RELATIONSHIPS BETWEEN METABOLIC-CLEARANCE RATE OF INSULIN AND BODY-MASS INDEX IN A FEMALE-POPULATION RANGING FROM ANOREXIA-NERVOSA TO SEVERE OBESITY

Changes in the metabolic clearance rate of insulin (MCR(I)) have been described in several pathological conditions. Conflicting data suggest that they may be related to either body mass index (BMI) or body comp osition. This study aimed to investigate the relationship between the MCR(I) and BMI in an exclusively female population showing a wide rang e of BMI. For that purpose, hyperinsulinemic normoglycemic glucose cla mps were performed in nine anorectic subjects (BMI: 14.5 +/- 0.8 kg/m2 ), 11 healthy volunteers (BMI: 20.3 +/- 0.5 kg/m2) and 12 obese patien ts (BMI: 33.0 +/- 0.9 kg/m2). To exclude any influence of the menstrua l cycle on the MCR(I), five healthy women underwent three tests at dif ferent days of the menstrual cycle: menstruation period, late follicul ar pre-ovulatory phase and luteal phase, in random order. The MCR(I), which was quite reproducible in a given subject, was not significantly modified by the menstrual cycle. In the premenopausal female populati on studied, the mean (+/- s.e.m.) MCR(I) normalized for body weight (k g) were 35.4 +/- 3.4, 24.7 +/- 1.8 and 14.0 +/- 1.0 ml/kg/min (P < 0.0 1) for anorectic subjects, healthy volunteers and obese patients, resp ectively. These differences were maintained when the MCR(I) was normal ized according to corporeal surface (m2) (1018 +/- 75, 859 +/- 67, 638 +/- 40 ml/m2/min, P < 0.01) or lean body mass (kg) (37.1 +/- 3.4, 32. 6 +/- 2.7 and 24.1 +/- 0.5 ml/kg(LBM)/min, P < 0.01), but disappeared when MCR(I) was expressed per kg of ideal body weight (24.6 +/- 2.2, 2 4.6 +/- 2.1 and 22.4 +/- 1.4 ml/kg(IBW)/min, n.s.). There was a signif icant negative correlation between MCR(I) and BMI whatever the mode of expression of MCR(I), i.e. body weight (kg) (r = -0.785, P < 0.001), corporeal surface (m2) (r = -0.619, P < 0.001) or lean body mass (kg) (r = -0.543, P < 0.01). This correlation was not statistically signifi cant when the MCR(I) was expressed per kg of ideal body weight (r = -0 .121). The MCR(I) was also significantly correlated to both basal plas ma insulin levels (r = -0.400, P < 0.05) and glucose metabolic clearan ce rate (MCR(G)), as an index of insulin sensitivity, obtained in the clamp (r = -0.622, P < 0.001). These results suggest that differences in insulin metabolism are linked to differences in body weight by a st ill unknown mechanism and that variations in insulin metabolism may pa rtly account for differences in circulating insulin levels.