Effect of apixaban versus vitamin K antagonist and aspirin versus placebo on days alive and out of hospital: An analysis from AUGUSTUS

Abstract

Clinical trials of antithrombotic agents typically use separate time-to-event analyses for bleeding and ischemic events, but this framework has limitations. Days alive and out of hospital (DAOH) is an alternative that may provide additional insight. We assessed the utility of DAOH as a clinical trial endpoint among patients with atrial fibrillation and acute coronary syndrome or percutaneous coronary intervention METHODS: AUGUSTUS, a randomized clinical trial, compared apixaban with warfarin and aspirin with placebo in 4614 patients with atrial fibrillation and acute coronary syndrome or percutaneous coronary intervention. We used Poisson regression with a robust variance estimate to compare DAOH by treatment group. Mean (SD) DAOH was 168 (31); median (IQR) was 177 (169-180); 75% of patients neither died nor were hospitalized. Mean (SD) DAOH was 169 (28) with apixaban + placebo, 168 (29) with apixaban + aspirin, 168 (33) with warfarin + placebo, and 167 (33) with warfarin + aspirin. There were no significant differences in the rate ratio for DAOH for apixaban vs. warfarin (RR 1.00, 95% CI 0.99-1.01) or aspirin vs. placebo (RR 1.00, 95% CI 1.00-1.01). Compared with warfarin, apixaban increased the proportion of patients who neither died nor were hospitalized during follow-up (76.8 vs. 73.3%; OR 0.83, 95% CI 0.73-0.95). In this analysis of AUGUSTUS, there was no difference in DAOH by treatment arm. These findings contrast with time-to-event analyses, which showed lower rates of major bleeding and hospitalization with apixaban and placebo. DAOH may not be very a useful measure of effects of antithrombotic therapies in this population. clinicaltrials.gov; NCT02415400; https://clinicaltrials.gov/study/NCT02415400.