Hypothyroidism during pregnancy: Clinical manifestations, diagnosis, and treatment

INTRODUCTION — The evaluation and treatment of pregnant women with hypothyroidism parallels that of nonpregnant individuals but presents some unique problems. There are several important issues that must be considered when hypothyroidism occurs during pregnancy or when women with preexisting treated hypothyroidism become pregnant. The clinical manifestations, diagnosis, and treatment of hypothyroidism during pregnancy are reviewed here. Other aspects of thyroid disease during pregnancy or in women attempting pregnancy are reviewed elsewhere:

●(See "Overview of thyroid disease and pregnancy".)

●(See "Hyperthyroidism during pregnancy: Clinical manifestations, diagnosis, and causes".)

●(See "Hyperthyroidism during pregnancy: Treatment".)

●(See "Subclinical hypothyroidism in nonpregnant adults", section on 'Reproductive abnormalities' and "Subclinical hypothyroidism in nonpregnant adults", section on 'Fertility'.)

CLINICAL FEATURES

Clinical manifestations — The range of clinical symptoms of hypothyroidism during pregnancy is similar to those that occur in nonpregnant patients and may include fatigue, cold intolerance, constipation, and weight gain. Symptoms may be overlooked or attributed to the pregnancy itself as some of the symptoms of hypothyroidism are similar to those of pregnancy (although cold intolerance is not a normal clinical manifestation of pregnancy). Many patients are asymptomatic. (See "Clinical manifestations of hypothyroidism".)

Laboratory findings — To meet the increased metabolic needs during a normal pregnancy, there are changes in thyroid physiology that are reflected in altered thyroid function tests. These changes include an increase in thyroxine (T4)-binding globulin (TBG), which results in total T4 and triiodothyronine (T3) concentrations that are higher than in nonpregnant women. In addition, high serum human chorionic gonadotropin (hCG) levels, particularly during early pregnancy, result in a reduction in first trimester serum thyroid-stimulating hormone (TSH) concentrations. (See "Overview of thyroid disease and pregnancy", section on 'Thyroid adaptation during normal pregnancy'.)

Because of the changes in thyroid physiology during normal pregnancy and because there are substantial population differences in the TSH upper reference limit, thyroid function tests should be interpreted using population-based, trimester-specific reference ranges for TSH and assay method and trimester-specific reference ranges for serum free T4. (See "Overview of thyroid disease and pregnancy", section on 'Trimester-specific reference ranges'.)

●If the laboratory does not provide population and trimester-specific reference ranges for TSH, an upper reference limit of approximately 4.0 mU/L can be used.

●Trimester-specific reference ranges for free T4 should be provided with the assay kits. If not available (and particularly if free T4 values are discordant from serum TSH), measurement of total T4 may be superior to free T4 in the second and third trimesters. Total T4 levels during later pregnancy are approximately 1.5-fold higher than in nonpregnant women.

Thyroid peroxidase (TPO) antibodies are elevated in 30 to 60 percent of pregnant women with an elevated TSH [1,2]. Women who have subclinical hypothyroidism with positive TPO antibodies have a higher risk of pregnancy complications than those whose TPO antibodies are negative [3]. (See 'Subclinical hypothyroidism' below.)

DIAGNOSIS — The diagnosis of primary hypothyroidism during pregnancy is based upon the finding of an elevated serum TSH concentration, defined using population and trimester-specific TSH reference ranges for pregnant women [4]. (See "Overview of thyroid disease and pregnancy", section on 'Trimester-specific reference ranges'.)

TSH should be measured in any women with symptoms of hypothyroidism. Screening of asymptomatic women is reviewed below. (See 'Screening' below.)

For women with a TSH above the population and trimester-specific upper limit of normal (or above 4.0 mU/L when local reference ranges are not available), we also measure a free T4 (or total T4, if trimester-specific reference range for free T4 is not provided or if free T4 measurements appear discordant with TSH measurements). In addition, we agree with the American Thyroid Association (ATA) recommendation to measure thyroid peroxidase (TPO) antibodies in pregnant women with TSH >2.5 mU/L to inform treatment considerations [4]. (See 'Indications for treatment' below.)

●Overt primary hypothyroidism is defined as an elevated trimester-specific TSH concentration in conjunction with a decreased free T4 concentration (below assay normal using reference range for pregnant women).

●Subclinical hypothyroidism is defined as an elevated trimester-specific serum TSH concentration and a normal free T4 concentration.

Women with central hypothyroidism from pituitary or hypothalamic disease will not have elevated TSH concentrations during pregnancy. (See "Central hypothyroidism", section on 'Diagnosis'.)

PREGNANCY COMPLICATIONS — Hypothyroidism can have adverse effects on pregnancy outcomes, depending upon the severity of the biochemical abnormalities:

●Overt hypothyroidism

●Subclinical hypothyroidism

●Maternal hypothyroxinemia (isolated low maternal free T4)

Overt hypothyroidism — Overt hypothyroidism (elevated TSH, reduced free T4) complicating pregnancy is unusual (0.3 to 0.5 percent of screened women). Two factors contribute to this finding; some hypothyroid women are anovulatory [5], and hypothyroidism (new or inadequately treated) complicating pregnancy is associated with an increased rate of first trimester spontaneous abortion [6-8].

