Risk stratification and effectiveness of implantable cardioverter-defibrillator (ICD) therapy are unresolved issues in hypertrophic cardiomyopathy (HCM), a cardiac disease that is associated with arrhythmias and sudden death. We assessed ICD therapy in 132 patients with HCM: age at implantation was 34 +/- 17 years, and 44 (33%) patients were aged </= 20 years. Indications were sustained ventricular tachycardia (VT) or cardiac arrest (secondary prevention) in 47 (36%) patients, and clinical features associated with increased risk for sudden death (primary prevention) in 85 (64%) patients. There were 6 deaths and 55 appropriate interventions in 27 (20%) patients during a mean follow-up period of 4.8 +/- 4.2 years: 5-year survival and event-free rates were 96%+/- 2%and 75%+/- 5%, respectively. ICD intervention-free rates were significantly less for secondary than for primary prevention:64%+/- 7%versus 84%+/- 6%at 5 years,P = 0.02. Notably, 59 of 67 events (cardiac arrest and therapeutic ICD interventions), or 88%, occurred during sedentary or noncompetitive activity. Incidence of therapeutic shocks was related to age but not to other reported risk factors, including severity of cardiac hypertrophy, nonsustained VT during Holter monitoring, and abnormal blood pressure response to exercise. ICD related complications occurred in 38 (29%) patients, including 60 inappropriate ICD interventions in 30 (23%) patients. However, 8 (27%) of the patients with inappropriate shocks also had therapeutic interventions. ICD is effective for secondary prevention of sudden death in HCM. However, selection of patients for primary prevention of sudden death, and prevention of device related complications require further refinement.