Subclinical Hypothyroidism:

April 1, 2004

ABSTRACT: Before a diagnosis of subclinical hypothyroidism is made, other causes of elevated thyroid-stimulating hormone (TSH), such as recovery from a serious acute illness, must be ruled out. Diagnosis is based on at least 2 assays performed over a period of several weeks. If the screening TSH levels are abnormal, measurement of serum free thyroxine is needed to confirm the diagnosis. Although patients who have TSH levels in the upper limit of the traditional normal range may be at higher risk for progression to hypothyroidism, no evidence exists for adverse health consequences. For patients with serum TSH levels in the range of 5.1 to 10 mIU/L, consider such factors as age, comorbidities, symptoms of hypothyroidism, pregnancy or anticipation of pregnancy, and the presence of goiter or antithyroid antibodies in the decision to treat. Thyroxine therapy is recommended for patients with serum TSH levels above 10 mIU/L.

As serum thyroid-stimulating hormone (TSH) assays have become standard for the evaluation of thyroid function, subclinical hypothyroidism is more frequently detected, especially among elderly persons. A recent report has outlined the recommendations of an expert panel on the basis of a critical review of the literature (Table 1).1

Yet controversy still persists and questions remain about the best approach to this disorder. What are the metabolic and clinical consequences-and who should be treat-ed? Here I address these and other questions that clinicians often ask about the management of subclinical hypothyroidism.

How is subclinical hypothyroidism defined?

The reference range for peripher- al thyroid hormone levels is rela- tively wide. The values for this range are derived from measure- ment in apparently healthy persons with no known thyroid disease; the 95th percentile was selected as the cutoff. Thus, 2.5% of values may fall either below or above the normal range. If the prevalence of an occult thyroid condition is higher than 5% in the population, many affected persons will have levels within the normal range.

The individual variation in peripheral thyroid hormone levels is much narrower than the laboratory reference range; serial measurements in the same person vary little over a 12-month period.2Thus, the serum free thyroxine level may be significantly decreased to below a person's specific normal range and still be in the normal reference range.

The feedback response of TSH to peripheral thyroid hormone variation is more sensitive and is log-linear. A slight reduction in serum thyroxine levels is associated with an elevation of serum TSH, despite a peripheral hormone level that is in the normal laboratory reference range. This condition is called subclinical hypothyroidism or, more appropriately, mild thyroidfailure.3,4

What is the normal range for TSH levels?

2The traditionally accepted reference range of normal serum TSH levels is 0.3 to 5 mIU/L. A number of professional organizations have suggested that the normal range should be 0.3 to 2.5 or 3 mIU/L.5,6 The argument in favor of this change is that TSH values in the population are not normally distributed, and when persons with antithyroid antibodies, goiter, or a strong family history of thyroid disease are excluded from normal value studies, the 95% TSH reference range shrinks to 0.3 to 2.5 or 3 mIU/L.5 Moreover, some epidemiologic studies have shown that patients whose TSH level is 3 to 5 mIU/L are more likely to test positive for antithyroid antibodies and are at higher risk for progression to clinical hypothyroidism.7

Nevertheless, in our Mayo Clinic practice, if this narrow TSH range were implemented, the result would be that 1 in 5 patients with no history of thyroid disease in whom TSH was measured would be classified as biochemically hypothyroid.8 In the absence of convincing evidence of metabolic abnormalities or clinical symptoms in patients with serum TSH values between 3 and 5 mIU/L-and the absence of benefits of lowering TSH levels in this range with thyroxine therapy9-such a change in the definition of the normal level will not be helpful to patients and may result in unnecessary therapies and investigations.

In my experience, for patients who are already being treated for hypothyroidism, the optimal TSH levelis 0.5 to 3 mIU/L. However, in asymptomatic persons-especially older ones-with serum TSH levels between 3 and 5 mIU/L, microtitration to lower levels of TSH is not recommended because of practical management issues, the risk of suppressing TSH production, and the adverse effects of subclinical hyperthyroidism.

What are the other causes of elevated TSH?

