• Atrial fibrillation
• device closure
• percutaneous intervention
• stroke prevention

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, affecting 3–5% of the population aged between 65 and 75 years and increasing to 8% for those over 80 years of age.1–4 The major cause of serious disability and death in patients with AF is embolic stroke. The annual stroke rate varies from 2% to more than 20% depending on the age and comorbidity of the population studied.1 Most ischaemic strokes associated with AF originate in the atria, and transoesophageal echocardiographic (TOE) studies suggest that the vast majority (>90%) of all thrombi related to stroke in patients with AF originate in the left atrial appendage (LAA).5

Current UK National Institute for Health and Clinical Excellence (NICE) guidelines recommend oral anticoagulants as the mainstay of therapy to reduce stroke risk in AF.6 Adjusted-dose warfarin reduces stroke by approximately 60% and death by approximately 25% compared with no antithrombotic treatment.7 However, warfarin therapy is markedly underused in patients with AF for several reasons, including compliance issues, contraindications, the need for regular international normalised ratio (INR) measurements, the narrow therapeutic window and also a fear of bleeding complications.8–13 In practice, the estimated number of patients with AF adequately receiving anticoagulant therapy is less than 50%,14 and may be still lower in developing countries where the management of anticoagulation is often problematical.15

More recently dedicated percutaneous devices designed to occlude the LAA have been introduced. The hypothesis for their use is that occlusion of the LAA may prevent thrombus formation in the appendage and thus reduce the risk of thromboembolic stroke. Initial registry data and randomised studies of these devices have shown encouraging results as an alternative to warfarin therapy for selected patients. In this paper we will review the currently available LAA occlusion devices and the evidence …