In continuing pregnancies, hypothyroidism has been associated with an increased risk of several complications, including [9-18]:

●Preeclampsia and gestational hypertension

●Placental abruption

●Nonreassuring fetal heart rate tracing

●Preterm delivery, including very preterm delivery (before 32 weeks)

●Low birth weight (which was likely due to preterm delivery for preeclampsia in one study [13] but not in a second study where the rate of preeclampsia was negligible [17])

●Increased rate of cesarean section

●Postpartum hemorrhage

●Perinatal morbidity and mortality

●Neuropsychological and cognitive impairment in the child

Subclinical hypothyroidism — Subclinical hypothyroidism (elevated TSH, normal free T4) is more common than overt hypothyroidism, occurring in 2.0 to 2.5 percent of screened pregnant women in the United States (iodine-sufficient region) [1,19].

Subclinical hypothyroidism in women seeking fertility is reviewed separately. (See "Subclinical hypothyroidism in nonpregnant adults", section on 'Reproductive abnormalities' and "Subclinical hypothyroidism in nonpregnant adults", section on 'Fertility'.)

Adverse pregnancy outcome — The risk of complications during pregnancy is lower in women with subclinical, rather than overt, hypothyroidism. However, in some [3,13,20-26], but not all [27], studies, women with subclinical hypothyroidism were also reported to be at increased risk for severe preeclampsia, preterm delivery, placental abruption, neonatal respiratory distress syndrome, and/or pregnancy loss compared with euthyroid women.

In a meta-analysis of individual participant data from 19 cohort studies, the risk of preterm birth was higher in women with subclinical hypothyroidism compared with euthyroidism (6.1 versus 5 percent, odds ratio [OR] 1.29, 95% CI 1.01-1.64) [28]. In an earlier systematic review and meta-analysis of 18 cohort studies, pregnant women with subclinical hypothyroidism were at higher risk for placental abruption (relative risk [RR] 2.14), pregnancy loss (RR 2.01), premature rupture of membranes (RR 1.43), and neonatal death (RR 2.58) compared with euthyroid women [29]. The risks of preterm delivery and preeclampsia did not reach or were of borderline statistical significance, respectively (RRs 1.20, 95% CI 0.97-1.50 and 1.30, 95% CI 1.00-1.68). Limitations of this analysis included variability among the studies in the upper reference limit of TSH used to define subclinical hypothyroidism and inconsistent accounting for thyroid antibody status.

Assessment of antibody status is important because women with subclinical hypothyroidism and positive anti-thyroid peroxidase (TPO) antibodies tend to have the highest risk of adverse pregnancy outcomes, and adverse outcomes occur at a lower TSH than in women without TPO antibodies [3]. In the American Thyroid Association (ATA) systematic review (ATA guidelines on thyroid disease during pregnancy), the risk of pregnancy-specific complications was apparent in TPO-positive women with TSH >2.5 mU/L but was not consistently apparent in TPO-negative women until TSH values exceeded 5 to 10 mU/L [4]. (See "Overview of thyroid disease and pregnancy", section on 'Thyroid peroxidase antibodies in euthyroid women'.)

In addition, limited data suggest that pregnancy outcome for women undergoing in vitro fertilization may be worse among those with preconception TSH levels higher than 2.5 mU/L. As an example, in one study of delivery outcomes after in vitro fertilization, gestational age and birth weight were higher for 150 deliveries where preconception TSH was <2.5 mU/L compared with 45 deliveries where TSH was >2.5 mU/L [30].

Cognitive impairment — It is uncertain if the children of women with subclinical hypothyroidism are at risk for neuropsychological impairment. Some [15,31-33], but not all [34], observational studies suggest an association between subclinical hypothyroidism in pregnancy and impaired cognitive development in children. The different outcomes in these studies may be related to differences in the degree of TSH elevation, maternal iodine status, and cognitive tests performed.

In one report of seven- to nine-year-old children, the mean intelligence quotient (IQ) score at age five years was slightly lower in 62 children whose mothers had high serum TSH concentrations (above 98^th percentile for pregnancy, mean 13.2 mU/L) during the second trimester than in 124 children of mothers who had normal serum TSH concentrations (103 versus 107, p = 0.06) [15]; 15 percent of the former had a score of 85 or lower, as compared with 5 percent of the latter. Some experts speculate that preterm delivery may explain some of the neurocognitive dysfunction (when found) in the children of women with subclinical hypothyroidism [20]. However, an analysis of maternal thyroid function at delivery of preterm infants (born ≤34 weeks) and neurodevelopmental outcome assessed at 5.5 years of age demonstrated significant decrements in general cognition and verbal and perceptual performance subscales for each mU/L increment in maternal TSH [35].

In another report, 54 children born to mothers with mean TSH 7.81 mU/L in the first trimester (all treated with levothyroxine) had smaller hippocampal volume and lower scores on memory testing, suggesting either no effect of levothyroxine or that the levothyroxine was not initiated sufficiently early [18].

