Author + information
- Received December 5, 2017
- Revision received March 5, 2018
- Accepted April 5, 2018
- Published online August 20, 2018.
- Rahul Bhardwaj, MDa,b,
- Aditi Naniwadekar, MDa,
- William Whang, MDa,
- Alexander J. Mittnacht, MDa,
- Chandrasekar Palaniswamy, MDa,c,
- Jacob S. Koruth, MDa,
- Kamal Joshi, MDa,
- Aamir Sofi, MDa,
- Marc Miller, MDa,
- Subbarao Choudry, MDa,
- Srinivas R. Dukkipati, MDa and
- Vivek Y. Reddy, MDa,∗ ()
- aHelmsley Electrophysiology Center, Icahn School of Medicine at Mount Sinai, New York, New York
- bDepartment of Cardiology, Loma Linda University, Loma Linda, California
- cDepartment of Cardiology, University of California, San Francisco–Fresno, Fresno, California
- ↵∗Address for correspondence:
Dr. Vivek Y. Reddy, Helmsley Electrophysiology Center, Icahn School of Medicine at Mount Sinai, One Gustave L. Levy Place, Box 1030, New York, New York 10029.
Objectives The goal of this study was to determine the safety and feasibility of a novel esophageal balloon retractor (DV8) for MED during PVI.
Background The authors previously showed that mechanical esophageal deviation (MED) is feasible using an off-the-shelf metal stylet to allow uninterrupted ablation along the posterior left atrium during pulmonary vein isolation (PVI). Although it is an attractive strategy to avoid esophageal thermal injury, this technique was hampered by both the propensity for oropharyngeal trauma from the stiff stylet and the limited lateral esophageal displacement.
Methods In 200 consecutive patients undergoing atrial fibrillation ablation, the DV8 balloon retractor was used for MED; contrast was instilled into the esophagus to accurately delineate the trailing esophageal edge. Deviation was performed to maximize the distance from the trailing esophageal edge to the closest point of the ablation line (MEDEffective) and correlated to occurrences of luminal esophageal temperature elevation.
Results In patients undergoing MED during a first-ever PVI of 304 vein pairs, the MEDEffective during right and left PVI were 21.2 ± 8.7 mm and 15.5 ± 6.8 mm, respectively. Deviation of at least 5 mm of MEDEffective was achievable in 97.7%. Luminal esophageal temperature increases universally occurred (100%) at MEDEffective <5 mm, less often (28%) at MEDEffective 5 to 20 mm, and rarely (1.9%) at MEDEffective >20 mm. There were no esophageal complications, but 2 patients experienced oropharyngeal bleeding due to trauma related to device placement.
Conclusions MED with the balloon retractor safely moved the esophagus away from the site of energy delivery during atrial fibrillation ablation.
Catheter ablation is an effective and widely adopted treatment for symptomatic atrial fibrillation (AF). The dominant strategy for AF ablation is pulmonary vein isolation (PVI), either alone or in combination with other lesion sets in the left atrium. The effectiveness is largely dependent on the contiguity and transmurality of the circumferential lesions around the pulmonary vein. However, esophageal injury is a significant potential complication that can occur with ablation along the posterior left atrial (LA) wall, and fear of injury can be a limiting factor in performing effective ablation (1,2). In patients undergoing radiofrequency catheter ablation of AF, the reported incidence of any esophageal changes ranges from 2.2% to 48%. In a recent, large, single-center experience, the incidence of the most dreaded complication, esophageal perforation/fistula formation, occurred in 0.6% of patients undergoing AF ablation (3). Less often recognized, functional impairment of the gastrointestinal system related to damage of the esophagus and vagal plexus also occurs in up to 74% of patients, which may manifest as esophageal dysmotility, delayed gastric emptying times, or abnormal sham feeding times (4).
Minimizing damage to the esophagus during AF ablation has been explored through various methods, including monitoring the luminal temperature in the esophagus, reducing power or delivering shorter duration lesions along the posterior left atrium, and irrigating the esophagus with cool water (1,5). Despite use of these approaches, patients have nonetheless developed esophageal fistulas, and these strategies may negatively affect the long-term PVI rates and clinical success of the ablation procedure. Studies have shown that the distance between the esophagus and the ablated point is the most important predictor of esophageal temperature elevation (6,7).
