Prevalence, phenotype and cardiometabolic risk of polycystic ovary syndrome under different diagnostic criteria

What is the prevalence, phenotype and metabolic features of polycystic ovary syndrome (PCOS) in the same population according to three different diagnostic criteria? The prevalence of PCOS under National Institutes of Health (NIH), Rotterdam and Androgen Excess and PCOS (AE-PCOS) Society criteria was 6.1, 19.9 and 15.3, respectively. PCOS carried a 2-fold increased risk of metabolic syndrome regardless of the diagnostic criteria used. The prevalence rates of PCOS differ depending on the diagnostic criteria used to define the syndrome. The current paper gives the prevalence rates of the component and composite phenotypes of PCOS in the same population and reports similar rates of metabolic syndrome in women with PCOS under contrasting diagnostic criteria. In this cross-sectional study, 392 women between the ages of 18 and 45 years were analyzed. When the prevalence of PCOS according to NIH was set to 8 with a precision of 2.2 and confidence interval of 95, the sample size required for a prevalence survey was found to be 400 subjects. The study was carried out in the General Directorate of Mineral Research and Exploration, a government-based institute, in which the largest number of female staff (n 527) are employed within a single institute in Ankara, Turkey. The study was performed between 7 December 2009 and 30 April 2010. All female subjects between the ages of 18 and 45 years were invited to participate. Women older than 45 or younger than 18 years, post-menopausal women, women with a history of hysterectomy or bilateral oopherectomy and pregnant women were excluded. Totally, 392 of the employees were recruited for the final analyses. The prevalence of PCOS under NIH, Rotterdam and AE-PCOS Society criteria were 6.1, 19.9 and 15.3, respectively. While the prevalence of metabolic syndrome was 6.1 in the whole study group, within the patients diagnosed as PCOS according to NIH, Rotterdam and AE-PCOS Society criteria, it was 12.5, 10.3 and 10.0, respectively. Even though we have included women working at a single institution with a high response rate for the participation, we cannot exclude potential selection bias due to undetermined differences between our sample and background community. We might have underestimated actual prevalence of metabolic syndrome in PCOS due to lack of oral glucose tolerance test 2 h glucose data. Current results can be generalized to Caucasian populations and may present variations in other populations according to race and ethnicity. This work was, in part, sponsored by Merck Serono. Not applicable.