Prevalence and Consequences of Thyroid Dysfunction
The prevalence of thyroid dysfunction in adults in the general population ranges from 1 to 10 percent, and is even higher in selected groups (Samuels, 1998; Vanderpump, 2000; Wang and Crapo, 1997). Reasons for the variation include the testing site (community, health fair, medical clinic), age and sex of people tested, and method of assessment. In addition, it is not always clear whether people with known thyroid disease, those taking thyroid hormone or antithyroid therapy, or those who had other disorders that might affect thyroid function were included or excluded from the study group.
Subclinical hypothyroidism is defined as a high serum thyroid stimulating hormone (TSH) concentration and a normal serum free T4 concentration, and overt hypothyroidism as a high serum TSH concentration and a low serum free T4concentration. Subclinical hyperthyroidism is defined as a low serum TSH concentration and a normal serum free T4 concentration, and overt hyperthyroidism as a low serum TSH concentration and a high serum free T4 concentration. These definitions of “subclinical” and “overt” dysfunction are made purely by biochemical criteria. Persons with “subclinical” hyperthyroidism or hypothyroidism may display clear symptoms or signs of thyroid dysfunction while those with “overt” hyperthyroidism or hypothyroidism may show no other evidence of thyroid dysfunction. In all population surveys, most of the abnormal serum TSH concentrations are just outside the normal range, indicative of very mild thyroid dysfunction, and the frequency of more abnormal values is considerably lower.
Screening for thyroid dysfunction by measurement of serum TSH may yield several possible results (Figure 3-1). In the context of this volume, candidates for screening are people who have no recognized symptoms or signs of thyroid
dysfunction. High serum TSH values indicate the presence of subclinical or overt hypothyroidism, and low serum TSH values indicate the presence of subclinical or overt hyperthyroidism. The distinction between subclinical and overt thyroid dysfunction is made on the basis of measurements of serum free T4, as described in the preceding paragraph. Among people who are found by screening to have abnormal serum TSH concentrations, careful evaluation will often reveal symptoms or signs of thyroid dysfunction, as described in the preceding chapter. In general, those people whose screening tests indicate subclinical thyroid dysfunction more often have none or fewer of these symptoms and signs than do those people whose screening tests indicate overt thyroid dysfunction. However, there are no symptoms or signs that are unique to either group or that reliably distinguish persons with any of these forms of thyroid dysfunction from normal people.
In this chapter we summarize the results of studies using currently available tests done in people outside of hospitals, mostly in the United States. Prevalence rates have been rounded to the nearest 1 percent (unless <2 percent) for simplicity. The emphasis is on the results in older people. In making its assessments the committee relied to a much greater degree on information related to subclinical dysfunction than overt dysfunction because the population of screening positives much more resembles populations with subclinical dysfunction than populations with overt dysfunction. There are several reasons for this:
The vast majority (90%-95%) of persons with positive screening tests have subclinical dysfunction. The classification into overt and subclinical thyroid dysfunction is an arbitrary splitting of a continuum. The two groups are not necessarily distinct. Many subjects in the subclinical population will resemble subjects in the overt group.
The literature on patients with biochemically overt thyroid dysfunction is less relevant to the population of screened subjects who have biochemically overt dysfunction. Most of the information on biochemically overt dysfunction comes from patients who have marked symptoms or a long history of known thyroid disease. There is little information on patients with overt dysfunction who are discovered without prior suspicion. While not perfectly applicable, the literature on biochemically subclinical hypothyroidism is more relevant to the screening population; the subject became of interest over the question of whether an abnormal serum TSH level is significant in persons who have normal serum thyroxine levels.
The distinction between biochemically defined “subclinical” or “overt” thyroid dysfunction is less likely to be meaningful in terms of either burden of disease or potential to benefit from treatment in the screening population than in the entire population. In the overall population of persons with thyroid dysfunction defined by abnormal serum TSH concentrations, those who also have abnormal serum T4 will be much more likely to have clinically recognizable morbidity and potential to benefit from treatment. The screening population, however, is a much more homogeneous group; subjects, by definition, lack recognized morbidity. This does not mean that there are not people with “overt” dysfunction in the screening population whose symptoms were grossly overlooked and/or who would benefit significantly from treatment; it just means that a similar problem exists among people with biochemically “subclinical” dysfunction.
