Polycystic ovary syndrome (PCOS) is a common endocrine disorder that affects about 1 in 15 women of reproductive age.1-3 The multifaceted syndrome of PCOS varies substantially among affected women. Primary features include androgen excess (hirsutism and acne), menstrual irregularity, and metabolic dysfunction (obesity, insulin resistance, acanthosis nigricans).1-3 PCOS is also one of the most common causes of female infertility, affecting 6% to 12% of reproductive-age women (as many as 5 million) in the United States.4

The exact causes of PCOS are unknown but are thought to involve complex genetic and environmental interactions. Insulin resistance, accompanied by compensating hyperinsulinemia, is believed to play a key role in PCOS pathophysiology by increasing ovarian and adrenal androgen production.1-3 Clinical presentations of PCOS encompass a wide spectrum, ranging from a mild phenotype (normal androgens with oligomenorrhea and polycystic ovaries) to a severe phenotype (marked hirsutism, alopecia, and obesity).1-3

This Clinical Focus discusses the important role that laboratory testing plays in the diagnosis and management of PCOS. This information is not intended as medical advice. Test selection and interpretation, diagnosis, and patient management decisions should be based on the physician’s education, clinical expertise, and assessment of the patient.

Diagnosis of PCOS

An estimated 75% of individuals with PCOS remain undiagnosed because of the variability in presentation and differences in published clinical diagnostic criteria.5 Notably, a systematic review of clinical practice guidelines for the diagnosis of PCOS found differences in diagnostic criteria and recommended testing.6 Overall, women eventually diagnosed with PCOS report that they are dissatisfied with the care that they received.7 The diversity among diagnostic approaches used for PCOS in clinical practice can delay diagnosis.6,8 More than one-third of women with PCOS spend >2 years and see ≥3 healthcare professionals before the condition is diagnosed.7

No single criterion or test can be used to diagnose PCOS—diagnosis is primarily based on the exclusion of diseases with similar symptoms.1-3 The 2003 Rotterdam criteria are the most-used approach for diagnosis of PCOS; the Rotterdam consensus defines PCOS based on the presence of at least 2 of 3 criteria: (1) hyperandrogenism; (2) ovulatory dysfunction; and (3) polycystic ovarian morphology (PCOM).1-3 Four phenotypes of PCOS are recognized that differ based on the presence or absence of 1 of these criteria (Table 1).9 Notably, PCOM on ultrasound is not necessary for diagnosis of PCOS phenotype B.

Table 1. Polycystic Ovary Syndrome Phenotypes Based on Rotterdam Criteria [return to contents]

The criteria used to diagnose PCOS are also satisfied by many other disorders (eg, congenital adrenal hyperplasia [CAH]), nonclassic CAH due to 21-hydroxylase deficiency, hyperprolactinemia, obesity, and hypothyroidism).1-3 Some similar disorders have potentially distinguishing characteristics that can help distinguish them from PCOS (Table 2). Importantly, PCOS can only be diagnosed once other disorders have been excluded.1-3,10

Table 2. Potentially Distinguishing Characteristics Present in Disorders With Similar Clinical Features to PCOS [return to contents]

In 2018, the International Evidence-Based Guideline for the Assessment and Management of Polycystic Ovary Syndrome (hereafter referred to as the International Evidence-Based Guideline) published 31 evidence-based recommendations, 59 clinical consensus recommendations, and 76 clinical practice points related to assessment and management of PCOS.9 The advisory board concluded that evidence in the assessment and management of PCOS is generally of low to moderate quality.11

As summarized below, the guideline endorsed the Rotterdam criteria for diagnosis of PCOS but also provided (1) refinement of individual diagnostic criteria, with a focus on improving the accuracy of diagnosis; (2) recommendations to reduce unnecessary testing; and (3) increased focus on education, lifestyle modification, emotional well-being, and quality of life.9

Hyperandrogenism

Clinical hyperandrogenism

Features of clinical hyperandrogenism include9

• Hirsutism (excess facial or body hair)
• Acne
• Alopecia ranging from vertex, crown, and diffuse pattern to bitemporal and frontal hairline loss

Standardized instruments can be used to evaluate hirsutism (eg, modified Ferriman Gallwey score) and alopecia (eg, Ludwig visual score).9 In addition, diagnosis of clinical hyperandrogenism should take into account natural variations in presentation by race and ethnicity.9

