Higher incidence of esophageal lesions after ablationof atrial fibrillation related to the use of esophagealtemperature probesPatrick Müller, MD,*† Johannes-Wolfgang Dietrich, MD,† Philipp Halbfass, MD,*

Aly Abouarab,* Franziska Fochler,* Atilla Szöllösi,* Karin Nentwich, MD,* Markus Roos, MD,*

Joachim Krug, MD,* Anja Schade, MD,* Andreas Mügge, MD,† Thomas Deneke, MD*†

From the *Heart Center Bad Neustadt, Clinic for Interventional Electrophysiology, and †University HospitalBergmannsheil, Ruhr University Bochum, Bochum, Germany.

BACKGROUND Endoscopically detected esophageal lesions (EDELs)have been identified in apparently asymptomatic patients after catheterablation of atrial fibrillation (AF). The use of esophageal probes tomonitor luminal esophageal temperature (LET) during catheter ablationto protect esophageal damage is currently controversial.

OBJECTIVE The purpose of this study was to investigate theimpact of the use of esophageal temperature probes during AFcatheter ablation on the incidence of EDELs.

METHODS Eighty consecutive patients (mean age 63.8 � 11.36years; 68.8% men) with symptomatic, drug-refractory paroxysmal (n¼52, 65%) or persistent AF who underwent left atrial radiofrequencycatheter ablation were prospectively enrolled. Posterior wall ablationwas power limited (r25W). In the first 40 patients, LET was monitoredcontinuously (group A), whereas no esophageal temperature probe wasused in group B (n¼ 40 patients). Assessment of EDEL was performedby endoscopy within 2 days after radiofrequency catheter ablation.

RESULTS Overall, 13 patients (16%) developed EDELs after AFablation. The incidence of EDELs was significantly higher in group A

Address reprint requests and correspondence to: Dr Patrick Müller,Clinic for Interventional Electrophysiology, Heart Center Bad Neustadt, BadNeustadt an der Saale, 97616 Germany. E-mail address: Patrick.Mueller-3@ruhr-uni-bochum.de.

1547-5271/$-see front matter B 2015 Heart Rhythm Society. All rights reserved.

than group B (30% vs 2.5%, Po .01). Within group A, patients whodeveloped EDEL had higher maximal LET during AF ablation thanpatients without EDEL (40.97� 0.921C vs 40.14� 1.11C, P¼ .02).Multivariable logistic regression analysis revealed the use of anesophageal temperature probe as the only independent predictorfor the development of EDEL (odds ratio 16.7, P o .01).

CONCLUSION The use of esophageal temperature probes in thesetting of AF catheter ablation per se appears to be a risk factor forthe development of EDEL.

KEYWORDS Atrial fibrillation; Atrioesophageal fistula; Catheterablation; Esophageal lesion; Esophageal temperature monitoring

ABBREVIATIONS AF ¼ atrial fibrillation; BMI ¼ body mass index;EDEL ¼ endoscopically detected esophageal lesion; LA ¼ leftatrial; LET ¼ luminal esophageal temperature; PPI ¼ proton pumpinhibitor

(Heart Rhythm 2015;0:1–6) I 2015 Heart Rhythm Society. Allrights reserved.

IntroductionAtrial fibrillation (AF) is the most common sustainedarrhythmia. Currently, AF affects 5 million Americans and6 million Europeans, and its prevalence is expected toincrease over the next several decades.1,2 In considerationof the rising prevalence of AF, the use of catheter ablation totreat patients with symptomatic drug-refractory AF hasgrown rapidly during the past decade.1 Although majorcomplications in the context of AF ablation are rare,atrioesophageal fistula is the most devastating complicationof left atrial ablation and often is fatal.3 In contrast to a lowincidence of atrioesophageal fistula (0.03%–0.05%),4,5

recent studies revealed a high incidence of asymptomatic

endoscopically detected esophageal lesions (EDELs) inpatients after AF ablation that ranged from 10% to 48%.6–11

A lower body mass index (BMI), the use of generalanesthesia, maximal energy at the posterior left atrial wall,maximal esophageal temperature during ablation, and type ofablation performed appear to be predictors for the develop-ment of postablation EDEL.6–12 To avoid esophageal injuryduring AF ablation, many centers routinely use continuousluminal esophageal temperature (LET) monitoring by esoph-ageal temperature probe, but to date, the impact of the use ofan esophageal temperature probe by itself on the incidence ofEDEL has been controversial.

