Page 7
The Coronary Artery Bypass Graft Patch (CABG Patch) trial enrolled subjects with decreased ejection fractions (35 percent or less) and abnormalities on signal-averaged electrocardiography who were scheduled for coronary revascularization. Patients were randomly assigned in the operating room either to implantation of a defibrillator with epicardial leads or to no defibrillator therapy. The CABG Patch trial was terminated when an interim analy-sis of data from 900 subjects who were followed for more than a mean (SD) of 3216 months showed no potential benefit in the defibrillator group. There were 101 deaths among the 454 patients who re-ceived defibrillators, and 95 deaths among the 446 control patients. The hazard ratio for death in the defibrillator group was 1.07 (95 percent confidence interval). Several possible explanations for this result have been proposed. Most of the deaths occurred in the hospital, with a large proportion of them occurring early in the postoperative period.37 About 10 per-cent of the control group crossed over early to im-plantation of a defibrillator, and these patients may have been at the highest long-term risk for death from cardiovascular causes. The mortality rate was lower than anticipated, possibly owing to improve-ments in early surgical management and to the ben-efits of revascularization for patients with a low ejec tion fraction. Epicardial-lead systems were used in the patients who received defibrillators and may have had an adverse effect on early mortality.
The second Multicenter Automatic Defibrillator Implantation Trial (MADIT II)23 enrolled 1232 patients with coronary artery disease, a history of myocardial infarction, and an ejection fraction of 30 percent or less. Documentation of spontaneous or inducible arrhythmias was not required. Patients were randomly assigned to either defibrillator therapy or conventional medical therapy. Antiarrhythmic therapy was used in less than 20 percent of patients in both groups. During an average of 20 months of follow-up, mortality from all causes was 19.8 percent in the control group and 14.2 percent in the defibrillator group. The hazard ratio for death in the defibrillator group was 0.69.