Accepted Manuscript
Feasibility, Safety and Efficacy of a Novel Pre-Shaped Nitinol Esophageal Deviatorto Successfully Deflect the Esophagus and Ablate Left Atrium without EsophagealTemperature Rise during Atrial Fibrillation Ablation – The DEFLECT GUT study
Valay Parikh, MD, Vijay Swarup, MD FHRS, Jacob Hantla, CRNA FHRS, VenkatVuddanda, MD, Tawseef Dar, MD, Bharath Yarlagadda, MD, Luigi Di Biase,MD FHRS, Amin Al-Ahmad, MD FHRS, Andrea Natale, MD FHRS, DhanunjayaLakkireddy, MD FHRS
PII: S1547-5271(18)30362-X
DOI: 10.1016/j.hrthm.2018.04.017
Reference: HRTHM 7561
To appear in: Heart Rhythm
Received Date: 23 February 2018
Please cite this article as: Parikh V, Swarup V, Hantla J, Vuddanda V, Dar T, Yarlagadda B, Di BiaseL, Al-Ahmad A, Natale A, Lakkireddy D, Feasibility, Safety and Efficacy of a Novel Pre-Shaped NitinolEsophageal Deviator to Successfully Deflect the Esophagus and Ablate Left Atrium without EsophagealTemperature Rise during Atrial Fibrillation Ablation – The DEFLECT GUT study, Heart Rhythm (2018),doi: 10.1016/j.hrthm.2018.04.017.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
1
Feasibility, Safety and Efficacy of a Novel Pre-Shaped Nitinol Esophageal Deviator to 1
Successfully Deflect the Esophagus and Ablate Left Atrium without Esophageal 2
Temperature Rise during Atrial Fibrillation Ablatio n – The DEFLECT GUT study 3
4
Short Title: Esophageal deviation during atrial fibrillation ablation 5
6
Valay Parikh MD1, Vijay Swarup MD FHRS2, Jacob Hantla CRNA FHRS2, Venkat Vuddanda 7
MD1, Tawseef Dar MD1, Bharath Yarlagadda MD1, Luigi Di Biase MD FHRS3, Amin Al-8
Ahmad MD FHRS4, Andrea Natale MD FHRS4, Dhanunjaya Lakkireddy MD FHRS1 9
10
1 University of Kansas Health System, Kansas City, Kansas. 11
2Arizona Heart Rhythm Center, Phoenix, AZ. 12
3Albert Einstein College of Medicine at Montefiore Hospital, New York. 13
4Texas Cardiac Arrhythmia Institute, St. David's Medical Center, Austin, Texas. 14
15
Section: Original Research 16
Word Count: 4998 words 17
Corresponding Author: 18
Dhanunjaya Lakkireddy, MD FHRS 19
Professor of Medicine 20
University of Kansas Medical Center 21
3901 Rainbow Boulevard; MS 3006 22
Kansas City, KS 66160, USA 23
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
2
Phone: 913-588-9406; Fax: 913-588-9770 24
Email: [email protected] 25
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
3
Disclosures 26
27
V. Parikh, V. Vuddanda, T. Dar, B. Yarlagadda: None. 28
V. Swarup: Consultant: Abbott, Biosense Webster. 29
Speaker: Boston Scientific, Janssen, Pfizer. 30
J. Hantla: Consultant: EP Reward. 31
L. Di Biase: Consultant: Stereotaxis, Biosense Webster, Abbott. 32
Speaker: Biotronik, Medtronic, Boston Scientific, Janssen, Pfizer, 33
Epi EP. 34
A. Al-Ahmad: Speaker: Medtronic, Boston Scientific, Biosense Webster, Abbott. 35
A. Natale: Consultant: Stereotaxis, Biosense Webster, Abbott. 36
Speaker: Medtronic, Boston Scientific, Janssen, Pfizer. 37
D. Lakkireddy: Consultant: Abbott, North-East Scientific, Biosense Webster. 38
Speaker: Boston Scientific, Janssen, Pfizer. 39
40
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
4
Abstract 41
42
Background: Esophageal thermal injury is a feared complication of radiofrequency ablation 43
(RFA) for atrial fibrillation (AF). Rise in luminal esophageal temperature (LET) limits the ability 44
to deliver RF energy on posterior wall of LA. 45
46
Objective: The aim of this registry was to evaluate feasibility, safety and efficacy of a 47
mechanical esophageal deviation (ED) tool during AF ablation. 48
49
Methods: We evaluated 687 patients who underwent RFA for AF. In 209 patients, EsoSure® 50
was used to deflect esophagus away from ablation site. Propensity-score matching was 51
performed to obtain 180 patients each in ED and non-ED arms. ED was used for LET rise seen 52
in 61.7% (111/180) patients and was used if esophagus was in the line of ablation on fluoroscopy 53
in 38.3% (69/180) patients. 54
55
Results: The mean deviation of trailing edge of esophagus with EsoSure® was 2.45 ± 0.9 cm 56
(range: 1-4.5 cm). LET rise >1°C was significantly lower in ED than non-ED group (3% vs 57
79.4%, p<0.001). Mean LET rise (ED 0.34 ± 0.59 vs non-ED 1.66 ± 0.54, p<0.001). Intra-58
procedural success of PVAI, was slightly improved in ED arm than in non-ED arm without 59
statistical significance. AF recurrence was lower in ED arm at 3-month, 6-month and 1-year 60
follow-up than non-ED arm. No ED-related complications were noted. 61
62
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
5
Conclusions: Mechanical displacement of esophagus with EsoSure® appears to be feasible, safe 63
and efficacious in enabling adequate RF energy delivery to posterior wall of LA without 64
significant LET rise and obvious clinical signs of esophageal injury. 65
66
67
Key Words: Atrial Fibrillation; Atrio-Esophageal Fistula; Catheter Ablation; Esophageal 68
