Hyperthyroidism during pregnancy: Treatment

INTRODUCTION — The treatment of pregnant women with hyperthyroidism parallels that of nonpregnant women but presents unique problems. There are several important issues to be considered when hyperthyroidism occurs during pregnancy. These include understanding indications for treatment, therapeutic options for pregnant women, and goals of antithyroid drug therapy.

The treatment of hyperthyroidism during pregnancy is presented here. The clinical manifestations, diagnosis, and causes of hyperthyroidism and other aspects of thyroid disease during pregnancy are discussed separately. (See "Hyperthyroidism during pregnancy: Clinical manifestations, diagnosis, and causes" and "Overview of thyroid disease and pregnancy".)

OVERVIEW OF TREATMENT — The Endocrine Society and the American Thyroid Association (ATA) have published clinical guidelines for the management of thyroid dysfunction during pregnancy that are similar to the approach outlined here [1-3]. Treatment recommendations are predominantly based upon observational studies and clinical experience.

Goals — A good fetal and maternal outcome depends upon controlling the mother's hyperthyroidism. There are changes in thyroid physiology during normal pregnancy that are reflected in altered thyroid function tests. Thyroid function tests, therefore, should be interpreted with an understanding of trimester-specific reference ranges for the individual tests. (See "Overview of thyroid disease and pregnancy", section on 'Trimester-specific reference ranges'.)

The goal of treatment is to maintain persistent but mild hyperthyroidism in the mother in an attempt to prevent fetal hypothyroidism since the fetal thyroid is more sensitive to the action of antithyroid drugs [4]. Overtreatment of maternal hyperthyroidism with thionamide antithyroid drugs can cause fetal goiter and primary hypothyroidism. On the other hand, transient central hypothyroidism may be seen in infants whose mothers had poorly controlled hyperthyroidism during pregnancy, presumably due to suppression of the fetal pituitary-thyroid axis [5].

To attain the goal of mild hyperthyroidism, the mother's serum free thyroxine (T4) concentration should be maintained at or just above the trimester-specific normal range for pregnancy or (especially if the trimester-specific reference range is not available) the total T4 and triiodothyronine (T3) should be maintained at 1.5 times above the nonpregnant reference range. The serum thyroid-stimulating hormone (TSH) concentration should be below the reference range for pregnancy (eg, goal TSH approximately 0.1 to 0.3 mU/L), using the lowest possible dose of medication [2].

Attaining these goals requires assessment of thyroid function frequently (ie, at four-week intervals) with appropriate adjustment of medication. (See 'Monitoring and dose adjustments' below.)

Indications for treatment — Women with symptomatic, moderate to severe, overt hyperthyroidism due to Graves' disease, toxic adenoma, toxic multinodular goiter, or gestational trophoblastic disease require treatment of hyperthyroidism. Such patients will almost always have TSH values below 0.05 mU/L and elevations in trimester-specific free T4 concentrations and/or total T4 and T3 concentrations that exceed 1.5 times the upper limit of normal for nonpregnant patients.

Although hyperthyroidism due to gestational trophoblastic disease resolves with treatment of the underlying gestational trophoblastic disease and subsequent normalization of human chorionic gonadotropin (hCG) levels, symptomatic women require treatment prior to surgery. (See "Nonthyroid surgery in the patient with thyroid disease", section on 'Preoperative preparation for urgent surgery' and "Hydatidiform mole: Treatment and follow-up".)

Not all patients with biochemical, overt hyperthyroidism require treatment if the hyperthyroidism is mild, since the goal of treatment is to maintain mild maternal hyperthyroidism.

Treatment of hyperthyroidism is not required in the following settings:

●Transient, subclinical hyperthyroidism (normal serum total or free T4 and T3 concentrations for pregnancy in the presence of a subnormal TSH) in the first trimester of pregnancy, because it is considered a normal physiologic finding and therefore does not require therapy. (See "Overview of thyroid disease and pregnancy", section on 'Thyroid adaptation during normal pregnancy'.)

●hCG-mediated, overt hyperthyroidism (also called gestational transient thyrotoxicosis), because it is usually transient and mild. (See "Hyperthyroidism during pregnancy: Clinical manifestations, diagnosis, and causes", section on 'Gestational transient thyrotoxicosis'.)

●Hyperemesis gravidarum-associated hyperthyroidism, because it is usually mild and subsides as hCG production falls (typically by 16 to 18 weeks gestation). Women with severe hyperemesis, however, require treatment of dehydration with intravenous fluids. (See "Nausea and vomiting of pregnancy: Treatment and outcome".)

