When Thyroidologists Agree to Disagree: Comments on the 2012 Endocrine Society Pregnancy and Thyroid Disease Clinical Practice Guideline

The thyroid gland and its function are profoundly impacted by pregnancy. Typically, thyroid hormone production and daily iodine intake increase by approximately 50% in the first few weeks of pregnancy (1). These increasing demands may result in hypothyroidism later in pregnancy in women who are iodine-deficient, even if they were euthyroid during the first trimester. To determine thyroid status, clinicians depend on thyroid function tests, which are among the most commonly requested laboratory tests (2). Normally, the evaluation of thyroid function begins with measuring TSH. During pregnancy, the typical increase in placental human chorionic gonadotropin concentrations result in lower, pregnancy-specific TSH concentrations in the first trimester and higher total T₄ and T₃ concentrations overall (3). Other important laboratory assessments during pregnancy include thyroid autoantibodies such as autoantibodies to thyroid peroxidase (TPOAb), thyroglobulin (TgAb) and the TSH receptor (TRAb). Advancements and innovations in bioanalytical technologies and the growing interest in thyroid-related research, especially during pregnancy, have immensely improved our diagnostic capacity and knowledge regarding the management of thyroid dysfunction.

The stated objective of the new “Management of Thyroid Dysfunction during Pregnancy and Postpartum: An Endocrine Society Clinical Practice Guideline” published in this issue of the JCEM (4) is to update the previous Endocrine Society guidelines published in 2007 (5). I believe this mission is well accomplished. The monumental task of developing the updated guideline was given to the 13 members of the Task Force, with the help of a medical writer, and it was reviewed and commented on by members of The Endocrine Society, the Asia and Oceania Thyroid Association, and the Latin American Thyroid Society, and approved by The Endocrine Society Council.

In October 2011, the “Guidelines of the American Thyroid Association (ATA) for the Diagnosis and Management of Thyroid Disease during Pregnancy and Postpartum” were published in the journal Thyroid (6). Four of the 13 members (30%) of The Endocrine Society's Task Force served on both of these committees and authored both guidelines. As one of the authors of the ATA guidelines, I will focus on several important differences between the two documents (4, 6).

The new Endocrine Society guideline is comprehensive and reflects some of the recent explosion of knowledge in the field. A full 15% (21 publications) of the 141 references are updates from the previous Endocrine Society guideline and were published after 2007. Surprisingly, however, there is not a section in the 2012 guideline dedicated to thyroid function tests, the foundation for any diagnosis and treatment of thyroid disease. Furthermore, there is no update on the most recent advancements in thyroid function testing. Indeed, the 2012 guideline recommends caution in the interpretation of serum free T₄ (FT4) levels during pregnancy, and that each laboratory should establish trimester-specific reference ranges for pregnant women if using a FT4 assay (4). However, the effects of ethnicity and trimester-specific ranges for TSH are only briefly discussed (in Sections 1.2.5 and 8.4b). Trimester-specific FT4 ranges should be method-specific, and it is important to use a reference population with optimal iodine intake.

Assay methodologies continue to evolve as performance standards are established and new technology and instrumentation are developed. The specificity and sensitivity of thyroid function tests have progressively improved from the competitive immunoassays of the early 1970s and noncompetitive immunometric assays to the more recent liquid chromatography tandem mass spectrometry (LC/MS/MS) (7). Unfortunately, during pregnancy the diagnostic accuracy of serum-free thyroid hormone methods is not predictable from their assay methodology (8). Direct serum-based assays of free thyroid hormones (FT4 and FT3), total thyroid hormones (total T₄ and T₃), T₄ binding globulin (TBG), transthyretin (TTR/prealbumin), and albumin are available (9), in addition to more sensitive and specific tests for TPOAb, TgAb, and TRAb (10).

HPLC coupled to a tandem mass spectrometer is the current analytical method of choice to measure analytes in biological matrices. Because LC/MS/MS methods have high specificity, sensitivity, precision, and throughput, this technology is increasingly being used in clinical laboratories. These methods are especially important in states of disease and during pregnancy when protein binding is a factor that interferes with other methods for thyroid hormone analysis. It is inevitable that these powerful analytical methods will eventually be routinely available for clinical use.

Currently, the accepted technique for FT4 measurement is separation of the free hormone from binding globulins by equilibrium dialysis or ultrafiltration of serum, performed at 37 C, followed by thyroid hormone measurement by either immunoassays or LC/MS/MS. This is the proposed international reference method for measuring FT4 concentrations of dialysates and ultrafiltrates (7, 11, 12). However, sample separation using equilibrium dialysis is labor-intensive, technically demanding, time consuming (17–24 h), and expensive, and it is not readily available in most clinical laboratories. Ultrafiltration is faster and less expensive but comes with its own series of issues (15, 16). Granted, direct methods that employ sample preparation using equilibrium dialysis or ultrafiltration are not foolproof, and technical problems that relate to dilution, adsorption, temperature, or the influence of endogenous binding protein inhibitors remain (15, 16). These methods are costly, necessitate highly skilled operators, and are mostly found in reference or research laboratories. Nevertheless, they are the new “gold standard” and therefore deserve a mention in the 2012 Endocrine Society guideline.

