M. Hermann, P.

Berger / Experimental Gerontology 36 (2001) 10751082

1075

Experimental Gerontology 36 (2001) 10751082 www.elsevier.nl/locate/expgero

St. Petersburg Proceedings

Hormonal changes in aging men: a therapeutic indication?

Martin Hermann, Peter Berger*
Institute for Biomedical Aging Research, Austrian Academy of Sciences, Endocrinology Unit, Rennweg 10 A6020 Innsbruck, Austria Received 13 November 2000; received in revised form 26 January 2001; accepted 26 January 2001

Abstract The rise in male life expectancy is paralleled by increased age-related clinical signs and symptoms such as muscle weakness, osteoporosis, benign prostatic hyperplasia, changes in body composition, fatigue, decreased sexual interest and activity, and increased prevalence of erectile dysfunction, all of which limit the quality of life. Many of these symptoms are similar to those of clinically well-dened hormone deciencies, e.g. Kallman syndrome, PraderLabhartWilli syndrome or deciencies due to treatment of pituitary tumors. Three male endocrine axes are characterized by age-related changes in concentrations of circulating hormones: (i) the hypothalamicpituitarytesticular axis with lower serum levels of testosterone (T) and higher serum levels of luteinizing and follicle-stimulating hormone, (ii) the hypothalamicpituitaryadrenal axis with its gradual decline in dehydroepiandrosterone (DHEA) and DHEA-sulfate (DHEAS), (iii) the growth hormone (GH)/insulin-like growth factor I (IGF-I) axis showing decreased hormone production concomitant with symptoms similar to those of GH-decient adults. The benecial effects of hormone replacement in nonelderly hormone-decient individuals and in postmenopausal women raised hope that hormone substitution might prevent or even reverse some of the symptoms of male aging. However, this approach is hampered by the lack of individual age-related hormone reference values and reliable clinical read-out parameters. The ndings so far do not support the need for widespread hormone replacement in elderly men. Larger long-term prospective studies are needed to identify clinically useful read-out parameters, and then demonstrate that hormone replacement can translate into functional parameters, thereby providing the individual benet of treatment for aging men. q 2001 Elsevier Science Inc. All rights reserved.
Keywords: Aging; Hormone replacement; Male; T; DHEA; GH; Melatonin

* Corresponding author. Tel.: 143-512-583919-24; fax: 143-512-583919-8. E-mail address: peter.berger@oeaw.ac.at (P. Berger). 0531-5565/01/$ - see front matter q 2001 Elsevier Science Inc. All rights reserved. PII: S 0531-556 5(01)00113-9

1076

M. Hermann, P. Berger / Experimental Gerontology 36 (2001) 10751082

1. Introduction Progress in medical care such as vaccines and the advent of antibiotics, have dramatically increased the average life expectancy in most industrialized countries. Unfortunately this progress is paralleled by age-associated morbid and premorbid changes such as muscle weakness, general frailty, osteoporosis or prostate hyperplasia (BPH), that limit free and independent life for elderly men. As a consequence, a major task of modern society should not be to merely prolong but to extend independent and healthy life, thereby increasing the quality of life and, as a byproduct, lowering the costs of care for the elderly (rev. in Hermann et al., 2000; Vermeulen, 2000). During aging and intercurrent pathologic processes, physical performance gradually declines in both genders. Amongst other changes, aging in men is characterized by decreased sexual interest and acitvity, and an increased prevalence of erectile dysfunction, as shown by the Massachusetts Male Aging Study (Feldman et al., 1994). In addition, elderly men may suffer from several physical and emotional symptoms paralleled in some individuals by hormonal changes, reminiscent of those in women during perimenopausal transition. Loss of female cyclicity and reproductive function, i.e. menopause, is a universal nding by the age of approx. 50. Already as early as age 40, the frequency of ovulation decreases, and reproductive ovarian function ceases within the next 15 years. Since men generally do not experience a rapid decline of gonadal function or irreversible arrest of reproductive capacity in old age, the term `andropause' is not appropriate. Nevertheless, gradual declines in hormone serum levels can be observed (Fig. 1) (Gray et al., 1991; Gooren, 1996; Zadik et al., 1985). The decrease in serum Testosterone (T) levels during aging is characterised by high interindividual variability, thus, by far not all ageing men will become hypogonadal to a clinically signicant degree.

