Abstract
Background: Dose adjustments of direct-acting oral anticoagulants (DOACs) for atrial fibrillation are based on pivotal clinical trials assessing their effectiveness and safety in controlled settings. However, the appropriateness of these dosing strategies in real-world practice is uncertain. The purpose of this study is to compare the effectiveness and safety of dose-specific DOACs with those of warfarin. Methods: This study retrieved articles from MEDLINE, Embase, and CENTRAL until March 5, 2024. Primary outcomes were the incidence of stroke/systemic embolisms (S/SEs) and major bleeding (MB). Direct pairwise meta-analyses compared each dose-specific DOAC with warfarin. Heterogeneity was assessed using Higgin's I (2) and Q statistics, while publication bias was evaluated through funnel plots and Begg's and Egger's tests, with adjusted pooled estimates calculated via trim-and-fill and precision-effect estimate with standard error (PET-PEESE) methods. A network analysis was conducted, with additional comparisons made using a Bayesian random-effects model for indirect evidence. Results: A total of 32 studies with 2,332,770 patients were included. Both standard-dose (SD) and low-dose (LD) DOACs significantly reduced S/SE, except for LD apixaban and LD edoxaban. Rivaroxaban did not show significant difference in MB compared to warfarin. In East Asian patients, all doses of DOACs exhibited lower hazard ratios (HRs) for S/SE and MB than those observed in the primary analysis, with LD rivaroxaban significantly reducing MB, a finding not observed in the primary analysis. Rank probability analysis indicated that the dose-specific DOACs had different safety profiles and small but meaningful differences in effectiveness. SD apixaban (S/SE: second, MB: second) and edoxaban (S/SE: first, MB: fourth) and LD edoxaban (S/SE: fourth, MB: first) had high ranks. LD apixaban had the most significant difference in rank for S/SE from SD apixaban, ranking eighth compared to second. Conclusions: This study found that all DOACs provided comparable or superior effectiveness and safety to warfarin. SD apixaban, SD edoxaban, and LD edoxaban achieved a favorable balance between preventing S/SE and MB risk.