CHAPTER 47—ENDOCRINE AND METABOLIC DISORDERS
DISORDERS OF PARATHYROID AND CALCIUM METABOLISM
HORMONAL REGULATION OF WATER AND ELECTROLYTE BALANCE
DISORDERS OF THE ADRENAL CORTEX
Impaired homeostatic regulation, a hallmark of aging, occurs in many endocrine systems but may become manifest only during stress. For example, fasting blood glucose levels change little with normal aging, increasing 1 to 2 mg per dL per decade of life. In contrast, glucose levels after a glucose challenge increase much more in healthy older persons than in young adults. In some cases, a loss of function in one aspect of endocrine function may result in a compensatory change in endocrine regulation and associated alterations in catabolism that maintain homeostasis. For example, the reduction in testicular testosterone production that occurs in many older men may be partially compensated for by an increase in pituitary luteinizing hormone secretion and a decrease in testosterone metabolism. In other instances, compensatory changes or alterations in hormone catabolism do not fully offset age-related impairment in endocrine functions, as illustrated by the age-related decline in basal serum aldosterone levels. In this case, a decline in aldosterone clearance fails to offset the decrease in aldosterone secretion.
As with diseases in other organ systems, endocrine disorders in older adults often present with nonspecific, muted, or atypical symptoms and signs. Some of these presentations are well-defined syndromes that are seen almost exclusively in older adults, such as apathetic thyrotoxicosis or hyperosmolar nonketotic state in patients with diabetes mellitus. However, more commonly, endocrine disorders present with subtle, nonspecific symptoms, such as cognitive impairment, or an absence of any complaints. Indeed, the diagnosis of endocrinopathies such as hyperparathyroidism, diabetes mellitus, hypothyroidism, and hyperthyroidism in older adults is commonly established as a result of abnormalities found on routine laboratory screening.
Laboratory evaluation of older adults for endocrine disorders may be complicated by coexisting medical illnesses and medications. For example, the presence of serious acute nonthyroidal illness may lead to the mistaken impression of a thyroid disorder because of the reduction in free thyroxine (T4) levels and sometimes increased or decreased thyrotropin (TSH) levels in sick but euthyroid older patients. Furthermore, ranges of normal laboratory values for endocrine testing are commonly established in younger adults, and even age-adjusted norms for laboratory tests may be confounded by the inclusion of older adults who are ill. Therefore, normal ranges for healthy older people are not available for most laboratory tests.
With aging, a reduction in T4 secretion is balanced by a decrease in T4 clearance, resulting in unchanged circulating T4 levels. Triiodothyronine (T3) levels are unchanged until extreme old age, when they decrease slightly. However, T3 levels are commonly reduced in the setting of nonthyroidal illness because of decreased T4-to-T3 conversion. TSH levels are unchanged or minimally changed in healthy older people.
Nonspecific, atypical, or asymptomatic presentations of thyroid disease are common in older adults. In the outpatient setting, laboratory testing in the stable outpatient is the most reliable way to identify hypothyroidism or hyperthyroidism in older adults who are not acutely ill. Given a 1.4% prevalence of thyroid disease in ambulatory women aged 50 and over, some clinicians recommend routine screening with a highly sensitive TSH test, although treatment may not affect outcomes. In addition, the prevalence of hypothyroidism or hyperthyroidism is sufficiently high to warrant TSH testing in all older adults with a recent decline in clinical, cognitive, or functional status, or upon admission to the nursing home. However, the results of thyroid function testing may be confusing in euthyroid patients with significant concurrent illnesses, as discussed below.
Most prevalence estimates of hypothyroidism in older adults range from 0.5% to 5% for overt disease, and from 5% to 10% for subclinical hypothyroidism, depending on the population studied. As in younger people, most cases of hypothyroidism in older people are due to chronic autoimmune thyroiditis.
Symptoms of hypothyroidism are often atypical in older adults. Some clinical features of hypothyroidism (eg, dry skin, decreased skin turgor, slowed mentation, weakness, constipation, anemia, hyponatremia, arthritis, paresthesias, gait disturbances, elevated myocardial band of creatine phosphokinase) may misleadingly suggest other diseases. Furthermore, these symptoms usually have an insidious onset and a slow rate of progression. As a result, the diagnosis of hypothyroidism is recognized on clinical examination in only 10% to 20% of cases in older adults, and laboratory screening is necessary to detect most cases of hypothyroidism in this population. In addition, older patients with mild hypothyroidism who develop serious nonthyroidal illness may rapidly become severely hypothyroid, and older adults are more susceptible to myxedema coma in this setting. Demented older people with hypothyroidism rarely recover normal cognitive function with thyroid replacement, but cognition, functional status, and mood may improve with treatment of the hypothyroidism.
Subclinical hypothyroidism, with elevated serum TSH and normal free T4 levels, occurs in up to 15% of people aged 65 and over, and is more common in women. Subclinical hypothyroidism is associated with impaired left ventricular diastolic and systolic function, and is an important risk factor for atherosclerosis and myocardial infarction in older women. An approach to the management of subclinical hypothyroidism is outlined in Figure 47.1. The presence of elevated thyroid peroxidase antibody titers portends the eventual development of thyroid failure and overt hypothyroidism, and it is appropriate to initiate T4 replacement therapy in these patients. In addition, T4 replacement is indicated in older adults with progressively increasing TSH levels or a TSH level persistently above 10 mIU/L.
