Silent cerebral events/lesions after second-generation cryoballoon ablation: How can we reduce the risk of silent strokes?

Background Atrial fibrillation (AF) ablation is associated with a substantial risk of silent cerebral events/lesions (SCEs/SCLs) detected on magnetic resonance imaging (MRI). Objective The purpose of this study was to investigate the factors associated with the incidence of SCEs/SCLs during second-generation cryoballoon ablation. Methods Two hundred fifty-six AF patients underwent brain MRI 1 day after pulmonary vein (PV) isolation using second-generation cryoballoons with a single 28-mm balloon and short freeze strategy. Results Overall, 991 of 1016 PVs (97.5%) were successfully isolated by 4.9 ± 1.3 cryoballoon applications per patient, and 25 PVs required touch-up radiofrequency ablation. The total procedure time was 72.7 ± 26.1 minutes. SCEs and SCLs were detected in 68 (26.5%) and 27 (10.5%) patients, respectively. None of the patients reported any neurologic symptoms. Reinsertion of once withdrawn cryoballoons and subsequent applications significantly increased the incidence of SCEs (odds ratio [OR] 2.057; 95% confidential interval [CI] 1.051–4.028; P = .035), and additional left atrial mapping with a multielectrode catheter significantly increased the incidence of SCLs (OR 3.317; 95% CI 1.365–8.056; P = .008). Transient coronary air embolisms were significantly associated with the incidence of SCLs (OR 3.447; 95% CI 1.015–11.702; P = 0.047). On the contrary, an uninterrupted anticoagulation regimen, use of radiofrequency deliveries for transseptal access, cryoballoon air removal with extracorporeal balloon inflations, strength of the MRI magnet, internal electrical cardioversion, and touch-up ablation were not associated with the incidence of SCEs/SCLs. Conclusion A significant number of SCE/SCL occurrences was observed after second-generation cryoballoon ablation procedures. These results suggest that air embolisms are the main mechanism of SCEs/SCLs, and the injected air volume might determine the lesion type.