BACKGROUND - In patients with non-valvular atrial fibrillation, embolic stroke is thought to be associated with left atrial appendage (LAA) thrombi. We assessed the efficacy and safety of percutaneous closure of the LAA for prevention of stroke compared with warfarin treatment in patients with atrial fibrillation.

METHODS - Adult patients with non-valvular atrial fibrillation were eligible for inclusion in this multicentre, randomised non-inferiority trial if they had at least one of the following: previous stroke or transient ischaemic attack, congestive heart failure, diabetes, hypertension, or were 75 years or older. 707 eligible patients were randomly assigned in a 2:1 ratio by computer-generated randomisation sequence to percutaneous closure of the LAA and subsequent discontinuation of warfarin (intervention; n=463) or to warfarin treatment with a target international normalised ratio between 2.0 and 3.0 (control; n=244). Efficacy was assessed by a primary composite endpoint of stroke, cardiovascular death, and systemic embolism. We selected a one-sided probability criterion of non-inferiority for the intervention of at least 97.5%, by use of a two-fold non-inferiority margin. Serious adverse events that constituted the primary endpoint for safety included major bleeding, pericardial effusion, and device embolisation. Analysis was by intention to treat. This study is registered with Clinicaltrials.gov, number NCT00129545.

FINDINGS - At 1065 patient-years of follow-up, the primary efficacy event rate was 3.0 per 100 patient-years (95% credible interval [CrI] 1.9-4.5) in the intervention group and 4.9 per 100 patient-years (2.8-7.1) in the control group (rate ratio [RR] 0.62, 95% CrI 0.35-1.25). The probability of non-inferiority of the intervention was more than 99.9%. Primary safety events were more frequent in the intervention group than in the control group (7.4 per 100 patient-years, 95% CrI 5.5-9.7, vs 4.4 per 100 patient-years, 95% CrI 2.5-6.7; RR 1.69, 1.01-3.19).

INTERPRETATION - The efficacy of percutaneous closure of the LAA with this device was non-inferior to that of warfarin therapy. Although there was a higher rate of adverse safety events in the intervention group than in the control group, events in the intervention group were mainly a result of periprocedural complications. Closure of the LAA might provide an alternative strategy to chronic warfarin therapy for stroke prophylaxis in patients with non-valvular atrial fibrillation.

FUNDING - Atritech.