By itself, an increased TSH level is usually due to primary hypothyroidism, but TSH levels may be transiently elevated during recovery from acute illnesses. Therefore, the diagnosis of hypothyroidism should be confirmed with the combination of an elevated TSH level and a decreased free T4 or free T4 index, or by the demonstration of a persistently increased TSH level, or both. Other potentially confusing scenarios in the diagnosis of hypothyroidism include the low T4 syndrome, seen in euthyroid patients with severe nonthyroidal illnesses and presenting with a decreased free T4 index without an increase in TSH levels. Free T4 levels are usually normal in the low T4 syndrome, with elevated levels of reverse T3. Thyroid hormone supplementation has not been shown to be beneficial in these patients. A normal (or low) TSH together with a low free T4 level may also suggest secondary hypothyroidism, which is differentiated from the low T4 syndrome by the presence of hypopituitarism (deficiencies in other pituitary hormones) and decreased reverse T3 levels. Rarely, older people with primary hypothyroidism may also present with inappropriately normal TSH levels resulting from suppression of TSH by fasting, acute illnesses, and medications such as dopamine, phenytoin, or glucocorticoids.
T4 replacement is usually initiated at a low dosage (eg, 25 μg per day) in older adults, increasing the dose every few weeks until TSH levels normalize. However, in patients with cardiac disease, it is prudent to begin replacement therapy at even lower dosages (eg, 12.5 μg per day). In these patients, thyroid replacement should not be withheld for fear of exacerbating cardiac disease; instead, the goal is to reduce or eliminate symptoms of hypothyroidism without causing intolerable exacerbation of cardiac symptoms, such as angina. Older adults who are severely hypothyroid at presentation should receive larger initial T4 replacement doses of 50 to 100 μg, or as high as 400 μg intravenously for those with myxedema stupor or coma, even if there is preexisting heart disease. Such patients should also receive testing to exclude concomitant adrenal insufficiency as well as stress doses of glucocorticoids before receiving T4 to avoid precipitating an adrenal crisis with T4 replacement.
Thyroid hormone requirements decrease with aging because of a reduction in clearance rate, and T4 replacement doses are as much as a third lower in older than in younger adults. The average T4 replacement dosage in older adults is approximately 110 μg per day. Overreplacement of thyroid hormone should be avoided, because osteopenia related to increased bone turnover and exacerbation of heart disease may occur. With correction of the hypothyroid state, the clearance rate of medications such as anticonvulsants, digoxin, and opiate analgesic agents may be affected, necessitating dosage adjustments. T4 supplementation may have beneficial effects on some parameters of cognitive and cardiac function in some older adults with subclinical hypothyroidism, although randomized trials of such treatment have yielded mixed results. Finally, elevations in total and low-density lipoprotein cholesterol levels in hypothyroid patients may resolve with restoration of the euthyroid state, even in those with subclinical hypothyroidism, suggesting that T4 replacement may reduce the risk of atherosclerotic vascular disease in older adults with good long-term survival prospects.
Hyperthyroidism develops in 0.5% to 2.3% of older people, and 15% to 25% of all cases of thyrotoxicosis occur in adults aged 60 and over. In the United States, most cases in older adults are due to Graves’ disease, but toxic multinodular goiter and autonomously functioning adenomas are more common in older than in young adults, especially in populations with low iodine intake.
Hyperthyroidism often presents with vague, atypical, or nonspecific symptoms in frail older patients. Many findings that are common in younger adults (eg, tremor, heat intolerance, tachycardia, ophthalmopathy, increased perspiration, goiter, brisk reflexes) are less common or absent in older persons, whereas other manifestations, such as atrial fibrillation, heart failure, constipation, anorexia, muscle atrophy, and weakness, are more common in older adults. Older persons may present with apathetic thyrotoxicosis, a well-known clinical presentation of hyperthyroidism that is rarely seen in younger persons, in which the usual hyperkinetic presentation is replaced by depression, inactivity, lethargy, or withdrawn behavior, often in association with symptoms such as weight loss, muscle weakness, or cardiac symptoms. A low TSH level is associated with a threefold higher risk of developing atrial fibrillation within 10 years, and hyperthyroidism is present in 13% to 30% of older people with atrial fibrillation. Hyperthyroidism is a cause of secondary osteoporosis and should be considered in the evaluation of patients presenting with decreased bone mass.
A highly sensitive TSH test is adequate as an initial test for hyperthyroidism in relatively healthy older patients, but the diagnosis should be confirmed with a free T4 test. Figure 47.2 illustrates the approach to diagnosing the cause of a low TSH level. Most asymptomatic older adults with low serum TSH levels are clinically euthyroid and have normal T4 and T3 levels, with normal TSH on repeat testing 4 to 6 weeks later. T3 thyrotoxicosis, with elevated T3 but normal T4 levels, occurs in a minority of hyperthyroid patients, but it is more common with aging, especially in patients with toxic adenomas or toxic multinodular goiter. However, in contrast to young adults, many older persons with hyperthyroidism do not have increased T4 or T3 levels, probably because of decreased conversion of T4 to T3 associated with aging and nonthyroidal illness. Diagnostic confusion may occasionally occur in euthyroid patients with conditions or medications causing elevated T4 levels (high T4 syndrome). The high T4 syndrome may occur with drugs or illnesses that decrease T4-to-T3 conversion (high-dose glucocorticoids or β-blocking agents, acute fasting) or that increase circulating levels of thyroid-binding globulin (estrogens, clofibrate, hepatitis).
