Testosterone replacement: when is there a role?

Transcription

1 (2000) 12, Suppl 4, S112±S118 ß 2000 Macmillan Publishers Ltd All rights reserved /00 $ Testosterone replacement: when is there a role? 1 * 1 Queens University, Department of Urology, Kingston General Hospital, Kingston, Canada Hypogonadism is an uncommon cause of erectile dysfunction. Unfortunately, hypogonadal states in adult males are dif cult to diagnose on purely clinical grounds and it is necessary to seek biochemical support. The simplest way to establish the diagnosis of hypogonadism is by determination of serum testosterone levels. Several methods exist but total testosterone determination plus assessment of sex hormone-binding globulin or bio-available testosterone appear to be the most reliable and accessible. Once a diagnostic of hypogonadism has been established in a man with erectile dif culties, a trial of androgen supplementation is warranted if no contraindications exist. Knowledgeable monitoring is essential. In the absence of an adequate response, co-morbidities should be diligently sought out. In the absence of reliable guidelines for androgen administration to patients with erectile failure, a set of recommendations are provided. (2000) 12, Suppl 4, S112±S118. Keywords: erectile dysfunction; pharmacotherapy; androgens; review Introduction The aging process in men is associated with variable but sometimes profound hormonal alterations and a decrease in sexual performance. 1 This should not be construed to imply that the latter is exclusively a consequence of the endocrine changes. The Massachusetts Male Aging Study (MMAS) was unable to associate the coexistence of ED and changes in the serum levels of sex hormones with the notable exception of a direct correlation between sexual dysfunction and a serum de cit in dehydroepiandrosterone (DHEA) and its sulfated form (DHEAS). This nding is dif cult to interpret because, although hypotestosteronemia and aging show a very marked inter-individual variability, the association of low DHEA and advancing age exhibits a more predictable, constant and profound association. In other words, ED and the decease in serum levels of DHEA is so prevalent in the elderly that their relationship may simply be coincidental, and not necessarily related. It is widely accepted that androgens are important for general sexual function and the role of androgens in erectile physiology is particularly fundamental. Current estimates indicate that 1 in 200 male adults have abnormally low levels of testosterone (T) and most of these men are candidates for androgen supplementation therapy. 2 *Correspondence:, Queens University, Department of Urology, Kingston General Hospital, Kingston, ON K7L 2V7 Canada. Moralesa@post.queensu.ca Testosterone and sexual function Although it is possible to obtain erections and maintain some semblance of libido with minimal levels of androgens, in general castrate levels of T result in suppression of both sexual interest and performance. 3 Nevertheless, the prevalence of hypogonadism in men with ED is low 4 and its presence may not re ect causality. Solid evidence has shown that profound hypogonadism results in the early abolition of erections associated with rapid eye movement (REM) sleep while erotically stimulated erectile function may be preserved during wakefulness. 5 In addition to having a detrimental effect on erectile activity, hypogonadism brings a diminution in the frequency of sexual thoughts and intercourse. Its adverse effect on the volume of the ejaculate and semen quality are also well recognized. 6 Should men with ED be investigated for hypogonadism? A great deal of controversy has existed in the literature regarding this issue. Those opposing the routine hormonal assessment argue that hypogonadism is a rare cause of ED, the cost of the testing is considerable and, above all, it can be suspected by its effect on libido. Let us examine these points individually. The prevalence of hypogonadism in the age group with the highest incidence of ED (middle age and

2 beyond) is higher than initially suspected. 7 Morley et al 8 reported a signi cant incidence of androgen de ciency in males, as measured by serum levels of bioavailable T. Nevertheless, it is recognized that hypogonadism is rarely the main etiological factor in ED, a view further supported by the limited success of adequate exogenous T supplementation in the treatment of ED. 9,10 The cost of determining serum T is dif cult to assess and depends on a number of variables. These variables include: the degree of laboratory sophistication, the type of T determination (total, free or bioavailable), the number of tests requested and the type and number of supplementary tests ordered (determination of sex hormone-binding globulin, gonadotropins) to facilitate accurate interpretation of the results. The basic determination of total serum T is inexpensive (under $10.00). The view that the diagnosis of hypogonadism can be established on the basis of history and physical examination is a fallacy. Adult hypogonadism, except in the most extreme cases, is dif cult to diagnose accurately without biochemical investigations. 