Clark T. Sawin*
It is natural to wonder, as we get older, whether or not there is some change with age that might be reversed and thereby prolong our lives or, at least, make the later years more pleasant. Some of the ancients thought that the changes seen in older persons were due to a gradual drying out or cooling of the body; possible remedies were thus the addition of moisture or warmth to one's daily regimen, although neither seemed to actually slow the aging process (1).
This type of reasoning by analogy persisted into the late 19th and early 20th centuries. A prime example is the similarity and equation of the clinical findings in severe hypothyroidism and the changes that occur as we age. In the 1870s and 1880s, many saw the new and striking description of myxoedema (2,3), then the term for what we now call hypothyroidism, as a kind of premature aging. When, in 1888, the Clinical Society of London published its famous report (4), which concluded that the cause of this peculiar disorder was a “destructive change of the thyroid gland,” Victor Horsley, who had done the experimental thyroidectomies for the report, had already concluded that “the symptoms of mere senility may be accounted for by the loss of the functions of the thyroid body” (5). With this background, it is no great surprise that, after the discovery of the successful therapy of hypothyroidism with thyroid extract in 1891 (6), others used the same therapy to try to prevent aging or at least to prevent one of its concomitants, atherosclerotic disease, until well into the 20th century (7,8,9,10). When there were no specific tests of thyroid function, such attempts were reasonable.
The underlying assumption of this type of thinking is that hypothyroidism is a natural consequence of becoming older. If this were so, then one ought to be able to show a clear decrease in thyroid function in all, or almost all, older persons. If, on the other hand, the presence of hypothyroidism in older persons is an incidental disease and so a confounder in the study of the aging process rather than its consequence, then there may or may not be a decrease in thyroid function as we age. As with most studies of aging, the reality may be that the problem of whether or not thyroid secretion falls with age is not, strictly speaking, soluble. For example, if there are two variables, age and disease, neither of which can be held constant and each of which might affect the thyroid gland, any change in glandular function might be due to either or both variables, and one may not be able to distinguish the effects of the two. The problem is common in epidemiology (e.g., the definition of endemic goiter largely depends on an arbitrary selection of a prevalence rate above which goiter is considered endemic, i.e., all are subject to it, and below which it is not, i.e., goiter is sporadic and peculiar to an individual).
Why should any thyroid changes be a problem if the thyroid gland slows down only in old persons? The answer is clear: life span is increasing in many countries, and any systematic or episodic change in thyroid function that affects longevity or the quality of life is important. Note, for example, that a person 80 years old will, on average, live 7 or 9 more years (men or women, respectively) (11).
The remainder of this chapter discusses issues related to changes in thyroid function with increasing age in humans. The conclusions need to be somewhat tentative because of the confounders just noted; further, almost all the data are cross-sectional, the definition of “normal” or “healthy” is rarely explicit, and cohort or survival effects are usually not addressed. Better answers would come from careful longitudinal studies of the same persons followed for many years, but as yet there are no such studies.
SERUM THYROXINE CONCENTRATIONS
The serum concentration of thyroxine (T4), the principal circulating thyroid hormone, seems to change little with increasing age. This was first shown indirectly with measurement of the serum concentration of protein-bound iodine (PBI), which measures mainly T4 (12,13), and then with assays of serum total T4 and estimates of serum free T4 concentrations (13,14,15). The data are sometimes confounded by the inadvertent inclusion of those with subclinical hypothyroidism, with a resulting decrease in the mean serum T4 or free T4 concentration (16,17,18). Other data that show a slight increase in serum T4 in women (19) or in free T4 estimates in men and women (14,20,21) are less easy to explain, but are likely of little clinical relevance in view of the generally unchanged or perhaps slightly lower serum thyrotropin (thyroid-stimulating hormone, TSH) concentrations in older persons (see later in the chapter). In a few instances of a lower serum free T4 index in older persons, the hypothesis has been raised of a resetting downward of the feedback threshold by which thyroid hormones shut off the secretion of TSH (22); however, the evidence for this idea is thin.
