Author + information
- David S. Frankel, MD∗ ()
- Cardiovascular Division, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania
- ↵∗Address for correspondence:
Dr. David S. Frankel, Cardiovascular Division, Perelman School of Medicine at the University of Pennsylvania, 3400 Spruce Street, 9 Founders Pavilion, Philadelphia, Pennsylvania 19104.
Atrial esophageal (AE) fistula is perhaps the most feared complication of atrial fibrillation (AF) ablation. Although the incidence is low, mortality remains strikingly high. For example, a global survey of 405 cardiac electrophysiologists performing nearly 200,000 AF ablations identified 31 patients with esophageal perforations, including 22 with AE fistula (1). Despite urgent surgery or esophageal stenting in the majority, only 4 patients (18%) survived without significant neurological injury. In a more recent single-center registry, 832 patients underwent surveillance esophageal endoscopy shortly following AF ablation (2). Esophageal lesions were detected in 150 patients (18%), including 98 with just erythema or erosion and 52 with ulceration. Of the 52 ulcerations, 5 progressed to esophageal perforations; 2 of these 5 patients died.
Thus protecting the esophagus remains an essential priority in AF ablation. Strategies used to reduce the risk of esophageal thermal injury include decreasing ablation lesion power and duration along the posterior left atrium, monitoring the luminal esophageal temperature (LET), and irrigating the esophagus with cooled water (3). It is humbling that the majority of AE fistulas reported in the global survey occurred despite monitoring the LET and that all the esophageal thermal lesions reported in the single-center registry occurred despite reducing ablation lesion power along the posterior left atrium. Further, when LET rises are detected, the operator typically waits for the temperature to return to baseline and further decreases ablation lesion power and/or duration, thus prolonging the procedure and potentially decreasing efficacy.
Laterally displacing the esophagus from the path of ablation is therefore an appealing option. This was initially accomplished using familiar tools, including endoscopes and transesophageal echocardiography probes (4,5). Koruth et al. (6) subsequently reported intubating the esophagus with a standard thoracic chest tube and a malleable, endotracheal stylet. Although these investigators were able to deviate the esophagus and decrease LET rises, oropharyngeal trauma occurred in 63% of patients. Thus the need was established for a safer tool to displace the esophagus mechanically.
To meet this clinical need, the DV8 inflatable balloon retractor (Manual Surgical Sciences, Inc., Minneapolis, Minnesota) was developed. When deflated, the balloon is straight, facilitating esophageal intubation. When inflated, the midportion of the balloon deviates laterally. In this issue of JACC: Clinical Electrophysiology, Bhardwaj et al. (7) present a series of 200 consecutive patients undergoing AF ablation, with the DV8 balloon retractor used to mechanically deviate the esophagus away from ablation lesion sets. Contrast material was injected to delineate the esophageal contour, and then a multisensor temperature probe was inserted along the trailing edge of the esophagus (i.e., as close as possible to the ablation lesion set) to monitor the LET. Next, the balloon was inflated and rotated to deviate the esophagus maximally away from the intended lesion set. Left atrial geometry was then created, and the first pair of pulmonary veins was isolated. The balloon was then repositioned to deviate the esophagus maximally away from the intended lesion set around the remaining pulmonary vein pair. Left atrial geometry was recreated with the new balloon position, and the second pulmonary vein pair was isolated.
The MEDEffective was defined as the minimum distance between the trailing edge of the esophagus and the closest point along the ablation line. Using the inflatable balloon retractor, the operators were able to achieve MEDEffective of 21 mm for the right pulmonary veins and 16 mm for the left pulmonary veins. This extent of lateral esophageal deviation was superior to that achieved in the previous study using a malleable stylet (6). Not surprisingly, the greater the achieved MEDEffective, the less likely the LET was to rise. Specifically, when the MEDEffective was 0 to 5 mm, the LET universally rose. Conversely, when MEDEffective was >20 mm, the LET rose during only 1.9% of lesion sets.
Several limitations to this study should be considered. Esophageal endoscopy was not routinely performed, and thus we must rely on the surrogate endpoints of extent of esophageal deviation and LET rise. Nevertheless, these surrogates are appropriate and likely to correlate with the risk of AE fistula. Additionally, the operators did not decrease ablation lesion settings along the posterior wall, delivering 35 to 40 W for 20 to 40 s. It is therefore likely that with lower-power ablation lesions, less mechanical esophageal deviation would be required to prevent LET rises. Two patients developed oropharyngeal bleeding following endotracheal extubation, a substantial improvement over the stiff, endotracheal stylet. Perhaps additional operator experience and design optimization will further decrease oropharyngeal trauma. Finally, the additional time required to create new left atrial geometries with each balloon retractor position must be considered. This cost in time may well be offset by time saved during ablation, if the operator can avoid repeatedly waiting for LET to return to baseline following rises.
The present study by Bhardwaj et al. (7) is an important step forward in esophageal protection during AF ablation, reporting the results of a tool specifically designed for this purpose. The investigators should be congratulated not only for this contribution, but also for an entire body of work on this important topic. Nevertheless, questions remain to be answered. Is it better to reduce ablation lesion power and monitor the LET or mechanically deviate the esophagus? Although reducing ablation lesion power may protect the esophagus, it could also lead to higher rates of chronic pulmonary vein reconnection (8). The impact on procedural workflow should also be examined, balancing the need for creating multiple left atrial geometries with decreasing waiting periods for LET to return to baseline. What is the best tool to displace the esophagus mechanically? A recent series of 704 AF ablations using a transesophageal echocardiography probe to move the esophagus reported even greater lateral displacements and comparable safety (9). What is the role for medical esophageal protection, such as proton pump inhibitors, following ablation? What is the long-term impact of esophageal thermal injuries that do not progress to AE fistula? Further efforts towards answering these questions, protecting the esophagus, and minimizing the rare but lethal complication of AE fistula are warranted.
↵∗ Editorials published in JACC: Clinical Electrophysiology reflect the views of the author and do not necessarily represent the views of JACC: Clinical Electrophysiology or the American College of Cardiology.
Dr. Frankel has reported that he has no relationships relevant to the contents of this paper to disclose.
The author attests he is in compliance with human studies committees and animal welfare regulations of the author’s institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- 2018 American College of Cardiology Foundation
- Barbhaiya C.R.,
- Kumar S.,
- Guo Y.,
- et al.
- Halbfass P.,
- Pavlov B.,
- Muller P.,
- et al.
- Bhardwaj R.,
- Naniwadekar A.,
- Whang W.,
- et al.
- Tran V.N.,
- Kusa S.,
- Smietana J.,
- et al.
- Mateos J.C.,
- Mateos E.I.,
- Pena T.G.,
- et al.