Atrioesophageal fistula formation with cryoballoon ablation is most commonly related to the left inferior pulmonary vein

Atrioesophageal fistula formation with cryoballoon ablation is most commonly related to the left inferior pulmonary vein Roy M. John, MBBS, PhD, FHRS, Sunil Kapur, MD, Kenneth A. Ellenbogen, MD, FHRS, Jayanthi N. Koneru, MD From the Brigham and Women s Hospital, Boston, Massachusetts, and VCU School of Medicine and the Medical College of Virginia Hospital, Richmond, Virginia. BACKGROUND Collateral damage has been reported with use of the cryoballoon for pulmonary vein isolation. OBJECTIVE The purpose of this study was to determine the incidence and characteristics associated with atrioesophageal fistula (AEF) after cryoballoon use. METHODS Cases of AEF reported with use of the cryoballoon since 2011 were collected from the Manufacturer and User Facility Device Experience (MAUDE) database, publications, and the manufacturer s database. Lowest balloon temperatures were compared with matched control patients undergoing cryoballoon ablation without AEF formation. Location of AEF was compared with AEF associated with radiofrequency ablation. RESULTS A total of 11 cases of AEF were identified from a worldwide experience that exceed 120,000 cases. Mean age was 60 (range 31 78 years), and 80% of patients were male. Although mean lowest balloon temperatures were no different between patients with AEF and those with no AEF ( 58.5 C ± 7.2 C vs 56 C ± 2.6 C, P ¼ NS), balloon inflation times were longer in patients with AEF (238.8 ± 54.8 seconds vs 178.1 ± 37.5 seconds in the non-aef group, P.001) All cases of AEF for which location was identified occurred in relation to the left pulmonary veins. The left inferior pulmonary vein (LIPV) was involved in 8 of 10 patients with cryoballoon compared to 0 of 11 patients in the radiofrequency group (P o.05). Mortality for cryoballoonassociated AEF was 64%. CONCLUSION AEF after cryoballoon use is rare (o1 in 10,000) and most commonly was identified near the LIPV. Proximity of the esophagus to the LIPV and evidence of esophageal luminal cooling should be considered indications to limit cryoablation at this vein. KEYWORDS Atrioesophageal fistula; Atrial fibrillation ablation; Cryoballoon; Pulmonary vein isolation; Catheter ablation; Ablation complications (Heart Rhythm 2017;14:184 189) I 2016 Heart Rhythm Society. All rights reserved. Introduction The Arctic Front cryoballoon (Medtronic Inc, Minneapolis, MN) has emerged as a safe and effective tool for electrical isolation of the pulmonary veins (PVs) in the treatment of atrial fibrillation. 1 5 The initial market-approved iteration of the cryoballoon the Arctic Front I (AF I), had a design in which cooling was primarily limited to the equatorial surface of the balloon. Modifications in the Arctic Front II (AF II) resulted in a larger cooling surface involving the entire leading hemisphere of the balloon surface. Use of this balloon for pulmonary vein isolation (PVI) shows a 1-year freedom from atrial fibrillation of 70% 90%. 6 8 Major procedure-related adverse events with the Arctic Front II Dr. John has received research support from St. Jude Medical and Biosense Webster. Dr. Ellenbogen has received consulting fees and speaking honoraria from Medtronic. Drs. Kapur and Koneru report they have no relationships relevant to the contents of this paper to disclose. Address reprint requests and correspondence: Dr. Roy M. John, Arrhythmia Section, Brigham and Women s Hospital, 75 Francis St, Boston, MA 02115. E-mail address: rjohn2@partners.org. have been reported in 5.3% of patients, with phrenic nerve injury being the most important complication. Unresolved phrenic nerve injury at hospital discharge has been documented in up to 4% of patients. 9 Atrioesophageal fistula (AEF) is a well-recognized complication of radiofrequency (RF) ablation in the left atrium. It is estimated to occur with a frequency of 1 in 500 1000 cases. 10 After initial experience with the cryoballoon suggested a low risk for AEF, several reports of this complication emerged in the literature. 11 14 Given the low incidence of AEF, factors that predispose to it are difficult to define. The purpose of this study was to analyze the anatomic and procedural factors associated with this rare but devastating complication of PVI with the cryoballoon. Methods All cases of AEF reported since the launch of the Arctic Front cryoballoon in 2011 were identified from published reports in the literature, the Manufacturer and User Facility Device Experience (MAUDE) database, and the internal 1547-5271/$-see front matter B 2016 Heart Rhythm Society. All rights reserved. http://dx.doi.org/10.1016/j.hrthm.2016.10.018

