Current guideline-based management of thyroid dysfunction in pregnancy and puerperium, from trimester-specific reference ranges to levothyroxine dosing, from thionamide selection to TRAb-based fetal monitoring.
⚕️ Target audience: This clinical review is intended for physicians and healthcare professionals; It is a summary of current literature and guidelines for educational purposes. Clinical decisions should be made based on patient-specific evaluation and applicable institutional protocols.
There are two facts that make thyroid management during pregnancy a separate topic: normal pregnancy itself changes thyroid function tests — the result interpreted at non-pregnancy intervals can lead to misdiagnosis — and the treatment decision concerns both organisms; Both the obstetric consequences of undertreatment and the fetal effects of overtreatment are real. This review discusses hypothyroidism and hyperthyroidism during pregnancy, thyroid storm and postpartum thyroiditis in a single framework.
The increased metabolic demand during pregnancy creates two fundamental changes in thyroid physiology. Under the influence of estrogen, the concentration of thyroxine binding globulin (TBG) approximately doubles; The result of this is that total T4 and total T3 levels are higher than in the non-pregnant population — free hormone levels are not directly affected by this increase. The second change is due to hCG: due to the homology between hCG and the beta subunits of TSH, hCG has a weak thyrotropic effect, causing a physiological fall in first-trimester TSH, particularly when hCG peaks at 10–12 weeks (Lee & Pearce 2022, Nat Rev Endocrinol).
Because the TSH upper reference limit varies significantly between populations, thyroid function tests should be interpreted with population-based, trimester-specific, and assay-specific reference ranges; Both ATA 2026 and ACOG adopt this approach (ATA 2026 Guidelines, Thyroid 2026; ACOG Practice Bulletin 223, 2020). The problem is that most laboratories do not publish the range for their pregnant population. The practical solution used in such cases is to take approximately 4.0 mU/L as the upper limit of TSH; a range of 0.1–4.0 mIU/L can be used in early pregnancy (typical range in the non-pregnant adult is 0.5–4.5 mIU/L) (Lee & Pearce 2022).
In what context is 2.5 mU/L valid: 2.5 mU/L in pregnancy is not a diagnostic threshold. This value appears in two separate places: the preconception TSH target in a woman with hypothyroidism (between the lower reference limit and 2.5 mU/L) and the treatment target in a pregnant woman who has started levothyroxine (the lower half of the trimester range; if there is no range, <2.5 mU/L is reasonable). The diagnostic threshold is the trimester-specific upper limit or ~4.0 mU/L (ATA 2026).
Direct free T4 immunoassays may be misleading in pregnancy due to binding protein changes; Finding free T4 at or below the lower limit of the nonpregnant reference range along with normal TSH is not uncommon, especially in the second and third trimesters (Lee 2009, Am J Obstet Gynecol). If there is no trimester-specific free T4 range—and especially if free T4 and TSH are discordant—measuring total T4 or free T4 index is superior to direct free T4 assay. In late pregnancy, total T4 levels are approximately 1.5 times non-pregnant values (Lee & Pearce 2022).
| Test | Situation during pregnancy | Rule of thumb |
|---|---|---|
| TSH | It decreases physiologically in the 1st trimester due to the effect of hCG | Trimester-specific interval; otherwise the upper limit is ~4.0 mU/L (ATA 2026; ACOG PB 223, 2020) |
| Free T4 | May be misleading due to increased TBG and assay limitations | Trimester-specific kit range; Otherwise, total T4 or free T4 index is preferred (Lee 2009, Am J Obstet Gynecol) |
| Total T4 / total T3 | Increases due to increase in TBG | During weeks 7–16, the nonpregnant range is increased by 5% per week; after week 16, 150% of the upper limit is used (Lee & Pearce 2022) |
| TPO antibody | Indication of thyroid autoimmunity | Positive in 30–60% of pregnant women with elevated TSH (ATA 2026); indication for measurement is controversial (see below) |
| TRAb | Specific to Graves' disease | Positive in more than 95% of Graves' cases (UpToDate, hyperthyroidism, 2026); Used to predict fetal risk during pregnancy |
Overt hypothyroidism is defined as low free or total T4 with TSH above the population and trimester-specific range; In most patients with overt primary hypothyroidism, TSH is >10 mU/L, frequently >20 mU/L. Subclinical hypothyroidism is the combination of persistently high trimester-specific TSH and normal free/total T4. Isolated maternal hypothyroxinemia is free T4 with normal TSH in the lower 2.5–5% percentile of the reference range (UpToDate, hypothyroidism, 2026).
