1057
Novel Technique to Prevent Phrenic Nerve Injury During Pulmonary Vein Isolation Using Preprocedural Imaging ATTILA ROKA, M.D., Ph.D., E. KEVIN HEIST, M.D., Ph.D., MARWAN REFAAT, M.D., JEREMY RUSKIN, M.D., and MOUSSA MANSOUR, M.D. From the Heart Center, Massachusetts General Hospital, Boston, Massachusetts, USA
Prevention of Phrenic Injury Using Imaging. Introduction: Phrenic nerve (PN) injury is one of the major complications of pulmonary vein isolation (PVI). Pace mapping for PN capture is routinely used to identify areas with high risk for injury along the anterior border of the right pulmonary veins (PVs). Our aim was to evaluate the feasibility of using preprocedural imaging to identify areas where no PN capture is possible along the anterior border of the right PVs, thus avoiding the need for pace mapping during PVI. Methods and Results: It was hypothesized that PN capture along the anterior border of the right PVs does not occur in the area where the right and left atria overlap. Three-dimensional segmentation of both atria was performed on preprocedural magnetic resonance and computed tomography angiograms in 40 patients before undergoing PVI. The area of overlap between the right and left atria was delineated. Image registration was performed during the procedure. Using pacing, regions with and without PN capture were marked along the anterior border of the right PVs. A total of 361 points were tested for PN stimulation (9 ± 4 points/patient). PN capture occurred in 97 out of the 189 points (51%) in the area with no overlap between the right and left atria. No PN capture occurred in the area of overlap (172 points, P< 0.001). Conclusion: Delineation of the area of overlap between the right and left atria derived from preprocedural imaging reliably identifies regions where PN pace capture does not occur. Testing for PN stimulation before ablation may not be necessary in these regions. (J Cardiovasc Electrophysiol, Vol. 26, pp. 1057-1062, October 2015) atrial fibrillation, electroanatomical mapping, phrenic nerve, pulmonary vein isolation, radiofrequency ablation Introduction Injury to the phrenic nerve (PN) is one of the major complications of pulmonary vein isolation (PVI). The incidence ranges between 0.48% and 11%, depending on the type of ablation and energy source used.1-3 It is more likely to occur with cryoballoon PVI than with radiofrequency (RF) energy. Recent studies demonstrated that PN injury can occur even when wide area circumferential ablation (WACA) approach is used.1 One possible explanation is the lack of an objective definition of WACA. The location of the ablation lesions anterior to the right pulmonary veins (PVs) may vary among different operators. High output pacing in the region targeted for ablation allows the identification of the location of the PN. Avoiding ablation in these regions can prevent PN injury.1,4 Although this approach may decrease
the risk of PN injury during RF ablation, pacing in multiple locations can result in prolongation of the procedural time and interruption of ablation if a dragging ablation approach is used. The PN runs in a craniocaudal direction as part of the pericardiacophrenic bundle, anterior to the superior vena cava (SVC), right PVs, and medial aspect of the left atrium (LA). Along its course, it gets separated from the LA by the right atrium (RA). In this study, we aimed to investigate whether atrial anatomy reconstructed from preprocedural imaging could be used to identify the regions where the PN is protected by the RA, and thus unlikely to be injured if ablation is performed there during PVI, thus eliminating the need for pacing in these regions. Methods
This work was supported by the Deane Institute for Integrative Research in Atrial Fibrillation and Stroke at Massachusetts General Hospital (Boston, MA). E. Kevin Heist reports research grants from St. Jude Medical, and serving as a consultant to Biosense Webster and St. Jude Medical. Jeremy Ruskin reports Fellowship support from St. Jude Medical, and serving as a consultant to Biosense Webster. Moussa Mansour reports research grants from Biosense Webster and St. Jude Medical, and serving as a consultant to Biosense Webster and St. Jude Medical. Other authors: No disclosures. Address for correspondence: Moussa Mansour, M.D., Heart Center, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA. Fax: +1-617-726-3852; E-mail: [email protected] Manuscript received 22 April 2015; Revised manuscript received 19 June 2015; Accepted for publication 23 June 2015. doi: 10.1111/jce.12758
Patients and Imaging Techniques In 40 patients undergoing PVI, magnetic resonance angiogram (MRA) and computed tomography angiogram (CTA) were obtained before the procedure. The MRI protocol consisted of multiple localizing images, followed by 0.2 mmol/kg gadolinium-enhanced end expiration breathhold three-dimensional (3D) MRA of pulmonary vessels, using timing bolus method. The images were imported into the electroanatomical mapping system. Threedimensional segmentation of the RA and LA was performed, and areas of overlap between the RA and LA were delineated on the imported CT or MRI images before the start of the ablation procedure using the mapping system working station. This was performed by a technologist at a time that does not interfere with the flow
1058
Journal of Cardiovascular Electrophysiology
Vol. 26, No. 10, October 2015
Figure 1. Pacing for phrenic nerve mapping inside and outside the area of overlap (white line) of the right and left atria in a right anterior oblique (RAO) view. A: Location of pacing points resulting in phrenic nerve capture. Sites with phrenic nerve capture are shown in blue dots. B: The right atrium is rendered transparent and sites with phrenic nerve capture are shown in blue dots. Sites with no phrenic nerve capture are shown in white dots. LA = left atrium; RA = right atrium. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
of the procedure, usually during the transseptal puncture. Shallow RAO, with slight caudal angulation, was the projection that allowed visualization and evaluation the RA borderLA surface and its relationship to the ostia of the right-sided PVs. The imaging data were merged with mapping data obtained during the ablation procedure. Areas with and without PN capture during pace mapping were marked along the anterior border of the right PVs. Ablation Procedure The ablation procedures were performed under general anesthesia and high-frequency jet ventilation in all patients. Double transseptal access was obtained with fluoroscopic and intracardiac echo guidance. A circular mapping catheter was used for left atrial and PV mapping (Lasso, BiosenseWebster, Diamond Bar, CA, or A-Focus, St. Jude Medical, Minneapolis, MN, USA). A 3.5 mm externally irrigated tip catheter (Thermocool SF, Biosense Webster; Themocool SMARTTOUCH Biosense Webster; Tacticath Quartz, St. Jude Medical) was used for ablation and testing for PN stimulation. When contact force sensing was used, the target force range was 10–40 x g for both mapping and ablation. Wide area PVI was performed in all patients. Patients with persistent AF underwent additional ablation.
