Abstract TP394: Comparison of Major Bleeding Endpoints Between Non-vitamin K Antagonist Oral Anticoagulants and Aspirin: A Network Meta-Analysis

Abstract

Background: The favorable risk-benefit of Non-vitamin K Antagonist Oral Anticoagulants (NOACs) vs. warfarin among individuals at high thrombotic risk is well established. However, relatively few trials have directly compared NOACs to aspirin with regard to bleeding outcomes. A Network meta-analysis integrates data from direct comparisons of treatments within trials and from indirect comparisons of interventions assessed against a common comparator in different trials, to compare all investigated treatments. Objective: To conducted a network meta-analysis to comprehensively investigate the risks of ICH, major bleeding, and fatal bleeding, for NOACs vs. aspirin. Methods: Pubmed and EMBASE from 1966 to June 2017 were searched to identify relevant studies. We included randomized controlled trials that included a comparison of (1) NOACs vs placebo, (2) NOACs vs aspirin, and (3) warfarin vs aspirin, and in which the number of patients and ICH or major bleeding events were reported. Only trials with treatment duration of 3 months or longer were included. We conducted the network meta-analysis utilizing a frequentist model with Stata version 14. Results: Literature search identified 34 trials with 110,127 participants. The population varied (e.g. atrial fibrillation, stroke, ischemic heart disease, venous thromboembolism) across trials. Two trials compared NOACs with aspirin, 10 trials compared NOACs with warfarin, and 22 trials compared warfarin with aspirin. Direct comparison showed that both NOACs and aspirin had lower risks of ICH than warfarin (NOAC vs warfarin: OR 0.45, 0.39-0.53; aspirin vs warfarin: OR 0.53, 0.37-0.76) while NOACs and aspirin had similar risk of intracranial hemorrhage (OR 0.87, 0.42-1.80). Network meta-analysis combining direct and indirect comparisons showed NOACs and aspirin had similar risks of intracranial hemorrhage (OR 0.98, 0.32-3.00). Network meta-analysis combing direct and indirect comparisons also showed NOACs and aspirin had similar risks of major bleeding (OR 0.99, 0.20-4.94) and fatal bleeding (OR 0.99, 0.55-1.78). Conclusions: This network meta-analysis of randomized controlled trials showed that risks of major bleeding endpoints were not different between NOACs and aspirin across broad populations.