Testosterone treatment of male hypogonadism

INTRODUCTION — Hypogonadism in a male refers to a decrease in either of the two major functions of the testes: sperm production or testosterone production. These abnormalities can result from disease of the testes (primary hypogonadism) or disease of the pituitary or hypothalamus (secondary hypogonadism). The use of testosterone to treat hypogonadism in adult men, primary or secondary, is reviewed here. The clinical manifestations and diagnosis of male hypogonadism, induction of spermatogenesis in men with secondary hypogonadism, and induction of puberty with testosterone are discussed elsewhere. (See "Clinical features and diagnosis of male hypogonadism" and "Induction of fertility in males with secondary hypogonadism" and "Approach to the patient with delayed puberty", section on 'Testosterone therapy'.)

MECHANISMS OF TESTOSTERONE ACTION — Testosterone has many different biologic effects, at least in part because it can act as three hormones. It can act directly by binding to the androgen receptor. It can also act in tissues that express the enzyme 5-alpha-reductase, via conversion to dihydrotestosterone, which binds more avidly to the androgen receptor than testosterone itself. Finally, it can act as an estrogen following conversion by aromatase to estradiol, which binds to the estrogen receptor.

●Testosterone requires conversion to dihydrotestosterone for its action on the external genitalia (which include the prostate gland) and sexual hair. This mechanism provides the basis for the use of the 5-alpha-reductase inhibitor, finasteride, to treat benign enlargement of the prostate and male pattern baldness. (See "Male pattern hair loss (androgenetic alopecia in males): Management", section on 'Oral finasteride' and "Medical treatment of benign prostatic hyperplasia", section on '5-alpha reductase inhibitors'.)

●Testosterone requires conversion to estradiol for much of its action on bone. This effect is illustrated by the rare condition of aromatase deficiency in men, which results in failure of epiphyseal closure and severe osteoporosis. Treatment with estradiol corrects both. (See "Etiology of osteoporosis in men".)

Testosterone also appears to require conversion to estradiol to stimulate normal sexual function and decrease body fat in men, as shown by an experiment in which men 20 to 50 years old were treated with a gonadotropin-releasing hormone (GnRH) agonist to suppress testosterone and estradiol secretion and then replaced with testosterone, with or without an aromatase inhibitor [1]. Addition of the aromatase inhibitor partially blocked testosterone from increasing libido and erectile function and from decreasing subcutaneous and intraabdominal fat.

Testosterone does not appear to require conversion to estradiol to affect muscle. In the experiment above [1], treatment with the aromatase inhibitor did not block the effect of testosterone on lean muscle mass, thigh muscle area, or leg press strength.

GENERAL PRINCIPLES — The following principles should guide testosterone therapy:

●Testosterone should be administered only to a man who is hypogonadal, as evidenced by clinical symptoms and signs consistent with androgen deficiency and a distinctly subnormal serum testosterone concentration. In comparison, increasing the serum testosterone concentration in a man who has symptoms suggestive of hypogonadism, but whose testosterone concentration is already normal, will not relieve those symptoms. (See "Clinical features and diagnosis of male hypogonadism", section on 'Symptoms'.)

●Symptoms and signs suggestive of androgen deficiency include low libido, decreased morning erections, loss of body hair, low bone mineral density, gynecomastia, and small testes. Symptoms and signs such as fatigue, depression, anemia, reduced muscle strength, and increased fat mass are less specific [2].

●Testosterone can be replaced satisfactorily whether the testosterone deficiency is due to primary or secondary hypogonadism.

●The principal goal of testosterone therapy is to restore the serum testosterone concentration to the normal range. It is not yet known if restoring the normal circadian rhythm of testosterone is important. (See "Male reproductive physiology".)

●The role of testosterone replacement to treat the decline in serum testosterone concentration that occurs with increasing frequency in men above age 60 years, in the absence of identifiable pituitary or hypothalamic disease, is uncertain. (See "Approach to older males with low testosterone".)

●Testosterone therapy is indicated only for testosterone deficiency, not for impaired spermatogenesis. Testosterone therapy impairs spermatogenesis further by suppressing pituitary gonadotropin secretion.

CLINICAL BENEFITS OF TESTOSTERONE THERAPY IN HYPOGONADAL MEN — The desirable effects of testosterone administration include the development or maintenance of secondary sexual characteristics and increases in libido, muscle strength, fat-free mass, and bone density. Undesirable effects related directly to testosterone include acne, prostate disorders (such as benign prostatic hyperplasia [BPH] symptoms), sleep apnea, and erythrocytosis. (See 'Potential adverse effects' below.)

Virilization/sexual function — Normalization of the serum testosterone concentration should lead to normal virilization in men who are not virilized and maintenance of virilization in those who already are. Men who become hypogonadal in adulthood and are still normally virilized, but whose hypogonadism is manifested by a decrease in libido and energy, should note a marked improvement in these symptoms. Failure of improvement when the serum testosterone concentration has been restored to normal suggests another cause of the symptoms.

Muscle strength/fat-free mass — Testosterone replacement also leads to substantial improvements in muscle strength and fat-free mass in hypogonadal men. In one report, for example, the administration of 100 mg of testosterone enanthate once a week for 10 weeks to hypogonadal men increased their strength in the bench press by 22 percent, their squat strength by 45 percent, and fat-free mass by 5 percent [3].