In screened pregnant women, overt hypothyroidism has a frequency of 0.3–0.5%, and subclinical hypothyroidism has a frequency of 2.0–2.5% in the iodine sufficient region (USA) (UpToDate, hypothyroidism, 2026). A significant portion of mild TSH elevation is temporary: in observational studies, in 50–75% of women with elevated TSH in early pregnancy, the value returns to normal after two weeks. In a study involving 1466 pregnant women in Denmark, only 44 (49%) of 89 women with high TSH levels in the first trimester continued to have high levels after two weeks; If baseline TSH was >5 mU/L, ~65% were permanent, and if >6 mU/L, 80% were permanent (UpToDate, hypothyroidism, 2026). Therefore, with mild TSH elevation, the test is repeated before starting treatment.
Untreated overt hypothyroidism has been associated with preeclampsia and gestational hypertension, placental abruption, nonreassuring fetal heart rate, preterm birth (including very premature birth before 32 weeks), low birth weight, increased cesarean section, postpartum hemorrhage, perinatal morbidity-mortality, first trimester spontaneous abortion, and neuropsychological impairment in the child. The increase in risk in subclinical hypothyroidism is much smaller: in meta-analyses the absolute risk difference in complications compared to euthyroid women is in the range of 1–5%, and the risk appears proportional to the degree of TSH elevation (UpToDate, hypothyroidism, 2026).
| Result | Data | Source |
|---|---|---|
| Preterm birth | 6.1% vs 5.0%; OR 1.29 (95% CI 1.01–1.64) | IPD meta-analysis of 19 cohorts (Korevaar 2019, JAMA) |
| Preeclampsia | 3.6% vs 2.1%; OR 1.53 (95% CI 1.09–2.15) | IPD meta-analysis (Toloza 2022, Lancet Diabetes Endocrinol) |
| Preterm birth — TSH level | TSH ≤4–6 mU/L: 5.4% >6 mU/L: 7.8% >10 mU/L: 11.4% | Danish retrospective cohort |
| Preeclampsia — TSH level | TSH <2.5 mU/L: 2.3% >6 mU/L: 6.2% | Danish retrospective cohort |
| Pregnancy loss (with levothyroxine) | 1.3% vs 2.5%; RR 0.51 (95% CI 0.25–1.05) | Meta-analysis of 6 randomized trials (UpToDate, hypothyroidism, 2026) — small number of events |
The picture is unclear in terms of the child's neurocognitive development: while some observational studies show a relationship (average IQ at age five in children of mothers with second trimester TSH average of 13.2 mU/L, mean IQ 103 vs 107, p=0.06), the Avon Longitudinal Study of Parents and Children analysis found no relationship between subclinical hypothyroidism in early pregnancy and exam success between the ages of 4 and 15. (Nelson 2018, BMJ ) — median TSH in this cohort was 3.22 mU/L, which was very slightly elevated.
Randomized evidence has two common limitations: treatment was started after the first trimester in most studies (the period when the potential benefit is thought to be greatest may have been missed) and the TSH elevation in the women included was very mild. On the benefit side: In 131 TPO antibody-positive women, levothyroxine in the first trimester reduced preterm birth, particularly in those with TSH ≥4 mU/L (5.3% vs 29.4%); In a subsequent study of the same group in women with TPO negative, TSH >4 mU/L subclinical hypothyroidism, preterm birth was 7.3% vs 19% (HR 0.38; 95% CI 0.15–0.98); In those with recurrent pregnancy loss, levothyroxine reduced pregnancy loss but did not improve any other outcomes (Leng 2022). On the no-benefit side: there was no difference in preterm birth, preeclampsia, or miscarriage between levothyroxine and placebo in 677 pregnant women with subclinical hypothyroidism (median TSH 4.4 mU/L) enrolled at a mean of 16.6 weeks; Median IQ at age five in children was 97 vs 94 (Casey 2017, N Engl J Med), and a meta-analysis of nine randomized studies and 13 cohorts found no benefit.
Monitored levothyroxine therapy is safe during pregnancy; however, overtreatment is not innocent—it has been associated with an increase in preterm birth (OR 2.14; 95% CI 1.51–2.78) (UpToDate, hypothyroidism, 2026).