In contrast to these reports, an analysis from the Avon Longitudinal Study of Parents and Children did not show an association between maternal subclinical hypothyroidism in early pregnancy and childhood performance on National Curriculum Tests administered from ages 4 to 15 years [34]. In this analysis, 4616 mothers had thyroid tests performed in the first trimester (median 10 weeks gestation), and 166 had subclinical hypothyroidism (defined as a TSH above the 97.5^th centile with a free T4 2.5^th to 97.5^th centile). The median TSH in mothers with euthyroidism and subclinical hypothyroidism was 0.97 and 3.22 mU/L, respectively.

Effect of thyroid hormone replacement — It is uncertain if thyroid hormone replacement reduces the risk of adverse pregnancy, neonatal, and/or childhood cognitive outcomes in women with subclinical hypothyroidism. The limitations of the existing trial data include initiation of levothyroxine after the first trimester, which may be too late, and the mild degree of TSH elevation in the mothers participating in the studies.

●Pregnancy outcomes – Studies assessing the benefit of levothyroxine therapy in reducing adverse pregnancy outcomes show conflicting results [36-39]. As examples:

•In a trial of 131 women with positive TPO antibodies (euthyroidism or subclinical hypothyroidism) randomly assigned to treatment with levothyroxine or no treatment, treatment with levothyroxine significantly decreased the rate of preterm delivery, particularly in women with TSH ≥4 mU/L (5.3 versus 29.4 percent in control group) [36].

•In a randomized trial that was designed to assess a case-finding versus a universal thyroid screening strategy, over 4500 women in their first trimester of pregnancy were randomly assigned to universal screening or case-finding groups (see 'Screening' below) [38]. All patients in the universal screening group and all high-risk patients in the case-finding group were tested for free T4, TSH, and TPO antibody levels in the first trimester, and those with positive TPO antibody titers were treated if the serum TSH was greater than 2.5 mU/L. Low-risk women in the case-finding group had their first trimester blood samples assayed postpartum. Overall, there was no significant difference in the total number of adverse outcomes between the case-finding and universal screening groups.

However, in a secondary analysis, low-risk women in the universal screening group who were found to have subclinical hypothyroidism (TSH greater than 2.5 mU/L and positive antibody titers) and were treated with thyroid hormone had fewer adverse outcomes (miscarriage, hypertension, preeclampsia, gestational diabetes, preterm labor, preterm delivery, and many others) than the low-risk patients in the case-finding group with subclinical hypothyroidism who were not treated (OR 0.43, 95% CI 0.26-0.70) [38].

•In a multicenter trial, 677 pregnant women (mean 16.6 weeks gestation) with subclinical hypothyroidism (median TSH 4.4 mU/L with normal free T4) were randomly assigned to levothyroxine or placebo [39]. There were no significant differences in the frequencies of preterm delivery, preeclampsia, gestational hypertension, miscarriage rate, or other maternal or fetal outcomes. There was no interaction according to TPO antibody positivity.

•In a meta-analysis of six trials, there were fewer pregnancy loss events in the group treated with levothyroxine compared with no treatment (1.3 versus 2.5 percent; RR 0.51, 95% CI 0.25-1.05) [40]. There were a small number of events overall, which reduced the precision of the analysis.

●Cognitive development – There are insufficient data to determine whether there is an effect of maternal levothyroxine treatment on neurocognitive outcomes in the children of mothers with subclinical hypothyroidism. The limitations of the existing data include initiation of levothyroxine after the first trimester, which may be too late, and the mild degree of TSH elevation in the mothers participating in the studies. As examples:

•In a randomized trial comparing screening for and treatment of thyroid dysfunction in early pregnancy with a control group (serum samples from the control group were stored and assayed after delivery), there were no differences between the two groups in the neurocognitive outcomes of the children [37,41]. Specifically, in the mothers who had tested positive for thyroid dysfunction (TSH >3.65 mU/L, serum free T4 below the 2.5^th percentile, or both), approximately half of whom received levothyroxine, there were no differences in:

-The IQ of the children at 3 or 9.5 years of age

-The proportion of children with IQ score <85 at 3 or 9.5 years of age

The mean serum TSH was 3.8 mU/L (compared with 13.2 mU/L in one observational study [15]), and treatment was initiated at a median gestational age of 13 weeks. Approximately 25 percent of children in each group did not complete psychological testing. It is uncertain if treatment earlier in gestation or testing of children at an older age would change the outcome. It is also possible that the study population included women with very mild subclinical hypothyroidism, where an effect on cognitive development would be less likely to have been observed.

•In a multicenter trial, 677 pregnant women (mean 16.6 weeks gestation) with subclinical hypothyroidism (median TSH 4.4 mU/L, normal free T4) were randomly assigned to levothyroxine or placebo [39]. There were no significant differences in neurodevelopmental (median IQ scores 97 and 94, respectively) or behavioral outcomes in the children at five years of age. There was no interaction according to TPO antibody positivity.