Bone density — Testosterone replacement improves bone density in male hypogonadism, as illustrated in a study of 72 such men receiving testosterone replacement therapy [4]. The increase in bone density averaged 39 percent in the first year of testosterone replacement, and it eventually reached and was maintained in the normal range. The response was greatest in the first year in previously untreated patients and was most pronounced in those with lowest bone density measurements at baseline. (See "Treatment of osteoporosis in men", section on 'Congenital hypogonadism'.)

Other possible effects — Testosterone has also been postulated to improve mood and cognition, but these effects have not been demonstrated convincingly:

●Mood – Data on the impact of testosterone therapy on mood are inconsistent. In men with moderately severe hypogonadism, treatment with long-acting testosterone esters for 30 weeks significantly improved several measures of mood compared with pretreatment, but the study was neither controlled nor blinded [5]. In young and old men made hypogonadal by a gonadotropin-releasing hormone (GnRH) analog, graded doses of testosterone enanthate did not improve two measures of mood [6,7]. In contrast, two small trials suggest a beneficial effect of testosterone on mood in older men with subclinical depression. (See "Approach to older males with low testosterone", section on 'Testosterone therapy'.)

●Cognition – In older men with slightly low testosterone concentrations, administration of testosterone enanthate every two weeks did not significantly change any of several measures of cognition [8]. Other data in older men have not shown an effect of testosterone therapy on cognition. (See "Approach to older males with low testosterone", section on 'Testosterone therapy'.)

Time course of effects — The time course of the effects of testosterone replacement is variable. This was illustrated in a three-year study of physiologic transdermal testosterone replacement in 18 previously untreated hypogonadal men [9]. Increases in fat-free mass, prostate volume, erythropoiesis, energy, and sexual function occurred within the first three to six months. In contrast, the full effect on bone mineral density did not occur until 24 months.

GENERAL APPROACH TO TREATMENT

Appropriate candidates

Hypogonadal adult men — Testosterone should be administered only to an adult male who is hypogonadal, as evidenced by clinical symptoms and signs consistent with androgen deficiency and a subnormal morning (8 to 10 AM) serum testosterone concentration on three separate occasions. Ideally, samples should be drawn fasting; in one report, an oral glucose load suppressed serum testosterone concentrations acutely [10]. Symptoms and signs suggestive of androgen deficiency include low libido, decreased morning erections, loss of body hair, low bone mineral density, gynecomastia, and small testes. Symptoms and signs such as fatigue, depression, anemia, reduced muscle strength, and increased fat mass are less specific. The beneficial effects of testosterone in these men are clear, and they should be treated with testosterone to raise their serum testosterone concentrations to normal.

Older men with low testosterone and symptoms of hypogonadism — The role of testosterone replacement to treat the decline in serum testosterone concentration that occurs with aging in men, in the absence of identifiable pituitary or hypothalamic disease, has been unclear. However, results from the Testosterone Trials suggest that testosterone has a beneficial effect on sexual function, mood, possibly walking, bone density, and anemia [11]. Results of these trials are reviewed separately. (See "Approach to older males with low testosterone".)

Inappropriate use of testosterone in healthy, middle-aged men — There has been a dramatic increase in inappropriate use of testosterone therapy in healthy, middle-aged and older men. It is important for clinicians to understand that the diagnosis of testosterone deficiency should be made only on the basis of clinical symptoms and signs consistent with androgen deficiency and consistently subnormal serum testosterone concentrations at 8 to 10 AM on three occasions.

For men with vague symptoms that could be the result of low testosterone (eg, fatigue) and a single, but not repeatedly, subnormal serum testosterone concentration, we recommend strongly against testosterone use. The Endocrine Society's guidelines and others also strongly advocate this approach [2,12]. Treatment should not be prescribed on a "trial" basis, because the results are not interpretable and because once men have started treatment, they may find cessation difficult because of the prolonged period of hypogonadism during recovery of the pituitary-testicular axis. In addition, testosterone therapy eventually results in suppression of spermatogenesis and decreased testicular size. (See 'Suppression of spermatogenesis' below.)

The rise in testosterone prescriptions in healthy, middle-aged men is likely due, at least in part, to direct-to-consumer advertising (DTCA) encouraging use of testosterone products for nonspecific symptoms, such as decreased energy and sexual interest [13-16]. In one study of 75 designated market areas with high rates of DTCA of specific testosterone products and/or testosterone deficiency ("low T"), approximately 1 million of 17 million men (6 percent) had themselves tested for low testosterone for the first time between 2009 and 2013 [17]. Approximately 280,000 men (1.6 percent) started testosterone therapy during that interval. The study authors calculated that each exposure to DTCA was associated with a 0.6, 0.7, or 0.8 percent increase in testosterone testing, initiation of therapy, or initiation of testosterone therapy without baseline testing, respectively. Although the percentage changes are small, in large populations, the impact is large.

Choice of testosterone regimen — Choosing among the different testosterone preparations requires an understanding of their pharmacokinetics. Native testosterone is absorbed well from the intestine, but it is metabolized so rapidly by the liver that it is difficult to maintain a normal serum testosterone concentration in a hypogonadal man with oral testosterone. The solutions to this problem that have been developed over many years involve modifying the testosterone molecule, changing the method of testosterone delivery, or both.

A number of testosterone preparations are currently available or are under development for treating testosterone deficiency (table 1).

We usually suggest testosterone gels because they typically result in normal and relatively stable serum testosterone concentrations, and most patients prefer them to other preparations. However, other factors affect choice of regimen, including patient preference, cost, convenience, and insurance coverage, which varies by plan and regimen. In general, the newest preparations (the gels) cost the most and injectable esters cost the least.