Should TPO antibody be measured in pregnant women with high TSH? Here experts differ. With the ATA 2026 guideline, an author and an editor at UpToDate adopt the measurement: high TPO antibody levels have been reported with an increased risk of fetal loss and preterm birth, antibody positivity may be useful in predicting the development of hypothyroidism, risk of miscarriage, and postpartum thyroid dysfunction. The other editor of the same issue does not measure it on the grounds that there is not enough evidence that knowing antibody status and treating it accordingly improves pregnancy outcomes. The prognostic weight of the antibody is nevertheless undeniable: the risk of developing overt hypothyroidism in antibody positives is 2.4–7%, while in negatives it is 0.1–0.2%; the same rates for subclinical hypothyroidism are 1.9–29% and 2–5.7% (Osinga 2024, Thyroid).
| Finding | Approach | Starting dose |
|---|---|---|
| TSH >10 mU/L (overt or subclinical) | Levothyroxine is started; In most cases, the elevation is permanent. If TSH is >10, it can be started without waiting for the free T4 result. | Near complete replacement if obvious: ~1.6 mcg/kg/day (ATA 2026) |
| TSH >4 – ≤10 mU/L, low free T4 (overt) | Confirm by repeating TSH and free T4 within one week; If confirmed, treatable. | ~1.6 mcg/kg/day |
| TSH >4 – ≤10 mU/L, free T4 normal (subclinical) | Post-verification decision is individualized — institutions and experts are separated (see below) | If treatment decision is made, intermediate dose: ~1 mcg/kg/day (ATA 2026; ASRM 2024 against treatment) |
| Low free T4, normal TSH (isolated hypothyroxinemia) | Typically not treated; no evidence of benefit | — |
| TSH normal again, TPO antibody positive | No treatment; TSH every 4–6 weeks until mid-pregnancy, then once in the 3rd trimester | — |
Guidelines differ on subclinical hypothyroidism. The UpToDate author and one section editor favor initiating levothyroxine as soon as possible after confirmation, reasoning that the potential benefit is greatest in the first trimester (potential reduction in preterm birth and recurrent pregnancy loss), and use shared decision-making when diagnosed in the second or third trimester. Another editor of the same topic concurs with the American Society for Reproductive Medicine and does not treat subclinical hypothyroidism in pregnancy due to insufficient evidence of benefit (ASRM, Fertil Steril 2024). In untreated patients, TSH is repeated every 4–6 weeks until mid-pregnancy, and then once in the third trimester.
Levothyroxine is taken on an empty stomach, ideally one hour before breakfast. TSH is measured four weeks after starting treatment. The goal is to maintain TSH in the lower half of the trimester-specific reference range; If there is no interval, a trimester-specific lower limit of 2.5 mU/L is reasonable. If TSH remains above the range, the dose is increased by 12–25 mcg/day (UpToDate, hypothyroidism, 2026). Since dose adjustment is required frequently in the first half of pregnancy, TSH is measured every four weeks; As long as the dose does not change in the second half, at least one measurement in each trimester is sufficient (ATA 2026).
In the hypothyroid woman planning pregnancy, the dose is optimized preconception; The target serum TSH is between the lower reference limit and 2.5 mU/L. This goal has clinical relevance: approximately 30% of women taking levothyroxine present to the first prenatal visit with a TSH >4 mU/L, and a TSH of 4.5–10 mU/L at presentation is predictive of miscarriage compared with <2.5 mU/L (OR 1.80; 95% CI 1.03–3.14); The risk is higher if TSH is >10 mU/L (OR 3.95; 95% CI 1.87–8.37) (UpToDate, hypothyroidism, 2026).
Levothyroxine requirement during pregnancy increases in 50–85% of patients; the increase may begin as early as the fifth week, with a median onset of the eighth week and plateauing around the 16th week (Alexander 2004, NEJM). Therefore, the recommended approach is to increase the preemptive dose as soon as the pregnancy test becomes positive:
The average dose increase was 15% in Hashimoto's and 50% in athyreotic patients in one study (Alexander 2004, NEJM). An alternative approach is to measure TSH as soon as pregnancy is confirmed without a preemptive increase, repeat four weeks later and four weeks after each dose change, and check at least every trimester. After delivery, the dose is reduced to pre-pregnancy levels and confirmed by TSH after 4–6 weeks (ATA 2026).
Should levothyroxine be continued postpartum? Since treatment criteria in pregnant women are different from non-pregnant women, it is not always necessary to continue levothyroxine started for subclinical hypothyroidism after birth. 75% of women with subclinical hypothyroidism during pregnancy had normal thyroid function five years later (UpToDate, hypothyroidism, 2026). Because overt hypothyroidism may affect milk production, it is prudent to postpone evaluation until breastfeeding is completed and continue treatment until then.