Fig. 1. Changes of hormone serum levels in aging men. Annual changes in percentage of mean values between the second and sixth decade of life for growth hormone (GH) and the third and seventh decade for total testosterone (total T), bioavailable T (bio T), sex hormone binding globulin (SHBG), dehydroepiandrosterone (DHEA), DHEA-sulfate (DHEAS), luteinising hormone (LH), follicle-stimulating hormone (FSH) (adapted from Gray et al., 1991; Zadik et al., 1985).

M. Hermann, P. Berger / Experimental Gerontology 36 (2001) 10751082

1077

In addition to T, three other hormones have received attention as likely candidates for hormone replacement therapy in aging men: Growth hormone (GH), dehydroepiandrosterone (DHEA) and melatonin, their prevalence in the popular press is in an inverse relationship to their scientically proven benet for elderly men. In 1990, Rudman suggested that GH may reverse some symptoms associated with aging; unfortunately, his longer study showed that side effects, such as the carpal tunnel syndrome, may limit long-term use of GH. To circumvent this complication oral GH secretagogues have been developed that release physiological amounts of GH from the pituitary. The clinical usefulness of the GH secretagogues has yet to be proven. Issues addressing the safety of GH administration are of special importance, since it was shown that middle-aged persons with higher GH levels are likely to die earlier than those with lower GH levels. Moreover, a study using mice with genetic disorders of GH secretion showed that reduced GH secretion may prolong survival (rev. in Morley and Perry, 2000). The results of the controlled clinical trials that have been reported so far do not support the concept that GH replacement would materially benet the daily function of older men (rev. in Marcus and Hoffman, 1998). There are only a few scientic reports dealing with DHEA replacement in aging men. Due to the small number of subjects and the short duration of hormone replacement, insufcient data have been gathered with respect to long term safety or benecial effects of DHEA (rev. in Marcus and Hoffman, 1998; Hermann and Berger, 1998). In spite of its impressive age-dependent decrease it is until now impossible to determine whether this contributes to the clinical symptomatology of elderly men. Concerning melatonin, there are too few clinical studies to allow a meaningful assessment of its role as a candidate for hormone replacement therapy. Melatonin may have some benecial effects in elderly men via its ability to modulate sleep. Its in vitro activities as an antioxidant or neuroimmunoendocrine modulator are currently under investigation (rev. in Brzezinski, 1997; Hermann and Berger, 1999). 2. Age-related changes in the hypothalamicpituitarygonadal axis Testosterone and, to a greater extent, free T and bioavailable (bio) or weakly bound T (albumin-bound T) serum levels decline with age in men (for reviews see Morley and Perry, 2000; Hermann and Berger, 1999), as shown by a decrease of approx. 35% of total and of 50% of free T levels between the age of 20 and 80 years. Plasma levels of free T below the lower normal limit (,7 ng/dl) occur in 7% of elderly men aged 4060, 20% are affected in the age group 6080, and 35% in the age group over 80 years (Vermeulen and Kaufman, 1995). There is no clear consensus about the endocrine causative mechanisms of this rather modest age-associated decline (Fig. 1) the origin of which may be at all three levels of the hypothalamopituitarytesticular axis (Morley and Perry, 2000; Hermann and Berger, 1999; Gooren, 1996). At the testicular level, decreased numbers of Leydig cells (Neaves et al., 1984), impaired testicular perfusion (Suoranta, 1971) and a reduced release of T upon stimulation by hCG (Rubens et al., 1974; Harman and Tsitouras, 1980) have been reported (i.e. primary hypogonadism). Although elevated LH serum levels are common in elderly

