Abstract 12197: Causes of Death in Patients With Symptomatic Peripheral Artery Disease After Lower Extremity Revascularization: Insights From Voyager Pad

Abstract

Background: In trials of patients with coronary artery disease (CAD), unknown deaths (e.g. out of hospital) are often categorized as sudden/presumed cardiovascular (CV) based on observations that recurrent coronary events are a major cause of mortality. Such conventions are applied broadly to CV trials, including those in peripheral artery disease (PAD). Recent trials in patients selected on the basis symptomatic PAD, such as EUCLID trial, report that only ~1/3 have known CAD. Therefore, understanding the causes of death in patients in trials of symptomatic PAD may help elucidate whether different conventions should be considered. Hypothesis and Methods: VOYAGER PAD enrolled patients with symptomatic PAD after lower extremity revascularization. Source documents were collected for all deaths and cause was adjudicated by an independent blinded CEC using accepted CV trial conventions for those dying out of hospital. For this post-hoc analysis, source documents were re-reviewed to provide further details regarding specific causes of deaths. Results: 6564 patients were randomized and followed for a median of 28 months. A total of 637 deaths occurred, and of these, 35% were CV, 22% were unknown (e.g. out of hospital, sudden deaths), and 43% were non-CV. Of the CV deaths, 40% (14% of all deaths) were atherothrombotic (e.g. acute MI, stroke, PE), while 60% (21% of all deaths) were not atherothrombotic (e.g. heart failure, shock). Deaths due to cancer (17%) and infection (15%) were more frequent than atherothrombotic deaths. Conclusions: Patients recruited into trials for symptomatic PAD die of diverse causes. Atherothrombosis caused a minority of deaths, while heart failure, cancer, and infection are as or more frequent. These observations suggest that causes of death differ in populations selected on the basis of PAD versus CAD and that assumptions underlying categorization of deaths of unknown etiology should be carefully considered in these distinct populations.