John et al Atrioesophageal Fistula and Cryoballoon Ablation 185 complaint database of the manufacturer (Medtronic Inc). Demographics, procedural data, and mode and time of presentation of AEF were collated when available. Five cases were reported in the literature. Details of the remaining five patients who were not included in the case reports were obtained from the Medtronic and MAUDE databases. One case that was included in the MAUDE database as a possible case of AEF has been included in this series because of its typical clinical features that made AEF highly likely. For comparison of procedural variables, data were extracted from 67 consecutive uncomplicated cases of cryoballoon ablation for 1 year preceding and 1 year after a case of AEF at our institution. For comparison of AEF locations between cryoballoon and RF ablation, we found a total of 10 cases of AEF related to RF ablation in which the anatomic locations of the AEF were identified by computed tomographic imaging, surgery, or at autopsy. These cases were identified from the literature and through survey of physicians in the New England area. Data on the number of cryoballoon inflations, inflation times, and balloon temperatures for the comparator group were obtained from the Medtronic CryoConsole. Esophageal temperature monitoring was performed using a disposable 18Fr esophageal stethoscope Vital Signs model 1011 EU (CareFusion, Vernon Hills, IL). The lowest endoluminal esophageal temperature (LET) was defined as the nadir of temperature recordings with the temperature probe positioned at the level of the inflated balloon in the left and right anterior oblique fluoroscopic views. The study was approved by the institutional review board. All patient information has been de-identified. Statistical analysis Categorical variables are expressed as mean ± SD. Continuous variables were compared using the unpaired Student t test. P o.05 was considered significant. Results A total of 10 definite and 1 probable AEF cases related to use of Arctic Front cryoballoons since the commercial launch in 2011 were identified. Mean patient age was 58.9 ± 13.7 years, with 8 males and 2 females (no details in 1 case). The AF I and AF II cryoballoons were used in 3 and 8 cases, respectively. Table 1 lists procedural details. The 28-mm balloon was used for all but 1 patient in the AEF group. Mean number of balloon inflations lasting 430 seconds was 14 ± 7.6 in the AEF group and 12 ± 3.2 in the comparator group (P ¼ NS). Mean lowest balloon temperature recorded in the AEF cases was 58.5 C ± 7.2 C and was not significantly different from that in the comparator group ( 56 C ± 2.6 C, P ¼.14). Lowest balloon temperature for ablations in the left inferior pulmonary vein (LIPV) was available for 6 patients and was not significantly different from that in the comparator group ( 58.3 C ± 5.6 C vs 52.6 C ± 4.4 C, P ¼.34). However, mean total balloon inflation times at the PVs were significantly longer in the AEF group at 238.8 ± 54.8 seconds vs 178 ± 37.5 seconds for the comparator group (P o.001) (Figure 1). Duration of inflation in the LIPV was available for 5 cases in the AEF group and was again significantly longer for the AEF group (260 ± 86 seconds vs 172.1 ± 39 seconds; P o.001) (Figure 1). Eight of 10 cases of AEF were recorded in relation to the LIPV. In 1 patient, vegetations were seen endocardially in relation to the left superior PV. In another patient, the fistula was recorded between the left superior and inferior veins. No AEFs were registered in relation to the right veins (Figure 2). In contrast, known locations of AEF related to RF ablation cases were distributed between the right PVs, mid LA, and left superior PV locations (Figure 3). The occurrence