●Subclinical and mild, asymptomatic, overt hyperthyroidism due to Graves' disease, toxic adenoma, or toxic multinodular goiter.

Subclinical hyperthyroidism (subnormal TSH and free T4 within the trimester-specific reference range or total T4 and T3 <1.5 times above the upper limit of normal for nonpregnant patients) never requires treatment during pregnancy. Biochemical, overt hyperthyroidism (subnormal TSH and free T4 above the trimester-specific reference range or total T4 and T3 >1.5 times above the upper limit of normal for nonpregnant patients) may not require treatment if mild and asymptomatic (or minimally symptomatic), since the goal of treatment is to maintain mild maternal hyperthyroidism. (See 'Goals' above.)

In women who are being monitored without therapy, we measure TSH, free T4 (if there is a trimester-specific reference range), and/or total T4 or total T3 every four to six weeks.

Therapeutic options — The therapeutic options for hyperthyroid pregnant women are limited because of the potential adverse fetal effects of the available treatments. Most women are treated with thionamides. Thyroidectomy in the second trimester is an option for women who are unable to take thionamides.

●Thionamides – Thionamides are the primary treatment of hyperthyroidism due to Graves' disease, toxic adenoma, or toxic multinodular goiter during pregnancy. They are actively transported into the thyroid gland, where they inhibit both the organification of iodine to tyrosine residues in thyroglobulin and the coupling of iodotyrosines. (See 'Choice of thionamide' below.)

●Beta blockers – Beta blockers, such as metoprolol or propranolol (but not atenolol), can be used to treat tachycardia and tremor. They are the primary treatment for patients with hydatidiform mole or gestational trophoblastic neoplasia who generally cannot wait for three to six week for thionamides to control hyperthyroidism prior to surgery.

In general, long-term treatment with beta blockers (longer than two to six weeks) should be avoided in pregnant women because of concerns regarding fetal growth retardation and hypoglycemia, especially with atenolol [6-8].

●Thyroidectomy – Thyroidectomy during pregnancy is rarely necessary but is an option for women who cannot tolerate thionamides because of allergy or agranulocytosis. (See 'Thionamide intolerance' below.)

Plasmapheresis has also been used to rapidly control hyperthyroidism in women with trophoblastic disease and severe hyperthyroidism [9,10].

OUR APPROACH — Our approach outlined here is similar to that of the Endocrine Society and the American Thyroid Association (ATA) clinical guidelines for the management of thyroid dysfunction during pregnancy [1-3]. The majority of the discussion is focused on the treatment of Graves' disease during pregnancy as toxic adenoma, toxic multinodular goiter, and gestational trophoblastic disease are relatively less common.

Control of symptoms — In pregnant women with symptomatic, moderate to severe hyperthyroidism, beta blockers may be given to ameliorate symptoms. We typically start with metoprolol 25 to 50 mg daily. Propranolol, 20 mg every six to eight hours, is an alternative option. The dose can be increased as needed to control symptoms. However, if possible, beta blockers should be weaned as soon as the hyperthyroidism is controlled by thionamides because occasional cases of neonatal growth restriction, hypoglycemia, respiratory depression, and bradycardia have been reported after maternal administration [6,7,11].

In a retrospective analysis, women with chronic hypertension who used atenolol in early (15 weeks gestation) compared with late (30 weeks gestation) pregnancy had significantly smaller babies [6]. There has been one report suggesting a higher rate of spontaneous abortion for hyperthyroid women treated with both a thionamide and propranolol compared with a thionamide alone [12]. In most patients, beta blockers can and should be tapered and discontinued within two to six weeks. (See "Beta blockers in the treatment of hyperthyroidism" and "Graves' hyperthyroidism in nonpregnant adults: Overview of treatment", section on 'Symptom control: Beta blockers'.)

Decrease thyroid hormone synthesis — For pregnant women with moderate to severe hyperthyroidism due to Graves' disease, toxic adenoma, or toxic multinodular goiter, thionamides are our first choice of treatment to decrease thyroid hormone synthesis. Both methimazole and propylthiouracil (PTU) probably cross the placenta with equal transfer kinetics [13] and have similar effects on the fetal thyroid [14,15]. In one report of 77 newborns of euthyroid mothers treated with PTU or methimazole, there were no significant differences in TSH concentrations measured in cord blood at birth [15]. However, compared with control neonates who were not exposed in utero to thionamides, mean cord TSH levels were higher in neonates exposed to PTU but not methimazole.

Low thyroid function at birth is found in approximately one-half of neonates whose mothers received PTU or methimazole during pregnancy and who had serum T4 concentrations within the normal (nonpregnant) range [4]. In spite of these observations, two studies reported that the intelligence quotient (IQ) scores of children who were exposed to thionamides in utero (but were euthyroid at birth) were normal [16,17].