Screening for thyroid dysfunction in the first trimester in asymptomatic pregnant women is controversial; There is insufficient data to suggest that universal or targeted screening and treatment of maternal subclinical hypothyroidism improve outcomes at the population level. The approach in place is risk-based case finding — this is the consensus of the ATA 2026 guidance, ACOG Practice Bulletin 223 (2020) and the UpToDate author (UpToDate's rating for this recommendation: Grade 2C).
Risk factors defined by the ATA 2026 guideline: living in an area of moderate to severe iodine deficiency without using iodine-containing supplements or iodized salt; family or personal history of thyroid disease; history of subclinical hypothyroidism or postpartum thyroiditis; TPO antibody positivity; goiter; autoimmune disease (type 1 diabetes, celiac); high dose irradiation to the head and neck area; radioiodine therapy; previous thyroid surgery; history of recurrent miscarriage or preterm birth; infertility; use of medications associated with thyroid dysfunction (amiodarone, lithium, recently iodinated contrast); Turner or Down syndrome (ATA 2026).
The strongest objection to this approach comes from the list itself: in a meta-analysis of 25 prospective cohorts (65,559 participants) risk factors captured only 59% of women with hypothyroidism (Osinga 2024, Thyroid)—the ATA updated the 2026 list based on this data. Similarly, in a study that randomized more than 4,500 pregnant women to universal screening or case finding, case finding missed 34 of 54 hypothyroid women; however, there was no significant difference in overall adverse outcome between the two groups, with only secondary analysis showing fewer adverse outcomes in women with low-risk subclinical hypothyroidism detected and treated in the universal screening group (OR 0.43; 95% CI 0.26–0.70) (Negro 2010, J Clin Endocrinol Metab ). On the cost-effectiveness side, one analysis attributed an incremental cost-effectiveness ratio of $7258 per quality-adjusted life year to universal screening compared to risk-based screening (UpToDate, hypothyroidism, 2026).
If screening is to be performed, serum TSH is measured in the first trimester; If the value is in the trimester-specific range or ≤4.0 mU/L, most women do not require further testing. If there are autoimmunity risk factors (type 1 diabetes, celiac, history of postpartum thyroiditis), TPO antibody can be added. In women with a history of radioiodine therapy, hemithyroidectomy, or high-dose irradiation of the head and neck, repeat TSH is recommended approximately every four weeks during the first trimester (ATA 2026).
Overt hyperthyroidism (low TSH, high free T4 and/or T3) occurs in 0.1–0.4% of pregnancies. Two causes predominate: Graves' disease (0.1–1% of all pregnancies) and hCG-mediated hyperthyroidism—especially gestational transient thyrotoxicosis (1–3% of pregnancies). Silent or subacute thyroiditis, toxic adenoma, toxic multinodular goiter, and factitious thyrotoxicosis are less common during pregnancy.
The first step in the management of hyperthyroidism in pregnancy is to distinguish the condition that requires treatment (Graves) from the condition that does not (hCG-mediated); This distinction can usually be made with clinical findings and laboratory.
| Feature | Graves' disease | hCG-mediated (GTT / hyperemesis) |
|---|---|---|
| Frequency | 0.1–1% of pregnancies | GTT 1–3%; hyperemesis gravidarum 0.1–0.2% of pregnancies (UpToDate, hyperthyroidism, 2026) |
| Timing | During pregnancy; Typically eases in the 3rd trimester | Occurs with the hCG peak at 10–12 weeks and subsides at 14–18 weeks |
| Goiter / ophthalmopathy | Goiter is common; ophthalmopathy Graves specific | Goiter is not a classic finding (may be found); There is no ophthalmopathy |
| Vomiting | Not deterministic | is prominent in the hyperemesis chart |
| Hormone level | Free T4 and T3 are usually significantly elevated | Free T4 only slightly elevated; T3 may not rise |
| TRAb | Positive in more than 95% of cases | Negative |
| Doppler USG | Thyroid blood flow increased | Diagnostic value unknown (low flow in thyroiditis) |
| Treatment | Thionamide is indicated | No thionamide required; supportive therapy |
Women with hyperemesis gravidarum have elevated hCG and estradiol levels, and their hCG carries stronger thyroid-stimulating activity; In one study, low TSH was detected in 60% of these women, while the rate was 9% in normal pregnant women. Free T4 is also high in a small portion. If overt hyperthyroidism persists for more than a few weeks or extends beyond the first trimester, the condition is most likely not mediated by hCG. The frequency of hyperthyroidism in trophoblastic disease has decreased with early diagnosis: whereas clinical hyperthyroidism was reported in 55–60% in the past, in a UK series of 196 patients treated between 2005–2010, biochemical hyperthyroidism was 7% and clinical hyperthyroidism was only 2% (UpToDate, hyperthyroidism, 2026).