Although existing evidence argues against neurocognitive benefit associated with treating subclinical hypothyroidism in pregnancy, randomized trials are needed to determine whether earlier initiation of treatment (prior to 13 weeks) improves outcomes.

In a prospective cohort study that assessed the initiation of levothyroxine preconception or during the first trimester in patients with overt or subclinical hypothyroidism, there was no difference in performance on neurocognitive development testing between the groups at 6, 12, and 24 months of age [42].

Low maternal free T4 — Isolated maternal hypothyroxinemia (low T4) is defined as a maternal free T4 concentration in the lower 2.5^th to 5^th percentile of the reference range, in conjunction with a normal TSH. The effect of isolated maternal hypothyroxinemia on perinatal and neonatal outcome is unclear [2,14,27,31,43,44].

●Pregnancy outcomes – In one study, maternal serum free T4 concentrations below the 2.5^th percentile (with normal TSH) were not associated with adverse pregnancy outcomes [2]. However, in the First and Second Trimester Evaluation of Risk (FASTER) consortium, among the women with hypothyroxinemia and normal TSH (232 and 247 women in the first and second trimesters, respectively), there was an increased OR for preterm labor (1.62, 95% CI 1.00-2.62), macrosomia (1.97, 95% CI 1.37-2.83), and gestational diabetes (1.70, 95% CI 1.02-2.84) [27]. In the Generation R study and a subsequent meta-analysis, maternal hypothyroxinemia was associated with an increased risk of premature delivery (7.1 versus 5 percent in euthyroid women, OR 1.46, 95% CI 1.12-1.90) [23,28].

●Cognitive outcomes – In some studies, infants and toddlers whose mothers had reduced serum free T4 concentrations (with normal TSH) during gestation (12 to 20 weeks) had lower mean intelligence, psychomotor, or behavioral scores compared with children born to women with normal thyroid function during gestation [14,31,43-45]. As an example, in one study of 3727 mother-child pairs, the children of mothers whose free T4 was in the lowest 5 percent during the first trimester had IQ scores at six years of age that were 4.3 points lower than the children of mothers with higher free T4 concentrations [46]. Other studies have shown an increased frequency of autism and attention deficit disorder in offspring of hypothyroxinemic women [47,48].

In contrast, a case-control study that examined children at age two years born to mothers who had second trimester free T4 levels <3^rd centile versus those with free T4 levels between the 10^th and 90^th centile did not find any differences in neurocognitive development [49]. In the Avon Longitudinal Study of Parents and Children, there was also no difference in National Curriculum Test scores between children born to mothers with first trimester free T4 levels <2.5^th centile and those with free T4 levels in the 2.5^th to 97.5^th centile [34].

●Effect of thyroid hormone replacement – In two randomized trials, there was no difference in the IQ of children of mothers with low free T4 who did or did not receive T4 treatment at a median gestational age of 13 or 16 weeks [37,39]. As an example, in a multicenter trial, 526 pregnant women (mean 17.8 weeks gestation) with isolated maternal hypothyroxinemia (median free T4 0.83 ng/dL with normal TSH [median 1.5 mU/L]) were randomly assigned to levothyroxine or placebo [39]. There were no significant differences in neurodevelopmental (median IQ scores 94 and 91, respectively) or behavioral outcomes in the children at five years of age. In addition, there were no significant differences in the frequencies of preterm delivery, preeclampsia, gestational hypertension, miscarriage rate, or other maternal or fetal outcomes.

SCREENING — Because overt and subclinical hypothyroidism are associated with pregnancy complications, including pregnancy loss, and thyroid tests are widely available and easy to perform, there is interest in screening for thyroid dysfunction in asymptomatic pregnant women. The universal screening of asymptomatic pregnant women for thyroid dysfunction during the first trimester of pregnancy is controversial, however, because of insufficient data showing a benefit of thyroid hormone replacement. Thus, there is wide variation in screening practices [50-53].

Whom to screen — Because of insufficient evidence to support universal TSH screening in the first trimester, we and others [4,54,55] prefer a targeted approach to screening (ie, "case finding"). Pregnant women with any of the following are candidates for screening:

●Living in an area of moderate to severe iodine insufficiency

●Symptoms of hypothyroidism

●Family or personal history of thyroid disease

●Personal history of:

•Thyroid peroxidase (TPO) antibodies

•Goiter

•Age >30 years

•Type 1 diabetes

•Head and neck irradiation

•Recurrent miscarriage or preterm delivery

•Multiple prior pregnancies (two or more)

•Class 3 obesity (body mass index [BMI] ≥40 kg/m²)

•Infertility

•Prior thyroid surgery

•Use of amiodarone, lithium, or recent administration of iodinated radiologic contrast agents

The results of observational studies suggest that assessment of thyroid function only in women at high risk for thyroid or other autoimmune disease (targeted screening) will miss up to one-third of women with subclinical or overt hypothyroidism (TSH >3.5 to 4.2 mU/L) [52,56,57]. However, in prospective trials, universal screening compared with a targeted approach or with no screening did not improve pregnancy outcomes [37,38].

