Response to Comment on Narayan. Type 2 Diabetes: Why We Are Winning the Battle but Losing the War? 2015 Kelly West Award Lecture. Diabetes Care 2016;39:653-663.

In his Kelly West Award Lecture, Narayan suggested that the pathogenesis of type 2 diabetes may be fundamentally different in high-risk populations who vary dramatically in degree of adiposity (1). For example, he raised the possibility that type 2 diabetes was due to “poor insulin secretion” in thin individuals (Asian Indians), whereas the culprit was “poor insulin action” in overweight/obese individuals (Pima Indians). This suggestion does not seem to acknowledge the published evidence from our group and others of the significant differences in degree of insulin resistance that exist among various racial/ethnic groups that is independent of obesity. For example, we quantified insulinmediated glucose disposal with the insulin suppression test and compared the values in nonobese men and women of South Asian versus European ancestry (2). The experimental groups were matched for age, BMI, waist-to-hip ratio, alcohol intake, and physical activity. Despite being matched for these relevant demographic characteristics, South Asians were significantly more insulin resistant, with the predictable abnormalities of glucose intolerance, hyperinsulinemia, and dyslipidemia. In another study, the insulin suppression test was used to compare insulinmediated glucose in four groups of Taiwanese men: normal, nonobese (BMI 23.5 kg/m); normal, obese (30.4 kg/m); type 2 diabetes, nonobese (23.8 kg/m); and type 2 diabetes, obese (31.8 kg/m) (3). The two groups with diabetes, nonobese and obese, were significantly more insulin resistant when compared with weight-matched normal control groups. Indeed, the nonobese groupwith diabetes was more insulin resistant than the normal, obese group. These examples do not imply that excess adiposity is without an adverse impact on insulin sensitivity but do point out that it need not be present to account for insulin resistance. Indeed, measurements of insulin action (hyperinsulinemiceuglycemic clamp) showed that differences in adiposity, independent of differences in maximal aerobic capacity, accounted for approximately 25% of the several-fold variation in insulin action in glucosetolerant individuals of either Caucasian or Pima Indian ancestry (4). Furthermore, insulin action in Pima Indians was “about half of that in Caucasians at similar degrees of obesity,” and “there appeared to be a racially determined defect in the Indians as compared to the Caucasians (4).” There may be differences in the relative importance of “poor insulin action” versus “poor insulin secretion” in thepathophysiology of type 2 diabetes in thin (Asian Indians) versus obese (Pima Indians) people, but the fact that insulin resistance can vary dramatically in different ethnic/ racial groups, independent of adiposity, should not be ignored.