One conceptually attractive method to avoid esophageal damage would be to laterally displace the esophagus away from the site of ablation. We previously reported on the feasibility and effectiveness of mechanical esophageal deviation (MED) using an off-the-shelf malleable metal stylet within a plastic tube to laterally displace the esophagus during PVI procedures (7,8). We found that the MEDEffective (defined as the distance from the trailing esophageal edge to the closest point of the ablation line) directly correlated with the incidence of esophageal heating. Although it was overall effective in most patients, this approach is limited by the following: 1) mechanical oropharyngeal trauma related to the stiff metal stylet; and 2) the limited lateral excursion of MED in a significant minority of patients (MEDEffective was <5 mm and <10 mm in 13.9% and 27.1% of patients, respectively) (7). Indeed, there was a strong relationship between safety and efficacy: if a stiffer stylet was used, a greater amount of MEDEffective was achieved but at the expense of more oropharyngeal trauma. Thus, although the concept of esophageal displacement was shown to be valid, technical advances were required to achieve adequate MED but with fewer safety concerns.
The present article reports our initial experience on the feasibility of esophageal deviation during PVI procedures using the DV8 balloon retractor, a device designed to mechanically displace the esophagus laterally away from the site of ablation.
Consecutive patients undergoing LA catheter ablation for AF as either a first-time procedure or a redo procedure (including ablation of atrial tachycardia/flutter) between March 2016 and August 2017 at our institution were prospectively studied. The full cohort included all 200 consecutive patients undergoing ablation while using the esophageal balloon retractor. The data were collected as part of the Mount Sinai Hospital Quality Assurance process, and the study was approved by the institutional review board.
Patients were excluded if there was a history of esophageal strictures or previous esophageal surgery. As per our usual practice, all patients underwent ablation under general anesthesia with a strategy of uninterrupted oral anticoagulation with either warfarin or a non-warfarin oral anticoagulant.
Intravenous unfractionated heparin was administered to maintain an activated clotting time of 300 to 400 s, and 2 transseptal punctures were performed. For first-time procedures, ablation was performed by using either of the following: 1) an approved force-sensing irrigated tip radiofrequency ablation catheter, either the ThermoCool SmartTouch SurroundFlow catheter with the CARTO3 mapping system (Biosense Webster Inc., Diamond Bar, California) or the TactiCath catheter with the NavX mapping system (Abbott, St. Paul, Minnesota); or 2) the irrigated flexible-tip radiofrequency ablation catheter (FlexAbility, Abbott). For standard ablation, the power used during ablation along the posterior left atrium was not different than used at other locations: typically, 35 to 40 W with the SmartTouch and FlexAbility catheters and 25 to 30 W with the TactiCath catheter.
As per our usual practice, circumferential PVI lesion sets were placed to encircle each ipsilateral pair of veins. Additional ablation was performed at the discretion of the operator. Esophageal temperature was monitored throughout the procedure by placing a multielectrode temperature probe (CIRCA S-Cath, CIRCA Scientific, Inc., Englewood, Colorado) along the trailing edge of the esophagus. Radiofrequency energy delivery on the posterior wall was terminated if the esophageal temperature was >38°C.
Esophageal deviation balloon retractor
The DV8 inflatable balloon retractor (Manual Surgical Sciences, Minneapolis, Minnesota) (Figure 1) is composed of a polyurethane balloon wrapped with a silicone sleeve. The device is manually inserted through the mouth into the esophagus, similar to using a conventional orogastric tube. There is a port to inject contrast external to the balloon to opacify and delineate the esophageal lumen. Dilute contrast is then injected via a second port into the balloon itself using an ordinary pressure syringe. As the balloon inflates, it takes its predefined deflection shape. A standard insufflator is used to increase the pressure in the balloon until the trailing edge of the esophagus is laterally displaced away from the anticipated points of ablation. The balloon reassumes its straight configuration upon relieving the pressure. The balloon retractor is ∼10 mm and ∼17 mm in diameter when uninflated and inflated, respectively.
