Introduction

Aging in men leads to decline in multiple physiologic and psychological parameters. The physical effects of aging broadly recognized include weight gain, loss of muscle mass, and skin changes. The decline is a universal event for men as they age and is due in part to decreasing androgen production. The reduction of serum levels of total, free, and bioavailable testosterone throughout aging has been well documented [1–3]. While the majority of men remain asymptomatic, some complain of symptoms that are attributed to this drop in androgens. The symptoms are often vague and nonspecific like depression, loss of energy, and sexual dysfunction. The symptoms secondary to decreased androgen production are so common that some would say that they are not reflective of a medical condition, but rather the normal consequence of aging. This perspective may cause clinicians to question the wisdom of androgen replacement therapy in aging men, even in men suffering a significant decrease in quality of life [2]. Occasionally, skeptics of androgen replacement therapy point to the surprising results of the Women’s Health Initiative (WHI) studies to warn of similar, possibly unrecognized risks associated with hormone replacement therapy (HRT) in aging men. One may ask whether such comparison is appropriate. To address this question, a brief review of HRT in women and the WHI studies is necessary [4].

Menopause, women, and hormone replacement therapy

Menopause is the permanent cessation of the menstrual cycle. Menopause is tied to a specific date, defined as the day after the final menstrual period, and determined retrospectively once 12 months have gone by with no menstrual flow at all. In the United States, the average age of a woman having her last period, menopause, is 51; however, some women have their last period in their forties, though most have it later in their fifties. This transition is typically neither sudden nor abrupt; rather it tends to occur over a period of years, and is a natural consequence of aging. For some women, the accompanying signs and symptoms which may occur during the transition period can significantly disrupt daily activities and sense of well-being [5]. In most women, hormone replacement therapy (HRT) can alleviate the symptoms of menopause.

The issue with HRT is not with its use for the relief of menopausal symptoms, but with the contention that it may have additional health benefits for women. For many decades, HRT was widely recommended not only for the treatment of menopause and menopausal symptoms, but also for the primary prevention of osteoporosis and heart disease. The wide use of HRT in postmenopausal women was bolstered by observational studies that demonstrated that postmenopausal users of HRT had substantially lower rates of cardiovascular events than non-users [6]. HRT became the standard of care, as a form of ‘‘preventive medicine,’’ with many physicians believing it reduced risk of cardiovascular disease and improved women’s well-being and quality of life. This “primary prevention” aspect was considered especially beneficial, given the longer life expectancy and greater risk of disabling disease after menopause. Despite concerns about possible bias in the observational studies toward healthy women and reports of increased breast cancer risk among women taking HRT, many physicians nevertheless accepted the argument that women should take HRT because of the much higher prevalence of cardiovascular disease compared with the risk for breast cancer [7].

The Women’s Health Initiative

The Women’s Health Initiative (WHI) was a long-term national health study that has focused on strategies for preventing heart disease, breast and colorectal cancer, and osteoporotic fractures in postmenopausal women [4]. This 15-year project involved 161808 women aged 50–79, and is still ongoing. Randomized placebo-controlled clinical trials evaluating hormone therapy were a major part of the WHI. There were two such trials. The “WHI-E+P” compared conjugated equine estrogen, plus progestin (Prempro, Wyeth) to placebo in healthy postmenopausal women. The second, the “WHI-CEE” trial, enrolled menopausal women without a uterus and compared conjugated equine estrogen versus placebo.

The estrogen plus progestin trial was stopped early in July 2002 after investigators concluded that the risk–benefit profile was not consistent with the requirements for a viable intervention for primary prevention of chronic diseases. They reported an increased risk of myocardial infarction, breast cancer, stroke, deep venous thrombosis (DVT), and pulmonary embolism (PE). There was a decreased risk of colorectal cancer and fewer fractures. In announcing the decision, Dr. Claude Lenfant added that “the cardiovascular and cancer risks of estrogen plus progestin outweigh any benefits – and a 26 percent increase in breast cancer risk is too high a price to pay, even if there were a heart benefit. Similarly, the risks outweigh the benefits of fewer hip fractures” [4].

