New oral anticoagulants in atrial fibrillation: a reappraisal of trial results looking at absolute figures

Three new oral anticoagulant agents were tested versus warfarin in separate, large phase III randomized clinical trials for prevention of any stroke and systemic embolism in atrial fibrillation. Dabigatran, a direct thrombin inhibitor, is at 110 mg bid non-inferior and at 150 mg bid superior to warfarin; rivaroxaban, a factor X inhibitor, is also non-inferior, and apixaban, also a factor X inhibitor, is superior to warfarin on the same efficacy end point. Statistical analysis of subgroups does not suggest, for any of the tested drugs, major differences in relation to different risk levels and history of previous stroke/TIA. This re-appraisal of data was undertaken in search for possible additional information, by considering the absolute differences in efficacy and safety events versus warfarin and the corresponding efficiency and number needed to treat, also with regard to secondary versus primary prevention. By this approach, it appears that for all drugs, equivalence or advantage versus warfarin on the efficacy end point is largely driven by a reduction in hemorrhagic rather than ischemic strokes. Dabigatran shows a balanced effect on ischemic and hemorrhagic strokes, and apixaban is most effective in sparing intracranial bleeding versus warfarin. In secondary prevention, better efficiency is shown by dabigatran 150 and apixaban, versus rivaroxaban, despite the higher proportion of post-stroke/TIA patients (55 %) in the ROCKET AF trial of rivaroxaban seemed to favor better results of this drug in secondary prevention. These and other results of our approach should not be directly translated into clinical practice. They may supply useful suggestions to be subsequently tested in specific trials, although head-to-head comparative studies of the three drugs remain unlikely.