Dyslipidaemia is associated with insulin resistance in women with polycystic ovaries

OBJECTIVE Polycystic ovary syndrome (PCOS) is characterized by hyperinsulinaemia and insulin resistance. Previous reports of lipid abnormalities in the syndrome have produced conflicting results which may, in part, be related to the lack of appropriate controls for the obese women with PCOS. Only one study has related lipid levels to insulin sensitivity. The objective of this study was to assess lipids and lipoproteins in women with PCOS, to compare the results with weight matched controls, and to relate the findings to indices of insulin secretion and action, and to menstrual history. DESIGN A cross-sectional study of insulin sensitivity and lipids in a cohort of PCO subjects compared to weight and ethnic group matched controls. PATIENTS AND METHODS We have therefore investigated glucose tolerance, plasma lipids and lipoproteins In 19 lean (LP) and 55 obese (OF) patients with PCO and compared the results with those in 22 lean (LC) and 15 obese (OC) control women. Insulin sensitivity was measured in the same subjects with a short insulin (0.05 U/kg i.v. insulin) tolerance test (LP, n = 18; OP, n = 20; LC, n = 19; OC, n = 11). RESULTS Results are expressed as mean+/-SEM or median (interquartile range). Pasting plasma glucose levels were similar in the four groups but the plasma glucose area was higher after oral glucose (75 g) in both the lean and obese PCOS groups than in their controls (LC 32.4 +/- 0.7 vs LP 35.2 +/- 1.2, P < 0.01; OC 34.7 +/- 1.8 vs OP 37.8 +/- 1.5 mmol/l/3 h, P < 0.01). Insulin sensitivity was significantly reduced in obese PCOS women (LC 196 +/- 9 vs LP 179 +/- 9, NS; OC 168 +/- 12 vs OP 133 +/- 9 mmol/l/min, P < 0.01). Total serum cholesterol levels were similar in the lour groups but HDL(2)-cholesterol was reduced in both obese and lean PCOS (LC 0.42 (0.38-0.62), LP 0.31 (0.26-0.44), P < 0.05; OC 0.34 (0.21-0.47), OP 0.21 (0.12-0.32)mmol/l, P < 0.01). Total HDL-cholesterol was decreased significantly only in the obese PCOS group. Body mass index correlated significantly and negatively with total HDL-cholesterol and with HDL-cholesterol levels both within the PCOS group and the control women. Using multiple regression insulin insensitivity contributes significantly beyond BMI to the low HDL-cholesterol in women with polycystic ovaries. CONCLUSION Polycystic ovary syndrome is associated with biochemical risk factors for premature vascular disease, which cannot be explained by obesity alone.