Insulin sensitivity in anorexia nervosa: a mirror image of obesity?

Although, in many respects and from a metabolic point of view, obesity and AN are clearly two opposite pathological conditions, the available data concerning insulin sensitivity in these two syndromes are not so obviously opposite. Indeed, whereas everybody is convinced that obesity is characterized by an increased insulin resistance, the papers reporting insulin sensitivity parameters in AN contain some apparently contradictory results. The observations of simultaneously low fasting blood glucose and plasma-insulin levels in anorectic patients could suggest increased insulin sensitivity in AN. However, if this is the case, it would be present despite other metabolic and hormonal changes (increased plasma concentrations of free fatty acids, cortisol, and growth hormone) which are known factors of insulin resistance. During an oral glucose-tolerance test, an impaired glucose-tolerance occurring despite sustained insulin response to glucose is usually found in anorectic patients before treatment; these abnormalities are, at least partially, reversed after successful refeeding. From these results, such conclusive, if indirect, evidence exists for relative insulin insensitivity in untreated AN. Similar results were initially reported with the intravenous glucose-tolerance test. Typically, the coefficient of glucose assimilation K was reduced in anorectic patients before treatment and increased after realimentation. This seemed to occur despite a relative increase in insulin response to glucose, which again may be related to insulin resistance in these undernourished subjects. However, more recent data demonstrated that the early insulin response is significantly lower in anorectic patients than in controls and that more than half of these patients have normal glucose-tolerance despite decreased peripheral plasma insulin levels. These latter observations, on the contrary, suggest an increased insulin sensitivity, at least in some patients with AN. Only the recently developed minimal model method allows us to discriminate between changes in insulin secretion and action after intravenous glucose injection and thus to infer accurately the sensitivity of the tissues to insulin. Unfortunately, this technique has not been applied to anorectic patients, until now, to solve the controversy. The simplest way to assess the action in vivo of insulin is to perform an intravenous insulin-tolerance test. However, the initial findings with this test, which showed exaggerated fall in plasma-glucose values and delayed return to basal levels after intravenous injection of insulin in AN, do not necessarily mean increased insulin sensitivity in these self-starved patients.(ABSTRACT TRUNCATED AT 400 WORDS)