Topical/transdermal delivery

●Three testosterone gels (AndroGel, Testim, and Fortesta) and one solution (Axiron) are available. AndroGel is supplied in both 1% and 1.62% concentrations. The 1% concentration was the first to become available. It is still available in 2.5 and 5 g packets, which contain 25 and 50 mg of testosterone, respectively [18,19], and a metered-dose pump that delivers 1.25 g of gel (containing 12.5 mg of testosterone) per pump depression (although the metered-dose pump bottle is no longer available in the United States). When this preparation is applied to the skin once a day in doses of 5 to 10 g (delivering 50 to 100 mg of testosterone), the serum testosterone concentrations usually reach the normal male range within a month and remain steady throughout 24 hours (figure 1). The serum concentrations of testosterone throughout the 24 hours from one application to the next are similar at one, three, and six months [20]. Occasional local skin irritation occurs but usually does not necessitate discontinuation of therapy (table 1).

The 1.62% preparation of AndroGel was introduced later and is available in pump bottles. Each depression of the pump bottle yields 1.25 g of gel containing 20.25 mg of testosterone. A study comparing the bioavailability of this preparation at different application sites demonstrated 30 percent lower availability when it was applied to the abdomen than to the arms and shoulders [21]. In another study, administration of 1.25 to 5.0 g a day to men with slightly low serum testosterone concentrations increased the values to within the normal range [22]. In addition to the pump dispenser, AndroGel 1.62% is also available in 20.25 and 40.5 mg packets.

Testim (1% testosterone gel) is supplied in tubes containing doses of 5 and 10 g, which contain 50 and 100 mg of testosterone, respectively, and when applied daily usually results in normal serum concentrations of testosterone [23]. Anecdotal reports suggest that this preparation gives a musk-like odor (table 1).

Fortesta (2% testosterone gel) is also supplied in a metered-dose pump, with each pump depression delivering 0.5 g of gel (containing 10 mg of testosterone) [24]. The recommended starting dose is 40 mg, applied to the front and inner thighs, and adjusted to a minimum of 10 mg and a maximum of 70 mg, as determined by the serum testosterone concentration. A study in hypogonadal men showed that after 90 days of daily use, the mean serum testosterone concentration before gel application was in the low end of the normal range, after application reached a peak in the mid-normal range in approximately four hours, and fell to the baseline level 12 hours after application (table 1) [25].

Axiron (2% testosterone solution) is a solution of testosterone that also comes in a metered-dose pump with applicator. Each depression yields 30 mg (1.5 mL) of testosterone. The package insert suggests a starting dose of 30 mg applied to each axilla (total of 60 mg) once a day and adjustment of the dose as low as 30 mg and as high as 120 mg once a day, as judged by the serum testosterone concentration. A study in hypogonadal men treated with 30 to 90 mg of this preparation increased the mean serum testosterone concentration to normal (table 1) [26].

Generic versions of most gels are also now available.

●Transdermal delivery of testosterone first became available in 1994 with the introduction of a scrotal patch (chosen because drugs are absorbed readily across scrotal skin). Since then, body patches and gels have also become available, but the scrotal patch is no longer available in the United States. The major advantage of transdermal administration is maintenance of relatively stable serum testosterone concentrations, resulting in maintenance of relatively stable energy, mood, and libido (table 1).

●One patch, Androderm, is available in some countries. It is no longer available in the United Kingdom. Androderm relies upon chemical means to increase the absorption of testosterone across nongenital skin, and it is meant to be worn on the arm or torso. It delivers approximately 2 or 4 mg of testosterone per 24 hours and results in normal serum testosterone concentrations in the majority of hypogonadal men (figure 1) [27-30]. Anecdotal reports suggest that as many as one-third of men who try this preparation cannot continue it, because of severe skin rash. A matrix transdermal testosterone patch is available in countries outside of the United States [31].

Parenteral testosterone

Intramuscular injections

Long-acting — Testosterone enanthate and testosterone cypionate are esters of testosterone that have been used for many years for the treatment of testosterone deficiency. The rationale for their use is that esterification of a lipophilic fatty acid to the 17-beta hydroxyl group of testosterone (figure 2) makes testosterone even more lipophilic than the native molecule. Intramuscular (IM) injection of testosterone esters results in their storage in and gradual release from the oil-based vehicle in which they are administered, thereby prolonging the presence of testosterone in the blood (table 1) [32,33].

The therapeutic characteristics of testosterone enanthate are relatively well described. In one report, for example, 100 mg of testosterone enanthate was administered once a week for 12 weeks to 12 men with primary hypogonadism [34]. The mean serum testosterone concentration increased to slightly higher than the upper limit of normal one to two days after the injection and gradually decreased to the mid-normal range by the time of the next injection.

When the dose of testosterone was increased to 200 mg in an attempt to prolong the dosing interval to every two weeks, the peak serum testosterone concentration increased further, and the nadir, just before the next injection, decreased to the low-normal range (figure 1). Regimens of 300 mg every three weeks and 400 mg every four weeks increased the peaks and decreased the nadirs further.

The serum concentrations of luteinizing hormone (LH), which were initially above normal in all of the men, decreased gradually after initiation of the 100 mg/week regimen and reached the normal range by six to eight weeks. Serum LH concentrations also decreased to normal with the 200 mg per two-week regimen, barely to normal with the 300 mg per three-week regimen, and remained supranormal with the 400 mg per four-week regimen [34].

