Insulin resistance is the best predictor of the metabolic syndrome in subjects with a first-degree relative with type 2 diabetes.

Although obesity is associated with insulin resistance and the metabolic syndrome (MetS), some obese individuals are metabolically healthy. Conversely, some lean individuals are insulin resistant (IR) and at increased cardiometabolic risk. To determine the relative importance of insulin sensitivity, BMI and waist circumference (WC) in predicting MetS, we studied these two extreme groups in a high-risk population. One thousand seven hundred and sixty six subjects with a first-degree relative with type 2 diabetes were stratified by BMI and homeostasis model assessment of insulin resistance (HOMA(IR)) into groups. IR groups had higher triglycerides, fasting glucose, and more diabetes than their BMI-group insulin sensitive (IS) counterparts. Within both IS and IR groups, obesity was associated with higher HOMA(IR) and diastolic blood pressure (BP), but no difference in other metabolic variables. MetS (Adult Treatment Panel III (ATPIII)) prevalence was higher in IR groups (P < 0.001) and more subjects met each MetS criterion (P < 0.001). Within each BMI category, HOMA(IR) independently predicted MetS (P < 0.001) whereas WC did not. Within IS and IR groups, age and WC, but not BMI, were independent determinants of MetS (P < 0.001). WC was a less meaningful predictor of MetS at higher values of HOMA(IR). HOMA(IR) was a better predictor of MetS than WC or BMI (receiver operating characteristic (ROC) area under the curve 0.76 vs. 0.65 vs. 0.59, P < 0.001). In conclusion, insulin sensitivity rather than obesity is the major predictor of MetS and is better than WC at identifying obese individuals with a healthier metabolic profile. Further, as many lean individuals with a first-degree relative with type 2 diabetes are IR and metabolically unhealthy, they may all benefit from metabolic testing.