It is also possible that a lessened secretion of T4 in older persons might be due to a fall in the ability of the thyroid gland to respond to stimulation by TSH. Thyroid cell membranes from old rats, for example, seem to have a decreased response of adenylyl cyclase to TSH that matches the lower response of thyroid hormone secretion to TSH in this species (23). However, in humans, the response of serum thyroid hormones to TSH is the same in older and younger adults (24), and there is no decrease in the binding of TSH to human thyroid tissue from older persons (25). Further, if there were a fall in thyroid sensitivity to TSH in older persons, one would expect a rise in serum TSH, which seems not to occur in the absence of hypothyroidism (see later in the chapter).
Overall, there is no clinically important change in the serum concentration of T4 in older persons.
THYROID RADIOIODINE UPTAKE AND THYROXINE TURNOVER
Despite the stability of serum T4 concentration throughout the human age span, thyroid uptake of radioiodine is lower in older compared with younger men (26). This finding suggests that thyroid hormone secretion decreases with advancing age, a conclusion supported with the demonstration of a decrease in T4degradation in older men in the presence of a constant serum concentration of T4 (12). The only way the serum level could remain constant in the face of a decreased metabolic clearance of T4 have a concomitant decrease in T4 secretion from the thyroid gland. An early systematic review confirmed these conclusions for both men and women (27). The clinical corollary that one would expect with a decreased secretion of thyroid hormone with older age is that the replacement dose of T4 in patients with primary hypothyroidism should be lower, on average, in older than in younger persons. In fact, this is so (28,29,30), although the phenomenon may be somewhat dependent on the cause of the hypothyroidism (31).
In sum, T4 secretion decreases with older age but the serum concentration does not.
SERUM TRIIODOTHYRONINE CONCENTRATIONS
If there is less T4 secreted in older persons, the serum concentration of triiodothyronine (T3) might also be lower because most of the circulating T3 derives from circulating T4. Much would depend on the conversion rate of T4 to T3 and on the clearance of T3 itself.
At first, the data showed that older persons did have a lower serum concentration of T3 and, when estimated, of free T3 (16,17,32,33,34,35), although perhaps the findings might have applied only to men (36). Then came the recognition that illness sufficient to cause admission to a nursing home or to hospital is a common cause of a low serum concentration of T3 whereas, when selected for good overall health, older persons have no decrease in the serum level of T3 (37,38). More recent data indicate that, while there is a slightly lower serum concentration of total T3 in older men (though perhaps not in older women), age has no effect on the concentration of free T3 (20) and few healthy older persons have a value of serum free T3 outside the reference range for younger adults (39).
Data on healthy centenarians indicate that these truly old persons do indeed have a lower level of serum T3 but that the decrease does not occur until about the age of 80 years (40,41) and stays within the reference range, although even at this age some have not found such a decrease (42). Even here, however, the interpretation of the data is not clear, because there is almost certainly selection bias in comparing centenarians to middle-aged adults. Those who reach 100 years of age are a small group of survivors and are not typical of the population of persons over age 60 years. For example, there is a clear increase in the prevalence of thyroid antibodies above the age of 60 to 70 years, although this is not found in centenarians (43); the best interpretation is that few with thyroid antibodies reach 100 years of age.
On balance, there is probably a slight fall in the serum concentration of T3 with increasing age, but it is unlikely to be of clinical importance. If an individual has a clearly low serum T3 concentration, it is more likely due to nonthyroid illness and the associated blunting of the conversion of serum T4 to T3.
SERUM THYROTROPIN CONCENTRATIONS
Because part of the definition of “normal” often includes a serum concentration of TSH that is within the reference range for younger adults, it is not easy to say whether or not older persons have a high or low serum TSH level; persons with such levels are often excluded in a population study. Further, the prevalence of subclinical hypothyroidism rises with increasing age, especially in women; unless hypothyroidism is specifically sought and excluded prior to a study, the results will likely show a higher serum concentration of TSH in older compared with younger persons. Ideal studies might examine the serum TSH concentrations in older persons matched with younger persons on the basis of comparable serum concentrations of T4 and T3 and would follow, as noted earlier, the same population for some years; such studies are not available.