Pretreatment evaluation — Prior to initiating thionamides, we obtain baseline blood tests, including a complete blood count (white count with differential) and a liver profile (bilirubin and transaminases). We do not use thionamides in patients with a baseline absolute neutrophil count <1000 cells/microL or elevated liver transaminases (more than fivefold the upper limit of normal) except in selected patients after careful assessment of alternatives and risks.

Choice of thionamide — The thionamide drugs available in the United States are PTU and methimazole. Carbimazole is available in other countries. Carbimazole is completely metabolized to methimazole, although the carbimazole dose required to yield an equivalent dose of methimazole is approximately 40 percent higher. The choice of thionamide depends upon which trimester the drug is being initiated. Methimazole is preferred to PTU except during the first trimester of pregnancy.

●Diagnosed prior to pregnancy – Women diagnosed with Graves' disease prior to pregnancy who are taking methimazole could [2,3]:

•Elect to have definitive therapy with surgery or radioiodine prior to pregnancy. Women should then postpone pregnancy until they have become euthyroid following definitive treatment and on replacement therapy. This option is recommended for women who are requiring high doses of methimazole to maintain a euthyroid state.

•Switch to PTU before trying to conceive. This option is most reasonable in younger women with normal periods who are expected to conceive within one to three months.

•Switch to PTU as soon as the pregnancy test is confirmed. This option is more reasonable for older individuals having difficulty conceiving. It is recommended that a pregnancy test be obtained weekly.

•Discontinue methimazole with careful monitoring of thyroid function tests (weekly throughout the first trimester, then monthly). This option is best chosen for women who have already been treated with methimazole for 12 to 18 months, have a normal TSH level on low-dose therapy, and are thyrotropin receptor antibody (TRAb) negative. If hyperthyroidism recurs after discontinuation, the patient should be treated with PTU (if relapse occurs in the first trimester) or methimazole (if relapse occurs after the first trimester).

This approach is supported by a prospective study of 63 pregnant Chinese women who discontinued antithyroid drugs during early pregnancy [18]. Normal TSH and negative TRAb at the time of antithyroid drug discontinuation was associated with a lower risk of recurrence than when TSH was subnormal and TRAB positive (13 versus 83.3 percent).

Patients taking PTU during the first trimester may continue it for the remainder of pregnancy or switch back to methimazole at the start of the second trimester, as described immediately below.

●Diagnosed during the first trimester – Women diagnosed with symptomatic, moderate to severe hyperthyroidism during the first trimester of pregnancy should take PTU. Patients may continue PTU for the remainder of pregnancy or switch back to methimazole at 16 weeks. Clinical practice varies, and the decision should be made after discussing the options with patients and considering their values and preferences. Since methimazole is more potent than PTU, switching may increase the risk of maternal or fetal hypothyroidism. While switching to methimazole reduces the exposure to PTU, which has more serious hepatotoxicity than methimazole, most cases of liver failure occur within the first 60 days of treatment.

●Diagnosed after the first trimester – Women diagnosed with symptomatic, moderate to severe hyperthyroidism after the first trimester should take methimazole.

All three antithyroid drugs have been associated with possible teratogenic effects, but teratogenic effects are more severe with methimazole and carbimazole compared with PTU. There are numerous case series of aplasia cutis, a scalp defect, in newborns exposed to methimazole in utero [19-21] (see "Vesicular, pustular, and bullous lesions in the newborn and infant", section on 'Aplasia cutis congenita'). More serious congenital malformations, such as tracheoesophageal fistulas, patent vitellointestinal duct, choanal atresia, omphalocele, and omphalomesenteric duct anomaly, have also been observed with maternal use of methimazole and carbimazole but not PTU [21-31]. However, mild congenital anomalies, including preauricular sinuses and cysts and urinary tract abnormalities, have been observed after PTU [31]. In two studies, the prevalence of congenital anomaly was similar after exposure to either PTU or methimazole/carbimazole, but the PTU-associated abnormalities were less severe [31,32], while in another study, the prevalence was higher with exposure to methimazole compared with PTU [33]. (See "Thionamides: Side effects and toxicities", section on 'Potential teratogenicity'.)