Biochemical hyperandrogenism

Assessment of biochemical hyperandrogenism is useful in establishing the diagnosis of PCOS when clinical signs of hyperandrogenism (in particular hirsutism) are unclear or absent.9

Testing to assess biochemical hyperandrogenism in the diagnosis of PCOS includes9

• Free testosterone or calculated bioavailable testosterone
• Androstenedione or dehydroepiandrosterone (DHEA) sulfate if bioavailable testosterone is normal and hyperandrogenism is highly suspected

The International Evidence-Based Guideline recommends that high-quality assays, including liquid chromatography mass spectrometry-based assays, be used for the most accurate assessment of total or free testosterone in PCOS.9

Specific testing available at Quest Diagnostics for the assessment and diagnosis of hyperandrogenism associated with PCOS are presented below, in the section Laboratory Testing for PCOS.

Ovulatory dysfunction

Ovulatory dysfunction manifests as irregular menstrual cycles and amenorrhea. The International Evidence-Based Guideline defines irregular menstrual cycles as9

• Normal in the first-year postmenarche as part of the pubertal transition
• 1 to <3 years postmenarche: menstrual cycles <21 days or >45 days
• 3 years postmenarche to perimenopause: menstrual cycles <21 or >35 days or <8 cycles per year
• 1 year postmenarche: >90 days for any 1 cycle
• Primary amenorrhea by age 15 or >3 years after thelarche (breast development)

In patients with irregular menstrual cycles, a diagnosis of PCOS should be considered after other more common causes of irregular cycles (eg, stress and lifestyle factors) are ruled out.9

Diagnosis of PCOS in an adolescent with irregular menstrual cycles can be challenging. Many symptoms of PCOS are typical for reproductive maturation during puberty (eg, anovulation). Ovulatory dysfunction may be present even with regular menstrual cycles. Anovulation, if suspected, can be evaluated by measuring serum progesterone level in the luteal phase.9 Notably, the use of different diagnostic criteria can markedly affect the diagnosis of PCOS in adolescents12 as well as perimenopausal women.9

Polycystic ovarian morphology (PCOM)

Ultrasound is used to identify PCOM.9 Although an ovarian ultrasound is not required to diagnose PCOS in women with irregular menstrual cycles and hyperandrogenism,9 it can be useful for identifying the complete PCOS phenotype.9,13

Considerations for the use of ultrasound for diagnosis of PCOM in PCOS include the following9

• High-resolution, transvaginal ultrasound should be performed.
• Follicle count per ovary ≥20 or ovarian volume ≥10 mL is diagnostic of PCOM.
• Ultrasound should not be used for the diagnosis of PCOS in adolescents who are <8 years after menarche (gynecological age of <8 years), because high incidence of multi-follicular ovaries is high.

Notably, anti-Müllerian hormone (AMH) measurement is not recommended as an alternative to ultrasound for identification of PCOM, nor as a single test for the diagnosis of PCOS.9 Improved standardization of AMH assays and large-scale validation studies are required to demonstrate their clinical utility in the diagnosis of PCOS.9

Comorbidities and other considerations

Women with PCOS should be assessed for a number of comorbidities associated with this condition.1-3

About half of all women with PCOS are obese.1 Compared to the general population, women with PCOS are 4 times as likely to develop type 2 diabetes mellitus1 and 2 times as likely to develop metabolic syndrome.3 Notably, more than half of women with PCOS develop type 2 diabetes by age 40.4 Women with PCOS are also at increased risk for hypertension, cardiovascular disease (CVD), and metabolic syndrome and should be monitored regularly for CVD risk.1-3

Women with PCOS are also at increased risk for abnormal uterine bleeding, endometrial cancer, infertility, and pregnancy complications.1-3 Women with PCOS who present with oligomenorrhea and hyperandrogenism (a severe phenotype) are especially at risk for the aforementioned conditions.1-3

Individuals with PCOS, particularly adolescents, also have higher rates of anxiety-related disorders and depression compared with the general population.14 The International Evidence-Based Guideline recommends that all women, especially adolescents, be screened for symptoms of anxiety and depression at the time of diagnosis; those with positive screening results should be referred to an appropriate healthcare provider.11