These data suggest that testosterone enanthate can be administered once every one to two weeks in most men, every three weeks in a few, but not every four weeks. The doses used in this study were all somewhat excessive; we recommend for most men doses of 50 to 100 mg every week or 100 to 200 mg every two weeks.

Less information is available for testosterone cypionate, but the few studies that have been performed suggest that its characteristics are similar to those of testosterone enanthate. Testosterone enanthate has been in short supply in the United States [35], so the cypionate may be the only long-acting ester available in some settings.

An advantage of testosterone enanthate and cypionate over other testosterone preparations is that they are biologically effective in initiating and maintaining normal virilization in all hypogonadal men. Another advantage to some men is freedom from daily administration. The disadvantages are the need for deep IM administration of an oily solution every one to three weeks and fluctuations in the serum testosterone concentration, which result in fluctuations in energy, mood, and libido in many patients. These fluctuations are more pronounced as the dosing interval is increased.

Extra-long-acting — An IM formulation of another ester of testosterone, testosterone undecanoate, is available in several countries, including the United States (table 1) [36].

The prescribing information from the manufacturer for the preparation available in countries other than the United States recommends a dose of 1000 mg administered as a deep IM injection, with a second dose six weeks after the first and subsequent doses every 10 to 14 weeks. One study demonstrated that 1000 mg every three months maintained the mean serum testosterone concentration within the normal range for 18 months [37].

The dosing for the United States preparation is slightly different: each dose is 750 mg in 3 mL of oil injected only into the buttocks. The initial dose is followed by a second dose four weeks later and by subsequent doses every 10 weeks. The extra-long-acting preparations have been associated with rare cases of pulmonary oil microembolism (POME) and anaphylaxis (1.5 and 0.4 cases per 10,000 injections, respectively) [38]. In the United States, the drug is available only through a restricted program called the AVEED Risk Evaluation and Mitigation Strategy (REMS) Program. All injections must be administered in an office or hospital setting by a trained and registered health care provider and monitored for 30 minutes afterwards for adverse reactions.

Data provided by the manufacturer demonstrate that, after the third injection, the average peak serum testosterone value occurs approximately one week after an injection and is followed by a gradual decline until the next injection. The serum testosterone concentration at approximately five weeks would provide an approximate average for the interdosing period.

Because of the inconvenience of injecting a large volume of oil in a restricted setting and the lack of flexibility in dosing, we suggest this preparation only for patients who do not have access to, or do not care for, other forms of testosterone treatment and who do have access to a registered clinic and provider.

Subcutaneous injection — A formulation of testosterone enanthate for subcutaneous injection by autoinjector once a week is now available [39]. Insufficient information is available about its use in men with marked hypogonadism to recommend its routine use yet. Of note, subcutaneous testosterone has been used for the management of transgender men. (See "Transgender men: Evaluation and management", section on 'Testosterone therapy'.)

Oral preparations

●Methyltestosterone – Several 17-alpha alkylated androgens (eg, methyltestosterone) have been available for oral use for many years, but we suggest not using them. Decades ago, investigators discovered that adding an alkyl group in the 17-alpha position of the testosterone molecule slowed its catabolism by the liver (figure 2). Many endocrinologists who treat male hypogonadism think that these preparations are not fully effective in producing virilization, although no studies have tested these observations. In addition, several reports have described hepatic side effects with these preparations, including cholestatic jaundice, a hepatic cystic disease called peliosis hepatis, and hepatoma [40-44]. For both of these reasons and because better preparations are available, the 17-alpha alkylated androgens should generally not be used to treat testosterone deficiency.

●Oral testosterone undecanoate – Several oral testosterone preparations (testosterone undecanoate) are now available in a number of countries [45,46]. They are lipophilic, absorbed through the intestinal lymphatic system, and, unlike previous oral testosterone preparations, bypasses the first-pass hepatic effect. The product includes a warning about possible increases in blood pressure and cardiovascular events (myocardial infarction, stroke). A major concern is its potential inappropriate use in older men with age-related low testosterone; the risks likely outweigh the benefits in these patients (table 1).

Other

●A buccal tablet (Striant SR) is no longer available in the United States (table 1).

●A subcutaneous testosterone pellet (Testopel) is available. The recommended dosing is two to six 75 mg testosterone pellets every three to six months. The pellets are implanted into the subdermal fat of the buttocks, lower abdominal wall, or thigh with a trocar under sterile conditions using a local anesthetic [47]. Adverse events include pellet extrusion, infection, and fibrosis. Because of the need for surgery to insert them and the limited data on the serum testosterone concentrations during treatment, we do not routinely recommend this preparation (table 1).

●A nasal testosterone gel (Natesto) is now approved in the United States for the treatment of male hypogonadism [48]. The gel is administered into the nostrils via a metered-dose pump applicator. One pump actuation delivers 5.5 mg of testosterone; the recommended dose is 11 mg (two pump actuations, one in each nostril), three times daily (total 33 mg/day). One advantage over other formulations is the minimal risk of gel transfer to a partner or child. On the other hand, some men may find the three times daily regimen inconvenient, and men with allergies or underlying nasal or sinus pathology may have trouble tolerating the formulation as ≥3 percent of subjects in clinical trials experienced rhinorrhea, epistaxis, nasopharyngitis, sinusitis, and nasal scab. Lastly, data in mice show brain levels of testosterone that are twice as high with the nasal gel as with intravenous testosterone. It is not known if this occurs in men treated with the nasal gel. Until further published data are available, we suggest using the available testosterone gels, patch, or injectable esters over this new formulation.