The purpose of this study was to investigate whether theuse of an esophageal temperature probe per se affects theincidence of EDEL after AF ablation. For this reason,2 ablation groups were compared prospectively in our center:group A, with continuous LET using a temperature probe,and group B, in which LET monitoring was excluded.

http://dx.doi.org/10.1016/j.hrthm.2015.04.005

Figure 1 The SensiTherm probe. This esophageal temperature probe hasa 7F catheter, is not steerable, and has an atraumatic silicone tip and 3 middleelectrodes to monitor esophageal temperature in 3 locations, whereasproximal and distal electrodes can be used for pacing and sensing.

Heart Rhythm, Vol 0, No 0, Month 20152

MethodsStudy populationIn this study, we recruited 80 patients presenting for radio-frequency catheter ablation of symptomatic, drug-resistantparoxysmal (65%) or persistent (35%) AF. Transesophagealechocardiography was performed on the day before theablation procedure to rule out left atrial thrombi in everypatient. Patients were excluded from the study if they hadknown history of esophageal disease or symptoms of gastro-intestinal disease (dysphagia, heart burn, abdominal pain). Inthe first 40 patients, AF ablation was monitored by LET(group A); the next consecutive 40 patients underwentablation without esophageal temperature control (group B).

This study was approved by the local institutional reviewboard and conforms to the declaration of Helsinki of 2008.All authors had full access to the data and have read andagreed to the manuscript as written. All procedures followedwere in accordance with the ethical standards of theresponsible committee on human experimentation (institu-tional and national) and with the Helsinki Declaration of1975, as revised in 2000. Written informed consent wasobtained from all patients to be included in this study.

Electrophysiological study and ablation procedureIrrigated-tip radiofrequency ablation was performed with theCARTO 3 (Biosense Webster, Diamond Bar, CA) or NavX(St. Jude Medical, St. Paul, MN) electroanatomic mappingsystem with a ThermoCool Smart Touch catheter (BiosenseWebster). Ablations were only performed with a contactpressure of 10 g to a maximum of 40 g. All left atrial ablationprocedures were performed with the patient in deep sedationusing propofol infusion without general anesthesia and viasingle transseptal access to the left atrium by use of thestandard technique. A circular diagnostic pulmonary veinmapping catheter was used to create a 3-dimensional shell ofthe left atrium, and prior computed tomography images wereintegrated by merging. Isolation of the ipsilateral pulmonaryveins was performed en bloc in a point-by-point fashion andstarted only after the activated clotting time reached Z300seconds. Intraprocedurally, activated clotting times werecontrolled every 20 minutes during the course of theprocedure. Additional ablations were performed if low-voltage areas were identified on bipolar voltage maps(o0.5 mV). Ablation patterns were at the discretion of theoperator, and low-voltage areas were usually encircled.Ablation was performed with a maximum of 35 W with atarget temperature of 431C and a maximum irrigation rate of30 mL/min. Maximal energy delivery at the posterior wallwas reduced to a maximum of 25 W. To monitor LET, anintraluminal temperature probe (SensiTherm, FIAB, Firenze,Italy) was used in the first 40 patients (group A). Thistemperature probe has a 7F body and 5 olive-shapedelectrodes. This multisensor probe has 3 middle electrodesto monitor esophageal temperature in 3 locations, whereasproximal and distal electrodes can be used for pacing andsensing. This probe is not steerable and has an atraumatic

silicone tip to simplify esophageal insertion (Figure 1). Thelocation of the probe was adapted with regard to the site ofablation. In addition, radiofrequency energy was discontin-ued when the temperature of the esophagus probe reached39.51C (group A). Within the next 40 consecutive patients,no temperature probe was used (group B). Procedures wereperformed either under continued oral anticoagulation withwarfarin and a therapeutic international normalized ratio(2.0–2.5) or with continued novel oral anticoagulation(Figures 2 and 3).

Postablation esophageal endoscopyIn brief, esophageal endoscopy was performed 1 to 2 daysafter ablation in all patients to assess for the presence andextent of esophageal endoluminal thermal injury. Findingswere photo documented and classified as (1) no lesion, (2)erosion (erythema with intact mucosa), (3) ulceration, or (4)perforation. Investigators of postablation esophageal endos-copy were blinded to the use of the temperature probe.

