How PMOS (PCOS) Is Diagnosed in 2026: Criteria, Tests, and What You Actually Need

What changed in the 2023 guideline

The 2023 International Evidence-Based Guideline retained the Rotterdam framework but made three substantive changes that many patients — and some clinicians — have not yet incorporated.

1. AMH is now an accepted alternative to ultrasound. Serum anti-Müllerian hormone may be used in place of pelvic ultrasound to establish polycystic ovarian morphology in adults. AMH is less invasive, more accessible, and less expensive than transvaginal ultrasonography — a meaningful advantage in Indian practice, where transvaginal scanning is often declined by unmarried patients and where scan quality varies widely.

2. The diagnostic pathway is simplified when two clinical criteria are met. Where irregular cycles and hyperandrogenism coexist in an adult, imaging and AMH are unnecessary. This is intended to reduce cost, delay and over-investigation.

3. Adolescents are handled entirely differently. In adolescents, both hyperandrogenism and ovulatory dysfunction are required, and neither ultrasound nor AMH is recommended — their specificity is too poor in the years after menarche. This is covered in detail in our separate article on adolescent diagnosis.

---

Step 1: Establishing ovulatory dysfunction

Irregular cycles are defined by time since menarche — this precision matters, because applying adult definitions to a recently menstruating teenager is a common source of overdiagnosis.

Anovulation may also be confirmed biochemically with a mid-luteal progesterone (day 21 of a 28-day cycle, or 7 days before the expected period). A level below approximately 3–5 ng/mL indicates an anovulatory cycle. This is a useful, cheap test that is under-used in practice — particularly in patients who report apparently regular cycles but are not conceiving, since regular bleeding does not guarantee ovulation.

---

Step 2: Establishing hyperandrogenism

Clinical hyperandrogenism is assessed by examination, not by patient report alone:

• Hirsutism — terminal hair in androgen-dependent areas, scored with the modified Ferriman-Gallwey (mFG) system across 9 body regions. Thresholds are ethnicity-dependent. For South Asian women a cut-off of approximately ≥ 4–6 is generally applied, rather than the ≥ 8 historically used in European populations. Note that hair removal prior to consultation frequently masks the finding — ask directly.
• Androgenic alopecia — Ludwig pattern crown thinning with preserved frontal hairline.
• Acne — inflammatory acne, particularly persisting into adulthood or distributed along the jawline, lower face and neck.

Isolated mild acne alone is weak evidence of hyperandrogenism. Hirsutism is the most specific clinical marker.

Biochemical hyperandrogenism requires the right test:

• Calculated free testosterone or free androgen index (FAI) is preferred. Total testosterone alone is insensitive, because SHBG is suppressed by hyperinsulinaemia — so a patient may have a normal total testosterone with a clearly raised free fraction.
• FAI = (total testosterone nmol/L ÷ SHBG nmol/L) × 100
• Direct free testosterone immunoassays are unreliable and should not be used for diagnosis. Equilibrium dialysis or LC-MS/MS is the reference standard; calculated free testosterone from total testosterone and SHBG is the practical alternative.
• Timing: measure in the early follicular phase where possible, in the morning.
• Combined oral contraceptives suppress androgens. If the patient is on a COC, biochemical assessment requires a washout of at least 3 months. This is a very common cause of falsely reassuring results.

---

Step 3: Ovarian morphology or AMH (only if needed)

Ultrasound criteria (adults, ≥ 8 years post-menarche only):

• Using modern transducers (≥ 8 MHz): ≥ 20 follicles of 2–9 mm in either ovary, and/or ovarian volume ≥ 10 mL with no corpus luteum, cyst or dominant follicle present.
• Transvaginal is preferred. If the transabdominal route is used — which is frequently the case in Indian practice for unmarried patients — ovarian volume ≥ 10 mL is the appropriate criterion, as follicle counting is unreliable transabdominally.

AMH may be used instead. Its principal caveats:

• Assay-specific cut-offs — there is no single universal threshold, and results are not interchangeable between platforms. Use your laboratory's stated reference range.
• AMH declines with age; interpret against age-appropriate ranges.
• AMH is not valid for diagnosis in adolescents.
• AMH is a marker of ovarian reserve, not a fertility prediction. A high AMH in PMOS does not mean high fertility.

---

Step 4: Exclusion — the step most often skipped

PMOS is a diagnosis of exclusion. All patients require, at minimum:

Escalate further if red flags are present. Rapid-onset virilisation, clitoromegaly, deepening voice, or total testosterone > 5 nmol/L (approximately > 150 ng/dL) suggests an androgen-secreting tumour and requires urgent imaging of ovaries and adrenals — not a PMOS label. Similarly, features of Cushing's syndrome (violaceous striae, proximal myopathy, easy bruising, dorsocervical fat pad) warrant an overnight dexamethasone suppression test or 24-hour urinary free cortisol.

Non-classic CAH deserves particular attention in the Indian population: it is more prevalent in some South Asian and Middle Eastern groups than in European populations, it is clinically indistinguishable from PMOS on presentation, and it has entirely different implications for genetic counselling and for management in pregnancy. A morning follicular-phase 17-OHP should be routine, not optional.

---

Step 5: Metabolic assessment — required, not optional

Once diagnosed, every patient needs baseline metabolic assessment, regardless of BMI:

• 75 g oral glucose tolerance test (OGTT) — this is the recommended screen. Fasting glucose and HbA1c alone will miss a substantial proportion of impaired glucose tolerance in this population. Repeat every 1–3 years depending on risk.
• Fasting lipid profile
• Blood pressure — at every visit
• BMI and waist circumference, using Asian-specific cut-offs: overweight ≥ 23 kg/m², obese ≥ 25 kg/m²; waist circumference ≥ 80 cm in women
• Screening for depression and anxiety — the guideline explicitly recommends this. In a global survey of ethnic Indian women with the condition, 64% reported at least one co-morbidity, with anxiety and depression most common.
• Sleep apnoea screening where symptoms suggest it

---

For gynaecologists: the four most common diagnostic errors

1. Diagnosing on ultrasound alone. PCOM is present in up to 20–25% of asymptomatic women of reproductive age. Ultrasound appearance in isolation is not a diagnosis. This is the single most frequent error in routine practice.

2. Excluding the diagnosis because the scan is normal. The mirror image of the above, and equally common. Where oligomenorrhoea and hirsutism coexist, no imaging is required and a normal scan is irrelevant.

3. Relying on total testosterone. With SHBG suppressed by hyperinsulinaemia, total testosterone is frequently normal in genuinely hyperandrogenic patients. Always calculate FAI.

4. Omitting the OGTT in lean patients. Insulin resistance in PMOS is independent of BMI, and South Asian patients develop it at lower BMI thresholds. Lean patients are systematically under-screened.

A fifth, more subtle error: the LH:FSH ratio is not a diagnostic criterion. It is elevated in perhaps 50–60% of patients and is not part of any current guideline. It should not be used to confirm or exclude the diagnosis, though it remains widely requested.