At a Glance
Amenorrhea is the absence of menstrual blood flow. Primary amenorrhea should be considered in any patient with secondary sex characteristics who has not experienced periodic menstruation by 15 years of age or 5 years after breast development. Patients who have not developed secondary sex characteristics, especially the absence of breast development, and have not established periodic menstruation by 13 years of age should also be worked up for primary amenorrhea.
Polycystic ovary syndrome (PCOS) is the most common cause of amenorrhea in women with evidence of androgen excess. Although, PCOS most often occurs concurrently with onset of menses, it can also cause primary amenorrhea. The Androgen Excess Society defines PCOS as the presence of clinical and/or biochemical hyperandrogenism plus either abnormal ovulatory cycles and/or polycystic ovaries. This diagnosis can only be made after excluding related syndromes.
The current definition includes a broad spectrum of patients making the diagnosis challenging. A patient with primary amenorrhea due to PCOS may have signs of hyperandrogenism (hirsutism, male pattern baldness, acne) and/or hyperandrogenemia (elevated serum testosterone). Patients with PCOS and primary amenorrhea may also be infertile and insulin resistant (central obesity, hypertension, acanthosis nigricans). Hormonal dysregulation in the hypothalamic-pituitary-gonadal (HPG) axis, including elevated androgens and estrogens, hypersecretion of luteinizing hormone (LH), and reduced follicle stimulating hormone (FSH) synthesis, leads to menstrual irregularities and amenorrhea in women with PCOS.
The work-up for primary amenorrhea and suspected PCOS begins with a careful history and physical exam to look for anatomical defects, development of secondary sexual characteristics, and/or a personal or family history of hirsutism, obesity, metabolic syndrome, cardiovascular disease (CVD), diabetes, infertility, and/or amenorrhea. A personal history of slow onset of hirsutism is common in patients with PCOS. Rapid onset of symptoms suggests an androgen secreting adenoma. Physical exam of patients with PCOS often reveals a woman with excess facial and/or body hair and sometimes alopecia and/or acne. Women are often obese. The presence of moon facies, buffalo hump, thinned skin, and myopathy suggest Cushing’s syndrome (see chapter on Cushing’s Syndrome).
Although not essential for the diagnosis, ultrasonographic evidence of polycystic ovaries is present in more than 80% of patients with PCOS. Identification of at least 12 ovarian follicles measuring 2–9 mm and increased ovarian size, along with hyperandrogenism/hyperandrogenemia, suggests that amenorrhea is due to PCOS.
The laboratory work-up should begin with measurement of human chorionic gonadotropin (hCG) to rule out pregnancy. Many patients with PCOS ovulate intermittently, and, thus, pregnancy is possible. Nonpregnant patients with clinical suspicion should be evaluated for thyroid dysfunction by measuring TSH. Although rare in patients with hyperandrogenism, a high or low thyroid stimulating hormone (TSH) result suggests the amenorrhea is due to primary hypothyroidism or hyperthyroidism, respectively. Abnormal TSH results should be followed with fT4 analysis. A reduced fT4 confirms hypothyroidism, whereas and elevated result confirms hyperthyroidism.
Despite its rarity in patients with androgen excess (<1%), prolactin should also be measured to rule out prolactinoma. An elevated prolactin result should prompt a physician to perform an MRI in search of a pituitary adenoma. Prolactin inhibits gonadotropin function, thus, causing amenorrhea in nursing mothers and patients with prolactinomas.
Patients with normal TSH and prolactin should be evaluated for gonadotropic function by measuring LH and FSH. Patients with amenorrhea due to PCOS often have gonadotropic abnormalities. Increases in the GnRH/LH pulse frequency lead to elevated LH in the context of normal to low FSH. In lean patients, an LH/FSH ratio greater than 2-3 during the follicular phase of the menstrual cycle is indicative of PCOS.
Patients with primary amenorrhea, hirsutism, and gonadotropic abnormalities should be evaluated for hyperandrogenemia. The gold standard for assessing androgen concentrations is serum free testosterone. Quantitate total testosterone by LC/MS/MS and Sex hormone binding globulin (SHBG) by immunoassay, and then calculate free testosterone using a predetermined mathematical model. Elevated total and/or free testosterone confirms hyperandrogenemia. Excess concentrations of adrenal androgens are also present in a portion of patients with PCOS. Measurement of DHEA and/or DHEA-S may be helpful to assess adrenal gland involvement in amenorrhea.