Clinical course of patients with EDELIn case of EDEL, esophageal endoscopy was repeated within2 weeks. Patients with EDEL received both a liquid diet andproton pump inhibitors (PPIs) at double the standard doseuntil repeat endoscopy revealed healing of the esophagus.If EDEL persisted at second endoscopy, PPIs in double thestandard dose were continued until repeat endoscopyrevealed healing of the esophagus.

Statistical analysisData are expressed as mean � SD or median (interquartilerange) for continuous variables or as numbers and propor-tions for categorical variables. Depending on the class ofanalyzed data, univariate analysis was performed withunpaired Student t test, Wilcoxon-Mann-Whitney U test, orχ2 test, respectively. For multivariable logistic regression,

Figure 2 Fluoroscopic views illustrating the location of the esophageal temperature probe in relation to the ablation catheter during ablation at the rightsuperior pulmonary vein (A) and the right inferior pulmonary vein (B). Abl¼ irrigated-tip radiofrequency ablation catheter; Esi¼ esophageal temperature probe;Lasso ¼ Lasso catheter.

3Müller et al Temperature Probe and Esophageal Lesions After AF Ablation

outcome variables were selected that differed between theEDEL and non-EDEL group with P o .1 in univariateanalysis. An initial maximal model was successively sim-plified by eliminating nonsignificant parameters and predic-tors with an odds ratio around 1. Statistical analyses wereperformed with SPSS and custom S scripts written for theenvironment R 2.10.1 on a Mac OS X operating system.

ResultsPatient characteristicsA total of 80 patients undergoing radiofrequency ablationfor AF were included in this study. All patients under-went esophageal endoscopy within 2 days after AF abla-tion. The clinical characteristics of the study population are

Figure 3 Endoscopic image of an esophageal ulcer (encircled) aftersingle-tip catheter radiofrequency ablation in a patient with an esophagealtemperature probe (group A).

summarized in Table 1. The mean age of the studypopulation was 63.8 � 11.36 years, and 68.8% were men.Patients with EDEL had lower BMI than patients withoutEDEL, but without statistical significance (26.65 � 1.87 vs27.76 � 2.43 kg/m2, P ¼ .07).

Incidence of EDEL after AF ablationOverall, 13 patients (16%) developed asymptomatic EDELafter AF ablation. Specifically, 10 patients developed esoph-ageal erosion, and 3 presented with esophageal ulceration.No esophageal perforation or atrioesophageal fistula occur-red in the study cohort during the 3 months after the ablationprocedure.

Ablation parameters and impact of use ofesophageal temperature probe on EDELAlthough clinical and procedural parameters were compara-ble between group A and group B, the incidence of EDELwas significantly higher if an esophageal temperature probewas used during left atrial ablation (group A; 30% vs 2.5%;P o .01) (Table 1). Neither maximal energy applicationduring left atrial ablation nor additional ablation lines had animpact on the incidence of EDEL (Table 2). Within group A,patients who developed EDEL had higher maximal LETduring AF ablation than patients without EDEL (40.97 �0.921C vs 40.14 � 1.11C, P ¼ .02), whereas a temperaturerise Z39.51C during ablation showed no impact on theincidence of EDEL (Table 3).

Logistic regression model for predictors ofesophageal lesionsThe initial maximal multivariable logistic regression modelcontained BMI and the use of an esophageal temperatureprobe as possible predictors of EDEL. After elimination ofnonsignificant parameters, the use of an esophageal probe

Table 1 Characteristics of patients with and without temperature monitoring with esophageal probe

Patients without temperaturemonitoring (n ¼ 40)

Patients with temperaturemonitoring (n ¼ 40) P value

Age (y) 63.75 � 10.72 63.83 � 12.25 1.00Men:women (n) 28:12 27:13 .98Body mass index (kg/m2) 27.94 � 2.35 27.22 � 2.37 .17CHA2DS2VASc score 2 (1.25) 2 (2) .867PPI pre-AF ablation 9/40 7/40 .78PPI post-AF ablation 39/40 40/40 1.00Procedural time (min) 126.93 � 39.82 119.43 � 27.03 .328Ablation time (min) 30.09 � 10.4 31.11 � 10.74 .685Ablation time at left atrial posterior wall (min) 14.03 � 4.43 14.30 � 4.83 .791Mean energy (W) 33 � 2.95 33.13 � 2.45 .837Additional ablation line 16/40 12/40 .482EDEL after AF ablation 1/40 12/40 o.01

Values are mean � SD, or counts, respectively.AF ¼ atrial fibrillation; EDEL ¼ endoscopically detected esophageal lesion; PPI ¼ proton pump inhibitor.