3Before a diagnosis of subclinical hypothyroidism is made, other causes of elevated TSH must be ruled out. These are listed in Table 2.In addition, in hospitalized patients with acute illness, moderate transient elevation of TSH may occur after an initial suppressed TSH in the recovery phase. Certain medications and medical conditions cause mild elevation of serum TSH. The effect of interference from heterophile antibodies is corrected in most assays. In thyroid-hormone resistance and TSH-producing tumors, the peripheral thyroid hormone levels are elevated, whereas in subclinical hypothyroidism free thyroxine levels are usually in the normal or low-normal range.

Diagnosis of subclinical hypothyroidism is not based on a single TSH measurement; at least 2 values within a period of several weeks are required. In addition, when screening serum TSH levels are abnormal, serum free thyroxine values are needed to confirm the diagnosis. An abnormally low serum thyroxine level usually results in significant elevation of TSH (usually higher than 20 mIU/L), which indicates overt hypothyroidism. If free thyroxine levels are low and TSH levels are not elevated, assay errors or pituitary hypothalamic disease may be a factor.

What causes subclinical hypothyroidism?

4The most common causes are autoimmune thyroid disease (Hashimoto thyroiditis) and inadequately treated hypothyroidism. Other causes include a history of therapy for hyperthyroidism, radiation therapy, cytokine therapy, and administration of iodine, lithium, or amiodarone. Subclinical hypothyroidism may also occur during the hypothyroid phase of silent, postpartum, or subacute thyroiditis.

Certain conditions increase the thyroxine requirements of hypothyroid patients who are taking this hormone. Elevated thyroxine requirements in pregnancy may result in subclinical hypothyroidism in hypothyroid patients who are receiving maintenance therapy. Concomitant administration of oral contraceptives, iron or calcium supplements, resines, or sucralfate, or consumption of a high-fiber diet or soy products interferes with absorption of administered thyroxine. Inflammatory bowel disease or any intercurrent malabsorption syndrome may also be a problem and will increase the requirement for administered thyroxine. Certain medications, such as hydantoin and carbamazepine, increase thyroxine metabolism; if the dosage is not adjusted, subclinical hypothyroidism may result. Therefore, TSH levels in patients taking these medications should be monitored 6 to 8 weeks after initiation of therapy and the thyroxine dosage adjusted if necessary.

How common is subclinical hypothyroidism?

5A screening of large numbers of patients at a health fair, reports on large numbers of patients tested in tertiary care centers, and epidemiologic studies show that mildly elevated levels-5.1 to 10 mIU/L-are the most common serum TSH abnormality.10-12 One third of all patients seen at the Mayo Clinic have serum TSH measured. Of those patients, 8% have serum TSH levels in that range; 3.3% have levels above 10 mIU/L.11 This is similar to findings in the Colorado Health Fair Study and other studies.3,10 The National Health and Nutrition Examination Survey (NHANES III) found that the prevalence of hypothyroidism was 4.6% (0.3% clinical and 4.3% subclinical) (Figure).12 The prevalence of subclinical hypothyroidism increases with age and is more common in women than men.

Is subclinical hypothyroidism a risk factor for cardiovascular disease?

6Overt hypothyroidism is a risk factor for coronary heart disease because of associated metabolic abnormalities. The atherogenic implications of subclinical hypothyroidism are less clear. Besides hyperlipidemia, other possible atherogenic factors, such as increased homocysteine and lipoprotein (a) (Lp[a]) levels, have been suggested.13 Homocysteine levels are elevated in overt-but not subclinical-hypothyroidism. Evidence for increased Lp(a) is also insufficient.14 A recent report indicates that levels of C-reactive protein are elevated in subclinical hypothyroidism, which raises the possibility of an associated inflammatory process.15

Adverse effects of subclinical hypothyroidism on cardiovascular function were also reported in a meta-analysis that included studies of patients who had hyperthyroidism.13 Among the effects on the heart and vascular smooth muscle that may contribute to left ventricular (LV) systolic dysfunction on effort are:

Slowed LV relaxation at rest.

