Estimated Effectiveness and Safety of Nonvitamin K Antagonist Oral Anticoagulants Compared With Optimally Acenocoumarol Anticoagulated "Real-World" in Patients With Atrial Fibrillation

Nonvitamin K antagonist oral anticoagulants (NOACs) have been proposed as an alternative to vitamin K antagonists in atrial fibrillation (AF) patients but the comparative benefits between NOACs and optimally anticoagulated patients is unknown. We estimated the absolute benefit in clinical outcomes rates of real-world effect of NOACs in optimally anticoagulated AF patients with acenocoumarol. We included 1,361 patients stable on acenocoumarol with time in therapeutic range of 100% and 6.5 years of follow-up. Estimation of clinical events avoided was calculated applying hazard ratio, absolute and relative risk reduction from the real-world meta-analysis. Compared with an optimally anticoagulated population, dabigatran 110 mg had the highest estimated stroke reduction (0.97%/year vs 1.47%/year; p = 0.002), and the benefit was higher than in RE-LY trial. For major bleeding, apixaban showed the highest estimated reduction (1.81%/year vs 2.83%/year; p <0.001). For mortality, the largest estimated reduction was with apixaban (2.68%/year). For gastrointestinal bleeding, only apixaban had a significant reduction compared with acenocoumarol (0.69%/year vs 1.10%/year; p = 0.004), and the reduction was significantly higher than in ARISTOTLE trial. All NOACs showed significantly lower rates for intracranial hemorrhage and had a positive Net Clinical Benefit compared with acenocoumarol. Apixaban showed the highest extended estimated Net Clinical Benefit 2.64 (95%CI 2.34 to 2.96). In conclusion, in optimally acenocoumarol anticoagulated AF patients, estimated reductions in all clinical outcomes with various NOACs are evident, with the best effectiveness and safety profile with apixaban. Indeed, the estimated effect with "real world" NOACs would probably be higher than that seen in phase-III clinical trials.