What is the comprehensive workup for Polycystic Ovary Syndrome (PCOS)?

PCOS Workup

The comprehensive workup for PCOS requires documenting two of three Rotterdam criteria: hyperandrogenism (clinical or biochemical), ovulatory dysfunction, and polycystic ovarian morphology, with total testosterone and free testosterone as first-line biochemical tests measured by LC-MS/MS, and transvaginal ultrasound showing ≥20 follicles per ovary or ovarian volume >10 mL in adults. 1, 2

Clinical History and Physical Examination

Essential History Components

• Menstrual history: Cycle length >35 days indicates chronic anovulation; cycles 32-35 days require assessment for ovulatory dysfunction 2, 3
• Onset and duration of androgen excess signs: Gradual onset with weight gain suggests PCOS; rapid onset with clitoromegaly suggests virilizing tumor 2, 3
• Medication review: Document use of exogenous androgens or hormonal contraceptives 2
• Family history: Cardiovascular disease and diabetes 2
• Lifestyle factors: Diet, exercise, alcohol use, smoking 2

Physical Examination Findings

• Hyperandrogenism signs: Hirsutism (use Ferriman-Gallwey score), acne (severe or isotretinoin-resistant acne has 40% PCOS likelihood), androgenic alopecia (vertex/crown/diffuse pattern), clitoromegaly 2, 3
• Anthropometric measurements: BMI and waist-hip ratio (central obesity is primary mediator of PCOS complications) 2, 4
• Signs suggesting alternative diagnoses: Buffalo hump, moon facies, hypertension, abdominal striae (Cushing's syndrome) 2

Laboratory Testing

First-Line Androgen Assessment

Total testosterone (TT) is the single best initial test with 74% sensitivity and 86% specificity 1, 5:

• Must use LC-MS/MS method (not direct immunoassay) for superior accuracy: 92% specificity vs 78% with immunoassay 1, 5
• Direct immunoassays are inaccurate at low female testosterone ranges and should be rechecked by LC-MS/MS if inconsistent with clinical presentation 1

Free testosterone (FT) has highest sensitivity at 89% with 83% specificity 1, 5:

• Calculate using Vermeulen equation from high-quality TT and SHBG measurements, OR measure by equilibrium dialysis or ammonium sulfate precipitation 1, 2
• Direct FT immunoassays are highly inaccurate due to low serum concentrations and should never be used 1

Free Androgen Index (FAI) has 78% sensitivity and 85% specificity 1, 5:

• Calculated as: (Total Testosterone / SHBG) × 100 1, 2
• Caution when SHBG <30 nmol/L as FAI becomes less reliable 2

Second-Line Androgen Tests (If TT/FT Not Elevated and Clinical Suspicion High)

Androstenedione (A4): 75% sensitivity, 71% specificity; particularly useful when SHBG is low 1, 2

DHEAS: 75% sensitivity, 67% specificity; most reliable for adrenal androgen production, particularly valuable in women <30 years 1, 2

Differential Diagnosis Laboratory Tests

• 17-hydroxyprogesterone (early morning): Rule out non-classic congenital adrenal hyperplasia 2, 3
• TSH: Exclude thyroid disease 2
• Prolactin: Rule out hyperprolactinemia 2
• 24-hour urinary free cortisol or overnight dexamethasone suppression test: If Cushing's syndrome suspected 2

Metabolic Screening (All PCOS Patients)

• Fasting glucose and 2-hour oral glucose tolerance test: Screen for type 2 diabetes and impaired glucose tolerance 2, 5
• Fasting lipid panel: Screen for dyslipidemia 2
• Consider HbA1c for ongoing diabetes surveillance 2

Imaging Assessment

Transvaginal Ultrasound (Adults ≥18 Years)

Technical requirements: Use ≥8 MHz transducer frequency for optimal resolution 1, 2

Gold standard marker - Follicle Number Per Ovary (FNPO): ≥20 follicles (2-9 mm diameter) has 87.64% sensitivity and 93.74% specificity 1, 2, 5

Alternative markers when accurate follicle counting impossible:

• Ovarian volume (OV) >10 mL: Comparable diagnostic accuracy to FNPO 1, 2
• Follicle Number Per Single cross-section (FNPS): Secondary alternative 1, 2

Additional findings to document: Presence of dominant follicle, corpus luteum, increased central stromal echogenicity (subjective but sensitive/specific sign) 1, 6

Imaging in Adolescents (<20 Years, <8 Years Post-Menarche)

Do NOT use ultrasound as first-line diagnostic tool due to poor specificity and high false-positive rates from normal multi-follicular ovaries 1, 2, 5

• Diagnosis relies on persistent oligomenorrhea 2-3 years beyond menarche PLUS biochemical/clinical hyperandrogenism 2, 3

Alternative Imaging

MRI pelvis without contrast: Consider in obese patients when transvaginal ultrasound unacceptable and transabdominal ultrasound limited; provides reproducible ovarian volume but only moderate interobserver agreement for follicle counts 1

Emerging Biomarkers

Anti-Müllerian Hormone (AMH): Significantly elevated in PCOS but lacks standardization across assays and established diagnostic cut-offs; not yet recommended for routine clinical use 1, 2

Diagnostic Algorithm

1. Document clinical hyperandrogenism (hirsutism, acne, alopecia) and/or biochemical hyperandrogenism (TT and FT by LC-MS/MS as first-line; consider A4/DHEAS if negative with high suspicion) 1, 2

2. Confirm ovulatory dysfunction (menstrual cycles >35 days or irregular cycles 2-3 years post-menarche in adolescents) 2, 3

3. Assess polycystic ovarian morphology (transvaginal ultrasound with FNPO ≥20 or OV >10 mL in adults; avoid ultrasound in adolescents) 1, 2

4. Exclude mimics (17-OHP for CAH, TSH for thyroid disease, prolactin for hyperprolactinemia, consider Cushing's workup if rapid-onset severe hyperandrogenism) 2, 3

5. Screen for metabolic complications (OGTT, lipid panel) in all confirmed PCOS patients 2, 5

Critical Pitfalls to Avoid

• Never rely on direct immunoassays for testosterone measurement - they have poor accuracy at female ranges and will miss cases or create false positives 1, 5
• Do not use ultrasound in adolescents <8 years post-menarche as primary diagnostic criterion - normal pubertal ovaries appear polycystic 2, 3
• Do not diagnose PCOS without excluding other causes of hyperandrogenism and menstrual irregularity 2, 3
• FAI becomes unreliable when SHBG <30 nmol/L - use alternative androgen markers 2
• Transabdominal ultrasound is inadequate for accurate follicle counting - use only for ovarian volume when transvaginal approach impossible 1