Testosterone Testing

The Endocrine Society defines male hypogonadism as “a clinical syndrome that results from failure of the testis to produce physiological levels of testosterone (androgen deficiency) and a normal number of spermatozoa due to disruption of one or more levels of the hypothalamic-pituitary-testicular (HPT) axis.”¹

Many men have total testosterone levels below the lower limit of the reference range with no clinical symptoms. Other men have symptoms with total testosterone levels above the lower limit of the reference range. Men may develop different symptoms at different ages, and the sequence of symptom presentation may also vary. Thus, there is no single clinical presentation that fits all men.

Various symptoms of hypogonadism are also observed in other medical conditions. For example, fatigue may be due to diabetes, iron deficiency, depression, or hyperthyroidism. Bone loss may be secondary to vitamin D deficiency. These alternative medical conditions may be underdiagnosed and undertreated.

With aging, the responsiveness to pituitary hormones declines. The number of Leydig cells in the testes decline in number and sensitization. Luteinizing hormone levels become more erratic and affect the availability of testosterone. Growth hormone and dehydroepiandrosterone (DHEA) levels decline with aging, contributing to declines in muscle mass, strength, and overall well-being.

Total testosterone levels decrease by approximately one-third between ages 20-30 years and age 75 years.² The onset of decline is insidious, and the decline is gradual. In contrast, free testosterone declines by approximately 60% in the same time span.² Other hormone levels drop as well. The interaction among these hormones may affect their actions and the development of symptoms.

Hypogonadism is diagnosed based on clinical symptoms and testosterone measurements. One common approach to evaluating symptoms is use of a questionnaire such as the Androgen Deficiency in Aging Men (ADAM) questionnaire.^3,4 This questionnaire has a reported sensitivity of 88% and specificity of 60%.⁴ Thus, eight of nine men with hypogonadism will be identified with ADAM. However, four of ten men will be falsely identified as having hypogonadism. If the hypogonadism prevalence is 10%, then 20% of those positively identified by ADAM will really have hypogonadism (true positives). On the other hand, 98% of those who are negative by ADAM will not have hypogonadism (true negatives).

The components of the ADAM questionnaire include:

• Changes in mood (fatigue, depression, anger)
• Decreased body hair (feminization)
• Decreased bone mineral density and possible resulting osteoporosis
• Decreased lean body mass and muscle strength
• Decreased libido and erectile quality
• Increased visceral fat
• Oligospermia or azoospermia

There is no consensus on the degree of these signs or symptoms required for diagnosis.

An alternative questionnaire is the Massachusetts Male Aging Survey (MMAS) questionnaire.⁵ This survey has far better sensitivity than specificity.

Scores derived from these questionnaires do not predict or correlate well with measured total testosterone.⁶ Specimens for testosterone measurement should be collected between 7 and 10 a.m., because levels show a circadian rhythm; peak levels occur in the morning, especially among younger men. Also, because levels can fluctuate day-to-day, repeat testing is recommended by the Endocrine Society prior to the initiation of treatment.¹

Free and bioavailable testosterone measurements may be helpful when the total testosterone concentration is near the decision level or when perturbations in sex hormone binding globulin (SHBG) are likely. Longitudinal studies such as the Massachusetts Male Aging Study suggest that total testosterone decreases at a rate of about

1.6% annually, with a concomitant 1.3% annual increase in SHBG after age 40 years.⁷ An estimated 30% of men aged 70–79 years have low serum total testosterone, and approximately 70% have low bioavailable testosterone levels.² Free and bioavailable testosterone can be measured or calculated based on the total testosterone, SHBG, and albumin concentrations.

Luteinizing hormone (LH) testing may be useful in determining if a patient’s hypogonadism is primary (elevated LH) or secondary (LH within range or low). Prolactin testing is used to rule out hyperprolactinemia.