Heart Rhythm, Vol 0, No 0, Month 20154

remained the only independent predictor for the developmentof EDEL (odds ratio 16.7, P o .01) in the resulting minimalmodel of logistic regression (Table 4).

DiscussionOur study evaluated whether the use of an esophageal probeto monitor esophageal temperature during AF ablation itselfaffects the development of EDEL. With a total EDELincidence of 16% in our patient cohort, we report an eventrate that is well in line with previous published studies.To the best of our knowledge, the present study is the firststudy to investigate the impact of an esophageal temperatureprobe by itself on the incidence of EDEL after AF ablationwith irrigated radiofrequency ablation catheters.

Main findingsThe main findings were as follows: (1) The incidence ofEDEL after AF ablation is high (16%); (2) patients withEDEL tend to have a slightly lower BMI; (3) when anesophageal temperature probe was used, patients with EDELshowed higher maximal LET during AF ablation; and (4) theuse of an esophageal temperature probe in itself significantlyincreased the risk of esophageal damage.

Table 2 Clinical characteristics of the study population

PatieAF a

Age (y) 63.Men:women (n)Body mass index (kg/m2) 27.CHA2DS2VASc scorePPI pre-AF ablationPPI post-AF ablationProcedural time (min) 123.Ablation time (min) 30.Ablation time at left atrial posterior wall (min) 13.Mean energy (W) 32.Additional ablation lineUse of esophageal temperature monitoring during AF ablation

Values are mean � SD, or counts, respectively.AF ¼ atrial fibrillation; EDEL ¼ endoscopically detected esophageal lesion; PP

Incidence and predictors of EDEL after AF ablationAlthough the development of an atrioesophageal fistula inthe context of AF ablation is rare, its clinical outcome isdevastating and associated with high lethality.3 In consid-eration of this complication, the goal should be to understandthe pathophysiology and minimize esophageal thermalinjury during AF ablation.

Recent studies have reported clinically silent EDEL afterAF ablation using esophageal endoscopy or capsule endos-copy in up to 48% of cases.6–11 In the present study, 13 of 80patients (16%) developed EDEL after AF ablation.

Previous studies have identified a lower BMI, the use ofgeneral anesthesia, maximal energy at the posterior left atrialwall, maximal esophageal temperature during ablation, andtype of ablation performed as potential confounders of esoph-ageal injury in the context of AF ablation.6–12 The present studyindicates that the use of LET monitoring in and of itself maycontribute to esophageal injury during AF ablation.

Esophageal temperature monitoring during AFablation: benefit or risk?Tools to prevent esophageal damage during AF ablation aregreatly needed. To avoid esophageal wall injury during AF

nts without EDEL afterblation (n ¼ 67)

Patients with EDEL afterAF ablation (n ¼ 13) P value

75 � 10.51 64 � 15.92 .95748:19 7:6 .347

76 � 2.43 26.65 � 1.87 .0782 (2) 2 (3) .98413/67 3/13 .93966/67 13/13 .35757 � 33.86 121.125 � 36.2 .82719 � 9.58 32.38 � 14.23 .60591 � 4.16 15.46 � 6.49 .4299 � 2.76 33.46 � 2.4 .53123/67 5/13 .97528/67 12/13 o.01

I ¼ proton pump inhibitor.

Table 3 Ablation data for patients with temperature monitoring, with and without EDEL after AF ablation

Patients without EDEL (n ¼ 28) Patients with EDEL (n ¼ 12) P value

Temperature rise Z39.51C 19/28 (68) 11/12 (92) .23Mean maximal esophageal luminal temperature (1C) 40.14 � 1.1 40.97 � 0.92 o.05

Values are mean � SD or counts (%).AF ¼ atrial fibrillation; EDEL ¼ endoscopically detected esophageal lesion.

