Adiponectin is independently associated with insulin sensitivity in women with polycystic ovary syndrome

Summary objective The polycystic ovary syndrome (PCOS) is associated with obesity and insulin resistance predisposing to diabetes mellitus type 2 and atherosclerosis. Adiponectin is a recently discovered adipocytokine with insulin‐sensitizing and putative antiatherosclerotic properties. The aim of the study was to elucidate determinants of circulating adiponectin levels and to investigate the potential role of adiponectin in insulin resistance in PCOS women. patients and measurements Plasma adiponectin and parameters of obesity, insulin resistance and hyperandrogenism were measured In 62 women with PCOS and in 35 healthy female controls. results Both in PCOS and controls, adiponectin levels were lower in overweight or obese women than in normal‐weight women, without any difference between PCOS and controls after adjustment for body mass index (BMI). In PCOS and in controls there was a significant correlation of adiponectin with BMI ( r = −0·516, P < 0·001), fasting insulin ( r = −0·404, P < 0·001), homeostasis model sensitivity (HOMA %S) ( r = −0·424, P < 0·001) and testosterone ( r = −0·279, P < 0·01), but no correlation with androstenedione ( r = −0·112, P = 0·325), 17‐OH‐progesterone ( r =−0·031, P = 0·784) or the LH/FSH ratio ( r =−0·033, P = 0·753). Multiple linear regression analysis revealed that BMI and HOMA %S but not testosterone were independently associated with adiponectin plasma levels, explaining 16% (BMI) and 13% (HOMA %S) of the variability of adiponectin, respectively. In PCOS patients insulin sensitivity, as indicated by continuous infusion of glucose with model assessment (CIGMA %S) was significantly correlated with adiponectin ( r = 0·55; P < 0·001), BMI ( r =−0·575; P < 0·001), waist‐to‐hip ratio (WHR) ( r =−0·48; P = 0·001), body fat mass assessed by dual‐energy X‐ray‐absorptiometry (DEXA) [Dexa‐fat (total) ( r = −0·61; P < 0·001) and Dexa‐fat (trunk) ( r = −0·59; P < 0·001)] and with testosterone ( r = −0·42; P = 0·001). Multiple linear regression analysis demonstrated that markers of obesity such as BMI, total or truncal fat mass, age and adiponectin were independently associated with CIGMA %S, and that circulating adiponectin accounted for about 18% of the degree of insulin resistance in PCOS. By contrast, testosterone was not a significant factor, suggesting that PCOS per se did not affect insulin sensitivity independent from obesity, age and adiponectin. Metformin treatment for 6 months in insulin‐resistant PCOS women ( n = 9) had no effect on plasma adiponectin ( P = 0·59) despite significant loss of weight and fat mass and improvement in hyperandrogenaemia. conclusions PCOS per se is not associated with decreased levels of plasma adiponectin. However, circulating adiponectin is independently associated with the degree of insulin resistance in PCOS women and may contribute to the development and/or maintenance of insulin resistance independent from adiposity.