Gestational weeks 6 to 10 is the period of highest risk for congenital anomalies from exposure to thionamides. Therefore, PTU is preferred during the first trimester. However, reports of severe PTU-related liver failure have raised concerns about the routine use of PTU, including the use of PTU in pregnancy [34]. Methimazole has also been associated with liver disease, but it is typically due to cholestatic dysfunction in the United States, and methimazole is less likely to cause liver failure than PTU. Certainly, a woman with even mild elevations in transaminases while taking PTU during the first trimester should be transitioned to methimazole during the second trimester, but because of the concern that methimazole (which is more potent than PTU) might result in transient maternal or fetal hypothyroidism, endocrinologists treating patients who are stable on PTU might prefer to continue PTU throughout pregnancy [2]. (See "Thionamides: Side effects and toxicities", section on 'Hepatotoxicity'.)

The following studies illustrate the range of reported teratogenic effects of thionamides:

●In the United Kingdom, spontaneous reporting of adverse drug reactions over a 47-year period found 57 cases with 97 anomalies for carbimazole and only six cases with 11 anomalies for PTU; however, the relative use of these two drugs in pregnant women is unknown [29].

●In a population-based study from Taiwan, there was no increased risk of congenital anomalies among babies of hyperthyroid women treated with either PTU (n = 630 women) or methimazole (n = 73 women) [35]. The small number of methimazole-treated women may have made this study underpowered to detect a teratogenic effect.

●In a Japanese study of pregnant women with Graves' disease, 50 of 1231 patients (4.1 percent) treated with methimazole, 12 of 1399 (1.9 percent) treated with PTU, and 40 of 1906 (2.1 percent) who did not receive antithyroid drugs had congenital malformations [30].

●In a registry-based, cohort study from Denmark (817,093 live-born infants, 1820 exposed to antithyroid drugs), the prevalence of congenital anomaly was significantly higher in children exposed to any antithyroid drug (8.0, 9.1, 10.1, and 5.4 percent for PTU, methimazole/carbimazole, both PTU and methimazole/carbimazole, and no antithyroid drug use in early pregnancy, respectively) [31]. Methimazole/carbimazole and PTU were associated with urinary system malformation and PTU with malformations in the face and neck region. Choanal atresia, esophageal atresia, omphalocele, omphalomesenteric duct anomalies, and aplasia cutis were common in methimazole/carbimazole-exposed children. The congenital anomalies associated with PTU were less severe, but 9 of those 14 children underwent surgical correction [36].

●In a similarly designed registry analysis in Sweden, the prevalence of congenital anomaly was similar between children exposed to methimazole (6.8 percent) and PTU (6.4 percent), and it was not higher than the prevalence of congenital anomaly in nonexposed children (8 percent) [32].

●In the Korean National Health Insurance database of 2,886,970 pregnancies, 12,891 were exposed to antithyroid drugs. The absolute increases in the prevalence of malformations per 1000 live births were 8.81 for PTU alone versus 17.05 for methimazole alone and 16.53 for exposures to both antithyroid drugs during the first trimester [33]. Higher doses of methimazole were associated with a higher risk of congenital anomaly.

Initial dosing — To minimize the risk of hypothyroidism in the fetus, we give the lowest dose of thionamide necessary to control thyroid function:

●PTU 50 mg two to three times daily

●Methimazole 5 to 10 mg daily, or

●Carbimazole 5 to 15 mg daily

However, in patients with severe hyperthyroidism, full initial doses of PTU (100 mg three times per day) or methimazole (10 to 30 mg daily) may be required to control hyperthyroidism.

For patients switching between PTU and methimazole, the closest dose approximation should be made. Although the ratio of potencies of PTU and methimazole is uncertain, clinical experience suggests that methimazole is 20 to 30 times as potent on a milligram-to-milligram basis. Therefore, 300 mg of PTU would be roughly equivalent to 10 or 15 mg of methimazole.

Monitoring and dose adjustments — Graves' disease frequently ameliorates in the third trimester. Whenever possible, based on thyroid function tests and assessment of TRAb measurements, thionamides should be tapered and potentially discontinued during the third trimester.

Toxic adenoma and toxic multinodular goiter are unlikely to remit during pregnancy, and therefore, women with hyperthyroidism due to these disorders are usually maintained on thionamides throughout pregnancy.

●Thyroid function tests – Thyroid function tests (TSH and free T4 or total T4 if a trimester-specific reference range is not available for free T4) should be obtained every four weeks throughout pregnancy. If thionamides are discontinued in early pregnancy, thyroid tests should be checked weekly throughout the first trimester, then monthly.

Extra caution is necessary after switching thionamides. Thyroid tests should be performed two weeks after switching to be sure that a euthyroid state has been maintained and to avoid maternal overtreatment and fetal hypothyroidism.