Serum TSH is the first test in diagnosis; If TSH is <0.1 mU/L, free T4 (or free T4 index) and total T3 (or free T3) are requested. The diagnosis of overt hyperthyroidism is based on hormone levels that exceed the normal range for pregnancy with a low (<0.1 mU/L) or undetectable (<0.01 mU/L) TSH. If the diagnosis is uncertain, TRAb is measured with a third-generation TBII or TSI assay. Radionuclide imaging is contraindicated during pregnancy.
In women monitored without treatment, TSH, free T4 (if there is a trimester interval) and/or total T4-T3 are measured every 2–4 weeks (UpToDate, hyperthyroidism, 2026).
Beta blocker is recommended for symptomatic tachycardia and tremor due to moderate-severe hyperthyroidism (UpToDate Grade 2B). If there are no contraindications, start with the lowest effective dose: metoprolol 25–50 mg/day or alternatively propranolol 20 mg every 6–8 hours. As soon as hyperthyroidism is controlled with thionamides or resolves spontaneously, the beta blocker is tapered and discontinued; In most patients, it can be discontinued within 2–6 weeks. Long-term use of beta blockers during pregnancy should be avoided due to concerns about fetal growth restriction and hypoglycemia; Atenolol is not specifically used.
Before starting a thionamide, obtain a baseline complete blood count (with differential) and liver profile (bilirubin, transaminases). Thionamides are not used in patients with a baseline absolute neutrophil count <1000/microL (except those with known stable chronic neutropenia), or in patients with transaminases more than five times the upper limit of normal (UpToDate, hyperthyroidism, 2026).
The choice of medication depends on which trimester it is started: propylthiouracil (PTU) is preferred in the first 16 weeks, methimazole is preferred after the 16th week (ATA 2026; UpToDate Grade 2C). The rationale is twofold—PTU carries less severe teratogenic potential than methimazole, and maternal fulminant hepatic necrosis, a rare complication that has not been reported with methimazole, can be seen with PTU. Clinical practice diverges at this point: a patient well controlled with PTU in the first trimester may remain on PTU throughout pregnancy (methimazole is more potent, creating a risk of temporary maternal or fetal hypothyroidism during the transition), or may be switched to methimazole at week 16 (PTU exposure is shortened). A patient with mild elevations in transaminases while taking PTU should be switched to methimazole in the second trimester; In a patient who develops a rash after switching to methimazole, methimazole is discontinued and returned to PTU.
There are three options for a Graves' patient taking methimazole before pregnancy: definitive treatment before pregnancy (surgery or radioiodine—the recommended route for patients requiring high doses of methimazole), switching to PTU before trying to conceive (reasonable in young, regularly cycling women expected to conceive within 1–3 months), or discontinuing methimazole and monitoring closely. The last option is for patients with normal TSH and TRAb negative who have been using low dose methimazole for more than six months: In a prospective study conducted on 63 pregnant women, when the drug was discontinued, relapse was found to be 13% in those with TSH normal + TRAb negative, 83.3% in those with TSH suppressed + TRAb positive, and 36% in those with normal TSH but TRAb positive (UpToDate, hyperthyroidism, 2026).
All three antithyroid drugs have been associated with a possible teratogenic effect, but the effects are more severe with methimazole and carbimazole than with PTU: aplasia cutis, tracheoesophageal fistula, choanal atresia, omphalocele, and omphalomesenteric duct anomalies have been reported with methimazole exposure; Those reported with PTU (preauricular sinus and cysts, urinary system anomalies) are milder. Gestational weeks 6–10 are the period when the risk of anomalies is highest (UpToDate, hyperthyroidism, 2026).