11 Thus, sexual interest is a notoriously unreliable symptom that may be prominent in men with normal T levels but affected by other conditions (eg depression 12 ). Testicular size and consistency in the adult male show marked inter-individual variability and the appearance of secondary sex characteristics are not suf cient to suspect or rule out the presence of hypoandrosteronemia. 13 The clinician is, therefore, obliged to seek biochemical assistance before embarking upon treatment of ED. It should be noted, however, that the diagnosis of a medically signi cant hypogonadism should be based on a comprehensive evaluation of the patient and his biochemical test results. Finally, the author's opinion is that it is clinically parochial and inappropriate to consider hypogonadism only from a purely sexual perspective. After all, as noted above, men seeking professional help for ED belong to an age group in which there is clear and signi cant increase in the prevalence of deterioration in the function of the hypothalamic-pituitary-gonadal axis. 1,6,9,12 Unquestionably, it is the responsibility of the clinician (in general) and the urologist (speci cally) to rule out the presence of a hypogonadal state with its subtle but dire consequences beyond sexual problems (ie, depression, osteoporosis, detrimental alterations in lipid pro le, etc). 14 The minimal hormonal evaluation Measuring total serum T levels, preferably in the morning between 8 and 10 am can determine the androgenic milieu. This is the simplest, least expensive and readily accessible test method to rule out an abnormality in serum T. However, the results may be misleading, particularly in obese and in elderly men because this subset of patients commonly exhibit increased levels of sex-hormone binding globulin (SHBG), which prevents T from becoming metabolically active. Free T levels may provide a better estimation of testicular endocrine function but the fastidiousness required in the technique results in inter- and intra-laboratory variability, casting reservations on the accuracy of the test. Measuring bioavailable T (the sum of free albumin-bound T) may be the most reliable measurement because it determines the amount of T which is available to target tissues. Unfortunately, determinations of bioavailable T are not widely performed and are more costly. An appropriate compromise that takes into consideration the accuracy of test results, cost and availability is the simultaneous determination of total T and SHBG by immunoassay. 15 Regardless of the test employed, if an abnormal value is obtained with the initial determination, the results should be con rmed with a second test, particularly when the abnormality is borderline. If the abnormality is con rmed, it may then be appropriate to further investigate the hypothalamus ± pituitary ± gonadal axis by measuring levels of gonadotropins (FSH and LH). The measurement of prolactin may also be appropriate in this case, since hyperprolactinemia is frequently seen in the presence of hypotestosteronemia. As opposed to the clinical manifestations of low T levels (where libido is a poor clinical marker), this author has not yet observed a patient with a functional prolactinoma who did not also present with a marked decrease in sexual desire. Other hormones Further hormonal evaluation depends on the interest, expertise and commitment of the physician and, obviously, the patient's clinical picture. Despite overwhelming evidence to support its existence, androgen decline in the aging male (ADAM) also known as the andropause, continues to be debated. Part of the controversy is based on the vagaries and inconsistencies of the syndrome 16 and production of several other hormones are frequently affected during the aging process. For instance, diminution in muscle mass and strength have long been recognized as a sequelae of hypogonadism, 17 more recently, similar alterations were documented in adults with growth hormone de ciency. Disturbances in sleep patterns may equally be attributed to low levels of T or melatonin. These close interactions between endocrine systems and their target organs are poorly understood but our limitations in knowledge should not justify the denial of their existence. The alterations in the hormonal milieu occurring with age are incontrovertible. Basic biochemical endocrine evaluation in a man complaining of ED should include serum T. Initially, a more comprehensive assessment may not be justi ed, but further S113