The esophageal balloon retractor was used to move the esophagus laterally away from the target area of ablation after administration of intravenous heparin and transseptal puncture and before the creation of geometry. The esophageal retractor was inserted into the patient’s esophagus adjacent to the multisensory temperature probe (CIRCA S-Cath) by the operator or by anesthesiology staff. The temperature probe was positioned on the trailing edge of the esophagus to continuously measure the luminal esophageal temperature (LET) during ablation. Once the system was sufficiently advanced into the esophagus, based on fluoroscopy by radiopaque markers on the system, either liquid barium contrast (Liquid E-Z-Paque, Bracco Diagnostics Inc., Monroe Township, New Jersey) or standard iodinated contrast (Omnipaque, GE Healthcare, Chicago, Illinois) was instilled into both the esophagus (to delineate the esophageal borders) and into the balloon itself (to mechanically displace the esophagus). The temperature probe was manipulated to ensure that it was positioned lateral to the DV8 retractor closest to the esophageal trailing edge. An insufflator was used to inflate the balloon until adequate deviation was achieved, typically at ∼4 to 5 atm.
The esophagus was displaced away from the area of ablation with directionality determined by the operator based on the initial location of the esophagus with relation to the targeted pulmonary veins. After deviation, the LA geometry was created and ablation performed. After ablation of the first pulmonary vein pair, the esophageal balloon was deflated and rotated to move the esophagus to the contralateral side before ablation of the second pulmonary vein pair. As previously described, the extent of effective lateral esophageal displacement was defined as the distance between the PVI lesion set and the esophageal reference points corresponding to the trailing esophageal edge (Figure 2); the distance between the ablation point closest to the esophageal trailing edge was used. Using fluoroscopy in the anteroposterior view, reference points were placed on the electroanatomic map corresponding to the trailing esophageal edge. The esophagus could also be visualized by using intracardiac ultrasound imaging and integrated with the electroanatomic map (Figure 3).
After completing LA ablation, the esophageal balloon retractor was deflated and removed from the esophagus. The patients were extubated and transferred to the recovery area, and observed overnight on a telemetry unit. Any complications such as oral or pharyngeal bleeding, reintubation, fever, upper gastrointestinal bleeding, or any other unexpected events were recorded.
Summary statistics for demographic characteristics and stroke risk factors were calculated for the overall cohort (N = 200), and the extent of deviation was calculated within the sample of patients who underwent a first-ever PVI (n = 182). For the analysis of extent of deviation, the unit of observation was pulmonary vein pair, and left- and right-sided pulmonary vein pairs were analyzed together. The frequency of esophageal temperature increases >38°C was calculated according to the extent of deviation. We compared the extent of deviation by using the balloon retractor in the current sample versus that achieved in a previously published series of 114 consecutive patients undergoing PVI with esophageal deviation using a malleable metal stylet (7). Unpaired Student t tests and chi-square tests were used to compare continuous and categorical variables, respectively, between the 2 deviation methods. All analyses were performed by using SPSS version 22 (IBM SPSS Statistics, IBM Corporation, Armonk, New York).
Baseline patient characteristics are displayed in Table 1. Consistent with most AF ablation studies, the 200-patient cohort had a mean age of 64.2 years, and 64% were male. The majority of patients (91%) were undergoing a first-ever AF ablation procedure, whereas 9% were undergoing a redo ablation procedure. The majority of the patient population (97%) had AF, approximately equally split between paroxysmal and nonparoxysmal AF; atrial tachycardia/flutter was targeted in 3%. The mean CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke/transient ischemic attack, vascular disease, age 65 to 74 years, sex category) score was 2.2, with hypertension occurring in one-half of the group. All patients were receiving anticoagulation, with >95% receiving non-warfarin oral anticoagulant agents; the procedures were performed with a strategy of uninterrupted anticoagulation.
Force-sensing irrigated radiofrequency ablation catheters were used in >99% of patients; the cryoballoon was used in <1%. As shown in Table 2, the majority of the patients underwent electrical PVI (95.5%). As is our usual practice, the majority of the cohort also received a cavotricuspid isthmus ablation (85.5%). One quarter of patients underwent additional LA or right atrial ablation (25%).