| Study | PTU | Methimazole / carbimazole | Not exposed |
|---|---|---|---|
| Danish registry study, 817,093 live births (Andersen 2013, JCEM) | 8.0% | 9.1% (both: 10.1%) | 5.4% |
| Swedish registry analysis | 6.4% | 6.8% | 8.0% |
| Japan, pregnant women with Graves' disease | 1.9% (12/1399) | 4.1% (50/1231) | 2.1% (40/1906) |
| Korean national insurance database (2,886,970 pregnancies, 12,891 exposures) | +8.81 per 1000 live births | +17.05 per 1000 live births (both: +16.53) | reference |
Registry studies do not point in one direction: in some, the prevalence of anomalies was found to be similar between the two drugs, but anomalies associated with PTU were milder, and in others, the prevalence was higher with methimazole. In Korean data, higher methimazole dosage was associated with higher risk; In the Danish series, although anomalies due to PTU were milder, 9 of 14 children underwent surgical correction (Andersen 2013, JCEM).
Dosing is an attempt to balance between undertreatment (risk of miscarriage) and overtreatment (fetal hypothyroidism) of maternal hyperthyroidism. The risk of fetal hypothyroidism becomes particularly important in the late second and early third trimesters, as the fetal thyroid is not functional until the late second trimester; The incidence of neonatal hypothyroidism is best related to the thionamide dose in the third trimester. In the study covering 305 pregnancies, the rate of neonatal hypothyroidism was 19% with methimazole and 13% with PTU; Neonatal hypothyroidism was not observed when the methimazole dose was 10 mg or less and the PTU dose was 150 mg or less (UpToDate, hyperthyroidism, 2026).
Target: "high-normal", not "normal" in hyperthyroidism. The thionamide dose is adjusted to maintain free T4 at or just above the upper third of the trimester-specific reference range; If there is no trimester interval, the upper limit for total T4 and T3 is accepted as 1.5 times the non-pregnant value and the same target is applied. Serum TSH may remain below the pregnancy reference range (approximately <0.08 mU/L)—this is expected. The free T4 target is more important than TSH in dose adjustment. The aim is the lowest effective dose: overtreatment can lead to fetal goiter and fetal primary hypothyroidism because the fetal thyroid is more sensitive to antithyroid drugs than the mother's (Momotani 1986, NEJM).
Thyroid function tests are measured every 2–4 weeks throughout pregnancy and also within 1–2 weeks after medication is changed. Routine liver enzyme monitoring has not been found effective in reducing the risk of liver failure due to PTU and is not recommended by the ATA and the FDA; If monitoring is preferred, PTU is discontinued when transaminases exceed three times the upper limit of normal. All patients should be instructed to discontinue the medication and consult a physician if they develop fatigue, nausea-vomiting, jaundice, dark urine or light-colored stools. Leukocyte count is not routinely monitored, but is requested at the onset of febrile illness and pharyngitis; In this case, the patient should stop taking the medication until the result is seen.
TRAb monitoring serves two purposes. In women diagnosed with Graves' disease during pregnancy and started on thionamide, TRAb is measured at diagnosis and, if elevated, repeated at 18–22 and 30–34 weeks; The disappearance of the antibody is an indicator of remission and allows the drug to be reduced or discontinued. The second is predictive of fetal risk: 1–5% of mothers with Graves' disease (active or previously treated) develop hyperthyroidism in the fetus or newborn; This probability increases when maternal TRAb exceeds 3–5 times the upper limit of normal (ATA 2026).
Graves' disease often goes into remission during the third trimester; Depending on TRAb decline, thionamide may be discontinued at the end of the second trimester or in the third trimester in at least one third of women. Remission is not expected in toxic adenoma and toxic multinodular goiter. If thionamide is stopped in the third trimester, a thyroid test is performed two weeks later and monthly thereafter; if it is discontinued earlier, a thyroid test is performed every 1–2 weeks during the first trimester and then monthly (UpToDate, hyperthyroidism, 2026).
A Graves' patient who cannot use thionamide due to allergy, agranulocytosis, or hepatotoxicity may require thyroidectomy during pregnancy; Surgery during pregnancy is associated with an increased risk of spontaneous abortion and preterm birth and higher rates of surgical complications. Preoperative preparation is done with beta blocker (metoprolol or propranolol) and potassium iodide solution (35–50 mg iodine per drop; 1–3 drops per day) for 7–10 days (UpToDate, hyperthyroidism, 2026).
