Background: Invasive treatments for atrial fibrillation (AF) pose a risk of ischemic stroke due to periprocedural brain embolization, which may be manifest or silent. The primary aim of our study was to compare the rate of silent strokes after percutaneous catheter-based and thoracoscopic epicardial ablation for AF. The secondary aim was to evaluate the development of silent strokes over time.Methods: We included 39 subjects (aged 64.1±8.9 years) treated for persistent symptomatic AF with thoracoscopic ablation and 30 subjects (aged 64.1±10.5 years) treated for paroxysmal or persistent symptomatic AF with catheter ablation. Subjects underwent brain MRI before and early after the ablation, moreover, the surgical group underwent late MRI 6 months after therapy. On early MRI, the presence of silent strokes and their number and size were evaluated. On late MRI, transformation of previously-detected acute ischemic lesions into chronic infarction or their reversibility were assessed.Results: Initially, different chronic ischemic findings were found in 64% of patients from the surgical group and in 70% from catheter group. Early MRI results: acute ischemic lesions were detected in 2 (6.7%) subjects (overall 3 lesions sized <5 mm) in the catheter group and in 17 (43.6%) subjects in surgical group.Most subjects in the surgical group showed multiple lesions (88%); 195 lesions were detected, a median 6 (IQR 8) lesions per case. Eighty-two percent of lesions were <5 mm, 12% 5-10 mm, 5% 10-30 mm, and 2% were large territorial ischemia. Only 1 case was symptomatic, the rest were silent strokes. On late MRI, 53.5% of all acute lesions were reversible. Lesions <5 mm were reversible in 63.1% of cases, lesions 5-10 mm were reversible in 21.7% and all lesions larger than 10 mm persisted. In 29.4% of patients all acute ischemic lesions were fully reversible.Conclusions: Periprocedural silent strokes were significantly more common after thoracoscopic epicardial ablation compared to catheter ablation considering both the number of affected patients and number of lesions. The majority of acute ischemic brain lesions were small, up to 5 mm in diameter, roughly half of which were reversible. Reversibility of acute ischemic lesions decreased with size. However, in 29.4% of affected patients, all lesions were fully reversible.