1078

M. Hermann, P. Berger / Experimental Gerontology 36 (2001) 10751082

men, these levels in response to the decline of T levels with aging are lower than those observed in younger men with similarly decreased T (Korenman et al., 1990). These ndings speak in favour of alterations in the hypothalamicpituitary unit (i.e. secondary hypogonadism). Above the age of 80, elevated LH levels may also be explained by the loss of the increased opioid tone of middle aged men (rev. in Morley and Perry, 2000). Nevertheless, elderly men fail to sufciently increase LH secretion despite having a lower opioid tone (Vermeulen et al., 1989). A hypothalamic impairment of stimulating LH, and subsequently androgen secretion is inferred from the loss of circadian rhythm of LH and T levels (Bremner et al., 1983; Pincus et al., 1996). The nycthemeral variations in plasma T levels are signicantly decreased in elderly men (Deslypere and Vermeulen, 1984). The reduced number of spontaneous high amplitude LH pulses in elderly men does not seem to be a consequence of decreased sensitivity of the gonadotrophs to gonadotropin releasing hormone (GnRH), but of the release of smaller amounts of GnRH at each pulse (Kaufman et al., 1991). In addition to aging per se, other hereditary, environmental (obesity, stress), psychosocial (depression, smoking, drugs) or socioeconomical (diet, hygiene) factors may result in even lower circulating T levels in elderly men (Vermeulen and Kaufman, 1995). The differential diagnosis of androgen deciency is generally rst classied by primary testicular failure (hypergonadotropic or primary hypogonadism) vs. a malfunction at the hypothalamic/pituitary level (hypogonadotropic or secondary hypogonadism). Both, testicular and hypothalamic/pituitary levels, may be involved in age-related changes of T levels (Fig. 2). 3. Is the decrease in androgen levels in elderly men of clinical relevance? Many of the clinical features accompanying the aging process, such as decreased muscle mass, strength and energy, decreased virility, libido and sexual activity, increased frequency of impotence, decreased cognitive function and decreased general well-being, are also observed in androgen-decient young men. This led to speculations that lower androgen levels in aging men may be responsible for some of these changes, and that androgen replacement therapy may prevent, retard or even reverse some of these agerelated clinical features. Although one would expect a signicant correlation between these clinical features and androgen levels, this is not the case, since most of the clinical features accompanying the aging process are non-specic and of multifactorial origin (Vermeulen, 2000). Testosterone levels required to sustain sexual interest in young men are rather low, T replacement therapy at a dose maintaining T levels at half the baseline value being sufcient (Bagatell et al., 1994). In a retrospective study in which elderly hypogonadal men (bio T ,72 ng/dl) were treated with T the self-assessment of libido was dramatically improved (Hajjar et al., 1997). Although the incidence of erectile dysfunction increases with age, androgen deciency is rarely the major cause of impotence in elderly men. Since the androgen levels necessary to sustain libido are higher than those for erectile function, the patients consulting their physician presumably do have a normal libido, and thus T levels high enough to sustain

M. Hermann, P. Berger / Experimental Gerontology 36 (2001) 10751082

1079

Fig. 2. Changes with age in the hypothalamic pituitary gonadal (HPG) axis in men. Gradual changes with age in the regulation of the HPG axis result in a mixture of partial dysfunctions at each organic level. Primary gonadal failure leading to partial hypergonadotropic hypogonadism (left scheme) is intermingled with central defects of the hypothalamicpituitary unit usually causing only mild secondary gonadal failure or hypogonadotropic hypogonadism (right scheme). Age-associated changes at the testicular level are reductions in the numbers of Leydig and Sertoli cells, thickening and hernia-like protrusions of the basal membrane of the seminiferi tubules, development of vacuolization and accumulation of lipofuscin within the Leydig cells and a decrease in the T production capacity when the testes are stimulated by human chorionic gonadotropin; moreover SHBG levels in serum are upregulated with age: The sum of all these changes is meant to result in diminished levels of bioavailable T. This in turn should lead to compensatory upregulation of the hypothalamicpituitary unit but concomitant changes at these levels, i.e. failure of the hypothalamus to generate appropriate amplitudes of the pulsatile secretion bursts of gonadotropin-releasing hormone (GnRH), higher sensitivity to the suppressive effects of T, and smaller amounts of GnRH released at each pulse, seem to be responsible for age-associated inadequate T production and availability. Moreover, despite declining opioidergic suppression of GnRH secretion, old men fail to adequately restore GnRH and LH synchronous pulsatility and amplitudes to compensate for mostly mild primary hypoganadism reected by too low T serum levels (rev. in Hermann and Berger, 1998; Morley and Perry, 2000; Plas et al., 2000).