Deviation; Esophageal Injury. 69
70
Funding: This research did not receive any specific grant from funding agencies in the public, 71
commercial, or not-for-profit sectors. 72
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
6
Introduction 73
Radiofrequency catheter ablation (RFA) is increasingly being performed for the 74
management of atrial fibrillation (AF). One of the feared complications of this procedure is 75
thermal esophageal injury and atrio-esophageal fistula (AEF) [1,2]. Esophageal injury of 76
varying severity has been reported, anywhere between 2-48 % [3-7]. Although the overall 77
incidence of AEF is low (0.03-0.1%), it is associated with very high morbidity and mortality 78
[2,8,9]. A rise in luminal esophageal temperature during the procedure has been identified as a 79
surrogate marker for thermal injury [10-13]. Despite the limitations of this method [14-16], 80
luminal esophageal temperature monitoring during AF-RFA is routinely performed in most 81
centers across world. A rapid and significant rise in luminal esophageal temperature may indicate 82
significant heating of esophageal tissue from adjacent RFA on left atrial(LA) posterior wall. 83
Typically, RFA is more tempered on posterior wall with lower power, shorter duration and lower 84
contact force to minimize collateral damage to esophagus. If esophagus lies exactly posterior to 85
target ablation site, frequent temperature rises limit ability to deliver adequate lesions, increase 86
procedure time and potentially result in higher PV reconnections. 87
88
Mechanical deflection of esophagus away from ablation site may enable more effective 89
RF lesion delivery while avoiding esophageal thermal injury. Previous attempts of esophageal 90
displacement with transesophageal echocardiography(TEE) probe, endoscope and an 91
endotracheal stylet placed in thoracic chest tube have been reported with modest but promising 92
success [17-19]. However, their routine use has been limited due to their complexity, resource 93
constraints and concerns about esophageal erosion with bulky heat-retaining probes. Recently, a 94
novel pre-shaped nitinol esophageal retractor (EsoSure®) has been available for use in US for 95
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
7
esophageal deviation. Limited experience in animal models has suggested safety of esophageal 96
deviation with this device [20]. However, there are no clinical data available for EsoSure® use 97
for deflecting esophagus during AF-RFA to minimize esophageal injury. We performed a multi-98
center study using a prospective registry involving 4 participating centers to evaluate safety and 99
efficacy of EsoSure® use during AF-RFA. 100
101
102
Methods 103
Study Population 104
All patients who had EsoSure® device for esophageal deviation (EsoSure® group) use 105
during AF-RFA between January 2016 and August 2017 were included in study. EsoSure® was 106
used by operator when a) there was a luminal esophageal temperature rise (>1ºC from baseline or 107
>38.5ºC), or b) esophagus was lying exactly posterior to target site while ablating on the 108
posterior wall of LA. Patients with pre-existing esophageal stricture, esophageal varices, prior 109
esophageal dilatation and other intervention were excluded. Patients undergoing AF-RFA 110
without EsoSure® use between August 2015 and August 2017, served as a control group (non-111
EsoSure® group). 112
113
Ablation Procedure 114
Patients underwent pulmonary vein antral isolation (PVAI) with a double transseptal 115
approach as described in detail elsewhere [21]. All ablations were performed under general 116
anesthesia on uninterrupted anticoagulation. LA was mapped using mapping catheter (Lasso®, 117
Biosense Webster, California or Advisor®, Abbott, Minnesota). Electrical isolation was 118
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
8
accomplished by ablating PV antra with 3.5 mm/4 mm open irrigated tip, contact force sensing 119
catheter (ThermoCool SmartTouch® SF, Biosense Webster/TactiCath™, Abbott). All site 120
monitored esophageal temperature by 9F single thermistor esophageal probe (ER 400-9; Smiths 121
Medical, Ohio). It was adjusted dynamically along trailing edge of esophagus to approximate 122
lesion delivery site as close as possible throughout the procedure. 123
124
The endpoint for PVAI was achievement of both entrance block and exit block. 20 125
mcg/min of isoproterenol was given for 15 minutes to identify non-PV triggers. For paroxysmal 126
AF, only PVAI was performed. For persistent AF, additional ablations were done at the 127
operator’s discretion. 128
129
Ablation Parameters 130
A maximum of 40W on anterior wall and 25 W for 20 seconds (contact force <15 gm) on 131
posterior wall was used. Ablation was stopped if the luminal esophageal temperature increased 132