Figure 2. Measuring the distance between the right superior pulmonary veins ostium and the line of RA-anterior LA overlap. Black dots = pacing sites with phrenic nerve capture; LA = left atrium; RA = right atrium; red and pink dots = ablation lesions; white dots = pacing sites with no phrenic nerve capture. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
Measurements Testing for PN Stimulation Testing for PN capture was performed with high energy stimulation (20 mA at 2 ms pulse duration),1 with the ablation catheter along the anterior border of the right PVs (Fig. 1A,B). The muscle paralysis was weaned off during isolation of the right PVs and confirmed by thumb adduction in response to train-of-four stimulation of the ulnar nerve. Detection of diaphragmatic stimulation was verified by fluoroscopy and direct observation of the abdomen. Pacing points were tagged on the electroanatomical maps with different colors depending on whether or not PN capture occurred. Ablation was avoided in areas where PN capture was present. Pacing from the SVC was also performed at the end of the procedure to check for PN function.
Offline measurements were performed on the electroanatomical maps using the imported CTA and MRA images. The ostia of the right superior and inferior PVs were identified and delineated as previously described.5 The area of overlap between the RA and LA was also marked. The shortest distances between the ostia of the right superior and inferior PVs and the line of overlap between the RA and LA were measured (Fig. 2). Statistical Analysis Paired comparison of categorical variables was performed with the χ 2 test. Logistic regression analysis was performed to assess the relationship of PN capture with clinical and anatomical data. Significance level was determined at
Roka et al. Prevention of Phrenic Injury Using Imaging
1059
Figure 3. Location of ablation lesions. Ablation lesions at the anterior antrum of the right pulmonary veins deployed proximal to the area of overlap (white line) between the right atrium and left atrium. A: A patient with a wide distance between the ostia of the right pulmonary veins and the area of the overlap between the right and left atria. B: A patient with a narrow distance between the ostia of the right pulmonary veins and the area of the overlap between the right and left atria. The line of overlap crosses the ostium of the right superior pulmonary vein. LA = left atrium; RA = right atrium; red and pink dots = ablation lesions. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
P < 0.05. IBM SPSS 20 software (IBM Corp., Armonk, NY, USA) was used for statistical computations. Results
TABLE 1 Distance Between the Line Demarcating the Area of Overlap (RA–LA Border) Along the Anterior Wall of the LA and the Right Pulmonary Vein Ostia
Number of RPVs
n
Distance of Line from RSPV (mm)
1 (right common PV) 2 (RSPV, RIPV) 3 (RSPV, RMPV, RIPV)
1 31 8
1 (from right common PV) 2±3 3±3
Patient Characteristics Forty patients were studied. Seventy-five percent were males, and the average age was 65.0 ± 8.6 years. The left atrial size was 45 ± 7 mm, the BMI 27.6 ± 4.7, and the BSA 2.00 ± 0.36. The indication for PVI was symptomatic drugrefractory paroxysmal AF in 21 (52.5%) and persistent AF in 19 (47.5%). Preprocedural CTA and MRA were obtained in 37 (92.5%) and 3 (7.5%) patients, respectively. The median number of right PV ostia was 2 (range 1–3). The median number of left PV ostia was 2 (range 1–2). The CARTO electroanatomical mapping system (Biosense-Webster) was used in 26 patients (65%) and the NavX system (St. Jude Medical) in 14 patients (35%). Contact force sensing was used in 15 out of 40 (38%) patients. All patients underwent PVI during the ablation procedure. In addition to the PVI, 50% of the patients had posterior wall ablation, 15% focal trigger ablation, 10% mitral isthmus line, and 35% cavotricuspid isthmus ablation. Location of the Area of Overlap The line demarcating the area of overlap (RA–LA border) ran along the anterior wall of the LA. This line was proximal to the ostium of the right superior PV in 23/40 (57.5%). In the remaining 17 (42.5%) patients, the line crossed the ostium of this vein (Fig. 3A,B). In all patients, this line of overlap was proximal to the ostium of the right inferior PV. When the line did not cross the ostia of the PVs, the distance separating it from the right superior PV and the right inferior PV ostia was 4 ± 3 and 13 ± 6 mm, respectively (Fig. 2, Table 1). The distance between the line and both ostia was more than 5 mm in 6 patients (15%), and less than or equal to 5 mm in 4 patients (10%). The distance was more than 5 mm only from the inferior PV ostia in 30 patients (75%). In no patient, the distance was more than 5 mm from the right superior, but less than or equal to 5 mm from the right inferior PV ostia.
Distance of Line from RIPV (mm) – 12 ± 6 17 ± 5
The distance was greater from the RIPV, than the RSPV. In patients where the line crossed the ostium, the distance was determined as 0 mm when calculating the means for each group. RIPV = right inferior pulmonary vein; RMPV = right middle pulmonary vein; RSPV = right superior pulmonary vein.
Incidence and Factors Affecting PN Capture A total of 361 points in the LA (9 ± 4/patient) were tested for PN stimulation. Of these, 172 were located in the area of RA and LA overlap, and 189 were not. PN capture occurred in 27% of all points. Sixty-five percent of the patients had at least 1 site where PN capture was present. PN capture occurred in none of the points with RA and LA overlap, but in 97 out of the 189 (51%) points in the area of no overlap (P < 0.001). There was no significant association between any of the clinical parameters and the incidence of PN capture (Table 2). Procedural Outcomes The total procedure duration including general anesthesia induction and extubation was 241±60 minutes. The fluoroscopy time was and 22 ± 6 minutes. PVI was successfully performed with large circumferential antral ablation. The PVs were encircled in ipselateral pairs in all patients. Isolation of the right PVs was achieved with ablation lesions proximal to the line of overlap between the LA and RA in 30 patients (75%, Fig. 3A,B). In 10 patients (25%), ablation lesions needed to be delivered distal to the line of overlap in order to achieve isolation (Fig. 4).
