Prevalence of AD among the elderly

The prevalence of AD among older men is thought to be high. Studies with inclusive (ie, strictly biochemical vs clinical) definitions of hypogonadism report high figures. In a longitudinal study of an aging prospective cohort, Harman and colleagues cite the prevalence of low total serum T (<11.3 nM or 325 ng/dL) as 20% among men in their 60s and 50% among those more than 80 years of age. In another population-based study of community-dwelling men, Araujo and colleagues found that, although almost one-quarter of those aged 30 to 79 years had low total T levels (<10.4 nM or 300 ng/dL), more than 18% of men aged 70 to 79 years had hypogonadism in conjunction with at least 1 specific symptom (low libido, erectile dysfunction [ED], osteoporosis/osteoporotic fracture) or at least 2 nonspecific symptoms of low T (sleep disturbance, depressed mood, lethargy, diminished physical performance). Even when using a more strict definition of LOH, in which at least 3 sexual symptoms were required in the presence of low total T levels, Wu and colleagues showed that a significant proportion of older men had AD, with a prevalence of 3.2% in those aged 60 to 69 years and 5.1% in those aged 70 to 79 years. Given the variation in definitions and the need for uniformity, professional societies and consensus guidelines have recommended a standardization of LOH as total serum T less than the normal reference range of young, healthy men and clinical symptoms of AD.

At current trends of population growth, it is estimated that the number of individuals aged 65 years and older in the United States will more than double between 2000 and 2030 ; the proportion of older men with symptomatic and, thus, potentially treatable disease is considerable and likely to grow.

Conditions associated with AD

The human and economic burden of aging-related decline in health is substantial. From sarcopenia and osteoporosis to depression and cognitive disorders, such debilitative disease processes are costly and widespread among the elderly. Given the toll these often incapacitating afflictions take on the individual and on society, marked interest has developed in their prevention and treatment.

AD has been implicated in many of the conditions associated with aging in men. Changes in body composition, decreased bone density, depressed mood, and declines in cognition have been linked to low T, as have low libido and ED. Furthermore, maladies traditionally associated with cardiovascular and cerebrovascular disease, within a spectrum of disorders related to insulin resistance (variously known as the metabolic syndrome [MS], cardiometabolic syndrome, or syndrome X), have also been linked to AD. although not yet thoroughly elucidated, the relationship between AD and these risk factors for significant morbidity and mortality among the elderly presents both an intriguing opportunity for prevention and/or treatment of a wide variety of ailments as well as a cause for caution when considering T replacement for the androgen-deficient older man.

Conditions associated with AD

The human and economic burden of aging-related decline in health is substantial. From sarcopenia and osteoporosis to depression and cognitive disorders, such debilitative disease processes are costly and widespread among the elderly. Given the toll these often incapacitating afflictions take on the individual and on society, marked interest has developed in their prevention and treatment.

AD has been implicated in many of the conditions associated with aging in men. Changes in body composition, decreased bone density, depressed mood, and declines in cognition have been linked to low T, as have low libido and ED. Furthermore, maladies traditionally associated with cardiovascular and cerebrovascular disease, within a spectrum of disorders related to insulin resistance (variously known as the metabolic syndrome [MS], cardiometabolic syndrome, or syndrome X), have also been linked to AD. although not yet thoroughly elucidated, the relationship between AD and these risk factors for significant morbidity and mortality among the elderly presents both an intriguing opportunity for prevention and/or treatment of a wide variety of ailments as well as a cause for caution when considering T replacement for the androgen-deficient older man.

Body composition and sarcopenia

Decreasing muscle mass and strength are associated with the aging process and are often accompanied by deterioration in physical function, mobility disability, and frailty, states that have been shown to predict quality of life, morbidity, and mortality. In aging men, a concomitant decrease in T is observed, and lower T levels have been associated with decreased muscle mass in epidemiologic studies. This has led to the postulation that T plays a role in the development of sarcopenia, or loss of protein mass and function. Despite the known correlation between increased muscle mass and strength, whether the effects of T are direct or indirect is not certain, which may explain disparate findings in the literature regarding the hypothetical downstream effects of sarcopenia on physical performance.

