Fertility, androgen production and sensitivity, and sexual function in aging men

Menopause in women is a discrete event in the life cycle, and is marked by the cessation of menses, in association with a sharp fall in circulating estradiol levels and a rise in follicle-stimulating hormone (FSH). Although no single event delineates reproductive senescence in men, results of many investigations suggest that aging men experience reductions in androgen levels, virility, and fertility, along with related metabolic changes. Nonetheless, the question of a ``male menopause'' remains controversial, in part because of the difficulty in discriminating the effects of age-related confounding variables such as stress, nonendocrine illnesses, malnutrition, obesity and drug or medication use, from aging per se.

Aging of seminiferous tubules and fertility
With regard to fertility, despite occasional reports of paternity in men in their 90s, there is clearly a decrease in the rate of conception in old male/young female marriages. Semen analyses in elderly men reveal normal sperm numbers, but decreased sperm motility and increases in abnormal forms. Changes in seminiferous tubules with age include thickening of the basement membrane, peritubular fibrosis, sclerotic narrowing or collapse of the lumen, patchy impairment of germ cell maturation, and immaturity or degeneration of spermatocytes, as well as increases in multinucleated Sertoli cells. Studies have also revealed a small but measurable, decrease in average testis size with advancing age, whether studied at necropsy or in vivo. Functioning seminiferous tubules exercise negative feedback control on the pituitary gonadotrope by producing inhibin, a peptide that acts to reduce FSH production. Thus, with damage or destruction of the tubules, FSH increases, even if Leydig cell testosterone secretion remains normal. In fact, basal levels of FSH increase with age, most prominently in men with the most marked changes in seminiferous tubular morphology. There is a corresponding decrease in basal serum inhibin levels. Thus, evidence favors the hypothesis that an intrinsic age-related reduction in seminiferous tubular function leads to reduced inhibin secretion with secondary effects on pituitary function.

Aging effects on sex hormone secretion and bioavailability
As for sex hormone secretion, early studies demonstrated a reduction in bioassayable urinary androgen, and subsequent investigations of small numbers of older men showed decreased testicular vein testosterone levels, and reductions in both metabolic clearance and production rates of testosterone. Initially basal levels of total plasma testosterone were shown to decrease progressively after age 50. Because subjects in these studies were not always carefully screened for health factors, the confounding effects of illness, medications, etc. may have accounted for some of the results observed. Illness does affect reproductive function as is demonstrated by a report of decreased plasma free testosterone in men with benign lung disease and a reduced total and free hormone level in men with lung malignancy. In some studies of exceptionally healthy men, no age effect on circulating testosterone concentration was found. Subsequent investigations examining multiple samples over a 24 hour period demonstrated that morning peak (but not afternoon nadir) and 24 hour mean integrated testosterone levels were decreased in older men, but in a similar study no age effect on circadian levels was observed. Thus, the question whether aging per se significantly reduces morning peak testosterone secretion or total levels remains controversial.

The fraction of circulating testosterone bound to sex hormone binding globulin (SHBG) is considered to be biologically unavailable. Because an increase occurs with age in circulating SHBG, older men may exhibit reductions in bioavailable testosterone, and hence its effects, disproportionate to what would be expected from measurements of total testosterone. However, in healthy men the increase in plasma testosterone binding to SHBG appears insufficient to significantly alter the apparent free concentration.

Although there are no large longitudinal investigations of the effects of age on sex steroids in men, two cross-sectional studies, each examining more than 1000 individuals, have recently been reported. In one study there was a significant downward trend with age in both total and non-SHBG bound testosterone concentrations, while in another no significant trend in plasma testosterone was found. Neither study demonstrated an increase with age in the number of men with truly hypogonadal androgen levels. A recent meta-analysis of studies of androgens in aging men revealed a significant inverse correlation of total plasma testosterone with age which disappeared when reports which included men with ill health were omitted. This analysis also found that investigations which included ill or institutionalized subjects consistently showed lower levels of testosterone overall.

5alpha-Dihydrotestosterone (DHT) is produced from testosterone and is the "activated'' form which binds to cytoplasmic androgen receptor in most tissues. This testosterone metabolite can be formed in the liver and also ``leaks back'' from androgen target tissues, so that it circulates in plasma at about 20% of total testosterone levels. Both reduced and unaltered plasma levels of total or free DHT have been reported in older men. In one study of elderly men, many of whom had benign prostatic hyperplasia, there were high plasma levels of DHT, but subnormal levels of testosterone, suggesting an increase with age in peripheral 5alpha reduction of testosterone, possibly in prostate tissue.

Necropsy studies have generally revealed an age-related decrease in number, and an increase in morphological abnormalities, of the Leydig cells, but in some of these investigations patients had died of a malignant disease or other protracted illness. Mean basal plasma levels of luteinizing hormone (LH), as well as urinary excretion of bioassayable gonadotropins, increase progressively in men beyond the age of 50, and human chorionic gonadotropin (hCG) stimulation tests have uniformly revealed diminutions in the testosterone secretory response in older men, consistent with an age-related decrease in Leydig cell number and/or reserve secretory capacity. Thus, the evidence favors some degree of primary testicular failure in aging men.