1060
Journal of Cardiovascular Electrophysiology
Vol. 26, No. 10, October 2015
TABLE 2 Correlation Between Clinical and Anatomical Characteristics and Phrenic Nerve Capture
Parameter
Value
Male: 30 (75%) Female: 10 (25%) Age 65.0 ± 8.6 years LA size 45 ± 7 mm Height 178 ± 10 cm Weight 87.7 ± 20.1 kg Body mass index 27.6 ± 4.7 kg/m2 Body surface area 2.00 ± 0.36 m2 Paroxysmal: 21 (52.5%) Atrial fibrillation Persistent: 19 (47.5%) Prior cardiac surgery 2 (5%) Number of right PV ostia Median 2 (range 1–3) Number of left PV ostia Median 2 (range 1–2) RA–LA border 17 (42.5%) overlapping with RSPV ostium Gender
Correlation with Presence of Phrenic Nerve Capture (Univariate Logistic Regression) P = 0.251 P = 0.359 P = 0.175 P = 0.132 P = 0.191 P = 0.671 P = 0.121 P = 0.273 P = 0.287 P = 0.276 P = 0.507 P = 0.169
LA = left atrium; RA = right atrium; PV = pulmonary vein; RSPV = right superior pulmonary vein.
Figure 4. Ablation lesions distal to the border of the overlap (white line) between the right and left atria. Ablation lesions (black arrow) distal to the line were required in this patient to achieve isolation of the rightsided pulmonary veins. High output pacing (20 mA at 2 ms pulse width) was utilized to test for phrenic nerve capture in this region, ablation was avoided in areas where phrenic nerve capture was present (blue dots). Blue dots = pacing sites with phrenic nerve capture; LA = left atrium; RA = right atrium; red and pink dots = ablation lesions; white dots = pacing sites with no phrenic nerve capture. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
One patient had a persistent left-sided SVC, which was also isolated during the PVI. No patient in this study sustained PN injury as confirmed by pacing in the SVC at the end of the procedure. No other periprocedural complications occurred either. Discussion This study resulted in the following 3 important findings: first, PN capture with pacing in the antral region of the right-sided PVs does not occur in the area of overlap of the right and left atria; second, isolation of the right PVs can be achieved in the majority of the patients by limiting the anterior ablation lesions to the region proximal to the line demarcating the area of overlap; and third, there were no clinical or anatomical characteristics associated with higher incidence of PN capture with pacing. Incidence and Mechanism of Injury of the PN The location of the PN relative to the atria has been well described in anatomical studies.6 The PN runs in a craniocaudal direction as part of the pericardiacophrenic bundle, in close proximity of the superior vena cava, right PVs, and medial aspect of the LA. Injury to the PN during catheter ablation results from direct thermal injury. Experimental data show that the damage is axonal in nature and is characterized by Wallerian degeneration, with potential for recovery.7 This correlates well with clinical observations of cases of PN injuries resulting in transient diaphragmatic dysfunction. Damage to the PN has been reported with various energy modalities: RF, ultrasound, laser, and, most commonly, cryoablation.1-3,8,9 In the STOP AF study, the incidence of PN injury during cryoablation was 11.2%.10 This complication continues to occur despite preventive techniques including continuous PN pacing during cryoenergy delivery, and termination of freezing cycle in case of weakening or loss of diaphragmatic contraction. In a recent single-center study of 115 patients using cryoballoon for PVI, PN palsy occurred in 3.5%.11 The use of electromyographic monitoring and terminating the cryo lesion delivery with the reduction in diaphragmatic motion was found to reduce the risk of PN palsy.12 PN injury also occurs with RF ablation, although at a lower frequency. A large observational study of 3,755 patients reported an incidence of 0.5% of PN injury with RF ablation.3 Even when a WACA approach is used, this complication continues to occur.1 One possible explanation of the occurrence of PN injury despite wide ablation is the lack of a clear definition of a safe zone of ablation with the WACA approach. Prevention of Nerve Injury with Pace Mapping Various techniques have been proposed to avoid PN injury. High output pacing to delineate the course of the PN and avoiding the areas of capture is the most commonly used technique.1,4 This technique was investigated in a prospective study of 100 patients who underwent RF PVI using a WACA approach.1 PN capture occurred in 30% of patients leading to the modification of the ablation lines in order to avoid the areas of capture.1 Similar to our study, pacing to identify the location of the PN was performed at 20 mA and 2 ms.1 It
Roka et al. Prevention of Phrenic Injury Using Imaging
is conceivable that pacing at higher output may have led to phrenic capture in more areas. Another technique is PN pacing at a location cranial to the ablation site. With this technique, pacing is usually performed from the SVC to monitor the right PN, and from the left subclavian vein to monitor the left PN. Diaphragmatic contractions are monitored with fluoroscopy, palpation, or intracardiac echo, and ablation is stopped when diaphragmatic contractions are weakened or lost. All the studies mentioned above and others demonstrated that pace mapping may help reduce PN injury during ablation. However, pacing for PN capture adds time and complexity to the ablation procedure. A study of 18 patients demonstrated that pace mapping of the PN required 14 ± 6 minutes.13 This is a significant amount of time to add to an already long procedure. Role of Preprocedural Imaging Preprocedural CT imaging has been shown to allow the localization of the PN. The pericardiacophrenic bundle, including the nerve and the accompanying vessels, can be visualized from reconstructed 3D images.14,15 Using this method, the average distance between the phrenic bundle and the site of PN capture was found to be 8.7 ± 5.8 