As an example, in the trial described above (over 4500 women in their first trimester of pregnancy randomly assigned to universal screening or case-finding groups, and those with positive TPO antibody titers were treated if the serum TSH was greater than 2.5 mU/L), there were no significant differences in adverse outcomes between the case-finding and universal screening groups [38]. The majority of the women in the low-risk case-finding group were euthyroid (97.9 percent), whereas hypothyroidism was found in 34 (1.9 percent) and hyperthyroidism in five (0.2 percent) women. Because these samples were assayed postpartum, these women were not treated. The case-finding approach missed 34 of 54 hypothyroid women. (See 'Effect of thyroid hormone replacement' above.)

In another randomized trial, antenatal screening (median gestational age 12 weeks) and maternal treatment of subclinical hypothyroidism at a mean of 13 weeks did not result in improved pregnancy-related outcomes or cognitive function in children at three years of age compared with controls in whom blood was drawn and stored for testing after delivery [37]. (See 'Effect of thyroid hormone replacement' above.)

Limited data suggest that universal screening may be more cost effective than not screening [58,59]. One analysis found that universal screening compared with risk-based screening resulted in an incremental cost-effectiveness ratio of USD $7258 per quality-adjusted life-year [59].

Approach to screening — In women who meet the case-finding criteria, we suggest measurement of serum TSH during the first trimester as the screening test for hypothyroidism:

●If the serum TSH is between the trimester-specific lower limit of normal and 2.5 mU/L, most women require no further testing.

However, in women at particularly high risk for developing hypothyroidism during pregnancy (post-radioiodine treatment, post-hemithyroidectomy, history of exposure to high-dose irradiation of the head or neck region), we reassess TSH during pregnancy (eg, approximately every four weeks during the first trimester, and then once during each of the second and third trimesters).

●If the serum TSH is >2.5 mU/L, we measure TPO antibodies.

An increased rate of fetal loss and premature delivery has been reported in women with high serum anti-TPO antibody concentrations, and adverse outcomes occur at a lower TSH than in women without TPO antibodies. Therefore, the presence of TPO antibodies may be useful for making treatment decisions in women with borderline thyroid function tests (eg, TSH 2.5 to 4.0 mU/L) and in predicting the development of hypothyroidism and the risk of miscarriage and postpartum thyroid dysfunction. (See 'Indications for treatment' below and "Overview of thyroid disease and pregnancy", section on 'Thyroid peroxidase antibodies in euthyroid women' and "Postpartum thyroiditis".)

●If the TSH is >4 mU/L, we suggest measurement of free T4 to determine the degree of hypothyroidism.

TREATMENT — Our approach to the treatment of pregnant women outlined below is largely consistent with the Guidelines of the American Thyroid Association (ATA) and the Endocrine Society for the Diagnosis and Management of Thyroid Disease During Pregnancy and Postpartum [4,54].

Patients not currently taking thyroid hormone — (algorithm 1)

Indications for treatment

●TSH >4 mU/L

•Free T4 below the reference range, TPO antibodies positive or negative – All pregnant women with newly diagnosed, overt hypothyroidism (TSH above trimester-specific normal reference range [or above 4.0 mU/L if trimester-specific range unavailable]) with low free T4 should be treated with thyroid hormone (levothyroxine, T4) (algorithm 1). (See 'Levothyroxine initial dosing' below.)

•Free T4 within the reference range, TPO antibodies positive or negative – Because maternal euthyroidism is potentially important for normal fetal cognitive development, we and others [54] suggest treatment of pregnant women with newly diagnosed subclinical hypothyroidism (TSH above trimester-specific normal reference range [or above 4.0 mU/L if trimester-specific range unavailable], with normal free T4), regardless of thyroid peroxidase (TPO) antibody status (algorithm 1). The data assessing treatment with T4 in this subgroup of women are conflicting and limited by variability in the TSH criteria used to define hypothyroidism and the late initiation of thyroid hormone treatment (often late in the first trimester). (See 'Subclinical hypothyroidism' above.)

This approach differs slightly from the ATA guidelines (reviewed at the end of this section), in which treatment recommendations are based on TPO antibody status [4].

●TSH 2.6 to 4 mU/L – For pregnant women with TSH in this range, we individualize the decision to treat based upon patient characteristics, values, and preferences. Some experts, including one editor of this topic, do not treat euthyroid pregnant women. Other experts, including the author and another editor of this topic, offer treatment to selected pregnant women based on the presence or absence of TPO antibodies, which have been associated with adverse pregnancy outcomes (eg, early pregnancy loss) (algorithm 1). (See "Overview of thyroid disease and pregnancy", section on 'Pregnancy outcomes'.)

•TPO antibodies positive – For pregnant women with a TSH between 2.6 mU/L and the upper limit of the trimester-specific reference range (or 4.0 mU/L if trimester-specific range unavailable), positive TPO antibodies, and a history of recurrent miscarriage, many experts (including the author and one editor of this topic) offer treatment with T4 (50 mcg daily) to those who prefer this intervention. This is based on weak evidence in view of conflicting data regarding the efficacy of T4 for reducing the risk of miscarriage. However, carefully monitored thyroid hormone treatment is safe. (See "Overview of thyroid disease and pregnancy", section on 'Effect of T4 treatment'.)

