Female hirsutism is a relatively common problem in which the overriding concern of the physician is to rule out an ovarian or adrenal tumor. The type and distribution of hair can be used to differentiate physiologic hair growth from nonphysiologic growth (hirsutism). In females there are two types of hair: fine, thin, nonpigmented vellus hair and coarse, curly, pigmented terminal hair. Pubic and axillary terminal hair growth is induced by androgens in the physiologic range. Growth of such hair on the face (especially the chin), sternum area, and sacrum suggests male distribution and therefore excess androgens. Virilization goes one step further and is associated with definite hirsutism, acne, deepening of the voice, and hypertrophy of the clitoris.

The major etiologies of hirsutism are listed in the box. Hirsutism with or without other evidence of virilization may occur before or after puberty. When onset is prepubertal, the etiology is more often congenital adrenal hyperplasia or tumor. After puberty, PCO disease, Cushing’s syndrome, late-onset congenital adrenal hyperplasia, and idiopathic causes are more frequent. Findings suggestive of tumor are relatively sudden onset; progressive hirsutism, especially if rapidly progressive; and signs of virilization.

Laboratory tests

There is considerable disagreement among endocrinologists regarding which tests are most useful in hirsutism. The tests most often used are urine 17-KS, serum testosterone (total or free testosterone), DHEA-S, serum androstenedione, serum dihydrotestosterone, urine 3-a-androstanediol glucuronide, serum LH and FSH, serum prolactin, serum 17-OH-P, urine free cortisol, an ACTH stimulation (“Cortrosyn”) test, and the low-dose overnight dexamethasone suppression test. Each of these tests is used either to screen for one or more etiologies of hirsutism or to differentiate between several possible sources of abnormality. There is considerable disagreement concerning which tests to use for initial screening purposes, with some endocrinologists ordering only two or three tests and others using a panel of five to seven tests. Therefore, it may be worthwhile to briefly discuss what information each test can provide.

Some Conditions That Produce Hirsutism
Ovary
PCO disease
(Hyperthecosis)
Ovarian tumor
Adrenal
Congenital adrenal hyperplasia (CAH)
Cushing’s syndrome (nontumor)
Adrenal tumor
Testis
Leydig cell tumor
Other
Idiopathic hirsutism
Hyperprolactinemia
Starvation
Acromegaly (rare)
Hypothyroidism (rare)
Porphyria (rare)
Medications
Phenytoin (Dilantin)
Diazoxide (Hyperstat)
Minoxidil(Loniten)
Androgenic steroids
Glucocorticosteroids
Streptomycin

Urine 17-KS was one of the first tests in hirsutism or virilization. Elevated urine 17-KS levels suggest abnormality centered in the adrenal, usually congenital adrenal hyperplasia (CAH) or adrenal tumor. Ovarian tumors, PCO disease, or Cushing’s syndrome due to pituitary adenoma usually do not produce elevated urine 17-KS levels (although some exceptions occur). However, some cases of CAH fail to demonstrate elevated 17-KS levels (more often in the early pediatric age group and in late-onset CAH). Serum 17-OH-P in neonatal or early childhood CAH and 17-OH-P after ACTH stimulation in late-onset CAH are considered more reliable diagnostic tests. Also, a significant minority (16% in one series) of patients with adrenal carcinoma and about 50% of patients with adrenal adenomas fail to show elevated 17-KS levels. In addition, 17-KS values from adrenal adenoma and carcinoma overlap, although carcinoma values in general are higher than adenoma values. The overnight low-dose dexamethasone test is more reliable to screen for Cushing’s syndrome of any etiology (including tumor) than are urine 17-KS levels, and the 24-hour urine free cortisol assay is a little more reliable than the overnight dexamethasone test (see Chapter 30). At present most endocrinologists do not routinely obtain 17-KS levels.

