PCOS Panel Interpretation
Interpret a PCOS panel by first confirming at least two of three Rotterdam criteria: (1) oligo/anovulation (cycles >35 days), (2) clinical or biochemical hyperandrogenism, and (3) polycystic ovarian morphology on ultrasound (≥20 follicles per ovary or ovarian volume >10 mL), while excluding mimicking disorders through TSH, prolactin, and metabolic screening. 1, 2
Step 1: Assess Hyperandrogenism
Biochemical Markers (in order of diagnostic accuracy)
Total testosterone (TT) is the single best initial marker with 74% sensitivity and 86% specificity, measured via liquid chromatography-tandem mass spectrometry (LC-MS/MS) which shows superior specificity (92%) versus direct immunoassays (78%). 2
Calculated free testosterone (cFT) has the highest sensitivity at 89% with 83% specificity, calculated using the Vermeulen equation from high-quality TT and SHBG measurements. 2
Free androgen index (FAI) demonstrates 78% sensitivity and 85% specificity, but use cautiously when SHBG <30 nmol/L. 2
Androstenedione (A4) shows 75% sensitivity and 71% specificity—useful when SHBG is low or if TT/FT are normal but clinical suspicion remains high. 2
DHEAS has 75% sensitivity and 67% specificity, most reliable for adrenal androgen production, particularly valuable in women <30 years. 2
Clinical Hyperandrogenism
Look for hirsutism (Ferriman-Gallwey score), acne, androgenic alopecia, or clitoromegaly during physical examination. 1
Critical caveat: 30% of women with confirmed PCOS have normal testosterone levels, so clinical hyperandrogenism alone plus irregular cycles can establish diagnosis without abnormal labs. 2
Step 2: Confirm Ovulatory Dysfunction
Document menstrual history: cycle length >35 days suggests chronic anovulation. 1
Measure mid-luteal progesterone (days 21-23 of cycle): levels <6 nmol/L confirm anovulation. 2
LH/FSH ratio >2 (measured days 3-6 of cycle) suggests PCOS, but is abnormal in only 35-44% of cases—making it a poor standalone marker. 2
Step 3: Ultrasound Assessment (Age-Dependent)
Adults (≥18 years)
Use transvaginal ultrasound with ≥8 MHz transducer frequency for optimal resolution. 1, 3
Follicle number per ovary (FNPO) ≥20 follicles is the gold standard with 87.64% sensitivity and 93.74% specificity. 3, 1
Ovarian volume >10 mL serves as alternative when accurate follicle counting is impossible (81.48% sensitivity, 81.04% specificity). 3
Follicle number per single cross-section (FNPS) is a secondary alternative marker (81.07% sensitivity, 82.70% specificity). 3
Adolescents (<20 years, ≥1 year post-menarche)
Avoid ultrasound as first-line diagnostic tool due to poor specificity and high false-positive rates. 1
Rely on clinical and biochemical hyperandrogenism plus menstrual irregularity persisting 2-3 years beyond menarche. 1
If ultrasound is performed, ovarian volume shows 81.84% sensitivity and 83.54% specificity in this age group. 3
Step 4: Exclude Mimicking Disorders
Mandatory Exclusion Tests
TSH: Rule out thyroid disease causing menstrual irregularity. 2
Prolactin: Morning resting serum levels to exclude hyperprolactinemia (levels >20 μg/L abnormal). 2
17-hydroxyprogesterone: If DHEAS elevated, screen for non-classic congenital adrenal hyperplasia. 2
Clinical Red Flags Requiring Additional Workup
Cushing's syndrome: Buffalo hump, moon facies, hypertension, abdominal striae, central fat distribution, easy bruising, proximal myopathies. 1, 2
Androgen-secreting tumors: Rapid onset of symptoms, severe hirsutism, very high testosterone levels (>10.0 nmol/L androstenedione). 1, 2
Acromegaly: Coarse facial features, enlarged hands/feet. 1
Primary ovarian failure: Check FSH levels. 1
Step 5: Metabolic Screening (Mandatory for All PCOS Diagnoses)
Two-hour oral glucose tolerance test with 75g glucose load to screen for type 2 diabetes and glucose intolerance. 1, 2
Fasting lipid panel: Total cholesterol, LDL, HDL, triglycerides. 1, 2
BMI calculation and waist-hip ratio (WHR >0.9 indicates truncal obesity). 1, 2
Fasting glucose/insulin ratio: Ratio >4 suggests reduced insulin sensitivity. 2
Diagnostic Algorithm Summary
- First-line labs: TT or cFT via LC-MS/MS, TSH, prolactin
- If TT/cFT normal but high clinical suspicion: Add A4 and DHEAS
- Confirm ovulatory dysfunction: Menstrual history (cycles >35 days) or mid-luteal progesterone <6 nmol/L
- Ultrasound (adults only): FNPO ≥20 follicles (gold standard) or OV >10 mL (alternative)
- Metabolic screening: 2-hour OGTT, fasting lipids, BMI, waist-hip ratio
- Diagnosis confirmed: Any 2 of 3 Rotterdam criteria present after exclusion of mimics
Important Diagnostic Nuances
Rotterdam criteria sensitivity varies by geography: Studies using Rotterdam criteria show higher sensitivity (77-89%) but lower specificity (83%) compared to NIH criteria (sensitivity 51%, specificity 94%). 3
AMH is investigational only: While AMH ≥35 pmol/L shows 92% sensitivity and 97% specificity, it should not replace ultrasound or serve as a standalone diagnostic test due to lack of assay standardization. 3, 2
Phenotype documentation is essential: Clearly denote which of the four PCOS phenotypes the patient has, as each carries different long-term metabolic implications. 4