Hyperadiponectinaemia in anorexia nervosa

OBJECTIVE Adiponectin (ApN) is a fat-derived hormone that enhances insulin sensitivity, controls body weight, prevents atherosclerosis and negatively regulates haematopoiesis and immune functions. In contrast to many proteins secreted by adipose tissue, the circulating level of ApN falls in obesity and insulin resistance states. The influence of starvation-induced depletion of fat stores on ApN concentrations is yet unknown. We therefore investigated plasma ApN in anorexia nervosa (AN). PATIENTS AND DESIGN We measured plasma ApN in 26 female anorectic patients and examined its relationships to several anthropometric or metabolic parameters. Twenty-four age-matched healthy female controls (C) were also studied. RESULTS Body mass index (BMI) and fat mass were markedly decreased in AN. However, plasma ApN levels were 30% higher in anorectic than in control subjects (P < 0.01), while a reverse pattern was observed for leptin concentrations. When normalized for fatness, ApN values almost doubled in AN. ApN levels were negatively correlated with BMI and fat mass (P < 0.05 in the combined population, AN + C). Insulin sensitivity tended to be 40% higher in AN (n = 7) than in C (n = 12) subjects, and plasma ApN levels were positively correlated with insulin sensitivity (P < 0.05 in AN + C subgroups). Total and low density lipoprotein (LDL)-cholesterol were higher, or tended to be higher, in AN, but there were no correlations between plasma ApN and plasma lipids. By contrast, ApN was related to the lipid profile, in a manner consistent with its antiatherogenic role, in healthy controls [i.e. negatively correlated with triglycerides, total and LDL-cholesterol and total/high density lipoprotein (HDL) cholesterol; P < 0.05 or less for each parameter]. In a multiple regression analysis, BMI and insulin sensitivity in AN were independent determinants for ApN levels, explaining up to approximately 80% of the variance in this measure. CONCLUSIONS Plasma adiponectin levels are increased in anorexia nervosa. This may, at least in part, be due to the lack of negative feedback exerted by fat mass on adiponectin production and/or to enhanced insulin sensitivity. We speculate that hyperadiponectinaemia could, in turn, contribute to maintain a state of enhanced insulin sensitivity and possibly exacerbate haematological and infectious complications of anorexia nervosa.