of AEF in relation to LIPVs was significantly higher for cryoballoon ablation compared to RF ablation cases (P o.05). Endoluminal esophageal temperatures were monitored in only 3 cases of AEF with cryoballoon (Figure 4). In 1 case (case #9 in table), the balloon was deflated in the LIPV when the temperature fell from 35.8 C to 24 C. The temperature continued to drop to a lowest temperature of 20.9 C. Another case registered 20 C as the lowest endoluminal temperature. but the temperature probe used had a lower limit for registration of 20 C, so the lowest temperature is not known other than it likely was lower. A third case had a drop in esophageal temperature to 31 C, but in this case the temperature probe had malfunctioned. Seven of the 11 cases of AEF resulted in death, for a mortality rate of 64%. Of the 4 survivors, at least 1 patient was left with significant neurologic deficits. Discussion With the increasing use of the Arctic Front cryoballoon for PVI, the limitations and potential complications related to its use must be recognized. Phrenic nerve palsy, gastroparesis, bronchial damage, and AEFs have been reported. 1,9,14,15 This study collated the reported cryoballoon cases of AEF and found that AEF is primarily associated with the left PVs, particularly the LIPVs. In 8 of 10 cases in which the location of the AEF was evident on imaging, autopsy, or surgery, the location of the AEF was closely associated with the LIPV, suggesting an increased susceptibility to thermal damage to the esophagus from use of the cryoballoon in this vein. Of the available data on procedural variables, only the duration of balloon inflation in the AEF cases was associated with AEF and was found to be significantly longer compared to that in a non-aef group. Because uniform cooling is obtained circumferentially with the cryoballoon, structures posterior to the thinner LA wall remain susceptible to collateral damage with cooling. PVI with the cryoballoon has been reported to cause bronchial erosion and hemoptysis. 15 Ice formation in the left mainstream bronchus has been demonstrated in patients undergoing simultaneous bronchoscopy. 16 In a recent study, esophageal thermal lesions with the second-generation cryoballoon were detected by endoscopy in 20% of patients

186 Table 1 Case no. Age, sex Patient details, procedural findings, and outcome Arctic Front Generation I/II Balloon size (mm) No. of ablations Minimum balloon temperature ( C) Median ablation duration (sec) Location of AEF Findings and outcome 1 64, M I 23 13 75 240 TEE: Vegetation near left superior pulmonary vein 2 78, M I 28 10 67 (LIPV: 3 applications; lowest 67) 3 58, F I 28 9 55 (LIPV: 2 inflations for 240 s; lowest in LIPV 53) 4 75, F II 28 30 52 (LIPV: 3 for 240 s; lowest in LIPV 52) 5 48, M II 28 20 59 (LIPV: 2 for 180 s; lowest 59) AEF at surgery; fatal 360 LIPV 8-mm AEF in posterior wall of LIPV surgically repaired; left hospital alive but central nervous system defects 11 240 LIPV Presented with dysphagia 2 weeks later; died after surgery to close AEF at LIPV 14 240 LIPV LET 31 C but probe malfunction; presented with fever 17 days later; subsequent CVA; AEF near LIPV repaired at surgery; death from sepsis and ischemic bowel 14 240 LIPV LET 20 C but probe does not register below 20 C; presented with fever 2 weeks later; TEE thrombus at LIPV ostium; collapse, LAD thrombus, and death; LIPV thrombus with hole in esophagus at autopsy 14 270 LIPV Presented with CVA; CT showed AEF at LIPV; esophageal stent; eied 12 6 61, M II 28 8 61 (LIPV: 3 for 300 s; lowest 55) 7 31, M II 28 9 61 200 LIPV Presented with hemoptysis and fever; TEE followed by CVA; CT showed AEF at LIPV; surgical repair; died 13 8 66, M II 28 LIPV Surgical closure; survived 9 58, M II 28 8 60 (LIPV: 2 for 180 s; lowest 60) LET 20.9 C 180 LIPV Presented with fever and TIA 3 weeks later; CT showed air adjacent to LIPV; immediate surgery confirmed AEF to LIPV; alive with no deficits 10 50, M II 28 20 53 180 Between LUPV and Surgical closure; survived LIPV 11 II 28 Unknown Presented 6 weeks after cryoballoon PVI with fever and stroke; died; no autopsy; cause of death listed as stroke; AEF likely but not confirmed AEF ¼ atrioesophageal fistula; CT ¼ computed tomography; CVA ¼ cerebrovascular accident; LAD ¼ left anterior descending coronary artery; LET ¼ lowest esophageal temperature; LIPV ¼ left inferior pulmonary vein; LUPV ¼ left upper pulmonary vein; PVI ¼ pulmonary vein isolation; TEE ¼ transesophageal echocardiography; TIA ¼ transient ischemic attack. Heart Rhythm, Vol 14, No 2, February 2017

