Polycystic Ovaries Are Common in Women With Hyperandrogenic Chronic Anovulation But Do Not Predict Metabolic or Reproductive Phenotype

Abstract

Originally viewed as having only reproductive consequences, polycystic ovary syndrome (PCOS) now is recognized as frequently being associated with insulin resistance, and that the syndrome is a multisystem endocrinopathy characterized by hyperandrogenism and chronic anovulation. This study was done to determine whether the size and morphology of the ovaries in women with PCOS are associated with markers of insulin sensitivity. Participants were 88 women with PCOS and 21 control women ranging in age from 17 to 45 years. All were in the early follicular phase or its equivalent (no follicles more than 10 mm in diameter; a serum progesterone below 3 ng/mL). Blood samples were taken and transvaginal ultrasound and a 2-hour glucose tolerance test done on the same day. At least one ovary exceeded 10 cm 3 in size (PCOV), and/or there was polycystic ovary morphology (PCOM, at least 10 peripheral follicular cysts 8 mm or less in diameter as well as increased central ovarian stroma) in 84% of women with PCOS and in 48% of the control group. PCOM discriminated between the study and control women better than did PCOV. Women with PCOS were 50 times more likely to have POM compared with control women (odds ratio [OR], 50). The chance of PCOV being present was increased 5-fold in women with PCOS (OR, 4.6). There were no substantial differences between groups in fasting or 2-hour blood glucose values on tolerance testing, the insulin sensitivity index, or integrated measures of glucose and insulin. Associations with circulating androgen levels also were lacking. Women with PCOS and PCOM had lower levels of follicle-stimulating hormone (FSH) than those with PCOS and non-PCOM. In women with PCOS and PCOV, the luteinizing hormone (LH)-to-FSH ratio was higher than in women without PCOV and PCOS. Disregarding morphology, women with PCOS and control women exhibited no significant correlation between total ovarian volume and parameters of glucose metabolism or levels of reproductive hormones (except FSH). There also were no significant correlations between the PCOS and control groups for total ovarian volume, any aspect of glucose metabolism, or levels of reproductive hormones. These findings suggest that, in women with endocrine findings of PCOS, the size and mo'rphology of the ovaries are of little help in identifying specific metabolic or reproductive abnormalities. Routine ovarian ultrasonography may not be necessary in women having hyperandrogenic chronic anovulation.