Other Diagnostic Criteria

Why multiple criteria systems?

PCOS is a multifaceted syndrome. Over the years, different medical societies have proposed their own diagnostic criteria. These systems do not all give the same answer to the question "who has PCOS?" — which has real consequences for patients who do or do not receive a diagnosis.

NIH 1990 — The most restrictive

The NIH (National Institutes of Health, 1990) criteria were the first to be formalised. They require the simultaneous presence of:

• Hyperandrogenism (clinical or biochemical) — mandatory
• Oligo-anovulation — mandatory

These are the most restrictive criteria: they automatically exclude phenotypes C (ovulatory) and D (normo-androgenic). Approximately 30 to 40% of individuals who meet the Rotterdam criteria do not meet the NIH criteria. These criteria remain used in certain studies for comparing homogeneous populations.

AES 2006 — Hyperandrogenism as a mandatory criterion

The Androgen Excess Society (AES, 2006) criteria require:

• Hyperandrogenism (clinical or biochemical) — mandatory
• At least one of: oligo-anovulation OR PCOM

This system recognises phenotypes A, B and C, but excludes phenotype D (normo-androgenic). The AES position is that PCOS is primarily a disease of androgen excess — a view contested by studies showing that phenotype D also exhibits metabolic abnormalities.

PCOS-SOC 2013 — Adoption of Rotterdam

The PCOS-SOC (Androgen Excess and PCOS Society, 2013) revised its position and adopted the Rotterdam criteria as the standard, while producing practical recommendations for screening and treating metabolic complications. This was an important step towards international harmonisation.

ESHRE 2023 — The most comprehensive current recommendation

The ESHRE 2023 guidelines (Evidence-based Guideline for the Assessment and Management of Polycystic Ovary Syndrome) represent the most recent and most comprehensive update. They:

• Confirm the Rotterdam criteria as the international standard
• Update the PCOM threshold (AFC ≥ 20 follicles/ovary)
• Introduce AMH as a potential alternative to the ultrasound criterion
• Expand recommendations beyond diagnosis: metabolic workup, mental health, quality of life

What this means for patients

The choice of criteria directly determines who gets diagnosed. Under the NIH 1990 criteria, a person with phenotype D (irregular cycles + PCOM but no hyperandrogenism) will not receive a diagnosis. Under Rotterdam, they will — and can access appropriate monitoring and care. This is why Rotterdam is considered more inclusive and closer to clinical reality.

Key takeaways

• NIH 1990: HA + OA mandatory — excludes phenotypes C and D
• AES 2006: HA mandatory + (OA or PCOM) — excludes phenotype D
• Rotterdam 2003: 2/3 criteria — includes all phenotypes
• PCOS-SOC 2013: adopts Rotterdam
• ESHRE 2023: current recommendation, Rotterdam + updates