●Human chorionic gonadotropin (hCG), while not an androgen, stimulates the testes to make testosterone and is especially useful in stimulating both testosterone and sperm production. It is given by subcutaneous or IM injections two to three times weekly. (See "Induction of fertility in males with secondary hypogonadism".)

Suppression of spermatogenesis — Administration of testosterone exogenously suppresses LH secretion and thereby suppresses the high concentration of intratesticular testosterone essential for spermatogenesis [49]. Although the effect of replacement doses of testosterone in men with mild degrees of hypogonadism who still have some degree of spermatogenesis has not been studied, hypogonadal men who are being considered for testosterone replacement should be told of the suppressive effect of testosterone on spermatogenesis and asked if they desire fertility. If they do, other treatments should be considered. (See "Treatments for male infertility" and "Induction of fertility in males with secondary hypogonadism".)

Exogenous testosterone for hormonal contraception — Choices for male contraception are currently limited (condoms and vasectomy), but there have been ongoing attempts to develop a hormonal male contraceptive. Supraphysiologic doses of exogenous testosterone suppress spermatogenesis, but earlier and more complete suppression to azoospermia or near-azoospermia can be achieved with replacement doses of testosterone combined with a progestin [50]. In a new multicenter study of 320 normal men receiving a long-acting, combination regimen (testosterone undecanoate [1000 mg] and norethisterone enanthate [200 mg]) administered IM every eight weeks, 96 percent of continuing users suppressed to a sperm concentration ≤1 million/mL within 24 weeks [51]. Contraceptive efficacy in the men who suppressed sperm concentration ≤1 million/mL over 56 weeks was 98.4 percent (four pregnancies in 266 continuing participants), a rate similar to some female contraceptive methods (see "Contraception: Counseling and selection"). Recovery of normal spermatogenesis after 52 weeks was 95 percent in continuing users. An external safety review committee recommended early termination of this trial because it concluded the risks of continuing outweighed the possible benefits. The adverse events of concern were mood changes, depression, pain at the injection site, and increased libido.

Contraindications to use — Several possible contraindications to testosterone therapy should be evaluated before initiating testosterone treatment:

●Prostate cancer – A man who has a history of prostate cancer should generally not be treated with testosterone. A possible exception is a hypogonadal man who had a radical prostatectomy for cancer confined to the prostate and has been free of disease and has had an undetectable prostate-specific antigen (PSA) for at least two years [2,52].

The use of testosterone replacement therapy in prostate cancer survivors with hypogonadism due to prior androgen deprivation therapy is controversial. This topic is reviewed separately. (See "Overview of approach to prostate cancer survivors".)

Other men should be evaluated for the possibility of previously undiagnosed prostate cancer. A man over 50 years (or over 40 years if he is African American or has a history of prostate cancer in a first-degree relative) should have a digital rectal examination and a serum PSA measurement. If a prostate nodule is detected or the PSA is >4 ng/mL or >3 ng/mL in a man of high risk, the man should be referred for urologic consultation.

●Breast cancer – Testosterone is aromatized to estradiol, so men who have breast cancer should not be treated with testosterone.

●Lower urinary tract symptoms (LUTS) that are severe (table 2). LUTS should be assessed by the International Prostate Symptom Score (IPSS) (table 2) [53]. If warranted by symptoms, the urine flow rate [54] and post-void residual urine in the bladder by ultrasonography [55] should be measured before beginning treatment (see "Clinical manifestations and diagnostic evaluation of benign prostatic hyperplasia"). Moderate or severe LUTS (eg, score >19) due to benign prostatic hyperplasia (BPH) should be treated before beginning testosterone replacement. (See "Medical treatment of benign prostatic hyperplasia" and "Surgical treatment of benign prostatic hyperplasia (BPH)".)

●Erythrocytosis, eg hematocrit >50 percent. Testosterone stimulates erythropoiesis, so the hematocrit should be measured before initiating testosterone treatment, and if it is elevated, the cause should be sought and the condition treated before testosterone treatment is initiated.

●Sleep apnea that is severe and untreated (see "Clinical presentation and diagnosis of obstructive sleep apnea in adults") may be worsened [2]; the clinician should inquire about symptoms, such as excessive daytime sleepiness and apnea witnessed during sleep by a partner. If indicated, polysomnography should be performed (see "Clinical presentation and diagnosis of obstructive sleep apnea in adults"). Patients whose sleep apnea is well treated with continuous positive airway pressure (CPAP) may take testosterone treatment.

●Uncontrolled heart failure [56]. Testosterone has slight sodium retaining properties, so severe heart failure should be treated before beginning testosterone treatment.

Monitoring

Is the testosterone dose therapeutic?

Serum testosterone concentration — Patients who are treated with testosterone should be monitored to determine that normal serum testosterone concentrations are being achieved. Monitoring should be done two to three months after initiation of treatment and after changing a dose. When the dose appears to be stable, monitoring every 6 to 12 months should suffice. They should also be monitored for undesirable effects.

The timing of serum testosterone measurements varies with the preparation that is used.

●Serum testosterone should be measured midway between injections in men who are receiving testosterone enanthate or cypionate, and the value should be mid-normal, eg, 500 to 600 ng/dL (17.3 to 20.8 nmol/L). The dose should be reduced if higher values are obtained.

●The serum testosterone can be measured at any time in men who are using any of the transdermal preparations, with the recognition that the peak values occur six to eight hours after application of the patch.