The hyperandrogenism of PCOS can persist into menopause.9 A diagnosis of PCOS postmenopause is more likely if a patient has a past diagnosis of PCOS, a long-term history of irregular menstrual cycles and hyperandrogenism, and/or PCOM during the reproductive years.9 Postmenopausal individuals presenting with new-onset, severe, or worsening signs of hyperandrogenism (eg, hirsutism) require evaluation for androgen-secreting tumors and ovarian hyperthecosis.9

Test availability [return to contents]

Quest Diagnostics offers tests and panels for the differential diagnosis of PCOS and conditions with overlapping features (Table 3). Quest also offers tests and panels for the diagnosis and management of comorbidities associated with PCOS. For example, testing related to cardiometabolic comorbidities (diabetes and CVD risk) is presented in Table 4.

Table 3. Laboratory Tests for Differential Diagnosis of PCOS and Assessment of PCOS Criteria [return to contents]

Table 4. Laboratory Tests for the Diagnosis and Management of Cardiometabolic Comorbidities Associated with PCOS [return to contents]

Test selection and interpretation [return to contents]

The following section summarizes literature and guidelines for test selection and interpretation guiding differential diagnosis of PCOS from conditions with overlapping features, tests that help support a PCOS diagnosis, and diagnosis and management of comorbidities associated with PCOS.

Differential diagnosis of PCOS

Conditions that share many of the same features with PCOS and should be excluded before a diagnosis of PCOS is made. In addition to potentially distinguishing characteristics (Table 2), laboratory testing can be used for differential diagnosis (Figure 1 and discussed below). The conditions are listed below in alphabetical order.

Figure 1. Polycystic Ovary Syndrome (PCOS) Diagnostic Algorithm [return to contents]

Acromegaly

Acromegaly, like PCOS, can have symptoms of oligomenorrhea and hirsutism with insulin resistance. However, it is readily distinguished from PCOS by elevated serum insulin-like growth factor-1 (IGF-1)16 (see Appendix Table 1 for reference ranges).

Amenorrhea

Conditions characterized by amenorrhea include hyperprolactinemia and hyper- or hypothyroidism, as well as pregnancy, hypothalamic amenorrhea, and primary ovarian insufficiency.16

Pregnancy is distinguished from PCOS by the presence of human chorionic gonadotropin (hCG). As described below for thyroid-stimulating hormone (TSH), specimen pretreatment to remove human anti-mouse antibodies (HAMA) may be required for detection of hCG in some patients.

Hypothalamic amenorrhea (including functional hypothalamic amenorrhea) and primary ovarian insufficiency can be distinguished from PCOS in most patients by low serum levels of estradiol. Functional hypothalamic amenorrhea reference ranges have not been established for Quest assays; postmenopausal reference ranges are ≤0.38 pg/mL (free estradiol, test code 36169) and ≤10 pg/mL (ultrasensitive estradiol, test code 30289). Serum luteinizing hormone (LH) and follicle stimulating hormone (FSH) measurements are also useful in distinguishing these disorders. In PCOS, serum FSH levels are normal and LH levels are normal or elevated (LH levels are usually higher than FSH).16,19 In functional hypothalamic amenorrhea, these markers are low to low-normal; FSH levels are usually higher than LH. Reference ranges for luteal phase are FSH: 1.5-9.1 mIU/mL20 and LH: <0.5 mIU/mL (test code 7137). In primary ovarian insufficiency, FSH levels are elevated; postmenopausal levels are typically 23.0 to 116.3 mIU/mL (test codes 470 and 7137).

Cushing syndrome

Cushing syndrome has several symptoms that overlap with PCOS, including acne, hirsutism, and anovulation. Differential diagnosis involves measuring serum cortisol after dexamethasone suppression.2,16,21 Unlike PCOS, Cushing syndrome is associated with an elevated result. The Pituitary Society consensus guideline update indicates serum cortisol levels <1.8 μg/dL predict absence of Cushing syndrome.22 Values ≥1.8 μg/dL may be predictive of Cushing syndrome; the higher the value the less likely the elevation has another etiology, especially when both elevated cortisol (test code 6921) and dexamethasone (test code 29391) are are observed.22

