New American Thyroid Association and American Association of Clinical Endocrinologists Guidelines for Thyrotoxicosis and Other Forms of Hyperthyroidism: Significant Progress for the Clinician and a Guide to Future Research

T he authors of the new American Thyroid Association (ATA) and American Association of Clinical Endocrinologists (AACE) guidelines for hyperthyroidism and other forms of thyrotoxicosis, published in this issue (1), are to be congratulated. These guidelines are much needed. They serve to update separate previous guidelines from the ATA in 1995 (2) and from AACE in 2002 (3). The updated version is far more comprehensive than the older versions, including topics such as thyroiditis, gestational thyrotoxicosis, surgical preparation and extent, and radiation safety, which were not previously discussed. The new guidelines are also more modern in format, incorporating separate recommendations together with levels of evidence.

An important shortcoming of the new guidelines is that they are based on relatively limited available evidence. Of the 100 graded recommendations, 80 are strong recommendations (grade 1) and 20 are weak recommendations (grade 2). Seventy are based on low-quality evidence (+), 29 are based on moderate-quality evidence (++), and only one is based on high-quality evidence (+++). Of the 80 strong recommendations, 56 (70%) are based on low-quality evidence (+), 23 (29%) are based on moderate-quality evidence (++), and only one (1%) is based on high-quality evidence (+++). Of the 20 weak recommendations, 15 (75%) are based on low-quality evidence (+) and five (25%) are based on moderate-quality evidence (++). Therefore, most are strong recommendations that are based on low-quality evidence. This indicates that clinical management of hyperthyroidism is still largely rooted in expert opinion and personal experience. Given this, it is perhaps remarkable that 98 of the recommendations were unanimous, whereas two of the recommendations had a single dissenting member.

One goal of this editorial is to explore areas where we three editorialists disagreed with the recommendations and with each other. One minor area of disagreement among us relates to determining the etiology of thyrotoxicosis. All of us appreciate the guidelines' emphasis on the use of the radioactive iodine uptake when the etiology of hyperthyroidism is not clinically apparent. However, one of us feels that, given concerns about unnecessary medical radiation exposure and costs, thyroid receptor antibody (TRAb) testing should be considered as an alternative to nuclear medicine testing in all patients, not just pregnant and nursing women. All of us are united in the hope that these guidelines will help to decrease the ordering of unnecessary iodinated contrast imaging in hyperthyroid patients, an all-too-frequent practice that delays subsequent radioactive iodine imaging or treatment.

With regard to ruling out factitious thyrotoxicosis, we would add that the triiodothyronine (T₃)/thyroxine (T₄) ratio is typically <20 with excess L-thyroxine (LT₄) ingestion and >20 with excess T₃ ingestion, but it may be unhelpful in thyroid hormone extract ingestion because the T₄/T₃ ratios of thyroid hormone extract products may be inconsistent. Although the U.S. Food and Drug Administration has warned against this practice, manufacturers have previously marketed 3,5,3′-triiodothyroacetic acid (TRIAC)-containing supplements as weight loss aids. High TRIAC doses would be expected to cause low serum thyrotopin (TSH) and T₄ levels, but elevated serum T₃ due to cross-reactivity of TRIAC with T₃ assays (4). The suggestion to check fecal T₄ levels to rule out factitious thyrotoxicosis in thyroglobulin antibody positive patients is both novel and helpful.

The previous AACE guidelines noted that “in the U.S. radioactive iodine is currently the treatment of choice for Graves'” and largely ignored other treatment options. The new guidelines give more equal and balanced consideration to I-131, anti-thyroid drugs (ATDs), and surgery in the management of Graves' disease. We applaud Recommendation 4, with its emphasis on individualizing the management plan with proper attention paid to the preferences of each patient.

Although all therapeutic options are recognized, we believe that the duration of ATD use is appropriately limited under the current guidelines. Definitive therapy is preferred in Graves' disease patients whose disease does not remit after 12–18 months of ATD treatment and in all patients with toxic nodules, given the potential for ATD-related adverse effects even after prolonged use. Although the guidelines state that the prolonged use of ATDs is acceptable when patients “prefer this approach,” we all strongly discourage this, especially in patients taking higher ATD doses (>15 mg/d methimazole [MMI]). In this situation, we feel most comfortable documenting that another course has been suggested, but that the patient has refused definitive therapy. In this case we do not believe that the customer is always right.