●Serum testosterone concentrations vary substantially when a gel is used but not in a predictable way. Therefore, we suggest two serum testosterone measurements before making dose adjustments [57]. This was illustrated in a study of 47 hypogonadal men ≥65 years (27 receiving testosterone gel and 20 on placebo gel); two outpatient, two-hour post-application testosterone levels were significantly but not highly correlated (the correlation coefficient was 0.67). The therapeutic goal should be a testosterone value well within the normal range (400 to 700 ng/dL [13.9 to 24.3 nmol/L]). Patients can have the two samples obtained with as many days between as fits their schedules. Ideally, the two samples should be obtained before the outpatient visit, so the results are available at the time of the visit.

Although it is ideal to measure testosterone twice before making dose adjustments, a single measurement represents an improvement over typical practice as 50 percent of men prescribed testosterone have no biochemical monitoring for the first six months [58].

If the patient has primary hypogonadism, normalization of the serum LH concentration should also be used to judge the adequacy of the testosterone dose no matter which testosterone preparation is used.

Bone density — If a bone fracture or bone density in the osteoporotic range was a presenting finding of hypogonadism, bone density should be reassessed every two years until it becomes normal or stabilizes. If it stabilizes and is still in the osteoporotic range, pharmacologic treatment for osteoporosis should be initiated.

Potential adverse effects — Testosterone enanthate and the testosterone patch and gels have few side effects unrelated to the action of testosterone. Some of the actions of testosterone itself, while not side effects, are undesirable.

Prostate cancer — Prostate volumes and serum PSA increase in response to testosterone treatment [59-61]. On average, values increase to those of age-matched eugonadal men [62]. Because prostate cancer is, at least to some degree, testosterone dependent, it seems theoretically likely that the risk of prostate cancer is less in hypogonadal men than in eugonadal men and that the risk increases to normal, but not above, when testosterone is replaced. However, data to support or refute this assumption are limited [63,64]. It seems prudent, nonetheless, to monitor hypogonadal men for prostate cancer after beginning testosterone replacement, just as one would monitor a eugonadal man. (See "Approach to older males with low testosterone".)

We agree with the Endocrine Society guidelines [2] that men over age 50 years (or 40 years if they are at high risk) who begin testosterone treatment should be reevaluated for prostate cancer three months and one year after beginning treatment and thereafter according to the standard of care. A patient should be referred for urologic consultation if a prostate nodule is palpated or the serum PSA concentration (confirmed by a repeat value) rises by more than 1.4 ng/mL in any one-year period or to >4 ng/mL.

Benign prostatic hyperplasia (BPH) — Because BPH is a testosterone-dependent disease [65], there is a theoretical reason for concern that testosterone treatment might increase BPH and worsen urinary outflow obstruction and LUTS. (See "Clinical manifestations and diagnostic evaluation of benign prostatic hyperplasia".)

However, in a 2018 meta-analysis of trials of testosterone therapy in 1771 hypogonadal men, no significant effects of testosterone were seen on LUTS when compared with the placebo/nonintervention group [66]. In the 788 men 65 years and older enrolled in the Testosterone Trials who were treated with testosterone or placebo for one year, there was no difference in change in the IPSS between the two treatment groups [11].

Erythrocytosis — Erythrocytosis is a common adverse effect of testosterone administration, particularly with testosterone ester injections [67,68]. It is of concern because the risk of venous thromboembolic disease is directly related to hematocrit [69]. In a meta-analysis of three placebo-controlled clinical trials that enrolled a total of 1543 participants, erythrocytosis occurred in 16 men in the testosterone arms compared with one man in the placebo arms [66].

The hematocrit should be measured after three to six months after initiating testosterone treatment and then yearly. If it increases above the upper limit of normal, a cause should be sought, and if none is found, the dose of testosterone should be decreased or stopped. The Endocrine Society guidelines suggest stopping therapy if the hematocrit increases to ≥54 percent [2]. The hematocrit should be reevaluated two months after the decrease or discontinuation. If the hematocrit normalizes, a lower dose of testosterone should be continued or restarted. If the hematocrit cannot be kept below the upper limit of normal, even when the serum testosterone concentration is at the low end of the normal range during testosterone treatment, the patient should be evaluated for hypoxia and sleep apnea. If no treatable cause is found, phlebotomy can be considered.

Venous thromboembolism — Standard labeling of testosterone products in the United States has included information about the risk of venous thromboembolism (VTE) as a consequence of erythrocytosis (see 'Erythrocytosis' above). However, after reports of deep venous thromboses and pulmonary emboli unrelated to polycythemia in men taking testosterone [70], the US Food and Drug Administration (FDA) now requires a more general warning in the labeling of all approved testosterone products [71]. In one study, 39 of 40 cases of VTE in men taking testosterone were associated with a previously undiagnosed thrombophilia-hypofibrinolysis, highlighting the importance of a careful personal and family VTE history prior to initiating treatment. Routine screening for thrombophilias in men considering testosterone therapy is not currently suggested.

Cardiovascular risks — The Endocrine Society [72] and the FDA have issued statements alerting clinicians to the potential concerns about testosterone therapy and cardiovascular safety. Based upon available data, the FDA has concluded that, despite conflicting and inconclusive evidence, there is a possibility of increased cardiovascular risk associated with testosterone use. Therefore, they are requiring that manufacturers of approved testosterone products change their labeling to clarify the approved uses of testosterone (see 'Appropriate candidates' above) and to add information about the possibility of increased risks of myocardial infarctions (MIs) and strokes in patients taking testosterone [73].