mm. However, direct reconstruction of the course of the bundle by CT can be difficult to achieve, requires a high level of expertise, is time-consuming and more complicated in comparison to the method described in our study. The findings in our study suggest that a simple maneuver delineating the border between the RA and the LA can define a safe area of ablation and avoid the need for pace mapping. This delineation process can be performed in a short time, and it neither interferes with the flow of the ablation, nor does it result in the prolongation of the procedural time. This technique depends on accurate image integration as the CT and MRI images are obtained before the ablation procedure. Some changes in anatomy due to volume shifts or anesthesia/jet ventilation can occasionally occur, and caution should be exercised during the procedure to detect signs of misalignment of the integrated image. Despite this limitation, the technique was 100% accurate in identifying areas where PN capture was not possible. The line demarcating the area of overlap of the RA and LA is closer to the ostium of the right superior PV than it is to the ostium of the right inferior PV. In some patients, in our study this distance was found to be as long as 29 mm. Is such a situation, isolating the right PV can be challenging if ablations were to be limited to the wide area proximal to the line. As a result, ablation may need to be performed in some patients distal to the line or over the carina in order to achieve isolation. Only in this situation pacing for PN mapping would be necessary. Conclusion PN injury is one of the major complications of PVI. The WACA approach reduces but does not eliminate the risk of injury. One limitation of the WACA approach is the lack of objective definition of wide ablation. This study demonstrates that a safe zone for wide ablation can be easily indentified and clearly defined using routine preprocedural imaging data from CTA or MRA. Ablation in this area can be performed without the need for pacing to delineate the PN. Pacing can
1061
be reserved only to situations necessitating ablation distal to the area of overlap. The adoption of this technique provides an objective guide for ablation during WACA and may result in shortening and simplification of the ablation procedure. Limitations One limitation of this study is that the application of the technique is limited to ablation catheters that can be visualized on the 3D electroanatomical maps. Using this technique for current generation balloon-based catheters requires imaging tools that allow the integration of electroanatomical maps with X-ray imaging, which is not readily available in all electrophysiology laboratories. Another limitation is the unblinded design of this study. Pacing for the PN was performed after the demarcation of the line of overlap of the RA and LA. However, the difference in the rate of capture of the PN between the areas proximal and distal to the line of overlap was very large, and is unlikely to become less significant if a blinded design is used.
References 1. Yong Ji S, Dewire J, Barcelon B, Philips B, Catanzaro J, Nazarian S, Cheng A, Spragg D, Tandri H, Bansal S, Ashikaga H, Rickard J, Kolandaivelu A, Sinha S, Marine JE, Calkins H, Berger R: Phrenic nerve injury: An underrecognized and potentially preventable complication of pulmonary vein isolation using a wide-area circumferential ablation approach. J Cardiovasc Electrophysiol 2013;24:10861091. 2. Calkins H, Kuck KH, Cappato R, Brugada J, Camm AJ, Chen SA, Crijns HJ, Damiano RJ Jr, Davies DW, DiMarco J, Edgerton J, Ellenbogen K, Ezekowitz MD, Haines DE, Haissaguerre M, Hindricks G, Iesaka Y, Jackman W, Jalife J, Jais P, Kalman J, Keane D, Kim YH, Kirchhof P, Klein G, Kottkamp H, Kumagai K, Lindsay BD, Mansour M, Marchlinski FE, McCarthy PM, Mont JL, Morady F, Nademanee K, Nakagawa H, Natale A, Nattel S, Packer DL, Pappone C, Prystowsky E, Raviele A, Reddy V, Ruskin JN, Shemin RJ, Tsao HM, Wilber D: 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: Recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design: A report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial Fibrillation. Developed in partnership with the European Heart Rhythm Association (EHRA), a registered branch of the European Society of Cardiology (ESC) and the European Cardiac Arrhythmia Society (ECAS); and in collaboration with the American College of Cardiology (ACC), American Heart Association (AHA), the Asia Pacific Heart Rhythm Society (APHRS), and the Society of Thoracic Surgeons (STS).Endorsed by the governing bodies of the American College of Cardiology Foundation, the American Heart Association, the European Cardiac Arrhythmia Society, the European Heart Rhythm Association, the Society of Thoracic Surgeons, the Asia Pacific Heart Rhythm Society, and the Heart Rhythm Society. Heart Rhythm 2012;9:632-696. 3. Sacher F, Monahan KH, Thomas SP, Davidson N, Adragao P, Sanders P, Hocini M, Takahashi Y, Rotter M, Rostock T, Hsu LF, Clementy J, Haissaguerre M, Ross DL, Packer DL, Jais P: Phrenic nerve injury after atrial fibrillation catheter ablation: Characterization and outcome in a multicenter study. J Am Coll Cardiol 2006;47:2498-2503. 4. Fan R, Cano O, Ho SY, Bala R, Callans DJ, Dixit S, Garcia F, Gerstenfeld EP, Hutchinson M, Lin D, Riley M, Marchlinski FE: Characterization of the phrenic nerve course within the epicardial substrate of patients with nonischemic cardiomyopathy and ventricular tachycardia. Heart Rhythm 2009;6:59-64. 5. Mansour M, Refaat M, Heist EK, Mela T, Cury R, Holmvang G, Ruskin JN: Three-dimensional anatomy of the left atrium by magnetic resonance angiography: Implications for catheter ablation for atrial fibrillation. J CardiovascElectrophysiol 2006;17:719-723. 6. Lachman N, Syed FF, Habib A, Kapa S, Bisco SE, Venkatachalam KL, Asirvatham SJ: Correlative anatomy for the electrophysiologist, part II:
1062
7.
8.
9. 10.
11.
