The independent effects of hyperandrogenaemia, hyperinsulinaemia, and obesity on lipid and lipoprotein profiles in women.

We performed this study to clarify the independent effects of hyperandrogenaemia, hyperinsulinaemia, and obesity on lipid and lipoprotein levels in women with hyperandrogenaemia (HA) and anovulation which we designated as the polycystic ovary syndrome (PCO). We examined fasting lipid, lipoprotein, sex hormone and insulin levels in 38 women (21 obese (ob), 17 non-obese (nob] with HA and anovulation (PCO) and 38 normal ovulatory women (21 obese, 17 non-obese), matched for age and weight. The women with PCO had significantly increased androgen levels compared to the normal women. However, total oestradiol levels were similar in the PCO and normal women. Mean fasting insulin levels and 2-h glucose levels (both P less than 0.001) were significantly higher in ob PCO women. There were significant decreases (P less than or equal to 0.01) in high-density lipoprotein (HDL) levels in both the obese groups (ob PCO and ob normal) compared to the non-obese (nob PCO and nob normal) groups. Otherwise, mean lipid and lipoprotein levels did not differ in the ob or the nob PCO women compared to the control groups. The correlations between sex hormone, lipid and lipoprotein levels differed in the four groups of women. After statistical adjustment for potential hormonal interactions, nob PCO women had significant positive correlations between testosterone and LDL levels (R = 0.51, P less than 0.05) and insulin and TTG levels (R = 0.61, P less than 0.01). Ob normal women had a significant positive correlation between oestrone and TTG levels (R = 0.44, P less than or equal to 0.05). We conclude that (1) PCO women are in a low to risk for CVD primarily because of the increased prevalence of obesity rather than the reproductive hormone abnormalities associated with this disorder. However, by their lipid profiles, the PCO women were still in a low to intermediate risk group for CVD.