PREVALENCE OF SUBCLINICAL AND OVERT HYPOTHYROIDISM
The largest community-based study of thyroid function in the United States was carried out as part of the National Health and Nutrition Examination Survey (NHANES III) in 1988 to 1994 (Hollowell et al., 2002). Among 13,444 people aged 12 years and older remaining after all subjects with either known thyroid disease, antithyroid antibodies, or abnormal levels of serum T4 were excluded,
1.8 percent had a high serum TSH concentration. Exact percentages by age are not given, but based on extrapolation from a figure in the article, the prevalence of high serum TSH concentrations was 2 percent in people aged 60 to 69 years, 6 percent in those aged 70 to 79 years, and 10 percent in those aged 80 years and older. If only subjects with known thyroid disease are excluded from the NHANES sample, the prevalence of high serum TSH concentrations was 4.1 percent among 16,533 people. Extrapolated figures for the older population were 6.5 percent in people aged 60 to 69, 12 percent among ages 70 to 79, and 14 percent among subjects aged 80 years and older.
In the Colorado Thyroid Disease Prevalence Study, among 24,337 people (mean age 56 years) who attended a health fair, 2,067 (8 percent) had subclinical hypothyroidism and 103 (0.4 percent) had overt hypothyroidism (Canaris et al., 2000). Among those persons with high serum TSH concentrations, 74 percent had slightly high values (5.1 to 10 mU/L), and 26 percent had values >10 mU/L. The prevalence of high serum TSH values (independent of serum T4 values) was 16 percent among women aged 65 to 74 years and 21 percent among women aged 75 years and older; the comparable rates for men were 11 percent and 16 percent, respectively.
The results of other, smaller studies focused on older people were similar. For example, among 2,139 people aged 60 years or older enrolled in the Framingham Heart Study, 8 percent had subclinical hypothyroidism (women, 14 percent; men, 6 percent) (Sawin et al., 1985). Among 283 people aged 60 years and older attending a primary care geriatrics clinic in Oklahoma, the prevalence of subclinical hypothyroidism was 15 percent and that of overt hypothyroidism was 1 percent in both women and men (Bemben et al., 1994). Among 968 people aged 55 years or older attending a health fair in Michigan, 7 percent had subclinical hypothyroidism (the prevalence was 9 percent among those aged 65 to 74 years, and 10 percent among those aged 75 years and older) (Bagchi et al., 1990). With respect to patients enrolled in the Medicare program, 111 of 719 people (15 percent) living in New Mexico had high serum TSH concentrations (Lindeman et al., 1999).
CONSEQUENCES OF SUBCLINICAL HYPOTHYROIDISM
The consequences of subclinical hypothyroidism are symptoms attributed to thyroid deficiency, the presence of biochemical or physiological abnormalities that might be a threat to the person’s health and quality of life, and the risk of progression to more severe thyroid dysfunction. In the screening population, the consequences of biochemically overt hypothyroidism are similar, but presumably greater. We focus here on subclinical hypothyroidism because of its greater relevance to the screening population.
Symptoms and Signs of Subclinical Hypothyroidism
Patients with subclinical hypothyroidism may or may not have symptoms similar to those present in most patients with frank hypothyroidism. In the Colorado Thyroid Disease Prevalence Study, the frequency of several symptoms of hypothyroidism, such as dry skin, poor memory, and muscle weakness, was approximately 3 to 5 percent higher in those people with high serum TSH concentrations than in those with normal serum TSH concentrations, and those with a higher number of symptoms were more likely to have a high serum TSH concentration (Canaris et al., 2000). However, in the Oklahoma and New Mexico surveys cited earlier, there were few differences in symptoms or results of cognitive tests in the normal people and those with subclinical hypothyroidism (Bemben et al., 1994; Lindeman et al., 1999). Also, in the placebo-controlled studies of thyroid hormone treatment in patients with subclinical hypothyroidism, thyroid hormone treatment has proven little more effective than placebo in ameliorating any symptoms the patients might have had (summarized by Helfand in Appendix B).