After establishing hyperandrogenemia and/or hyperandrogenism and/or polycystic ovaries, other related disorders leading to amenorrhea and hirsutism must be ruled out to confirm a diagnosis of PCOS. Along with thyroid disease and hyperprolactinemia, diseases with overlapping clinical and biochemical components like late onset congenital adrenal hyperplasia (CAH), Cushing’s Syndrome, Androgen secreting neoplasms, and HAIR-AN must be excluded (see chapter on Congenital Adrenal Hyperplasia (CAH)).
Nonclassical CAH is associated with mutations in enzymes involved in steroid hormone synthesis especially 21-hydroxylase (21-OH). Decreased 21-OH activity leads to accumulation of precursor 17-hydroxyprogesterone (17-OHP). Basal elevations in serum 17-OHP suggest CAH. This diagnosis is confirmed by detecting elevated 17-OHP in response to ACTH stimulation testing.
Cushing’s syndrome is rare in patients with hyperandrogenemia and should only be ruled out if hallmark signs, such as moon facies and buffalo hump, are present. Detection of an elevated 24-hour urinary free cortisol is suggestive of Cushing’s syndrome. Cushing’s is confirmed with an abnormal response to an overnight dexamethasone suppression test (i.e., lack of suppression). If patients experience a rapid onset with significant degree of hyperandrogenism and hyperandrogenemia, patients should have an MRI to search for an ovarian androgen-secreting neoplasm.
Patient’s with PCOS-like symptoms and severe insulin resistance may have hyperandrogenic-insulin resistant-acanthosis nigricans (HAIR-AN) syndrome. Since these patients are predisposed to a higher risk for developing PCOS comorbidities, such as CVD and diabetes, they should be closely monitored. These patients may be distinguished from patients with PCOS by detecting marked elevations in fasting insulin (>80 mcU/mL) or post prandial insulin (>300 mcU/mL). After ruling out related disorders, PCOS is the most likely cause of primary amenorrhea in patients with hyperandrogenemia and/or hyperandrogenism.
Are There Any Factors That Might Affect the Lab Results? In particular, does your patient take any medications – OTC drugs or Herbals – that might affect the lab results?
False-positive and false-negative results to urine pregnancy tests can occur and may result in a misdiagnosis. False negative results occur is urine that is too dilute. To ensure an appropriate urine specimen, perform urine pregnancy testing on first morning voids and check the protein concentration by measuring the urine-specific gravity and/or urine creatinine. False-negative results may also be caused by the variant effect. This phenomenon occurs when high concentrations of hCG isoforms in urine (hCG beta core fragment) are not recognized by both antibodies in the assay. Instead, they interfere with one antibody and cause a false-negative result. One can test for the variant effect by diluting the urine sample and repeating the testing.
Prolactin is mildly elevated by stress, herpes simplex virus (HSV) infections in the chest wall, and numerous drugs, including dopamine agonists, proton pump inhibitors, antipsychotics (Risperidone, Phenothiazines, Haloperidol), antihypertensives (Methyldopa, reserpine, Verapamil), estrogens, and illicit drugs (amphetamines, cannabinoids, opiates, etc.). Any of these may lead to dysregulation of gonadotropins, amenorrhea, and infertility.
LH and FSH are episodically released from the pituitary, and concentrations may vary with time of measurement. First morning specimens are recommended. LH and FSH concentrations change throughout the menstrual cycle, even in amenorrheic patients. LH and FSH should be measured early in the follicular phase of the cycle, if possible.
Concentrations of LH and FSH change during puberty, results should be evaluated in the context of age and tanner stage specific reference intervals. Drugs, such as anticonvulsants, clomiphene, and naloxone, may falsely elevate LH, whereas smoking, cimetidine, clomiphene, digitalis, and levodopa may elevate FSH. Artificially low LH and FSH results may occur in patients taking oral contraceptives and hormone treatments. Phenothiazines reduce FSH concentrations, whereas digoxin decreases LH. LH pulsatility is reduced in obese patients, thus, the LH/FSH ratio is not diagnostic in the majority of PCOS patients. Although an elevated LH/FSH ratio is suggestive of PCOS, a nonelevated ratio is not informative.
Because of lack of precision and correlation with standards in women and children, the Androgen Excess Society recommends measurement of testosterone by a quantitative high performance liquid chromatography-Tandem Mass Spectrometry assay. Testosterone values are affected by age, race, body mass index (BMI), and time of day and menstrual cycle. Thus, results should be evaluated in the context of age-, race-, and BMI-specific reference intervals and collected in the morning and early in the menstrual cycle, if possible.
Marked elevations in serum testosterone (>2 ng/mL) are most likely due to androgen secreting neoplasms (ASN), not PCOS. However, not all ASNs show marked serum testosterone elevations. Testosterone concentrations are elevated by many drugs, including danazol, androgenic progestins, valproic acid, acetazolamide, minoxidil, and oral contraceptives. Testosterone concentrations may be reduced in patients taking anabolic steroids.