Journal of Cardiovascular Electrophysiology
Vol. 26, No. 10, October 2015
Cardiac ganglia, phrenic nerve, coronary venous system. J Cardiovasc Electrophysiol 2011;22:104-110. Andrade JG, Dubuc M, Ferreira J, Guerra PG, Landry E, Coulombe N, Rivard L, Macle L, Thibault B, Talajic M, Roy D, Khairy P: Histopathology of cryoballoon ablation-induced phrenic nerve injury. J Cardiovasc Electrophysiol 2014;25:187-194. Bunch TJ, Bruce GK, Mahapatra S, Johnson SB, Miller DV, Sarabanda AV, Milton MA, Packer DL: Mechanisms of phrenic nerve injury during radiofrequency ablation at the pulmonary vein orifice. J Cardiovasc Electrophysiol 2005;16:1318-1325. Andrie RP, Schrickel JW, Nickenig G, Lickfett L: Left phrenic nerve injury during cryoballoon ablation of the left superior pulmonary vein. Pacing Clin Electrophysiol 2012;35:e334-e336. Packer DL, Kowal RC, Wheelan KR, Irwin JM, Champagne J, Guerra PG, Dubuc M, Reddy V, Nelson L, Holcomb RG, Lehmann JW, Ruskin JN, Investigators SAC: 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;61:17131723. Metzner A, Rausch P, Lemes C, Reissmann B, Bardyszewski A, Tilz R, Rillig A, Mathew S, Deiss S, Kamioka M, Toennis T, Lin T, Ouyang F, Kuck KH, Wissner E: The incidence of phrenic nerve injury during
12.
13.
14.
15.
pulmonary vein isolation using the second-generation 28 mm cryoballoon. J Cardiovasc Electrophysiol 2014;25:466-470. Mondesert B, Andrade JG, Khairy P, Guerra PG, Dyrda K, Macle L, Rivard L, Thibault B, Talajic M, Roy D, Dubuc M, Shohoudi A: Clinical experience with a novel electromyographic approach to preventing phrenic nerve injury during cryoballoon ablation in atrial fibrillation. Circ Arrhythm Electrophysiol 2014;7:605-611. Schmidt B, Chun KR, Ouyang F, Metzner A, Antz M, Kuck KH: Three-dimensional reconstruction of the anatomic course of the right phrenic nerve in humans by pace mapping. Heart Rhythm 2008;5:11201126. Fukumoto K, Takatsuki S, Jinzaki M, Yamada M, Tanimoto K, Nishiyama N, Aizawa Y, Hagiwara Y, Fukuda Y, Kimura T, Miyoshi S, Kuribayashi S, Fukuda K: Three-dimensional imaging and mapping of the right and left phrenic nerves: Relevance to interventional cardiovascular therapy. Europace 2013;15:937-943. Horton R, Di Biase L, Reddy V, Neuzil P, Mohanty P, Sanchez J, Nguyen T, Mohanty S, Gallinghouse GJ, Bailey SM, Zagrodzky JD, Burkhardt JD, Natale A: Locating the right phrenic nerve by imaging the right pericardiophrenic artery with computerized tomographic angiography: Implications for balloon-based procedures. Heart Rhythm 2010;7:937-941.
Novel Technique to Prevent Phrenic Nerve Injury During Pulmonary Vein Isolation Using Preprocedural Imaging ATTILA ROKA, M.D., Ph.D., E. KEVIN HEIST, M.D., Ph.D., MARWAN REFAAT, M.D., JEREMY RUSKIN, M.D., and MOUSSA MANSOUR, M.D. From the Heart Center, Massachusetts General Hospital, Boston, Massachusetts, USA
Prevention of Phrenic Injury Using Imaging. Introduction: Phrenic nerve (PN) injury is one of the major complications of pulmonary vein isolation (PVI). Pace mapping for PN capture is routinely used to identify areas with high risk for injury along the anterior border of the right pulmonary veins (PVs). Our aim was to evaluate the feasibility of using preprocedural imaging to identify areas where no PN capture is possible along the anterior border of the right PVs, thus avoiding the need for pace mapping during PVI. Methods and Results: It was hypothesized that PN capture along the anterior border of the right PVs does not occur in the area where the right and left atria overlap. Three-dimensional segmentation of both atria was performed on preprocedural magnetic resonance and computed tomography angiograms in 40 patients before undergoing PVI. The area of overlap between the right and left atria was delineated. Image registration was performed during the procedure. Using pacing, regions with and without PN capture were marked along the anterior border of the right PVs. A total of 361 points were tested for PN stimulation (9 ± 4 points/patient). PN capture occurred in 97 out of the 189 points (51%) in the area with no overlap between the right and left atria. No PN capture occurred in the area of overlap (172 points, P< 0.001). Conclusion: Delineation of the area of overlap between the right and left atria derived from preprocedural imaging reliably identifies regions where PN pace capture does not occur. Testing for PN stimulation before ablation may not be necessary in these regions. (J Cardiovasc Electrophysiol, Vol. 26, pp. 1057-1062, October 2015) atrial fibrillation, electroanatomical mapping, phrenic nerve, pulmonary vein isolation, radiofrequency ablation Introduction Injury to the phrenic nerve (PN) is one of the major complications of pulmonary vein isolation (PVI). The incidence ranges between 0.48% and 11%, depending on the type of ablation and energy source used.1-3 It is more likely to occur with cryoballoon PVI than with radiofrequency (RF) energy. Recent studies demonstrated that PN injury can occur even when wide area circumferential ablation (WACA) approach is used.1 One possible explanation is the lack of an objective definition of WACA. The location of the ablation lesions anterior to the right pulmonary veins (PVs) may vary among different operators. High output pacing in the region targeted for ablation allows the identification of the location of the PN. Avoiding ablation in these regions can prevent PN injury.1,4 Although this approach may decrease
