Dear Editor,

Thyroid hormones are produced by the fetus from 11 weeks gestation[1] and are essential for normal fetal and neonatal development. Maternal autoimmune thyroid disease (Graves' disease) or its treatment may adversely affect this normal development by causing hypo- or hyperthyroidism in utero and/or in the early neonatal period. This case demonstrates that the fetus/infant may be susceptible to both mechanisms of disruption of thyroid function.

A 35-year-old woman in her fifth pregnancy with known Graves' disease presented with thyrotoxicosis at 7 weeks gestation (thyroid stimulating hormone (TSH) 0.01 mIU/L (0.3–4), fT3 50 pmol/L (3.5–6.5), fT4 95.3 pmol/L (10–23)) with positive thyroid antibodies (anti-thyroid peroxidase antibodies 190 IU/mL (<10) and thyroid stimulating immunoglobulin (TSI) >40 U/L (0–1). She was transferred from carbimazole to propylthiouracil (PTU) daily dose 150 mg. This was reduced to 100 mg by the end of the first trimester. Low TSH and high fT3 levels persisted throughout the second trimester (fT3 8.2–29.3 pmol/L) but levels of fT4 returned to the normal range.

On ultrasound at 26 weeks gestation, a fetal goitre was identified. The goitre appeared to have a uniform increase in vascularity, fetal growth, heart rate and skeletal ossification appeared normal. By 31 weeks, maternal thyroid function remained unchanged but the fetal goitre had enlarged with a circumference of 9.4 cm, >95th centile (95th centile by biparietal diameter 6.5 cm and by gestational age 5.2 cm).[2] Referral was made to a tertiary fetal therapy unit for fetal blood sampling (FBS) to determine fetal thyroid function and to assess the potential obstructive effect of the fetal neck mass and possible need for ex utero intrapartum treatment procedure.

The trachea appeared patent, fetal swallowing was observed and liquor volume was normal, and there was no significant change in the size of the goitre (Fig. 1). FBS revealed fT3 3.5 pmol/L (3.9–6.8), fT4 6.7 pmol/L (12–22) and TSH of >100 mIU/L (0.3–4), confirming severe fetal hypothyroidism. Treatment with 60 mcg intra-amniotic tri-iodothyronine (Liothyronine sodium, Goldshield Pharmaceuticals, Croydon UK) was given and maternal PTU decreased to 50 mg daily. By 35+3 weeks, maternal clinical and biochemical status was increasingly thyrotoxic (TSH 0.01 mU/L, fT3 19.1 pmol/L and fT4 22.9 pmol/L). Despite administration of propanolol, the woman was increasingly symptomatic and so a decision was made to proceed with delivery. A healthy male infant weighing 2850 g (63rd customised birthweight centile) was delivered by an uncomplicated lower segment caesaren section. There were no concerns with the neonate's airway and no palpable goitre.


Figure 1. Transverse view of bilobed fetal goitre with patent trachea (arrow) at 32 weeks gestation. Thyroid circumference margins identified by dotted line measures 10.7 cm circumference, >95th centile (95th centile by BPD 6.8 cm and by gestational age 5.5 cm).[2]

Download figure to PowerPoint

On day one, the infant's thyroid function was normal (TSH 11 mIU/L (0.01–10.5), fT3 5.5 pmol/L (3.4–7.2), fT4 15.3 pmol/L (10–40)). However, from day 10, the infant was thyrotoxic (TSH 0.09 mIU/L, fT4 61.1 pmol/L, fT3 14.7 pmol/L) with TSI level >40 U/L (0–1). Carbimazole (1.25 mg bd) and propranolol (1 mg tds) were commenced due to tachycardia and frequent wakening. There was a rapid response, with doses reduced, and finally stopped, at 3 weeks of age. At this stage, thyroid function remained normal with the infant settled, sleeping and feeding well.

The development of a fetal goitre is relatively rare but serial ultrasound evaluation of the fetal thyroid is now accepted practice for women with autoimmune thyroid disease in pregnancy. Both hypo- and hyperthyroidism may result in a fetal goitre. Maternal auto-antibodies may cross the placenta and stimulate the fetal thyroid TSH receptor, causing excess thyroid hormone production, fetal hyperthyroidism and tachycardia that may lead to fetal hydrops. Conversely, and more rarely, these auto-antibodies may also inhibit the fetal thyroid TSH receptor and cause hypothyroidism. In addition, antithyroid drugs cross the placenta and may inhibit fetal thyroperoxidase, causing inhibition of fetal thyroid hormone production, hypothyroidism and so potential neurological damage. Huel et al. have identified ultrasound features that may help differentiate thyroid function including thyroid vascular pattern (hypothyroid – peripheral vascularisation and hyperthyroid – central vascularisation), fetal heart rate (hyperthyroid – tachycardia) and bone maturation (hypothyroid – delayed and hyperthyroid – accelerated).[3] However, FBS should still be considered to elucidate the cause of a fetal goitre prior to commencing treatment.

Fetal hypothyroidism secondary to maternal antithyroid drug use treated with reduction of maternal drug dose and/or fetal therapy with T4 or T3 has previously been reported.[4, 5] In addition, fetal/neonatal hyperthyroidism as a consequence of maternal thyroid disease has been well described. However, reports of the fetus and infant being affected by both hypo- and hyperthyroidism are rare. Rosenfeld et al. report on a register of women treated with PTU in pregnancy and summarise two cases of fetal hypothyroidism and subsequent neonatal hyperthyroidism.[6] To our knowledge, this is the first case with clinical details of such a case with demonstrable maternal antibodies in the neonate.

In this case, despite a relatively low dose of PTU and very high maternal antibodies, the fetus became significantly hypothyroid in utero secondary to maternal PTU (inhibition by maternal antibodies is unlikely due to rapid onset hyperthyroidism in the postnatal period). As maternal PTU dosage was decreased and the drug has a short half life, it is likely that drug levels were negligible in the neonate shortly following delivery. Conversely, maternal auto-antibodies may persist for several months and this neonate had very high levels of maternal antibody soon after birth that lead to symptomatic hyperthyroidism once the effects of PTU were eliminated.

This case highlights the need to consider both hypo- and hyperthyroidism as causes of fetal goitre, the benefit of antenatal sampling to confirm the diagnosis, and the need for ongoing neonatal surveillance.


  1. Top of page
  2. References