Subclinical hyperthyroidism is present in less than 2% of older people and is associated with adverse cardiovascular events such as atrial fibrillation, increased left ventricular mass, and impaired ventricular relaxation; osteoporosis; neuropsychiatric problems including dementia; and excess cardiovascular and all-cause mortality. Accordingly, treatment for this condition may be justifiable, but there is a lack of data from randomized controlled trials to support this approach.
Thyroid scanning and measurement of radioactive iodine uptake may be useful in confirming hyperthyroidism and defining the cause (Figure 47.2). Radioactive iodine therapy is the treatment of choice for most older people with hyperthyroidism. Higher or repeated doses are often necessary for patients with toxic multinodular goiter. Antithyroid drugs such as methimazole or propylthiouracil are given before radioactive iodine, to control symptoms and to avoid a worsening of thyrotoxicosis due to transient release of thyroid hormone after radioactive iodine. β-Blocking agents are helpful to manage symptoms such as tachycardia, tremor, and anxiety, but patients should be monitored for changes in cardiopulmonary function. Following radioactive iodine therapy, patients should be followed with serial TSH levels for the eventual development of hypothyroidism, or persistent or recurrent hyperthyroidism. With resolution of hyperthyroidism, the clearance rate of other drugs may decrease, necessitating dosage adjustments to avoid excessive drug levels.
The incidence of multinodular goiter increases with aging, and approximately 90% of women aged 70 years and over, and 60% of men aged 80 years and over have thyroid nodules. Most of these are nonpalpable. Multinodular goiters often have autonomously functioning areas, so that administration of exogenous thyroid hormone to suppress these goiters may cause iatrogenic hyperthyroidism. Older persons with multinodular goiter may develop iodine-induced thyrotoxicosis after receiving radiocontrast or amiodarone.
Solitary thyroid nodules are more likely to be malignant in people over 60 years of age, especially men. The incidence of differentiated thyroid cancers is similar in older and younger adults, whereas anaplastic thyroid carcinomas occur almost exclusively in older adults. However, even well-differentiated papillary and follicular carcinomas are more aggressive and are associated with increased mortality in older persons. Accordingly, a new solitary nodule or an enlargement of an existing nodule warrants a careful evaluation, including a fine-needle aspiration. An approach to the management of a solitary thyroid nodule is outlined in Figure 47.3. Levothyroxine suppressive therapy is indicated to reduce the risk of cancer recurrence and mortality for some patients with thyroid cancer (eg, those at highest risk), but adverse effects on the heart and osteoporosis may occur with long-term thyroid suppression. β-Blocking agents and bone antiresorptive agents may be useful to minimize these untoward effects.
Important changes occur with aging in several systems that regulate calcium homeostasis, ultimately leading to a reduction in bone mass and in some cases osteoporosis in older people (Table 47.1). The net effect of these changes is to increase circulating levels of parathyroid hormone (PTH), which increases 30% between 30 and 80 years of age. Serum calcium levels remain normal as a result of the increase in PTH, but the balance between bone resorption and bone formation is altered in favor of resorption, resulting in a decrease in bone mass and an increased risk of osteoporosis with aging.
Dietary calcium intake is inadequate in most older people. However, as a consequence of factors mentioned in Table 47.1, older people are less able than younger adults to compensate by increasing their intestinal absorption of ingested calcium. In addition, vitamin-D deficiency is common in older adults, especially in hospitalized patients, nursing-home residents, and homebound community-dwelling people. Increased bone turnover and bone loss, especially of cortical bone, is a major consequence of secondary hyperparathyroidism in vitamin D-deficient older adults. Furthermore, vitamin-D deficiency is associated with muscle weakness and may contribute to fall risk in some patients.
Adequate dietary calcium and vitamin-D supplementation may reverse age-related hyperparathyroidism, increase bone mineral density, and reduce falls and osteoporotic fracture rates, although the effectiveness of vitamin D alone in preventing osteoporotic fractures is unclear. For many older people, an elemental calcium intake of at least 1000 to 1500 mg per day is desirable, together with at least 400 IU of vitamin D per day. High-dose supplementation may cause vitamin-D intoxication with hypercalcemia, hypercalciuria, impairment of kidney function, and bone loss. A few patients who are unable to take daily oral supplements may benefit from 100,000 IU of oral or parenteral vitamin D every 6 months with minimal risk of hypercalcemia. (See also the section on prevention and treatment of osteoporosis in Osteoporosis and Osteomalacia.)