the risk of PN injury during RF ablation, pacing in multiple locations can result in prolongation of the procedural time and interruption of ablation if a dragging ablation approach is used. The PN runs in a craniocaudal direction as part of the pericardiacophrenic bundle, anterior to the superior vena cava (SVC), right PVs, and medial aspect of the left atrium (LA). Along its course, it gets separated from the LA by the right atrium (RA). In this study, we aimed to investigate whether atrial anatomy reconstructed from preprocedural imaging could be used to identify the regions where the PN is protected by the RA, and thus unlikely to be injured if ablation is performed there during PVI, thus eliminating the need for pacing in these regions. Methods
This work was supported by the Deane Institute for Integrative Research in Atrial Fibrillation and Stroke at Massachusetts General Hospital (Boston, MA). E. Kevin Heist reports research grants from St. Jude Medical, and serving as a consultant to Biosense Webster and St. Jude Medical. Jeremy Ruskin reports Fellowship support from St. Jude Medical, and serving as a consultant to Biosense Webster. Moussa Mansour reports research grants from Biosense Webster and St. Jude Medical, and serving as a consultant to Biosense Webster and St. Jude Medical. Other authors: No disclosures. Address for correspondence: Moussa Mansour, M.D., Heart Center, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA. Fax: +1-617-726-3852; E-mail: [email protected] Manuscript received 22 April 2015; Revised manuscript received 19 June 2015; Accepted for publication 23 June 2015. doi: 10.1111/jce.12758
Patients and Imaging Techniques In 40 patients undergoing PVI, magnetic resonance angiogram (MRA) and computed tomography angiogram (CTA) were obtained before the procedure. The MRI protocol consisted of multiple localizing images, followed by 0.2 mmol/kg gadolinium-enhanced end expiration breathhold three-dimensional (3D) MRA of pulmonary vessels, using timing bolus method. The images were imported into the electroanatomical mapping system. Threedimensional segmentation of the RA and LA was performed, and areas of overlap between the RA and LA were delineated on the imported CT or MRI images before the start of the ablation procedure using the mapping system working station. This was performed by a technologist at a time that does not interfere with the flow
1058
Journal of Cardiovascular Electrophysiology
Vol. 26, No. 10, October 2015
Figure 1. Pacing for phrenic nerve mapping inside and outside the area of overlap (white line) of the right and left atria in a right anterior oblique (RAO) view. A: Location of pacing points resulting in phrenic nerve capture. Sites with phrenic nerve capture are shown in blue dots. B: The right atrium is rendered transparent and sites with phrenic nerve capture are shown in blue dots. Sites with no phrenic nerve capture are shown in white dots. LA = left atrium; RA = right atrium. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
of the procedure, usually during the transseptal puncture. Shallow RAO, with slight caudal angulation, was the projection that allowed visualization and evaluation the RA borderLA surface and its relationship to the ostia of the right-sided PVs. The imaging data were merged with mapping data obtained during the ablation procedure. Areas with and without PN capture during pace mapping were marked along the anterior border of the right PVs. Ablation Procedure The ablation procedures were performed under general anesthesia and high-frequency jet ventilation in all patients. Double transseptal access was obtained with fluoroscopic and intracardiac echo guidance. A circular mapping catheter was used for left atrial and PV mapping (Lasso, BiosenseWebster, Diamond Bar, CA, or A-Focus, St. Jude Medical, Minneapolis, MN, USA). A 3.5 mm externally irrigated tip catheter (Thermocool SF, Biosense Webster; Themocool SMARTTOUCH Biosense Webster; Tacticath Quartz, St. Jude Medical) was used for ablation and testing for PN stimulation. When contact force sensing was used, the target force range was 10–40 x g for both mapping and ablation. Wide area PVI was performed in all patients. Patients with persistent AF underwent additional ablation.
Figure 2. Measuring the distance between the right superior pulmonary veins ostium and the line of RA-anterior LA overlap. Black dots = pacing sites with phrenic nerve capture; LA = left atrium; RA = right atrium; red and pink dots = ablation lesions; white dots = pacing sites with no phrenic nerve capture. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
Measurements Testing for PN Stimulation Testing for PN capture was performed with high energy stimulation (20 mA at 2 ms pulse duration),1 with the ablation catheter along the anterior border of the right PVs (Fig. 1A,B). The muscle paralysis was weaned off during isolation of the right PVs and confirmed by thumb adduction in response to train-of-four stimulation of the ulnar nerve. Detection of diaphragmatic stimulation was verified by fluoroscopy and direct observation of the abdomen. Pacing points were tagged on the electroanatomical maps with different colors depending on whether or not PN capture occurred. Ablation was avoided in areas where PN capture was present. Pacing from the SVC was also performed at the end of the procedure to check for PN function.