John et al Atrioesophageal Fistula and Cryoballoon Ablation 187 Figure 1 Longest balloon application times during the case and in the left inferior pulmonary vein. AEF ¼ atrioesophageal fistula. at a mean of 1.5 days after ablation. 17 The exact mechanism of AEF formation is not defined, but direct cooling and necrosis with adherence and fistulous track formation likely play the major role. A second proposed mechanism is injury to the esophageal vasculature with late esophageal necrosis. The observed bimodal distribution of esophageal injury relating to RF ablation would be in keeping with direct injury being responsible for early esophageal ulcerations, whereas the later occurrence may be related to ischemic injury. 18 The role of esophageal temperature monitoring remains controversial because optimal probe position to record lowest esophageal temperature is not always achieved. In the present study, 3 patients had LET recorded. The LET in patient 9 was 20.9 C. Two inflations were performed in the LIPV in this patient. The initial inflation for 180 seconds resulted in LET of 25 C, but there was persisting veno atrial connections in the inferior aspect of the vein. A second inflation for 180 seconds after repositioning the guide circular catheter to a more superior branch of the vein resulted in a balloon temperature of 60 C and vein isolation. Deflation of the balloon was performed when the Figure 3 Locations of cryoballoon-related cases of atrioesophageal fistula compared to radiofrequency (RF) ablation-related atrioesophageal fistula. LA ¼ left atrium; LIPV ¼ left inferior pulmonary vein; LSPV ¼ left superior pulmonary vein; RIPV ¼ right inferior pulmonary vein; RSPV ¼ right superior pulmonary vein. esophageal temperature was 24 C, but a persistent fall to 20.9 C was seen in the minute after deflation (Table 1). Such thermal delay is analogous to that seen with RF ablation and is a feature of the thermal probes used for temperature monitoring. 17 In animal studies, evidence of esophageal damage was seen only below 10 C. In 1 study, 52% of 67 patients who underwent cryoballoon PVI with a first-generation cryoballoon underwent serial endoscopy. No cases of AEF occurred despite recording an LET of 0 C in 1 patient. 19 All esophageal ulcers healed within 3 months. Although no threshold temperature was predictive of esophageal ulceration, the 6 cases of esophageal ulceration had a lower nadir for LET and more LET recordings below 30 C. Studies that included EGD after use of the second-generation cryoballoon suggest an LET cutoff of 10 C 12 C to prevent thermal damage. 17,20,21 In a study using deflective esophageal probes, recorded LET was lower by 6 C in 85% of patients. 17 In the present series, we have documentation of AEF occurring with an LET of 20.9 C using a nondeflectable 18Fr esophageal stethoscope positioned opposite the site of the inflated balloon as determined by orthogonal fluoroscopic views. Hence, a clear cutoff for an LET beyond which thermal damage can be prevented remains in doubt. Our current practice is to deflate the balloon when the esophageal temperature falls to 30 C. The majority of AEFs have been documented with use of the Arctic Front II balloon because of the increased number Figure 2 Approximate locations of atrioesophageal fistulas depicted in the posteroanterior (PA) view of the left atrium. All 10 cases of atrioesophageal fistula with cryoballoon ablation were identified in relation to the left pulmonary veins (blue dots). By comparison, fistulas associated with radiofrequency (RF) ablation were distributed more randomly in the posterior left atrium (red dots). LIPV ¼ left inferior pulmonary vein; LSPV ¼ left superior pulmonary vein; RIPV ¼ right inferior pulmonary vein; RSPV ¼ right superior pulmonary vein. Figure 4 Lowest esophageal temperature (LET) recordings in cases of atrioesophageal fistula with cryoablation.