Radioiodine is absolutely contraindicated in pregnancy and lactation — fetal thyroid tissue begins to function at 10–12 weeks and can be ablated. The use of levothyroxine together with thionamide (block-replacement) is not recommended during pregnancy; Levothyroxine barely crosses the placenta, making it difficult to determine the lowest dose of thionamide that will keep the mother under control. There is insufficient evidence for the routine use of iodine as primary treatment in Graves' disease: In a preliminary study from Japan, 1333 women who continued antithyroid medication were compared with 283 women who switched to potassium iodide. Congenital anomalies were found to be lower in the iodine group (1.5% vs 4.1%), live births were found more frequently (92% vs 85%), but maternal hyperthyroidism was less well controlled.
Due to concerns of hepatotoxicity due to PTU, methimazole is preferred over PTU in breastfeeding mothers (Grade 2C); It is given after breastfeeding and in divided doses. Doses of methimazole up to 20 mg per day do not significantly affect the infant's thyroid function; If the dose is above 20 mg/day, the baby's thyroid tests should be checked in the first and third months (ATA 2026).
Graves' disease often flares or relapses in the postpartum period; In a study of 150 women, 84% (21/25) of those with Graves' disease in remission who had a pregnancy had a recurrence compared with 56% (70/125) of those who did not have a pregnancy—and 20 of 21 relapses occurred 4–8 months after birth; In other words, the risk belongs to the postpartum period rather than the pregnancy itself. One approach in women treated during pregnancy is to test the thyroid six weeks after birth, repeating every six weeks if dosage adjustment is necessary, and every four months if tests remain normal (UpToDate, hyperthyroidism, 2026).
Thyroid storm is an endocrine emergency that occurs with a triggering event, usually on the basis of uncontrolled hyperthyroidism, and is accompanied by multiple organ dysfunction and systemic decompensation. It is rare in pregnancy—prevalence reported as 2.97 per million pregnancies in a population-based cohort based on a Canadian database—but accounts for 5–16% of inpatients for thyrotoxicosis and in-hospital mortality is in the range of 3–11% (Vadini 2024, Thyroid Res ). Pregnancy-specific triggers include preeclampsia, anemia, placenta previa, labor induction or cesarean section, and chorioamnionitis.
Diagnosis is clinical: hyperpyrexia (>39.4 °C / 103 °F), tachycardia usually over 140/min, arrhythmia, signs of congestive heart failure, nausea-vomiting-diarrhea and neuropsychiatric disorders such as confusion, agitation, tremor, seizure. Burch-Wartofsky Point Scale (BWPS) and Japan Thyroid Association (JTA) measures have not been validated in pregnancy; BWPS has high sensitivity but low specificity, while JTA measures may have lower sensitivity. The degree of free T4 elevation does not correlate with the severity of the storm.
| Step | Drug and dosage | Pregnancy-specific note |
|---|---|---|
| Hormone synthesis | PTU 200–250 mg every 4 hours—maintenance dose is the same across sources. There is divergence in loading dose: Vadini 2024 recommends enteral 600–1000 mg loading, ATA 2016 recommends 500–1000 mg loading; UpToDate does not give a loading dose and starts directly with a maintenance dose (on the grounds that the initial dose is usually sufficient to block hormone synthesis). | Trimester-independent first choice: reduces peripheral T4→T3 conversion by also inhibiting type 1 deiodinase (45% decrease in T3 in 24 hours vs 10–12% in methimazole) (Vadini 2024). Methimazole 20 mg every 6 hours if PTU is contraindicated (UpToDate, thyroid storm, 2025) |
| Hormone release | SSKI 5 drops (250 mg) every 6–8 hours or Lugol's solution 10 drops (0.5 mL) every 8 hours | 🔴 Administered at least one hour after thionamide — otherwise Jod-Basedow effect. Lithium carbonate is contraindicated in pregnancy |
| Adrenergic control | Propranolol 60–80 mg enteral every 4–6 hours (IV route only under hemodynamic monitoring). There are two separate published regimens for esmolol: 250–500 µg/kg IV loading + 50–100 µg/kg/min infusion, or 1 mg/kg over 30 seconds + approximately 150 µg/kg/min in the JTA/JES 2016 guideline. Dose titrated according to heart rate | Divergence in preference: JTA/JES 2016 favors short-acting beta-1 selective agent (esmolol) over propranolol — citing retrospective analysis showing increased mortality with propranolol in heart failure. ATA 2016 and UpToDate prefer propranolol; It also suppresses peripheral T4→T3 conversion at high doses. In a subsequent large retrospective cohort, adjusted in-hospital mortality was similar for the two options (beta-1 selective 6.3% vs propranolol 7.4%; OR 0.85; 95% CI 0.57–1.26), including in those with acute heart failure (Matsuo 2024, Crit Care Med). Atenolol is not used |