Observational studies have shown that higher T levels are associated with increased lean mass and decreased fat mass. Interventional studies randomizing older men to exogenous T or placebo have also shown favorable changes in body composition with T, as has a trial with simultaneous 5-α reductase inhibition. However, results are mixed for strength and physical performance, although several studies found significantly increased muscle strength among older men who received exogenous T, investigations by Snyder and colleagues, Emmelot-Vonk and colleagues, and Kenney and colleagues showed no difference between the treatment and control groups in domains of strength or physical performance.

Likewise, several prospective cohort studies reveal conflicting results. In a longitudinal examination of 2 independent samples, 1 nationally representative and 1 selective, Schaap and colleagues showed no significant association between total or free T levels and either 3-year decline in physical performance or 3-year decline in muscle strength among older men. Conversely, Krasnoff and colleagues showed that higher free T levels among the Framingham Offspring were significantly correlated with better physical performance and faster walking speed, whereas an increase of free T 1 standard deviation more than the baseline was associated with lower odds of either development or progression of limited mobility. In light of findings in another prospective cohort that (1) T levels in elderly men were significantly associated with increased lean mass and decreased body fat; (2) such characteristics in body composition were significantly associated with strength and physical function; but (3) T levels and physical performance were only weakly correlated at best; Araujo and colleagues suggested that the influence of sex hormones on physical function may be mediated indirectly through body composition, indicating that the relationships between T, body composition, and physical function are not necessarily linear.

Bone mineral density

Osteoporosis and bone fracture among the elderly pose a formidable hazard to individuals and potentially a large burden to public health. Hip fracture has been well established as an independent predictor of morbidity and mortality among the elderly, with more than 85% occurring among individuals more than 65 years of age in the United States ; the associated costs of care in the United States are estimated to be substantial. Cohort studies of major bone fractures, in general, show older men to be at risk, including one that observed older men to be at particularly increased risk compared with women and younger men. An epidemiologic study of osteoporotic fractures suggests that the residual lifetime risk of fracture in men older than 60 years is high, approaching that of developing prostate cancer. The modulation of bone metabolism by sex hormones has offered a compelling theoretic option for the prevention of osteoporosis and bone fracture through androgen supplementation, particularly among older men with both low T and low bone mass density (BMD).

However, although observational studies have shown that low T levels among older men are associated with increased risk of fracture (hip hazard ratio 1.88, 95% confidence interval [CI] 1.24–2.82, nonvertebral 1.32, 95% CI 1.03–1.68), with increased falls, with bone resorption, and bone turnover products, no interventional studies to date have shown an effect of T on the incidence of fracture. Multiple small, randomized, controlled studies have shown an increase in BMD with exogenous T; using transdermal formulations, Snyder and colleagues showed a significant increase in lumbar spine density among treated older men with low baseline T levels, and Kenny and colleagues found significantly increased BMD at the femoral neck among those with a low baseline free T who received exogenous T. Likewise, Amory and colleagues observed increased BMD at hip and in spine with either T or T and 5-α reductase inhibition compared with placebo. Although a meta-analysis suggested that only intramuscular T was associated with increased lumbar BMD, this did not include recent double-blinded, randomized, placebo-controlled trials, one of oral T among healthy older men in which no change in BMD was noted, and one of transdermal T in a frail cohort of older men that showed an increase in axial BMD.

Because no investigations of T administration have examined fracture as an outcome, treatment of osteopenia or osteoporosis with T has not been recommended by consensus groups. In hypogonadal men, guidelines suggest monitoring of BMD and T at 2-year intervals. Nevertheless, low T may be considered another useful clinical indicator in the global assessment of potential fracture risk among older men.

