Atrial fibrillation (AF) is one of the most common cardiac arrhythmias and cardioembolic stroke due to AF is its major complication. Direct oral anticoagulants (DOAC) reduce the risk of cardioembolism in patients with AF. Despite DOAC therapy, there is a significant residual stroke risk of 1-2%/year. Recent data from the Swiss Stroke Registry found 38% of patients with AF and ischemic stroke were on prior anticoagulant therapy (approximately 400 patients per year in Switzerland). The investigators found in a prior observational study, that patients with AF who have ischemic stroke despite anticoagulation are at increased risk of having another ischemic stroke (HR 1.6; 95% confidence interval, CI 1.1-2.1). Combining observational data from 11 international stroke centres, the investigators found that the majority of ischemic strokes despite anticoagulation in patients with AF is "breakthrough" cardioembolism (76% of patients) and only a minority of 24% is related to other causes unrelated to AF. Optimal secondary prevention strategy is unknown. The investigators have conducted two independent observational studies including together >4000 patients but did not identify any strategy (e.g. switch to different DOAC, additional antiplatelet therapy) that seems superior. A recent randomized controlled trial on surgical occlusion of the left atrial appendage (LAAO) found that LAAO may provide additional protection from ischaemic stroke in addition to oral anticoagulation. Triggered by this finding, the investigators performed a matched retrospective observational study and found that patients with AF and stroke despite anticoagulation who received a combined mechanical-pharmacological therapy (DOAC therapy + LAAO) had lower rates of adverse outcomes compared to those with DOAC therapy alone. Therefore, the investigators hypothesize that in patients with AF and ischemic stroke despite anticoagulant therapy, LAAO in addition to anticoagulation with a DOAC is superior to DOAC therapy alone. The investigators propose an international, multi-center randomized controlled two-arm trial to assess the effect of LAAO in patients with AF suffering from strokes despite anticoagulation therapy and without competing stroke etiology. The investigators will use the PROBE design with blinded endpoint assessment. The investigators will enrol patients with non-valvular AF and a recent ischemic stroke despite anticoagulation therapy at stroke onset. Patients will be randomized 1:1 to receive LAAO + DOAC therapy (experimental arm) or DOAC therapy alone (standard treatment arm). The primary endpoint is the first occurrence of a composite outcome of recurrent ischemic stroke, systemic embolism and cardiovascular death during follow-up. Secondary outcomes include individual components of the primary composite outcome, safety outcomes (i.e. symptomatic intracranial haemorrhage, major extracranial bleeding, serious device- or procedure-related complication), functional outcome (modified Rankin Scale) and patient-oriented outcomes. The minimum follow-up is 6 months and all patients will receive follow-ups every 6 months until end of study, the maximal follow-up will be 48 months. Based on prior observational data from the investigators' group and others (5 observational studies, >5000 patients), the investigators estimate the proportion of patients with the primary outcome in the standard treatment arm to be 18% in the first year and 9% in the second year (=cumulative 27% after 2 years). A relative risk reduction of 40% at 2 years would be clinically relevant. Based on these assumptions and a log-rank test, the investigators would need 98 events for a power of 80% at an alpha-level of 5%. Assuming a recruitment rate of 52, 118, 156 and 156 patients in years 1 to 4, an additional 6 months of follow-up (mean follow-up time of 2.1 years) and a uniform drop-out rate of 7.5% per year, 482 patients would need to be enrolled. How to treat patients with an ischemic stroke despite anticoagulation is a major yet unresolved clinical dilemma. This trial has the potential to answer the question whether LAAO plus DOAC therapy is superior to current standard of care for patients with AF who have ischemic stroke despite anticoagulation.

Inclusion Criteria:

• Age ≥ 18 years
• Written informed consent
• Permanent, persistent, or paroxysmal spontaneous AF previously known or diagnosed during the index hospitalization.
• Recent (≤3 months) symptomatic ischemic stroke.
• Active and ongoing anticoagulation therapy at stroke onset assessed based on medical history (i.e. any therapeutic oral anticoagulation therapy [Vitamin K antagonist/DOAC according to prescription recommendations for AF; inadequate low-dose DOAC therapy allowed for inclusion] not stopped/paused for >48 hours due to any reason, i.e. medical intervention or non-adherence).
• Active or planned long-term therapy with DOAC

Exclusion Criteria:

• Contraindications to DOAC therapy
• Life expectancy <1 year according to the opinion of the investigator
• Stroke due to: Ipsilateral intra/extracranial high-grade stenosis, Isolated lacunar stroke, Other well-defined stroke aetiologies (i.e., endocarditis, vasculitis, Reversible Cerebral Vasoconstriction Syndrome [RCVS], Posterior Reversible Encephalopathy Syndrome [PRES], cerebral sinus venous thrombosis)
• Previous persistent foramen ovale or atrial septum defect closure.
• Rheumatic heart disease
• Severe heart valve disease that requires treatment (severe aortic stenosis or regurgitation, severe mitral stenosis or regurgitation).
• Contraindications for TEE (relevant esophageal varices, esophageal stricture, history of esophageal cancer).
• Cardiac or non-cardiac surgical procedure within 30 days of randomization
• Enrolled in another investigation of a cardiovascular device or investigating secondary prevention therapy.
• Severely reduced Left Ventricular Ejection Fraction (LVEF) <30%.
• Severe renal impairment as described in the summary of medicinal product characteristics for the chosen DOAC (e.g. rivaroxaban, apixaban and edoxaban creatinine clearance <15 ml/min; dabigatran creatinine clearance <30 ml/min).
• Hypertrophic cardiomyopathy
• Intracardiac tumor
• Ventricular thrombus
• Acute cardiac decompensation
• LAA is obliterated or surgically ligated
• Persistent proximal LAA thrombus despite 4 weeks of anticoagulation (if a proximal thrombus in the LAA is found, anticoagulation with vitamin K antagonist (INR 2.5-3.5) may be started, and if the thrombus disappears, the patient may be eligible for LAAO)
• Pregnancy or breastfeeding (pregnancy test in urine or blood to be performed at screening for women of childbearing potential)