rapidly upto 1ºC from baseline or reached a maximum of 38.5ºC. An esophagogram to define 133
borders of esophagus was performed with light barium prior to EsoSure® insertion. In control 134
group, ablation at that location was halted until luminal esophageal temperature returned to 135
baseline. Strategy to deliver further lesions at that or adjacent sites were determined by operator 136
based on factors such as proximity of esophagus, rapidity of luminal esophageal temperature rise 137
and feasibility of lesion delivery. 138
139
EsoSure® Device 140
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
9
EsoSure® (Northeast Scientific Inc., Waterbury, CT) is a 0.052 inches diameter 141
temperature programmed Nitinol stylet which is soft at room temperature and firm at body 142
temperature (Figure 1a-c). It is inserted into lumen of standard 18-Fr Oro-Gastric tube (OGT) 143
and assumes its temperature programmed “S” shaped curve. The esophagus has minimal 144
anatomical restraints in the posterior mediastinum which enables the device to create an 145
esophageal deviation behind LA. The stylet-OGT can be moved cranially-caudally to deviate 146
different sections of esophagus. It can also be rotated to achieve posterior deviation of 147
esophagus. It is a US Patented and FDA registered class I device under classification of a 148
retractor. The ability of EsoSure® to retract esophagus was well appreciated in human cadaver 149
and live human use (Supplemental video). 150
151
152
Esophageal Deviation and Luminal Esophageal Temperature monitoring 153
The EsoSure® was used in patients with a) temperature rise on the luminal esophageal 154
temperature probe (>1ºC from baseline or absolute temperature >38.5ºC) while ablating on the 155
posterior wall of LA or b) when esophagus was just posterior to the site of ablation on 156
fluoroscopy. A step-by-step guide as outlined below was used at all participating sites while 157
inserting EsoSure® (Supplemental Table-1). In brief, EsoSure® device was introduced into the 158
lumen of OGT following light barium esophagogram. The OGT was primed with lubricant (1-2 159
cc of olive oil) injected through the lumen prior to insertion in order to facilitate smooth 160
advancement of EsoSure® into OGT. Using fluoroscopy, EsoSure® stylet was adjusted to 161
deviate esophagus away from the planned RF lesion delivery site (Figure-2). Esophageal 162
temperature probe was adjusted to trailing edge of esophagus (as close as possible to RF lesion 163
site). Subsequent manipulations of EsoSure® were performed as needed to keep esophagus away 164
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
10
from ablation site. The goal was to achieve deflection of trailing edge of esophagus by at least 165
2cm from ablation site. In cases with <2cm deviation, we manipulated to achieve maximal 166
possible deviation away from intended ablation. All manipulations were done under fluoroscopy 167
and final position was confirmed with fluoroscopy with residual contrast in esophagus. In cases 168
where EsoSure® stylet was inserted due to rising temperature, repeat ablation at same spot was 169
performed after esophageal deviation. At the end of procedure, EsoSure® was withdrawn and 170
suction was applied through OGT to remove residual barium from stomach and esophagus. 171
Allied professionals scrubbing in case performed esophageal deviation after prior training with 172
assistance from operator as needed. 173
174
Data collection 175
Demographics, clinical, procedural and complications data were collected prospectively. 176
Efficacy endpoints: 1) successful esophageal deviation ≥2 cm on fluoroscopy 2) Acute 177
PVAI 3) Procedure time saved by moving esophagus 4) Number of sudden rapid luminal 178
esophageal temperature rises per case. 4) 3-month, 6-month and 12-month atrial arrhythmia 179
recurrences. 180
Safety endpoints: 1) Symptoms of potential complications related to esophageal injury 181
(dysphagia, odynophagia, early satiety, chest pain, melena, fever, chills, and stroke like 182
symptoms) -collected prospectively with 1-week phone call; and 1-month and 3-month clinic 183
visits. 2) Non-esophageal complications. 184
185
Statistical analysis 186
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
11
The protocol was approved by Institutional Review Board at each site. Categorical and 187
continuous variables were compared using chi-square/Fisher’s exact test and students t-test 188
respectively. 189
190
Potential confounding factors adjusted for in the multivariable analyses included age, 191
gender, type of AF, CHADS2VA2Sc score and gastroesophageal reflux disease (GERD). We 192
estimated a propensity score by fitting a logistic-regression model that adjusted for these items 193
above. One-to-one pair matching between groups was performed by nearest-neighbor matching 194
