Mortality as an endpoint in atrial fibrillation

The aim of this study was to evaluate the use of mortality as an appropriate endpoint in studies of atrial fibrillation. Almost none of the atrial fibrillation trials has shown a significant difference in mortality among the various treatment arms because the studies are small, and many patient-years are required to show significant differences. Differences can be found when mortality is used as part of a combined endpoint and in meta-analyses of atrial fibrillation trials. Some trials of atrial fibrillation in the setting of conditions predisposing to high mortality, such as heart failure, postmyocardial infarction, and perioperative state for surgical procedures, can have mortality differences. Nevertheless, mortality has a role in validating the safety of new therapies and can be used as a measure of the impact and prognostic significance of atrial fibrillation in various disease settings. (C) 2004 Heart Rhythm Society. All rights reserved.