Primary hyperparathyroidism and malignancy are the most common causes of hypercalcemia in older adults. The annual incidence of primary hyperparathyroidism is approximately 1 per 1000, and the disease is threefold more prevalent in women than in men. Most patients with primary hyperparathyroidism are asymptomatic, and the diagnosis is made after an incidental finding of hypercalcemia. When the disease is symptomatic, older persons are more likely than younger adults to present with neuropsychiatric symptoms such as depression and cognitive impairment, neuromuscular symptoms such as proximal muscle weakness, or osteoporosis. Typical laboratory findings in primary hyperparathyroidism and other common causes of hypercalcemia are described in Table 47.2. The diagnosis of primary hyperparathyroidism is confirmed with an elevated or high normal PTH level by the use of an assay for intact PTH, in the presence of hypercalcemia.
Surgery is the treatment of choice for primary hyperparathyroidism with total serum calcium levels more than 1 mg/dL above the normal range, 24-hour urine calcium levels above 400 mg, creatinine clearance reduced by more than 30% in comparison with normal persons of the same age, markedly decreased bone density (T score below −2.5 at any site on bone densitometry), or nephrolithiasis. Patients with serum calcium levels less than 1 mg/dL above the normal range who are asymptomatic and managed conservatively should avoid thiazide diuretics, dehydration, and immobilization. Baseline assessment in these patients should include blood pressure, serum calcium and creatinine, creatinine clearance, 24-hour urine calcium, abdominal radiography, and bone densitometry. Follow-up assessments should include blood pressure and serum calcium every 6 months, and serum creatinine and bone densitometry can be considered periodically. In addition, these patients should be followed clinically for the development of nephrolithiasis, minimal trauma fractures, and neuropsychiatric or neuromuscular symptoms. Medical management options for hyperparathyroidism also include β-blocking agents, oral phosphate in patients with low serum phosphate levels and good kidney function, and possibly bisphosphonates or raloxifene.
In hospitalized patients, the most common cause of hypercalcemia is a malignancy that produces PTH-related peptide, with hypercalcemia resulting primarily from increased net bone resorption. The presence of an underlying cancer is usually evident on examination and routine diagnostic testing. Squamous cell cancers of the lung or head and neck are common causes of hypercalcemia due to PTH-related peptide production. Other common malignancies associated with hypercalcemia include breast cancer, lymphoma, and myeloma, although the mechanism of the hypercalcemia is different for many of these cancers. Acute treatment for hypercalcemia includes volume replacement with intravenous saline, followed by diuresis with a loop diuretic when rehydration is complete. A parenteral bisphosphonate such as pamidronate should be given, along with treatment of the underlying malignancy, if possible. In addition to their usefulness in the treatment of hypercalcemia, bisphosphonates may decrease bone pain and the risk of pathologic fractures in patients with osteolytic bone metastases from a variety of cancers.
Paget’s disease is characterized by localized areas of increased bone remodeling, resulting in a change in bone architecture and an increased tendency to deformity and fracture. Its prevalence increases with aging, affecting 2% to 5% of people aged 50 years and over. Paget’s disease is usually asymptomatic and is often diagnosed as an incidental finding on radiographs or during evaluation for an unexplained elevation in serum alkaline phosphatase. The most commonly affected sites are the pelvis, spine, femur, tibia, and skull. When Paget’s disease is symptomatic, pain is the most common presenting symptom, either localized to the affected bones or resulting from secondary osteoarthritic changes, often in the hips, knees, and vertebrae. When bone deformities occur, the long bones of the lower extremities are usually affected, often with a bowing of the involved extremity. Skull involvement may result in compression of the eighth cranial nerve and sensorineural hearing loss. The most devastating complication of Paget’s disease is malignant transformation of the affected bone, especially osteosarcoma. Treatment is not usually necessary for asymptomatic disease, unless there is concern for hearing loss from skull involvement, nerve root or spinal cord compression from vertebral involvement, or hip fracture from femoral neck involvement. Bisphosphonates suppress the accelerated bone turnover and bone remodeling that is characteristic of this disease, and they are the treatment of choice. During treatment, patients should be monitored clinically for changes in bone pain, joint function, and neurologic status, and with biochemical indices of bone formation (eg, serum osteocalcin or bone-specific alkaline phosphatase) or resorption (eg, urinary N-telopeptide), or both.
Unlike young adults, older persons are predisposed to both volume depletion and free water excess. This impairment in regulation of volume status and osmolality is multifactorial, reflecting decreased total body water content as well as alterations in antidiuretic hormone (ADH) secretion, osmoreceptor and baroreceptor systems, urine-concentrating capability, renal hormone responsiveness, and thirst sensation. ADH secretion tends to be excessive in older people, with normal to elevated basal ADH levels, increased ADH responses to osmoreceptor stimuli such as hypertonic saline infusion, and decreased ethanol-induced inhibition of ADH secretion. This state of relative ADH excess with aging, together with the common occurrence of renal insufficiency, heart failure, hypothyroidism, and diuretic use, predisposes older adults to hyponatremia by impairing free water clearance. (Medications causing syndrome of inappropriate antidiuretic hormone or SIADH include the selective serotonin-reuptake inhibitors, sulfonylureas, carbamazepine, oxcarbazepine, and tricyclic antidepressants.)