Testosterone treatment of hypogonadal men has generally not been associated with increased cardiovascular risk, as shown by an observational study [74], a meta-analysis of 51 clinical trials that included seven trials with the outcome of MI (n = 1053 patients, relative risk [RR] 0.91) [67] and a randomized trial published after data were analyzed for the meta-analysis [75]. The meta-analysis, however, was done in men of variable ages and baseline risk for cardiovascular disease; some of the included trials were done in older men, while others were done in considerably younger human immunodeficiency virus (HIV)-positive men. A subgroup analysis in men over age 65 years was reassuring as no increase in cardiovascular events was noted in those taking testosterone compared with placebo [67].

The impact of testosterone on subclinical atherosclerosis in older men was addressed in the Testosterone's Effects on Atherosclerosis Progression in Aging Men (TEAAM) trial, a study of 308 men ≥60 years with low or low-normal serum testosterone concentrations (100 to 400 ng/dL [3.5 to 13.9 nmol/L], free testosterone <50 pg/mL [0.17 nmol/L]) who were randomly assigned to three years of testosterone gel (1%, 7.5 g) or placebo gel [76]. Mean serum testosterone concentration was significantly higher in the treatment group compared with the placebo group (565 and 330 ng/dL [19.6 and 11.4 nmol/L], respectively), but the rates of change in carotid intima-media thickness and coronary artery calcium scores (markers of subclinical atherosclerosis) were not significantly different. While it is reassuring that testosterone therapy did not accelerate progression of subclinical atherosclerosis, a much larger trial looking at clinical cardiovascular events would be needed to establish cardiovascular safety [77-79].

However, other evidence has raised concern for possible excess cardiovascular risk among some groups of men receiving testosterone therapy, as illustrated by the following findings:

●A randomized trial of testosterone therapy in older men (Testosterone in Older Men With Sarcopenia [TOM] trial) (mean age 74 years) with mobility limitation and a high prevalence of chronic disease was stopped early, due to the higher frequency of self-reported, cardiovascular-related adverse events (that were not prespecified outcomes) in men assigned to the testosterone arm compared with the placebo arm [80]. Because the trial was stopped early and because no single adverse event was predominant, it is difficult to draw firm conclusions about the effect of testosterone therapy on cardiovascular risks in this population. In addition, another randomized trial in a similar number of men, also with mild degree of frailty, reported no excess of cardiovascular events [75].

●In the 788 men of average age 72 years enrolled in the Testosterone Trials and treated with testosterone or placebo for one year, the number of major adverse cardiovascular events was similar in the testosterone and placebo arms [11]. However, in a subset of men who participated in the cardiovascular trial of the Testosterone Trials, noncalcified coronary artery plaque volume, as determined by computed tomographic angiography, increased by 40 mm³ (95% CI 23-56) in 73 men treated with testosterone compared with 4 mm³ (95% CI -14 to 22) (p = 0.003) in 65 men treated with placebo [81].

●In a retrospective cohort study of 8709 men ages 60 to 64 years with low serum testosterone concentrations (<300 ng/dL [10.4 nmol/L]) undergoing coronary angiography, men who were subsequently prescribed testosterone had a higher risk of a composite outcome of all-cause mortality, MI, and stroke than men who did not take testosterone (hazard ratio [HR] 1.29, 95% CI 1.04-1.58); the results were similar after adjusting for the presence of coronary artery disease (CAD) [82]. Although the difference was significant, the absolute risk difference after one year of treatment was only 1.3 percent (11.3 versus 10 percent of patients had events in the testosterone and nontestosterone groups, respectively). A high percentage of men in the cohort had comorbidities, including over 80 percent with CAD. Men were prescribed testosterone gels, injections, but most commonly, patches.

●A second retrospective cohort study examined records of 55,593 men who had been prescribed testosterone therapy. They found an increase in the RR of MI in men in the three months after starting testosterone therapy when compared with the rate of MI in the prior year (RR 1.36) [83]. However, the absolute excess risk was very low: an additional 1.25 cases per 1000 patient-years. The increase in MIs was more pronounced in men over age 65 years (RR 2.19, absolute increase 6.25 versus 0.5 additional cases per 1000 patient-years for those under age 65 years). In addition, men under age 65 years with known heart disease were at increased risk for MI (RR 2.07, absolute excess risk 10 additional cases per 1000 patient-years). Pretreatment serum testosterone concentration was not one of the factors for inclusion in this study, so it is not known if the men were hypogonadal.

Both retrospective studies [82,83] are limited by the inability to verify diagnoses, indications for testosterone therapy, testosterone dose, and information on serum testosterone concentrations immediately before initiating therapy and during therapy. The lack of documentation of antecedent hypogonadism is particularly important because of the widespread and often inappropriate prescription of testosterone therapy for men with normal age-associated symptoms who do not have hypogonadism. (See 'Inappropriate use of testosterone in healthy, middle-aged men' above.)

The route of testosterone administration may be associated with risk. In a retrospective analysis of three databases that included over 544,000 men receiving newly prescribed testosterone therapy, use of injectable testosterone preparations (which result in intermittent supraphysiologic serum testosterone concentrations) was associated with a greater risk of MI (HR 1.30, 95% CI 1.18-1.45) and stroke (HR 1.21, 95% CI 1.10-1.32), but not VTE (HR 0.92), when compared with testosterone gel use [84]. Of note, this study did not assess the safety of testosterone therapy in users versus nonusers, nor did it provide information on indications for testosterone therapy. The absolute increased risks of MI and stroke were low (1 and 5 additional events/1000 patient-years for men in their 50s or over age 65 years, respectively). These data do not change our management approach. We typically suggest testosterone gels, but some patients prefer injectables because of lower cost and convenience. (See 'Choice of testosterone regimen' above.)