3 S114 assessment is clearly indicated if hypogonadism is documented. For instance, the measurement of pituitary gonadotropins, follicle stimulating hormone (FSH) and luteinizing hormone (LH) can provide an indication whether primary, secondary or tertiary hypogonadism is the cause. However, it must be remembered that at middle age and beyond, there is a diminution in pituitary function that results in a attening in the circadian production of gonadotropins. 18 Therefore, it is not uncommon to encounter men with erectile problems (who are most commonly in these age groups) and primarily testicular failure resulting from a breakdown of the feedback mechanisms, ie, a low serum T with normal or only minimally elevated gonadotropins. Measurement of other endocrinological parameters (ie DHEA, growth hormone, melatonin, and leptin) is not indicated in the context of erectile dysfunction, but thyroid function may have an impact on sexual performance. Such an occurrence is exceedingly rare and the routine assessment of thyroid function is unlikely to be productive. However, hyprothyroidism may result in hyperprolactinemia, therefore, assessment of thyroid function in hyperprolactenemic patients is indicated. 19 On the other hand, levels of total testosterone may appear abnormally high in the presence of normal bioavailable T in patients with hyperthyroidism. 20 Justi cation for androgen supplementation In the context of ED, exogenous androgens are indicated if hypogonadism is documented with or without alteration in sexual desire. Libido is a poor indicator of androgenic homeostasis unless there is a profound hypotestosteronemia. Low serum T levels may co-exist with a variety of other important conditions which are well recognized causes of erectile failure (atherosclerosis, diabetes and a variety of erectolytic medications 21 ). However, in the presence of hypogonadism and ED it is justi ed to consider a trial of exogenous androgen therapy. The reasons for this view include the simplicity and safety of the treatment and the presence of a documented etiology. The response, in terms of recovery of erectile function and improvement in sexual interest, is not spectacular. Our experience and that of others clearly indicates that only a limited number of men undergo restoration of erections and improvement in libido secondary to androgen supplementation. 9,10,21 Patients not responding to T administration require reevaluation to rule out associated causes. A trial of androgen administration should be not shorter than 3 months. At the end of this period a re-assessment of sexual function and other parameters is indicated. Continuation of T administration would then depend on the clinical response and possible alter-ations in other organ systems resulting from treatment. It is evident that androgen supplementation for conditions such as osteoporosis requires a prolonged course of no less than several years before an objective improvement can be documented. 22 Treatment options Androgens can be administered by several routes, each one offering a different set of advantages and drawbacks. 23 Parenteral Injectable esters of T have been available for the longest period of time and their effects are well recognized. 24 They are inexpensive and safe but their use carries several major drawbacks which include: (a) the need for periodic (every 2 ± 3 weeks) deep intramuscular injection; (b) the administration of injectable preparations results, in the rst 72 hours, in supraphysiological levels of serum T followed by a steady decline over the next 10 ± 14 days; 25 (c) the steady decline in T levels frequently results in a very low nadir immediately before the next injection. This phenomenon translates into wide swings in mood and well being (the roller-coaster effect) which is disconcerting and upsetting to both patients and their partners; (d) parental androgens do not provide the normal circadian patterns of serum testosterone and the intermittent supra-physiological levels they produce may result in the development of breast tenderness and gynecomastia. The most widely used parenteral preparations are the 17-b-hydroxyl esters of testosterone which include the short acting propionate and the longer acting enanthate and cypionate. Currently, the propionate is rarely used because its short half-life requires that it be administered every other day. The enanthate and cypionate, on the other hand, can be administered in doses between 200 mg and 400 mg every 10 ± 21 d to maintain normal average testosterone levels. 26 Higher doses will not maintain T levels in the normal range beyond the 3-week limit. Appropriate treatment of hypogonadism with injectable esters of T has been shown to improve libido, sexual function, energy levels, mood and bone density if the symptoms are caused by an androgen de ciency. 18 Persistent supraphysiologic levels of serum T may result in infertility due to suppression of LH and FSH production. 27 Although concern exists about the psychosexual effects of markedly elevated levels serum T, published evidence indicates that, even in eugonadal men,