Extent of mechanical esophageal deviation
In the 182 patients undergoing a first-ever PVI procedure, while all 384 pulmonary vein pairs were electrically isolated by using the esophageal balloon retractor, data were collected and available for analysis during PVI in 158 patients, including 304 total pulmonary vein pairs, 155 right pulmonary vein pairs, and 149 left pulmonary vein pairs. The mean MEDEffective achieved was 18.4 ± 8.7 mm away from the closest ablation point of the PVI lesion set. With respect to the individual right and left pulmonary vein lesion sets, the mean MEDEffective were 21.2 ± 8.7 mm and 15.5 ± 6.8 mm, respectively. The extent of mechanical MEDEffective was 0 to 5 mm, 6 to 10 mm, 10.1 to 15 mm, 15.1 to 20 mm, or >20 mm in 7 (2.3%), 35 (11.5%), 76 (25%), 78 (25.7%), and 108 (35.5%) instances of MED, respectively (Table 3, Figure 4). Overall, esophageal temperature increases >38°C were observed during ablation of 62 (20.4%) of the PVI lesion sets and more frequently during isolation of the left pulmonary veins than the right pulmonary veins.
During those ablation lesions associated with esophageal temperature increases, the mean MEDEffective was 11.6 ± 5.1 mm. At MEDEffective <5 mm, which occurred in 7 pulmonary vein lesion sets, there were a total of 21 lesions with LET increases; this finding translates into an average of 3 lesions with LET increases per patient. At MEDEffective 5.1 to 10 mm, 10.1 to 15 mm, 15.1 to 20 mm, and >20 mm, there were1.31, 0.57, 0.58, and 0.05 lesion with LET increases per patient, respectively. The extent of MEDEffective achieved and the associated temperature increases are displayed in Table 3 and Figure 4. As shown, esophageal temperature increases universally occurred when the MEDEffective was <5 mm, with the incidence being of intermediate frequency (28%) between 5 and 20 mm and rarely (1.9%) when the MEDEffective was >20 mm.
Esophageal deviation achieved using the balloon retractor versus stylet technique
These data with the balloon retractor (304 pulmonary vein pairs) were compared with the previously published data using the malleable metal stylet technique (n = 225 pulmonary vein pairs) (7). The mean MEDEffective was 18.2 ± 8.4 mm and 14.6 ± 8.4 mm for the balloon retractor and stylet techniques, respectively (p < 0.0001). The distribution of the MEDEffective according to quintiles using each of the 2 techniques is shown in Figure 5. Importantly, the percentage of patients in whom MED was ineffective (i.e., MEDEffective <5 mm) was significantly less frequent with the balloon retractor compared with the stylet technique (2.3% vs 12.9%, respectively; p < 0.00001). Similarly, the percentage of patients with an ideal MED (i.e., MEDEffective >20 mm) was significantly greater with the balloon retractor (35.5% vs 21.8%; p = 0.0006). Overall, temperature increases were not as common using the balloon retractor compared with the stylet technique, occurring in 20.4% and 35.5% (p = 0.0001).
Safety of MED
There were no instances of clinically evident trauma to the esophagus. There were no instances of atrioesophageal fistula or clinical reports of gastric dysmotility during follow-up. However, 2 patients experienced oropharyngeal bleeding in the immediate post-procedural period before endotracheal extubation. In 1 patient, direct laryngoscopy revealed mucosal injury in the vallecula, believed to be secondary to mechanical trauma during esophageal intubation. This patient was treated with acetaminophen for throat discomfort. In the second patient, direct laryngoscopy revealed a small (1.5 to 2 cm) linear laceration in the posterior oropharynx controlled with topical silver nitrate and Surgicel (Ethicon Inc., Somerville, New Jersey) to achieve hemostasis. In both patients, esophageal endoscopy revealed no evidence of esophageal injury, and a contrast esophagram showed no evidence of esophageal or pharyngeal perforation. For both patients, the 2 electrophysiologists intubating the esophagus with the balloon retractor reported difficultly with placement of the device. In addition, in both instances, these cases were among the first 5 times these physicians had attempted to intubate the esophagus with the device. All patients were followed up clinically and experienced no long-term sequelae.