erectile function. Therefore, non-hormonal causes, such as atherosclerosis or polyneuropathy, may be involved (Morley, 1986). Secondary hypogonadism and impotence are two common, but independently distributed conditions of older men (Korenman et al., 1990). Aging is paralleled by a decrease in muscle mass and an increase in fat mass, predominantly visceral fat, which is the major cause for insulin resistance and the atherogenic rin et al., 1992). The correlation between muscle mass and free T levels, lipid prole (Ma and the increase of muscle mass/decrease of abdominal fat after androgen substitution in men underline the importance of androgen deciency for age-associated changes in body composition. Nevertheless, it is also clear that other factors, such as age-associated decline in GH levels and decreased physical activity, can be important co-determinants for these

1080

M. Hermann, P. Berger / Experimental Gerontology 36 (2001) 10751082

changes (rev. in Vermeulen, 2000). In a double blind study in which 108 men over 65 years old were randomized to wear either a T patch or a placebo patch for 36 months, fat mass decreased and lean mass increased signicantly in the T-treated group. Muscle strength of knee extension and exion was, however, not signicantly different in the two groups (Snyder et al., 1999a). Concerning atherosclerosis and coronary stenosis, two frequent age-associated diseases, most epidemiological studies surprisingly show an inverse correlation between T levels and the atherogenic lipid prole, atherosclerosis or the degree of coronary stenosis, respectively. Supplementation with androgens within the physiological range improves the lipid prole in elderly men, probably via an increase of insulin sensitivity. In this context, it is important to mention that non-aromatizable or alkylated androgens as well as pharmacological doses of androgens are clearly atherogenic, and can lead to cardiovascular accidents (rev. in Vermeulen, 2000). A positive correlation between bone mass and T levels has been observed in aging men (Orwoll and Klein, 1995). It was shown by Snyder and co-workers that increasing the serum T concentrations in men over 65 years of age did not increase lumbar spine bone density overall, except in those men with low pretreatment serum concentrations (Snyder et al., 1999b). Nevertheless, it has to be kept in mind that only a few percent of osteoporotic men are hypogonadal. It is evident that only men with subnormal free or bioavailable T levels are candidates for androgen substitution. The symptoms and clinical signs, such as decreased energy, mood, libido, muscle mass and strength, bone mass, the regression of secondary sex characteristics or the increase of abdominal fat mass (decreased sense of general wellbeing or libido are more difcult to objectify), can be used to suspect T deciency. Conrming this diagnosis is usually not easy since subnormal T levels are dened on the basis of data in young men, the relevance of which for elderly men is controversial. Exact criteria for placing an older man on T therapy remain to be determined. If total serum T levels are repeatedly below 350 ng/dl, current guidelines suggest treatment with T (Tenover, 2000). What are the contra-indications for an androgen supplementation/substitution? Concerning the prostate, an androgen-dependent organ, the presence of a BPH is not a contra-indication. In contrast, although the role of androgens for the development and progression of a subclinical prostatic carcinoma (an androgen-sensitive tumor) remains unknown, prostatic cancer is an absolute contra-indication and should be excluded by rectal examination and PSA measurement. During androgen treatment, any increase of PSA levels with more than 0.75 ng/ml in two consecutive controls requires further exploration. Although rare, the presence of a mammary carcinoma as well as of untreated prolactinoma are also contra-indications for T therapy since increases in serum estradiol levels due to T therapy may stimulate the growth of these tumors. A clearcut atherogenic lipid prole and polycythemia are two relative contra-indications. A dose reduction may be necessary in patients with cardiac decompensation or hypertension, since water and salt retention are inuenced by T (rev. in Vermeulen, 2000). Testosterone can be administered via i.m. injections, transdermally i.e. using patches, or orally. Intra-muscular injections of T enanthate or cypionate (200250 mg every two

