Analysis of the relationship between body mass index, insulin resistance, and beta-cell function: a cross-sectional study using the minimal model.

The objective of our research was to identify the mathematical model that would best define the relationship between obesity, insulin resistance (IR), and beta-cell function. Eighty-seven healthy subjects with a wide range of body mass index (BMI) were studied. Insulin sensitivity (IS) was calculated using Bergman's minimal model. Acute insulin response (AIRg) was calculated as the secretion of insulin during the first 10 minutes following a glucose bolus. IS x AIRg was used as an index of insulin-mediated glucose uptake (IMGU). The relationships among BMI, IS, fasting plasma insulin (FPI), and AIRg were studied in linear relationship terms and in terms of the hyperbolic function. Where the best fit was linear, the Jones and Molitoris method was used to investigate whether the 2-line fit was significantly better. The division of the population into BMI quartiles shows that from the third quartile, IS (12.4 +/- 6.0 v 11.0 +/- 6.4 v 4.8 +/- 1.8 v 3.2 +/- 2.0 E-5 min(-1)[pmol/L](-1), P < .01) diminishes. Nevertheless, a plateau was established between the last 3 quartiles for IS x AIRg. AIRg related to BMI via a breakpoint of 29.3 kg . m(-2). The best fits for both the BMI/IS and BMI/FPI relationships were hyperbolic. Our data indicate that obesity represents a continuum of IR, with severity increasing as BMI increases. Nevertheless, above a value of 29 kg . m(-2) and despite great increases in adiposity, IS tends to descend slowly. Moreover, there seems to be an IMGU threshold at a BMI value of approximately 27 kg . m(-2), above which an increase in adiposity leads to a greater fall in IS x AIRg. Furthermore, this threshold also appears to affect pancreatic response to a glucose stimulus.

Glucose metabolism in lean patients with mild type 2 diabetes mellitus: evidence for insulin-sensitive and insulin-resistant variants.

Obesity and type 2 diabetes mellitus (DM2) are 2 closely related syndromes, with obesity occurring in 70% to 80% of DM2 patients. Both syndromes are characterized by insulin resistance (IR). However, the metabolic characteristics of lean DM2 patients are not clearly defined, a fact attributed to the heterogeneity of the diabetes syndrome. Our objective was to study glucose metabolism in lean DM2 patients, in terms both of the basal and the insulin-stimulated states, and particularly, to investigate whether 2 subpopulations of diabetic patients are identifiable on the basis of degree of IR. Sixteen nonobese (body mass index [BMI] less than 27 kg. m(-2)) DM2 subjects with light to moderate fasting hyperglycemia were studied. Ten healthy subjects were used as a control group, with no family history of DM2 and matched by age, sex, and BMI in the diabetic group. All participants underwent an intravenous glucose tolerance test with frequent sampling over 180 minutes. Insulin sensitivity (IS) and glucose effectiveness at zero insulin (GEZI) were calculated using Bergman's minimal model. Non-insulin-mediated glucose uptakes (NIMGU) and insulin-mediated glucose uptakes (IMGU) were calculated for the basal (F) and insulin-stimulated states at 11.1 mmol/L of glucose (11.1). The beta-cell function was calculated via the acute insulin response to glucose (AIRg). Clustering techniques were used to identify subpopulations of DM2 patients on the basis of insulin sensitivity. The group of DM2 patients was characterized by both IR (IS index, 6.23 +/- 4.68 v 12.75 +/- 7.74 x 10(-5). min(-1). (pmol. L(-1))(-1), P <.01) and insulin secretion abnormalities (AIRg, 336 +/- 456 v 1,912 +/- 1,293 pmol/L. min, P <.0001), but showed similar values for GEZI (0.011 +/- 0.005 v 0.011 +/- 0.007 min(-1), not significant [NS]) in comparison to the control group. For the basal state, no differences were found between the DM2 patients and control subjects for NIMGU(F) (0.13 +/- 0.07 v 0.08 +/- 0.05 mmol/kg. min, NS) or for IMGU(F) (0.05 +/- 0.04 v 0.05 +/- 0.02 mmol/kg. min, NS). For the insulin-stimulated state, the DM2 patients showed a reduction of approximately 50% in the IMGU(11.1) value (0.20 +/- 0.17 v 0.38 +/- 0.24 mmol/kg. min, P <.05), but no significant differences were found for NIMGU(11.1) (0.19 +/- 0.09 v 0.20 +/- 0.12 mmol/kg. min, NS) in relation to the control group. Using the clustering technique, it was possible to identify 2 subpopulations of DM2 patients, a DM-IS group (n = 6) that was insulin sensitive (IS index, 11.70 +/- 2.40 x 10(-5). min(-1). (pmol. L(-1))(-1)) and a DM-IR group (n = 10) that was insulin resistant (IS index, 3.02 +/- 1.60 x 10(-5). min(-1). (pmol. L(-1))(-1)). The DM-IS group was characterized by an absence of IR, diminished GEZI, and a reduction in AIRg; whereas the DM-IR group was characterized by IR and a reduction in AIRg, but normal GEZI. We conclude that (1) as a group, DM2 patients are characterized by IR and beta-cell dysfunction, but normal NIMGU; (2) two subpopulations of DM2 patients can be identified on the basis of insulin sensitivity, with the DM-IS group further characterized by diminished GEZI; and finally, (3) deterioration in the pancreatic response to glucose stimulus is a sine qua non condition for a profound alteration in glucose metabolism in DM2 patients.