Are you sure the patient has subclinical hyperthyroidism?

Subclinical hyperthyroidism (SH) is defined biochemically by a low (or undetectable) thyroid stimulating hormone (TSH) level with a normal serum free T4 and normal serum total T3 levels due to thyroid disease or exogenous excess thyroid hormone administration. The prevalence of SH is approximately 1%, although it tends to be more common in areas with relative iodine deficiency. By comparison, overt hyperthyroidism occurs when the TSH level is low (or undetectable) and the free T4 level and/or total T3 level is elevated. It is important to distinguish between SH and overt hyperthyroidism because they may be managed differently in most cases.

SH can be due to exogenous or endogenous thyroid hormone excess. Exogenous SH occurs when a patient consumes excessive thyroid hormone, either intentionally or unintentionally. An example of intentional thyroid hormone excess is use of levothyroxine (LT4) to achieve suppressed TSH levels in patients with thyroid cancer. In patients taking thyroid hormone as replacement, some may simply be taking too much or have been prescribed a higher thyroid hormone dose than necessary and require an adjustment.

Endogenous SH can occur due to toxic multinodular goiter, solitary autonomously functioning nodule, Graves’ disease, and thyroiditis (i.e. subacute, silent, and postpartum thyroiditis). SH due to nodular disease tends to be persistent, whereas SH due to Graves’ disease can be either transient or persistent. SH related to Graves’ disease may go into a temporary or permanent remission. SH due to thyroiditis is usually transient.

Continue Reading

Patients with SH are often asymptomatic. When symptoms are present, they are similar to the symptoms in patients with overt hyperthyroidism, although they are usually milder. Symptoms of SH include, but are not limited to, fatigue, palpitations, anxiety or other changes in mood, heat intolerance, diaphoresis, tremor, weight loss, and loose stools or diarrhea.

On physical exam, patients with SH may have only minimal signs of hyperthyroidism, which include tachycardia, atrial fibrillation, warm and/or moist skin, tremor, hyperreflexia and thyroid gland enlargement with or without nodules. In patients with Graves’ disease, they may have additional signs such as Graves’ ophthalmopathy or Graves’ dermopathy. With Graves’ ophthalmopathy, patients can present on exam with proptosis, lid lag and/or stare. With Graves’ dermopathy, which is relatively rare, patients can present on exam with thickening of the skin, particularly in the lower tibia region. They may also have discomfort in the involved area(s).

Key laboratory findings for SH include thyroid function tests (TSH, free T4 and total or free T3), showing the pattern mentioned above with a low (or suppressed) TSH level, normal free T4 level and normal total T3 level.

What else could the patient have?

The differential diagnosis of subclinical hyperthyroidism (SH) includes any disease or process causing a low TSH level, such as overt hyperthyroidism, central hypothyroidism, non-thyroidal illness, medications, age-related changes, and pregnancy.

While overt hyperthyroidism does present with a low TSH, it can be easily distinguished from SH because in overt hyperthyroidism, the free T4 level and/or the total T3 level are elevated. Typically, but not always, patients with overt hyperthyroidism are more symptomatic than patients with SH. Elderly patients tend to be less symptomatic and can present with abnormal lab values but few, if any, symptoms.

Central hypothyroidism occurs due to pituitary or hypothalamic dysfunction. This causes a low TSH, but the free T4 and total T3 levels are also low or, at least, in the low-normal range. Central hypothyroidism should be suspected in patients with a history of sellar or suprasellar surgery or in patients with pituitary disorders.

Non-thyroidal illness is a common cause of low TSH levels in hospitalized patients. The typical pattern of thyroid function tests in patients with non-thyroidal illness is low TSH, normal free T4 and low total T3 levels. When this pattern is seen in patients who are sick, no intervention is usually needed. The thyroid function tests can simply be repeated when the patient has recovered from the acute illness. Rarely, a patient may have actual hyperthyroidism in the context of non thyroidal illness.