M. Hermann, P. Berger / Experimental Gerontology 36 (2001) 10751082

1081

weeks) have the disadvantage of supraphysiological T levels for the rst 23 days after injection. These should be avoided, since they may cause gynecomastia due to T conversion into estradiol insulin resistance and an atherogenic lipid prole (Bagatell et al., 1994). Between injections, T levels continously decline to low normal or even slightly subnormal levels. More physiological levels of T may be achieved with scrotal patches daily applied at 10 p.m., thereby mimicking the nycthemeral variations of T. Testosterone undecanoate, an orally active form, has the disadvantages that only variable and relatively unpredictable serum levels can be obtained. In summary, although it is well established that T levels decline with age in men, the contributions of this decline to deleterious age-associated symptoms, such as muscle dystrophy, decreased bone mass, libido, cognitive function and functional status are still not fully understood. A benecial effect of T replacement therapy is only present in men `truly` hypogonadal, but the exact criteria for placing elderly men on therapy have not yet been determined. Further long-term studies with larger patient populations are needed to address several unanswered questions, particularly those regarding the potential benet of hormone replacement in eldery men who suffer from several concomitant diseases. Acknowledgement The authors acknowledge the support of the Austrian Science Funds (P13652-GEN). References
Bagatell, C.J., Heiman, J.R., Matsumoto, A.M., Rivier, J.E., Bremner, W.J., 1994. Metabolic and behavioral effects of high dose exogenous testosterone in healthy men. J. Clin. Endocrinol. Metab. 79, 561567. Bremner, W.J., Vitiello, M.V., Prinz, P.N., 1983. Loss of circadian rhythmicity in blood testosterone levels with aging in normal men. J. Clin. Endocrinol. Metab. 56, 12781281. Brzezinski, A., 1997. Melatonin in humans. N. Engl. J. Med. 336, 186195. Deslypere, J.P., Vermeulen, A., 1984. Leydig cell function in normal men: effect of age, life-style, residence, diet, and activity. J. Clin. Endocrinol. Metab. 59, 955962. Feldman, H.A., Goldstein, I., Hatzichristou, D.G., Krane, R.J., McKinlay, J.B., 1994. Impotence and its medical and psychosocial correlates: results of the Massachusetts Male Aging Study. J. Urol. 151, 5461. Gooren, L.J.G., 1996. The age-related decline of androgen levels in men: clinically signicant?. Brit. J. Urol. 78, 763768. Gray, A., Feldmann, A., McKinlay, J.B., Longcope, C., 1991. Age, disease and changing sex hormone levels in middle-aged men: results of the Massachusetts Male Aging Study. J. Clin. Endocrinol. Metab. 73, 1016 1625. Hajjar, R.R., Kaiser, F.E., Morley, J.E., 1997. Outcomes of long-term replacement in older hypogonadal males: a retrospective analysis. J. Clin. Endocrinol. Metab. 82, 37933796. Harman, S.M., Tsitouras, P.D., 1980. Reproductive hormones in ageing men. Measurement of sex steroids, basal LH, and Leydig cell response hCG. J. Clin. Endocrinol. Metab. 51, 3540. Hermann, M., Berger, P., 1998. Ageing of the male endocrine system. Rev. Physiol. Biochem. Pharm. 139, 90 122. Hermann, M., Berger, P., 1999. Hormone replacement in the aging male?. Exp. Gerontol. 34, 923933. Hermann, M., Untergasser, G., Rumpold, H., Berger, P., 2000. Aging of the male reproductive system. Exp. Gerontol. 35, 12671279. Kaufman, J.M., Giri, M., Deslypere, J.M., Thomas, G., Vermeulen, A., 1991. Inuence of age on the

