Maternal thyroxine and risk of fetal malformations

Maternal thyroid hormone is essential for a normal pregnancy and fetal development. The fetal thyroid gland is not functionally mature until 18–20 weeks of gestation resulting in the fetal demand for the hormone in early pregnancy depending on the placental transfer of maternal thyroid hormone [1]. Therefore, maternal thyroid dysfunction might result in subsequent adverse consequences. For example, compared with the normal-range levels, maternal higher free thyroxine (FT4) levels in early pregnancy have been associated with lower birth weight and an increased risk of small gestational age newborns [2], while both high and low FT4 related to fetal brain underdevelopment (lower child intelligence quotient, gray matter, and cortex volume) [3].

Despite theoretical possibilities, evidence for the association between FT4 concentrations and the risk of fetal congenital malformations is largely absent due to inadequate statistical power. It is difficult to conduct a cohort study with high statistical power due to the fact not many pregnant women receive a thyroid function test since universal tests are not recommended, the prevalence of congenital malformations is low, and the correlation might be weak. However, early pregnancy tested thyroid dysfunction has been associated with a higher likelihood of specific congenital malformations (e.g. neural tube defects, anencephaly or microcephaly, and musculoskeletal malformations) in some case-control studies [4,5] and a small sample size cohort study [6].

Recently, we reported a significant association between FT4 concentrations and the risk of congenital heart defects (CHDs) in a hospital-based cohort study [7]. In this study, 52,047 women who received a universal thyroid function test were recruited in the initiation stage and 41,647 women with singleton pregnancies were included in the final analysis. Considering that the low prevalence of confirmed CHDs (96 confirmed cases, with a prevalence of 2.3 per 1,000 infants) diagnosed between the delivery and the 42nd day of birth might induce a potential bias, we also included fetuses who were suspected of CHD complications in their ultrasound examinations performed between 20 and 24 weeks of gestation (215 total CHD cases, 5.2 per 1,000 infants). After controlling for potential confounders, including maternal age, residence, parity, assisted conception, gestational diabetes, preeclampsia and fetal sex, adjusted odds ratios (OR) and 95% confidence interval [95% CI] for total CHDs were 1.04 (95% CI, 1.01–1.07) and 1.05 (95% CI, 1.02–1.08) for confirmed CHDs, respectively. In addition, a stronger association was found when restricted to women who had a thyroid function test between 12 and 18 weeks of gestation (OR = 1.05, 95% CI: 1.01–1.09 for total CHDs; and OR = 1.07, 95% CI: 1.03–1.11 for confirmed CHDs). Interestingly, a derived variable, the free-to-total thyroxine proportion (‰), which may reflect the status of binding proteins/globulins and the health status of the mother, was significantly related to an increased risk of total CHD and confirmed CHD (OR was 2.41 [95% CI: 1.27–4.59] and 4.34 [95% CI: 1.92–9.79], respectively). The association also became stronger when restricting to 12–18 weeks of gestation (OR = 3.32; 95% CI: 1.43–7.73 and 10.46 95% CI: 3.93–27.86, respectively).

These findings have several implications. Firstly, it seems that all pregnant women should be tested for their thyroid function to identify those at higher risk of having an infant with CHDs. However, we should also consider the cost of a universal thyroid function test in the context of the low prevalence of CHDs. Routine thyroid function tests are not currently recommended by the American Thyroid Association and the American College of Obstetricians and Gynecologists [8]. Therefore, further study on the cost-effectiveness analysis is warranted to address this issue.

Secondly, the findings between FT4 concentration and the risk of CHDs might raise concerns about the authenticity between antithyroid drugs exposure (such as propylthiouracil and/or methimazole) and the risk of congenital malformations. CHDs are the most common congenital malformations in newborns and the correlation between antithyroid drugs and risk of congenital malformations is weak (e.g. an adjusted OR of 1.19, 95% CI: 1.12–1.28 was estimated) [9]. The correlation might weaken if FT4 concentration is included as a potential confounder in the adjusted models.

Thirdly, the overlap between the time for thyroid function tests (e.g. 12–18 weeks of gestation) and the fetal heart maturation cycle (begins from 50 days of a human embryo to birth) [10], make it difficult to distinguish whether a higher FT4 concentration is involved in the regulation of CHDs or just as a coexisting condition. However, these associations may provide new insights into the mechanism of CHDs. Finally, the free-to-total thyroxine proportion may be a new indicator that needs more attention since it showed the strongest correlation to the risk of CHDs, which might indicate its high value in screening fetal CHDs between 12 to 18 weeks of gestation.

In summary, assuming that the study limitations did not affect the results [7], more investigations are warranted to evaluate the necessity of a universal thyroid function test, reveal the role of thyroid hormones in the mechanism of CHDs, and judge the value of the marker derived from 12–18 weeks tested thyroid function in screening for fetal CHDs.

Disclosure statement

No potential conflict of interest was reported by the author(s).