5Müller et al Temperature Probe and Esophageal Lesions After AF Ablation

ablation, it is common practice to limit radiofrequencygenerator power settings (25–30 W) and ablation duration(o30 seconds) when ablating at the posterior left atrial wall.Additionally, it is common to treat patients peri-interven-tionally with PPIs.13

The use of an esophageal probe for LET monitoringto avoid esophageal injury during AF ablation has beendiscussed controversially in the current guidelines; never-theless two-thirds of the members of the Heart RhythmSociety Task Force on Catheter and Surgical Ablation ofAtrial Fibrillation used esophageal temperature probes.14

Singh et al15 reported that the use of an esophagealtemperature probe might be associated with a significantlylower incidence of EDEL compared with patients with thesame energy settings but without LET measurement. Nota-bly, the group without LET monitoring was small (14 of 81patients), had significantly lower BMI (28� 3 vs 31� 6 kg/m2,P ¼ .05), and more often had general anesthesia during AFablation (43% vs 13%, P ¼ .01).

Continuous esophageal LET monitoring is able to detect asignificant rise of temperature and maximal temperaturelevels of the esophagus during AF ablation. Althoughprevious studies indicated no association between esoph-ageal temperature rise and esophageal mucosal injury, incongruence to our findings, mean maximal temperaturelevels were significantly higher in patients with EDEL(Table 3).7,15 Notably, Cumming and coworkers16,17 showedin an animal model that the rise of intraluminal esophagealtemperature was too low compared with esophageal tissuetemperature to be able to be detected in a timely manner andthus to prevent EDEL and that it was independent of energydelivery. This may give the operator a false impression ofsafety.

Deneke et al showed that LET monitoring duringpulmonary vein isolation using duty-cycled phased radio-frequency ablation was associated with the incidence ofesophageal damage and suggested that LET monitoring byitself may contribute to thermal effects on the esophagus.6

Additionally, the present study revealed a more than 10-foldelevated incidence of EDEL after radiofrequency ablationwith an irrigated-tip catheter if LET monitoring was used(30% vs 2.5%, P o .01).

Table 4 Minimal model of logistic regression analysis of risk of EDEL a

B SE Wald

Use of esophageal temperature probe 2.816 1.070 6.9

Independent variables were chosen by backward selection as described.AF ¼ atrial fibrillation; CI ¼ confidence interval; EDEL ¼ endoscopically detec

Pathophysiologic mechanismsAlthough the precise mechanism of EDEL is not fully under-stood, potential mechanisms include direct thermal injury, acidreflux, infection from the lumen, and ischemic injury throughthermal occlusion of end-arterioles, which may explain thedeferred development of atrioesophageal fistula. Because ofthe very small distance between the anterior esophagus walland the left atrial wall (2.5–4.5 mm),18 it is clear that collateralthermal injury to the esophagus during AF ablation may beable to trigger a cascade of events that potentially results in thedevelopment of atrioesophageal fistula.

Besides the effect of direct catheter ablation–induced esoph-ageal tissue damage, Sternick et al19 hypothesized that thedevelopment of a posterior esophageal lesion during AF ablationmay be related to radiofrequency inductive heating in thestainless steel olive-shaped thermocouple of the esophagealprobe, which acts as an antenna. Therefore, the metal thermistorelectrodes of the esophageal temperature probes used in our studycohort may interact with the radiofrequency current and may actas secondary antennas during the radiofrequency application.20,21

Study limitationsThe major limitations of the present single-center study are itsnonrandomized fashion and the rather small number of patientsenrolled, which limits its statistical power. This may be ofrelevance with regard to the fact that additional ablation lines(in 28 of 80 patients) did not affect the development of EDEL.

It cannot be excluded that esophageal damage might havebeen caused by the insertion and manipulations of the temper-ature probe in group A or the transesophageal echocardiog-raphy probe. Because no esophageal endoscopy was performedbefore AF ablation, we can only assume that the location in theesophagus makes the EDEL likely to be related to AF ablation.

We did not specifically document tissue contact pressureduring ablations at the posterior wall but have used contactforce measurement occasionally during this study. There-fore, differences in mean contact force between groups mayexist. In addition, no consequences were taken from potentialpain response of patients when ablating at the posterior wall.Patients were deeply sedated, and ablations were per proto-col not stopped if sensation of pain occurred.

fter AF ablation

Degrees of freedom P value OR (95% CI)

1 o.01 16.71 (2.05–136.1)

ted esophageal lesion; OR ¼ odds ratio; SE ¼ standard error.

