Objective: Central fat mass (CFM) correlates with insulin resistance and increases the risk of type 2 diabetes and cardiovascular complications. On the other hand, increased peripheral fat mass (PFM) is associated with higher insulin sensitivity. Thus, we examined the contribution of adipose tissue distribution, as assessed by the PFM/CFM ratio, to insulin sensitivity in overweight and obese postmenopausal women. Design and methods: A total of 124 nondiabetic overweight and obese postmenopausal women underwent an oral glucose tolerance test (OGTT) and a hyperinsulinemic/euglycemic (HI) clamp. Body composition was determined using computed tomography for visceral adipose tissue (VAT) and dual X-ray absorptiometry for fat mass, lean body mass and their respective proportions. Participants were divided by tertiles of the PFM/CFM ratio. Results: Participants with preferential CFM (group 1) had higher fasting insulin levels and insulin area under the curve (AUC) during OGTT, as well as lower glucose infusion rates during the HI clamp, whether it was expressed per kg of body weight (M) or per kg of fat-free mass (Mm), compared with the other two groups. The PFM/CFM ratio also correlated significantly with fasting insulin (r ¼ À0.32, Po0.001), the insulin AUC (r ¼ À0.42 Po0.001), M (r ¼ 0.39 Po0.001) and Mm (r ¼ 0.37 Po0.001). Using hierarchical regression, we demonstrated that the PFM/CFM ratio was an independent predictor of insulin AUC, M and Mm and that its sequential addition to CFM and VAT improved significantly the predictive value of the model for insulin sensitivity for all variables except fasting insulin. Conclusion: The PFM/CFM ratio, which integrates the antagonistic effects of both central and peripheral depots on insulin sensitivity, added substantially to the prediction of insulin sensitivity over VAT and CFM alone.