In the absence of a history of recurrent miscarriage, some experts (including one editor of this topic) also offer T4 (50 mcg daily) to those who prefer this intervention (based on weak evidence).

If a decision is made not to treat, TSH should be reassessed every four weeks during the first trimester and once each in the second and third trimester to monitor for the development of hypothyroidism. If TSH rises above the population and trimester-specific upper limit of normal (approximately 4 mU/L), we begin treatment with T4.

The management of women with TPO antibodies and normal thyroid function is reviewed in more detail elsewhere. (See "Overview of thyroid disease and pregnancy", section on 'Thyroid peroxidase antibodies in euthyroid women' and "Recurrent pregnancy loss: Evaluation" and "Recurrent pregnancy loss: Management", section on 'Thyroid dysfunction and diabetes mellitus'.)

•TPO antibodies negative – For pregnant women with a TSH between 2.6 mU/L and the upper limit of the trimester-specific reference range (or 4.0 mU/L if trimester-specific range unavailable), and negative TPO antibodies, we do not treat with T4.

For women at high risk for developing hypothyroidism (eg, history of radioiodine treatment, hemithyroidectomy, or exposure to high-dose irradiation of the head and neck), we monitor for the development of hypothyroidism by reassessing TSH approximately every four weeks during the first trimester and once during each of the second and third trimesters. If TSH rises above the population and trimester-specific upper limit of normal (approximately 4 mU/L), we begin treatment with T4.

●TSH between the trimester-specific lower limit of normal and 2.5 mU/L – These women are euthyroid and do not require T4 treatment.

●Low free T4, normal TSH (maternal hypothyroxinemia) – We do not typically treat pregnant women with isolated hypothyroxinemia (low free T4, normal TSH). (See 'Low maternal free T4' above.)

The 2017 ATA guidelines base their treatment recommendations on TPO antibody status [4]. They recommend measurement of TPO antibodies in pregnant women with TSH >2.5 mU/L and treatment as follows:

●Positive TPO antibodies – Thyroid hormone should be considered if TSH is above 2.5 mU/L and should be initiated if TSH is above the population and trimester-specific upper limit of normal (approximately 4.0 mU/L).

●Negative TPO antibodies – Thyroid hormone should be considered if the TSH is above population and trimester-specific upper limit of normal but <10 mU/L and should be initiated if the TSH is >10 mU/L.

●Maternal hypothyroxinemia – The ATA does not suggest treatment of pregnant women with isolated hypothyroxinemia (low free T4, normal TSH).

Levothyroxine initial dosing — The treatment of choice for correction of hypothyroidism in pregnancy is the same as in nonpregnant patients: synthetic levothyroxine (T4). Several formulations of T4 are available. Because there may be subtle differences in bioavailability between T4 formulations, some endocrinologists feel that it is preferable to stay with the same formulation whenever possible. When using generic preparations, the manufacturer can be identified from the prescription label, and the patient may request refills from the same generic pharmaceutical company.

The goal of T4 replacement in pregnancy is to restore euthyroidism as soon as possible. General dosing guidance is as follows:

●TSH >4 mU/L (or above population and trimester-specific upper limit of normal), with low free T4 (using assay method and trimester-specific reference range) – Close to full replacement dose (approximately 1.6 mcg/kg body weight per day)

●TSH >4 mU/L, with normal free T4 – Intermediate dose (approximately 1 mcg/kg per day)

●TSH 2.6 to 4 mU/L – If a decision has been made to treat euthyroid women with TPO antibodies, low dose (typically 50 mcg daily)

T4 should be taken on an empty stomach, ideally an hour before breakfast, but few patients are able to wait a full hour.

Monitoring and dose adjustments — After initiation of T4 therapy, the patient should be reevaluated and serum TSH measured in four weeks.

The goal is to maintain TSH in the lower half of the trimester-specific reference range. If not available, a goal TSH of <2.5 mU/L is reasonable.

If the TSH remains above the normal trimester-specific reference range, the dose of T4 can be increased by 12 to 25 mcg/day. TSH should be measured every four weeks during the first half of pregnancy because dose adjustments are often required. TSH can be monitored less often (at least once each trimester) in the latter half of pregnancy, as long as the dose is unchanged.

Overtreatment should be avoided. Overtreatment with levothyroxine during pregnancy has been associated with an increase in preterm delivery (OR 2.14, 95% CI 1.51-2.78) [60]. In addition, one study suggested an association between overtreatment during pregnancy (free T4 >97^th percentile with average TSH 0.05 to 0.08 mU/L) and behavioral difficulties in the children [61].