Heart Rhythm, Vol 0, No 0, Month 20156

Because LET monitoring in group A was only performedwith the SensiTherm probe, the higher risk of EDEL cannotbe generalized to other probes.

Finally, the present findings need to be validated in largerrandomized trials to substantiate our findings and to evaluatewhether these findings can be applied to other designs ofLET probes.

ConclusionIn patients undergoing radiofrequency ablation of AF, theincidence of collateral thermal injury of the esophagus (EDEL)is high. Our results suggest that although the informationprovided by maximal LET during AF ablation is helpful, itsbenefit may be outweighed by the use of the SensiThermesophageal temperature probe, which appears to be an inde-pendent risk factor for the development of EDEL. Further studiesare needed to substantiate the hypothesis that esophageal damageis linked to the use of esophageal temperature measurement byselecting different LET devices to reduce the incidence of EDEL.

References1. Camm AJ, Kirchhof P, Lip GY, et al. Guidelines for the management of atrial

fibrillation: the Task Force for the Management of Atrial Fibrillation of theEuropean Society of Cardiology (ESC) [published correction appears in EurHeart J 2011;32:1172]. Eur Heart J 2010;31:2369–2429.

2. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV, Singer DE.Prevalence of diagnosed atrial fibrillation in adults: national implications forrhythm management and stroke prevention: the AnTicoagulation and RiskFactors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285:2370–2375.

3. Mohanty S, Santangeli P, Mohanty P, et al. Outcomes of atrioesophageal fistulafollowing catheter ablation of atrial fibrillation treated with surgical repair versusesophageal stenting. J Cardiovasc Electrophysiol 2014;25:579–584.

4. Ghia KK, Chugh A, Good E, Pelosi F, Jongnarangsin K, Bogun F, Morady F,Oral H. A nationwide survey on the prevalence of atrioesophageal fistula after leftatrial radiofrequency ablation. J Interv Card Electrophysiol 2009;1:33–36.

5. Pappone C, Oral H, Santinelli V, et al. Atrioesophageal fistula as a complicationof percutaneous transcatheter ablation of atrial fibrillation. Circulation 2004;109:2724–2726.

6. Deneke T, BünzK,BastianA, PäslerM,AndersH, LehmannR,MeuserW, deGroot JR,Horlitz M, Haberkorn R, Mügge A, Shin DI. Utility of esophageal temperaturemonitoring during pulmonary vein isolation for atrial fibrillation using duty-cycledphased radiofrequency ablation. J Cardiovasc Electrophysiol 2011;3:255–261.

7. Rillig A1, Meyerfeldt U, Birkemeyer R, Wiest S, Sauer BM, Staritz M, Jung W.Oesophageal temperature monitoring and incidence of oesophageal lesions afterpulmonary vein isolation using a remote robotic navigation system. Europace2010;5:655–661.

8. Knopp H, Halm U, Lamberts R, Knigge I, Zachäus M, Sommer P, Richter S,Bollmann A, Hindricks G, Husser D. Incidental and ablation-induced findingsduring upper gastrointestinal endoscopy in patients after ablation of atrialfibrillation: a retrospective study of 425 patients. Heart Rhythm 2014;4:574–578.

9. Sohara H, Satake S, Takeda H, Yamaguchi Y, Nagasu N. Prevalence ofesophageal ulceration after atrial fibrillation ablation with the hot balloon ablationcatheter: what is the value of esophageal cooling? J Cardiovasc Electrophysiol2014;7:686–692.

10. Halm U, Gaspar T, Zachäus M, Sack S, Arya A, Piorkowski C, Knigge I,Hindricks G, Husser D. Thermal esophageal lesions after radiofrequency catheterablation of left atrial arrhythmias. Am J Gastroenterol 2010;105:551–556.

11. Di Biase L, Saenz LC, Burkhardt DJ, et al. Esophageal capsule endoscopy afterradiofrequency catheter ablation for atrial fibrillation: documented higher risk ofluminal esophageal damage with general anesthesia as compared with conscioussedation. Circ Arrhythm Electrophysiol 2009;2:108–112.