188 of cases performed with the Arctic Front II and possibly inadequate recognition of the extent of cooling obtained with the Arctic Front II. It is likely that once vein isolation is obtained within the first 40 seconds of application, ablation can be limited to 180 seconds and would not require a repeat inflation. Single-center studies have shown similar efficacy for single vs multiple inflations. 22 It also is possible that the second inflation may increase collateral damage without improving efficacy. However, no randomized studies exist, and the initial pivotal study (Sustained Treatment of Paroxysmal Atrial Fibrillation [STOP AF]) and other randomized studies used multiple inflations to effect PVI. 1,4 The latest iteration of the cryoballoon (Arctic Front Advance Short Tip) allows for better monitoring of vein potentials during cryoablation because the circular catheter through the central lumen can be withdrawn into the proximal vein. A singleshot PVI was more frequently achieved with this iteration compared to the second-generation balloon, reducing the left atrial dwell time without loss of efficacy. These features may improve its safety profile for rare complications such as AEF. 23 The difference in locations of AEF between catheterbased RF ablation and cryoballoon is striking, with a propensity for the latter technique to cause AEF when the esophagus is in close relation to the left veins. Point ablation with RF at the PV antrum tends to create a pouch that can extend to extracardiac structures such as the lungs and esophagus. In contrast, ice formation with a good balloon occlusion extends into the base of the PV antrum and into the PV itself. If the balloon is positioned deep in the vein, ice formation is faster and larger, increasing the risk of collateral damage. Hence, care must be exercised to position the balloon proximally. Deep positioning in the vein is often indicated by a precipitous drop in balloon temperatures to 60 C or lower. In addition, the proximal seal technique of withdrawing the balloon to allow a contrast leak at the balloon vein interface is helpful for visualizing the ostium of the vein. Minimum forward pressure from this point to achieve complete occlusion of the vein can prevent distal positioning. 24 The finding of the susceptibility of the LIPV to cryoballoon-related esophageal damage has been observed in previous case reports and endoscopic studies. 14,17 The LIPV ostium has the closest relationship to the esophagus, especially when the esophagus is buttressed between the left atrium, the descending aorta, and the spine. Preprocedural multidetector computed tomographic scans have shown an average distance between the LIPV ostium and esophagus of 7.2 mm compared to 27 mm for the right inferior PV ostium. 17 The coaxial force of the cryoballoon is directed leftward, superiorly and posteriorly. The left upper PV usually had adequate occlusion and rapid isolation is commonly evident in this vein. The left inferior vein often requires additional manipulations for adequate positioning for occlusion, and any forward thrust tends to direct forces toward the posterior LA. Such posteriorly directed forces may have a more direct impact on the esophagus, potentially Heart Rhythm, Vol 14, No 2, February 2017 causing both direct and ischemic damage. The right PVs, on the other hand, may be less susceptible to esophageal damage because the right inferior PV is difficult to engage fully and very low temperatures are difficult to achieve. The right upper PV runs the risk of phrenic nerve paralysis, particularly with ostial ablations that tend to produce lower temperatures. It is possible that careful monitoring and balloon deflation to avoid nerve injury offer some protection against esophageal damage. However, the low number of patients