The previous AACE guidelines did not comment on the duration of ATD therapy or on differences between propylthiouracil (PTU) and MMI. The 1995 ATA guidelines were more forthcoming about ATD use, but there have been several important changes in recommendations regarding anti-thyroid drug therapy since 1995. PTU is now considered to be a second-line agent except during the first trimester of pregnancy or in thyroid storm. All of us disagree with the implication that MMI should be avoided in favor of PTU in patients with thyroid storm because there is no clear evidence to indicate that PTU is more effective than MMI in this setting (5). There is always concern for adverse reactions or inadequate therapy when switching from one ATD to another. The guidelines recommend against changing from one ATD to another in the setting of serious adverse skin reactions or agranulocytosis. However, switching from PTU to MMI is recommended for patients with liver function abnormalities, as is changing from MMI to PTU for the first trimester of pregnancy.

The optimal duration of ATD therapy has been narrowed from the 6–24 months previously advocated to 12–18 months. Block-replace therapy, previously presented as a reasonable option, is now specifically proscribed.

A new recommendation (Recommendation 20) promotes measurement of TRAb prior to stopping ATDs. We find it somewhat perplexing that measurement of TRAb is recommended before weaning patients off ATD therapy, but not at diagnosis or prior to initiation of therapy. With regard to TRAb measurement prior to the discontinuation of ATDs, the guidelines do not clearly discuss the best course of action if the TRAb is clearly elevated. In this instance, should ATDs be continued? Should the dose be increased? Should definitive therapy be instituted? Should ATDs be discontinued with closer follow-up? We would favor discontinuation of ATD therapy and institution of definitive therapy but believe that further clinical studies are warranted to determine the best course of action.

The prior ATA guidelines mentioned the use of beta blockers for symptomatic relief but did not provide specific information about their most appropriate use. In the current guidelines, the difference between the strong recommendations for beta blockers (Recommendations 5 and 32) and the weak recommendations for methimazole (Recommendations 6 and 33) pretreatment prior to I-131 treatment in at-risk patients (not all patients) is the result of the dissent of just one panel member. We don't agree with this distinction. Furthermore, in Recommendation 60, both beta blockers and MMI are recommended prior to I-131 in children with severe hyperthyroidism. The reason for the difference in the recommendation strength between adults and children is not clear. While many patients require no pretreatment with either drug, all of us personally recommend pretreatment of all high risk patients with both beta blockers and MMI.

Plasmapheresis is a therapeutic option not mentioned at all in the guidelines. We would note that plasmapheresis has been reported to reduce or normalize thyroid hormone levels preoperatively in patients who cannot take ATD therapy (6,7). Its successful use has also been reported in patients with thyroid storm (8). As such, mention of plasmapheresis is warranted to assure awareness of this effective, albeit rarely applied, therapy.

A significant departure from previous guidelines is the explicit statement that the goal of I-131 therapy in Graves' disease is hypothyroidism. This is helpful in setting appropriate patient and provider expectations. We also find the discussion regarding the use of ATDs in toxic multinodular goiter to raise serum TSH and enhance uptake for enhanced tissue killing with 131 I to be particularly clinically useful.

The recommendations regarding safety precautions following radioactive iodine therapy are important, particularly in light of current variable practice patterns around the United States (9). A detailed ATA Best Practices statement regarding safety precautions following I-131 dosing has recently been published (10). One caveat with regard to the guidelines is that it is stated, based on Nuclear Regulatory Commission (NRC) rules, that discharge precautions after I-131 pertain only to patients with retained doses >33 mCi. Doses used in patients with hyperthyroidism are typically lower than this threshold. However, it would be unfortunate if this literal interpretation of the NRC rule were to downplay the importance of the radiation safety precautions following I-131 therapy for hyperthyroidism. Precautions remain important even when lower I-131 doses are administered. The retained radiation dose lasts longer in hyperthyroid patients than in thyroid cancer patients. Hyperthyroid patients are often young enough to have small children at home who are at particular risk from radiation exposure. Finally, the careful observation of precautions helps to alleviate fears about radiation exposure among the lay public.

One discrepancy in the guidelines is between Recommendation 37, which advocates following free T₄, total T₃, and TSH in the post I-131 monitoring of patients with multinodular goiter and toxic adenoma, and Recommendation 11, which advocates only free T₄ and total T₃, but not TSH testing, in the post I-131 follow-up of Graves' disease patients. Following I-131, TSH may be suppressed for up to 6 weeks after T₄ and T₃ normalize, but it is not clear that Graves' disease and toxic nodular disease are different in this regard. We would recommend free T₄, total T₃, and TSH testing in all these situations.

Detailed recommendations regarding the pre- and postoperative care of patients who undergo thyroidectomy are clinically valuable. An important distinction is made between Graves' disease patients, who should receive iodine preoperatively to decrease thyroid vascularity, and patients with autonomous nodular goiter, who should not receive preoperative iodine due to the risk of precipitating Jöd-Basedow disease.