The thionamide dose should be adjusted based on the results of the thyroid function tests to maintain serum free T4 concentrations at or just above the upper limit of normal, using a trimester-specific reference range, or the total T4 and T3 (if trimester-specific normal ranges for free T4 are not available) should be maintained at approximately 1.5 times the upper limit of normal for nonpregnant adults. Serum TSH concentrations should be maintained below the reference range for pregnancy.

Ultimately, low doses of PTU or methimazole (eg, 50 mg twice daily or less for PTU; 2.5 to 5 mg a day for methimazole) may be all that is required [37]. It is possible to discontinue the thionamide during the third trimester in at least one-third of women [37]. In this setting, thyroid tests should be obtained two weeks after discontinuation and then monthly. The amelioration of hyperthyroidism as pregnancy progresses is due to a fall in serum TRAb concentrations or, rarely, a rise in TSH receptor-blocking antibodies [38,39]. However, Graves' hyperthyroidism often worsens postpartum. (See 'Postpartum issues' below.)

Monitoring thyroid function tests throughout pregnancy is important because maternal hyperthyroidism in the third trimester may increase the risk of low birth weight (independent of the risk of neonatal Graves' disease). (See "Hyperthyroidism during pregnancy: Clinical manifestations, diagnosis, and causes", section on 'Pregnancy complications'.)

●TRAb – For women diagnosed with hyperthyroidism during pregnancy who will be taking thionamides, serum TRAb should be measured at diagnosis and, if elevated, again at 18 to 22 weeks and at 30 to 34 weeks of gestation.

Disappearance of TRAb indicates potential remission of Graves' disease, and the dose of thionamides can be reduced and potentially discontinued.

High TRAb levels in late pregnancy are associated with an increased risk of fetal and neonatal hyperthyroidism (see 'Fetal or neonatal hyperthyroidism' below). Neonatal hyperthyroidism is reviewed in detail separately. (See "Evaluation and management of neonatal Graves disease" and "Thyroid physiology and screening in preterm infants".)

●Adverse effects – PTU-associated liver failure, which can occur at any time during the course of treatment, has a sudden onset and a rapidly progressive course. Therefore, routine monitoring of liver function is not currently suggested by the ATA and US Food and Drug Administration (FDA). Patients should be advised to stop their medication and contact their clinician if they develop weakness, malaise, nausea and vomiting, jaundice, dark urine, or light-colored stools.

Some clinicians and their patients prefer to monitor liver function every four weeks when blood is being drawn to assess thyroid function. If this approach is chosen, PTU should be discontinued if serum transaminases are greater than three times the upper limit of normal. This approach has not been shown to reduce the risk of PTU-associated liver failure. Once the drug is discontinued, the patient should be followed closely with frequent transaminase measurements to be sure that they are returning to normal. If they progressively rise, immediate referral to a clinician with expertise in liver disease is recommended.

If the patient who was tolerating PTU during the first trimester and had no adverse side effects, such as abnormal liver function tests, develops a rash when switched to methimazole in the second trimester, methimazole should be stopped and PTU resumed. (See "Thionamides: Side effects and toxicities".)

Thionamide intolerance — For women with Graves' disease who cannot tolerate thionamides because of allergy or agranulocytosis, thyroidectomy during pregnancy may be necessary. Radioiodine therapy for pregnant women with hyperthyroidism is absolutely contraindicated.

The indications for surgery are similar to those in nonpregnant individuals. Surgery during pregnancy, however, is associated with an increased risk of spontaneous abortion or premature delivery [40] and significantly higher rates of surgical complications than nonpregnant women [41]. These risks are minimized by operating during the second trimester. (See "Nonobstetric surgery in pregnant patients: Patient counseling, surgical considerations, and obstetric management", section on 'Timing of surgery'.)

Prior to thyroidectomy, pregnant women with intolerance to thionamides should be pretreated with beta blockers (metoprolol or propranolol) and a short course (ie, 7 to 10 days) of potassium iodine solution (35 to 50 mg iodine per drop, 1 to 3 drops daily). Iodine lowers serum thyroid hormone concentrations acutely and, in addition, decreases thyroid gland vascularity. (See "Surgical management of hyperthyroidism", section on 'Preoperative preparation'.)

Is there a role for iodine as primary therapy for hyperthyroidism? — We believe there is insufficient evidence to recommend routine iodine for the primary treatment of pregnant women with Graves' disease.

Older reports suggested that prolonged, high-dose iodine therapy can cause fetal goiter [42]. However, a study from Japan of 35 women with mild to moderate Graves' hyperthyroidism suggested that low doses of iodine are safe during pregnancy [43]; only one neonate had subclinical hypothyroidism at birth. Another study from Japan compared 283 women who shifted from methimazole to iodine during the first trimester with 1333 women who continued thionamides [44]. The women taking iodine had fewer congenital anomalies and more live births than those continuing thionamides, although their hyperthyroidism was less well controlled. These data suggest a potential role for iodine (eg, potassium iodide solution [SSKI] 1 drop daily) in women who require therapy but are intolerant of thionamides.

