In order to test the hypothesis that abdominal adiposity measured by computed tomography is a better predictor of diabetes risk than body mass index or other measures of abdominal adiposity [i.e., waist circumference or the waist-to-hip ratio (WHR)], specific measurements of body composition, such as intra-abdominal (visceral) adipose tissue (AT) and subcutaneous AT, were performed in 1,106 men and women who participated in the Diabetes Prevention Program (DPP). The DPP is a prospective study that followed participants who were randomized to lifestyle therapy, metformin, or placebo. In men in the placebo group, the measure of abdominal adiposity that best predicted the development of diabetes over a 3.2-year follow-up was intra-abdominal AT (obtained at the L4-L5 level) [HR=1.79 (95% CI, 1.21-2.67)]. In men in the lifestyle group, WHR had the strongest association with diabetes incidence [HR=3.47 (95% CI, 1.36-8.89)]. In women in the placebo group, the ratio of intra-abdominal AT to subcutaneous AT best predicted diabetes incidence [HR=1.29 (95% CI, 1.02-1.64)], whereas intra-abdominal AT measured at the L2-L3 level showed the highest association in the lifestyle group [HR=2.12 (95% CI, 1.34-3.34)]. None of these measurements predicted diabetes incidence in the metformin group. Based on these results, DPP investigators suggested that direct measurements of central adiposity do not add more to the risk of diabetes associated with abdominal adiposity than the measurement of either waist circumference or WHR. The paper was accompanied by an editorial by Amalia Gastaldelli who suggested potential mechanisms linking increased intra-abdominal AT accumulation to diabetes development. Gastaldelli remarked that the absence of a relationship between abdominal obesity and diabetes risk in the metformin group might be explained by the directed effect of metformin on insulin sensitivity, which seems to be largely independent of weight loss. A major limitation of this study was also highlighted. Because lifestyle intervention improved fitness and weight loss, the changes in abdominal obesity during the intervention should have been taken into consideration to establish the relationship between abdominal obesity and diabetes risk.
Documentation centre