Sex hormone binding globulin (SHBG) is affected by age, sex, and pubertal status and should be evaluated with age-, sex-, and tanner stage-specific reference intervals. Obesity and/or insulin resistance, along with hypothyroidism, androgens, and Cushing’s disease, decrease SHBG synthesis. Hyperthyroidism, anorexia, oral contraceptives, and liver disease increase SHBG concentrations in serum.
Calculation of free testosterone with mathematical modeling requires accurate measures of total testosterone and SHBG and is site specific. Appropriate validation of the % free testosterone calculation requires analysis of several hundred patients.
Measurement of serum DHEA is contraindicated for the diagnosis of PCOS because of is short half-life and diurnal and intra-individual variation. DHEA-S is the preferred analyte to address adrenal androgen excess, as it is unchanged throughout the day and menstrual cycle. DHEA-S concentrations decrease with age and should be assessed with age-specific reference intervals.
Insulin assays are intended to measure endogenous insulin, but some assays also recognize human recombinant insulin. Insulin concentrations may be increased in patients on insulin therapy. Some insulin assays may recognize proinsulin, causing falsely elevated results. Commercially available insulin assays are not standardized and should not be used interchangeably across laboratories. Guidelines from the National Academy of Clinical Biochemistry suggest that measurement of fasting insulin to determine degree of insulin resistance provides no additional diagnostic information to clinical signs and symptoms, such as acanthosis nigricans and central obesity.
As is the case with many immunoassays, heterophilic antibodies can cause false-positive results. Therefore, caution should be taken when hCG, TSH, prolactin, LH, FSH, testosterone, SHBG, and insulin immunoassay results do not match the clinical picture.
What Lab Results Are Absolutely Confirmatory?
Work-up of a patient with primary amenorrhea due PCOS is diagnosis of exclusion. There is no confirmatory test for amenorrhea due to PCOS. The Androgen Excess Society currently defines PCOS as the presence of hyperandrogenemia (elevated testosterone) or hyperandrogenism (hirsutism, alopecia, and/or acne) with ovulatory dysfunction and/or polycystic ovaries on ultrasound and excluding related disorders.
Patients with suspected PCOS should be closely monitored for onset of comorbidities, including diabetes, CVD, and endometrial cancer. Fasting glucose should be measured in all patients with suspected PCOS. If normal, patients should be monitored every 2 years. If impaired (fasting glucose 100–125 mg/dL), patients should undergo an oral glucose tolerance test. An elevated glucose (>140 ng/dL) 2 hours after glucose load is considered abnormal. Alternatively, hemoglobin A1C can be measured to screen for diabetes. A1C values between 5.7 and 6.4% are considered prediabetic, whereas values greater than 6.5% confirm the diagnosis of diabetes.
Patients should be screened for CVD every 2 years. A detailed physical exam should include a check for BMI, central obesity, and hypertension. Abnormal results of a complete lipid profile, including fasting triglycerides greater than 150 mg/dL and/or HDL less than 50 mg/dL, are associated with an increased risk of CVD.
PCOS patients should be screened annually (or when symptoms arise) for endometrial cancer by pelvic exam and/or ultrasound.
Several diseases and medications affect fasting and post challenge glucose results. Acromegaly, stress, chronic renal failure, Cushing’s syndrome, hyperthyroidism, pancreatitis, and pancreatic cancer elevate glucose results and should be ruled out prior to diagnosing a patient with impaired fasting glucose. Drugs, such as corticosteroids, tricyclic antidepressants, diuretics, epinephrine, oral contraceptives, lithium, Dilantin, and salicylates, elevate glucose.
Hemoglobin A1C results reflect long-term glucose control and are not accurate in patients with abnormal red cell life spans, including those with acute blood loss, hemolytic anemia, a hemoglobinopathy, or thalassemia. Some hemoglobin variants may coelute with A1C and directly interfere with the assay.
Fasting plasma triglyceride (TG) concentrations are elevated in patients who are overweight or obese, smoke, drink alcohol, have hypothyroidism, and have inherited lipid disorders. Drugs, such as corticosteroids, beta blockers, and estrogens, increase TG results. The reference intervals established by the ATP III guidelines are not intended for
HDL and total cholesterol concentrations are reduced in critically ill patients.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- At a Glance
- Are There Any Factors That Might Affect the Lab Results? In particular, does your patient take any medications - OTC drugs or Herbals - that might affect the lab results?
- What Lab Results Are Absolutely Confirmatory?