Reduced Glucose-Induced Thermogenesis Is Present in Noninsulin-Dependent Diabetes Mellitus without Obesity*

Decreased glucose-induced thermogenesis has been observed in all forms of obesity. However, some studies implicate insulin resistance rather than obesity per se as the mechanism by which glucose-induced thermogenesis is reduced. To establish the role of insulin resistance in reduced thermogenesis independent of obesity, we compared energy expenditure in 9 nonobese individuals with noninsulin-dependent diabetes mellitus (NIDDM) to 16 nonobese control subjects using indirect calorimetry and the hyperinsulinemic clamp technique. To document the presence of insulin resistance and reduced glucose-induced thermogenesis in nonobese NIDDM, 6 individuals from each group were studied under identical conditions of hyperinsulinemia (120 mU/m2.min) and euglycemia (approximately 5 mmol/l). Both glucose uptake (0.482 +/- 0.042 vs. 0.737 +/- 0.040 g/min) and energy expenditure above basal (0.04 +/- 0.02 vs. 0.10 +/- 0.02 kcal/min) were decreased in nonobese NIDDM compared to control subjects (both P less than 0.05). To determine whether decreased glucose-induced thermogenesis could be overcome by correcting for reduced glucose uptake, the 9 nonobese NIDDM individuals were age and weight-matched to 9 control subjects and clamps were performed at matched rates of glucose uptake. During a 40 mU/m2.min insulin infusion, the nonobese NIDDM individuals were studied at hyperglycemia (17.5 +/- 1.9 mmol/L) and compared to the control subjects at euglycemia (5.1 +/- 0.1 mmol/L; P less than 0.05). Under these conditions, both groups achieved similar rates of glucose uptake (0.698 +/- 0.040 vs. 0.688 +/- 0.038 g/min, NIDDM and control subjects, respectively) and similar rates of energy expenditure above basal (0.08 +/- 0.03 vs. 0.06 +/- 0.02 kcal/min, P = NS). During 600 mU/m2.min clamps performed at hyperglycemia (19.0 +/- 1.2 vs. 14.5 +/- 1.1 mmol/L, NIDDM vs. control subjects, respectively; P less than 0.05), rates of maximal glucose uptake (1.538 +/- 0.093 vs. 1.518 +/- 0.047 g/min) and energy expenditure above basal (0.34 +/- 0.03 vs. 0.31 +/- 0.03 kcal/min) were also similar (P = NS). In conclusion nonobese NIDDM is associated with both decreased rates of glucose uptake and decreased glucose-induced thermogenesis. Decreased glucose substrate availability, due to impaired insulin action, appears to be the critical determinant of glucose-induced thermogenesis in nonobese NIDDM. These data indicate that decreased thermogenesis in NIDDM is a consequence of insulin resistance and can occur independent of obesity.