The estrogen only trial was discontinued in March 2004. In announcing the decision, Dr. Barbara Alving said that “after careful consideration of the data, NIH has concluded that with an average of nearly 7 years of follow-up completed, estrogen alone does not appear to affect (either increase or decrease) heart disease, a key question of the study. At the same time, estrogen alone appears to increase the risk of stroke and decrease the risk of hip fracture. It has not increased the risk of breast cancer during the time period of the study” [4].

A similar British study evaluated the risk–benefit ratio of HRT, which was used by around one-third of women aged 50–64 in Britain [8]. Data from the Million Women Study showed that current use of HRT was associated with an increased risk of incident and fatal breast cancer, and that the effect was substantially greater for estrogen–progestin combinations than for other types of HRT [9].These findings were similar to those observed in the WHI study. Following publication of the Million Women Study results, the use of HRT in Britain declined significantly from 107/1000 (95% confidence interval [CI] 104–110) in 2000 to 87/1000 (95% CI 84–89) in 2003 [10].

Following publication of the WHI data, physicians were urged to use HRT with caution. The investigators recommended that women with normal rather than surgical menopause should take the lowest feasible dose of HRT for the shortest possible time to minimize their risk. The United States Preventive Task Force concluded that the harms of long-term therapy are greater than the potential benefits [11].

Current expert opinion suggests that systemic hormone therapy is the most effective treatment for most menopausal symptoms, including vasomotor symptoms and vaginal atrophy. Although the risks of venous thromboembolic events and ischemic stroke increase with either estrogen therapy or estrogen and progestogen therapy, the risk is rare in the 50- to 59-year-old age group. An increased risk of breast cancer is seen with 5 years or more of continuous estrogen with progestogen therapy, possibly earlier with continuous use since menopause. While treatment duration may vary based on an individual’s risk and benefit assessment, in general, the lowest dose of hormone therapy should be used for the shortest amount of time to manage menopausal symptoms [12].

Androgen deficiency in the aging male

As in women, there is a decline in the production of testosterone as men age, though this decline differs from estrogen both in abruptness of onset and totality. The decline of serum testosterone levels is a gradual, age-related process resulting in an approximate 1% annual decline after age 30. In cross-sectional and longitudinal studies of men aged 30 or 40 years and above, total, bioavailable, and free testosterone concentrations fall with increasing age. The observed decline in bioavailable and free testosterone was even more pronounced than the decline in total testosterone levels [13–15].

Multiple mechanisms likely influence the decline of testosterone levels in aging men. Lower testosterone levels may result from reduced testicular responses to gonadotropin stimuli with aging, coupled with a progressive decline in the sensitivity of the hypothalamic–pituitary feedback mechanism, resulting in incomplete compensation for the fall in total and free testosterone levels. Whether the age-dependent decline in androgen levels leads to health problems in older men is being debated vigorously [16–19].

Three large population-based studies document the prevalence of hypogonadism in men. The Boston Area Community Health (BACH) Survey is a random community sample of men age 30–79, which showed that 24% had total testosterone (TT) <300 ng/dL, but only 5.6% of men were symptomatic [3]. The Hypogonadism in Males (HIM) study evaluated men aged 45 and older who came for care to participating primary care clinics. In this population, 38.7% of men had TT <300 ng/dL. Once again, many of these men were asymptomatic and the clinic visit was for other medical conditions [20]. The European Male Aging Study (EMAS) is a random sample of European men aged 40–79 that showed that almost 25% had some form of hypogonadism [21].

Testosterone replacement therapy (TRT) is available and was first approved in 1972 by the Food and Drug Administration (FDA) as a treatment for the signs and symptoms of male hypogonadism. Since the introduction of transdermal testosterone gel in 2000, the market for TRT has increased dramatically for men with age-related decline of testosterone. Is testosterone replacement therapy in the offing for all aging males with low testosterone? Is such a broad androgen replacement strategy based on a positive risk–benefit assessment? What information is available on the effects of TRT in men?

Unlike in women where estrogen replacement was widely believed to carry health benefits (until disproven by the WHI project), testosterone replacement has long been suspected of carrying a significant risk in men. The landmark paper of Huggins and Hodges associating testosterone and prostate cancer outcome in men with metastatic disease, for which they were awarded the Nobel Prize in Medicine, gave rise to fears that testosterone could induce or accelerate prostate cancer in hypogonadal men receiving replacement [22]. Despite some mitigating evidence from small observational series and clinical trials, continuing concern about risk, along with a rapidly expanding number of men receiving replacement therapy led to efforts to clarify the true risk/reward relationship for TRT in aging men.

