Is hormone level checking recommended for a female patient of reproductive age with Polycystic Ovary Syndrome (PCOS)?

Hormone Testing in PCOS: Evidence-Based Recommendations

Yes, hormone level checking is strongly recommended for diagnosing PCOS in reproductive-age women, with total testosterone and free testosterone serving as first-line laboratory tests to assess biochemical hyperandrogenism. 1

Primary Hormonal Tests to Order

First-Line Androgen Testing

• Total testosterone (TT) via LC-MS/MS is the single best initial biochemical marker, demonstrating 74% sensitivity and 86% specificity for PCOS diagnosis 1
• Calculated free testosterone (cFT) provides the highest diagnostic accuracy with 89% sensitivity and 83% specificity, and should be calculated using equilibrium dialysis or ammonium sulfate precipitation methods, or via the Vermeulen equation from high-quality TT and SHBG measurements 1
• Free androgen index (FAI) offers reasonable performance at 78% sensitivity and 85% specificity, though caution is warranted when SHBG <30 nmol/L 1
• Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is the preferred measurement method over direct immunoassay, providing superior specificity (92% vs 78%) and sensitivity (71% vs 74%) 1

Second-Line Androgen Testing

• Androstenedione (A4) can be considered when TT or cFT are not elevated, with 75% sensitivity and 71% specificity, particularly useful when SHBG is low 1
• DHEAS measurement has 75% sensitivity and 67% specificity, most reliable for assessing adrenal androgen production and particularly valuable in women <30 years 1

Gonadotropin Assessment

• LH and FSH levels should be measured, as elevated LH or an elevated LH/FSH ratio is characteristic of PCOS, though this finding is present in only 35-44% of cases 1, 2
• The LH/FSH ratio alone should not be used as a diagnostic criterion due to its low sensitivity, despite being commonly referenced 2

Essential Exclusionary Hormone Tests

Before confirming PCOS, the following must be checked to exclude other conditions:

• 17-hydroxyprogesterone to rule out non-classic congenital adrenal hyperplasia 3, 4
• TSH to exclude thyroid disease 3, 5
• Prolactin to rule out hyperprolactinemia 1, 3
• Morning cortisol or 24-hour urinary free cortisol if Cushing's syndrome is suspected based on clinical features (buffalo hump, moon facies, hypertension, abdominal striae) 3

Timing and Context Considerations

Hormonal Contraception Interference

• All hormonal contraception must be discontinued before testing, as progestin-containing methods (including Implanon) suppress the hypothalamic-pituitary-ovarian axis, making LH/FSH ratios unreliable and altering testosterone levels through SHBG changes 6
• Allow adequate washout period after contraceptive discontinuation before obtaining hormone levels 6

Menstrual Cycle Timing

• Hormone measurements should ideally be obtained in the early follicular phase (days 2-5) of the menstrual cycle in women with some menstrual activity 2
• In women with amenorrhea, testing can be performed at any time after excluding pregnancy 1

Additional Metabolic Screening

Beyond reproductive hormones, all women with PCOS require:

• Fasting glucose and HbA1c or oral glucose tolerance test to screen for type 2 diabetes and glucose intolerance 3, 7
• Fasting lipid panel for dyslipidemia screening 3, 7
• Fasting insulin levels may be considered to assess insulin resistance, though not required for diagnosis 1, 8

Anti-Müllerian Hormone (AMH): Current Status

• AMH is NOT recommended for clinical diagnosis of PCOS despite being elevated in PCOS patients, due to lack of assay standardization, absence of validated cut-offs, significant overlap between women with and without PCOS, and age-dependent variability 1, 3, 6
• AMH remains investigational and should not replace established diagnostic criteria 1, 3

Common Pitfalls to Avoid

• Do not rely on direct immunoassay methods for free testosterone, as these are highly inaccurate due to low serum concentrations; always use calculated free testosterone instead 1
• Do not use LH/FSH ratio as the sole biochemical criterion, as it has poor sensitivity (only 41-44% abnormal in confirmed PCOS) 2
• Do not order hormone panels while hormonal contraception is in place, as results will be unreliable and misleading 6
• Do not assume normal testosterone excludes PCOS—only 70% of PCOS patients have elevated total testosterone, so consider measuring A4 and DHEAS if clinical suspicion remains high 2
• Do not forget that PCOS is a clinical diagnosis requiring at least 2 of 3 Rotterdam criteria (hyperandrogenism, ovulatory dysfunction, polycystic ovarian morphology), not just abnormal hormone levels alone 1, 3

Practical Algorithm for Hormone Testing

1. Confirm patient is not on hormonal contraception; if present, discontinue and wait for washout 6
2. Order first-line tests: Total testosterone (via LC-MS/MS if available), SHBG to calculate free testosterone, LH, FSH 1, 2
3. Order exclusionary tests: TSH, prolactin, 17-hydroxyprogesterone 3, 4
4. If first-line androgens are normal but clinical suspicion high: Add androstenedione and DHEAS 1
5. Screen for metabolic complications: Fasting glucose/HbA1c, lipid panel 3, 7
6. Interpret results in context of clinical features and ultrasound findings to apply Rotterdam criteria 1, 3

The combination of testosterone, androstenedione, or LH (either alone or in combination) identifies 86% of women with PCOS when typical ovarian ultrasound appearances plus clinical features are present 2. This comprehensive hormonal assessment, combined with clinical evaluation and imaging, provides the most accurate diagnostic approach for PCOS in reproductive-age women.