There are 3 types of male hypogonadism: primary, secondary, or mixed. All 3 are characterized by deficiency of both testosterone and spermatozoa. In primary hypogonadism, testicular dysfunction leads to low levels of testosterone and high levels of LH and follicle stimulating hormone (FSH).¹ In secondary hypogonadism, dysfunction of the hypothalamic-pituitary axis results in low levels of testosterone, LH, and FSH^.1 Mixed hypogonadism manifests a mixture of primary and secondary hypogonadism.¹ It can be observed in older men.

In contrast, the European Male Aging Study (EMAS) defined testosterone deficiency as primary (low total testosterone with decreased LH), secondary (low total testosterone with elevated LH), or compensated (within range total testosterone with elevated LH).⁸ The prevalence of each among 3369 community-dwelling men ages 40-79 years was:

• Primary hypogonadism:            11.8%
• Secondary hypogonadism:        2.0%
• Compensated hypogonadism:    9.5%

The DETECT Study found low total testosterone (<300 ng/dL) was associated with obesity, metabolic syndrome, and ≥ 6 prescription medications. Very low levels of total testosterone (<100 ng/dL) were associated with advanced age, cancer, and liver disease.⁹

The HIM Study found the following clinical associations, reported as odds radio (with 95% confidence intervals).¹⁰ All were statistically significant.

• Obesity                         2.38 (1.93-2.93)
• Diabetes                       2.09 (1.70-2.58)
• Hypertension                 1.84 (1.53-2.22)
• Dyslipidemia                 1.47 (1.23-1.76)
• Asthma/COPD              1.40 (1.04-1.86)
• Prostatic disease          1.29 (1.03-1.62)

Araujo et al determined the prevalence of hypogonadism based on ≥3 related symptoms plus testosterone concentrations in a study of 1691 men aged 40-70 years. The overall prevalence was 6.0%.¹¹ The prevalence was 4.1%, 4.5%, and 9.4% in men 40-49 years, 50-59 years, and 60-70 years, respectively.¹¹

The Baltimore Longitudinal Study of Aging used a criterion of <325 ng/dL to define low testosterone; presence or absence of symptoms was not determined For men in their 60s, 70s, and 80s, the percent of men with low testosterone was approximately 20%, 30%, and 50%, respectively.²

Testosterone is the most abundant androgen. It is secreted by the testicular Leydig cells. In addition to its hormonal activity, testosterone is a prohormone that can be converted to dihydrotestosterone, a powerful androgen, and estradiol, an estrogen.

Testosterone secretion is dependent upon LH stimulation of the Leydig cells. Increasing levels of testosterone suppress secretion of LH and, conversely, decreasing levels of testosterone act to increase LH secretion. LH secretion from the pituitary is controlled by the hypothalamic gonadotropin-releasing hormone (GnRH).

Testosterone circulates in three major forms: unbound (free) testosterone, tightly-bound testosterone, and weakly-bound testosterone. The tightly-bound form is bound to sex hormone-binding globulin (SHBG), while the weakly-bound form is bound to albumin. Approximately two-thirds is tightly-bound, 30% to 32% is weakly-bound, and the remaining 0.5% to 3% is free. “Bioavailable” testosterone includes both unbound (free) and loosely-bound (to albumin) testosterone. Only bioavailable testosterone is able to bind to the androgen receptor.

Testosterone is metabolized by the 5-alpha-reductase enzyme to dihydrotestosterone, a biologically active androgen. In men, approximately 70% of dihydrotestosterone is derived from testosterone; in women, the primary prohormone for dihydrotestosterone is androstenedione. Androstenedione metabolism accounts for the majority of testosterone in women, but the ovaries and adrenal secrete small amounts.

Testosterone is important for maintaining muscle mass and strength, bone mass, fat distribution, sex drive, and sperm production in men. Low testosterone associations include low energy, reduced strength, decreased cognitive function, lower libido, increased breast size, and depressed mood.