Impaired LV diastolic filling with exercise.

Increased vascular tone at rest.

Impaired peripheral vasodilation with exercise.

In a 20-year follow-up of the population-based Whickham cohort study, the mortality from cardiovascular causes was no higher among participants who had subclinical hypothyroidism at baseline than among those who did not.16 (It should be noted that many of the patients with subclinical hypothyroidism were subsequently treated with thyroxine therapy and this may have changed the outcome.) However, in a case-control study, subclinical hypothyroidism in women whose mean age was 69 years was associated with a greater age-adjusted prevalence of aortic atherosclerosis (odds ratio, 1.7) and myocardial infarction (odds ratio, 2.3).17 The difference persisted after adjustment for body mass index, systolic and diastolic blood pressure, smoking, and total cholesterol and high-density lipoprotein levels.

However, the existing data are insufficient to formulate a conclusion, and carefully designed studies are necessary to settle this controversy.

What are the metabolic and clinical consequences of high normal TSH and subclinical hypothyroidism? Who should be treated?

7The consequences and need for treatment differ according to the degree of TSH elevation. Three distinct serum TSH categories are recognized.

Serum TSH levels of 3 to 5 mIU/L. Although persons who have serum TSH levels in the upper limit of normal may be at higher risk for progression to hypothyroidism,7,18 there is no firm evidence of adverse health consequences. In a 12-week randomized crossover study of patients with symptoms suggesting hypothyroidism who had TSH levels in the normal range, there was no difference between thyroxine-treated and control groups in cognitive and psychological function.9

Most experts do not recommend any intervention for this group of patients. A serum TSH in the upper limit of normal (higher than 3 mIU/L) may be an indication for follow-up serum TSH measurement in 1 year.

For symptomatic patients taking thyroxine, it is reasonable to attempt to lower serum TSH concentration to the 0.3 to 3 mIU/L range. However, there is no evidence that symptoms will improve. The benefits of increasing the dosage must be weighed against the possibility of adverse effects of overzealous thyroxine therapy, such as suppressed TSH levels and subclinical hyperthyroidism.

Serum TSH levels of 5 to 10 mIU/L. The Colorado Health Fair Study showed that there was a stepwise, but not statistically significant, increase in total cholesterol levels with increasing TSH level in this range.10 However, there are no randomized studies that show that levothyroxine therapy reduces cholesterol levels in this subgroup, and one study did not show any benefit.19 Most studies are not stratified for serum TSH level, and although improved symptoms and lipid levels have been shown in the group of patients with mild thyroid failure as a whole, results cannot be extended to the majority of subclinically hypothyroid patients in this subgroup.20,21 Some studies have shown cognitive, neuropsychiatric, cardiac, and muscular abnormalities in patients with subclinical hypothyroidism,13,22,23 but these findings need to be confirmed by further studies (Table 3).

The principal argument for treating patients in this group is that most will progress to overt hypothyroidism. There are no data to prove that therapy will result in an improved lipid profile, a reduction in cardiovascular risk, or an improvement in symptoms.1

However, treatment decisions need to be individualized. Consider such factors as the patient's age and comorbid conditions, as well as whether there is a gradual increase in TSH levels in follow-up and whether antithyroid antibodies, goiter, or symptoms of hypothyroidism are present. Factors that favor levothyroxine therapy in this group of patients are listed in Table 4.4 In symptomatic patients, a therapeutic trial of thyroxine is reasonable, but the likelihood of benefit is low.

In view of reports of reduced IQ in the offspring of subclinically hypothyroid pregnant women, thyroxine therapy is recommended for pregnant women and women who expect to become pregnant.24 I suggest thyroxine therapy for children and adolescents. I also advise therapy for patients with serum TSH levels above 8 mIU/L. The rationale for this approach is that these levels of serum TSH are associated with a 70% chance of progression to a TSH level of 10 mIU/L in 4 years. (V. Fatourechi, G. G. Klee, unpublished data.)