Frailty

No agreement exists on the definition of frailty, but it has been recognized broadly as a decline in multiple organ systems leading to functional loss, diminished physiologic reserve, and increased risk of morbidity and mortality. Frailty is generally acknowledged as common among the elderly, although not necessarily as an inexorable result of aging. When examining multiple parameters including strength, ambulation, fatigue, weight loss, and comorbidities, several prospective cohort studies have observed an association between lower T levels and frailty, in cross-sectional studies and over time. One study also reported significant correlation between sex hormone–binding globulin (SHBG) and frailty, although another examination of a larger cohort did not. Because frailty encompasses many different states and processes associated with both aging and AD, evidence may not yet have reached a level for treatment recommendations, but the relationship does provide further information on the emerging picture of the change from physiology to pathology in aging and hypogonadism.

Mood and cognition

Although depression and AD prevalence increase with aging, a direct and independent association between the two has never been proven. However, a close relationship has been intimated by both observational and interventional studies. Among the epidemiologic studies, low total T has been associated with more depressive symptoms in men of the Health Aging and Body Composition cohort ; low free T has been correlated both with increased risk of depression when adjusting for comorbid disease and with increased incident depressive symptoms in large samples of aged, community-dwelling men; and bioavailable T has been found to be 17% lower among men with depression compared with the rest of the Rancho Bernardo cohort.

Interventional studies have shown mixed results for exogenous T therapy and mood. Short courses of T (12 months or less) in small study samples have tended to reveal some effect on mood, although differences between treatment and placebo groups have not been significant. One study of treatment with low T doses (200 mg every 3 weeks for 12 weeks) showed no change in mood, although the putative aim of this particular pilot was to determine safety (ie, no other psychosocial effects of treatment) for therapy for cognitive decline in hypogonadal older men. Positive studies include longer courses of drug exposure (up to 42 months) or higher doses of drug. Although other factors such as degree of AD severity may affect the neurophysiology of disease and treatment and are difficult to identify, a meta-analysis of randomized, placebo-controlled androgen treatment trials for depression has suggested that T therapy is significantly associated with improved depression scores in both old and young hypogonadal men. This analysis did not include a recent randomized, placebo-controlled trial of intramuscular T in 184 men with hypogonadism and MS, in which 30 days of treatment showed significant improvement in depression scores using validated instruments. On balance, experts in psychiatry and neuroendocrinology have recommended T treatment of depressive mood associated with hypogonadism only in the research setting.

A link has also been established between cognitive function and T, although, again, causative direction is not clear and other mediators seem to be involved. Observational studies have linked low free T levels with poorer cognitive function, poorer memory, dementia, and Alzheimer disease, with genetic risk factors playing an interactive role with hypogonadal states in Alzheimer disease. A longitudinal study of older men revealed that those who developed a clinical diagnosis of AD also developed lower T levels in the previous 5 to 10 years, offering a potential prospect for treatment if AD precedes clinical manifestations of cognitive decline.

Again, treatment studies exhibit differing results. Among investigations of healthy, older men, some report improved spatial cognition and memory following therapy with T, whereas others show minimal to no effect on cognition with T or T with finasteride. Few small studies have examined treatment of men with mild cognitive impairment and/or Alzheimer disease; although 2 show some benefit with T therapy in visuospatial cognition and overall cognitive ability, another showed no significant difference between cases and controls. Once more, disparate findings convey the probable interplay between multiple observed and unobserved factors that contribute to the complex pathophysiology governing AD and cognition, as well as divergent methods in the studies themselves, including differing treatment duration, differing cognitive domains tested, and differing subject characteristics.

Libido and sexual function

For most urologists, the primary contact with hypogonadism is made through men presenting with low libido, ED, or the side effects of androgen deprivation therapy (ADT) as treatment of prostate cancer. Much evidence exists linking low libido, suboptimal or absent sexual function, and AD. Population-based and convenience samples have correlated lower T levels with decreased libido and sexual function. As noted by Wu and colleagues, symptoms of sexual dysfunction proved to be more sensitive indicators of LOH than other symptoms of hypogonadism, such as declining physical performance, depression, and fatigue. In this study of 3369 men aged 40 to 79 years, self-reported health status on multiple domains was correlated with serum T levels; symptoms that best predicted low serum T (<11 nM or 317 ng/dL) included low libido, poor morning erection, and ED.

