Abstract Number: Esoc2026a2118 Can Direct Oral Anticoagulants Replace Warfarin for Stroke Prevention in New-Onset Post-Operative Atrial Fibrillation? A Meta-Analysis and Trial Sequential Analysis

Background and aims New-onset postoperative atrial fibrillation (POAF) complicates up to 50% of post-cardiac surgery patients. Despite POAF’s high incidence, there are no clear guidelines regarding optimal anticoagulation and stroke prophylaxis regimen. Despite warfarin being the standard of care, its narrow therapeutic window and monitoring burdens are significant drawbacks. We investigated whether Direct Oral Anticoagulants (DOACs) provide a safe and effective alternative in this high-risk population. Methods A systematic search of major databases was conducted up to December 2025. Analysis was performed using the Mantel-Haenszel method with a random-effects model on RevMan. Results The analysis incorporated 36,825 patients from 13 studies (2 randomized-controlled trials + 11 observational studies). Analysis revealed no significant difference in stroke risk between Warfarin and DOACs [Relative Risk (RR) for warfarin: 1.22; 95% Confidence interval (CI): 0.96, 1.55; I^2 = 0%]. Trial sequential analysis (TSA) confirmed that the cumulative Z-curve remained within the non-significance zone, approaching the required information size and establishing the futility of further superiority testing. Furthermore, no significant differences were observed in the risk of all-cause mortality, readmission, systemic embolism, transient ischemic attack, major bleeding, gastrointestinal bleeding, or minor bleeding between warfarin and DOACs. Besides, there was no significant difference in post-operative length of stay between the two arms. Conclusions DOACs demonstrated an efficacy and safety profile comparable to warfarin for POAF. Reinforced by Trial Sequential Analysis and low heterogeneity despite mixed study designs, these results support DOACs as a non-inferior standard of care, allowing for simplified postoperative anticoagulation without compromising clinical outcomes.