In this consecutive series of 200 patients undergoing AF ablation, the DV8 esophageal balloon retractor was able to displace the esophagus laterally away from the point of posterior LA energy delivery during AF ablation procedures.
The most dreaded complication of AF ablation is an atrial-esophageal fistula, which has an incidence estimated to range from as low as 0.015% to as high as 0.3% (3,9,10). Although the initial instances of atrial-esophageal fistula were universally fatal, more recently, there has been a much greater awareness of this complication, with patients counseled to report suspicious symptoms, and with a proactive approach of early computed tomography scanning to look for evidence of mediastinal air (2,11). Despite this heightened awareness, proactive screening, and early surgical correction, atrial-esophageal fistula remains associated with a mortality rate of ∼50% in this “modern era” (11). Indeed, atrial-esophageal fistula accounted for 16% of all deaths associated with AF ablation (9). The mechanism of esophageal injury from AF ablation is believed to be a direct thermal injury to the esophagus, leading to necrosis and surrounding inflammation. Ischemic necrosis from injury to esophageal arteries, injury to the mucosa and intima of the esophagus, and damage to the vagal nerve and esophageal plexus may also play important roles in the fistula development. Although most of the reported cases of atrial-esophageal fistula have been associated with radiofrequency ablation, it has also been observed with cryoballoon-based PVI, particularly given the enhanced cooling power of the second-generation cryoballoon (12,13).
A human cadaveric study of the esophagus revealed loose fibrofatty connective tissue between the parietal pericardium and the esophagus, thereby allowing for esophageal mobility (14). In this study, the esophagus could be laterally displaced a mean of 4.56 cm. In patients undergoing AF ablation under conscious sedation, spontaneous mobility of the esophagus by ≥2 cm was observed (15). The feasibility of mechanically displacing the esophagus laterally away from the site of ablation was shown by using transesophageal echocardiography and endoscopy probes (16,17). Indeed, in the largest published series of MED, 704 consecutive patients safely underwent MED using a transesophageal echocardiography probe during AF ablation (18). Although use of transesophageal echocardiography probes for MED provides the proof-of-concept, such use is not practical for most electrophysiology laboratories. Accordingly, we previously evaluated the safety and feasibility of MED using a metal stylet, advanced into the esophagus through a plastic tube to minimize esophageal trauma (7,8). However, stiffer metal stylets had a tendency to cause oropharyngeal trauma, while more malleable stylets failed to achieve adequate lateral displacement.
As an alternative mechanical esophageal displacement strategy, the DV8 balloon retractor is intended to achieve the following: 1) minimize oropharyngeal trauma; 2) spread the lateral displacing force over a wide surface area so as to minimize pressure trauma; and 3) allow the operator to titrate the magnitude of lateral excursion by inflating or deflating the balloon. In the present series of 200 consecutive patients undergoing AF ablation while using this balloon retractor device, no instances of clinical esophageal injury or atrial-esophageal fistula occurred. It is also important to recognize that all of the procedures were performed with uninterrupted anticoagulation. However, in 2 patients, pharyngeal trauma occurred during attempts at esophageal intubation of the device; these instances were related to the lack of physician experience with esophageal intubation in the anesthetized patient (7). When the device was either placed by, or under the guidance of, physicians experienced with esophageal intubation (e.g., anesthesiologists), there were no instances of pharyngeal trauma.
Our group and others have shown that a greater distance between the esophagus and the site of ablation reduces esophageal heating; MEDEffective >20 mm largely eliminates LET >38°C (6,7). Using the DV8 balloon, MEDEffective ≤5 mm was invariably associated with significant LET increases. At MEDEffective between 5 and 20 mm, there was an intermediate frequency of LET increases: during 28% of these lesion sets, at least 1 LET increase was observed. It is also important to recognize that when using this balloon retractor, the amount of energy delivered on the posterior wall was not reduced. This energy titration strategy is unlike that used by most when only LET monitoring is employed.