Thus, it is curious that The Endocrine Society guideline recommends the use of the FT4 index (FT4I). This index is a calculated ratio, based on two estimates, namely an estimate T₃ resin uptake and immunoassay estimate of total T₄. The T₃ resin uptake test estimates TBG concentrations by adding a known, excess amount of ¹²⁵I-labeled T₃ to a measured volume of serum (17). Index tests (FT4I and FT3I) should probably be used only when there are no alternatives. Indeed, FT4 and FT3 immunoassays are protein-dependent to some extent, and most are prone to under- or overestimate FT4 in patients with significant abnormalities in thyroid hormone binding proteins. In addition, FT4I frequently fails to correct completely for the TBG-induced increase in total T₄. Moreover, no trimester-specific reference intervals are available for FT4I, whereas they are available for FT4 immunoassays. And finally, FT4I is rarely available in modern clinical laboratories because more precise methods have become available and because of this method's limitations.

Thyroid Autoantibodies

Pregnancy often results in hypothyroidism in women with limited thyroidal reserve or iodine deficiency, and postpartum thyroiditis can occur in women with underlying Hashimoto's disease even if they were euthyroid before conception. Ten to 20% of all pregnant women in the first trimester of pregnancy are thyroid antibody-positive and euthyroid. More than one third of the women who are euthyroid and thyroid antibody-positive will develop postpartum thyroiditis. It is therefore important to effectively use other laboratory tests, such as TRAb, TgAb, and TPOAb tests, to optimize treatment and pregnancy outcomes.

Gestational hyperthyroidism (transient hyperthyroidism) is associated with hyperemesis gravidarum. Although the ATA guidelines (no. 25 and 26) recommend supportive therapy, management of dehydration, and no antithyroid drugs, the new Endocrine Society guideline recommends serum total T₃ testing and the use of beta-blockers in women with significant thyrotoxicosis if their obstetricians agree (Section 3.2). It is important to mention the well-defined isolated hypothyroxinemia, which is also solely dependent on accurate laboratory testing (18).

Subclinical Hypothyroidism

Although it is well accepted that overt hypothyroidism and overt hyperthyroidism each have a deleterious impact on pregnancy, studies are now focusing on the potential impact of subclinical hypothyroidism and subclinical hyperthyroidism on maternal and fetal health, the association between miscarriage and preterm delivery in euthyroid thyroid antibody-positive women, and the prevalence and long-term impact of postpartum thyroiditis. Recently completed prospective randomized studies have begun to produce critically needed data on the impact of treating thyroid disease on the mother, fetus, and intellect of the unborn child. Indeed, the new Endocrine Society guideline recommends T₄ replacement in women with subclinical hypothyroidism who are TPOAb-positive and in women who are TPOAb-negative, although the evidence is fair to poor (Section 1.2.2).

Iodine Nutrition

In the new Endocrine Society Guideline, iodine nutrition during pregnancy is extensively discussed (Sections 6.0–6.6). However, it is not clear that the daily iodine intake should increase to 250 μg/d from supplements and diet in total. Section 6.3 mentions that if there is a significant concern about iodine insufficiency, TSH and thyroid hormone levels should be assayed. Iodine deficiency may, of course, cause hypothyroidism with an elevated TSH. However, if iodine deficiency is suspected, it should be noted that it has been long established that there is no quantitative relationship between TSH (or thyroid hormones) and urinary iodine concentrations (19, 20), and therefore TSH concentrations do not provide a measure for iodine sufficiency.

The final recommendation in this section for a “4-h separation of iron supplementation from levothyroxine administration” is an important one, but it would be more effective if discussed in the section on hypothyroidism (Section 1.0).

Screening

The ATA guidelines included an age of 30 yr or greater as a risk criterion (5). Although uncertainty remains regarding a universal screening mandate, the ATA guidelines strongly recommend targeted case finding (5). The agreement with The Endocrine Society guideline regarding the screening issue is encouraging, although not all obstetricians and gynecologists would agree (4).

Predictions on How the Current Science May Change These Guidelines

Prospective randomized trials currently under way worldwide will no doubt help elucidate some of the issues currently unresolved. Pharmacokinetic studies may provide helpful data for more individualized treatment of hypothyroidism in pregnancy, and cost-effectiveness studies could affect new screening recommendations. MicroRNA analysis studies of thyroid tissue and of fine-needle aspirates may advance capabilities for more accurate diagnosis and prognosis and the creation of new therapies. The creation of immortalized human thyroid cell lines opens endless possibilities for research on thyroid receptors, metabolism, and thyroid treatment in pregnancy.