The available information is conflicting. Basal serum concentrations of TSH can be unchanged in older persons (24,35,38,44), slightly higher in older women (45) or in older men (18), or slightly lower in both older men and women (21,41).
Perhaps more relevant to the issue of a change in TSH secretion with age is the finding that, with multiple measurements made over 24 hours, the mean serum concentration of TSH is slightly but clearly lower in older men compared to younger ones, although the values in the older men generally remained within the reference range for younger adults (44). Further, there is a blunting of the nocturnal rise in TSH secretion in older men (46).
What of the serum TSH response to thyrotropin-releasing hormone (TRH) in older persons? This would be important to study because the rise in serum TSH after giving TRH is proportional to the basal level of TSH. Because changes in the basal level may be difficult to detect because of excessive noise in the assay, the magnified concentrations of TSH after TRH may better reflect the true basal value. Here, most data indicate that at least some older persons have a lesser response of serum TSH to TRH than do younger persons (21,35), although the decreased response may be found only in older women (38) or only in older men (32,45). Occasionally, investigators have not found any difference in the TSH response to TRH in older men (18). Some have attributed a diminished response of serum TSH to TRH in older persons to the presence of unsuspected nodular goiter and a mild degree of thyroid autonomy rather than to older age per se (47).
That there is some decrease in TSH secretion in older age is supported by the observation that serum concentrations of TSH in primary hypothyroidism are not as high in older persons as in younger ones (48). That this in turn might be related to a decrease in TRH secretion by the hypothalamus has some experimental support, in that TRH secretion in vitro is lower than expected in hypothalamic tissue excised from older rats (49).
Physiologically, a lower TRH and TSH secretion in older age makes sense in view of the lesser need for T4 secretion as a result of the decreased T4 turnover noted earlier. A clinical corollary of this idea is seen in the small but significant percentage of older persons who have a clearly low serum TSH concentration compared to younger adults but who do not have thyrotoxicosis (50).
Thus, there is probably a small decrease in TSH secretion in older persons that may be related to the decreased clearance of T4. This slight fall in TSH secretion is generally within the reference range for younger adults and is probably of little clinical importance.
THYROID HORMONE ACTION
Whether or not there is some change in the action of thyroid hormone in older persons, at least on selected tissues, is an old speculation that has only modest support. The issue is raised because how else can one explain the notable differences in the clinical presentation of thyrotoxicosis in older persons compared to younger ones (51,52), wherein it is often difficult to detect the disorder in older persons despite comparable elevations in serum thyroid hormone concentrations?
There is some evidence in older rats that there is a decrease in the sensitivity of some tissues, especially the heart, to thyroid hormone, and that this decrease is related to a parallel age-related diminution in β-adrenergic responsiveness (53). It is possible, though by no means proven, that a similar mechanism operates in humans. The opposite speculation, i.e., that there is an increase in tissue sensitivity to thyroid hormones in older persons, has also been raised, as in the mention earlier of a possible resetting of the pituitary's response to thyroid hormones. While there is little evidence for a change in tissue sensitivity to thyroid hormones, why the clinical presentation of thyrotoxicosis differs in older persons remains unexplained.
SCREENING AND CASE-FINDING
Physicians rarely do true “screening,” which is the testing of an entire community or population; they usually do “case-finding,” which is testing for a disease in patients who come to them with a complaint of one kind or another. In the case of thyroid failure, the presence of a risk factor for thyroid dysfunction (Tables 11A.1,11A.2) is sufficient reason to test for such dysfunction at any age and, of course, age itself is a risk factor for hypothyroidism. Because the clinical findings of hypothyroidism in older persons are vague and nonspecific, because the only reliable way to make the specific diagnosis is to measure serum TSH, because the number of older persons with hypothyroidism is reasonably high, and because there is an effective treatment, one could argue that on these grounds alone all older persons should be tested for hypothyroidism, i.e., that they should be screened. Nevertheless, a recent complete assessment of the available evidence does not support such screening of all older persons as a routine practice (54).
TABLE 11A.1. RISK FACTORS FOR HYPOTHYRODISM
Age >60 y
Other autoimmune disease
Diabetes mellitus (type 1)
Subacute thyroiditis (overt or silent)
Head/neck cancer (treated)
Family member with thyroid disease
Iodine (any form)
Routine tests (if previously done):
Thyroid tests (if previously done):
Slightly raised serum TSH concentration
High serum antithyroid peroxidase antibody concentration
TABLE 11A.2. RISK FACTORS FOR THYROTOXICOSIS
Family member with thyroid disease
Cigarette smoking (Graves' disease)
Congestive heart failure
Iodine (any form)
Radiographic contrast agent
Low serum TSH concentration
In general, there is little systematic change in thyroid function with increasing age in hymans. In older persons, there is a modest decrease in T4 secretion without a change in the circulating level of the hormone, a slight fall in circulating T3 concentration after the age of 80 years or so, and perhaps a slight fall in the secretion of TSH. There is no change in the thyroid gland's response to TSH in older persons and no proven decrease in tissue response to thyroid hormones. All of these changes, except possibly the fall in serum T3 concentrations, are generally within the reference ranges for younger adults, which can therefore be used to diagnose thyroid dysfunction in older persons. Thus, age-related changes in thyroid function in older persons do not have major effects on the tests used to detect thyroid disease. When there are abnormalities in these tests that would indicate disease in younger adults, thyroid dysfunction should be presumed present in older persons unless proven otherwise. There is as yet no solid explanation for the different clinical appearances of thyroid dysfunction in older persons as compared to younger ones. There are no firm data to support routine testing (“screening”) of all older persons for thyroid dysfunction, principally because data to support benefit are lacking; however, vigorous case-finding is sensible.
When a younger adult is treated with T4 for hypothyroidism for many years, there should be consideration of a decrease in the dose when that person reaches the age of 60 years or so.
1. Birren JE. History of gerontology. In: Encyclopedia of gerontology, Vol. 1. New York: Academic Press, 1996:655.
2. Gull WW. On a cretinoid state supervening in adult life in women. Trans Clin Soc London 1874;7:180.
3. Ord WM. On myxoedema, a term proposed to be applied to an essential condition in the “cretinoid” affection occasionally observed in middle-aged women. Med-Chir Trans 1878; 61:57.
4. Ord WM, Cavafy J, Durham AE, et al. Report of a committee of the Clinical Society of London nominated December 14, 1883, to investigate the subject of myxoedema. Trans Clin Soc Lond 1888;21[Suppl]:1–215.
5. Horsley V. Relation of the thyroid gland to general nutrition. Lancet 1886;1:3.
6. Murray GR. Note on the treatment of myxoedema by hypodermic injections of an extract of the thyroid gland of a sheep. BMJ 1891;2:796.
7. Barnes BO. Prophylaxis of ischaemic heart-disease by thyroid therapy. Lancet 1959;2:149.
8. Wren JC. Symptomatic atherosclerosis: prevention or modification by treatment with desiccated thyroid. J Am Geriatr Soc 1971;19:7.
9. Starr P. Atherosclerosis, hypothyroidism, and thyroid hormone therapy. Adv Lipid Res 1978;16:345.
10. Lev-Ran A. Thyroid hormones and prevention of atherosclerotic heart disease: an old-new hypothesis. Perspect Biol Med 1994;37: 487.
11. Manton KG, Vaupel JW. Survival after the age of 80 in the United States, Sweden, France, England and Japan. N Engl J Med 1995;333:1232.
12. Gregerman RI, Gaffney GW, Shock NW, et al. Thyroxine turnover in euthyroid man with special references to changes with age. J Clin Invest 1962;41:2065.
13. Braverman LE, Dawber NA, Ingbar SH. Observations concerning the binding of thyroid hormones in sera of normal subjects of varying ages. J Clin Invest 1966;45:1273.
14. Blum CJ, Lafont C, Ducasse M, et al. Thyroid function tests in ageing and their relation to associated nonthyroidal disease. J Endocrinol Invest 1989;12:307.
15. Runnels BL, Garry PJ, Hunt WC, et al. Thyroid function in a healthy elderly population: implications for clinical evaluation. J Gerontol A Biol Sci Med Sci 1991;46:B39.
16. Herrmann J, Rusche HJ, Kroll HJ, et al. Free triiodothyronine (T3) and thyroxine (T4) serum levels in old age. Horm Metab Res 1974;6:239.
17. Lipson A, Nickoloff EL, Hsu TH, et al. A study of age-dependent changes in thyroid function tests in adults. J Nucl Med 1979;20:1124.
18. Harman SM, Wehmann RE, Blackman MC. Pituitary-thyroid hormone economy in healthy aging men: basal indices of thyroid function and thyrotropin responses to constant infusions of thyrotropin releasing hormone. J Clin Endocrinol Metab 1984; 58:320.
19. Britton KE, Ellis SM, Miralles JM, et al. Is “T4 toxicosis” a normal biochemical finding in elderly women? Lancet 1975;2: 141.
20. Hershman JM, Pekary AE, Berg L, et al. Serum thyrotropin and thyroid hormone levels in elderly and middle-aged euthyroid persons. J Am Geriatr Soc 1993;41:823.
21. Monzani F, Del Guerra P, Caraccio N, et al. Age-related modifications in the regulation of the hypothalamic-pituitary-thyroid axis. Horm Res 1996;46:107.
22. Lewis GF, Alessi CA, Imperial JG, et al. Low serum free thyroxine index in ambulating elderly is due to a resetting of the threshold of thyrotropin feedback suppression. J Clin Endocrinol Metab 1991;73:843.
23. Reymond F, Dénéréaz N, Lemarchand-Béraud T. Thyrotropin action is impaired in the thyroid gland of old rats. Acta Endocrinol 1992;126:55.
24. Szabolcs I, Szilágyi G, Góth M, et al. Plasma triiodothyronine response to thyrotropin releasing hormone, thyrotropin and propranolol in old age. Exp Gerontol 1981;16:309.
25. Durica S, Zakula Z, Isenovic E, et al. Age-related changes of TSH receptors in thyroid tissues obtained from euthyroid patients. Arch Gerontol Geriatr 1993;17:203.
26. Gaffney OW, Gregerman RI, Shock NW. Relationship of age to the thyroidal accumulation, renal excretion and distribution of radioiodide in euthyroid man. J Clin Endocrinol Metab 1962;22: 784.
27. Oddie TH, Meade JH, Fisher DA. An analysis of published data on thyroxine turnover in human subjects. J Clin Endocrinol 1966;26:425.
28. Rosenbaum RL, Barzel US. Levothyroxine replacement dose for primary hypothyroidism decreases with age. Ann Intern Med 1982;96:53.
29. Sawin CT, Herman T, Molitch ME, et al. Aging and the thyroid. Decreased requirement for thyroid hormone in older hypothyroid patients. Am J Med 1983;75:206.
30. Davis FB, LaMantia RS, Spaulding SW, et al. Estimation of a physiologic replacement dose of levothyroxine in elderly patients with hypothyroidism. Arch Intern Med 1984;144:1752.
31. Kabadi UM. Influence of age on optimal daily levothyroxine dosage in patients with primary hypothyroidism grouped according to etiology. South Med J 1997;90:920.
32. Snyder PJ, Utiger RD. Response to thyrotropin releasing hormone (TRH) in normal man. J Clin Endocrinol Metab 1972; 34:380.
33. Rubenstein HA, Butler VP, Werner SC. Progressive decrease in serum triiodothyronine concentrations with human aging: radioimmunoassay following extraction of serum. J Clin Endocrinol Metab 1973;37:247.
34. Caplan RH, Wickus G, Glasser JE, et al. Serum concentrations of the iodothyronines in elderly subjects: decreased triiodothyronine (T3) and free T3 index. J Am Geriatr Soc 1981;29:19.
35. Jacques C, Schlienger JL, Kissel C, et al. TRH-induced TSH and prolactin responses in the elderly. Age Ageing 1987;16:181.
36. Snyder PJ, Utiger RD. Thyrotropin response to thyrotropin releasing hormone in normal females over forty. J Clin Endocrinol Metab 1972;34:1006.
37. Olsen T, Laurberg P, Weeke J. Low serum triiodothyronine and high serum reverse triiodothyronine in old age: an effect of disease not age. J Clin Endocrinol Metab 1978;47:1111.
38. Kaiser FE. Variability of response to thyrotropin-releasing hormone in normal elderly. Age Ageing 1987;16:345.
39. Goichot B, Schilenger JL, Grunenberger F, et al. Thyroid hormone status and nutrient intake in the free-living elderly. Interest of reverse triiodothyronine assessment. Eur J Endocrinol 1994;130:244.
40. Pénzes L, Gergely I. Thyroid hormones in the light of general health status of Hungarian centenarians. Age 1989;12:137.
41. Mariotti S, Barbesino G, Caturegli P, et al. Complex alteration of thyroid function in healthy centenarians. J Clin Endocrinol Metab 1993;77:1130.
42. Maugeri D, Russo MS, Di Stefano F, et al. Thyroid function in healthy centenarians. Arch Gerontol Geriatr 1997;25:211.
43. Marioth S, Sansoni P, Barbesino G, et al. Thyroid and other organ-specific autoantibodies in healthy centenarians. Lancet 1992;339:1506.
44. Van Coevorden A, Laurent E, Decoster C, et al. Decreased basal and stimulated thyrotropin secretion in healthy elderly men. J Clin Endocrinol Metab 1989;69:177.
45. Erfuth EM, Norden NE, Hedner P, et al. Normal reference interval for thyrotropin response to thyroliberin: dependence on age, sex, free thyroxin index, and basal concentrations of thyrotropin. Clin Chem 1984;30:196.
46. Greenspan SL, Klibanski A, Rowe JW, et al. Age-related alterations in pulsatile secretion of TSH: role of dopaminergic regulation. Am J Physiol 1991;260:E486.
47. Croxson MS, Wilson TM, Ballantyne GH. TRH testing, T4-thyrotoxicosis and the aging thyroid gland. N Z Med J 1981;93: 417.
48. Wiener R, Utiger RD, Lew R, et al. Age, sex, and serum thyrotropin concentrations in primary hypothyroidism. Acta Endocrinol (Copenh) 1991;124:364.
49. Pekary AE, Mirell CJ, Turner LF, et al. Hypothalamic secretion of thyrotropin releasing hormone declines in aging rats. J Gerontol 1987;42:447.
50. Sawin CT, Geller A, Kaplan MM, et al. Low serum thyrotropin (thyroid-stimulating hormone) in older persons without thyrotoxicosis. Arch Intern Med 1991;151:165.
51. Trivalle C, Doucet J, Chassagne P, et al. Differences in the signs and symptoms of thyrotoxicosis in older and younger patients. J Am Geriatr Soc 1996;44:50.
52. Martin FIR, Deam DR. Thyrotoxicosis in elderly hospitalized patients. Clinical features and treatment outcomes. Med J Aust 1996;164:200.
53. Mooradian AD, Wong NCW. Age-related changes in thyroid hormone action. Eur J Endocrinol 1994;131:451.
54. Surks, MI, Ortiz E, Daniels GH, et al. Subclinical thyroid disease. Scientific review and guidelines for diagnosis and management. JAMA 2004;291:228–238.