Postpregnancy adjustments — Since the criteria for treating pregnant women differ from the criteria from treating nonpregnant women, it is not always necessary to continue levothyroxine after delivery. In one study, 75 percent of women with subclinical hypothyroidism during pregnancy had normal thyroid function five years postpartum [62]. Because overt hypothyroidism may interfere with milk production, it may be prudent to delay assessment until the completion of breastfeeding. Unless another pregnancy is imminent, however, the majority of women who were started on levothyroxine for TSH between 2.5 and 4.0 mU/L do not need to continue levothyroxine treatment.

Preexisting treated hypothyroidism

Goal preconception TSH — Women with preexisting hypothyroidism who are planning to become pregnant should optimize their thyroid hormone dose preconception. The goal preconception serum TSH level is between the lower reference limit and 2.5 mU/L [4,54]. However, some experts prefer a lower preconception TSH level (<1.2 mU/L).

Approximately 50 to 85 percent of women with preexisting hypothyroidism need more T4 during pregnancy [6,63-65]. In one study, only 17 percent of women with preconception TSH values <1.2 mU/L required a dose increase during the subsequent pregnancy, compared with 50 percent of women with preconception TSH levels between 1.2 and 2.4 mU/L [66]. Preconception counseling is important in this regard. Studies have shown that approximately 30 percent of women taking levothyroxine have a serum TSH >4 mU/L when they present for their first prenatal visit [67]. In such women, serum TSH of 4.5 to 10 compared with <2.5 mU/L at the time of presentation is a predictor of miscarriage (odds ratio [OR] 1.80, 95% CI 1.03-3.14) [68]. The risk of miscarriage was even higher in women with TSH >10 mU/L at presentation (OR 3.95, 95% CI 1.87-8.37).

Early dose adjustments — Given that T4 dose requirements may increase during pregnancy in women with preexisting hypothyroidism, hypothyroid women who are newly pregnant should preemptively increase their levothyroxine dose by approximately 30 percent and notify their clinician promptly. We typically accomplish this by increasing the dose from once-daily dosing to a total of nine doses per week (double the daily dose two days each week). Further dose changes are made based upon serum TSH concentrations measured every four weeks until the TSH becomes normal. Using such an approach, only 2 of 25 women in one randomized trial had TSH values greater than 5 mU/L during pregnancy, although eight women had TSH values greater than 2.5 mU/L during the first trimester or greater than 3.0 mU/L during the second or third trimesters, and two had TSH values <0.1 mU/L [69].

Another approach is to measure serum TSH as soon as pregnancy is confirmed, then again four weeks later, four weeks after any change in the dose of T4, and at least once each trimester [63]. The dose should be adjusted as needed every four weeks to achieve a normal TSH level. (See "Treatment of primary hypothyroidism in adults", section on 'Initial monitoring and dose adjustments'.)

Dose requirements may increase by as much as 50 percent during pregnancy, and the increase occurs as early as the fifth week of gestation as illustrated by the following:

●In a prospective study of 20 pregnancies in 19 hypothyroid women in whom serum TSH was measured every two weeks in the first trimester and then every four weeks thereafter, a T4 dose increase (on average by 47 percent) was necessary in 17 of the 20 pregnancies [64]. Although the median onset of the dose modification occurred at eight weeks with a plateau at 16 weeks of gestation, some women required an increase in dose as early as the fifth week. The higher dose was required until delivery.

●In a retrospective, population-based analysis of 950 pregnancies in hypothyroid women, 60 percent required a levothyroxine dose increase (34 percent during the first trimester) [65].

Unlike normal women, those with preexisting hypothyroidism or subclinical hypothyroidism are unable to increase thyroidal T4 and T3 secretion. This is especially true for women with thyroid cancer who have received radioiodine therapy [64] or patients with postablative or surgical hypothyroidism for Graves' disease or goiter [70]. Several factors have been thought to be responsible for the increased T4 requirement during pregnancy. They include weight gain and increased T4 pool size, high serum thyroxine-binding globulin (TBG) concentrations, placental deiodinase activity (which increases clearance of T4), transfer of T4 to the fetus, and reduced gastrointestinal absorption due to iron in prenatal vitamins [63].

Monitoring — The T4 dose should be reduced to prepregnancy levels after delivery, but serum TSH should be measured four to six weeks later to confirm that the reduction was appropriate [4,63,71].

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: Hypothyroidism" 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: Congenital hypothyroidism (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Clinical features – The range of clinical manifestations of hypothyroidism during pregnancy is similar to those that occur in nonpregnant patients and may include fatigue, cold intolerance, constipation, and weight gain. Symptoms may be overlooked or attributed to the pregnancy itself. Many patients are asymptomatic. (See 'Clinical features' above.)

●Diagnosis – The diagnosis of overt primary hypothyroidism during pregnancy is based upon the finding of an elevated population and trimester-specific thyroid-stimulating hormone (TSH) concentration (or above 4.0 mU/L when local reference range is not available) in conjunction with a decreased free thyroxine (T4) concentration (below assay normal using reference range for pregnant women). Subclinical hypothyroidism is defined as an elevated population and trimester-specific serum TSH concentration and a normal free T4 concentration. (See 'Diagnosis' above.)

