Aim. To study the outcomes frequency and structure in patients with atrial fibrillation (AF) depending on the cardioembolic events preventing method: left atrial appendage (LAA) isolation, direct oral anticoagulants (DOACs) or warfarin.Material and methods. A prospective observational study included patients with AF and high risk of cardioembolic complications and without contraindications to anticoagulants. Patients who refused long-term oral anticoagulants taking underwent LAA isolation, the rest of the patients received DOACs or warfarin. The observation period was 3 years. Mortality, cardioembolic complications and major bleeding (according to GARFIELD criteria) cumulative incidence was assessed.Results. We included 245 patients: 46 patients were treated with LAA isolation, 100 with warfarin, and 99 with DOACs. Multivariate regression analysis demonstrated a statistically significant advantage of LAA occluder in terms of combined endpoint achieving frequency compared to warfarin (hazard ratio [HR] 3.10; 95% confidence interval [CI] 1.01-9.54; p=0.049), and to DOACs (HR 3.44, 95% CI 1.15-10.29; p=0.027). A similar result was obtained for all-cause mortality (HR 5.24; 95% CI 1.12-24.55; p=0.036 and HR 5.58; 95% CI 1.22-25.49; p=0.027, respectively). There were no significant differences in bleeding rates between the groups.Conclusion. This observational study demonstrates the superiority of LAA isolation as a first-line therapy over DOACs and warfarin in patients with AF and high risk of cardioembolic complications. Randomized trials are required to confirm these observations.