Hyperthyroidism is a common disorder in pregnant women and is often caused by Graves disease. In pregnant women, the incidence of hyperthyroidism is 0.2% to 0.3%, and Graves disease accounts for 85% of cases. Untreated hyperthyroidism has serious risks, including preeclampsia, thyroid storm, heart failure, miscarriage and stillbirth, prematurity, intrauterine growth restriction, and neonatal thyrotoxicosis. Thyroid dysfunction is often treated with propylthiouracil (PTU) or methimazole (MMI), but both drugs readily cross the placenta with similar kinetics of placental transfer. Propylthiouracil has been the drug of choice during pregnancy because MMI is associated with teratogenic effects. Over the last 20 years, several reports have raised concerns about hepatotoxicity caused by PTU, and it has been the subject of a “black box warning.” This literature review was undertaken to assess the maternal and fetal risks of treatment with MMI and PTU during pregnancy.
PubMed, EMBASE, TOXNET, TOXLINK, DART, Medscape, EBSCO, and Google sought English and non-English publications, with searches limited to human data and clinical trials. Reference lists of relevant articles and all selected articles were hand-searched to identify additional trials.
Thirty-one of 2468 publications identified were reviewed to evaluate the adverse effects of MMI or PTU and included cohort and case-control studies and case reports or case series. No randomized controlled trials were identified. Several retrospective cohort studies addressed MMI teratogenicity, but only a few of these identified cases with one of the features associated with fetal MMI. Two case-control studies investigated an association between MMI and specific malformations. In 1 study, prenatal exposure to maternal hyperthyroidism treated with MMI was identified in 16.4% of cases (10/61) and only 1.1% (2/183) in the control group. The data suggested that prenatal exposure to maternal hyperthyroidism and treatment with MMI were associated with choanal atresia (CA). Methimazole-related adverse fetal outcomes were reported in numerous case reports/series. Fifteen cases with aplasia cutis congenita, 12 with CA, and 3 with other features of MMI embryopathy were described.
Several cohort studies of PTU exposure during pregnancy met inclusion criteria. One assessed 115 PTU-exposed pregnancies and 1141 controls exposed to nonteratogenic drugs and found comparable rates of major anomalies (1.3% for PTU exposure and 3.2% for controls; P = 0.5). Hypothyroidism was present in 9.5% of fetuses or neonates (56.8% of whom had goiter). Neonatal hyperthyroidism, possibly resulting from maternal disease, was found in 10.3%. Goiters diagnosed prenatally by ultrasound were successfully treated in utero by adjusting the maternal dose of PTU. Neonatal thyroid function generally became normal during the first month of life without treatment.
One study examined the intellectual development of 31 subjects aged 4 to 23 years, born to women with Graves disease; 15 were exposed to MMI (40–140 mg/wk), and 16 were exposed to PTU (250–1400 mg/wk) during pregnancy. The exposed and unexposed groups did not differ in total IQ. No differences were apparent between those exposed to MMI or to PTU or between those exposed to higher or lower doses. All children were euthyroid at birth, and none had goiter.
Over the past 20 years, 22 serious cases of liver injury in adults (resulting in 9 deaths and 5 liver transplants) have been reported to the Food and Drug Administration (FDA) and to the Adverse Event Reporting System. In 2010, the FDA made the following recommendations: (1) PTU should not be prescribed as the first-line treatment agent in children or adults; (2) because of MMI teratogenicity, PTU should be prescribed for hyperthyroidism during the first trimester of pregnancy, at least until more is known regarding the teratogenicity of MMI; (3) PTU is recommended over MMI in life-threatening thyrotoxicosis or thyroid storm because of its superiority in inhibiting peripheral conversion of T4 to T3; and (4) PTU can be used in individuals who have had adverse responses to MMI (other than agranulocytosis) and for whom radioiodine or surgery are not options. In neonates, the most significant case described neonatal hepatitis that was positive for newborn lymphocyte transformation test for PTU, which suggests an immunologic reaction to PTU through lymphocytes. Because PTU and MMI cross the placenta and are detected in fetal circulation, PTU treatment should be discontinued after the first trimester of pregnancy.
In conclusion, uncontrolled maternal hyperthyroidism may cause congenital malformations and adverse pregnancy outcomes. In most case reports of MMI-associated embryopathy, MMI exposure occurred in the first trimester and mostly in mothers with uncontrolled thyrotoxicosis. Choanal atresia and aplasia cutis congenita have not been reported in association with PTU treatment. The possible teratogenic effect of MMI cannot be ignored, and use of the drug should be avoided in the first trimester. Propylthiouracil may not be a human teratogen. However, a major concern with PTU treatment is the potential for serious hepatotoxicity. Pregnant women with hyperthyroidism should receive PTU during the first trimester and MMI thereafter.
Department of Pharmacology and Toxicology, Faculty of Medicine, University of Toronto; Division of Clinical Pharmacology and Toxicology, Hospital for Sick Children, Toronto, Ontario, Canada