Atrial fibrillation ablation using remote magnetic navigation and the risk of atrial-esophageal fistula: international multicenter experience.

PURPOSE: Remote magnetic navigation (RMN) has been used in various electrophysiological procedures, including atrial fibrillation (AF) ablation. Atrial-esophageal fistula (AEF) is one of most disastrous complications of AF ablation. We aimed to evaluate the incidence of AEF during AF ablation using RMN in comparison to manual ablation. METHODS: We conducted the first international survey among RMN operators for assessment of the prevalence of AEF and procedural parameters affecting the risk. Data from parallel survey of AEF among Canadian interventional electrophysiologists (CIE) using only manual catheters served as control. RESULTS: Fifteen RMN operators (who performed 3637 procedures) and 25 manual CIE operators (7016 procedures) responded to the survey. RMN operators were more experienced than CIE operators (16.3 ± 8.3 vs. 9.2 ± 5.4 practice years in electrophysiology, p = 0.007). The maximal energy output in the posterior wall was higher in the operator using RMN (33 ± 5 vs. 28.6 ± 4.9 W; p = 0.02). Other parameters including use of preprocedural images, irrigated catheter, pump flow rate, esophageal temperature monitoring, intracardiac echocardiography (ICE), and general anesthesia were similar. CIE operators administered proton-pump inhibitors postoperatively significantly more than RMN operators (76 vs. 35%, p = 0.01). AEF was reported in 5 of the 7016 patients in the control group (0.07%) but in none of the RMN group (p = 0.11). CONCLUSIONS: AEF is a rare complication and its evaluation necessitates large-scale studies. Although no AEF case with RMN was reported in this large study or previously on the literature, the rarity of this complication prevents firm conclusion about the risk.