Certain medications can also cause low TSH levels. The most frequent suspect is corticosteroid therapy, which is well-known to suppress TSH levels. Other medications include dopamine and dobutamine.

In some healthy elderly patients, TSH levels may be low due to a change in the set point of the hypothalamic-pituitary-thyroid axis or a decrease in thyroid hormone clearance. In these patients, the TSH level may be below normal, but it is not completely suppressed (undetectable).

During pregnancy, TSH reference ranges are different than in non-pregnant individuals. Due to normal physiologic changes, TSH levels (particularly in the first trimester) tend to be low or in the low-normal range with slightly elevated free T4 index and total T3 levels. The serum free T4 measured by some one-step analog assays are inappropriately low in pregnancy and should not be measured by these assays during pregnancy.

A full history and physical exam, along with a thorough review of the laboratory results can help to distinguish SH from these other conditions.

Key laboratory and imaging tests

The key laboratory tests needed for the diagnosis of subclinical hyperthyroidism (SH) are thyroid function tests, specifically TSH, free T4 and total or free T3. SH is associated with a low (or suppressed) TSH with normal free T4 and normal total T3. The laboratory studies should be repeated to confirm the diagnosis.

Once the diagnosis has been confirmed, then the etiology of the patient’s SH needs to be determined. Thyroid antibodies can help to distinguish Graves’ disease from other causes of SH. The most specific antibody for Graves’ disease is thyroid stimulating immunoglobulin (TSI), which can be measured by most reference laboratories.

In patients with suspected thyroiditis, it might be helpful to obtain thyroid peroxidase antibody and thyroglobulin antibody levels, as they are often positive in cases of thyroiditis. However, they are also usually positive in patients with Graves’ disease.

In addition to laboratory studies, imaging tests are useful in determining the etiology of SH. Obtaining a thyroid ultrasound offers information about the overall structure and characteristics of the thyroid gland. In patients with Graves’ disease or thyroiditis, the thyroid gland is usually enlarged and heterogeneous in appearance. In patients with multinodular goiter and/or solitary autonomously functioning nodule, the thyroid ultrasound can characterize the number and size of the nodules.

Even more specific than a thyroid ultrasound is a 4-6 and 24-hour radioactive iodine (I-123) uptake and scan of the thyroid gland. The results of the scan can be very helpful in elucidating the etiology of the patient’s SH. If the 24-hour uptake and scan shows diffuse uptake throughout the thyroid gland and an elevated uptake then the likely diagnosis is Graves’ disease. If the uptake and scan shows “hot nodule(s)” or specific areas of increased uptake, then this would fit a diagnosis of multinodular goiter or a solitary autonomously functioning nodule. Finally, if the uptake and scan shows decreased uptake in the thyroid, then the likely diagnosis is thyroiditis or excess exogenous thyroid hormone ingestion.

Other tests that may prove helpful diagnostically

In addition to the symptoms and signs mentioned above, subclinical hyperthyroidism (SH) can have specific effects on the cardiovascular system and on bone metabolism. Therefore, other testing that may be helpful in deciding how to manage a particular patient with SH would include a cardiovascular and a bone mineral density evaluation.

Several studies have identified an association between SH and atrial fibrillation. Patients with SH have a 2.8-5-fold increased risk of developing atrial fibrillation; this risk is higher in patients over the age of 60 and in patients with completely suppressed (undetectable) TSH levels. Treatment of SH has been found to improve this risk. The relationship is not as clear between SH and other types of cardiovascular disease. In patients with a history of atrial arrhythmias or underlying heart disease and in patients over the age of 60, it would be reasonable to consider performing a cardiac evaluation.

This evaluation could include any or all of the following: electrocardiogram, ambulatory Holter monitor, and echocardiogram. The findings of this evaluation would help to distinguish which patients are more likely to benefit from treatment.

There is evidence that postmenopausal women with SH have increased bone turnover and decreased bone mineral density. These changes may lead to an increase in fracture risk. Treatment of SH appears to improve bone mineral density. In women, especially those with risk factors for osteoporosis, performing a bone mineral density evaluation can help in deciding which patients are the best candidates for SH treatment.