without replacement, with use of caliper width equal to 0.2 of standard deviation of logit of 195
propensity score. Covariate balances before and after matching were checked by comparison of 196
standardized mean differences. A standardized mean difference of less than 10.0% was 197
considered to indicate a negligible imbalance between groups. 198
199
A p<0.05 (2-sided) was considered statistically significant. All analyses were 200
performed with IBM SPSS 24.0 (SPSS Inc., Chicago, IL). 201
202
Results 203
Our study consisted of 687 patients who underwent AF-RFA. Out of these, 209 patients 204
underwent EsoSure® use. Rest of 478 patients didn’t undergo EsoSure® use. Mean age in 205
EsoSure® group was 63.8 years, while it was 63.5 years in non-EsoSure® group. Males were 206
higher in non-EsoSure® group, but statistically not significant (69.9% vs 63.6%, p=0.13). 207
Baseline characteristics are shown in Table 1. Propensity-score matching yielded 180 patients 208
who underwent EsoSure® use to 180 patients who didn’t undergo EsoSure® use. Propensity-209
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
12
score matching improved the covariate balance considerably (Table-1). For the entire analysis 210
the matched groups were used. 211
212
Esophageal deviation and luminal esophageal temperature measurements: 213
Esophageal deviation was used for luminal esophageal temperature rise in 111(61.7 %) 214
patients early on and for close proximity of esophagus to ablation site in 69(38.3 %) patients 215
later in registry. Esophageal deviation was used primarily for left sided PVs in 133(74%) 216
patients, right sided PVs in 38(21%) patients and posterior wall in 9(5%) patients. There was no 217
statistical difference in location of esophagus between groups. Mean time duration for 218
esophagogram assessment, and EsoSure® placement and manipulation was 8±6 minutes. It was 219
reduced to 5±2 minutes in last 50 cases. The additional fluoroscopy time with esophageal 220
deviation process was 76±20 seconds. The mean displacement of trailing edge of esophagus was 221
2.45±0.9 cm on fluoroscopy (range of 1.0-4.5cm). In 36(17%) patients, the maximum esophageal 222
deviation was <2 cm. None of the clinical characteristics were predictive of deviation <2 cm. 223
224
Temperature rise >10C was significantly higher in Non-EsoSure® than EsoSure® group 225
(79.4% vs 3%, p<0.001). 3% EsoSure® group had esophageal deviation of 1cm from the trailing 226
edge with associated esophageal temperature rise. Non-EsoSure® group had a higher ∆ luminal 227
esophageal temperature rise (1.66 ± 0.54 vs 0.34 ± 0.59, p<0.001) than EsoSure® group (Table-228
2). Total procedure time was significantly lower in EsoSure® than non-EsoSure® group (182 ± 229
36 minutes vs 206 ± 32 minutes, p<0.001). There was no statistical difference in mean 230
fluoroscopy and RF ablation times. 231
232
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
13
Among the 111(61.7%) patients in whom EsoSure® was used due to increase in luminal 233
esophageal temperature, the mean peak difference in luminal esophageal temperature was 234
1.3±0.2ºC prior to EsoSure® use and 0.2±0.1ºC after EsoSure® use (p<0.001). In the remaining 235
69 (38.3%) patients esophagus was moved without waiting for temperature elevation if it was 236
within the line of ablation. In 97%(174/180) patients the peak difference in luminal esophageal 237
temperature was <0.5 ºC after using EsoSure®. 238
239
Ablation Efficacy endpoints: 240
There was no difference in intra-procedural PVAI rates between the groups 241
[EsoSure®:179/180 (99.4%) vs. non-EsoSure®:174/180 (96.7%), p=0.12]. However, AF 242
recurrence was significantly higher in non-EsoSure® at 3-month, 6-month and 12-month (Table-243
2). 244
245
Safety endpoints: 246
No patients were lost to follow-up. There was no difference in overall complications 247
between cohorts (3.8% vs 3.8%, p=1.0) (Table-3). Major and minor complications of the 248
procedure in EsoSure® group occurred in 7 (3.8%) of the patients. Two patients developed 249
pericardial effusion, of which one didn’t require intervention and the other needed cardiac 250
surgery. No patients had TIA, stroke or death. Eight patients (4.4%) complained of “sore-throat” 251
immediately post procedure which subsided within 3-days. At subsequent follow-ups, none of 252
the patients had symptoms suggestive of esophageal injury or had incidence of gastro-esophageal 253
(atrio-esophageal fistula, symptomatic esophageal ulcerations, symptomatic gastric dysmotility) 254
complications. 255
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
14
Discussion 256
Major findings: This is the first observational study to date evaluating the feasibility, safety and 257
efficacy of EsoSure®, an esophageal deviation tool, during AF-RFA. In our multi-center study, 258
