Graves Hyperthyroidism and Pregnancy: A Clinical Update

Komal Patil-Sisodia, MD; Jorge H. Mestman, MD


Endocr Pract. 2010;16(1):118-129. 

In This Article

Laboratory Tests

During pregnancy, normal values for serum TSH are trimester dependent. Because of the human chorionic gonadotropin effect in the first trimester, serum TSH values are lower than in nonpregnant persons, with a range of 0.2 to 2.5 mIU/L in the first trimester, 0.3 to 3.0 mIU/L in the second trimester, and up to 3.5 mIU/L in the third trimester.[19]

The assessment of free T4 in pregnancy, particularly in the second half of gestation, may be confusing for the attending physician. Laboratories often measure free T4 via radioimmunoassay, which provides estimated values, not direct measurement.[20] Changes in albumin and T4-binding globulin, such as those that are seen in pregnancy, impair the accuracy of these assays, thus providing inaccurate free T4 values.[21] Current reference methods for measurement of free T4 include equilibrium dialysis and liquid or gas chromatography/mass spectrometry.[20] These methods show less than a 10% decrease in plasma free T4 when comparing values from pregnant and nonpregnant patients.[20,22] Recent literature has confirmed that interpretation of free T4 values is inherently flawed in pregnant women if reference ranges for nonpregnant persons are used.[20,23] Free T4 levels decrease below the lower limits of normal for 2 commonly used immunoassays (Elecsys, Roche, Indiana; Tosoh, Ramsey, Minnesota) during the second and third trimesters of pregnancy, while remaining comparable or slightly lower in the first trimester.[20] Thus, the use of free T4 can be misleading in the care of pregnant women. Free T4 measurement by liquid chromatography/mass spectrometry has a strong correlation with free T4 values obtained by equilibrium dialysis, the criterion standard for free T4 measurement.[23] Furthermore, poor correlation has been documented between immunoassays and liquid chromatography/mass spectrometry for free T4 determination.[23] Assays using equilibrium dialysis are expensive, not readily available, and technically difficult, thus making this method an impractical choice for routine measurement of free T4 during pregnancy. Newer assays using liquid chromatography/mass spectrometry may be a better choice when compared with radioimmunoassay, but are not yet readily available.

Free T4 index, calculated by a product of serum total T4 and a test for assessing T4-binding globulin such as resin uptake, remains constant during pregnancy and prevents misdiagnosis of "hypothyroxinemia." Lee et al reported close to 30% of patients with "hypothyroxinemia" when the free T4 assay was used.[20] Furthermore, serum TSH correlated closely with free T4 index and not with free T4 by immunoassays (see Fig. 1 for a representative example). Serum T3 determination is justified only in cases of slight elevation or normal serum free T4 index in the presence of hyperthyroid symptoms and suppressed serum TSH. Serum thyroperoxidase antibodies are present in most persons with Graves hyperthyroidism and may be helpful in distinguishing the condition from gestational thyrotoxicosis (Table 2).

Figure 1.

A representative example of transient hyperthyroidism of hyperemesis gravidarum. By week 6, vomiting begins and becomes severe by week 10. Serum free thyroxine (T4) index is elevated and thyrotropin is suppressed. By weeks 16 to 18, vomiting subsides with marked improvement of the free T4 index value. During this period, the patient loses 3.6 kg. By week 18, the serum free T4 index returns to normal, but the serum thyrotropin remains suppressed until week 26. Patient regains and gains weight with a term delivery of a healthy infant. The gray band indicates reference range. LNMP, last normal menstrual period.

TRAb with stimulating activity is an important component in the evaluation of Graves hyperthyroidism during pregnancy. High levels of TRAb cross placental barriers and may affect fetal and neonatal thyroid function. TRAb is measured by several different methods. TRAb determination by bioassay measures cyclic adenosine monophosphate generation in TSH receptor–expressing cell lines. This method measures thyroid-stimulating immunoglobulins. Normal values are less than 130% of basal activity. The risk of fetal or neonatal hyperthyroidism is increased when greater than 500% of basal activity is detected. Radioreceptor assays measure TSH-binding inhibitory immunoglobulins that inhibit TSH binding to TSH receptor sites). Normal values are reported as less than 10%. Persons with Graves disease have inhibition that ranges from 10% to 100%. The risk of fetal or neonatal thyrotoxicosis increases when inhibition is greater than 50%, with a sensitivity of 100% and specificity of 76%.[24,25] TSH-binding inhibitory immunoglobulins can have either stimulating or blocking effects at the TSH receptor, although both have been detected in serum from patients with Graves hyperthyroidism. Changes in titers during pregnancy have been suggested to explain the improvement in hyperthyroidism with progression of pregnancy and recurrence in the postpartum period, although not all investigators concur.[26] Antibody titers decrease in the second half of pregnancy because of immunomodulation of pregnancy, followed by a possible flare during the postpartum period.[12] Therefore, to predict fetal or neonatal hyperthyroidism, the most appropriate gestational age to measure TSH-binding inhibitory immunoglobulins or thyroid-stimulating immunoglobulins is between 24 and 28 weeks. Both methods are equivalent, and each has advantages and disadvantages. The turnaround time for the thyroid-stimulating immunoglobulins assay is between 2 and 4 days, but it is a more expensive test. Assessing TSH-binding inhibitory immunoglobulins is less expensive, but the results can take up to 7 days. The clinician should follow-up on the turn-around time and cost of each assay. Recommendations for when TRAb should be assessed in pregnancy are listed in Table 3.