Management and treatment of the disease

When to consider treatment for subclinical hyperthyroidism

The first step in the management of subclinical hyperthyroidism (SH) is to repeat the thyroid function tests (TSH, free T4 and total T3) in perhaps 2 to 4 weeks to determine whether the SH is persistent. Once persistent SH has been established, then patients should be evaluated in an individual manner to determine treatment. Not all patients with SH require treatment. The decision is based on various factors, including TSH level, age of the patient, and coexisting conditions.

In most instances, the normal TSH range is approximately 0.4-4.5 mIU/L. A low TSH level is between 0.1-0.4 mIU/L and a suppressed TSH level is one below 0.1 mIU/L. In determining treatment for SH, it is important to know whether the patient’s TSH is low or suppressed.

Patients with TSH levels below 0.1 mIU/L are more likely to have complications due to their SH such as atrial fibrillation, bone loss and conversion to overt hyperthyroidism. Therefore, in certain groups of patients with TSH below 0.1 mIU/L, treatment should be strongly considered. These groups include patients age 60 or above, patients who have or are at risk for osteoporosis or heart disease, and patients with hyperthyroid symptoms.

In patients with TSH levels between 0.1-0.4 mIU/L, the link between SH and the complications listed above is less clear. Therefore, treatment is generally not recommended in these patients. However, if patients are age 60 or above, have cardiac disease or hyperthyroid symptoms, then treatment can be considered.

The decision to treat SH should be individualized and made only after a full discussion with the patient.

Even if patients are not being treated for SH, their thyroid function tests should be followed periodically to monitor for resolution or conversion to overt hyperthyroidism. In addition, they should avoid high doses of iodine, such as exposure to intravenous radiocontrast material, as this could exacerbate their hyperthyroidism.

How to treat patients with subclinical hyperthyroidism

Treatment varies based on the etiology of SH. In patients with SH due to subacute or postpartum thyroiditis, antithyroid drug treatment is contraindicated since the excess serum thyroid hormone concentrations result from release of stored hormone and not increased synthesis. Also, the SH is transient. However, if the patient is symptomatic, then beta-blockers can be started and titrated to the smallest dose needed to control the symptoms.

In patients with SH due to Graves’ disease, treatment options include thionamides or radioiodine. There are two readily available thionamides: methimazole and propylthiouracil (PTU). Because PTU has been linked to rare but fatal cases of hepatotoxicity (especially in children and pregnant women), methimazole is the first-line thionamide for use in patients with hyperthyroidism (except in pregnant patients during their first trimester). In treating SH, low-dose methimazole is usually sufficient.

Possible side effects of methimazole include neutropenia, abnormal liver enzymes (cholestatic hepatitis), and skin rash. Methimazole should not be used in the first trimester of pregnancy. If used during the first trimester of pregnancy, large doses of methimazole can cause aplasia cutis and esophageal or choanal atresia. Thus, PTU is the preferred thionamide for patients in their first trimester of pregnancy. However, pregnant patients with SH rarely need treatment with thionamides (please see below).

Once patients with SH are started on thionamide therapy, their laboratory studies (e.g., complete blood count [CBC], complete metabolic profile, FT4, TT3 or FT3, and TSH) should be routinely monitored to evaluate for changes in thyroid function tests and for potential side effects. After the patient becomes euthyroid, it would be reasonable to decrease or stop therapy with thionamides to evaluate whether the SH has resolved. Measurement of thyroid antibodies and thyroid stimulating immunoglobulins also may be useful to predict remission.

The other therapeutic option for patients with SH due to Graves’ disease is radioiodine therapy. It is more definitive than therapy with thionamides. The dose of radioiodine can be calculated from a 24-hour radioactive iodine uptake and scan of the thyroid gland. Once the dose is administered, it may take 2-3 months for the patient to become euthyroid. In the majority of cases, the patients eventually become hypothyroid and require supplementation with thyroid hormone, so this is necessary to discuss with the patient prior to therapy.

