Polycystic Ovary Syndrome - Diagnosis

Diagnosis

In 1990, a consensus workshop sponsored by the NIH/NICHD suggested that a patient has PCOS if she has all of the following:

1. oligoovulation
2. signs of androgen excess (clinical or biochemical). Androgen excess can be tested by measuring total and free testosterone levels. Other androgens, such as DHEA-S, may be normal or slightly above the normal range in patients with polycystic ovarian syndrome (PCOS) while levels of sex hormone–binding globulin (SHBG) are usually low in patients with PCOS. Androstenedione levels are also elevated in women with PCOS. This androgen precursor is 60% ovarian and 40% adrenal in derivation.
3. other entities are excluded that would cause polycystic ovaries

More recently, in 2003, a consensus workshop sponsored by ESHRE/ASRM in Rotterdam indicated PCOS to be present if any 2 out of the following 3 criteria are met and other entities are excluded that would cause these.

1. oligoovulation and/or anovulation
2. excess androgen activity
3. polycystic ovaries (by gynecologic ultrasound). According to available literature, at least one of the following criteria should be present to establish polycystic ovaries: either 12 or more follicles measuring 2–9 mm in diameter, or increased ovarian volume (>10 cm3). Not all women with PCOS have polycystic ovaries (PCO), nor do all women with ovarian cysts have PCOS; although a pelvic ultrasound is a major diagnostic tool, it is not the only one.

The initial NIH diagnostic criteria based on oligomenorrhoea/amenorrhoea and clinical or biochemical hyperandrogenism have been broadened in the 2003 Rotterdam diagnostic criteria to include polycystic ovaries (PCO) at ultrasound as a key diagnostic criteria. A total of 25% of young women have PCO on ultrasound and the inclusion of PCO in diagnostic criteria has increased the prevalence of PCOS. Earlier studies conducted in Greece, Spain and USA using the NIH criteria estimated the prevalence of PCOS at 4% to 8%. However, the first community-based prevalence study based on current Rotterdam diagnostic criteria conducted recently has shown the prevalence of PCOS to be 18%. The Rotterdam definition is wider, including many more patients, most notably patients without androgen excess. Critics say that findings obtained from the study of patients with androgen excess cannot necessarily be extrapolated to patients without androgen excess.

In 2006, the Androgen Excess & PCOS Society suggested a tightening of the diagnostic criteria to all of:

1. excess androgen activity
2. oligoovulation/anovulation and/or polycystic ovaries
3. other entities are excluded that would cause excess androgen activity

A recent study found that a questionnaire addressing the history of menstrual pattern, obesity and hirsutism can diagnose PCOS, according to a clinical prediction rule, with a sensitivity of 77.1% (95% confidence interval 62.7%–88.0%) and a specificity of 93.8% (95% CI 82.8%–98.7%). Some other blood tests are suggestive but not diagnostic. The ratio of LH (luteinizing hormone) to FSH (follicle stimulating hormone), when measured in international units, is greater than 1:1 (sometimes more than 3:1), as tested on Day 3 of the menstrual cycle. The pattern is not very specific and was present in less than 50% in one study.