●Pregnancy complications – Hypothyroidism can have adverse effects on the pregnancy, depending upon the biochemical severity of the hypothyroidism. (See 'Pregnancy complications' above.)

●Screening for hypothyroidism – The universal screening of asymptomatic pregnant women for thyroid dysfunction during the first trimester of pregnancy is controversial. We suggest a targeted approach (case finding) rather than universal screening (Grade 2C). We favor screening pregnant women if they are from an area of moderate to severe iodine insufficiency; have symptoms of hypothyroidism; have a family or personal history of thyroid disease or have a personal history of goiter, thyroid peroxidase (TPO) antibodies, type 1 diabetes or other autoimmune disorder, head and neck radiation, recurrent miscarriage or preterm delivery, class 3 obesity, infertility, multiple prior pregnancies (>2), use of amiodarone, lithium, or recent iodinated radiocontrast; or age >30 years. (See 'Whom to screen' above.)

In women who meet the case-finding criteria, we measure serum TSH during the first trimester as the screening test for hypothyroidism. (See 'Approach to screening' above.)

•If the TSH is above 2.5 mU/L, we measure TPO antibodies

•If the TSH is above the population and trimester-specific upper limit of normal (or >4 mU/L if local reference range is not available), we also measure free T4 to determine the degree of hypothyroidism.

●Treatment

•Indications for treatment – All pregnant women with newly diagnosed, overt hypothyroidism (TSH above population and trimester-specific normal reference range [or above 4.0 mU/L when local reference range is not available] with low free T4) should be treated with thyroid hormone (levothyroxine, T4). In addition, we suggest initiating T4 replacement in pregnant women with newly diagnosed subclinical hypothyroidism (TSH above population and trimester-specific normal reference range [or above 4.0 mU/L] with normal free T4) (algorithm 1) (Grade 2B). (See 'Indications for treatment' above and 'Subclinical hypothyroidism' above.)

For pregnant women with a TSH between 2.6 and 4 mU/L, we individualize the decision to treat based upon patient characteristics, values, and preferences. Some experts, including the author and one editor of this topic, offer T4 treatment (50 mcg daily) to patients with TPO antibodies who have a history of miscarriage and who prefer this intervention (algorithm 1). In the absence of a history of recurrent miscarriage, some experts (including one editor of this topic) also offer T4 (50 mcg daily) to TPO-positive women who prefer this intervention. Other experts, including the other editor of this topic, do not treat TPO-positive, euthyroid (TSH ≤4 mU/L) pregnant women. (See 'Indications for treatment' above and "Overview of thyroid disease and pregnancy", section on 'Thyroid peroxidase antibodies in euthyroid women' and "Recurrent pregnancy loss: Management", section on 'Thyroid dysfunction and diabetes mellitus'.)

In pregnant women who are not treated with thyroid hormone and who are at particularly high risk for developing hypothyroidism during pregnancy (TPO antibody-positive, post-radioiodine treatment, post-hemithyroidectomy, history of exposure to high-dose irradiation of the head or neck region), we reassess TSH during pregnancy (eg, approximately every four weeks during the first trimester, and then once during each of the second and third trimesters). If TSH rises above the population and trimester-specific upper limit of normal (approximately 4 mU/L), we begin treatment with T4. (See 'Approach to screening' above and 'Indications for treatment' above.)

•Initial dosing, monitoring, and dose adjustments – Patients with overt hypothyroidism should be started on close to full replacement doses (1.6 mcg/kg body weight per day), while patients with subclinical hypothyroidism may become euthyroid with lower doses and can therefore be started on approximately 1 mcg/kg daily. TSH should be measured every four weeks during the first half of pregnancy because dose adjustments are often required. The goal of treatment is to maintain TSH in the lower half of the trimester-specific reference range (or approximately <2.5 mU/L). (See 'Levothyroxine initial dosing' above and 'Monitoring and dose adjustments' above.)

●Preexisting treated hypothyroidism

•Goal preconception TSH – Women with preexisting hypothyroidism who are planning to become pregnant should optimize their thyroid hormone dose preconception. The goal preconception serum TSH level is between the lower reference limit and 2.5 mU/L. If possible, women already taking levothyroxine should have a normal serum TSH (ie, <2.5 mU/L) prior to becoming pregnant. (See 'Goal preconception TSH' above.)

•Early dose adjustments – T4 dose requirements may increase during pregnancy in women with preexisting overt or subclinical hypothyroidism. For treated hypothyroid women who are newly pregnant, we suggest preemptively increasing their levothyroxine dose at the time of the positive pregnancy test (Grade 2B). We typically accomplish this by increasing the dose from once-daily dosing to a total of nine doses per week (double the daily dose two days each week). (See 'Early dose adjustments' above.)

An alternative to preemptively increasing the dose is to measure serum TSH as soon as pregnancy is confirmed, then again four weeks later, four weeks after any change in the dose of T4, and at least once each trimester. The dose should be adjusted as needed every four weeks to achieve a normal TSH level, using a trimester-specific reference range. (See "Treatment of primary hypothyroidism in adults", section on 'Initial monitoring and dose adjustments'.)