Offline measurements were performed on the electroanatomical maps using the imported CTA and MRA images. The ostia of the right superior and inferior PVs were identified and delineated as previously described.5 The area of overlap between the RA and LA was also marked. The shortest distances between the ostia of the right superior and inferior PVs and the line of overlap between the RA and LA were measured (Fig. 2). Statistical Analysis Paired comparison of categorical variables was performed with the χ 2 test. Logistic regression analysis was performed to assess the relationship of PN capture with clinical and anatomical data. Significance level was determined at
Roka et al. Prevention of Phrenic Injury Using Imaging
1059
Figure 3. Location of ablation lesions. Ablation lesions at the anterior antrum of the right pulmonary veins deployed proximal to the area of overlap (white line) between the right atrium and left atrium. A: A patient with a wide distance between the ostia of the right pulmonary veins and the area of the overlap between the right and left atria. B: A patient with a narrow distance between the ostia of the right pulmonary veins and the area of the overlap between the right and left atria. The line of overlap crosses the ostium of the right superior pulmonary vein. LA = left atrium; RA = right atrium; red and pink dots = ablation lesions. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
P < 0.05. IBM SPSS 20 software (IBM Corp., Armonk, NY, USA) was used for statistical computations. Results
TABLE 1 Distance Between the Line Demarcating the Area of Overlap (RA–LA Border) Along the Anterior Wall of the LA and the Right Pulmonary Vein Ostia
Number of RPVs
n
Distance of Line from RSPV (mm)
1 (right common PV) 2 (RSPV, RIPV) 3 (RSPV, RMPV, RIPV)
1 31 8
1 (from right common PV) 2±3 3±3
Patient Characteristics Forty patients were studied. Seventy-five percent were males, and the average age was 65.0 ± 8.6 years. The left atrial size was 45 ± 7 mm, the BMI 27.6 ± 4.7, and the BSA 2.00 ± 0.36. The indication for PVI was symptomatic drugrefractory paroxysmal AF in 21 (52.5%) and persistent AF in 19 (47.5%). Preprocedural CTA and MRA were obtained in 37 (92.5%) and 3 (7.5%) patients, respectively. The median number of right PV ostia was 2 (range 1–3). The median number of left PV ostia was 2 (range 1–2). The CARTO electroanatomical mapping system (Biosense-Webster) was used in 26 patients (65%) and the NavX system (St. Jude Medical) in 14 patients (35%). Contact force sensing was used in 15 out of 40 (38%) patients. All patients underwent PVI during the ablation procedure. In addition to the PVI, 50% of the patients had posterior wall ablation, 15% focal trigger ablation, 10% mitral isthmus line, and 35% cavotricuspid isthmus ablation. Location of the Area of Overlap The line demarcating the area of overlap (RA–LA border) ran along the anterior wall of the LA. This line was proximal to the ostium of the right superior PV in 23/40 (57.5%). In the remaining 17 (42.5%) patients, the line crossed the ostium of this vein (Fig. 3A,B). In all patients, this line of overlap was proximal to the ostium of the right inferior PV. When the line did not cross the ostia of the PVs, the distance separating it from the right superior PV and the right inferior PV ostia was 4 ± 3 and 13 ± 6 mm, respectively (Fig. 2, Table 1). The distance between the line and both ostia was more than 5 mm in 6 patients (15%), and less than or equal to 5 mm in 4 patients (10%). The distance was more than 5 mm only from the inferior PV ostia in 30 patients (75%). In no patient, the distance was more than 5 mm from the right superior, but less than or equal to 5 mm from the right inferior PV ostia.
Distance of Line from RIPV (mm) – 12 ± 6 17 ± 5
The distance was greater from the RIPV, than the RSPV. In patients where the line crossed the ostium, the distance was determined as 0 mm when calculating the means for each group. RIPV = right inferior pulmonary vein; RMPV = right middle pulmonary vein; RSPV = right superior pulmonary vein.
Incidence and Factors Affecting PN Capture A total of 361 points in the LA (9 ± 4/patient) were tested for PN stimulation. Of these, 172 were located in the area of RA and LA overlap, and 189 were not. PN capture occurred in 27% of all points. Sixty-five percent of the patients had at least 1 site where PN capture was present. PN capture occurred in none of the points with RA and LA overlap, but in 97 out of the 189 (51%) points in the area of no overlap (P < 0.001). There was no significant association between any of the clinical parameters and the incidence of PN capture (Table 2). Procedural Outcomes The total procedure duration including general anesthesia induction and extubation was 241±60 minutes. The fluoroscopy time was and 22 ± 6 minutes. PVI was successfully performed with large circumferential antral ablation. The PVs were encircled in ipselateral pairs in all patients. Isolation of the right PVs was achieved with ablation lesions proximal to the line of overlap between the LA and RA in 30 patients (75%, Fig. 3A,B). In 10 patients (25%), ablation lesions needed to be delivered distal to the line of overlap in order to achieve isolation (Fig. 4).
1060
Journal of Cardiovascular Electrophysiology
Vol. 26, No. 10, October 2015
TABLE 2 Correlation Between Clinical and Anatomical Characteristics and Phrenic Nerve Capture
Parameter
Value
Male: 30 (75%) Female: 10 (25%) Age 65.0 ± 8.6 years LA size 45 ± 7 mm Height 178 ± 10 cm Weight 87.7 ± 20.1 kg Body mass index 27.6 ± 4.7 kg/m2 Body surface area 2.00 ± 0.36 m2 Paroxysmal: 21 (52.5%) Atrial fibrillation Persistent: 19 (47.5%) Prior cardiac surgery 2 (5%) Number of right PV ostia Median 2 (range 1–3) Number of left PV ostia Median 2 (range 1–2) RA–LA border 17 (42.5%) overlapping with RSPV ostium Gender
Correlation with Presence of Phrenic Nerve Capture (Univariate Logistic Regression) P = 0.251 P = 0.359 P = 0.175 P = 0.132 P = 0.191 P = 0.671 P = 0.121 P = 0.273 P = 0.287 P = 0.276 P = 0.507 P = 0.169
LA = left atrium; RA = right atrium; PV = pulmonary vein; RSPV = right superior pulmonary vein.