Therapies not recommended

●Radioiodine – Radioiodine is absolutely contraindicated during pregnancy (and breastfeeding) [45]. Fetal thyroid tissue begins to function by 10 to 12 weeks and therefore can be ablated by radioiodine. Many experienced clinicians, however, have encountered one or two women inadvertently treated with radioiodine during early pregnancy; the anecdotal impression is that radioiodine given before approximately 8 to 10 weeks of pregnancy does not cause fetal hypothyroidism or congenital anomalies. In one systematic review, the risks to the fetus were dependent upon the timing of exposure during pregnancy [46]. Spontaneous pregnancy loss was more likely when exposure (100 mGy) occurred during the first two weeks (prior to implantation). Exposure during organogenesis (from two weeks gestation) may result in congenital anomalies. After fetal thyroid has developed the capacity to capture iodine (12 to 14 weeks), there is a risk of fetal thyroid ablation and the attendant effects on neurocognitive development. If treatment is given during pregnancy, there needs to be full disclosure. Depending on the couple's wishes, termination of pregnancy might be considered, but the limited data that are available suggest normal outcomes, if the exposure is in the first trimester. (See "Radioiodine in the treatment of hyperthyroidism", section on 'Pregnancy and breastfeeding'.)

●Thyroxine – We do not recommend the use of T4 with thionamide therapy during pregnancy. Little T4 crosses the placenta, making it more difficult to determine the minimal dose of thionamide needed to control hyperthyroidism in the mother.

History of treated hyperthyroidism

●Euthyroid women with a history of remission after prior treatment of Graves' disease with thionamides have a small risk of recurrence when they become pregnant or postpartum. In such women, thyroid function tests (TSH and free or total T4) should be measured in early pregnancy, again if hyperthyroid symptoms appear, and postpartum. Abnormal tests require confirmation and treatment depending on the underlying abnormality.

●Euthyroid women taking thyroid hormone after definitive therapy with radioiodine or surgery have a small risk of fetal hyperthyroidism from transplacental transfer of TRAb [47]. TRAb should be measured during the first trimester and, if elevated, at 18 to 22 weeks of gestation.

Measurement of TRAb is not necessary for women in remission after stopping a course of thionamides, as thyroid function tests that confirm euthyroidism indicate the absence of elevated TRAb levels. (See 'Measurement of maternal antibodies' below.)

If TRAb levels are elevated, the fetus should be monitored for hyperthyroidism. (See 'Fetal or neonatal hyperthyroidism' below.)

FETAL OR NEONATAL HYPERTHYROIDISM — Approximately 1 to 5 percent of mothers with hyperthyroidism caused by Graves' disease (active or treated) have fetuses or neonates with hyperthyroidism [48]. Although rare, it can be severe (even life threatening) and have deleterious effects on neural development.

High fetal heart rate (>160 beats/minute), fetal goiter, advanced bone age, poor growth, and craniosynostosis are manifestations of fetal hyperthyroidism. Cardiac failure and hydrops may occur with severe disease. Neonatal hyperthyroidism is reviewed separately. (See "Evaluation and management of neonatal Graves disease", section on 'Clinical manifestations'.)

Measurement of maternal antibodies — Measurement of maternal serum thyrotropin receptor antibodies (TRAb) during the third trimester (24 to 28 weeks) helps to predict which infants are at higher risk for development of fetal and neonatal Graves' hyperthyroidism [49,50].

●For pregnant women with a past history of Graves' disease previously treated with radioiodine or surgery, we measure TRAb during the first trimester and, if elevated, at 18 to 22, and again at 30 to 34, weeks of gestation. For women with an intact thyroid in remission after stopping a course of thionamides, measurement of TRAb is not necessary. (See 'History of treated hyperthyroidism' above.)

●For pregnant women with a present history of hyperthyroidism, we measure TRAb at diagnosis and, if elevated, at 18 to 22 weeks and, if still elevated, at 30 to 34 weeks of gestation.

The fetus or infant is more likely to have Graves' hyperthyroidism when the maternal value is more than three to five times the upper limit of normal. If TRAb levels are over three times the upper limit of normal, increased monitoring is necessary. (See 'Fetal monitoring' below and "Evaluation and management of neonatal Graves disease", section on 'Testing of the neonate'.)

