Sudden cardiac death is often due to a ventricular arrhythmia. When a patient presents with a malignant arrhythmia unrelated to a transient reversible cause, there is a high probability of recurrent arrhythmia and sudden death. Clinical trials have shown a uniform survival benefit from implantable cardioverter-defibrillator (ICD) therapy in survivors of a malignant arrhythmia when compared with drug therapy. However, only 1% to 5% of patients survive an out-of-hospital cardiac arrest, emphasizing the need for primary prevention of sudden death. Clinical trial data available in this regard are largely limited to patients with coronary artery disease (CAD). Mortality can be reduced by the ICD in patients with CAD and depressed left ventricular ejection fraction (LVEF) less than 30%. If left ventricular function is only moderately depressed (LVEF between 30% and 40%), the presence of nonsustained ventricular tachycardia with inducible ventricular arrhythmia at electrophysiologic testing identifies patients who benefit from an ICD. The role of the ICD in primary prevention of sudden death in patients with nonischemic dilated cardiomyopathy is less clear at this time. Preliminary data indicate that the presence of heart failure symptoms in this population increases risk of sudden death that can be prevented by an ICD. Antiarrhythmic drugs have little role in prevention of sudden death; however, drugs that block the effects of beta-adrenergic stimulation, angiotensin, and aldosterone reduce mortality partly through their salutary effects on sudden death. Finally, a number of inherited defects of genes coding for ion channels, contractile sarcomeric proteins, and cell-to-cell junction proteins can result in primary electrical abnormalities and sudden death. The ICD is effective for secondary prevention, but its role in primary prevention is controversial and should be based on individual risk factors.