Disorders Secondary to Subclinical Hypothyroidism
Patients with subclinical hypothyroidism may have slightly high serum total or low-density-lipoprotein (LDL) cholesterol concentrations, and therefore might be at increased risk for atherosclerosis. The frequency of high serum total and LDL cholesterol concentrations in people with subclinical hypothyroidism has varied substantially in different studies, from no increase to a definite increase. In general, people with higher serum TSH concentrations have higher serum cholesterol concentrations (Canaris et al., 2000). Given the multiple factors (diet, activity, smoking status, heredity) that affect serum cholesterol values, perhaps more relevant is the extent to which serum cholesterol values change with thyroid hormone treatment in patients with subclinical hypothyroidism. A systematic review of the effect of thyroid hormone therapy in these patients revealed a mean decrease in serum cholesterol concentration of 7.9 mg/dl (0.2 mmol/L) and a mean decrease in serum LDL cholesterol concentration of 10 mg/dl (0.3 mmol/L) (Danese et al., 2000).
There is little evidence that patients with subclinical hypothyroidism have an increased risk of cardiovascular disease (Vanderpump et al., 1996). Some have abnormalities in left ventricular function, such a decrease in ejection fraction or prolonged diastolic relaxation time, as assessed by echocardiography (Biondi, 2002), but whether these changes have clinical consequences is not known. With respect to mortality from cardiovascular disorders (and all-cause mortality), the annual mortality rates for years 1 to 5 and the 10-year mortality rates among 76 people aged 60 years and older with subclinical hypothyroidism (some of whom
later received thyroid hormone) followed in a general practice in England were similar to the rates in England and Wales as a whole (Parle et al., 2001).
Risk of Progression of Subclinical Hypothyroidism to Overt Hypothyroidism
The risk of progression of subclinical hypothyroidism to overt hypothyroidism ranges from 3 percent to nearly 20 percent per year (Samuels, 1998; Vanderpump, 2000). In the only study in which adults living in the community (Whickham, United Kingdom) who had subclinical hypothyroidism were followed for a prolonged period (20 years), the average rate of progression to a clinical diagnosis of hypothyroidism in women was 3 percent per year in those with high serum TSH concentrations and 4 percent per year in those with high serum TSH concentrations and high serum antithyroid peroxidase antibody concentrations (Vanderpump et al., 1995). The risk of progression was higher in men, but there were far fewer cases at base line. In a follow-up study done in people being followed in a general practice in England, 13 of 73 patients (18 percent) aged 60 years and older who had subclinical hypothyroidism at the time of initial testing had biochemically overt hypothyroidism one year later (6 percent had normal concentrations at the latter time) (Parle et al., 1991).
The variation in rate of progression in different studies can be explained at least in part by variations in the cause of subclinical hypothyroidism. Although chronic autoimmune thyroiditis is undoubtedly the most common cause of both subclinical and overt hypothyroidism, other important causes are previous radioactive iodine (131I) or surgical treatment for hyperthyroidism; undertreatment of hypothyroidism; drugs, including lithium carbonate, iodine, and iodine-containing drugs; and the recovery phase of silent (painless) thyroiditis. The latter two situations are usually transient, as is chronic autoimmune thyroiditis occasionally. Treatment-related causes of hypothyroidism would not be relevant to screening except in cases where radioactive iodine was given years earlier and forgotten. There are no studies of people with subclinical hypothyroidism in which the causes were carefully investigated.
The degree of elevation in serum TSH concentration is an important determinant of progression. In the Whickham study cited earlier, the probability of clinically diagnosed hypothyroidism during the 20-year follow-up period was 0.3 percent per year among women with a baseline serum TSH concentration of 6 mU/L and 1 percent per year among those with a baseline serum TSH concentration of 12 mU/L (Vanderpump et al., 1995).
Thyroid hormone therapy prevents hypothyroidism but, based on the above rates of progression, many patients will never need therapy. This applies particularly to those people with only slightly high serum TSH concentrations, who constitute the largest subgroup of people at potential risk.