Serum TSH levels higher than 10 mIU/L. These patients may have elevated serum lipid levels. A meta-analysis and other studies have shown that thyroxine therapy results in an 8-mg reduction in low-density lipoprotein cholesterol levels.21,25,26 The factors that affect the response of lipid levels to levothyroxine therapy include higher levels of TSH, insulin resistance, higher levels of pre-therapy serum cholesterol, type III hyperlipidemia, and possibly genetic factors.

There is some evidence that mild thyroid failure exacerbates bipolar disorder and depression.22 There is also evidence that abnormalities of muscle function, nerve conduction, cardiac function,13 and cognitive function23 improve after normalization of TSH with thyroxine therapy. Sever- al randomized trials have shown benefits in various metabolic parameters, including cardiac, cognitive, and psychometric ones, after levothyroxine therapy.19-21,27,28

The consensus among thyroidologists is that patients with subclinical hypothyroidism and a serum TSH level above 10 mIU/L should be treated with thyroxine.29,30 The likelihood of improvement in lipid profile, sense of well-being, and cardiac function is high in this group.

What is the most appropriate therapy for subclinical hypothyroidism?

8For patients with serum levels of TSH in the range of 5 to 10 mIU/L in whom a decision is made to initiate therapy, and for all patients with subclinical hypothyroidism who have serum TSH concentrations above 10 mIU/L, levothyroxine is the therapy of choice. In my experience, the usual required daily thyroxine dose is 50 to 75 µg. Some endocrinologists recommend a full replacement dose because of anticipation that thyroid failure will progress. I prefer to start with 25 to 75 µg, depending on the patient's age, level of free thyroxine, and serum TSH level. Check the serum TSH level in 8 weeks; once a normal level is achieved, obtain another measurement in 6 months, and then annually. Once therapy is started, it is continued for life. The possibility of spontaneous normalization of thyroid function is 5% to 10%. In an occasional rare patient, autoimmune thyroid disease may change direction and hyperthyroidism may develop.31 n



1. Surks MI, Ortiz E, Daniels GH, et al. Subclinical thyroid disease: scientific review and guidelines for diagnosis and management. JAMA. 2004;291:228-238.

2. Andersen S, Pedersen KM, Bruun NH, Laurberg P. Narrow individual variations in serum T4 and T3 in normal subjects: a clue to the understanding of subclinical thyroid disease. J Clin Endocrinol Metab. 2002;87:1068-1072.

3. Cooper DS. Clinical practice. Subclinical hypothyroidism. N Engl J Med. 2001;345:260-265.

4. Fatourechi V. Mild thyroid failure [subclinical hypothyroidism]: to treat or not to treat? Compr Ther. 2002;28:134-139.

5. Baloch Z, Carayon P, Conte-Devolx B, et al. Laboratory medicine practice guidelines. Laboratory support for the diagnosis and monitoring of thyroid disease. Thyroid. 2003;13:3-126.

6. AACE Thyroid Task Force. American Association of Clinical Endocrinologists medical guidelines for clinical practice for the evaluation and treatment of hyperthyroidism and hypothyroidism. Endocr Prac. 2002;8:457-467.

7. Vanderpump MP, Tunbridge WM, French JM, et al. The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham Survey. Clin Endocrinol (Oxf). 1995;43:55-68.

8. Fatourechi V, Klee GG, Grebe SK, et al. Effects of reducing the upper limit of normal TSH values. JAMA. 2003;290:3195-3196.

9. Pollock MA, Sturrock A, Marshall K, et al. Thyroxine treatment in patients with symptoms of hypothyroidism but thyroid function tests within the reference range: randomised double blind placebo controlled crossover trial. BMJ. 2001;323:891-895.

10. Canaris GJ, Manowitz NR, Mayor G, Ridgway EC. The Colorado thyroid disease prevalence study. Arch Intern Med. 2000;160:526-534.

11. Fatourechi V, Lankarani M, Schryver PG, et al. Factors influencing clinical decisions to initiate thyroxine therapy for patients with mildly increased serum thyrotropin (5.1-10.0 mIU/L). Mayo Clin Proc. 2003;78:554-560.

