The aim of this study was to assess the relation between QRS duration and mortality in patients with known or suspected coronary artery disease, after adjustment for myocardial functional abnormalities, as assessed by exercise echocardiography. We studied 4,033 patients (age 62 +/- 12 years; 2,360 men, 18% with previous myocardial infarction) who underwent symptom-limited exercise echocardiography. The QRS duration was electronically measured from the 12-lead electrocardiogram. The incremental value of the QRS duration for predicting mortality was assessed by adding the QRS duration at the end of each of these modeling steps: clinical data, exercise electrocardiographic, and exercise echocardiographic variables. The QRS duration correlated positively with age, the wall motion score index at rest, and percentage of ischemic segments and negatively with workload (p = 0.0001). Of the 4,033 patients, 252 died during a median follow-up of 3 years. The QRS duration was univariately associated with an increased risk of death (relative risk 8.5, 95% confidence interval CI 4.4 to 16.4, p <0.0001). In an incremental multivariate model, the clinical predictors of mortality were age, male gender, previous infarction, and diabetes mellitus (chi-square 122). Workload was incremental to clinical data in the exercise test model (chi-square 193, p <0.0001). The exercise wall motion score index was incremental to both models (chi-square 211, p <0.001). The QRS duration was associated with an incremental risk of death when added to the clinical model (chi-square 133, p = 0.009), exercise test model (chi-square = 203, p = 0.002), and echocardiographic model (chi-square = 216, p = 0.03). A QRS duration > or =105 ms best identified patients at increased risk. In conclusion, QRS duration is associated with an increased risk of death, even after adjustment for clinical factors, exercise capacity, left ventricular function, and exercise-induced myocardial ischemia.