PREVALENCE OF SUBCLINICAL AND OVERT HYPERTHYROIDISM
The prevalence of hyperthyroidism is lower than that of hypothyroidism. In the NHANES III study, 1.6 percent of the 13,444 people aged 12 years and older had low serum TSH concentrations (Hollowell et al., 2002). Based on extrapolation from a figure in the article, the prevalence of low serum TSH concentrations was 1 percent in people aged 60 to 69 years and 70 to 79 years, and 3 percent in those aged 80 years and older. In the Colorado Thyroid Disease Prevalence Study, 219 of 24,337 people (0.9 percent) had subclinical hyperthyroidism and 22 (0.1 percent) had biochemically overt hyperthyroidism (Canaris et al., 2000). Among 2,007 people aged 60 years and older enrolled in the Framingham Heart Study, 248 (12 percent) had subclinical hyperthyroidism (Sawin et al., 1994), whereas among 968 people aged 55 years and older attending a health fair in Michigan, only 33 (3 percent) had subclinical hyperthyroidism (Bagchi et al., 1990).
CONSEQUENCES OF SUBCLINICAL HYPERTHYROIDISM
The burdens of subclinical hyperthyroidism are symptoms attributed to thyroid hormone excess, the presence of biochemical or physiological abnormalities that might be a threat to the person’s health and quality of life, and the risk of progression to more severe disease. The burdens of overt hyperthyroidism detected by screening are similar, but presumably greater. We focus on subclinical hyperthyroidism because of greater relevance to the screening population.
Symptoms and Signs of Subclinical Hyperthyroidism
People with subclinical hyperthyroidism may or may not have the symptoms that are present in most patients with frank hyperthyroidism or some impairment in physical and mental activity (Biondi et al., 2000; Marqusee et al., 1998). In the only trial of antithyroid drug therapy of subclinical hyperthyroidism, in 20 people (age range, 24 to 48 years), symptoms of hyperthyroidism improved in the treated patients, but untreated patients did not receive a placebo and the observers were probably aware of treatment group assignment (Yonem, 2002).
Disorders Secondary to Subclinical Hyperthyroidism
In people with subclinical hyperthyroidism, the values for measurements of cardiovascular function such a pulse rate, cardiac ejection fraction, and ventricular contractility and mass are intermediate between the values in normal subjects and patients with overt hyperthyroidism (Biondi, 2002). Cardiovascular function becomes more normal in people with subclinical hyperthyroidism during antithyroid therapy (or a reduction in dose of thyroid hormone). The only defined
long-term cardiovascular risk of subclinical hyperthyroidism is an increase in risk of atrial fibrillation. Among 2,007 people aged 60 years or older followed for 10 years, 28 percent of those with serum TSH concentrations =0.1 mU/L at base line later had atrial fibrillation, as compared with 12 percent of those with serum TSH concentrations of >0.1 to 0.4 mU/L and 8 percent of those with normal serum TSH concentrations (>0.4 to 5 mU/L) (Sawin et al. 1994).
There is a tendency for a decrease in bone density in people with subclinical hyperthyroidism, especially in postmenopausal women (Greenspan and Greenspan, 1999). This decrease is minimal in those women with only slightly low serum TSH concentrations, and is more marked in those with lower concentrations. The clinical correlate of low bone density is fracture. In a prospective study of 686 white women aged 66 years and older followed for 6 years, those with serum TSH concentrations of =0.1 mU/L at baseline had a threefold increase in risk of hip fracture (hazard ratio 3.6). Those with serum TSH concentrations of >0.1 to <0.5 mU/L had a smaller increase (hazard ratio 1.9), as compared with women with normal serum TSH concentrations (0.5 to 5.5 mU/L) (Bauer et al., 2001). The women who had serum TSH concentrations of =0.1 mU/L also had an increase in vertebral fracture. A large majority (86 percent) of the women with low serum TSH concentrations at baseline in this cohort were taking thyroid hormone.
