Proposed criteria for the identification of polycystic ovary syndrome following menopause: An ancillary study of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).

To propose plausible criteria with which to identify menopausal women with PCOS. A cross-sectional study involving the baseline data of 713 menopausal women at admission to the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) in Salvador, Bahia, Brazil. PCOS was identified by the presence of two of three criteria. (1) A history of amenorrhea or oligomenorrhea (OL) (regular intermenstrual intervals ≥35 days during reproductive life); (2) clinical or biochemical hyperandrogenism (HA), identified by a score ≥5 points in a hirsutism questionnaire constructed and validated for women in this age group, or total or free testosterone ≥ the 95th percentile for women considered normal; (3) insulin resistance (IR) (a homeostatic model assessment [HOMA] index≥2.2). Validation was performed using probable epidemiological endpoints. According to these criteria, 7.6% of the women in the sample had PCOS. Of these, 7.4% had HA and OL, 72.2% had HA and IR, 14.8% had OL and IR and 5.6%, had HA, OL and IR. Women with PCOS were younger, had had fewer pregnancies and entered menopause earlier. Positive associations were found between PCOS and overweight (PR: 1.31; 95%CI: 1.18-1.46), obesity (1.44; 1.01-2.06), carbohydrate metabolism disorders (impaired glucose tolerance and diabetes mellitus) (1.30; 1.03-1.65), and with diabetes alone (1.41; 0.83-2.39), although this latter association failed to reach statistical significance. The women selected in accordance with these criteria had the characteristics of PCOS that are not only expected, but also widely associated with this disorder.