Abstract 16411: Incidence of Atrial-Esophageal Fistula in Cryoballoon Ablation for Atrial Fibrillation is Dose Dependent
Introduction: Cryoballoon ablation has proven effectiveness in treating atrial fibrillation, yet the iterations of Medtronic cryoballoon first generation (CB1) and second generation (CB2) have raised concerns about safety. While Atrial-Esophageal fistula (AEF) has a historically low incidence, its near universally fatal outcome underscores the importance of effective and safe use of the cryoballoon technology. We hypothesized that incidence of AEF in cryoablation is dose dependent.
Methods: Medtronic is the manufacturer of the cryoballoon system, and maintains an active (albeit proprietary) database of use. Data were obtained from this database with the assistance of Medtronic staff. Data included cryoballoon type, duration and quantity of ablation, and reported AEF incidence for each group. Incidence of AEF was compared between groups, comparing CB1 (4 minutes duration, 2 applications), CB2 longer duration (4 minutes, 2 applications), and CB2 shorter duration (3 minutes, 2 applications). Statistical analyses with Relative Risk and chi-squared testing were performed.
Results: From January 2006 to May 2015, approximately 200,000 cryoablation cases were performed: 50,000 using the CB1 and 150,000 using CB2. For CB2 30,000 were performed using a recommended 4-minute ablation time (freeze-thaw-refreeze technique), and 120,000 were performed with a 3-minute or less ablation time (freeze-thaw-refreeze technique). A total of 11 incidents of AEF were reported. CB1 was noted to have 3/50,000 cases of AEF (0.006%). CB2 was noted to have 7/30,000 (0.023%) with the 4-minute ablation time, and 1/120,000 (0.0008%) cases with the 3-minute or less ablation time. Relative risk of CB2 (4 minute) compared to CB2 (3 minute) was calculated at 28, with chi squared p<0.001.
Conclusion: While other comorbid conditions may play a role in development of AEF (such as gastroesophageal reflux disease, esophageal trauma, and perforation), duration of ablation appears to have a role as well. While data is approximate, owing to use in less developed parts of the world, the magnitude of AEF reduction is impossible to ignore. In conclusion, dose reduction in cryoballoon ablation appears to have a significant impact on reduction of AEF and must be investigated further.
Author Disclosures: A. Williams: None. Z. Ahmad: None. M. Al-Zubaidi: None. W. Su: Research Grant; Modest; St. Jude Medical. Research Grant; Significant; Medtronic Inc. Honoraria; Modest; St Jude Medical. Honoraria; Significant; Medtronic Inc.
- © 2016 by American Heart Association, Inc.