we noted that mechanical displacement of esophagus with EsoSure® is not only feasible but 259
effective in displacing esophagus and enabling successful delivery of RF energy to achieve 260
ablation end-points without significant luminal esophageal temperature rise. Clinically, there 261
were no safety issues noted. 262
263
Esophageal injury leading to AEF is one of the rare but dreaded complications of AF 264
ablation procedures [1,2,8]. Intra-procedural luminal esophageal temperature measurement and 265
subsequent modification of ablation lesion locations and energy delivery settings became 266
standard practice. Often RF ablation is halted after luminal esophageal temperature rise for 267
esophagus to cool down and subsequently use lower energy and/or briefer applications at site of 268
interest prolonging procedure times. Luminal esophageal temperature may rise again during 269
repeat energy application needing re-cooling of esophagus. Repeated on-off, low efficiency 270
ablation could lead to sub-optimal lesion formation and in turn result in non-durable lesions and 271
increase in arrhythmia recurrence. Moreover, acceptable upper limit of luminal esophageal 272
temperature rise is not standardized and varies significantly across operators and institutions. 273
Very low incidence of AEF makes it very difficult to come up with an optimal cut-off luminal 274
esophageal temperature. The interference of luminal esophageal temperature rise during RF 275
ablation during index ablation procedure may persist during a ‘re-do’ ablation procedure also. 276
[22] Other approaches such as esophageal cooling, although conceptually viable, haven’t been 277
tested rigorously in clinical practice[23,24]. Therefore, mechanical displacement of esophagus 278
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
15
from the desired ablation site is a very useful alternative and in vast majority of cases can 279
become an important part of workflow, if performed safely and effectively. 280
281
Current data on esophageal deviation: 282
Owing to minimal attachments to mediastinal structures, esophagus can be moved away 283
from posterior wall of LA (Figure-3) with a deviating tool like EsoSure® but can quickly move 284
back to its approximate original location once the deviator is removed. Esophageal displacement 285
using a TEE probe by Mateos et al was found to be effective with average esophageal 286
displacement of 5.9±0.8 cm, sufficient to provide effective lesions without esophagus 287
overlapping[18]. However, it needs an additional procedure, and an experienced operator to 288
manipulate the probe. In addition, TEE probe is bulkier, significantly stiffer and could potentially 289
act as a heat retainer contributing to both mechanical and thermal esophageal injury. Similarly, 290
endoscope and an endotracheal stylet through a thoracic tube have been used to a limited extent 291
and faced logistic challenges, insufficient deviation and safety concerns[17,19,25]. 292
293
Our study shows that mechanical displacement of esophagus using EsoSure® is 294
technically feasible and clinically safe. The most common complaint was transient symptom of 295
“sore throat” seen in 8(4.4%) patients, which can be related to endotracheal intubation. It 296
resolved within 3-days in all cases without any clinical sequelae. With esophageal temperature 297
probe lined up along trailing edge of esophagus, we did not notice any significant temperature 298
elevation (>0.5ºC) in almost all patients while delivering effective lesions. We did not perform a 299
routine post-procedural endoscopy to evaluate for possible esophageal lesions with device use. 300
However, limited data from animal studies with endoscopy suggested safety of device [20]. 301
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
16
These evidences, in addition to our clinical experience, are reassuring that we don’t routinely 302
perform endoscopy. However, further prospective studies with post-procedural endoscopy will 303
be needed for a definitive objective evidence. 304
305
Another notable finding of this study is that we were able to mobilize esophagus from the 306
original location in most of the cases. Average displacement was 2.45±0.9 cm, that is consistent 307
with previous clinical studies and possibly similar to natural physiological migration [17,19,26]. 308
This displacement distance is sufficient enough to deflect esophagus from the area of ablation, 309
which decreases risk of luminal esophageal temperature and possibly esophageal injury as its 310
consequence. Displacement in our study is considerably less than 5.9±0.8 cm shown with TEE 311
probe[18]. This in part adds to safety profile of device. Extreme esophageal displacement is 312
unlikely going to be seen by this device due to its shallow and smooth deviation curves. Unlike 313