Mortality may be increased in people with subclinical hyperthyroidism in the first years after detection. Among 70 people aged 60 years and older with sub-clinical hyperthyroidism followed in a general practice in England, the 10-year cardiovascular and overall mortality rates were similar to the rates in England and Wales as a whole, but the numbers of deaths were higher than expected at 2, 3, 4, and 5 years of follow-up (standardized mortality ratios 1.7 to 2.2) (Parle et al., 2001).
Risk of Progression of Subclinical Hyperthyroidism to Overt Hyperthyroidism
The overall rate of progression of subclinical hyperthyroidism to overt hyperthyroidism ranges from 1 percent to 15 percent per year (summarized in Samuels, 1998, and Marqusee et al., 1998). The studies were all small (the largest was of 66 people), and the percentage of people who had normal serum TSH concentrations when retested varied, ranging from 14 percent to 61 percent (summarized in Marqusee et al., 1998). These wide variations undoubtedly reflect differences in the cause of the subclinical hyperthyroidism, and indeed the likelihood that in some people, the low serum TSH concentration was caused by transient non-thyroidal illness.
The causes of subclinical hyperthyroidism are the same as the causes of overt hyperthyroidism. They are Graves’ disease (autoimmune hyperthyroidism), nodular goiter, silent (painless) thyroiditis, subacute (painful) thyroiditis, iodine- and
drug-induced hyperthyroidism, and thyroid hormone therapy. However, their relative frequency differs considerably. Among people with subclinical hyperthyroidism, the proportion taking thyroid hormone is high. For example, in the Colorado Thyroid Disease Prevalence Study, 316 of the 535 people with subclinical hyperthyroidism (59 percent) were taking thyroid hormone; in other studies the proportion ranged from 5 percent to 62 percent (summarized in Marqusee et al., 1998). These data are not relevant to screening, but they indicate that a high proportion of people with subclinical hyperthyroidism can be treated by a reduction in thyroid hormone dose. In the only study in which the cause of spontaneously occurring subclinical hyperthyroidism was carefully sought, 12 of 24 people (50 percent) had silent thyroiditis or iodine-induced hyperthyroidism, and presumably recovered soon thereafter (Charkes, 1996). These people were identified in the course of medical practice, not by screening or clinic survey. Nonetheless, the results would seem to make documentation of persistent subclinical hyperthyroidism mandatory before considering whether intervention is warranted.
Thyroid dysfunction is common, especially in elderly people. Most people found to have thyroid dysfunction in surveys have subclinical thyroid dysfunction, in particular subclinical hypothyroidism. Among people with subclinical thyroid dysfunction, most have very small increases or decreases in serum TSH concentrations. When asked, some of these people with subclinical thyroid dysfunction have symptoms that are compatible with, though not specific for, thyroid dysfunction or have another indication for testing for thyroid dysfunction. Some people have biochemical or physiological abnormalities that are ameliorated by thyroid hormone therapy, in the case of people with subclinical hypothyroidism, or antithyroid therapy, in the case of subclinical hyperthyroidism. Among people with thyroid dysfunction, therapy may have beneficial effects on intermediate outcomes, such as reduction in serum lipid concentrations and improvement of myocardial contractility. However, appropriate therapy has not been proven to alter long-term morbidity or mortality in people with subclinical thyroid dysfunction. Similarly, while it is accepted that treatment will benefit patients with biochemically overt thyroid dysfunction who present with significant symptoms or complications, the lack of well designed studies makes it difficult to determine whether treatment would provide significant net benefit in persons who have biochemically defined overt thyroid dysfunction but little evidence of illness; the potential for harms is similar but potential for benefit is less. These uncertainties contribute to the difficulty in assessing the value of a screening program for thyroid dysfunction.
Bagchi N, Brown TR, Parish RF. 1990. Thyroid dysfunction in adults over age 55 years. A study in an urban US community. Arch Intern Med 150(4):785–787.
Bauer DC, Ettinger B, Nevitt MC, Stone KL, for the Study of Osteoporotic Fractures Research Group. 2001. Risk for fracture in women with low serum levels of thyroid-stimulating hormone. [Article]. Annals of Internal Medicine 134(7):561–568.