The Institute of Medicine (IOM) in 2002 conducted an independent assessment of clinical research on testosterone therapy and made recommendations for future direction for the field of TRT for the aging male [1]. In its review of the literature the committee identified “only” 31 placebo-controlled trials of testosterone therapy in older men. The placebo-controlled trial with the largest sample size involved 108 participants and the duration of therapy in 25 of the 31 trials was 6 months or less. Only one placebo-controlled trial lasted longer than a year. Therefore, assessments of risks and benefits have been limited, and uncertainties remain about the value of this therapy for older men.

The paucity of credible information is reflected in current practice guidelines informing physicians on hypogonadism and testosterone replacement therapy. Unlike HRT (that prior to WHI was recommended for all menopausal women), the Endocrine Society recommends against screening for androgen deficiency in the general population, and clinicians should measure serum testosterone levels in patients with clinical manifestations or symptoms of low testosterone.

Hypogonadism is a clinical condition characterized by consistent symptoms and signs and unequivocally low serum testosterone levels. Patients with hypogonadism are treated with testosterone to reverse the signs and symptoms of androgen deficiency; there is no indication for testosterone to be used as primary prevention of other medical conditions. These recommendations are similar to those of HRT for menopausal women issued after the WHI results became public [23].

As with other therapies, careful evaluation of patients is necessary before initiating a TRT regimen. Patients with breast or prostate cancer should not be treated with TRT. It is recommended that TRT not be given to patients with hematocrit above 50%, untreated severe obstructive sleep apnea, severe lower urinary tract symptoms (International Prostate Symptom Score [IPSS] score of ≥19), uncontrolled or poorly controlled heart failure, or in those desiring fertility. Urologic consultation is recommended in patients with palpable prostate nodule or induration, prostate-specific antigen (PSA) ≥4 ng/mL, or PSA ≥3 ng/mL in men at high risk of prostate cancer [23].

Symptoms and signs of hypogonadism

Testosterone has different functional effects; the exact expression depends on the target tissue and the point in the male life cycle that testosterone exposure occurs. In fetal life, testosterone is responsible for male sexual development, while during adolescence testosterone plays a role in male pubertal development. In contrast, symptoms of androgen deficiency in adult men are nonspecific and are often modified by age, comorbid illnesses, severity and duration of androgen deficiency, and variations in androgen sensitivity. The commonly described signs and symptoms of testosterone deficiency may represent the more severe end of the androgen deficiency spectrum since they are based on the clinician’s experience and derived from men who have sought medical assistance. In population-based surveys of community-dwelling, middle-aged and older men, low libido, erectile dysfunction (ED), and hot flushes, as well as less specific symptoms such as fatigue or lack of energy, mood changes, and reduced physical performance, were associated with low testosterone levels. In these surveys the prevalence of symptomatic androgen deficiency was around 6% [3,23–25].

Androgens exert their biological effect through a single intracellular receptor that is present in the reproductive tract as well as in many non-reproductive tissues, including bone, skeletal muscle, brain, liver, kidney, and adipocytes. Some actions of androgens are mediated by local enzymes such as 5α-reductase and aromatase [26].

An association between low testosterone (low T) and impaired fasting glucose, insulin resistance, type 2 diabetes, and the metabolic syndrome was seen in patients both with and without ED. Low T is associated not only with the metabolic syndrome, but also with each of its individual components, i.e., type 2 diabetes, visceral obesity, insulin resistance, dyslipidemia, and high blood pressure. The trinucleotide repeat sequence (CAG repeat) length could modulate the impact of testosterone on the metabolic risk factors, and men with longer CAG length and low testosterone concentrations show the highest risk of incident metabolic syndrome. Prospective studies demonstrate that a low T at baseline predicts the development of type 2 diabetes. Similarly, diabetes or metabolic syndrome at baseline may herald the onset of hypogonadism [27].

