Nonvitamin-K-antagonist oral anticoagulants versus warfarin in patients with atrial fibrillation and previous stroke or transient ischemic attack: An updated systematic review and meta-analysis of randomized controlled trials.

Background In a previous systematic review and meta-analysis, we assessed the efficacy and safety of nonvitamin-K antagonist oral anticoagulants versus warfarin in patients with atrial fibrillation and stroke or transient ischemic attack. Since then, new information became available. Aim The aim of the present work was to update the results of the previous systematic review and meta-analysis. Methods We searched PubMed until 24 August 2016 for randomized controlled trials using the following search items: "atrial fibrillation" and "anticoagulation" and "warfarin" and "previous stroke or transient ischemic attack." Eligible studies had to be phase III trials in patients with atrial fibrillation comparing warfarin with nonvitamin-K antagonist oral anticoagulants currently on the market or with the intention to be brought to the market in North America or Europe. The outcomes assessed in the efficacy analysis included stroke or systemic embolism, stroke, ischemic or unknown stroke, disabling or fatal stroke, hemorrhagic stroke, cardiovascular death, death from any cause, and myocardial infarction. The outcomes assessed in the safety analysis included major bleeding, intracranial bleeding, and major gastrointestinal bleeding. We performed fixed effects analyses on intention-to-treat basis. Results Among 183 potentially eligible articles, four were included in the meta-analysis. In 20,500 patients, compared to warfarin, nonvitamin-K antagonist oral anticoagulants were associated with a significant reduction of stroke/systemic embolism (relative risk reduction: 13.7%, absolute risk reduction: 0.78%, number needed to treat to prevent one event: 127), hemorrhagic stroke (relative risk reduction: 50.0%, absolute risk reduction: 0.63%, number needed to treat: 157), any stroke (relative risk reduction: 13.1%, absolute risk reduction: 0.7%, number needed to treat: 142), and intracranial hemorrhage (relative risk reduction: 46.1%, absolute risk reduction: 0.88%, number needed to treat: 113) over 1.8-2.8 years. Conclusions This updated meta-analysis in 20,500 atrial fibrillation patients with previous stroke or transient ischemic attack shows that compared to warfarin non-vitamin-K antagonist oral anticoagulants are associated with a significant reduction of stroke, stroke or systemic embolism, hemorrhagic stroke, and intracranial bleeding.