Serum testosterone was discussed earlier. Testosterone levels are elevated (usually to mild degree) in more than one half of the patients with PCO disease and in patients with Leydig cell tumors or ovarian testosterone-producing tumors. Elevated serum testosterone suggests a problem of ovarian origin because the ovaries normally produce 15%-30% of circulating testosterone; and in addition, the ovary produces about one half of circulating androstenedione from adrenal DHEA and this androstenedione is converted to testosterone in peripheral tissues. On the other hand, testosterone is produced in peripheral tissues and even in the adrenal as well as in the ovary. Serum testosterone levels therefore may be elevated in some patients with nonovarian conditions such as idiopathic hirsutism. These multiple origins make serum testosterone the current best single screening test in patients with hirsutism. Several investigators have found that serum free testosterone levels are more frequently elevated than serum total testosterone levels in patients with hirsutism, although a lesser number favor total testosterone. Possible false results in total testosterone values due to changes in testosterone-binding protein is another point in favor of free testosterone. Some investigators have reported that serum androstenedione and DHEA-S levels are elevated in some patients with free testosterone values within reference range and that a test panel (e.g., free testosterone, androstenedione, dihydrotestosterone, and DHEA-S) is the most sensitive means of detecting the presence of excess androgens (reportedly with a sensitivity of 80%-90%).

Dehydroepiandrosterone-sulfate (DHEA-S) is produced entirely in the adrenal from adrenal DHEA. Therefore, elevated DHEA-S levels suggest that at least some excess androgen is coming from the adrenal. The DHEA-S test is also more sensitive for adrenal androgen excess than the urine 17-KS test. A minor difficulty is that a certain number of patients with PCO disease (in which the ovary is supposed to be the source of excess androgen production) and also certain patients with idiopathic hirsutism have some DHEA-S evidence of adrenal androgen production. For example, one third of PCO disease patients in one series were found to have increased DHEA-S levels. Although this was interpreted to mean that some adrenal factor was present as well as the ovarian component, it tends to confuse the diagnosis. Another difficulty is that some infertile women without hirsutism have been reported to have elevated DHEA-S levels. Androstenedione, as noted previously, is a metabolite of adrenal DHEA that is produced about equally in the adrenal and in the ovaries but then reaches peripheral tissues where some is converted to testosterone. Therefore, elevated serum androstenedione levels indicate abnormality without localizing the source. Dihydrotestosterone (DHT) is a metabolite of testosterone and is formed mainly in peripheral tissues. Therefore, elevated DHT levels suggest origin from tissues other than the ovaries or adrenals. Androstanediol glucuronide is a metabolite of DHT and has the same significance. Some investigators report that it is elevated more frequently than DHT. Serum prolactin levels are usually elevated in prolactin-producing pituitary tumors (prolactinoma). These tumors are said to be associated with hirsutism in about 20% of cases. The mechanism is thought to be enhancement of ACTH effect on formation of DHEA. However, it has also been reported that up to 30% of patients with PCO disease have mildly elevated serum prolactin levels (elevated < 1.5 times the upper limit of the reference range). Prolactinomas are more likely in patients with irregular menstrual periods.

Luteinizing hormone and FSH are useful in diagnosis of PCO disease, which typically (although not always) shows elevated LH levels, with or without elevated FSH levels.

Urine free cortisol or low-dose overnight dexamethasone test are both standard tests used for diagnosis of Cushing’s syndrome, which is another possible cause of hirsutism.

Serum 17-OH-P is used to diagnose CAH. Levels of the 17-OH-P specimens drawn between 7 A.M. and 9 A.M. are elevated in nearly all patients with CAH who have symptoms in the neonatal period or in early childhood. However, in late-onset CAH that becomes clinically evident in adolescence or adulthood, 17-OH-P levels may or may not be elevated. The most noticeable symptom of late-onset CAH is hirsutism. Reports indicate that about 5%-10% (range, 1.5%-30%) of patients with hirsutism have late-onset CAH. The most effective test for diagnosis of late-onset CAH is an ACTH (Cortrosyn) stimulation test. A baseline 17-OH-P blood specimen is followed by injection of 25 units of synthetic ACTH, and a postdose specimen is drawn 30 minutes after IV injection or 60 minutes after intramuscular injection. Exact criteria for interpretation are not frequently stated. However, comparison of test results in the literature suggests that an abnormal response to ACTH consists of 17-OH-P values more than 1.6 times the upper limit of the normal 17-OH-P range before ACTH. An “exaggerated” response appears to be more than 3 times the upper limit of the pre-ACTH normal range. In homozygous late-onset CAH there is an exaggerated 17-OH-P response to ACTH. CAH heterozygotes may have a normal response to ACTH or may have an abnormal response to ACTH that falls between a normal response and an exaggerated response.