Hyperprolactinemia

Hyperprolactinemia can present with amenorrhea, hirsutism, and galactorrhea. It is distinguished from PCOS by higher levels of serum prolactin (reference range: 3-30 ng/mL in nonpregnant women, test code 746).2,15,16 About 14% of young women (21 to 30 years) with amenorrhea have hyperprolactinemia.23 In women with amenorrhea and galactorrhea, prevalence estimates for hyperprolactinemia are as high as 70%.24 In addition, certain medications, especially psychotropic medications, can cause elevated prolactin levels.24,25

In rare cases, elevated prolactin test results are due to macroprolactin, which is usually an inactive form of prolactin bound to an IgG. While most women with macroprolactinemia are asymptomatic, some may experience menstrual disorders (17%-39%), infertility (29%-35%), and galactorrhea (33%-46%).25,26 One possible reason for the variability is that antibodies may bind epitopes that do not fully inactivate prolactin.9 Macroprolactin can be distinguished from prolactin by a polyethylene-glycol precipitation test (Table 3).

Finally, elevated prolactin levels may be caused by a noncancerous tumor called a prolactinoma, leading to anovulation, galactorrhea, and hyperandrogenic symptoms such as acne and hirsutism.10

Nonclassic CAH

Nonclassic CAH is often confused with PCOS because of the overlap in symptoms, such as hyperandrogenemia, oligo/amenorrhea, and PCOM.2,11,23 Unlike PCOS, the disease is exclusively caused by genetic defects in the steroidogenic enzyme 21-hydroxylase.2,11,23 Nonclassic CAH is distinguished from PCOS by increased levels of 17-hydroxyprogesterone (17-OHP) caused by partially diminished 21-hydroxylase activity (see Appendix Table 2 for reference ranges).

Measuring 17-OHP is useful when distinguishing PCOS from nonclassic CAH (Figure 2).18 High values (>1,000 ng/dL for 17-OHP for the early follicular phase) indicate 21-hydroxylase deficiency and thus nonclassic CAH, whereas values <200 ng/dL exclude nonclassic CAH.18 For values 200 to 1,000 ng/dL, further testing with cosyntropin stimulation is suggested: increased 17-OHP in response to stimulation (>1,000 ng/dL) suggests 21-hydroxylase deficiency and nonclassic CAH rather than PCOS.18 Genetic testing may be considered if the CAH diagnosis is uncertain.18 Elevated 11-deoxycortisol can differentiate CAH caused by 11-β-hydroxylase deficiency vs 21-hydroxylase deficiency.18

Figure 2. Laboratory Testing in the Differential Diagnosis of Polycystic Ovary Syndrome (PCOS) and Nonclassic Congenital Adrenal Hyperplasia (CAH) [return to contents]

Thyroid disease

Thyroid disease may lead to menstrual disorders similar to those of PCOS and can be distinguished by measuring serum TSH.1 The pituitary gland produces more TSH when blood levels of thyroid hormones T3 and T4 are low (hypothyroidism) and less when they are high (hyperthyroidism). TSH values (test code 899) above the upper limit of normal (>4.5 mIU/L) suggest hypothyroidism, whereas values below the lower limit of normal (<0.4 mIU/L) may suggest hyperthyroidism and the need for further evaluation.

Falsely elevated or depressed TSH values may occur in specimens from patients who have (1) received mouse monoclonal antibody preparations during diagnosis or therapy, or (2) been exposed to mice.27 Such patients may have developed HAMAs that interfere with accurate analysis. Testing specimens from such patients involves pretreatment to inhibit possible interference. In addition, TSH and immunoassays are affected by biotin intake; biotin supplements should be avoided for at least 24 hours, preferably several days, before these tests.28

Virilizing tumors

Adrenal-virilizing tumors may be accompanied by large elevations in androgens: total testosterone levels >150 ng/dL and DHEA sulfate levels >700 μg/dL.10,16,17 However, no specific cutoff levels define the presence of a tumor versus PCOS.

Ultrasound imaging of ovaries and magnetic resonance imaging or computed tomography of the adrenal glands are useful in the differential diagnosis (Figure 1) when there are marked elevations of serum testosterone and DHEA sulfate2,10,21 (for reference ranges see Appendix Table 1).

Laboratory testing useful for diagnosis of PCOS

Laboratory testing can help diagnose PCOS by providing biochemical evidence of hyperandrogenism, assessing ovulatory function, and detecting polycystic ovaries (Table 2).