Multiple studies have shown that T supplementation improves sexual desire and enjoyment, both in the short term and with long-term therapy. Moreover, it seems that T supplementation can improve response to phosphodiesterase type 5 inhibition. In one investigation of diabetic patients with ED, a trial of sildenafil citrate revealed lower total T in those who did not respond to treatment; 70% of these nonresponders were converted to responders with coadministration of oral T. T as monotherapy for ED has been shown to be effective in young men for whom hypogonadism is the primary cause of ED, but it was not in those with LOH. Among hypogonadal men with major depressive disorder, T did seem to improve sexual function in both treatment and placebo groups such that the difference between the two was not significant. However, older hypogonadal men receiving T showed significantly improved libido in a retrospective analysis, whereas sexual function along numerous parameters improved with T compared with placebo in randomized treatment trials. Similarly, significantly improved sexual function in the domains of spontaneous erection, sexual desire, and sexual motivation were observed in those receiving higher doses of T in a graded treatment trial compared with the placebo group.

Although it is tempting to consider T therapy for such a well-established association as AD and ED, multiple studies and commentaries have noted that ED may be the consequence not only of low sex steroid levels but also of cardiovascular and metabolic disease. As evidence builds suggesting a closer relationship between them without, as yet, a clear-cut sense of causative direction, so does concern given the theoretic increased risk of morbidity and realized adverse events recently posed by exogenous T exposure among those with cardiometabolic risk factors.

MS

Variously known as cardiometabolic syndrome, syndrome X, and Reaven syndrome, MS is distinguished by a constellation of hemodynamic and metabolic abnormalities, including maladaptive responses such as the glycemic dysregulation associated with visceral obesity. Hypertension and dyslipidemia are among the disorders represented that also increase risk for cardiovascular disease. Although the respective magnitudes of that risk contributed by each in this clustering of factors is still uncertain, the cluster itself has, and continues to, become better characterized since the 1988 American Diabetes Association Banting Lecture in which Reaven coined the term syndrome X. The 2 most current and widely accepted definitions of MS have been put forth by the International Diabetes Federation and the US National Cholesterol Education Program, both of which require increased waist circumference, blood pressure, triglycerides, and fasting blood glucose as well as decreased high-density lipoprotein (HDL) to establish disease.

At present, the prevalence of cardiometabolic risk factors among the elderly is high and seems to be increasing. In a 6-year period, the proportion of obese (body mass index [BMI] ≥30) US adults aged 60 years and older increased to 31% by 2004. In another study of this nationally representative sample, greater than 15% of Americans more than 60 years of age had diabetes mellitus (DM), a likely underestimate of true prevalence given the high suspected rate of undiagnosed disease ; examination of a Medicare cohort suggested an increase of 23% in DM incidence between 1994 and 2004, with an increase in prevalence of 62% in the same period. In a population-based national sample of US adults, the prevalence of MS was found to be greater than 40% in persons aged 60 years or older, with increasing prevalence over time.

The correlation between low T and MS has become the focus of intense investigation. Although the relationship is complex and the directions of causation not clear, advances in understanding have offered insights that ultimately present the clinician with a more difficult challenge regarding management of AD in an elder population given current evidence. On a biochemical level, the leading hypothesis for the underlying mechanism is the induction of hypogonadism by visceral fat through endocrine and cell-signaling functions. Negative feedback on the pituitary is postulated via aromatization of T to estradiol and production of the proinflammatory cytokines tumor necrosis factor a (TNF-a), interleukin (IL)-1, IL-6, and C-reactive protein (CRP). Furthermore, leptin (a protein hormone produced by fat) seems to interfere with luteinizing hormone stimulation of androgen production. Increased insulin likewise has been implicated in decreased testicular steroidogenesis.

In vivo, 3 areas of research support a connection between T and MS: studies correlating low T with any or all of the cardiometabolic cluster risk factors; studies showing a link between ADT and MS; and studies showing that treatment with T may ameliorate many of the metabolic derangements associated with MS.