Insulin Action in Black Americans With NIDDM

OBJECTIVE: To assess the influences that obesity and hyperglycemia have on insulin action in black NIDDM patients. RESEARCH DESIGN AND METHODS: Thirty-nine subjects were studied who had normal GHb levels and/or FPG less than 6.4 mM and who had not taken pharmacological agents for 2-91 mo before the study. Insulin action was studied using the euglycemic insulin clamp with a D-[3-3H]glucose infusion. Degree of obesity was assessed with BMI. During carefully monitored follow-up, 9 patients relapsed into a hyperglycemic state, and insulin action was restudied after acute reregulation of their plasma glucose. RESULTS: Insulin action was related to the degree of obesity at the extremes of BMI: 7 of 8 patients (87.5%) with a BMI less than 24.0 kg/m2 were insulin sensitive, and 8 of 9 patients (88.9%) with a BMI greater than 28.5 kg/m2 were insulin resistant. In the midrange BMI (24.0-28.5 kg/m2), patients were equally likely to be insulin resistant or insulin sensitive. A plot of frequency versus glucose disposal in those patients was compatible with a bimodal distribution (P less than 0.025): 12 of 22 patients were normally insulin sensitive (glucose disposal 6.1-9.4 mg.kg-1.min-1), and 10 were insulin resistant (glucose disposal 2.4-4.8 mg.kg-1.min-1). Analysis of this midrange BMI group showed that in the insulin-sensitive group, an inverse relationship existed between BMI and glucose disposal (r = -0.64, P less than 0.05), whereas no such relationship was found in the insulin-resistant group. The clinical characteristics of the midrange BMI group indicated that fasting plasma insulin, total cholesterol, and triglycerides were higher; whereas BMI, age, and FPG were not different in the insulin-resistant compared with the insulin-sensitive group. With the development of hyperglycemia, insulin action in the insulin-sensitive group. With the development of hyperglycemia, insulin action in the insulin-sensitive group was decreased, independent of obesity, whereas it was unchanged in the insulin-resistant group. CONCLUSIONS: Insulin resistance exists in only approximately 50% of black NIDDM patients. The relationship between obesity and insulin resistance is not a simple one. The data can be explained by one of two hypotheses: 1) insulin resistance in black NIDDM patients is an acquired defect related to the development of obesity and is modulated by hyperglycemia, or 2) NIDDM exists as two variants, one with primary insulin resistance and one with normal insulin sensitivity, and that insulin resistance causes central and/or generalized obesity.