in this report and the limitations intrinsic to autopsies preclude any definite conclusions regarding the safety of the cryoballoon in the right veins, and the potential for esophageal damage should always be a consideration when the esophagus is located in close proximity to the location of cryoapplication. Study limitations The small number of patients in this series limits the power of our conclusions. However, AEF is a rare and sporadic occurrence, and systematic data collection and follow-up are not available. The present study provides data on the largest collection of this complication with cryoballoon PVI. The comparator group for cryoballoon used in this study is from a single center and may not be applicable to general practice. We followed standard recommendations for use of the cryoballoon, so our data likely closely mirror those of other operators. Similarly, rigorous data collection is not available for RF-related AEFs and our data may be specific to a region because some of the data were derived from surveys of electrophysiologists from the region. Conclusion AEF associated with cryoballoon ablation is rare, with an estimated incidence of o1 in 10,000. To date, all documented cases of cryoballoon-associated AEFs have been in relation to the left PVs, particularly the LIPV. In this limited series, balloon inflation times in the LIPV were longer in the AEF cases compared with non-aef cases. Although the role of LET remains controversial, rapid decline in endoluminal temperature when the esophagus is located in immediate proximity to the LIPV should prompt limitation of cryoenergy delivery at this vein. References 1. Packer DL, Kowal RC, Wheelan KR, Irwin JM, Champagne J, Guerra PG, Dubuc M, Reddy V, Nelson L, Holcomb RG, Lehmann JW, Ruskin JN. Cryoballoon ablation of pulmonary veins for paroxysmal atrial fibrillation: first results of the North American Arctic Front (STOP AF) pivotal trial. J Am Coll Cardiol 2013;61:1713 1723. 2. Vogt J, Heintze J, Gutleben KJ, Muntean B, Horstkotte D, Nölker G. Long-term outcomes after cryoballoon pulmonary vein isolation: results from a prospective study in 605 patients. J Am Coll Cardiol 2013;61:1707 1712. 3. Neumann T, Vogt J, Schumacher B, et al. Circumferential pulmonary vein isolation with the cryoballoon technique: results from a prospective 3-center study. J Am Coll Cardiol 2008;52:273 278. 4. Kuck KH, Brugada J, Furnkranz A, et al. Cryoballoon or radiofrequency ablation for paroxysmal atrial fibrillation. N Engl J Med 2016;374:2235 2245. 5. Luik A, Radzewitz A, Kieser M, et al. Cryoballoon versus open irrigated radiofrequency ablation in patients with paroxysmal atrial fibrillation: the

John et al Atrioesophageal Fistula and Cryoballoon Ablation 189 prospective randomized, controlled, non-inferiority Freeze AF Study. Circulation 2015;132:1311 1319. 6. DE Regibus V, Mugnal G, Moran D, Hunuk B, Stroker E, Hacioglu E, Ruggiero D, Cotino-Moreno HE, Takarada K, Brugada P, DE Asmundis C, Chiercha GB. Secondgeneration cryoballoon ablation in the setting of lone paroxysmal atrial fibrillation: single procedural outcome at 12 months. J Cardiovasc Electrophysiol 2016;6:677 682. 7. Okunura K, Matsumoto K, Kobayashi Y, Nogami A, Hokanson RB, Kueffer RB, Kueffer F, CRYO-Japan PMS Study Investigators. Safety and efficacy of cryoballoon ablation for paroxysmal atrial fibrillation in Japan: results from the Japanese prospective post-market surveillance study. Circ J 2016;8:1744 1749. 8. Tscholl V, Lsharaf AK, Lin T, Bellmann B, Biewener S, Nagel P, Suhail S, Lenz K, Landmesser U, Roser M, Rillig A. Two year outcome in patients with persistent atrial fibrillation after pulmonary vein isolation using the secondgeneration 28 mm cryoballoon. Heart Rhythm 2016;9:1817 1822. 