Recognized risk factors account for only a small portion of the variance in the 4% to 10% incidence of major ischemic events associated with percutaneous coronary intervention. Body mass index (BMI) (body weight in kg/[height in m]2) is a clinically useful estimate of body fat and has been shown to correlate with mortality from several causes. We sought to evaluate the effect of BMI as a potential risk factor for the complications of percutaneous coronary intervention in 3,571 consecutive percutaneous coronary intervention patients treated at a single referral center. Patients were prospectively divided into the nonobese (BMI < or = 25), mildly obese (BMI 26-35), and very obese (BMI > 35), based on accepted definitions. Multiple logistic regression analyses were used to determine the correlates of major complications from 25 candidate variables, including BMI < or = 25 (n = 614 patients) and BMI > 35 (n = 275 patients), recorded prospectively in a relational database. Death occurred in 2.8% of the BMI < or = 25 group, in 3.7% of the BMI > 35 group, and in 0.9% of the BMI 26-34 group (p < 0.001), but there was no difference in the incidence of other ischemic events. Blood product transfusion was required in 12% of the BMI < or = 25 group, in 7% of the BMI 25-34 group, and in 8% of the BMI > 35% group (p = 0.003). Multivariate analysis, after adjustment for other significant correlates, demonstrated that both BMI < or = 25 (odds ratio [OR] = 2.7, p = 0.005) and BMI > 35 (OR = 7.4, p < 0.001) were independent correlates of death. Low-normal or high BMI is a newly described and powerful risk factor for in-hospital death after percutaneous coronary intervention.