Patients with SH due to nodular disease, either multinodular goiter or solitary autonomously functioning nodule, are good candidates for radioiodine therapy. Radioiodine therapy is preferred because nodular disease tends to be persistent, making definitive therapy beneficial. In addition, the vast majority of patients with nodular goiter have resolution of their hyperthyroidism after one dose of radioiodine therapy. The possibility of thyroid cancer should be ruled out by a fine needle aspiration biopsy of hypofunctioning nodules. If the nodular thyroid is very large, definitive thyroid surgery remains an option.

As mentioned above, pregnant patients have different normal reference ranges for TSH based on trimester. The reference ranges should be provided by individual laboratories. If they are not provided, then it is reasonable to use the following:

  • first trimester reference range for TSH, 0.1-2.5 mIU/L
  • second trimester reference range for TSH, 0.2-3.0 mIU/L
  • third trimester reference range for TSH, 0.3-3.0 mIU/L

In addition, pregnant patients also tend to have slightly high or high-normal free T4 index and total T3 levels due to physiologic changes. Because of these alterations in thyroid function tests, it is unusual for pregnant patients to be diagnosed with SH. If they do have laboratory and clinical results suggestive of SH and symptoms of SH, then it is reasonable to start treatment with low-dose beta-blockers.

Again, not all patients with SH require treatment. Deciding whether to treat and how to treat is a choice that should be made in collaboration with the patient. Ultimately, the most important step in the management of SH is monitoring for transition to overt hyperthyroidism, which occurs in approximately 0.5-1% of cases. In patients with overt hyperthyroidism, treatment is necessary.

What’s the Evidence?/References

Cooper, DS. “Approach to the patient with subclinical hyperthyroidism”. J Clin Endocrinol Metab. vol. 92. 2007. pp. 3-9. (The author provides a practical approach to managing a patient with subclinical hyperthyroidism. The article reviews the effects of subclinical hyperthyroidism on the cardiovascular system, bone health and quality of life. All of this information contributes to a useful management algorithm, which is provided at the end of the article.)

Bahn, RS, Burch, HB, Cooper, DS, Garber, JR, Greenlee, MC. “Hyperthyroidism and other causes of thyrotoxicosis: Management guidelines of the American Thyroid Association and American Association of Clinical Endocrinologists”. Thyroid. vol. 21. 2011. pp. 593-646. (This article comprehensively reviewed the evidence for the management of hyperthyroidism. They combined this evidence with clinical experience, and they present a total of 100 recommendations, looking at all aspects of hyperthyroidism management. The authors also make a point of mentioning the current controversies in the management and treatment of hyperthyroidism, and they give appropriate justification for all of their recommendations and guidelines.)

Surks, MI, Ortiz, E, Daniels, GH, Sawin, CT, Col, NF. “Subclinical thyroid disease. Scientific review and guidelines for diagnosis and management”. JAMA. vol. 291. 2004. pp. 228-38. (This article provides a thorough review of subclinical thyroid disease including definitions of subclinical hypothyroidism and subclinical hyperthyroidism, data on epidemiology, recommendations on how to appropriately evaluate and treat patients, analysis of the risk and benefits of treatment, and an expert opinion on whether population-based screening is needed. The expert panel, using the US Preventive Task Force criteria, recommends against population-based screening for thyroid disease.)

Biondi, B, Cooper, DS. “The clinical significance of subclinical thyroid dysfunction”. Endocrine Reviews. vol. 29. 2008. pp. 76-131. (The authors of this article present a detailed look at subclinical thyroid disease, both subclinical hypothyroidism and subclinical hyperthyroidism. They address a variety of topics. For subclinical hyperthyroidism, the authors describe the etiology, differential diagnosis, prevalence, natural history, symptoms, risks, effects of treatment, and ultimately, treatment guidelines on this subject.)

