The relative impact of body mass index (BMI) and of exercise capacity on mortality was investigated in an intermediate to high risk of coronary artery disease (CAD) population. The present study examined 2,119 patients without known CAD who were referred for treadmill exercise stress myocardial perfusion imaging. During the 8.4±1.4 years of follow-up, there were 183 deaths. Age, Bruce protocol exercise time, BMI, male gender, and diabetes were all significant correlates of all-cause mortality. A multivariate analysis indicated that both exercise capacity and BMI correlated negatively with mortality, suggesting that increased exercise capacity and BMI were associated with lower mortality in patients with an intermediate to high risk of CAD. Although these results sustain the “obesity paradox”, this conclusion must be drawn carefully due to the limited number of individuals in some BMI and exercise capacity subgroups.