As one of the oldest marketed drugs in medicine, testosterone has a long record of safe and effective use in androgen replacement therapy. It was first characterized in the mid-1930s, earning the fathers of its discovery the 1939 Nobel Prize in Chemistry. By now, testosterone has been in clinical use for nearly seven decades, mainly for the treatment of androgen deficiency in men. This lack may be caused by testicular (primary) or pituitary/ hypothalamic (secondary) lesions and affects about 20% of men aged 50 and older. Only a fraction (5-10%) of hypogonadal men, however, are diagnosed and treated accordingly.

Despite the extensive history of testosterone prescription, this treatment option continues to be challenging. To assist in recognizing hypogonadism and choosing an appropriate approach, several international bodies have established guidelines that are based on systematic reviews of available evidence (see references for examples).

How is hypogonadism diagnosed?
The 2006 Endocrine Society (ES) guidelines recommend making a diagnosis of testosterone deficiency only in men with consistent symptoms and signs and unequivocally low serum testosterone levels, and suggest that a morning serum total testosterone be the initial test ordered. This advice is in contrast to the guidelines of the International Society for the Study of the Aging Male (ISSAM), which advocate measurement of bio-available or free testosterone, and are more in keeping with what most physicians feel are far superior tests of androgen effect. The ES guidelines do, however, allow for the measurement of bio-available or free testosterone in "some patients."

True free testosterone assays are not commonly available. The analog free testosterone assay, which was widely used, has been criticized for its inadequate representation of true free testosterone and an associated poor performance in clinical assessments. Assays for bioavailable testosterone (a measurement of free and loosely protein-bound testosterone) or calculated free testosterone are more reliable and clinically useful. Still, there's room for improvement: reproducible assays from reputable laboratories should be more widely accessible.

Symptoms prompting measurement of testosterone include decreased libido, reduced muscle bulk and strength, and hot flushes. Table 1A lists some of the signs more specifically related to testosterone deficiency. It remains challenging for the physician to detect androgen deficiency as many of the symptoms are non-specific on their own and gain more significance by the company they keep. These less specific symptoms are itemized in Table 1B, and should prompt you to investigate for testosterone deficiency when they occur in concert with those from Table 1A.

Testosterone levels fluctuate widely, even in healthy men. To avoid confusion and the potential for over-diagnosis of androgen deficiency, it's important not to measure hormones during the transient testosterone suppression that's common when an acute or subacute illness is present. Measure testosterone after the illness has passed, and make sure to draw the blood sample before 10 am to avoid the physiological drop that occurs in the mid-day and evening due to testosterone's normal circadian rhythm. When low testosterone levels are found in a patient, they should be confirmed with repeat testing. The physician may also use this opportunity to measure luteinizing hormone (LH), follicle stimulating hormone (FSH) and prolactin levels.

How do we treat androgen deficient men?
The guidelines recommend testosterone therapy for symptomatic men with classic androgen deficiency syndrome to induce and maintain the patient's secondary sex characteristics, sexual function, sense of well-being and bone mineral density (BMD). The formulations of choice include testosterone or testosterone esters. Serum testosterone levels should be restored to a range that is mid-normal for healthy, young men. There's no longer a role for the older alkylated forms of testosterone such as methyl-testosterone in the treatment of hypogonadal men; these formulations have a risk of liver toxicity and are obsolete now that safer intramuscular (IM), oral and transdermal forms are available.

A number of conditions disqualify a patient for testosterone therapy. Breast or prostate cancer, erythrocytosis, hyperviscosity, untreated obstructive sleep apnea, symptoms of severe benign prostatic hyperplasia (BPH; American Urological Association [AUA] prostate symptom score > 19), or uncontrolled serious heart failure are all examples. A palpable prostate nodule or induration, or prostate-specific antigen (PSA) levels > 3 µg/L also preclude testosterone therapy until further urologic evaluation.

See Table 2 for the potential adverse effects of testosterone replacement.

How should we start testosterone therapy?
No one formulation of testosterone is ideal for all men. We are fortunate in Canada to have three approved modes of delivery including IM, oral and transdermal. Some dose titration is often required to compensate for patient-to-patient variability of absorption and clearance.

Here are some recommended regimens for testosterone replacement therapy:

• 75-100 mg of testosterone enanthate or cypionate administered weekly, or 150-200 mg given every other week
• one or two 5-mg nongenital testosterone patches applied nightly on the skin of the back, thigh or upper arm -- away from pressure areas
• 5-10 g of testosterone gel spread daily
• 80 mg of testosterone undecanoate orally twice a day with meals.