Symptoms of low testosterone (low libido, erectile dysfunction, osteoporosis or fracture, sleep disturbance, depressed mood, lethargy, or diminished physical performance) are nearly universal, especially in older men.

Zitzman et al¹² and others suggest that the specific symptoms associated with hypogonadism develop at different total testosterone concentrations. Loss of vigor and libido may occur at approximately 350-430 ng/dL, whereas depression, disturbed sleep, lack of concentration, and diabetes are associated with total testosterone levels of approximately 230-290 ng/dL. Other hormones, including estrogen, may also influence pattern of symptoms.

The Table below lists the clinical indications for testosterone testing, based on references 1, 13-15. A physician’s test selection and interpretation, diagnosis, and patient management decisions should be based on his/her education, clinical expertise, and assessment of the patient.

There is no universally accepted lower limit. Studies have used different methods, different populations, different times of the day for specimen collection, and different statistical methods. The Quest Diagnostics assays have a reportable lower limit of 250 ng/dL which is consistent with that reported by other laboratories. This is based on the 2.5th percentile of a distribution of results, the approach used to define most reference ranges. In a study by Mohr et al. (2005), the 2.5th percentile for men in their 40s was 251 ng/dL¹⁶ virtually identical to what is commonly reported by laboratories.

The cutpoint below which treatment is recommended, however, is controversial. Symptoms are more likely to appear once the concentration drops below 300 ng/dL; however, testosterone treatment effects may not be evident unless the pretreatment concentration is below 200 ng/dL.¹

Some have suggested “optimal health” is defined as when we are at our peak level of health such as when we are 25 years old. This approach implies that changes observed with aging are potentially preventable or can be rectified through medical interventions. Alternatively, we may have sufficient reserve capacity when we are at our peak health such that we can continue to enjoy good health in subsequent decades. For example, our renal function gradually declines with advancing age. Few of us will develop end stage renal disease despite “suc” deterioration in renal function. Likewise, most men appear to have sufficient capacity to maintain good health even as total testosterone levels decline with age. Although age-based reference ranges for testosterone are available, medical organizations, including the Endocrine Society, continue to promote a single set of criteria for all men, irrespective of age.

Studies differ on this question and may be affected by time of day the specimen was collected, age of subjects, and BMI (adiposity). For example, data from college-age students may not translate to observations among older men who are being evaluated for hypogonadism. If ethnic differences do exist, they are likely of modest clinical significance.

Miller et al found testosterone levels did not differ by ethnic group.¹⁷ Ellis and Nyborg found African-American Vietnam veterans had a 3% higher level of total testosterone than white Vietnam veterans.¹⁸ More recently, the Third National Health and Nutrition Examination Survey (NHANES III) found no difference in total testosterone levels between non-Hispanic black and white men.¹⁹ 

Free and bioavailable testosterone testing is indicated when the total testosterone is near the lower limit of the adult male reference range, and diagnosis of hypogonadism is being considered in men and mild

hyperandrogenism is being considered in women. In boys of pubertal age, free and bioavailable testosterone levels may be a better indicator of hypogonadism than total testosterone levels. Low concentrations of the free and bioavailable forms may indicate delayed puberty.

Typically, free and bioavailable testosterone levels parallel the total testosterone level. However, this may not be the case when sex hormone-binding globulin (SHBG) concentration is affected by certain medical conditions or medicines, including:

• Liver disease and severe systemic illnesses
• Inherited abnormalities in SHBG
• Treatment with corticosteroids or sex steroids, eg, oral conjugated estrogen

Women with polycystic ovarian syndrome (PCOS) may have insulin resistance that is associated with low levels of SHBG. In this circumstance, free and bioavailable testosterone may be elevated more than the total testosterone level.

Both free and bioavailable testosterone measurements are reliable. They almost always correlate well with each other (correlation coefficient = 0.96). However, there are times when the two results vary.