In a study examining the relation of the esophagus-to-ablated point distance and any corresponding LET increase, the odds ratio for LET increase ≥38°C was 2.28 (p = 0.004) comparing <4.0 mm versus >5.0 mm distance; the longest distance for a ≥38°C increase was 18.5 mm (6). Together, these data indicate that a distance of <5 mm from the esophageal edge universally results in esophageal heating. At the optimal separation of MEDEffective >20 mm, LET increases rarely occurred; this target was achieved in 35.5% of targeted vein pairs with the balloon retractor in our study. By contrast, we previously reported that esophageal deviation outcome was only achieved in ∼22% of vein pairs using the malleable metal stylet technique (7). A comparison between the 2 methods is shown in Figure 5. As would be expected, this also translated to fewer occurrences of temperature increase with the DV8 balloon retractor.
It should also be recognized that by defining MEDEffective as the closest distance between the esophageal trailing edge to the closest ablation point, a relatively high bar was used. That is, the effectiveness of MED is underestimated because only the temperature changes of the closet ablation point is evaluated; if there were no LET increases in the remaining posterior wall lesions, this would not be appreciated. Accordingly, another measure of the effectiveness of MED would be the number of LET increases observed per patient. In our study, there were more temperature increases observed with MEDEffective <5 mm compared with the other groups. Finally, it should be recognized that an LET increase with the ablation catheter away from the esophageal edge may not have as negative a consequence as the same LET increase with the ablation catheter directly atop the esophagus.
In theory, the displacement of the esophagus away from the site of ablation should decrease the incidence of atrial-esophageal fistula. However, given the low incidence of atrial-esophageal fistula, it is highly unlikely that this theory could ever be proven in any practical clinical trial. For example, even if one expected an atrial-esophageal fistula rate as high as 0.3% in the control group, and assumed that MED could decrease the fistula rate by 99% to 0.003%, a total of 5,378 patients would still need to be randomized to treatment to have an 80% chance of detecting a statistically significant decrease, as significant at the 5% level. Conversely, because of the potential concern of esophageal damage, in virtually every patient, one’s ablation strategy is modified to truncating ablation lesion duration so as to avoid excessive esophageal heating. In addition to the disruption in procedural workflow, some data also suggest that interruptions of energy delivery with esophageal heating may even affect the durability of PVI. In a retrospective series of patients presenting for redo ablation because of arrhythmia recurrence, there was a relationship between esophageal heating during the initial procedure and subsequent pulmonary vein reconnections at the posterior LA ablation lesion set (19).
One important clinical question is whether LET monitoring should be used in conjunction with the DV8 retractor. In the present study, LET was used in most of the cases as a means to characterize the temperature on the esophageal trailing edge. However, in routine clinical practice, we do not always use an LET probe. One approach supported by the data would be to make this decision based on the distance from the point of ablation to the esophageal edge: 1) attempt to avoid ablation at distances <5 mm from the esophageal edge; 2) if ablation is absolutely necessary at distances <5 mm, either place an LET probe to help titrate energy or use alternative energy delivery strategies to minimize lesion depth (e.g., high power with short duration, low-flow saline irrigation); 3) at intermediate distances (5 to 20 mm), use these alternative energy delivery strategies to minimize lesion depth; and 4) at distances >20 mm, ablate as normal.
Although the technique appeared safe in this relatively large series of patients, introduction of the device did result in clinically significant trauma in 2 patients; thus, esophageal intubation of the balloon retractor (just as with other devices such as nasogastric tubes, transesophageal echocardiography, or esophagogastroduodenoscopy probes) should be performed carefully, particularly in anesthetized patients who are unable to express discomfort. It is unknown how safe the retractor would be in high-risk populations, such as those with pre-existing active esophageal ulcers, significant mucosal friability, or esophageal varices.
In the present study, MED effectiveness was defined by the magnitude of lateral displacement from the ablation lesion set. This definition was based on our experience, as well as others, that greater lateral displacement results in less esophageal heating. However, other esophageal imaging technologies, such as esophagogastroduodenoscopy, barium swallow, or magnetic resonance imaging, were not used.