| Peripheral conversion / relative adrenal insufficiency | Hydrocortisone 300 mg IV loading followed by 100 mg IV every 8 hours | Hydrocortisone is preferred because dexamethasone (2–4 mg IV every 8 hours) crosses the placenta (Vadini 2024) |
| Enterohepatic cycle | Cholestyramine 4 g orally every 6 hours | Impairs the absorption of fat-soluble vitamins; routine prenatal supplementation may be inadequate (FDA category C) |
| Refractory case | Plasmapheresis; emergency thyroidectomy as a last resort | Effect lasts 48–72 hours; After the procedure, birth is postponed for at least 24 hours if possible. The patient is placed in the left lateral decubitus position. |
Supportive treatment is carried out in the intensive care unit, with continuous cardiac and fetal monitoring, in collaboration with endocrinology and perinatology. Paracetamol, not aspirin, is used to reduce fever — aspirin can theoretically increase the level of free hormone by separating the hormone from its binding site. Since hepatic glycogen stores are depleted, close monitoring is required for hypoglycemia; Prenatal vitamins are maintained with forms that do not contain iodine. Giving birth without correcting the condition is avoided (ACOG). The trigger is quickly sought; If an infection such as chorioamnionitis is suspected, antibiotics are started early.
Postpartum thyroiditis is a destructive thyroiditis that occurs in 5–8% of women in the first 12 months after birth. Its pathogenesis is explained as the rebound of the underlying thyroid autoimmunity with the end of the immune tolerance created by pregnancy; As a matter of fact, it is more common in TPO antibody-positive women and those with a personal or family history of autoimmune diseases. It may also occur after miscarriage or pregnancy loss, or in women taking levothyroxine for Hashimoto's thyroiditis (Lee & Pearce 2022, Nat Rev Endocrinol).
The classic course is three-phase: 4–8 weeks after birth, the thyrotoxic phase begins with the release of ready hormone from the inflamed gland and typically lasts 1–2 months; This is followed by a period of temporary hypothyroidism lasting 4–6 months. However, the course is variable: approximately 25% of women have only a thyrotoxic phase, and 50% have only a hypothyroid phase. Although the majority recover fully and return to a euthyroid state within a year, up to 50% may develop permanent hypothyroidism even after the initial recovery; This risk is higher in TPO antibody positive women. Postpartum thyroiditis recurs in up to 70% of subsequent pregnancies (Lee & Pearce 2022).
Separating the thyrotoxic phase from Graves changes treatment. The thyrotoxic phase of postpartum thyroiditis is temporary and is caused by the release of ready-made hormone—there is no increase in synthesis, so antithyroid medications do not work. In distinction: TRAb is sensitive and specific for Graves (TPO antibody may be high in both conditions); The T3/T4 ratio is often higher in Graves (>20:1); Timing is also helpful—postpartum thyroiditis typically occurs 1–4 months after birth, Graves' disease 6–12 months. Radioactive iodine uptake differentiates if necessary (low in thyroiditis, high in Graves') but is not appropriate during breastfeeding and TRAb measurement is usually sufficient.
Treatment is symptomatic. Significant thyrotoxic symptoms, such as palpitations and anxiety, are controlled with lactation-safe beta blockers (metoprolol, propranolol). During the hypothyroid phase, levothyroxine can be started if the woman is symptomatic or actively trying to become pregnant; Since hypothyroidism is usually temporary, tapering off is attempted 6–12 months after diagnosis. Since the risk of permanent hypothyroidism is high in recovered women, annual TSH control and follow-up for recurrence after subsequent pregnancies are recommended. Selenium supplementation has been investigated for its anti-inflammatory effect in autoimmune thyroiditis, but studies in pregnancy have yielded mixed results; Selenium is not recommended for preventing or treating postpartum thyroiditis.
Routine thyroid function screening is not recommended in asymptomatic women in the postpartum period. On the other hand, testing should be considered 6–12 weeks after birth in high-risk groups: previous postpartum thyroiditis or Graves' disease history, Hashimoto's thyroiditis under levothyroxine, known TPO antibody positivity, type 1 diabetes, other non-thyroidal autoimmune diseases and chronic hepatitis C. Serum TSH should also be checked in cases of postpartum depression or breastfeeding difficulty (Peng & Pearce 2022, J Endocrinol Invest ).
Last updated: .