The reference method for measuring total testosterone is liquid chromatography–tandem mass spectrometry (LC/MS/MS).¹³ Though immunoassays are widely used, they cannot reliably quantify low concentrations (ie, those below the lower limit of the reference interval for men). They are therefore inappropriate for quantifying testosterone in children, women, and hypogonadal men but may be used to screen men with suspected hypogonadism.

Direct measurement of free and bioavailable testosterone is not available, so concentrations are estimated from calculations. Free testosterone is best calculated based on the LC/MS/MS-derived total testosterone and

equilibrium dialysis-derived percent free levels. Alternatively, free testosterone can be calculated based on total testosterone, SHBG, and albumin measurements. Bioavailable testosterone is calculated similarly. This method, however, does not account for estrogens or other compounds that displace testosterone from SHBG.

Using the LC/MS/MS and equilibrium-based dialysis methods, free testosterone is calculated as follows:

Free testosterone = (total testosterone) (% free)

The concentration of free (and bioavailable) testosterone can also be calculated by use of one of several published equations. The two most widely used equations for calculating free and bioavailable testosterone are those described by Vermeulen et al²⁰ and Sodergard et al.²¹ Both equations are based on the law of mass action. These equations assume that when the concentrations of total testosterone, SHBG, and albumin are known, free and bioavailable testosterone can be calculated using known constants for the binding of testosterone to SHBG and albumin. Accordingly, these equations are accurate as long as competition for SHBG and albumin binding sites is limited. Quest Diagnostics employs a modified Vermeulen equation which takes into account the dimeric SHBG measured by our assay:

Free testosterone = total T – (SHBG bound T) – (albumin bound T)

Bioavailable testosterone = total testosterone – (SHBG bound T)

Where…

T = testosterone

SHBG = sex hormone binding globulin

Total Testosterone

The Endocrine Society recommendations for monitoring TRT include measurement of total testosterone concentrations three to six months after initiation of therapy.¹ Precise timing and frequency vary based on the method of administration (eg, injection, patch, pellets, etc.). The intent is to ensure levels are in the mid-normal adult male range.

Hematocrit

The hematocrit should be measured at baseline (initiation of therapy), three to six months later, and annually thereafter.¹

Bone mineral density

Bone mineral density is suggested for the lumbar spine and/or femoral neck after one to two years of TRT in hypogonadal men with osteoporosis or low trauma fracture.¹

Prostate

Men who are ≥40 years of age and have a baseline prostate specific antigen (PSA) >0.6 ng/mL should have a PSA test and a digital rectal examination (DRE) before initiation of treatment, three-to six months later, and then in accordance with screening guidelines.¹ The Endocrine Society guidelines provide additional guidance regarding TRT, PSA, and prostate disease.

Contraindications

TRT is not recommended for men “with breast or prostate cancer, a palpable prostate nodule or induration or prostate-specific antigen greater than 4 ng/mL or greater than 3 ng/mL in men at high risk for prostate cancer such as African Americans or men with first-degree relatives with prostate cancer without further urological evaluation, hematocrit >50%, untreated severe obstructive sleep apnea, severe lower urinary tract symptoms with International Prostate Symptom Score (IPSS) >19, or uncontrolled or poorly controlled heart failure.”¹

So far the data are suggestive, but not definitive. The Endocrine Society has recommended that until evidence from large clinical trials is available, “patients should be made aware of the potential risk of cardiovascular events in middle-aged and older men.”²² In March of 2015, the U.S. Food and Drug Administration (FDA) required manufactures to add a warning to their product labeling. The new information warns the user about “a possible increased risk of heart attacks and strokes in patients taking testosterone.”²³ Further, the FDA is requiring manufacturers to conduct a clinical trial specifically evaluating the effect of TRT on cardiovascular events (eg, heart attack and stroke).