Wartofsky, L. “Management of subclinical hyperthyroidism”. J Clin Endocrinol Metab. vol. 96. 2011. pp. 59-61. (The author provides a brief but effective editorial on the treatment of subclinical hyperthyroidism and stresses the importance of considering whether treatment will benefit the patient.)

Sawin, CT, Geller, A, Wolf, PA, Belanger, AJ, Baker, E. “Low serum thyrotropin concentrations as a risk factor for atrial fibrillation in older persons”. N Engl J Med. vol. 331. 1994. pp. 1249-52. (This article studied the risk of developing atrial fibrillation in patients age 60 or older with low serum thyrotropin levels. They followed 2,007 participants for 10 years and found that in persons with serum thyrotropin levels equal to or below 0.1 mU per L, the incidence of atrial fibrillation was 28 percent compared with 11 percent in persons with normal thyrotropin levels. Based on this data, the authors conclude that persons age 60 or older with low serum thyrotropin levels have an approximately 3-fold higher relative risk for developing atrial fibrillation compared to those with normal thyrotropin levels.)

Auer, JA, Scheibner, P, Mische, T, Langsteger, W, Eber, O. “Subclinical hyperthyroidism as a risk factor for atrial fibrillation”. Am Heart J. vol. 142. 2001. pp. 838-42. (The authors of this article also investigated the risk of developing atrial fibrillation in more than 20,000 participants, assigned to one of three groups based on their thyroid function tests. Both the overt hyperthyroidism group and the subclinical hyperthyroidism group had higher percentages of participants with atrial fibrillation compared with the euthyroid group (13.8% and 12.7% respectively compared with 2.3%), and there was no significant difference between the overt hyperthyroidism group and the subclinical hyperthyroidism group. Overall, participants with overt hyperthyroidism and subclinical hyperthyroidism showed a 5-fold higher risk of atrial fibrillation than the euthyroid participants.)

Cappola, AR, Fried, LP, Arnold, AM, Danese, MD, Kuller, LH. “Thyroid status, cardiovascular risk, and mortality in older adults”. JAMA. vol. 295. 2006. pp. 1033-41. (This article examined multiple cardiovascular outcomes in participants age 65 and older, comparing groups based on their thyroid status. Outcomes included atrial fibrillation, coronary heart disease, cerebrovascular disease, cardiovascular death and all-cause death. The group with subclinical hyperthyroidism had a higher incidence of atrial fibrillation than the euthyroid group. No other differences were seen between the subclinical hyperthyroidism group and the euthyroid group for coronary heart disease, cerebrovascular disease, cardiovascular death or all-cause death.)

Bauer, DC, Ettinger, B, Nevitt, MC, Stone, KL. “Risk for the study of osteoporotic fractures. Risk for fracture in women with low serum levels of thyroid-stimulating hormone”. Ann Intern Med. vol. 134. 2001. pp. 561-8. (The authors of this article evaluated the effects of low thyroid-stimulating hormone levels on fracture risk in women age 65 and older. Women with TSH levels equal to or below 0.1 mU per L had an increased risk (3-fold) for hip fracture and an increased risk (4-fold) for vertebral fracture compared with women who had normal TSH levels.)

Faber, J, Jenson, IW, Petersen, L, Nygaard, B, Hedegus, L. “Normalization of serum thyrotropin by means of radioiodine treatment in subclinical hyperthyroidism: Effect on bone loss in postmenopausal women”. Clin Endocrinol (Oxf). vol. 48. 1998. pp. 285-90. (This article investigated whether treatment of subclinical hyperthyroidism by radioiodine therapy in postmenopausal women with nodular goiter prevents bone loss. While the study was non-randomized, they did evaluate two groups of women: one group was treated with radioiodine and the other group was followed without treatment. Assessment of bone mineral density showed that the untreated group had a continued fall in bone mass (approximately 2% per year), whereas the treated group had normalization of their TSH levels and did not show this same fall in bone mass.)