Update on MADIT: The multkenter automatic defibrillator implantation trial

I n 1990, we developed the Multicenter Automatic Defibrillator Implantation Trial (MADIT) hypothesis, that an implanted cardioverter-defibrillator (ICD) would improve survival in high-risk coronary patients when compared with conventional medical management. In the subsequent trial we did not prespecify a comparison drug; management in the conventional arm was left to the discretion of the attending cardiologist, provided the drug was FDA approved. There were 32 participating centers, 30 in the United States and 2 in EuroDe. We enrolled the first patient into MADIT on Didember 27, 1990. The safety monitoring committee recommended termination of the trial on March 2 1, 1996, and the investigators and executive committee met on March 24, 1996, and agreed to terminate the trial. By that time we had randomized 196 patients. The average duration of follow-up was 27 months. The endpoint of this study was total mortality; there were 54 deaths at the time the trial was stopped. Patient enrollment Gteria were nonsustained ventricular tachycardia, prior Q-wave infarction, ejection fraction <35%, and electrophysiologic induction of ventricular tachycardia nonsuppressible with intravenous, procainamide. Thus, all subjects were at high risk. Average ejection fraction turned out to be 26% and the average age was 67 years. Randomization into defibrillator versus best conventional treatment arms tias well balanced. The major dmg used in the nondevice arm (74% of patients) was amiodarone. There were very few patients on type IA agents and none on type IC (Table I). The Kaplan-Meier camulative survival curve for the 196 patients after randomization showed im-