The above guidelines were instigated by multiple studies. For example, in 2010 Basaria et al reported that a clinical trial of TRT was halted early owing to a high rate of cardiovascular events in the treatment group.²⁴ Finkle et al published an observational study in 2014 that found a higher risk of acute myocardial infarction in both younger and older men with pre-existing heart disease.²⁵ Among men less than 65 years of age, the study found an almost 3-fold increase in myocardial infarction events within the first 90 days after filling a prescription for TRT. For older men, the increased risk was two-fold. In contrast, two other observational studies found a significant reduction in the all-cause mortality rate associated with TRT.^26,27 A meta-analysis of placebo-controlled studies reported that TRT was associated with an increase in cardiovascular events.²⁸

Given the demographics of patients receiving TRT, it is not surprising that many have cardiovascular disease. Of the 2.3 million patients prescribed TRT in 2013, 1.3 million, or 57%, had one or more concurrent claims for cardiovascular disease medication.²⁹ Physicians should discuss the benefits and risks, including possible risks, with patients before prescribing TRT.

TRT has been approved in the U.S. since the 1950s for conditions associated with deficiency or absence of endogenous testosterone. TRT is effective for both primary (testicular) and secondary (hypothalamic/pituitary) conditions. TRT was intended to restore testosterone levels into the “normal” adult male range.

In the past decade, more men, especially middle-aged men, were provided TRT. These men may have nonspecific symptoms of aging that overlap with symptoms associated with “classic” hypogonadism. It has not been established that these symptoms are the consequence of the age-related decline in testosterone. Hypogonadism may have been under-recognized, or it may be that there are many more men with evidence of hypogonadism than in prior decades.

The volume of TRT prescriptions grew greatly from 2010 to 2013. In 2010, there were 1.3 million prescriptions and this grew to 2.3 million in 2013, representing a 76% increase.³⁰ This rise in prescriptions coincided with the advent of direct-to-consumer advertisements that described the benefits of TRT.

According to a report based on health insurance claims, the largest group receiving TRT was composed of men ages 40-60 years.³⁰ An FDA analysis of 243,091 claims for TRT showed that approximately 70% of the men prescribed TRT were ages 40-64 years.³¹

The same FDA analysis revealed that only about 50% of men prescribed TRT had been diagnosed with hypogonadism. This suggests that approximately half of the men are being treated and monitored based on nonspecific symptoms. Consistent with this finding is the U.S. office-based physician practices survey that showed the most common ICD-9 code associated with TRT is 257.2, “other testicular function.” The FDA study further found that 25% of men prescribed TRT had not had their testosterone level measured prior to initiating treatment; 6% had a claim for testosterone testing only after initiation of therapy; and 21% had no evidence of testing anytime during treatment. Without baseline and follow-up testosterone measurements and consideration of nonhypogonadal causes of low testosterone, assessment of TRT effectiveness is impossible.

Early Development

Testosterone is necessary in utero for the development of male genitalia in 46,XY fetuses. In the absence of testosterone, the fetus tends to develop as a female. Thus, a 46,XY newborn with a sexual development disorder may present with external genitalia ranging from nearly normal female to nearly normal male, depending on the severity of the disorder. Total testosterone levels can range from absent to increased, depending on the condition. Male infants with hypogonadism or hypopituitarism may display micropenis or cryptorchidism.

Puberty

Delayed puberty and hypogonadism in boys can be associated with primary or secondary testicular failure. Elevated luteinizing hormone (LH) and follicle stimulating hormone (FSH) are consistent with primary hypogonadism, whereas decreased levels are consistent with secondary or tertiary hypogonadism.

Girls/Women

Measurement of free or bioavailable testosterone in females offers greater sensitivity for evaluation of mild androgen excess than does total testosterone.^14,15 In girls and women, excess androgen production is associated with premature adrenarche (ie, appearance of pubic and/or axillary hair before age 8), oligo/amenorrhea, and clinical features of hyperandrogenism (eg, alopecia, severe acne, hirsutism). These features are associated with polycystic ovary syndrome (PCOS), the most common endocrinopathy in women of reproductive age. Free testosterone levels are elevated in ~70% of PCOS cases.¹⁵