique opportunity to examine a large number of pregnant patients with Graves’ disease in our hospital, we attempted to determine the relative risk of neonatal

Correspondence: Naoko Momotani, Ito Hospital, 4-3-6, Jingumae, Shibuya-ku, Tokyo 150, Japan.

695

696 Naoko Momotani et al.

malformation in this disease when the mother is hyperthyroid or receiving antithyroid drug therapy during the sensitive period of organogenesis.

MATERIALS A N D METHODS

Six hundred and forty-three infants were born alive between August 1965 and May 1980 to mothers who were seen at Ito Thyroid Clinic and Hospital and diagnosed as having Graves’ disease on the basis of clinical examination and laboratory data. All were delivered at their own hospitals by their own practitioners, and all of the births were attended by obstetricians. Since the prevalence of minor malformations as well as malformations of internal organs can be influenced by the definition and diagnostic ability, we considered in this study only the major structural malformations of external organs which are readily detectable by surface examination, including the oral cavity. At their first visit after delivery, a doctor interviewed the mothers about the congenital malformations diagnosed by the obstetricians. Throughout this study the same single doctor interviewed all the patients in the same fashion. Afterwards the doctor conducted examinations of the malformed infants.

The mothers were treated with MMI or underwent subtotal thyroidectomy before and/or during pregnancy. Thyroid function was assessed in most cases by determination of the Free T4 index; in some cases the total T4, T3 and/or PBI concentrations was measured, due allowance being made for the influence of pregnancy on the normal range of values.

The subjects were classified into four groups on the basis of maternal therapy and thyroid status during the first trimester, since this period is recognized as a critical period for organogenesis: Group 1, infants whose mothers were not treated with MMI and were hyperthyroid; Group 2, infants whose mothers were not treated with MMI and were euthyroid; Group 3, infants whose mothers were treated with MMI and were hyperthyroid; Group 4, infants whose mothers were treated with MMI and were euthyroid. Neither infants born to mothers who received radiation or drugs other than MMI nor multiple births were included. Infants born to diabetic mothers were excluded.

Statistical significance was assessed by Fisher’s exact test.

RESULTS

Table 1 lists the six infants with major malformations of external organs found in the series. These comprised malformations of the ear-lobe, omphalocele, imperforate anus, anencephaly, harelip and polydactyly-one case each. All of these except the omphalocele are common malformations among Japanese newborns. There was no consistent pattern of infant malformation prevalence and maternal age; one of the mothers was 33 years of age, while the others were all in their twenties at the time of delivery.

Relationship of prevalence of infant malformation to maternal thyroid function and therapy

The number and prevalence of malformed infants in each study group are summarized in Table 2. The overatl frequency of malformed infants born to hyperthyroid mothers was 3.0% (five of 167), whereas the corresponding rate of malformed infants born to euthyroid mothers was 0.2% (one of 476). The prevalence in the former was significantly higher than

Hyperthyroidism and malformation

Table 1 . Details of six neonates with major malformations of external organs in the present series

697

Maternal age

Patient at delivery No. Type of malformation (years)

1 Malformation of the ear-lobe 28 2 Omphalocele 25 3 Imperforate anus 29 4 Anencephaly 29

6 Polydactyly 29 5 Harelip 33

Maternal thyroid status

during the first trimester

Hyperthyroid Hyperthyroid Hyperthyroid Hyperthyroid Hyperthyroid

Euthyroid

Maternal medication during the

first trimester

Methimazole Methimazole

-

- -

-

Table 2. Number and prevalence of malformed neonates in relation to maternal methimazole therapy and thyroid status during the first trimester

Hyperthyroid Euthyroid Total

Malformed Malformed Malformed Total Total Total

Neonates No. Incidence Neonates No. Incidence Neonates No. Incidence

Methimazole 117 2 1.7% 126 0 0.0% 243 2 0.8% None 50 3 6.0* 350 1 0.3 400 4 1.0 Total 167 5 3.0* 476 1 0.2 643 6 0.9

* P < 0.01, hyperthyroid vs. euthyroid subjects.

that in the latter by Fisher’s exact test (P < 0.01). On the other hand, the overall prevalence rates for malformed infants in MMI-treated and in non-treated mothers were 0.8% (two of 243) and 1.0% (four of 400), respectively. These values are statistically similar. It was worthy of note that malformed infants in the treated group were both in the hyperthyroid group. Furthermore, three of four malformed infants in the non-treated group were also in the hyperthyroid group. In the non-treated hyperthyroid group, three of 50 infants were malformed (i.e. 6.0%), whereas in the non-treated euthyroid group, only one of 350 infants was malformed (i.e. 0.3%). Fisher’s exact test revealed a significant difference in the incidences between these two groups (P < 0.01).

Dose eflect of MMI on the occurrence of malformations

The total doses of MMI administered to each mother during the first trimester are plotted in Fig. 1 . The highest dose was 1680 mg and the lowest dose was 10 mg. The doses administered to mothers who delivered malformed infants were 160 mg and 240 mg, respectively. Table 3 shows the highest dose administered daily to each mother during the first trimester. Of the two malformed infants in the MMI-treated group, one was born to a

698 Naoko Mornotani et al.

1 SO(

n Q, loo( E

0 Q,

0 Q

3 75t

Fig. I . The total doses of methimazole administered to 243 mothers during the first trimester. (0)

Not malformed; (0) malformed. The doses administered to mothers who delivered malformed infants (solid circles; 160 mg and 240 mg, respectively) were rather lower than the mean.

Table 3. Frequency of mal- formed neonates in relation to the highest daily doses of

methimazoie

Number Dose

(mgiday) Total malformed ~~

230 31 0 20 31 1 15 25 0 10 62 0 5 94 1

mother treated with 20 mg daily and the other to a mother treated with 5 mg daily. No malformed infants were born to mothers who had received 30 mg or more of MMI daily.