12. MartinekM, Bencsik G, Aichinger J, Hassanein S, SchoeflR, Kuchinka P, Nesser HJ,Purerfellner H. Esophageal damage during radiofrequency ablation of atrial fibrilla-tion: impact of energy settings, lesion sets, and esophageal visualization. J CardiovascElectrophysiol 2009;7:726–733.

13. Zellerhoff S, Lenze F, Eckardt L. Prophylactic proton pump inhibition after atrialfibrillation ablation: is there any evidence? Europace 2011;13:1219–1221.

14. Calkins H, Kuck KH, Cappato R, et al. Heart Rhythm Society Task Force onCatheter and Surgical Ablation of Atrial Fibrillation. 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrialfibrillation: recommendations for patient selection, procedural techniques,patient management and follow-up, definitions, endpoints, and research trialdesign: a report of the Heart Rhythm Society (HRS) Task Force on Catheter andSurgical Ablation of Atrial Fibrillation. Developed in partnership with theEuropean Heart Rhythm Association (EHRA), a registered branch of theEuropean Society of Cardiology (ESC) and the European Cardiac ArrhythmiaSociety (ECAS); and in collaboration with the American College of Cardiology(ACC), American Heart Association (AHA), the Asia Pacific Heart RhythmSociety (APHRS), and the Society of Thoracic Surgeons (STS). Endorsed by thegoverning bodies of the American College of Cardiology Foundation, theAmerican Heart Association, the European Cardiac Arrhythmia Society, theEuropean Heart Rhythm Association, the Society of Thoracic Surgeons, the AsiaPacific Heart Rhythm Society, and the Heart Rhythm Society. Heart Rhythm2012;4:632–696.

15. Singh SM, d'Avila A, Doshi SK, Brugge WR, Bedford RA, Mela T, Ruskin JN,Reddy VY. Esophageal injury and temperature monitoring during atrialfibrillation ablation [published correction appears in Circ Arrhythm Electro-physiol 2012;5:e30]. Circ Arrhythm Electrophysiol 2008;3:162–168.

16. Cummings JE, Schweikert RA, Saliba WI, Burkhardt D, Brachmann J, Guenther J,Schibgilla V, Verma A, Dery MA, Drago JL, Kilicas-lan F, Natale A. Assessmentof temperature, proximity, and course of the esophagus during radiofrequencyablation within the left atrium. Circulation 2005;112:459–464.

17. Cummings JE, Barrett CD, Litwak KN, et al. Esophageal luminal temperaturemeasurement underestimates esophageal tissue temperature during radiofre-quency ablation within the canine left atrium: comparison between 8 mm tipand open irrigation catheters. J Cardiovasc Electrophysiol 2008;19:641–644.

18. Rolf S, Boldt LH, Parwani AS, Wutzler A, Huemer M, Blaschke D, Herberger E,Haverkamp W. Findings and outcome of fluoroscopic visualization of theoesophageal course during catheter ablation of atrial fibrillation. Europace2011;6:796–802.

19. Sternick EB, Persiano AC, Arantes V. Is it safe to monitor oesophagealtemperature during AF ablation? Europace 2012;10:1432.

20. Boll DT, Lewin JS, Duerk JL, Merkle EM. Do surgical clips interfere withradiofrequency thermal ablation? AJR Am J Roentgenol 2003;180:1557–1560.

21. Goldberg SN, Ryan TP, Hahn PF, SchimaW, Dawson SL, Lawes KR,Mueller PR,Gazelle GS. Transluminal radiofrequency tissue ablation with use of metallicstents. J Vasc Interv Radiol 1997;8:835–843.

CLINICAL PERSPECTIVESAlthough catheter ablation of AF has evolved rapidly during the past decade, recent attention has been focused on ablation-related complications. A rare but devastating complication of left atrial ablation is the development of atrioesophagealfistula. To prevent esophageal injury, continuous LET monitoring during left atrial ablation procedures has been used. Thisstudy compared the use and nonuse of continuous LET monitoring during AF ablation to investigate the impact of theesophageal temperature probe in and of itself on the development of asymptomatic EDELs. Our results indicate that the useof LET monitoring, with the SensiTherm probe, itself appears to involve a risk for the development of EDEL. On the basisof these results, validation of our findings in a prospective randomized trial is needed. Moreover, because in this study onlythe SensiTherm probe was used for LET monitoring, and other esophageal probes are available, the impact of differentesophageal probes on the development of EDEL should be investigated.