TEE probe, EsoSure® is a Nitinol stylet that sits in the OGT creating gentle displacement in an 314
arc mimicking physiologic migration without dramatic stress on esophagus. The spherical tip of 315
the stylet further helps improve the safety profile of the device by preventing inadvertent 316
puncture of OGT. It is important to note that we did not use the device in patients with prior 317
esophageal interventions including fundoplication surgeries and esophageal dilations. The safety 318
of this device in this subgroup of patients is currently unknown. 319
320
In our study, intra-procedural efficacy measured as acute PVAI was similar in both arms. 321
This was associated with significantly longer procedure time for non-EsoSure® arm without 322
statistical differences in fluoroscopy or RF time. Brief lesions resulting in rapid rise in luminal 323
esophageal temperature and subsequent wait time for luminal esophageal temperature to return to 324
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
17
nadir may account for this finding. AF recurrences at 3-month, 6-month and 12-month were 325
significantly higher in non-EsoSure® arm. These findings of similarities in acute PVAI with 326
higher recurrence rate at subsequent follow-ups in non-EsoSure® arm can be hypothesized due 327
to repeated delivery of non-permanent lesions resulting in local tissue edema and inflammation, 328
which may result in acute achievement of PVAI, but subsequent reconnection at later point of 329
time. At this moment, we don’t have definitive data to support whether esophageal deviation can 330
result in good lesion formation and aid durable PVI. However, subsequent studies comprising 331
patients undergoing a re-do ablation may help us understand this phenomenon. 332
333
In summary, esophageal deviation with EsoSure® device during AF ablation appears to 334
be safe and effective in enabling effective lesion delivery to the LA without significant rise in 335
luminal esophageal temperature. This adds to the existing armamentarium of tools to minimize 336
esophageal injury during AF ablation. A prospective trial to further validate our findings and also 337
to correlate with post-procedural endoscopic findings is the next step. 338
339
Limitations and Future directions 340
The current study has all the inherent limitations of an observational registry. We did not 341
perform endoscopic evaluations in these patients to assess if esophageal lesions were minimized 342
or avoided. Nevertheless, there was no clinical occurrence of any major esophageal injury related 343
complications in our study, which should be considered as a positive finding and should provide 344
impetus to further studies evaluating esophageal displacement strategies. Endoscopic evaluation 345
of esophagus post ablation with esophageal deviation may help confirm the improved safety 346
attributed to use of this device. A prospective randomized control trial with and without 347
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
18
esophageal deviation assessing the anatomic and functional integrity of the upper gastrointestinal 348
system and acute successful PVI and long-term arrhythmia freedom may help us understand its 349
true utility. Whether esophageal deviation helps in preventing esophageal fistula cannot be 350
assessed by this small study. However, posterior wall ablation can be done effectively improving 351
the acute PVI success without significant luminal esophageal temperature rise which is a marker 352
for esophageal injury. We should also consider adding these data points to the NCDR AF 353
ablation registry to be able to assess the impact of esophageal deviation on the outcomes of 354
esophageal injury going forwards. 355
356
Conclusion 357
Esophageal deviation using EsoSure® is feasible, safe and efficacious during RFA for AF 358
in patients in whom luminal esophageal temperature rise prevents delivering effective RF lesion 359
or in patients where esophagus is directly posterior to the intended ablation site. Future 360
randomized studies with a protocol-based endoscopic evaluation may give definitive evidence of 361
safety and efficacy. 362
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
19
References 363
364
1 Calkins H, Kuck KH, Cappato R et al.: 2012 HRS/EHRA/ECAS Expert Consensus 365
Statement on Catheter and Surgical Ablation of Atrial Fibrillation: recommendations for patient 366
selection, procedural techniques, patient management and follow-up, definitions, endpoints, and 367
research trial design. Europace 2012;14:528-606. 368
2 Pappone C, Oral H, Santinelli V et al.: Atrio-esophageal fistula as a complication of 369
percutaneous transcatheter ablation of atrial fibrillation. Circulation 2004;109:2724-2726. 370
3 Di Biase L, Dodig M, Saliba W, Siu A, Santisi J, Poe S, Sanaka M, Upchurch B, Vargo J, 371