9. Knight BP, Novak PG, Sangrigoli R, Essebag V, Champagne J, Adler SW, Dubuc M, Jain SK, Svinarich T, John RM, Mansour M. 12-month clinical outcomes following pulmonary vein isolation using the Arctic Front Advance second generation cryoballoon: results from the multicenter STOP-AF Post Approval Study. Heart Rhythm 2015;12:S120. 10. Cappato R, Calkins H, Chen SA, Davies W, Iesaka Y, Kalman J, Kim Y, Klein G, Natale A, Packer D, Skanes A, Prevalence and causes of fatal outcome in catheter ablation of atrial fibrillation. J Am Coll Cardiol 2009;53:1798 1803. 11. Stockigt F, Schrickel JW, Andrie R, Lickfelt L. Atrioesophageal fistula after cryoballoon pulmonary vein isolation. J Cardiovascular Electrophysiol 2012;23:1254 1257. 12. Lim HW, Cogert GA, Cameron CS, Cheng VY, Sandler DA. Atrioesophageal fistula during cryoballoon ablation for atrial fibrillation. J Cardiovasc Electrophysiol 2014;25:208 213. 13. Medel DV, Marti-ALmor J, Serrano JM, Sionis A, Petracca RL. Atrioesophageal fistula secondary to pulmonary vein cryoablation. Eur Heart J Cardiovasc Imaging 2014;15:116. 14. Kawasaki R, Gauri A, Eimouchi D, Duggal M, Bhan A. Atrioesophageal fistula complicating cryoballoon pulmonary vein isolation for paroxysmal atrial fibrillation. J Cardiovasc Electrophysiol 2014;25:784 782 15. van Opstal JM, Timmermans C, Blaauw Y, Pison L. Bronchial erosion and hemoptysis after pulmonary vein isolation by cryoballoon ablation. Heart Rhythm 2011;8:1459. 16. Verma N, Gillespie CT, Lin AC, Knight BP. Ice formation in the left mainstem bronchus during cryoballoon ablation for the treatment of atrial fibrillation. Heart Rhythm 2016;13:814 815. 17. Miyazaki S, Nakamura H, Taniguchi H, Takagi T, Iwasawa J, Watanabe T, Hachiya H, Kirao K, Iesaka Y. Esophagus-related complications during secondgeneration cryoballoon ablation: insight from simultaneous esophageal temperature monitoring from 2 esophageal probes. J Cardiovasc Electrophysiol 2016;27: 1038 1044. 18. Barbahiya CR, Kumar S, Guo Y, Zhong H, John RM, Tedrow UB, Koplan BA, Epstein LM, Stevenson WG, Michaud GF. Global survey of esophageal injury in atrial fibrillation ablation: characteristics and outcomes of esophageal perforation and fistula. J Am Coll Cardiol Electrophysiol 2016;2:143 150. 19. Ahmed H, Neuzil P, d Avila A, Cha YM, Laragy M, Mares K, Brugge WR, Forcione DG, Ruskin JN, Packer DL, Reddy VY. The esophageal effects of cryoenergy during cryoablation for atrial fibrillation. Heart Rhythm 2009;6:962 969. 20. Metzner A, Vurchard P, Wohlmuth P, et al. Increase incidence of esophageal thermal lesions using the second generation 28 mm cryoballoon. Circ Arrhythm Electrophysiol 2013;6:769 775. 21. Fumkranz A, Bordignon S, Schmidt M, Bohmig M, Bohmer MC, Bode F, Schulte-Hhn B, Nowak B, Dignab AU, Chun JK. Luminal esophageal temperature after second generation cryoballoon pulmonary vein isolation. Heart Rhythm 2013;10:789 793. 22. Miyazaki S, Hachiya H, Nakamura H, Taniguchi H, Takagi T, Hirao K, Iesaka Y. Pulmonary vein isolation using a second generation cryoballoon in patients with paroxysmal atrial fibrillation: one year outcome using a single big-balloon 3-minute freeze technique. J Cardiovasc Electrophysiol. 2016 Aug 18. http://dx. doi.org/10.1111/jce.13078. 23. Aryana A, Kowalski M, O Neill PG, Koo CH, Lim HW, Khan A, Hokanson RB, Bowers MR, Kenigsberg DN, Ellenbogen KA, Cryo-DOSING Investigators. Catheter ablation using the third-generation cryoballoon provides an enhanced ability to assess time to pulmonary vein isolation facilitating the ablation strategy: short and long-term results of a multicenter study. Heart Rhythm 2016;13: 2306 2313. 24. Su W, Kowal R, Kowalski M, Metzner A, Svinarich JT, Wheelan K, Wang P. Best practice guide for cryoballoon ablation in atrial fibrillation. The compilation experience of more than 3000 procedures. Heart Rhythm 2015;12:1658 1666.