Because there is controversy regarding the clinical relevance of exercise-induced ventricular arrhythmias, we analyzed their significance in 383 patients who had undergone both exercise thallium-201 stress-testing and cardiac catheterization. Two-hundred twenty-one patients (58%) had no exercise-induced ventricular arrhythmias while 162 (42%) did. There was no difference between patients with and without exercise-induced ventricular arrhythmias in terms of previous myocardial infarction (p = 0.61), incidence of fixed thallium-201 defects (0.06), number of diseased vessels (p = 0.09) and resting left ventricular ejection fraction (p = 0.06). In contrast, evidence of provocable ischemia (redistribution on thallium-201 and ST-segment depression on the electrocardiogram) were more likely (p less than 0.02) to be seen in patients with exercise-induced ventricular arrhythmias. Discriminant function analysis revealed that these 2 variables best separated patients with and without exercise-induced ventricular arrhythmias. In a 4- to 8-year follow-up, 89 patients had adverse cardiac events. Of these 89, there were 41 deaths, 9 nonfatal myocardial infarctions and 39 coronary revascularization procedures performed later than 3 months after catheterization. Patients with exercise-induced ventricular arrhythmias were more likely (p = 0.01) to have these events than those without these arrhythmias. Moreover, these arrhythmias provided independent prognostic information beyond that provided by the thallium-201 stress test and coronary angiography. We conclude that exercise-induced ventricular arrhythmias are associated with exercise-induced ischemia and provide prognostic information which adds marginally to that provided by other noninvasive and invasive parameters in ambulatory patients being evaluated for chest pain.