It is interesting to note that total thyroidectomy is preferred in toxic adenoma patients when any abnormality requiring future surveillance is present in the contralateral lobe. We often have difficulty convincing patients with known thyroid cancer that completion thyroidectomy is their best option, whereas here the intentional resection of benign tissue is, justifiably, advocated.

Regarding postoperative care, Recommendation 26 suggests that either serum calcium or parathyroid hormone (PTH) should be monitored in the postoperative setting. We do not feel that these tests are interchangeable. Serum calcium values (corrected for albumin levels) are both less expensive and more useful than serum PTH for acute clinical decision-making. There is a risk that appropriately low serum PTH levels in those discharged on calcium and calcitriol may be misinterpreted as permanent hypoparathyroidism. In addition, serum parathyroid hormone levels within the normal range may be misleading (11).

The section on the management of subclinical hyperthyroidism is a particularly welcome addition to the new guidelines. This topic was not addressed at all in the 1995 ATA guidelines. Previous consensus statements (12,13) and the previous AACE guidelines noted that this area is controversial and therapeutic options should be individualized. The prior publications all recommended consideration of treatment for subclinical hyperthyroidism (when the serum TSH is persistently <0.1 mIU/L) for those with symptoms, atrial fibrillation, osteoporosis, or osteopenia. The new guidelines are more aggressive in adding consideration of treatment in any patient if the underlying etiology is toxic nodular disease, and in all individuals aged ≥65 and all postmenopausal women not on bisphosphonates or estrogen.

Another useful addition to the new guidelines is the discussion regarding iodine-induced hyperthyroidism. The guidelines state that “iodine-containing contrast agents may be an additional factor in the development of thyroid storm in patients with illnesses unrelated to thyroid disease,” implying that contrast administration to patients without underlying thyroid disease may lead to storm. However, the majority of case reports about the induction of storm with contrast are in subjects with underlying subclinical thyrotoxicosis from toxic adenoma or toxic multinodular goiter. It would be unusual to induce severe thyrotoxicosis in otherwise normal thyroids with iodinated contrast. In addition, the guidelines may downplay the time to clearance of an iodinated contrast load. We have observed that after contrast administration, especially with repeated catheterization or angiography or in the setting of mild renal compromise, the iodine load may take up to several months to clear. While the guidelines note that a urinary iodine concentration may be used to determine whether the iodine load is too high for radioactive iodine scanning or therapy, no threshold values are provided. Based on anecdotal evidence we would suggest that it is reasonable to proceed with radioactive iodine scanning or treatment after a known iodine load if the urinary iodine concentration is under 200 μg/L.

Although commonly encountered in clinical practice, the treatment of thyroid disease in pregnancy is clearly an understudied topic. In the current guidelines, 11 of 12 recommendations (86%) focused on hyperthyroidism in pregnancy were strong recommendations despite all 12 recommendations being based on low-quality evidence. A significant departure from previous guidelines is the recommendation for the use of PTU as a first-line ATD only in the first trimester of pregnancy; previously, PTU had been preferred throughout gestation. We all share concerns about Recommendation 68, which advocates the use of trimester-specific free T₃ or free T₄ assays in the diagnosis of hyperthyroidism. Both measurements, but particularly commercially available free T₃ assays, are problematic in pregnant women, and trimester-specific ranges are not widely available.

Recommendations for the diagnosis and treatment of postpartum thyroiditis are included in these guidelines for the first time. However, the guidelines are somewhat unclear regarding the expected duration of thyroid dysfunction, the potential duration of LT₄ therapy for the symptomatic hypothyroid phase, and the likelihood that permanent hypothyroidism might be present after withdrawal. New ATA guidelines for the management of thyroid disease in pregnancy and the postpartum period are currently being considered for publication and should provide more detailed recommendations. The current guidelines do have useful information regarding the use of Tc99 or I-123 (never I-131) in the diagnosis of postpartum thyroiditis. Specific advice about the number of days to avoid breastfeeding following nuclear medicine studies would have been helpful.

Finally, the new guidelines include a useful section on Graves' orbitopathy. The previous ATA guidelines recommended corticosteroids with I-131 treatment in all patients with established ophthalmopathy, regardless of severity or activity. The previous AACE guidelines noted that “systemic glucocorticoids have been used … in patient with active ophthalmopathy … particularly after I-131 therapy, but their efficacy is not fully established.” The updated guidelines now, more helpfully, state that those with mild active Graves' orbitopathy who are smokers or those with moderate to severe active Graves' orbitopathy should receive glucocorticoids with I-131 treatment.

The lack of high-quality evidence regarding the management of hyperthyroidism underscores the need for more clinical and translational research. It is hoped that, in addition to serving as a clinical tool, these guidelines may be used to help identify gaps in the literature and to direct research that will provide answers to persistent management questions in this important disease that affects 1%–2% of the population.