Dyslipidaemia in polycystic ovarian syndrome: different groups, different aetiologies?

The objective was to study the pathophysiology of the dyslipidaemia in polycystic ovarian syndrome (PCOS) patients, and to determine how it is related to hyperinsulinaemia, hyperandrogenism and dehydroepiandrosterone sulphate (DHEA-S) concentrations. The lipoprotein lipid profile, anthropometric measurements, endocrine profile and the presence of insulin resistance were evaluated in 31 PCOS patients and 20 age-matched healthy women, who served as controls. PCOS patients had higher fasting insulin concentrations, higher body mass indexes (BMI) and were hyperlipidaemic, with higher total cholesterol, low density lipoprotein (LDL) and triglyceride (TG) concentrations. There were no relationships between plasma lipids and anthropometric variables in the patient group as a whole. Insulin-resistant (IR) and non-IR (NIR) PCOS patients were then evaluated separately. Obesity with marked hyperandrogenism were the predominant features in patients with IR. NIR patients were not obese and had significantly less hyperandrogenism. The adrenal androgen DHEA-S was at the upper limit of its normal range in both groups. However, both PCOS subgroups exhibited similar significant abnormalities in terms of their lipid parameters. Insulin and DHEA-S concentrations were positively correlated with total cholesterol, LDL and TG, and negatively correlated with high density lipoprotein, in IR patients. In NIR subjects, insulin was not correlated with any of the lipids and DHEA-S was negatively related to cholesterol and LDL. Anthropometric variables were related to lipids in only the NIR patients. Thus PCOS subjects as a group exhibit dyslipidaemia, characterized by increased total cholesterol, LDL and TG concentrations. When divided into IR and NIR subjects, there were no differences in the degree of lipid abnormalities, despite significant variations in the BMI and androgen status. Thus, in PCOS subjects, dyslipidaemia may occur irrespective of insulin resistance. Insulin and DHEA-S concentrations were positively correlated with an atherogenic lipid profile in the IR group only. As distinct from syndrome X when IR was present, dyslipidaemia was not related to body weight or the waist:hip ratio. In the NIR group there was no relationship between lipids and insulin; DHEA-S, on the other hand, was negatively related to cholesterol and LDL concentrations. Thus, dyslipidaemia in PCOS patients may occur irrespective of insulin resistance, and may have different metabolic aetiologies depending on DHEA-S metabolism. It remains to be seen whether the two types of PCOS are associated with different risks for ischaemic heart disease.