4 amounts up to ve times the normal physiological replacement doses of T cypionate have only minimal psychosexual effects. 28,29 result in various degrees of dermatitis, occasionally producing signi cant chemical burns. S115 Oral preparations These demand special consideration because they undergo rapid hepatic metabolism and may, therefore, fail to result in adequate serum levels of androgens. In order to prevent metabolism by the liver, oral agents available in the USA are alkylated preparations. However, these alkylated preparations generally provide erratic androgenic activity and exhibit a potential for signi cant liver toxicity which has included: hepatocellular adenomas, cholestatic jaundice and hemorrhagic liver cysts. 30 Testosterone undecanoate is yet another oral preparation that is not, at present, available in the USA but it is widely used throughout the world. As the only effective oral testosterone ester, when taken with meals it circumvents the rst passage through the liver. It is free of liver toxicity and brings serum T levels within physiological range. It is liposoluble and for this reason it must be taken with meals. None of the oral medications cited above results in a faithful re ection of the circadian level variations. However, careful selection of the timing and amount of the dosing may ameliorate this problem. Transdermal T therapy Transdermal T therapy (TTT) results in serum T levels within the range observed in normal men over the 24 h circadian cycle. TTT is available in both scrotal and non-scrotal patches. The former has lost appeal due to inconveniences such as its ability to remain in place and the need for frequent shaving of the scrotal skin. In addition, due to the high concentrations of 5-alpha reductase in the scrotal skin, abnormally high serum levels of dihydrotestosterone (DHT) have been reported. 31 Either patch applied at bedtime will result in peak T levels in the early morning and a nadir just prior to their replacement. Transdermal non-scrotal patches also produce normal levels of estradiol, but, unlike the scrotal patches, these do not result in abnormal levels of DHT. 31 In addition to producing physiologically appropriate serum levels of T, they actually lower the level of sex hormone binding globulin (SHBG), promote virilization and increase bone mineral density. 32 Finally, the T patches, compared to injectable forms of T, minimize excessive eryth-ropoiesis and suppression of gonadotropins. 33 The most common side effects of the non-scrotal transdermal patches are related to the enhancers used to facilitate absorption; these enhancers frequently Other preparations Delivery forms not widely prescribed but worth mentioning include long acting implantable subcutaneous pellets and microcapsules. 34,35 The subcutaneous implants are not associated with high DHT levels, which is desirable when treating older hypogonadal man in whom concerns regarding the prostate gland are of paramount importance. The signi cance of elevated levels of DHT, which are more common with oral or scrotal administration of T still remains to be clari ed. Similarly, MENT (7a-methyl-19 nortestosterone), which is not reduced to DHT but aromatized to estradiol, would be more desirable in the presence of urinary obstructive symptoms. New frontiers in androgen therapy It is recognized that androgen receptors (AR) are widely distributed among reproductive and nonreproductive tissues in the body. Of interest in the treatment of sexual dysfunction is their ubiquitous presence in the pineal gland and brain cortical and subcortical regions and male external genitalia. This wide distribution permits envisioning the hypothetical exploitation of selective androgen receptor modulators (SARMS) 36 with a speci c activity in the target tissue of interest. Since SARMs are not substrates for 5a-reductase activity, 35 it would be possible to achieve adequate levels of androgens in the brain and genitalia without a concomitant increase in the prostate or seminal vesicles. The development of such molecules is currently becoming a reality. 37 Responsibility with treatment Male hypogonadism, as mentioned before, affects a variety of organs and the limited view of its exclusive importance in sexual function is too con ning and may be detrimental to patient care. However, for the purposes of this review, it can be categorically established that low T levels in a man with sexual dysfunction are a clear indication for hormone replacement therapy (HRT). If there are no correctable causes attributable to the endocrine dysfunction, a limited period of hormonal treatment (ie, 3 ± 6 months) is usually suf cient to establish a cause ± symptom relationship. Obviously, if adequate HRT does not result in an appropriate response, a search for other co-morbidities becomes mandatory.