Figure 4. Ablation lesions distal to the border of the overlap (white line) between the right and left atria. Ablation lesions (black arrow) distal to the line were required in this patient to achieve isolation of the rightsided pulmonary veins. High output pacing (20 mA at 2 ms pulse width) was utilized to test for phrenic nerve capture in this region, ablation was avoided in areas where phrenic nerve capture was present (blue dots). Blue dots = pacing sites with phrenic nerve capture; LA = left atrium; RA = right atrium; red and pink dots = ablation lesions; white dots = pacing sites with no phrenic nerve capture. For a high quality, full color version of this figure, please see Journal of Cardiovascular Electrophysiology’s website: www.wileyonlinelibrary.com/journal/jce
One patient had a persistent left-sided SVC, which was also isolated during the PVI. No patient in this study sustained PN injury as confirmed by pacing in the SVC at the end of the procedure. No other periprocedural complications occurred either. Discussion This study resulted in the following 3 important findings: first, PN capture with pacing in the antral region of the right-sided PVs does not occur in the area of overlap of the right and left atria; second, isolation of the right PVs can be achieved in the majority of the patients by limiting the anterior ablation lesions to the region proximal to the line demarcating the area of overlap; and third, there were no clinical or anatomical characteristics associated with higher incidence of PN capture with pacing. Incidence and Mechanism of Injury of the PN The location of the PN relative to the atria has been well described in anatomical studies.6 The PN runs in a craniocaudal direction as part of the pericardiacophrenic bundle, in close proximity of the superior vena cava, right PVs, and medial aspect of the LA. Injury to the PN during catheter ablation results from direct thermal injury. Experimental data show that the damage is axonal in nature and is characterized by Wallerian degeneration, with potential for recovery.7 This correlates well with clinical observations of cases of PN injuries resulting in transient diaphragmatic dysfunction. Damage to the PN has been reported with various energy modalities: RF, ultrasound, laser, and, most commonly, cryoablation.1-3,8,9 In the STOP AF study, the incidence of PN injury during cryoablation was 11.2%.10 This complication continues to occur despite preventive techniques including continuous PN pacing during cryoenergy delivery, and termination of freezing cycle in case of weakening or loss of diaphragmatic contraction. In a recent single-center study of 115 patients using cryoballoon for PVI, PN palsy occurred in 3.5%.11 The use of electromyographic monitoring and terminating the cryo lesion delivery with the reduction in diaphragmatic motion was found to reduce the risk of PN palsy.12 PN injury also occurs with RF ablation, although at a lower frequency. A large observational study of 3,755 patients reported an incidence of 0.5% of PN injury with RF ablation.3 Even when a WACA approach is used, this complication continues to occur.1 One possible explanation of the occurrence of PN injury despite wide ablation is the lack of a clear definition of a safe zone of ablation with the WACA approach. Prevention of Nerve Injury with Pace Mapping Various techniques have been proposed to avoid PN injury. High output pacing to delineate the course of the PN and avoiding the areas of capture is the most commonly used technique.1,4 This technique was investigated in a prospective study of 100 patients who underwent RF PVI using a WACA approach.1 PN capture occurred in 30% of patients leading to the modification of the ablation lines in order to avoid the areas of capture.1 Similar to our study, pacing to identify the location of the PN was performed at 20 mA and 2 ms.1 It
Roka et al. Prevention of Phrenic Injury Using Imaging
is conceivable that pacing at higher output may have led to phrenic capture in more areas. Another technique is PN pacing at a location cranial to the ablation site. With this technique, pacing is usually performed from the SVC to monitor the right PN, and from the left subclavian vein to monitor the left PN. Diaphragmatic contractions are monitored with fluoroscopy, palpation, or intracardiac echo, and ablation is stopped when diaphragmatic contractions are weakened or lost. All the studies mentioned above and others demonstrated that pace mapping may help reduce PN injury during ablation. However, pacing for PN capture adds time and complexity to the ablation procedure. A study of 18 patients demonstrated that pace mapping of the PN required 14 ± 6 minutes.13 This is a significant amount of time to add to an already long procedure. Role of Preprocedural Imaging Preprocedural CT imaging has been shown to allow the localization of the PN. The pericardiacophrenic bundle, including the nerve and the accompanying vessels, can be visualized from reconstructed 3D images.14,15 Using this method, the average distance between the phrenic bundle and the site of PN capture was found to be 8.7 ± 5.8 mm. However, direct reconstruction of the course of the bundle by CT can be difficult to achieve, requires a high level of expertise, is time-consuming and more complicated in comparison to the method described in our study. The findings in our study suggest that a simple maneuver delineating the border between the RA and the LA can define a safe area of ablation and avoid the need for pace mapping. This delineation process can be performed in a short time, and it neither interferes with the flow of the ablation, nor does it result in the prolongation of the procedural time. This technique depends on accurate image integration as the CT and MRI images are obtained before the ablation procedure. Some changes in anatomy due to volume shifts or anesthesia/jet ventilation can occasionally occur, and caution should be exercised during the procedure to detect signs of misalignment of the integrated image. Despite this limitation, the technique was 100% accurate in identifying areas where PN capture was not possible. The line demarcating the area of overlap of the RA and LA is closer to the ostium of the right superior PV than it is to the ostium of the right inferior PV. In some patients, in our study this distance was found to be as long as 29 mm. Is such a situation, isolating the right PV can be challenging if ablations were to be limited to the wide area proximal to the line. As a result, ablation may need to be performed in some patients distal to the line or over the carina in order to achieve isolation. Only in this situation pacing for PN mapping would be necessary. Conclusion PN injury is one of the major complications of PVI. The WACA approach reduces but does not eliminate the risk of injury. One limitation of the WACA approach is the lack of objective definition of wide ablation. This study demonstrates that a safe zone for wide ablation can be easily indentified and clearly defined using routine preprocedural imaging data from CTA or MRA. Ablation in this area can be performed without the need for pacing to delineate the PN. Pacing can
1061
be reserved only to situations necessitating ablation distal to the area of overlap. The adoption of this technique provides an objective guide for ablation during WACA and may result in shortening and simplification of the ablation procedure. Limitations One limitation of this study is that the application of the technique is limited to ablation catheters that can be visualized on the 3D electroanatomical maps. Using this technique for current generation balloon-based catheters requires imaging tools that allow the integration of electroanatomical maps with X-ray imaging, which is not readily available in all electrophysiology laboratories. Another limitation is the unblinded design of this study. Pacing for the PN was performed after the demarcation of the line of overlap of the RA and LA. However, the difference in the rate of capture of the PN between the areas proximal and distal to the line of overlap was very large, and is unlikely to become less significant if a blinded design is used.
References 1. Yong Ji S, Dewire J, Barcelon B, Philips B, Catanzaro J, Nazarian S, Cheng A, Spragg D, Tandri H, Bansal S, Ashikaga H, Rickard J, Kolandaivelu A, Sinha S, Marine JE, Calkins H, Berger R: Phrenic nerve injury: An underrecognized and potentially preventable complication of pulmonary vein isolation using a wide-area circumferential ablation approach. J Cardiovasc Electrophysiol 2013;24:10861091. 2. Calkins H, Kuck KH, Cappato R, Brugada J, Camm AJ, Chen SA, Crijns HJ, Damiano RJ Jr, Davies DW, DiMarco J, Edgerton J, Ellenbogen K, Ezekowitz MD, