This trial is designed to examine the hypothesis that thoracoscopic LAA clipping is superior to NOACs for stroke, systemic embolism, all-cause mortality, major bleeding events and clinically relevant nonmajor bleeding events in AF patients at high risk of embolism (CHA2DS2-VASc ≥2 in men and ≥3 in women) that are not undergoing ablation.

Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia disease, and the incidence of AF increases markedly with age and approximately doubles with each decade. According to the previous study, the incidence of stroke were five times in patients with AF than those in the general population. Systemic oral anticoagulant is a well-established, guideline-recommended therapy for the prevention of ischemic stroke in patients with nonvalvular AF at high risk of embolism (CHA2DS2-VASc scores ≥2 in men and ≥3 in women), and the guidelines recommend that the novel oral anticoagulants (NOACs) be preferred (Class I, Level of evidence A). However, a significant proportion of patients with nonvalvular AF have difficulties in long-term oral anticoagulant therapy, due to medication adherence and contraindications to oral anticoagulants. Also, several randomized controlled trials indicated that bleeding risk remained high with novel oral anticoagulants. Therefore, it is essential to explore alternative treatment strategies for stroke prevention in patients with nonvalvular AF.

It has been reported that the left atrial appendage (LAA) is suspected as a vital source of cerebral emboli and may lead to ischemic stroke, removal or closure of the LAA may be an alternative to oral anticoagulants. Various surgical or interventional approaches have been developed to close or occlude LAA to prevent stroke in AF patients, such as percutaneous LAA occlusion, suture ligation, and surgical excision. However, these techniques suffer from incomplete LAA closure or the presence of residual blood flow, which can lead to thrombosis and stroke. Thoracoscopic LAA clip, on the other hand, cloud block blood flow between the LAA and the left atrium (LA), achieving isolation of LAA and preventing thrombi and strokes. A previous study has demonstrated a high 95% success rate of LAA clipping without operation-related complications, and freedom from stroke was 99.1% at a median follow-up of 20 months. Therefore, LAA clipping is an effective and durable method in stroke prevention. However, currently high-quality RCTs are lacking to support the superiority of LAA clipping compared with NOACs in terms of stroke prevention and safety. In this trial, the investigators designed a multicenter prospective RCT to compare the efficacy and safety of thoracoscopic LAA clipping and NOACs in patients with non-paroxysmal AF.

Inclusion Criteria:

• Age ≥ 18 years.
• Persistent or long-standing persistent AF documented by medical history or direct electrocardiogram.
• CHA2DS2-VASc ≥2 in men and ≥3 in women.
• Agree to perform thoracoscopic LAA occlusion procedure.

Exclusion Criteria:

• With electrical cardioversion or ablation intent.
• Other heart diseases with surgical indications.
• Ischemic stroke and other cardiac embolic events within 30 days.
• Major clinical bleeding event within 30 days.
• Contraindications to anticoagulation.
• Intracardiac thrombus.
• Left ventricular ejection fraction (LVEF) < 30%.
• Active systemic infection or infective endocarditis or pericarditis
• Severe liver disease (acute clinical hepatitis, chronic active hepatitis, cirrhosis) or alanine transaminase (ALT)/ aspartate transaminase (AST) greater than 3 times the upper limit of normal value.
• Severe renal insufficiency (eGFR ≤ 30mL/min).
• Other diseases requiring oral anticoagulants.
• Active aortic plaque.
• Acute coronary syndrome within 3 months.
• Symptomatic carotid artery stenosis.
• Patients requiring dual antiplatelet drug therapy.
• Previous cardiac and left lung surgery.
• Severe left pleural and pericardial adhesions.
• Pregnant or breastfeeding patients.
• Metal allergies.
• Terminal illness with a life expectancy of less than 2 years.
• Participation in other clinical studies at the time of enrollment.
• Refuse to participate in this study.