Bemben DA, Winn P, Hamm RM, Morgan L, Davis A, Barton E. 1994. Thyroid disease in the elderly. Part 1. Prevalence of undiagnosed hypothyroidism. J Fam Pract 38(6):577–582.
Biondi B, Palmieri EA, Fazio S, Cosco C, Nocera M, Sacca L, Filetti S, Lombardi G, Perticone F. 2000. Endogenous subclinical hyperthyroidism affects quality of life and cardiac morphology and function in young and middle-aged patients. J Clin Endocrinol Metab 85(12):4701–4705.
Biondi B, Palmieri EA, Lombardi G, Fazio S. 2002. Effects of subclinical thyroid dysfunction on the heart. Ann Intern Med 137:904–914.
Canaris GJ, Manowitz NR, Mayor G, Ridgway EC. 2000. The Colorado thyroid disease prevalence study. Arch Intern Med 160(4):526–534.
Charkes ND. 1996. The many causes of subclinical hyperthyroidism. Thyroid 6(5):391–396.
Danese MD, Ladenson PW, Meinert CL, Powe NR. 2000. 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 85(9):2993–3001.
Greenspan SL, Greenspan FS. 1999. The effect of thyroid hormone on skeletal integrity. Ann Intern Med 130(9):750–758.
Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA, Braverman LE. 2002. Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 87(2):489–499.
Lindeman RD, Schade DS, LaRue A, Romero LJ, Liang HC, Baumgartner RN, Koehler KM, Garry PJ. 1999. Subclinical hypothyroidism in a biethnic, urban community. J Am Geriatr Soc 47(6):703–709.
Marqusee E, Haden ST, Utiger RD. 1998. Subclinical thyrotoxicosis. Endocrinol Metab Clin North Am 27(1):37–49.
Parle JV, Franklyn JA, Cross KW, Jones SC, Sheppard MC. 1991. Prevalence and follow-up of abnormal thyrotrophin (TSH) concentrations in the elderly in the United Kingdom. Clin Endocrinol 34(1):77–83.
Parle JV, Maisonneuve P, Sheppard MC, Boyle P, Franklyn JA. 2001. Prediction of all-cause and cardiovascular mortality in elderly people from one low serum thyrotropin result: A 10-year cohort study. Lancet 358(9285):861–865.
Samuels MH. 1998. Subclinical thyroid disease in the elderly. Thyroid 8(9):803–813.
Sawin CT, Castelli WP, Hershman JM, McNamara P, Bacharach P. 1985. The aging thyroid. Thyroid deficiency in the Framingham Study. Arch Intern Med 145(8):1386–1388.
Sawin CT, Geller A, Wolf PA, Belanger AJ, Baker E, Bacharach P, Wilson PW, Benjamin EJ, D’Agostino RB. 1994. Low serum thyrotropin concentrations as a risk factor for atrial fibrillation in older persons. N Engl J Med 331(19):1249–1252.
Vanderpump MP, Tunbridge WM, French JM, Appleton D, Bates D, Clark F, Grimley Evans J, Hasan DM, Rodgers H, Tunbridge F, et al. 1995. The incidence of thyroid disorders in the community: A twenty-year follow-up of the Whickham Survey. Clin Endocrinol 43(1):55–68.
Vanderpump MP, Tunbridge WM, French JM, Appleton D, Bates D, Clark F, Grimley Evans J, Rodgers H, Tunbridge F, Young ET. 1996. The development of ischemic heart disease in relation to autoimmune thyroid disease in a 20-year follow-up study of an English community. Thyroid 6(3):155–160.
Vanderpump MP, Tunbridge WM. 2000. The epidemiology of thyroid diseases. In Braverman LE Utiger RD, eds. The Thyroid: a Fundamental and Clinical Text. 8th ed. Philadelphia: Lippincott Williams and Wilkins. Pp. 467–473.
Wang C, Crapo LM. 1997. The epidemiology of thyroid disease and implications for screening. Endocrinol Metab Clin North Am 26(1):189-218.
Yonem O, Dokmetas HS, Aslan SM, Erselcan T. 2002. Is antithyroid treatment really relevant for young patients with subclinical hyperthyroidism? Endocr J 49:307–314.