Under other circumstances, older people are at increased risk of volume depletion. With aging, basal aldosterone secretion declines disproportionately to the decrease in clearance, with a net reduction in circulating aldosterone levels of about 30% by the age of 80 years. At the same time, atrial natriuretic hormone secretion (and renal responsiveness to this hormone) increases with aging. Atrial natriuretic hormone inhibits aldosterone production and causes natriuresis and diuresis through its effects on the kidneys. Taken together, these changes predispose older people to volume depletion by decreasing the ability of the kidneys to conserve sodium under conditions of fluid deprivation. Baroreceptor ADH responses to hypotension and hypovolemia are decreased in older people, placing them at additional risk of dehydration. Moreover, renal responsiveness to ADH is decreased with aging, resulting in a decreased ability of the kidneys to maximally concentrate urine. Finally, even healthy older adults have decreased thirst sensation and may not be aware that they are becoming dehydrated. Demented and immobile older people are at the highest risk for severe dehydration.
In addition to predisposing to volume depletion, age-related hyporeninemic hypoaldosteronism also increases the risk of hyperkalemia, especially in patients with diabetes mellitus or renal insufficiency. The addition of angiotensin-converting enzyme inhibitors, nonsteroidal anti-inflammatory drugs, β-blocking agents, and diuretics with aldosterone-antagonist properties may lead to potentially lethal hyperkalemia in some of these patients.
Basal serum cortisol levels do not change with aging, because decreased cortisol secretion is balanced by a decrease in clearance. Adrenocorticotropic hormone (ACTH) stimulation of cortisol production is unchanged, and cortisol and ACTH responses to stress and secretagogues are unimpaired with aging. Clinically, acute cortisol responses to stress may be higher and more prolonged in older than in younger adults. Accordingly, unless it is emergent, adrenal function testing should be deferred at least 48 hours after major stressors, such as surgery or trauma. In older patients with a normal ACTH stimulation test in whom adrenal insufficiency is suspected, endocrinology consultation is recommended to assist with further testing.
Chronic glucocorticoid therapy is also the most common cause of adrenal failure in older adults, because of chronic suppression of adrenal function. Recovery of adrenal axis function is variable and may take several months to occur. Autoimmune-mediated adrenal failure is less common in older than in younger adults, but tuberculosis, adrenal metastases, and adrenal hemorrhage in anticoagulated patients are more common causes of adrenal insufficiency in older persons. Older patients with chronic adrenal insufficiency may present with nonspecific symptoms such as anorexia, weight loss, or impaired functional status, and hyperkalemia may not be present initially. Accordingly, a high index of suspicion is required to make the diagnosis. When adrenocortical insufficiency is suspected, the ACTH stimulation test should be performed and therapy initiated, although the best test to exclude secondary adrenal insufficiency (decreased pituitary ACTH secretion) is controversial. In older people who are stopping chronic glucocorticoid therapy, the replacement regimen should be tapered gradually, and stress dose coverage should be given for major surgery and other acute physiologic stresses until adrenocortical function has normalized.
Exogenous glucocorticoids are the most common cause of Cushing’s syndrome in older adults, often causing adverse effects, including psychiatric and cognitive symptoms, osteoporosis, myopathy, and glucose intolerance. For patients beginning long-term glucocorticoid therapy, baseline and follow-up bone densitometry measurements are indicated, and calcium, vitamin D, and antiresorptive treatments such as bisphosphonates should be initiated as appropriate. Management of subclinical glucocorticoid hypersecretion is discussed in the following section.
In autopsy studies, the prevalence of clinically inapparent adrenal masses (adrenal incidentalomas) ranges from less than 1% of people younger than 30 years of age to 7% of persons older than 70 years of age. Most adrenal incidentalomas are benign adrenocortical adenomas, although pheochromocytomas and adrenocortical carcinomas also occur.
The goals of assessment are to determine whether the tumor is functional (hormone-secreting) (Table 47.3), and whether it is benign or malignant. Screening for subclinical glucocorticoid hypersecretion is controversial. Many adrenocortical adenomas have a degree of functional autonomy, and some patients may develop hypertension, insulin resistance, and other metabolic derangements. However, it is unclear whether subclinical glucocorticoid hypersecretion is associated with long-term morbidity, or whether adrenalectomy or medical management of metabolic derangements improves outcomes. Moreover, screening all older adults with adrenal incidentalomas for glucocorticoid hypersecretion would yield a high proportion of false-positive results. Accordingly, it may be prudent to limit testing to patients with a symptom complex suggesting Cushing’s syndrome and patients scheduled for major surgery who are at risk for postoperative adrenal crisis.
The assessment of malignancy risk in adrenal incidentaloma patients is based on lesion size, its imaging characteristics, and its rate of growth. The prevalence of adrenal cortical carcinoma in these patients increases from 2% of lesions smaller than 4 cm to 25% of lesions larger than 6 cm. Surgical excision is generally recommended for adrenal masses larger than 6 cm, and for imaging findings including rapid growth rate that suggest the mass is not an adenoma. However, the individual’s treatment preferences and clinical condition must be taken into account before recommending treatment.
In contrast to cortisol, circulating levels of the principal adrenal androgen, dehydroepiandrosterone (DHEA), decline progressively with aging and in octogenarians are only 10% to 20% of young adult levels. Low DHEA levels are associated with poor health, whereas DHEA levels are positively correlated with some measures of longevity and functional status. Given these associations, there has been interest in the potential therapeutic effects of DHEA administration in older adults.
