New options for anticoagulant therapy in non-valvular atrial fibrillation: Data from ARISTOTLE, ROCKET-AF and RE-LYThe role of Vitamin K antagonists such as warfarin as the management of choice for patients with non-valvular atrial fibrillation (AF) at high risk for thrombo-embolism has been central to therapeutics for many years. However, the use of warfarin is associated with the problem of the need for individual effect monitoring, difficulties in achieving good control in some patients, and substantial risk of bleeding and other complications.In recent years two major classes of oral anticoagulants, direct thrombin inhibitors and factor Xa inhibitors, have been the subject of comparative clinical trials against warfarin in patients with non-valvular AF. Prior to the 2011 ESC meeting key published data on these agents could be summarized as follows:-(1) RE-LY trial. This study was a large (n018,113) randomized open-label comparison of two doses of the direct thrombin inhibitor dabigatran (in doses of either 110 or 150 mg BID) against warfarin, with blinded adjudication of end-points [1]. Patients with severely impaired renal function were excluded. The results surprisingly showed substantial differences in efficacy vs haemorrhagic risk according to dabigatran dosage. The primary efficiency end-point of stroke/systemic embolism occurred less frequently in the high-dose dabigatran group than with warfarin therapy, without an excess of major haemorrhagic complications, while the lower (100 mg BID) dabigatran dosage regimen was associated with a significantly lower rate of haemorrhagic complications than that associated with warfarin treatment. Furthermore, both dabigatran regimens were associated with significantly lower rates of intracranial haemorrhage than those occurring with warfarin treatment. Minor concerns were an excess of gastrointestinal tract bleeds and a trend towards an increased rate of acute myocardial infarction. An analysis of the impact of variable warfarin control, as measured by time in the therapeutic range, on the results of RE-LY [2], suggested that these results were large, but not entirely, independent of individual centres' quality of INR control.(2) ROCKET-AF. This study was a double-blind randomized comparison of the Factor Xa inhibitor rivaroxaban against warfarin, conducted in a population of 14,264 relatively high risk patients with non-valvular AF [3]. Importantly, rivaroxaban was given as a once-daily dosage of 20 mg (or 15 mg in patients with moderate impairment of renal function). Patients with CHADS 2 scores of 1 were excluded: 87% of patients had CHADS 2 scores of ≥ 3.The primary end-point of the study (stroke) systemic embolism occurred at a rate of 1.7% per annum on rivaroxaban and 2.2% per annum on warfarin, clearly establishing noninferiority of rivaroxaban. However, this primary analysis was a per-protocol, on-treatment method. While the incidence of intracranial haemorrhage was significantly lower with rivaroxaban therapy, there was no overall reduction in the principal...