5 S116 Other sequelae of hypogonadism may coexist with ED; in this situation, prolonged T administration is indicated even if the sexual problems fail to show improvement. There is a persistent and somewhat justi ed concern about undesirable side effects that may result when some of the currently available androgen preparations are prescribed ethically or when abused by athletes and body builders. The potential for adverse effects is of particular concern in the liver, the prostate, lipid pro le and cardiovascular system, sleep patterns and social behavior and emotional state. These considerations have been reviewed recently. 16 Recommendations The current state of knowledge within the narrow con nes of androgen therapy in men with ED permits the development of a set of recommendations (not yet guidelines) which appear appropriate until a pertinent scienti c society or regulatory agency develops guidelines or standards with universal credibility and appeal. They are as follows: (1) The diagnosis of adult hypogonadism based on a history of low sexual desire is notoriously unreliable. The physical examination is also unreliable except in the most profound cases. Biochemical determinations are, therefore, recommended to establish a rm diagnosis. Either a total T and SHBG or bioavailable T are the most dependable tests. (2) Due to the pulsatile nature of T production, it is prudent to con rm an abnormal result. (3) T administration must be supported by a clear indication (clinical judgement and biochemical information). (4) Borderline levels of serum T may be of limited relevance and may not correlate well with sexual desire and performance. Such cases demand careful consideration of the advantages and drawbacks of androgen therapy. Nevertheless, a limited therapeutic trial is usually warranted. (5) In a man with concomitant hypogonadism and ED, a 3 ± 6 month period of hormonal supplementation is indicated. Failure to respond to therapy despite reaching eugonadal serum T levels is an indication to search diligently for co-morbidity. In the absence of a response and if no other sequelae of hypotestosteronemia exist, discontinuation of treatment should be considered. (6) Men with secondary hypogonadism (of hypothalamic-pituitary origin) may experience bene t by treatment with gonadotropins or clomiphene. 38 (7) Although not yet proven, in hypogonadal men, co-administration of T with an erectogenic drug may show synergistic or supra-additive ef cacy. 39 (8) In men over the age of 40 y a digital rectal examination (DRE) and prostatic speci c antigen (PSA) determination are mandatory. Biopsy of the prostate is reserved for those in whom an abnormality is detected. (9) The suspicion or presence of cancer of the prostate or breast are absolute contraindications for androgen therapy. (10) The T preparation, dose and route of administration need to be selected carefully depending on the individual needs and circumstances of the patient. (11) For the rst year after the onset of therapy, patients should be followed quarterly to assess response to therapy (clinical and biochemical) including hemoglobin, DRE and PSA. Patients who remain stable may subsequently be followed annually, at which time other tests of liver function and lipid pro le should be included. (12) During therapy serum T levels may uctuate considerably, particularly with the use of the intramuscular preparations. In the speci c situation of the hypogonadal impotent man, the clinical response is the most reliable guide to dose requirements. Conclusions Hypogonadism is not a common cause of ED. However, its clinical diagnosis in the adult is dif cult and normally requires biochemical investigations. A serum T determination is justi ed in men complaining of ED with or without alterations in sexual desire. The prevalence of ED is particularly common at a point in life when alterations occur in men's hormonal environment. Even though ED may not be a consequence of the changing hormonal mileu, the consequences of hypogonadism in other organ systems are increasingly recognized and timely diagnosis and appropriate treatment may signi cantly improve the patient's quality of life. References 1 Feldman HA et al. Impotence and its medical and psychosocial correlates: Results of the Massachusetts Male Aging Study. J Urol 1994; 151: 54 ± Replacing testosterone in men. Drug Ther Bull 1999; 37: 3 ± 6. 3 Bancroft J, Wu, CFW. Changes in erectile responsiveness during androgen replacement therapy. Arch Sex Behav. 1983; 12: 59.