Fetal monitoring — All fetuses of women with Graves' disease should be monitored for signs of fetal thyrotoxicosis by determination of [51]:

●Fetal heart rate

●Fetal growth rate

Ultrasound — Fetal thyroid ultrasound monitoring should be performed in pregnant women with active Graves' hyperthyroidism and/or women with serum TRAb levels greater than two to three times the upper limit of normal [49]. In one report, half of the goiters associated with fetal hyperthyroidism had central color flow Doppler versus none of the goiters associated with fetal hypothyroidism [52]. (See "Overview of the clinical utility of ultrasonography in thyroid disease", section on 'Ultrasonography of the thyroid in the fetus and neonate'.)

Fetal blood sampling — Because of the potential risk of fetal loss, we suggest not performing umbilical vein sampling routinely in pregnant women with Graves' disease. Fetal blood for thyroid function tests can be obtained by percutaneous umbilical vein sampling after 20 weeks of gestation [53,54]; however, this procedure is associated with a 1 to 2 percent risk of fetal loss. Umbilical vein sampling should be considered only when fetal goiter is present on ultrasound and there is clinical uncertainty as to whether the fetus is hyperthyroid or hypothyroid because of maternal thionamide treatment for Graves' disease.

Fetal blood sampling can provide useful clinical information when fetal thyroid function is uncertain. As an example, in one study of 24 pregnancies (26 fetuses) among 18 mothers with Graves' disease, fetal blood sampling was performed when maternal anti-TRAb levels were high or there were sonographic signs of fetal thyroid disease [53]. Five fetuses were found to be either hypothyroid (n = 3) or hyperthyroid (n = 2). In this setting, knowledge of fetal thyroid status and medical intervention can improve fetal/neonatal outcome and thus justify the risk of the procedure, which should only be performed by experienced obstetricians [53,54]. (See "Fetal blood sampling".)

Thionamides may be given to mothers, regardless of maternal thyroid status, to treat the hyperthyroid fetus. If maternal hypothyroidism develops, she should be treated with levothyroxine.

POSTPARTUM ISSUES

Breastfeeding — Given the concerns about potential propylthiouracil (PTU)-associated hepatotoxicity, we suggest methimazole rather than PTU for nursing mothers. Methimazole should be administered following a feeding in divided doses. When the maternal dose of methimazole is >20 mg daily, infants should have thyroid function tests assessed after one and three months.

PTU is less soluble than methimazole and is more bound to plasma proteins, whereas methimazole is free in serum, so that relatively more methimazole reaches the infant via breast milk [55]. Nonetheless, there is little difference in serum thyroid hormone concentrations or thyroid function in infants of mothers given moderate doses of either drug. As an example, in a study of 139 mothers taking up to 20 mg methimazole daily, thyroid function, growth, and development of their breastfed infants were normal [56]. Similar results were seen in a study of mothers taking PTU [57]. There has yet to be a report of agranulocytosis or liver disease in an infant who was nursed by a woman taking PTU or methimazole.

Radioiodine therapy for the treatment of Graves' disease is contraindicated during breastfeeding. (See "Radioiodine in the treatment of hyperthyroidism", section on 'Pregnancy and breastfeeding'.)

Relapse — Postpartum hyperthyroidism may be due to a relapse of Graves' disease or to postpartum thyroiditis. The disorders can be differentiated from one another based upon the clinical presentation, the number of months postpartum (earlier onset favors thyroiditis), thyrotropin receptor antibody (TRAb) measurement, and evaluation of serum T3-to-T4 ratio. (See "Hyperthyroidism during pregnancy: Clinical manifestations, diagnosis, and causes", section on 'Graves' disease' and "Postpartum thyroiditis", section on 'Differential diagnosis'.)

Women with Graves' disease who have been treated during pregnancy need careful monitoring during the postpartum period as they may experience an exacerbation. One approach is to measure thyroid function tests (TSH, free T4) six weeks postpartum, then every six weeks if an adjustment in thionamide dose is needed or every four months if thyroid tests remain normal [58].

In addition, women with Graves' disease who have been in remission are at risk for having a relapse during this period. This was illustrated in a study of 150 women with Graves' disease in remission after antithyroid drug therapy [59]. Of the 25 women who had a subsequent pregnancy, 21 (84 percent) experienced a relapse, compared with 70 of 125 (56 percent) who did not have a pregnancy. The risk of relapse appeared to be related to the postpartum period rather than the pregnancy itself as 20 of the 21 relapses occurred four to eight months postpartum.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Hyperthyroidism" and "Society guideline links: Thyroid disease and pregnancy".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Hyperthyroidism (overactive thyroid) (The Basics)" and "Patient education: Hyperthyroidism (overactive thyroid) and pregnancy (The Basics)")

●Beyond the Basics topics (see "Patient education: Hyperthyroidism (overactive thyroid) (Beyond the Basics)" and "Patient education: Antithyroid drugs (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Indications for treatment – Women with symptomatic and/or moderate to severe, overt hyperthyroidism due to Graves' disease, toxic adenoma, toxic multinodular goiter, or gestational trophoblastic disease require therapy for the treatment of hyperthyroidism.