Hyperandrogenism

High-sensitivity methods, such as liquid chromatography/tandem mass spectrometry (LC/MS/MS), should be used for the most accurate assessment of total or free testosterone in PCOS. Direct free testosterone assays, such as radiometric or enzyme-linked assays, are not recommended, as they demonstrate poor sensitivity, accuracy, and precision at the levels required for detection in women.25

Elevated testosterone levels provide biochemical evidence for hyperandrogenism associated with PCOS (see Appendix Tables 1 and 2 for reference ranges). Total testosterone reflects all forms of testosterone: free, bioavailable, and bound to proteins such as sex hormone-binding globulin (SHBG) and albumin.21,29 For diagnosis of PCOS, measuring free testosterone using LC/MS/MS and equilibrium dialysis is preferred over measuring total testosterone alone, because it is a more sensitive test for hyperandrogenism.21 Calculated bioavailable testosterone, as determined by LC/MS/MS, can also be used to diagnose hyperandrogenism.9

In hyperandrogenemic women, decreased SHBG levels can elevate free testosterone levels even when total testosterone is normal (see Appendix Table 1 for reference ranges). In women with PCOS, low SHBG levels are associated with obesity, insulin resistance, hyperandrogenemia, anovulation, and PCOM.29 Measuring SHBG level is useful in the evaluation of testosterone levels, and in the future SHBG level may become a useful biomarker for diagnosis of PCOS.29

Measuring levels of other androgens, such as DHEA, DHEA sulfate, and androstenedione, may also help diagnose PCOS. Elevated levels of these markers are consistent with hyperandrogenemia30 and should be measured if hyperandrogenism is highly suspected and testosterone is normal.9 However, the value of routinely measuring these androgens is low because relatively few additional patients would receive a diagnosis.30

Ovulatory dysfunction

Ovulatory dysfunction (oligo- or anovulation) can occur with regular menstrual cycles, as well as with irregular bleeding. The International Evidence-Based Guideline recommends determining if ovulation is occurring by measuring serum progesterone level9 (eg, 7 days before the next menses are expected31). The American College of Obstetrics and Gynecology (ACOG) considers a single progesterone value of >3 ng/mL as evidence of ovulation.15 The progesterone cutoff value for ovulation can vary with the type of assay: Quest’s LC/MS/MS assay (test code 17183) reference range for midluteal progesterone level is 6.7 to 22.2 ng/mL.

Polycystic ovaries

PCOM is assessed with ultrasound; laboratory testing has limited value. Although an AMH value of ≥4.5 ng/mL has been proposed as a diagnostic threshold for PCOM, it is not considered a substitute for ultrasound.21,32 However, baseline values of AMH may predict the likelihood of women with PCOS achieving a live birth, and also help identify those who are less likely to ovulate and conceive on clomiphene therapy.32

Comorbidities and other considerations

The International Evidence-Based Guideline for the Assessment and Management of Polycystic Ovary Syndrome provides guidance for monitoring comorbidities, which includes testing related to diabetes and CVD risk.9 For more information on test selection and interpretation, see Table 4 and related test guides:

• CardioIQ Insulin Resistance Panel With Score (https://testdirectory.questdiagnostics.com/test/test-guides/TS_CardioIQ_IR/cardioiq-insulin-resistance-panel-with-score)
• Metabolic Risk Panel (https://testdirectory.questdiagnostics.com/test/test-guides/TS_MetabolicRiskPnl/metabolic-risk-panel)
• LDL Cholesterol (https://testdirectory.questdiagnostics.com/test/test-guides/TS_LDL_Cholesterol/ldl-cholesterol)
• Laboratory Testing for Diabetes Diagnosis and Management (https://testdirectory.questdiagnostics.com/test/test-guides/TG_Diabetes/laboratory-testing-for-diabetes-diagnosis-and-management)

Appendix. Reference Ranges of Tests Used in the Differential Diagnosis of PCOS [return to contents]

Appendix Tables 1 and 2 are provided for informational purposes only and are not intended as medical advice. 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.

Appendix Table 1. Reference Ranges for DHEA Sulfate, IGF-1, Free Testosterone and SHBG in Females [return to contents]

Appendix Table 2. Reference Ranges for 17-OHP, Androstenedione, and Total Testosterone [return to contents]