We continue to recommend that testosterone be administered only to men who are hypogonadal, as evidenced by clinical symptoms and signs consistent with androgen deficiency and a subnormal morning serum testosterone concentration on three occasions, as outlined above. The beneficial effects of testosterone in these men are clear, and there is no reason to suspect that increasing the serum testosterone to normal would increase the risk of any condition above normal. (See 'Appropriate candidates' above.)

Skin irritation — The patch Androderm often causes skin rashes, some very mild and others quite severe, requiring discontinuation of this treatment. The rash may sometimes be prevented by pretreatment of the skin with a corticosteroid cream [85]. This patch has been taken off the market in some countries.

Local skin irritation occasionally occurs with testosterone gels (AndroGel and Testim) but usually is not severe and does not necessitate discontinuation of therapy.

Secondary exposure: Minimal risk with precautions — With testosterone gels, the possibility of skin transfer to another person is very low if the patient follows the package insert directions that include:

●Washing hands thoroughly after application

●Avoiding skin contact until the gel has dried completely

●Avoiding getting the site wet for approximately five hours after application

In a study of healthy male volunteers receiving testosterone gel, intense skin contact with a second male volunteer (pretreated with norethisterone [400 mg IM] to suppress endogenous testosterone) did not result in an increase in serum testosterone concentrations [86].

However, the FDA has mandated a boxed warning for all testosterone gel products after receiving reports of secondary exposure in eight children ages nine months to five years [87,88]. Exposure resulted in penile or clitoral enlargement, premature development of pubic hair, advanced bone age, increased libido, and aggressive behavior, most of which regressed when the child was no longer exposed to testosterone (with the exception of penile/clitoral enlargement).

These complications are extremely rare (8 cases per 1.8 million prescriptions), and therefore, testosterone gel use should still be considered safe [88]. However, it is extremely important to take precautions to avoid contact transfer of testosterone, eg, by washing hands after application and keeping the application site covered with clothing.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Testosterone therapy in men" and "Society guideline links: Male infertility or hypogonadism".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Androgen replacement in men (The Basics)" and "Patient education: Low testosterone in men (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Candidates for testosterone: Hypogonadal men

•Hypogonadal men – For hypogonadal men (with signs and symptoms of testosterone deficiency such as low libido, decreased morning erections, loss of body hair, low bone mineral density, gynecomastia, and small testes) and a subnormal morning (8 to 10 AM) serum testosterone concentration on two to three separate occasions, we suggest testosterone replacement therapy (Grade 2B). (See 'Appropriate candidates' above.)

The main goal of testosterone therapy is to restore the serum testosterone concentration to the normal (eugonadal) range. (See 'General principles' above.)

•Inappropriate use of testosterone therapy – There has been a dramatic increase in inappropriate use of testosterone therapy in healthy, middle-aged and older men. Testosterone should not be prescribed for men with vague symptoms (such as fatigue) (see 'Inappropriate use of testosterone in healthy, middle-aged men' above) and a single low serum testosterone concentration (rather than two to three as recommended).

●Testosterone preparations – A number of testosterone preparations are currently available for treating testosterone deficiency (table 1).

We usually start with testosterone gels because they typically result in normal and relatively stable serum testosterone concentrations, and most patients prefer them to other preparations. However, some patients prefer parenteral formulations. Other factors that affect choice of regimen include patient preference, cost, convenience, and insurance coverage, which is variable. (See 'Choice of testosterone regimen' above.)

Dosing for different formulations is described above (table 1). (See 'Choice of testosterone regimen' above.)

●Potential adverse effects

•Erythrocytosis – Testosterone therapy is associated with erythrocytosis, more often with parenteral preparations. A hematocrit should be measured three to six months after initiating therapy, and if normal, it can be measured yearly. Clinical guidelines suggest stopping therapy if the hematocrit increases to ≥54 percent; the hematocrit should be reevaluated two months after the decrease or discontinuation. If the hematocrit normalizes, a lower dose of testosterone can be continued or restarted. (See 'Erythrocytosis' above.)

•Other – The impact of testosterone therapy on prostate cancer risk and symptoms of benign prostatic hyperplasia are somewhat unclear. In addition, cardiovascular risks are uncertain. Theoretically, restoring normal levels of testosterone in a formerly hypogonadal male should not increase these risks above those observed in eugonadal men. (See 'Potential adverse effects' above.)

●Monitoring treatment – Patients who are treated with testosterone should be monitored for any adverse effects and to determine that normal serum testosterone concentrations are being achieved. Monitoring should be done two to three months after initiation of treatment and after changing a dose. When the dose appears to be stable, monitoring can be done every 6 to 12 months. (See 'Is the testosterone dose therapeutic?' above.)

Serum testosterone should be measured midway between injections in men who are receiving parenteral testosterone (testosterone enanthate or cypionate), and the value should be mid-normal, eg, 500 to 600 ng/dL (17.3 to 20.8 nmol/L). The dose should be reduced if higher values are obtained.

Serum testosterone concentrations vary substantially when a gel is used. Ideally, two samples for testosterone should be measured after a dose before making any adjustments. However, in practice, many clinicians are only able to obtain one sample.