Ventricular fibrillation (VF) is the first recorded arrhythmia in 75% of patients who have a sudden cardiovascular collapse. Rarely (1%) does sustained ventricular tachycardia (VT) alone cause collapse and unconsciousness. Whether all VF begins as VT is unknown. Early application of cardiopulmonary resuscitation and rapid defibrillation are essential to ensure survival and satisfactory neurologic recovery. During the last 2 years in Seattle, the initial resuscitation rate for VF was 269 of 447 patients (60%), with 114 of 447 patients (26%) surviving long-term. Survivors of VF have a high overall risk of recurrent VF, with many univariate risk factors identified: evidence of poor left ventricular function (history of congestive heart failure, prior myocardial infarction [MI] or low ejection fraction), extensive coronary artery disease, absence of a new MI (either Q wave or non-Q wave) with VF, male gender, advanced age, complex or high-frequency ventricular ectopy on Holter recording, inducibility at electrophysiologic study, exercise-induced angina or hypotension, and smoking. Classification of cardiac deaths as arrhythmic or nonarrhythmic is important in interpreting the therapeutic response. However, because many patients have chronic symptoms, timing of the onset of a new event is difficult. Furthermore, accurate timing of an event does not guarantee correct classification. Sudden death is not necessarily arrhythmic, nor is all arrhythmic death sudden. Total cardiac mortality may be a simpler and more relevant end point to measure the overall effect of antiarrhythmic therapy.