Pregnant women with subclinical hyperthyroidism (low thyroid-stimulating hormone [TSH], normal free thyroxine [T4]) or total T4 and triiodothyronine (T3) <1.5 times the upper limit of normal for nonpregnant adults) and pregnant women with asymptomatic and/or mild, overt hyperthyroidism may be followed with no treatment. In women who are being monitored without therapy, we measure TSH, free T4 (if there is a trimester-specific reference range), and/or total T4 or total T3 every four to six weeks. (See 'Indications for treatment' above.)

●Treatment

•Beta blocker – Assuming there are no contraindications to its use, we suggest using a beta blocker for pregnant women with moderate to severe hyperthyroidism and hyperadrenergic symptoms (Grade 2B). We typically start with metoprolol 25 to 50 mg daily or propranolol 20 mg three to four times daily. In general, long-term treatment with beta blockers (longer than two to six weeks) should be avoided in pregnant women because of concerns regarding neonatal growth restriction, hypoglycemia, respiratory depression, and bradycardia, especially with atenolol. (See 'Control of symptoms' above and "Beta blockers in the treatment of hyperthyroidism".)

•Thionamide – For pregnant women with moderate to severe hyperthyroidism due to Graves' disease, toxic adenoma, or toxic multinodular goiter, we suggest a thionamide as our first choice of treatment (Grade 2B). (See 'Decrease thyroid hormone synthesis' above.)

We suggest using propylthiouracil (PTU) rather than methimazole in the first trimester and using methimazole for women who present in the second or third trimester (Grade 2C). Patients taking PTU during the first trimester can either switch to methimazole after 16 weeks or continue PTU throughout pregnancy. (See 'Choice of thionamide' above.)

●Monitoring and dose adjustments – Thyroid function tests (TSH and free T4 or total T4 if a trimester-specific reference range is not available for free T4) should be obtained every four weeks throughout pregnancy. If thionamides are discontinued in early pregnancy, thyroid tests should be checked weekly throughout the first trimester, then monthly. (See 'Monitoring and dose adjustments' above.)

The thionamide dose should be adjusted based on the results of the thyroid function tests to maintain serum free T4 concentrations at or just above the upper limit of normal, using a trimester-specific reference range, or total T4 and T3 (if trimester-specific normal ranges are not available) at approximately 1.5-fold above the upper limit of normal for nonpregnant patients. Serum TSH concentrations should be maintained below the reference range for pregnancy. (See 'Monitoring and dose adjustments' above.)

●Thionamide intolerance – For women with symptomatic, moderate to severe, overt hyperthyroidism who cannot tolerate thionamides because of allergy or agranulocytosis, thyroidectomy during pregnancy may be necessary (see 'Thionamide intolerance' above). Such women should be treated with a short course (7 to 10 days) of beta blockers and iodine in preparation for thyroidectomy. Radioiodine therapy for pregnant women with hyperthyroidism is absolutely contraindicated. (See "Surgical management of hyperthyroidism", section on 'Preoperative preparation' and 'Therapies not recommended' above.)

●Fetal or neonatal hyperthyroidism

•All fetuses of women with Graves' disease should be monitored for signs of fetal thyrotoxicosis by determination of fetal heart rate and assessment of fetal growth. (See 'Fetal or neonatal hyperthyroidism' above.)

•If fetal thyrotoxicosis is suspected, we suggest prenatal thyroid ultrasound to rule out fetal goiter. (See 'Fetal monitoring' above.)

•Because of the potential risk of fetal loss with fetal blood sampling, we suggest not performing this procedure routinely in pregnant women with Graves' disease. Specific indications for fetal blood sampling are outlined above. (See 'Fetal monitoring' above.)

●Postpartum considerations

•Breastfeeding – Given the concerns about potential PTU-associated hepatotoxicity, we suggest methimazole rather than PTU for nursing mothers (Grade 2C). Radioiodine therapy for the treatment of hyperthyroidism is absolutely contraindicated during breastfeeding. (See 'Breastfeeding' above.)

•Monitoring for relapse – Women with Graves' disease who have been treated before or during pregnancy need careful monitoring during the postpartum period as they may experience an exacerbation. (See 'Relapse' above.)