Suppression tests, such as a modified dexamethasone suppression test with suppression extended to 7-14 days, have been advocated in the past to differentiate between androgens of adrenal and ovarian origin. Dexamethasone theoretically should suppress nontumor androgen originating in the adrenal. However, studies have shown that the extended dexamethasone suppression test may be positive (i.e., may suppress androgen levels to < 50% of baseline) in some patients with PCO disease, and the test is no longer considered sufficiently dependable to localize the origin of increased androgen to either the adrenal or the ovary alone.

Radiologic visualization of abdominal organs is helpful if PCO disease, ovarian tumor, or adrenal tumor is suspected. Ultrasound examination of the ovaries and CT of the adrenals are able to detect some degree of abnormality in most (but not all) patients.

Polycystic ovary (PCO) disease

PCO disease is considered by some to be the most common cause of female postpubertal hirsutism. It is also an important cause of amenorrhea, oligomenorrhea, and female sterility. PCO disease is defined both clinically and by histopathology. The classic findings at operation are bilaterally enlarged ovaries (about 65%-76% of cases), which on pathologic examination have a thickened capsule and numerous small cysts (representing cystic follicles) beneath the capsule. However, PCO disease is considered to have a spectrum of changes in which there is decreasing ovarian size and a decreasing number of cysts until the ovaries are normal in size (about 25%-35% of cases) with few, if any, cysts but with an increased amount of subcapsular and interstitial stroma. There is also a condition called “hyperthecosis” in which the thecal cells of the stroma are considerably increased and have a luteinized appearance. Some consider hyperthecosis a separate entity; some include it in PCO disease but consider it the opposite end of the spectrum from the polycystic ovary type; and some combine PCO disease and hyperthecosis together under a new name, “sclerocystic ovary syndrome.”

Clinically, there is considerable variety in signs and symptoms. The classic findings were described by Stein and Leventhal, and the appellation Stein-Leventhal syndrome can be used to distinguish patients with classic findings from patients with other variants of PCO disease. The Stein-Leventhal subgroup consists of women who have bilaterally enlarged polycystic ovaries and who are obese, are hirsute without virilization, have amenorrhea, and have normal urine 17-KS levels. Other patients with PCO disease may lack one or more of these characteristics. For example, only about 70% of patients have evidence of hirsutism (literature range, 17%-95%). Some patients with PCO disease have hypertension, and some have abnormalities in glucose tolerance. A few have virilization, which is said to be more common in those with hyperthecosis.

Laboratory tests. Laboratory findings in PCO disease are variable, as are the clinical findings. In classic Stein-Leventhal cases, serum testosterone levels are mildly or moderately elevated in about 50% of patients, and other androgen levels are elevated in many patients whose serum testosterone level remains within reference range. Free testosterone levels are elevated more frequently than total testosterone levels. Higher testosterone values tend to occur in hyperthecosis. Most investigators consider the increased androgen values to be derived mainly from the ovary, although an adrenal component has been found in some patients. Although urine 17-KS levels are usually normal, they may occasionally be mildly increased. The most characteristic finding in PCO is an elevated serum LH level with FSH levels that are normal or even mildly decreased. However, not all patients with PCO show this gonadotropin pattern.

Summary of tests in hirsutism

Most endocrinologists begin the laboratory investigation of hirsutism with a serum testosterone assay. Many prefer free testosterone rather than total testosterone. The number and choice of additional tests is controversial. Additional frequently ordered screening tests include serum DHEA-S, serum DHT, and the ACTH-stimulated 17-OH-P test. If abnormality is detected in one or more of these tests, additional procedures can help to find which organ and disease is responsible.