Period at which MMI was administered 11s. the type of malformations

Figure 2 shows the period at which MMI was administered to mothers who delivered

Hyperthyroidism and malformation 699

5 mg i" -- Om pho loce I e

1 8 h c a o I I t I , , ,

0 I 2 3 4 5 6 7 8 9 10 I I 12 Weeks of gestation

Fig. 2. Dose of methimazole administered and the period vs. the type of malformation. Top, malformation of the ear-lobe; bottom, omphalocele. In the case with malformation of the ear-lobe, daily administration of 20mgof MMI was begun at the end of the 10th week of gestation when the ear-lobe would be expected to have already formed. In the other case with omphalocele, methimazole was administered throughout most of the first trimester, but the dose was only 5 or 10 mg daily.

malformed infants. In one case with malformation of the ear-lobe, daily administration of 20 mg of MMI was started at the end of the 10th week of pregnancy when the ear-lobe is known to have been formed already. In the other case with omphalocele, MMI was administered throughout most of the first trimester at a dose of 5 or 10 mg daily.

DISCUSSION

In the present study the infants born to mothers who were hyperthyroid during the sensitive period of organogenesis revealed a significantly higher incidence of malforma- tions of external organs than those whose mothers were euthyroid during that period. This suggests that a maternal hyperthyroid state or some other factor related to this condition may cause congenital malformations. The prevalence of malformed infants in the hyperthyroid group was significantly higher by Poisson probability test (P < 0.09, than in the the general population of Japan, reported by Japan Association of Maternal Welfare (JAMW). Since 1972, JAMW has been conducting annual surveys in malforma- tions of external organs by asking maternity clinics or hospitals to check the malformations found in the newborns according to the check lists. Their statistical report in 1980 at 238 institutions in Japan shows that 1400 newborns were externally malformed among 158 145 births (i.e. 0.89%). Comparison of the prevalence in the present study with that reported by JAMW may be handicapped by the fact that the two data bases were not assembled in the same fashion. However, prevalence figures probably will tend to be lower when based on information from mothers than when based on check lists recorded by doctors attending the births. Moreover in our study, where we considered only major defects, we excluded stillbirths which are known to have a much higher prevalence of congenital malformations than live births. Stillbirths as well as both major and minor defects were included in the survey of JAMW.

In the present study, we did not find a statistically significant correlation between

700 Naoko Momotani et al.

frequencies of malformations and the levels and duration of hyperthyroidism, presum- ably due to the relatively small absolute number of malformed infants involved (only six).

The types of malformations were not specific. Furthermore, only a small number of thyrotoxic mothers delivered malformed babies and mothers whose thyroid hormone levels were high did not always deliver malformed infants. These facts suggest that there are individual differences in susceptibility and that maternal hyperthyroidism is only one of several factors involved in the aetiology of malformations. The hyperthyroid state may act as a trigger for abnormal genetic factors.

We did not observe the congenital scalp defects previously reported in the newborns of mothers who had received MMI during pregnancy (Milham & Elledge, 1972; Mujtaba & Burrow, 1975). In these earlier reports of scalp defects, maternal thyroid status was not specified. In our study, the mothers who had been treated with MMI during the first trimester delivered malformed infants even less often than those who had received no medication; morover, there was no correlation between the MMI dose and the prevalence of malformed infants. Thus, it seems reasonable to conclude that MMI may protect against congenital malformations rather than exert a teratogenic effect in man; this protective effect may be mediated via control of maternal thyrotoxicosis.

Our results suggest that thyrotoxic patients who expect to become pregnant should be euthyroid before pregnancy and that they should be treated with MMI in sufficient dose to control the hyperthyroid state during the first trimester.

ACKNOWLEDGEMENT

We are indebted to Dr Yoshio Sumiyoshi for helpful advice concerning definition and incidence for malformations of external organs. Portions of this manuscript were presented at the 11th Annual Meeting of the European Thyroid Association in September, 1981 and published in abstract form (66 A, 1981).

REFERENCES

HAWE, P. & FRANCIS, H.H. (1962) Pregnancy and thyrotoxicosis. British Medical Journal, 2, 817-822. HEPNER, W.R. (1952) Thiourea derivatives and the fetus. American Journal of Obstetrics and Gynecology, 63,

HOLT. J.H. (1959) Hyperthyroidism in pregnancy. The Journal of the Kansas Medical Society, 60, 537-539. MILHAM, S. & ELLEDGE, W. (1972) Maternal methimazole and congenital defects in children. Teratology, 5, 125. MUJTABA, Q. & BURROW, G.N. (1975) Treatment of hyperthyroidism in pregnancy with propylthiouracil and

ORLOVA, 0.1. (1966) The health and development of children born of mothers with thyrotoxicosis. Excerpta

TALBERT, L.M., THOMAS, C.G., HOLT. W.A. & RANKIN, P. (1970) Hyperthyroidism during pregnancy.

WALLACE, J.T. (1933) Thyrotoxicosk in its relation to pregnancy. American Journal of Obsfetrics and

WALLACE, J.T. (1940) Thyrotoxicosis in pregnancy. Medical Times (Nen' York), 68, 274277. WHITELAW, M.J. (1947) Thiouracil in the treatment of hyperthyroidism complicating pregnancy and its effect of

869-874.

methimazole. Obstetrics and Gynecology, 46, 282-286.

Medica. 6 , 609.

Obstetrics and Gynecology, 36, 779-785.

Gynecology, 26, 77-83.

the human fetal thyroid. Journal of Clinical Endocrinology, 7, 767-773.