Even with this balloon retractor, the MEDEffective was <5 mm in 3% of patients, a distance that invariably resulted in esophageal heating. In addition, even with a reasonable amount of lateral excursion (MEDEffective between 5 and 20 mm), there was a significant minority of patients (28%) in whom at least 1 LET increase was observed. Of course, this finding must be interpreted in the context that in our study, the amount of energy delivered on the posterior wall was not reduced; that is, if in addition to MED with the DV8 retractor, the energy titration strategy on the posterior wall was also reduced, the incidence of LET increase may be substantially reduced at MEDEffective between 5 and 20 mm.
Finally, the fully inflated 17-mm diameter of the balloon retractor causes an anterior distortion of the contralateral posterior left atrium and pulmonary veins. Accordingly, it has been our usual practice (even when using the malleable stylet technique) to re-create a focused left atrium/pulmonary vein anatomy when the deviation balloon is rotated to the contralateral direction. Perhaps these challenges may be further mitigated in future device iterations.
In this study, we reported the safety and effectiveness of a novel method of esophageal deviation using a balloon retractor system. The balloon retractor represents an attractive method to minimize the risk of thermal injury to the esophagus, improve procedural workflow, and, theoretically, allow the delivery of more effective lesions by mitigating the need to interrupt ablation for esophageal temperature elevation.
COMPETENCY IN MEDICAL KNOWEDGE: Esophageal injury during AF ablation is largely thermal and related to proximity of the esophagus to the point of ablation along the posterior left atrium. Current strategies to avoid injury include monitoring LET and stopping ablation when heating is detected. Mechanical deviation is an attractive alternative strategy to minimize esophageal damage and improve workflow by reducing temperature increases that lead to interruption of ablation. We previously described the efficacy of an off-the-shelf stylet technique to deviate the esophagus. In the present series of 200 patients, we reported the safety and efficacy of a dedicated esophageal retractor. Greater lateral esophageal deviation was achieved by using this retractor compared with the stylet, and this technique was associated with fewer temperature increases.
TRANSLATIONAL OUTLOOK: Esophageal temperature increases were more common with closer proximity of the ablation site to the esophageal lumen. When the esophagus was >20 mm away laterally, there was a significantly lower rate of LET increase. The dedicated esophageal retractor is an effective tool to mechanically displace the esophagus, but a learning curve remains as trauma with insertion was initially observed. The device should be placed by physicians accustomed to intubating the esophagus.
Drs. Dukkipati, Mittnacht, and Reddy hold equity interest in Manual Surgical Sciences Inc. Dr. Dukkipati has also received research grants from Biosense Webster. Dr. Reddy holds equity interest (stock options) in Circa Scientific, Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial fibrillation
- left atrial
- luminal esophageal temperature
- mechanical esophageal deviation
- the distance from the trailing esophageal edge to the closest point of the ablation line
- pulmonary vein isolation
- Received December 5, 2017.
- Revision received March 5, 2018.
- Accepted April 5, 2018.
- 2018 American College of Cardiology Foundation
- Singh S.M.,
- d’Avila A.,
- Doshi S.K.,
- et al.
- Halbfass P.,
- Pavlov B.,
- Müller P.,
- et al.
- Lakkireddy D.,
- Reddy Y.M.,
- Atkins D.,
- et al.
- Palaniswamy C.,
- Koruth J.S.,
- Mittnacht A.J.,
- et al.
- De Vasconcelos J.T.,
- Filho S.S.,
- Atie J.,
- et al.
- Metzner A.,
- Burchard A.,
- Wohlmuth P.,
- et al.
- Marar D.,
- Muthusamy V.,
- Krishnan S.C.
- Good E.,
- Oral H.,
- Lemola K.,
- et al.
- Herweg B.,
- Johnson N.,
- Postler G.,
- Curtis A.B.,
- Barold S.S.,
- Ilercil A.
- Mateos J.C.,
- Mateos E.P.,
- Peña T.G.,
- et al.
- Tran V.N.,
- Kusa S.,
- Smietana J.,
- et al.