6 4 Nickel JC et al. Endocrine dysfunction in impotence. Incidence, signi cance and cost-effective screening. J Urol 1984; 130: 40 ± Nieschlag E, Behere HM. Pharmacological and clinical uses of testosterone. In: Nieschlar A, Behere, HM, (eds). Testosterone: action, de ciency, substitution. Springer-Verlag: Berlin, 1998, pp 294 ± Burris AS et al. A long-term prospective study of the physiologic and behavioural effects of hormone replacement in untreated hypogonadal men. J Androl 1992; 13: 297 ± Vermeulen A. Androgens in the aging male. J Clin Endocrinol Metab 1991; 73: 221 ± Morley JE. Clinical diagnosis of age-related testosterone de ciency. Aging Male 2000; 3(Suppl. 1): Bancroft J, Wu FC. Changes in erectile responsiveness during androgen replacement therapy. Arch Sex Behav. 1983; 12: 59 ± Morales A, Johnston B, Heaton JPW, Lundi M. Testosterone supplementation for hypogonadal impotence: assessment of biochemical measurements and therapeutic outcomes. J Urol 1997; 157: 849 ± Ansong KS, Punwaney RB. An assessment of clinical relevance of testosterone level determination in the evaluation of men with low sexual drive. J Urol 1999; 162: 719 ± Christiansen K. Behavioural correlates of testosterone. In: Nieschlan E, Behere HE, (eds). Testosterone: Action, De ciency, Substitution. Springer-Verlag: Berlin, 1998, pp 107 ± Montague DK. Editorial: medical therapies for erectile dysfunction. J Urol 1999; 162: Mauras N et al. Profound hypogonadism has signi cant negative effects on calcium balance in males: a calcium kinetic study. J Bone Miner Res 1999; 14: 577 ± Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999; 84: 3666 ± Morales A, Heaton JWP, Carson CC III. Andropause: a misnomer for a true clinical entity. J Urol 2000; 163: Bhasin S, Tenover JS. Sarcopenia: issues in testosterone replacement of older men. J Clin Endocrinol Metab 1997; 82: 1659 ± Russell-Aulet M, Jaffe CA, Demott-Frieberg R, Barkan AL. In vivo semiquanti cation of hypothalamic growth hormonereleasing hormone (GHRH) output in humans: evidence for relative GHRH de ciency in aging. J Clin Endocrinol Metab 1999; 84: Grubb MR, Checkers D, Malarkey WB. Patients with primary hypothyroidism presenting as prolactinomas. Am J Med 1987; 83: Vermeulen A, Stoica T, Verdonck L. The apparent free testosterone concentration and index of androgenicity. J Clin Endocrinol Metab 1971; 3: 759 ± Morales A, Heaton JPW, Johnston B, Adams M. Oral and topical treatment of erectile dysfunction: present and future. Urol Clin NA 1995; 22: 879 ± Guo C-Y, Jones TH, Eastell R. Treatment of isolated hypogonadotropic hypogonadism. Effect on bone mineral density and turnover. J Clin Endocrinol Metab 1997; 82: 658 ± Morales A, Baines J, Ruijs A, Tremblay RR. Clinical practice guidelines for screening and monitoring male patients receiving testosterone supplementation therapy. Int J Impot Res 1996; 8: Thompson, WO. Uses and abuses of the male sex hormone. JAMA 1946; 132: 185 ± Sokol RZ, Palacios A, Camp eld LA. Comparison of the kinetics of injectable testosterone in eugonadal and hypogonadal men. Fertil Steril 1982; 37: 425 ± Bhasin S. Androgen treatment of hypogonadal men. J Clin Endocrinol Metab 1992; 74: 1221 ± Bhasin S, Bremner WJ. Emerging issues in androgen replacement therapy. J Clin Endocrinol. Metab 1997; 82: 3 ± Yates WR et al. Psychosexual effects of 3 doses of testosterone cycling in normal men. Biol Psychiat 1999; 45: 254 ± Bagatell CJ, Bremner WJ. Drug therapy: androgens in men ± uses and abuses. New Engl J Med 1996; 334: 707 ± Bradwin SW, Swerdloff RS, Santen RJ. Androgens: risks and bene ts. J Clin Endocrinol Metab 1991; 73: 4±7. 