Objective: To report appropriate (AT) and inappropriate (IT) ICD therapies in patients with ischemic and nonischemic heart disease, as well as early and late procedure-related complications. Methods: One hundred and fifty-five patients 0 19 mate and 36 female), mean age 4 7 years (21-88), who underwent ICD implantation between 1994 and March 2003 were analyzed. Patients were divided into the following groups: Group I - Pbst-AMI patients (n = 80); Group II - Patients with nonischemic heart disease and LV ejection fraction < 40% (n 45), Chagas disease (n = 18), idiopathic dilated cardiomyopathy (n = 12), hypertensive disease (n = 8), hypertrophic cardiomyopathy (n 4) and valvular heart disease (n = 3); Group III - Patients with arrhythmogenic right ventricular dysplasia (n = 13); and Group IV - Patients with channelopathies: Brugada Syndrome (n = 8) and idiopathic ventricular arrhythmias (n = 9). All patients underwent electrophysiological study (EPS) with induction of sustained ventricular arrhythmia ICD implantation. Results: During the 26-month mean follow up, a high rate of appropriate ICD therapies (antitachycardia pacing and/or shock) was observed (46%) in the four groups, with no statistically significant difference. The four groups did not differ in either overall (8.4%) or arrhythmic mortality (1.3%). There was no correlation between appropriate ICD therapies and initial clinical presentation or inducible ventricular arrhythmia at EPS, and a 4% incidence of early and late procedure-related complications was found. Conclusion: The high incidence of appropriate ICD therapy and low rate of sudden death in the patients studied suggest that ICD is a valuable strategy in the management of ischemic and nonischemic patients previously selected by means of EPS.