Abstract

Patients at thromboembolic risk with non-valvular atrial fibrillation (AF) can now be managed either with a vitamin K antagonist (VKA) or with a fixed dose of a non-VKA oral anticoagulant (NOAC), while patients with valvular AF have been restricted to VKAs on the basis of a potentially higher risk and different mechanism of thrombosis, and the lack of sufficient data on the efficacy of NOACs. The terms 'non-valvular AF' and 'valvular AF' have not been however consistently defined. 'Valvular' AF has included any valvular disorder, including valve replacement and repair. In AF with rheumatic mitral disease, observational studies strongly suggest that VKA treatment is valuable. These patients have not been included in NOAC trials, but there is also no stringent argument to have excluded them. This is at sharp variance from patients with mechanical valves, also excluded from the pivotal Phase III trial comparing warfarin with NOACs, but in whom a single Phase II trial of dabigatran etexilate against VKA treatment was stopped prematurely because of increased rates of thromboembolism as well as increased bleeding associated with dabigatran. Until more data are available, such patients should be therefore managed with VKAs. We here propose an open-label randomized trial of one of the NOACs against the best of treatment available in regions of the world in which rheumatic heart disease is still highly prevalent, aiming at showing the superiority of the NOAC used against current standard treatment.