There is also evidence for an effect of aging on hypothalamic-pituitary function. For example, there is a pattern of low or normal LH in a significant fraction of older men with diminished testosterone levels. Stimulation of the pituitary with exogenous gonadotropin-releasing hormone (GnRH) has revealed decreases in the magnitude of LH and/or FSH responses in older men. Clomiphene citrate treatment also results in less gonadotropin response in older men. The finding that there is an attenuation of the amplitude of spontaneous LH secretory bursts in normal older men also provides evidence of altered hypothalamic-pituitary function. The relative contribution of hypothalamic versus pituitary dysfunction remains uncertain, but in two recent studies, repeated pulsing of GnRH appeared to restore LH secretory responsiveness in older men, suggesting that the decrease in pituitary gonadotropin secretion is mainly due to reduced hypothalamic GnRH production.

Aging effects on androgen target tissues
Aging might also reduce androgen effect by causing a loss of sensitivity of target tissues to
testosterone or DHT. Both decreased and increased sensitivity of pituitary gonadotropin secretion to feedback regulation by androgens have been reported in older men. Binding of DHT to sex hormone responsive skin is also decreased with age, suggesting that an age-related reduction in responsivity to androgens may result from alterations in receptor number or affinity. To date, there are no published reports regarding effects of aging on specific post-receptor actions of sex steroids.

Reproductive aging and sexual function, body composition and metabolism
It is not known whether the changes in androgen levels or action in aging men have any deleterious clinical effects. Many studies have recorded progressive declines in male sexual interest, activity and performance with age, and there is a striking increase in the prevalence of impotence to as much as 50-75% in men over 75. However, after adjustment for age and body mass index (BMI = weight/height squared), there is no difference in bioavailable testosterone levels in potent versus impotent old men, suggesting that hypogonadism and impotence are independently distributed conditions. Because impotence in older men is likely to be due mainly to neurologic or vascular changes, the value of replacing testosterone to improve erectile function in the elderly is questionable.

In young hypogonadal men reduced sex drive rather than impotence is the primary symptom of diminished androgen action. The most important predictors of sexual interest and activity in old men are their characteristic level of sexual activity in youth, their health, and the health of the spouse or partner. Nonetheless, old men with relatively high sexual activity levels have been reported to have greater total or bioavailable plasma testosterone than age-matched men with less sexual activity. Other studies have shown weak, but statistically significant, inverse correlations of free or bioavailable testosterone levels with sexual thoughts, sexual activity, and morning erections in aging men, although statistical significance may be lost when data are adjusted for the effects of age. Thus, while decreases in serum testosterone may contribute to the diminished sexual activity in older men, this effect is probably minor compared with the contributions of age-related alterations in psychological, social, neurological, vascular and health factors. If it is to be used at all, testosterone replacement should probably be reserved for older men who are frankly hypogonadal.

In old age, there are decreases in muscle and bone mass and increases in body fat, with fat redistribution from peripheral to central depots. The latter changes are associated with altered glucose and lipid metabolism and increased risk of diabetes mellitus and cardiovascular disease. Because men have lower levels of high density lipoprotein (HDL)-cholesterol and a greater risk of coronary vascular disease than do women, it might be expected that a decrease in serum testosterone would beneficially affect atherosclerotic risk. However, in one recent study HDL-cholesterol levels were positively correlated with serum free testosterone in men aged 30-79 years, and in another study middle-aged men treated with testosterone had decreases in intra-abdominal fat, as measured by computed tomography, and in insulin resistance, as measured by the glucose clamp technique. Testosterone treatment also produced decreases in fasting glucose, diastolic blood pressure and serum cholesterol. Although testosterone treatment has been shown to improve bone density and skeletal muscle mass and strength in older hypogonadal men, the contribution of diminished testosterone to the loss of muscle or bone mass during normal aging is unknown.

Summary
Although there is no inevitable event leading to age-related hypogonadism in men, there is evidence that a significant number of older men have modest reductions in androgen levels. The metabolic and clinical sequelae of this change remain to be defined as does the risk/benefit ratio of androgen supplementation.

Suggested Reading

Gary A, Berlin JA, McKinlay JB, Longcope C. An examination of research design effects on the association of testosterone and male aging: results of a meta-analysis. J Clin Epidemiol 1991;44:671-684.

Vermeulon A. Clinical review 24: androgens in aging male. J Clin Endocrinol Metab 1991;73:221-224.

Blackman MR, Elahi D, Harman SM. Endocrinology and aging. In: DeGroot LJ et al, eds. Endocrinology (3rd ed.) New York: Grune and Stratton, 1995:Chap. 147, 2702-2730.

Veldhuis JD, Urban RJ, Lizarralde G, Johnson ML, Iranmanesh A. Attenuation of luteinizing hormone secretory burst amplitude as a proximate basis for the hypoandrogenism of healthy aging in men. J Clin Endocrinol Metab 1992;75(52-58).

Korenman SG, Morley JE, Mooradian AD. et al. Secondary hypogonadism in older men: Its relation to impotence. J Clin Endocrinol Metab 1990;71:963-969.

Marin P, Holmang S, Jonsson L, et al. The effects of testosterone treatment on body composition and metabolism in middle-aged obese men. Int J Obes 1992;16:991-997