Most studies involving physiologic to mildly supraphysiologic DHEA supplementation in middle-aged and older persons have not found beneficial effects on body composition, although bone mineral density was found to increase modestly in postmenopausal women after 1 year of DHEA. DHEA improved mood and subjective well-being in patients with primary adrenal insufficiency and mid-life dysthymia, but improvements in mood and well-being were not consistently observed in healthy middle-aged and older persons. DHEA decreased circulating high-density lipoprotein cholesterol levels in older women, suggesting potential long-term atherogenic effects. Furthermore, DHEA is metabolized to estrogens and to androgens, including testosterone and dihydrotestosterone, and its effects on the risk of breast cancer in women and prostate cancer in men are unknown. Finally, higher doses of DHEA may cause androgenization in some women and gynecomastia in men. Thus, the safety and efficacy of DHEA supplementation in older adults have not been established, and its use is inappropriate outside of clinical studies.
Despite former controversy, there is now general agreement that total and free testosterone levels and testosterone secretion are lower in healthy older men than in younger men. Many healthy older men exhibit moderate primary testicular failure, with decreased sperm production, testosterone levels, and testosterone secretory responses to gonadotropin administration. In addition, many of these men have inappropriately normal (ie, not increased) gonadotropin levels in the presence of low testosterone levels, suggesting secondary (hypothalamic or pituitary) testicular failure. Overt testicular failure is common in chronically ill and debilitated older men, manifested by total testosterone levels well below the normal range and symptoms suggesting androgen deficiency, including decreased libido and potency, gynecomastia, and hot flushes. Testosterone replacement therapy is generally warranted in these patients, as in hypogonadal young men. However, it is more common to encounter older men with low-normal or mildly decreased serum testosterone levels and nonspecific manifestations, such as decreased libido, weakness, decreased muscle mass, osteopenia, and memory loss. In most cases, these manifestations have multiple causes, but it has been hypothesized that declining testosterone levels with aging contribute to their development, and that testosterone supplementation may help to prevent or treat these disorders.
Men with suspected hypogonadism should be evaluated with a serum free or bioavailable (non–sex hormone–binding globulin-bound) testosterone level, either measured by equilibrium dialysis or calculated from measurements of total testosterone and sex hormone–binding globulin. Concentrations of sex hormone–binding globulin, the main circulating binding protein for testosterone, increase with age. Therefore, the age-related decline in serum free or bioavailable testosterone is greater than that of total testosterone, and total testosterone measurements do not accurately reflect the decrease in biologically active testosterone with aging. Direct radioimmunoassays using “analog” kits for free testosterone are widely used but are not recommended because they may underestimate androgen deficiency in older men and overestimate androgen deficiency in men with low sex hormone–binding globulin (eg, moderately obese men). Luteinizing hormone and follicle-stimulating hormone levels should be obtained. In addition, a review (and if possible, discontinuation) of medications that may suppress gonadotropins (eg, glucocorticoids and central nervous system–active drugs) and a prolactin level are indicated if gonadotropins are low-normal or low in the presence of low testosterone levels. High prolactin levels inhibit gonadotropin secretion and could be due to either a pituitary adenoma or hypothalamic disorder. Further studies may be warranted in such patients, including magnetic resonance imaging of the pituitary fossa and assessment of other pituitary functions (eg, cortisol response to ACTH and T4). Baseline bone densitometry measurements should be obtained in men with decreased testosterone levels to exclude osteoporosis.
Data from controlled studies of testosterone supplementation of up to 3 years’ duration in older men with low-normal or mildly decreased serum testosterone levels are summarized in Table 47.4. However, it is unknown whether these potential benefits and risks are clinically important, or whether the benefits outweigh the risks. Bearing these uncertainties in mind, a trial of testosterone supplementation may be appropriate in older men with serum total testosterone levels below 3.0 ng/mL and clinical features suggesting hypogonadism (eg, osteoporosis, muscle wasting or weakness, mild anemia of unclear cause, loss of libido), although the U.S. Food and Drug Administration has not labeled the use of testosterone for this indication. Notably, bisphosphonates have clearly demonstrated efficacy in treating older men with osteoporosis. Androgen replacement therapy is inappropriate in asymptomatic older men with low-normal total testosterone levels who do not have clinical manifestations consistent with androgen deficiency. Men should be monitored closely for adverse androgenic effects of treatment, including erythrocytosis and potential exacerbation of prostatic disease. However, there is no direct evidence that testosterone therapy increases the risk of prostate cancer or symptomatic benign prostatic hyperplasia. (See also Disorders of Sexual Function.)
Many of the symptoms and signs of hormone deficiency mimic those classically associated with aging. The fact that many hormones also decline with aging has led to an enthusiasm for attempting to reverse unwanted changes associated with aging by the use of hormonal replacement. Based on very compelling epidemiologic data, replacement of estrogen, with or without progesterone, was once standard care for postmenopausal women, but this practice has fallen out of favor in the past several years given new data from randomized clinical trials. Estrogen replacement therapy (ERT) now is largely limited to treatment of menopausal symptoms (See the section on treatment of menopausal symptoms in Gynecologic Diseases and Disorders).