12. Hollowell JG, Staehling NW, Flanders WD, et al. Serum TSH, T4, and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab. 2002;87: 489-499.

13. Biondi B, Palmieri EA, Lombardi G, Fazio S. Effects of subclinical thyroid dysfunction on the heart. Ann Intern Med. 2002;137:904-914.

14. Yildirimkaya M, Ozata M, Yilmaz K, et al. Lipoprotein(a) concentration in subclinical hypothyroidism before and after levo-thyroxine therapy. Endocr J. 1996;43:731-736.

15. Christ-Crain M, Meier C, Guglielmetti M, et al. Elevated C-reactive protein and homocysteine values: cardiovascular risk factors in hypothyroidism? A cross-sectional and a double-blind, placebo-controlled trial. Atherosclerosis. 2003;166:379-386.

16. Vanderpump MP, Tunbridge WM, French JM, et al. The development of ischemic heart disease in relation to autoimmune thyroid disease in a 20-year follow-up study of an English community. Thyroid. 1996;6:155-160.

17. Hak AE, Pols HA, Visser TJ, et al. Subclinical hypothyroidism is an independent risk factor for atherosclerosis and myocardial infarction in elderly women: the Rotterdam Study. Ann Intern Med. 2000; 132:270-278.

18. Vanderpump MP, Tunbridge WM. Epidemiolo-

gy and prevention of clinical and subclinical hypothyroidism. Thyroid. 2002;12:839-847.

19. Kong WM, Sheikh MH, Lumb PJ, et al. A 6-month randomized trial of thyroxine treatment in women with mild subclinical hypothyroidism. Am J Med. 2002;112:348-354.

20. Cooper DS, Halpern R, Wood LC, et al. L-Thyroxine therapy in subclinical hypothyroidism. A double-blind, placebo-controlled trial. Ann Intern Med. 1984;101:18-24.

21. Meier C, Staub JJ, Roth CB, et al. TSH-controlled L-thyroxine therapy reduces cholesterol levels and clinical symptoms in subclinical hypothyroidism: a double blind, placebo-controlled trial (Basel Thyroid Study). J Clin Endocrinol Metab. 2001;86:4860-4866.

22. Haggerty JJ Jr, Prange AJ Jr. Borderline hypothyroidism and depression. Annu Rev Med. 1995; 46:37-46.

23. Volpato S, Guralnik JM, Fried LP, et al. Serum thyroxine level and cognitive decline in euthyroid older women. Neurology. 2002;58:1055-1061.

24. Haddow JE, Palomaki GE, Allan WC, et al. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med. 1999;341:549-555.

25. Caraccio N, Ferrannini E, Monzani F. Lipoprotein profile in subclinical hypothyroidism: response to levothyroxine replacement, a randomized placebo-controlled study. J Clin Endocrinol Metab. 2002; 87:1533-1538.

26. Danese MD, Ladenson PW, Meinert CL, Powe NR. Clinical review 115: effect of thyroxine therapy

on serum lipoproteins in patients with mild thyroid

failure: a quantitative review of the literature. J Clin Endocrinol Metab. 2000;85:2993-3001.

27. Jaeschke R, Guyatt G, Gerstein H, et al. Does treatment with L-thyroxine influence health status in middle-aged and older adults with subclinical hypothyroidism? J Gen Intern Med. 1996;11:744-749.

28. Nystrom E, Caidahl K, Fager G, et al. A double-blind cross-over 12-month study of L-thyroxine treatment of women with 'subclinical' hypothyroidism. Clin Endocrinol (Oxf). 1988;29:63-75.

29. Chu JW, Crapo LM. The treatment of subclinical hypothyroidism is seldom necessary. J Clin Endocrinol Metab. 2001;86:4591-4599.

30. McDermott MT, Ridgway EC. Subclinical hypothyroidism is mild thyroid failure and should be treated. J Clin Endocrinol Metab. 2001;86:4585-4590.

31. Fatourechi V, Gharib H. Hyperthyroidism following hypothyroidism. Data on six cases. Arch Intern Med. 1988;148:976-978.