Natale A: Capsule endoscopy in examination of esophagus for lesions after radiofrequency 372
catheter ablation: a potential tool to select patients with increased risk of complications. J 373
Cardiovasc Electrophysiol 2010;21:839-844. 374
4 Halm U, Gaspar T, Zachaus M, Sack S, Arya A, Piorkowski C, Knigge I, Hindricks G, 375
Husser D: Thermal esophageal lesions after radiofrequency catheter ablation of left atrial 376
arrhythmias. Am J Gastroenterol 2010;105:551-556. 377
5 Martinek M, Bencsik G, Aichinger J, Hassanein S, Schoefl R, Kuchinka P, Nesser HJ, 378
Purerfellner H: Esophageal damage during radiofrequency ablation of atrial fibrillation: impact 379
of energy settings, lesion sets, and esophageal visualization. J Cardiovasc Electrophysiol 380
2009;20:726-733. 381
6 Kiuchi K, Okajima K, Shimane A et al.: Incidence of esophageal injury after pulmonary 382
vein isolation in patients with a low body mass index and esophageal temperature monitoring at 383
a 39 degrees C setting. J Arrhythm 2015;31:12-17. 384
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
20
7 Schmidt M, Nolker G, Marschang H et al.: Incidence of oesophageal wall injury post-385
pulmonary vein antrum isolation for treatment of patients with atrial fibrillation. Europace 386
2008;10:205-209. 387
8 Mohanty S: Outcomes of atrio-esophageal fistula following catheter ablation of atrial 388
fibrillation treated with surgical repair versus esophageal stenting. J Cardiovasc Electrophysiol 389
2014;25:E6. 390
9 Sonmez B, Demirsoy E, Yagan N, Unal M, Arbatli H, Sener D, Baran T, Ilkova F: A 391
fatal complication due to radiofrequency ablation for atrial fibrillation: atrio-esophageal fistula. 392
Ann Thorac Surg 2003;76:281-283. 393
10 Ren JF, Lin D, Marchlinski FE, Callans DJ, Patel V: Esophageal imaging and strategies 394
for avoiding injury during left atrial ablation for atrial fibrillation. Heart Rhythm 2006;3:1156-395
1161. 396
11 Redfearn DP, Trim GM, Skanes AC, Petrellis B, Krahn AD, Yee R, Klein GJ: 397
Esophageal temperature monitoring during radiofrequency ablation of atrial fibrillation. J 398
Cardiovasc Electrophysiol 2005;16:589-593. 399
12 Singh SM, d'Avila A, Doshi SK, Brugge WR, Bedford RA, Mela T, Ruskin JN, Reddy 400
VY: Esophageal injury and temperature monitoring during atrial fibrillation ablation. Circ 401
Arrhythm Electrophysiol 2008;1:162-168. 402
13 Kiuchi K, Okajima K, Shimane A et al.: Impact of esophageal temperature monitoring 403
guided atrial fibrillation ablation on preventing asymptomatic excessive transmural injury. J 404
Arrhythm 2016;32:36-41. 405
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
21
14 Koranne K, Basu-Ray I, Parikh V, Pollet M, Wang S, Mathuria N, Lakkireddy D, Cheng 406
J: Esophageal Temperature Monitoring During Radiofrequency Ablation of Atrial Fibrillation: A 407
Meta-Analysis. 408
15 Muller P, Dietrich JW, Halbfass P et al.: Higher incidence of esophageal lesions after 409
ablation of atrial fibrillation related to the use of esophageal temperature probes. Heart Rhythm 410
2015;12:1464-1469. 411
16 Deneke T, Bunz K, Bastian A et al.: Utility of esophageal temperature monitoring during 412
pulmonary vein isolation for atrial fibrillation using duty-cycled phased radiofrequency ablation. 413
J Cardiovasc Electrophysiol 2011;22:255-261. 414
17 Chugh A, Rubenstein J, Good E et al.: Mechanical displacement of the esophagus in 415
patients undergoing left atrial ablation of atrial fibrillation. Heart Rhythm 2009;6:319-322. 416
18 Mateos JC, Mateos EI, Peña TG, Lobo TJ, Vargas RN, Pachón CT, Acosta JC: 417
Simplified method for esophagus protection during radiofrequency catheter ablation of atrial 418
fibrillation--prospective study of 704 cases. Rev Bras Cir Cardiovasc 2015;30:139-147. 419
19 Koruth JS, Reddy VY, Miller MA, Patel KK, Coffey JO, Fischer A, Gomes JA, 420
Dukkipati S, D'Avila A, Mittnacht A: Mechanical esophageal displacement during catheter 421
ablation for atrial fibrillation. J Cardiovasc Electrophysiol 2012;23:147-154. 422
20 Wang S SM, Lehmann I, Atsushi S, Rettmann M, Viker K, Parker K, Johnson S, Packer 423
D: The feasibility and safety of a novel displacement device in protecting the esophagus during 424
radiofrequency ablation of left atrial targets. Heart Rhythm 2017;14:S514–S563. 425
21 Di Biase L, Burkhardt JD, Mohanty P et al.: Left atrial appendage: an underrecognized 426
trigger site of atrial fibrillation. Circulation 2010;122:109-118. 427
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
22
22 Kennedy R, Good E, Oral H, Huether E, Bogun F, Pelosi F, Morady F, Chugh A: 428
Temporal stability of the location of the esophagus in patients undergoing a repeat left atrial 429
ablation procedure for atrial fibrillation or flutter. J Cardiovasc Electrophysiol 2008;19:351-355. 430
23 Arruda MS, Armaganijan L, Di Biase L, Rashidi R, Natale A: Feasibility and safety of 431