31 Arver S, Meikle AW, Dobbs AS, Maze NA. Permeation enhanced testosterone transdermal systems in the treatment of male hypogonadism: long term effects. J Endocrinol 1996; 148: 254 ± De Sanctis V et al. Clinical experience using Androderm testosterone transdermal system in hypogonadal adolescents and young men with beta-thalassemia major. J Ped Endocrinol Metab 1999; 11(Suppl 3): 891 ± Dobbs AS et al. Pharmacokinetics, ef cacy and safety of permeation-enhanced testosterone transdermal system in comparison with bi-weekly injections of testosterone enanthate for the treatment of hypogonadal men. J Clin Endocrinol Metab 1999; 84: 3469 ± Handelsman DJ et al. An analysis of testosterone implants for androgen replacement therapy. Clin Endocrin 1997; 47: 311 ± Hasin S, Swerdlof RS, Steiner B. A biodegradable testosterone microcapsule formulation provides uniform eugonadal levels of testosterone for 10 ± 11 weeks in hypogonadal men. J Clin Endocrinol Metab 1992; 74: 75 ± Negro-Vilar A. Selective androgen receptor modulators (SARMs): a novel approach to androgen therapy for the new millennium. J Clin Endocrin Metab 1999; 84: 3459 ± Moilanen AM et al. A testis-speci c androgen receptor coregulator that belongs to a novel family of nuclear proteins. J Biol Chem 1999; 274: 3700 ± Guay AT, Bansal S, Hodge MB. Possible hypothalamic impotence. Male counterpart to hypothalamic amenorrhea? Urology 1991; 38: 317 ± Heaton JPW, Adams MA, Morales A. A therapeutic taxonomy of treatments for erectile dysfunction: en evolutionary imperative. Int J Impot Res 1997; 9: 115 ± 119. S117

7 S118 Appendix example, polycythemias, and some people develop sleep apnea on testosterone replacement. Open discussion following Dr Morales' presentation Dr Pryor: In evaluating testosterone levels, we need to remember that it's a biorhythm that peaks in the morning and troughs in the afternoon. If you get a borderline or low level in a patient whose appointment is in the afternoon, you should repeat the test in the morning. Dr Morales: Yes. In fact, in older impotent men and in obese men, there is also a attening of the cycle. Testosterone may be normal, but these people have very high levels of sex hormone binding globulin. Always take measurements between 8:00 and 11:00 am. In the afternoon the levels are going to be lower. Dr Pryor: Another point I'd like to make is about the side effects that some are not as familiar with, for Dr Eid: Is there such a thing as a low testosterone level? Is it abnormal or normal? The concept of low testosterone, is that real or is that something we've made up? Dr Morales: I think it's real. Dr Lisa Tanover says that, in America if your testosterone is less than 250, that's de nitely abnormal; no question about it. You have to use clinical judgement. Given three months of testosterone, if the patient doesn't improve, you should look for comorbidity. One thing we have to eliminate is the urologists' parochial view that testosterone is sex. Testosterone has many other functions. I believe it's a central conditioner. I have a patient with a testosterone level of one, and cancer of the prostate. Two months ago I put a penile prosthesis in him and his libido is pretty good. But, because we don't understand many of these things, I don't think that's an argument for not treating these patients.