MRI-Detected Brain Lesions and Cognitive Function in Patients With Atrial Fibrillation Undergoing Left Atrial Catheter Ablation in the Randomized AXAFA-AFNET 5 Trial

Karl Georg Häusler*, Felizitas A Eichner, Peter U Heuschmann, Jochen B Fiebach, Tobias Engelhorn, Benjamin Blank, David Callans, Arif Elvan, Massimo Grimaldi, Jim Hansen, Gerhard Hindricks, Hussein R Al-Khalidi, Lluís Mont, Jens Cosedis Nielsen, Jonathan P Piccini, Ulrich Schotten, Sakis Themistoclakis, Johan Vijgen, Luigi Di Biase, Paulus Kirchhof

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


Background: We aimed to assess the prevalence of ischemic brain lesions detected by magnetic resonance imaging and their association with cognitive function 3 months after first-time ablation using continuous oral anticoagulation in patients with paroxysmal atrial fibrillation (AF). Methods: We performed a prespecified analysis of the AXAFA-AFNET 5 trial (Anticoagulation Using the Direct Factor Xa Inhibitor Apixaban During Atrial Fibrillation Catheter Ablation: Comparison to Vitamin K Antagonist Therapy), which randomized 674 patients with AF 1:1 to uninterrupted apixaban or vitamin K antagonist therapy before first-time ablation. Brain magnetic resonance imaging using fluid-attenuated inversion recovery and high-resolution diffusion-weighted imaging was obtained within 3 to 48 hours after AF ablation in all eligible patients enrolled in 25 study centers in Europe and the United States. Patients underwent cognitive assessment 3 to 6 weeks before ablation and 3 months after ablation using the Montreal Cognitive Assessment (MoCA). Results: In 84 (26.1%) of 321 patients with analyzable magnetic resonance imaging, high-resolution diffusion-weighted imaging detected at least 1 acute brain lesion, including 44 (27.2%) patients treated with apixaban and 40 (24.8%) patients treated with vitamin K antagonist (P=0.675). Median MoCA score was similar in patients with or without acute brain lesions at 3 months after ablation (28 [interquartile range (IQR), 26-29] versus 28 [IQR, 26-29]; P=0.948). Cerebral chronic white matter damage (defined as Wahlund score ≥4 points) detected by fluid-attenuated inversion recovery was present in 130 (40.5%) patients and associated with lower median MoCA scores before ablation (27 [IQR, 24-28] versus 27 [IQR, 25-29]; P=0.026) and 3 months after ablation (27 [IQR, 25-29] versus 28 [IQR, 26-29]; P=0.011). This association was no longer significant when adjusted for age and sex. Age was associated with lower MoCA scores before ablation (relative risk, 1.02 per 10 years [95% CI, 1.01-1.03]) and 3 months after ablation (relative risk, 1.02 per 10 years [95% CI, 1.01-1.03]). Conclusions: Chronic white matter damage as well as acute ischemic lesions detected by brain magnetic resonance imaging were found frequently after first-time ablation for paroxysmal AF using uninterrupted oral anticoagulation. Acute ischemic brain lesions detected by high-resolution diffusion-weighted imaging were not associated with cognitive function at 3 months after ablation. Lower MoCA scores before and after ablation were associated only with older age, highlighting the safety of AF ablation on uninterrupted oral anticoagulation. Registration: URL:; Unique identifier: NCT02227550.

Original languageEnglish
Pages (from-to)906-915
Number of pages10
Issue number12
Early online date9 Feb 2022
Publication statusPublished - 22 Mar 2022


  • anticoagulants
  • atrial fibrillation
  • magnetic resonance imaging
  • Anticoagulants
  • Magnetic resonance imaging
  • Atrial fibrillation

Cite this