Although current data suggest that the ICD is effective, the inherent unreliability of available nonrandomized observational studies make randomized trials of the ICD, such as AVID, CIDS, and CASH, ethically justified and highly desirable. Reasonable decisions regarding patient eligibility and the primary endpoint have been made by all three studies. When completed, these studies should have sufficient power to provide a clear answer as to whether or not the ICD reduces death in our highest risk patients.