Bone mass in men declines linearly with age starting at around age 30 and the decrease in trabecular mass is greater than that in cortical mass. Hypogonadism is an important risk factor for osteoporosis in men. Androgens stimulate the proliferation of bone cells in vitro [27]. The incidence of rapid hip-bone loss was higher in men with low testosterone level [27]. The role of aromatization of testosterone to estradiol (E₂) is predominant in the protective effect of testosterone on bone loss. The association of bone mineral density (BMD) with E₂ is much stronger than that of BMD with testosterone, probably due to the fact that bone has both androgen and estrogen receptors[ 27]. While BMD in the hip increased significantly in patients receiving testosterone, little is known about the amount of testosterone required to maintain bone mass in men. Furthermore, it is not known whether the beneficial effect of testosterone on bone metabolism is due to the androgen itself or to the estrogen produced from it [28].

Testosterone stimulates erythropoiesis. Administration of testosterone increases reticulocyte counts, hemoglobin concentrations, and bone marrow erythropoietic activity through direct and indirect mechanisms. Administration of androgens results in a significant increase in the levels of endogenous erythropoietin [26]. In addition, experimental data suggest that androgens may also directly enhance erythropoiesis by stimulating erythropoietic stem cells [26].

Frailty is a syndrome of aging characterized by loss of physiologic reserve and generalized vulnerability to disability. Testosterone retains nitrogen, which accounts for its anabolic properties. Prior to puberty, testosterone promotes muscle growth. After puberty, the androgen receptor in striated muscle is downregulated and is saturated with physiologic concentrations of circulating testosterone. Low T was associated with incident frailty as well as worsening frailty, suggesting that age-related changes in blood androgens may contribute to development and progression of frailty [27]. Another study reported that patients with low T demonstrated lower scores on vitality than eugonadal men, but no differences were detected in psychological well-being between hypogonadal and eugonadal men [2].

While androgens have a key role in stimulating and maintaining sexual function in men [29–30], the role of androgens in increasing the frequency and quality of erections is unclear [31–33]. In normal young men, suppression of serum testosterone concentrations to the range associated with castration reduces sexual desire, sexual fantasies, and spontaneous erections [34]. In a random sample of 414 community-dwelling men, bioavailable testosterone was significantly correlated with both erectile function and libido [2].

The threshold testosterone level below which symptoms of androgen deficiency and adverse health outcomes occur is not known. Individual symptoms may have different thresholds (i.e., low libido versus low energy), and individual men may not exhibit the same degree of symptomatology at similar testosterone levels. Such individual thresholds may be the result of variable androgen receptor sensitivity related to CAG repeats. The accepted threshold below which symptoms are more likely to occur is approximately 300 ng/dL and more men may become symptomatic as testosterone levels decline even further [23]. The existence of a measurable threshold at which symptoms appear was further supported by Kelleher et al. [31]. Men may reach a highly reproducible specific serum concentration of testosterone that triggers symptoms of androgen deficiency, though this level differs widely among individuals. Although this threshold varied from very low to values above the lower limit of the eugonadal reference range, on average it approximated the lower limit of the eugonadal reference range for young men, 300 ng/dL [23].

While most clinicians accept the existence of a threshold value for serum testosterone symptom improvement, it may not be 300 ng/dL for everyone. A recent study that investigated the relationship between symptoms and testosterone serum levels offered additional evidence; some improvement was seen in individuals receiving TRT who did not reach testosterone ≥300 ng/dL, suggesting that their individual threshold was reached even if they did not reach the prespecified threshold [27].

While no clear relationship between decreasing testosterone and symptoms has been determined, Zitzmann et al. did observe a general trend between decreasing testosterone and increasing prevalence of groups of symptoms [24]. For example: below testosterone concentrations of 432 ng/dL (15 nmol/L), loss of libido and loss of vigor increased; below testosterone concentrations of 288 ng/dL (10 nmol/L), depression, disturbed sleep, lack of concentration, and diabetes mellitus type 2 occurred in significantly more men; and only at testosterone concentrations below 230 ng/dL (8 nmol/L) did ED occur. These findings suggest that symptom-specific testosterone thresholds occur and that the spectrum of complaints of testosterone deficiency cannot be related to a uniform threshold of testosterone concentration, but that thresholds vary with the various symptoms of testosterone deficiency [24]. As previously mentioned, CAG repeats modulate the androgen receptor activity. Because there is a genetic heterogeneity, individual patients may have different thresholds at which the various symptoms of hypogonadism appear [27].