1082

M. Hermann, P. Berger / Experimental Gerontology 36 (2001) 10751082

responsiveness of the gonadotrophs to luteinizing hormone-releasing hormone in males. J. Clin. Enodcrinol. Metab. 71, 12551260. Korenman, S.G., Morley, J.E., Mooradian, A.D., Davis, S.S., Kaiser, F.E., Silver, A.J., Viosca, S.P., Garza, D., 1990. Secondary hypogonadism in older men: its relation to impotence. J. Clin. Endocrinol. Metab. 71, 963 969. Marcus, R., Hoffman, A.R., 1998. Growth hormone as therapy for older men and women. Annu. Rev. Pharmacol. Toxicol. 38, 4561. rin, P., Holma ng, S., Jo nnsen, L., Sjo stro m, L., Kvist, B., Holm, G., Lindstedt, G., Bjo rntorp, P., 1992. The Ma effects of testosterone treatment on body composition and metabolism in middle aged men. In. J. Obesity 16, 991997. Morley, J.E., 1986. Impotence. Am. J. Med. 80, 897906. Morley, J.E., Perry, H.M., 2000. Androgen deciency in aging men: role of testosterone replacement therapy. J. Lab. Clin. Med. 135, 370378. Neaves, W.B., Jonson, L., Porter, J.C., Parker Jr., C.R., Petty, C.S., 1984. Leydig cell numbers, daily sperm production and serum gonadotropin levels in aging men. J. Clin. Endocrin. Metab. 59, 756763. Orwoll, E., Klein, R., 1995. Osteoporosis in men. Endocrin. Rev. 16, 87116. Pincus, S.M., Mulligan, T., Iranmanesh, A., Gheorghiu, S., Godschalk, M., Veldhuis, J.D., 1996. Older males secrete luteinizing hormone and testosterone more irregularly, and jointly more asynchronously, than younger males. Proc. Natl. Acad. Sci. USA 93, 14,10014,105. ger, H., 2000. How is fertility affected in aging men?. Exp. Gerontol. 35, Plas, E., Berger, P., Hermann, M., Pu 543551. Rubens, R., Dhont, M., Vermeulen, A., 1974. Further studies on Leydig cell function in old age. J. Clin. Endocrinol. Metab. 39, 4045. Snyder, P.J., Peachey, H., Hannoush, P., Berlin, J., Loh, L., Lenrow, D.A., Holmes, J.H., Dlewati, A., Santanna, J., Rosen, C.J., Strom, B.L., 1999a. Effect of testosterone treatment on body composition and muscle strength in men over 65 years of age. J. Clin. Endocrinol. Metab. 84, 26472653. Snyder, P.J., Peachey, H., Hannoush, P., Berlin, J.A., Loh, L., Holmes, J.H., Dlewati, A., Staley, J., Santanna, J., Kapoor, S.C., Attie, M.F., Haddad, J.G., Strom, B.L., 1999b. Effect of testosterone treatment on bone mineral density in men over 65 years of age. J. Clin. Endocrinol. Metab. 84, 19661972. Suoranta, H., 1971. Changes in the small blood vessels of the adult human testis in relation to age and to some pathological conditions. Virch. Arch. Abt. A Path. Anat. 352, 165181. Tenover, J.L. 2000. Testosterone for older men. Endocrine Society, Toronto Ontario, Canada. Vermeulen, A., 2000. Andropause. Maturitas 34, 515. Vermeulen, A., Kaufman, J.M., 1995. Aging of the hypothalamopituitarytesticular axis in men. Horm. Res. 43, 2528. Vermeulen, A., Deslypere, J.P., Kaufman, J.M., 1989. Inuence of anti-opioids on luteinizing hormone pulsatility in aging men. J. Clin. Endocrinol. Metab. 68, 6872. Zadik, Z., Chalew, S.A., McCarter Jr., R.J., Meistas, M., Kowarski, A.A., 1985. The inuence of age on the 24 h integrated concentration of growth hormone in normal individuals. J. Clin. Endocrinol. Metab. 60, 513516.