using an esophageal protective system to eliminate esophageal thermal injury: implications on 432
atrial-esophageal fistula following AF ablation. J Cardiovasc Electrophysiol 2009;20:1272-1278. 433
24 Sohara H, Satake S, Takeda H, Yamaguchi Y, Nagasu N: Prevalence of esophageal 434
ulceration after atrial fibrillation ablation with the hot balloon ablation catheter: what is the value 435
of esophageal cooling? J Cardiovasc Electrophysiol 2014;25:686-692. 436
25 Palaniswamy C, Koruth J, Mittnacht A et al.: The Extent of Mechanical Esophageal 437
Deviation to Avoid Esophageal Heating During Catheter Ablation of Atrial Fibrillation, JACC: 438
Clinical Electrophysiology 3, pp 1146-1154. 439
26 Good E, Oral H, Lemola K et al.: Movement of the esophagus during left atrial catheter 440
ablation for atrial fibrillation. J Am Coll Cardiol 2005;46:2107-2110. 441
442
443
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
23
TABLES and FIGURES: 444
445
446
447
448
449
Table 1: Baseline Characteristics of the unmatched and matched cohorts. 450
451
452
Entire Cohort Propensity Score Matched Cohort
Characteristic EsoSure®
(n=209)
Non-EsoSure®
(n=478)
p value EsoSure®
(n=180)
Non-EsoSure®
(n=180)
p value
Age (mean ± SD, years) 63.8 ± 9.9 63.5 ± 11.3 0.79 62.91 ± 9.73 63.6 ± 11.52 0.51
Males (%) 133(63.6 %) 334(69.9 %) 0.13 124(68.9 %) 115(63.9 %) 0.37
Body Mass Index (mean ± SD, kg/m^2) 31 ± 7 32 ± 6 0.06 31 ± 4 32 ± 5 0.04
Paroxysmal Atrial Fibrillation (%) 135(64.6 %) 253(52.9 %) 0.01 108(60 %) 108(60 %) 1.0
CHA2DS2-VASc (mean ± SD) 2.3 ± 1.4 2.4 ± 1.5 0.22 2.5 ± 1.5 2.5 ± 1.3 1.0
Gastro Esophageal Reflux Disease (%) 55(26.3%) 231(48.3%) <0.001 55(30.6 %) 55(30.6 %) 1.0
Hiatal Hernia 22(10.5%) 62(13 %) 0.45 17(9.4%) 23(12.7%) 0.4
Prior Atrial Fibrillation ablation 74(35.4%) 187(39.1%) 0.39 41(22.7 %) 45(25 %) 0.71
Left Ventricular Ejection Fraction (%) 54 ± 10 55 ± 10 0.23 54.3 ± 9.4 54.2 ± 10.3 0.88
Left Atrial Volume (mean ± SD, in cc) 133 ± 47 131 ± 44 0.59 134 ± 49 129 ± 43 0.23
Primary Left atrial locations prompting
esophageal displacement
Left sided veins’ antra
Right sided veins’ antra
Posterior wall
142(68%)
53(25%)
14(7%)
339(71%)
116(24%)
23(5%)
0.47
133(74%)
38(21%)
9(5%)
129(72%)
43(24%)
8(4%)
0.72
Esophageal Displacement (mean ± SD) 2.61 ± 1 cm NA NA 2.45 ± 0.9 cm NA NA
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
24
Table 2: Differences in efficacy endpoints between EsoSure® and non- EsoSure® groups 453
Efficacy endpoint EsoSure®
(n=180)
Non-EsoSure®
(n=180)
p value
Mean Change in esophageal temperature 0.34±0.59 1.66±0.54 <0.001
Esophageal temperature rise
>1.0ºC(cases)
6(3%) 143(79.4%) <0.001
Total Procedure Time (minutes) 182±36 206±32 <0.001
Total Radiofrequency Duration (minutes) 62±21 65±26 0.22
Total Fluoroscopy Time (minutes) 34.5±13.7 33.1±13.2 0.32
Acute Pulmonary Vein Isolation 179(99.4%) 174(96.7) 0.12
AF recurrence at 3-months 29(16.1%) 51(28.3%) 0.008
AF recurrence at 6-months 37(20.6%) 55(30.6%) 0.04
AF recurrence at 12-months 42(23.3%) 60(33.3%) 0.05
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
25
Table 3: Differences in safety endpoints between EsoSure® and non- EsoSure®
groups
Efficacy endpoint EsoSure® (n=180) Non-EsoSure® (n=180) p value
GI symptoms 8(4.4%) 9(5.0%) 1.0
Major bleeding
complications
2(1.1%) 3(1.6%) 1.0
Minor bleeding
complications
5(2.7%) 4(2.2%) 1.0
TIA/Stroke 0 0 NA
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
26
Figure-1a-c: Features of EsoSure® Device. Figure-1a shows minimal and malleable
curves at room temperature. Figure-1b shows exaggerated and firm curves at body
temperature. Figure-1c shows additional features of device. Ball at the tip prevents
inadvertent puncture of OGT.
Figure-2: Esophageal deviation with EsoSure® (AP-view). Figure-2a: Pre-EsoSure®
esophagogram demonstrates exact relationship of trailing edge of esophagus in relation to
intended lesion delivery site indicated by position of ablation catheter. Temperature probe
in esophagus showing baseline position of esophagus in relation to lasso at ostium of
right superior pulmonary vein and ablation catheter in left superior pulmonary vein. Note
that catheter tip is in proximity of trailing edge of esophagus in this view near left sided
veins. Figure-2b: EsoSure® placed in OGT to deviate esophagus from intended site of
lesion delivery. The deviation trailing edge of esophagus from lesion delivery site was 1.8
cm.
Figure-3: Possible deflection courses for esophagus at displacement with EsoSure®.
This illustration on a CT scan image shows possible courses an esophagus can take on
deflection with EsoSure®. The most common courses are right and left posterior-lateral
which are best appreciated in LAO and RAO views respectively.
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT
MANUSCRIP
T
ACCEPTED
ACCEPTED MANUSCRIPT