How do we monitor treatment?
Men receiving testosterone therapy should be followed according to a standardized monitoring plan to make sure that any potential side effects are detected early:

• Evaluate the patient 3 months after beginning treatment, and follow up with annual exams to assess whether symptoms have improved and/or the patient is suffering from any adverse events.
• Monitor testosterone levels 2-3 months after initiating therapy; the goal is to raise serum testosterone levels into the mid-normal range (see Table 3 for details).
• Check hematocrit at baseline, at 3 months, and then annually. If hematocrit is > 52%, stop therapy until it decreases to a safe level and re-evaluate the patient for hypoxia and sleep apnea. If erythrocytosis develops during IM therapy, three options exist: the dose could be reduced, the dose interval can be contracted with a lower dose per injection but a similar overall total dose per month, or change formulation to a non-IM formulation. Erythrocytosis is much less common with transdermal formulations, and rare with oral testosterone undecanoate.
• Measure BMD of lumbar spine and/or femoral neck after 1-2 years of testosterone therapy in hypogonadal men with osteoporosis or low trauma fracture, consistent with regional standard of care.
• Perform a digital rectal examination (DRE) and check the PSA level before initiating treatment, at 3 months, and then in accordance with guidelines for prostate cancer screening depending on the age and race of the patient.
• Request a urologic consultation in the following cases:
  • verified serum PSA concentration > 4.0 µg/L
  • increase in serum PSA concentration > 1.4 µg/L within any 12-month period of testosterone treatment
  • PSA velocity > 0.4 µg/L/year, using the PSA level after 6 months of testosterone administration as a reference (only applicable if PSA data are available for >/- 2 years)
  • prostatic abnormality is detected on DRE
  • AUA or International Prostate Symptom Score (IPSS) > 19
• Evaluate formulation-specific adverse effects at each visit:
  • injectable testosterone esters (enanthate and cypionate) -- ask about fluctuations in mood or libido
  • testosterone patches -- look for skin reaction at the application site
  • testosterone gels -- advise patients to cover the application sites with a shirt and to wash the skin with soap and water before having skin-to-skin contact because this form leaves a residue on the skin that can be transferred to a woman or child who comes in close contact; serum testosterone levels are maintained when the site is washed 4-6 h after applying the gel.

Should testosterone be offered to aging men?
It's well established that the total and free testosterone levels decline modestly as men age. Both hypogonadism and naturally waning testosterone levels commonly cause symptoms in the following areas: sexual function, BMD, muscle and fat mass, muscle strength and cognitive function. You'll have to decide on a case-by-case basis whether low androgen levels are contributing to the condition, and if testosterone therapy could help to prevent or even reverse symptoms.

Because hypogonadism is so frequent in aging men and potential adverse side effects of testosterone treatment are a valid concern, some authors have questioned whether doctors should withhold androgen therapy in older male patients pending release of large, randomized controlled long-term studies, which would clarify the risks and benefits. A committee of the U.S. Institute of Medicine (IOM) of the National Academy of Sciences has in fact concluded that there's insufficient evidence for a well-established benefit of testosterone treatment in elderly men. While waiting for more robust studies in older men is a laudable ideal, and is well founded in the tenets of evidence based medicine, this stance does little to guide the present-day physician in how to deal with hypogonadal aging male patients.

It's also worthwhile to consider that young men with hypogonadism due to Klinefelter's Syndrome or bilateral anorchidism are universally offered androgen replacement without any long-term studies clarifying the effects on cardiovascular outcomes, bone fracture rates, prostate disease development or longevity. We've been content to treat these younger men based on short-term studies and immediate clinical response to therapy. Consequently, depriving older men of therapy options offered to the young, when the available evidence of treatment benefits and risks for both age groups is comparable, is prejudicial medicine based on age discrimination. Seniors deserve a well-thought-out assessment and frank discussion of the potential risks and benefits of testosterone therapy, and treatment may be initiated if desired.

For men with angina pectoris, it's been found that testosterone has a positive effect on endothelial dysfunction and that it may even act as an antianginal agent. In treadmill exercise testing, it significantly increased the time to 1 mm ST-segment depression compared to placebo (Malkin CJ et al. Heart 2004;90:871-6). Though the initial data are promising, further studies are needed before testosterone therapy can be potentially recommended for patients with angina pectoris.

Does testosterone replacement boost BMD?
Hypogonadism is a well-established cause of osteoporosis and fractures in men (Shahinian VB et al. N Engl J Med 2005;352:154-64), and several trials have shown testosterone replacement to increase BMD in such cases. Treatment can effectively increase bone density in several types of testosterone deficiency, including primary, secondary, congenital, idiopathic hypogonadotropic and hyperprolactinemic hypogonadism. Older men with low serum testosterone concentrations and those with glucocorticoid-induced hypogonadism have also profited from testosterone therapy. However, there haven't been any long-term studies with sufficient power to assess whether fracture rates are lower in osteoporotic, hypogonadal men treated with testosterone. Just as in cases of osteoporosis from other causes, patients should be supplemented with 1,000 mg/day of calcium, 800 IU of vitamin D and a weight-bearing exercise program.

What about androgen abuse in athletes?
Use of androgens by competitive and recreational athletes is quite common today, and physicians need to be alert to this misuse. Testosterone esters and synthetic testosterones are both used; the latter are often called "anabolic steroids" as they were developed to have a greater anabolic to androgenic effect compared to testosterone. This differential effect hasn't been proven in humans and is unlikely to exist as we have only one androgen receptor.

Red flags for androgen abuse are small testes, low sperm counts, high hematocrit and hemoglobin values, low serum sex hormone-binding globulin (SHBG), and immeasurable LH and FSH concentrations in male patients. Alkylated testosterone compounds may also elevate liver function tests, and recent use of injectable testosterone can hike creatine phosphokinase (CPK) levels. In women, androgen abuse might be suspected in an athlete with hirsutism, breast atrophy, balding or acne.