The aim of this study was to determine whether trunk fat mass, measured by dual-energy X-ray absorptiometry (DEXA), is predictive of insulin resistance and dyslipidemia, independently of arm and leg fat mass, in postmenopausal women. Total and regional body composition was measured by DEXA in 166 healthy, postmenopausal women (66 +/- 4 yr). Four primary markers of insulin resistance and dyslipidemia were assessed: 1) area under the curve for the insulin (INS(AUC)) response to an oral glucose tolerance test (OGTT), 2) product of the OGTT glucose and insulin areas (INS(AUC)xGLU(AUC)), 3) serum triglycerides (TG), and 4) high-density lipoprotein (HDL)-cholesterol. Trunk fat mass was the strongest independent predictor of each of the primary dependent variables. In multivariate regression models, trunk fat mass was associated with unfavorable levels of INS(AUC), INS(AUC)xGLU(AUC), TG, and HDL-C, whereas leg fat mass was favorably associated with each of these variables. Thus trunk fat is a strong independent predictor of insulin resistance and dyslipidemia in postmenopausal women, whereas leg fat appears to confer protective effects against metabolic dysfunction.