Three meta-analyses of observational studies have demonstrated an association of ERT use in women with a reduction in heart disease by half. Current use of estrogen is the factor most strongly associated with reducing risk, and prolonged use is the second most important factor. Preliminary studies suggest that the effects are similar when progesterone is added. However, a few long-term prospective studies are now available, and further studies are being completed to determine if the cardiovascular benefit of ERT is present, as suggested by epidemiologic studies. A randomized controlled trial of estrogen with progesterone replacement in women with established coronary disease did not find that estrogen improves cardiac outcome. In fact, the estrogen treatment group demonstrated increased mortality in the first year on therapy, with improved survival in years 2 through 5, leading to no net benefit. Another trial of women with coronary artery disease found no benefit from estrogen for angiographic changes of atherosclerosis.
The Women’s Health Initiative (WHI) is a set of clinical trails to test primary prevention of coronary artery disease with estrogen and estrogen-progesterone combinations. In the WHI, the estrogen-progesterone arm was discontinued early because of the increased risk of coronary disease, breast cancer, stroke, and deep-vein thrombosis in this group. Aspirin and lipid-lowering agents were not found to have an effect on the outcomes. The estrogen-alone arm of the study was discontinued early (February 2004 rather than March 2005) because the hormone increased the risk of stroke and did not reduce the risk of coronary heart disease.
Some observational studies also suggested that estrogen may have a role in preventing dementia. Studies of estrogen in women with mild dementia demonstrate improved memory, orientation, and calculation skills, but these studies are small and limited by selection bias. A placebo-controlled trial of estrogen replacement given for 1 year to 120 women with early to moderate Alzheimer’s dementia found no improvement in affective or cognitive outcomes. In the WHI, more women in the estrogen-progesterone arm of the trial had clinically important declines in their Mini–Mental State Examination scores. Similarly, in the estrogen-alone arm of the cognitive substudy, 76 women taking estrogen developed mild cognitive impairment, compared with 58 women receiving placebo (hazard ration [HR], 1.34; 95% confidence interval [CI], 0.95 to 1.89), and when this was combined with the estrogen-plus-progestin trial, the HR was 1.25 (95% CI, 0.97 to 1.60), not clear evidence of harm. Conjugated equine estrogen (CEE) alone did not protect against dementia, as 47 participants were diagnosed with probable dementia, 28 assigned to CEE, and 19 to placebo (HR, 1.49; 95% CI, 0.83 to 2.66). When the estrogen-alone trial was combined with the estrogen-plus-progestin trial, per the original WHI mental status protocol, the overall HR for probable dementia was 1/76 (95% CI, 1.19 to 2.60; P = .005).
The risks of breast cancer, endometrial cancer, and deep-vein thrombosis associated with the use of estrogen have been well established. Meta-analyses suggest that women who used estrogen for 5 years or less have no increased risk of breast cancer, but that treatment for 15 years or more is associated with a 30% increased risk. Information about estrogen-progesterone combinations, with various derivatives and dosing schedules, are incomplete; it is therefore not possible to conclude that any particular formulation may lessen these risks compared to other formulations. Unopposed estrogen increases the risk for endometrial cancer by 2% to 8%, and the risk increases with increasing duration of use. Risk may be decreased with lower doses than in standard use. Progesterone reduces or negates the risk of endometrial cancer associated with estrogen. Finally, data demonstrate a two- to fourfold increase in risk of thromboembolic disease with the use of estrogen. (See also the section on estrogen replacement therapy for osteoporosis prevention in Osteoporosis and Osteomalacia.)
Growth hormone secretion declines with aging, and by 70 to 80 years of age, about half of adults have no significant growth hormone secretion over 24 hours. A corresponding decline occurs in levels of insulin-like growth factor 1, which mediates most of the effects of growth hormone; it falls to levels comparable to those in growth hormone–deficient children in 40% of adults aged 70 to 80 years.
Adults with growth hormone deficiency due to hypothalamic-pituitary disease exhibit decreased muscle strength, lean body mass, and bone density; increased abdominal obesity; unfavorable lipid profiles; and an increased risk of cardiovascular disease. All improve with growth hormone replacement. Older adults without hypothalamic-pituitary disease have many of the same conditions, which leads to the hypothesis that growth hormone supplementation may have a beneficial effect on these clinically important age-related disorders.
Randomized controlled trials of growth hormone supplementation in older adults have reported increased lean body mass and bone density and decreased fat mass. However, growth hormone was not found to augment improvements in muscle strength achieved with exercise alone, and no improvements in functional status were demonstrated. Furthermore, significant adverse effects were common, including carpal tunnel syndrome, arthralgias, edema, and gynecomastia. The long-term efficacy and safety of growth hormone administration in older people are unknown. Short-term growth hormone supplementation may improve nitrogen balance in older persons with severe illness and catabolic states. However, growth hormone is very expensive, and at present it is not recommended for clinical use in older people who do not have established hypothalamic-pituitary disease.
Melatonin, a hormone secreted by the pineal gland, is thought to be involved in the regulation of circadian and seasonal biorhythms. Melatonin secretion is inhibited by exposure to light, resulting in a marked circadian variation in circulating melatonin levels, and its sedative effects suggest a role in sleep induction. Most studies show that plasma melatonin levels decline throughout life after early childhood, but the physiologic significance of this decline in melatonin secretion is unclear. Numerous claims have been made in the lay press regarding the “anti-aging” benefits of melatonin supplementation for various conditions, including insomnia, immune deficiency, cancer, and the aging process itself. Although melatonin may have sleep-inducing properties in older people with insomnia, the long-term risks and benefits of melatonin supplementation have not been established for insomnia or any other indication.
■ Bilezikian JP, Potts JT Jr, Fuleihan Gel-H, et al. Summary statement from a workshop on asymptomatic primary hyperparathyroidism: a perspective for the 21st century. J Clin Endocrinol Metab. 2002;87(12):5353–5361.
This statement summarizes the proceedings of a consensus conference on asymptomatic primary hyperparathyroidism held at the National Institutes of Health in April 2002. Covered topics included the epidemiology of primary hyperparathyroidism, pathophysiology, clinical presentations, differential diagnosis, bone mass measurements, fracture incidence, natural history, new surgical and localization procedures, and new approaches to specific medical management of primary hyperparathyroidism. Revised recommendations for surgical intervention in asymptomatic hyperparathyroidism and monitoring recommendations for patients who are followed without surgery are presented.
■ Grady D, Wenger NK, Herrington D, et al. Postmenopausal hormone therapy increases risk for venous thromboembolic disease. Ann Intern Med. 2000;132(9):689–696.
This analysis of venous thromboembolic events from 2763 women with known coronary artery disease receiving estrogen with progesterone or placebo for 4 years revealed that women receiving the estrogen and progesterone combination were more likely to experience a thrombotic event (34 events versus 13 events in the placebo group; relative hazard 2.7). There was no trend for decreasing relative hazard over time for deep-vein thrombosis (in contrast to the suggestion that any thrombotic event from estrogen would occur in the first few months following the initiation of therapy), but there was decreasing trend for pulmonary emboli after the first 2 years. Baseline characteristics that predicted venous thrombosis in multiple regression analysis were advanced age and assignment to hormone therapy. During the study, risk for thrombosis increased among women with a lower extremity fracture or cancer, and increased in the first 90 days following inpatient surgery or nonsurgical hospitalization and following myocardial infarction. Risk was substantially diminished in women using aspirin or statins. These findings are consistent with observational studies that suggest that estrogen therapy increases the risk of thrombotic events approximately two- to fourfold.
■ Gruenewald DA, Matsumoto AM. Aging of the endocrine system. In: Hazzard WR, Blass JP, Halter JB, et al., eds. Principles of Geriatric Medicine and Gerontology. 5th ed. New York: McGraw-Hill; 2004:819–835.
Clinical management issues and the physiology of aging are reviewed in this chapter on endocrine and metabolic disorders in older adults. Principles of geriatric endocrinology are covered, followed by age-related changes in neurotransmitter and neuropeptide regulation of endocrine systems. Hormonal changes in Alzheimer’s disease are briefly discussed. The remainder of the chapter delves into age-related disorders of the anterior and posterior pituitary, the hypothalamic-pituitary-adrenal axis, the sympathoadrenal system, the renin-angiotensin-aldosterone system, the growth hormone axis, and the hypothalamic-pituitary-testicular axis. The reader is referred to other chapters in the textbook for information on estrogen and androgen replacement therapy and for the management of thyroid disease and disorders of calcium and lipid regulation.
■ Gruenewald DA, Matsumoto AM. Testosterone supplementation therapy for older men: potential benefits and risks. J Am Geriatr Soc. 2003;51(1):101–115.
This is a qualitative systematic review of controlled trials of testosterone therapy in older men with low-normal to moderately decreased baseline testosterone levels.
■ Grumbach MM, Biller BM, Braunstein GD, et al. Management of the clinically inapparent adrenal mass (“incidentaloma”). Ann Intern Med. 2003;138(5):424–429.
This is a summary report from a National Institutes of Health consensus conference held February 2002. The panel developed recommendations for the biochemical evaluation of all patients with adrenal incidentaloma and criteria for surgical intervention. A summary of 10 “take-home points” is presented.
■ Rossouw JE, Anderson GL, Prentice RL, et al. Writing Group for the Women’s Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA. 2002;288(3):321–333.
The estrogen plus progestin component of the Women’s Health Initiative (WHI) reports its major findings following the early discontinuation of the trial in May 2002. The WHI is a randomized controlled primary prevention trial (planned duration, 8.5 years) in which 16,608 postmenopausal women aged 50 to 79 years with an intact uterus at baseline were recruited by 40 U.S. clinical centers from 1993 to 1998. The primary outcome was coronary heart disease (CHD), that is, nonfatal myocardial infarction and CHD death, and invasive breast cancer was the primary adverse outcome. The data and safety monitoring board recommended stopping the trial of estrogen plus progestin versus placebo because the test statistic for invasive breast cancer exceeded the stopping boundary for this adverse effect and the global index statistic supported risks exceeding benefits. The hazard ratios for the major clinical outcomes are summarized in the article. Overall, all-cause mortality was not affected during the trial, and the risk-benefit profile found was not believed to be consistent with the requirements for a viable intervention for primary prevention of chronic diseases.
David A. Gruenewald, MD
Alvin M. Matsumoto, MD
Anne M. Kenny, MD (estrogen replacement therapy section)