Aims: Non-vitamin K antagonist oral anticoagulants represent a new option for prevention of embolic events in patients with atrial fibrillation (AF). However, little is known about the impact of non-cardiac comorbidities on the efficacy and safety profile of these drugs.
Methods and results: In a post hoc analysis of the ENGAGE AF-TIMI 48 trial, we analysed 21 105 patients with AF followed for an average of 2.8 years and randomized to either a higher-dose edoxaban regimen (HDER), a lower-dose edoxaban regimen, or warfarin. We used the updated Charlson Comorbidity Index (CCI) to stratify the patients according to the burden of concomitant disease (CCI = 0, 1, 2, 3, and ≥4). The treatment groups were then compared for safety, efficacy, and net clinical outcomes across CCI categories. There were 32.0%, 7.3%, 42.1%, 12.7%, and 6.0% of patients with CCI scores of 0, 1, 2, 3, and ≥4, respectively. A CCI score ≥4 was associated with significantly higher rates of thromboembolic events, bleeding, and death compared to CCI = 0 (P < 0.05 for each). The annualized rates of the primary net clinical outcome (stroke/systemic embolism, major bleeding, or death) for CCI = 0, 1, 2, 3, or ≥4 were 5.9%, 8.7%, 6.6%, 10.3%, and 13.6% (Ptrend < 0.001). There were no significant interactions between treatment with HDER vs. warfarin and efficacy, safety, and net outcomes across the CCI groups (P-interaction > 0.10 for each).
Conclusion: Although increasing CCI scores are associated with worse outcomes, the efficacy, safety, and net clinical outcomes of edoxaban vs. warfarin were independent of the degree of comorbidity present.
Keywords: Anticoagulation; Atrial fibrillation; Comorbidity; Edoxaban; Warfarin.
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2019. For permissions, please email: journals.permissions@oup.com.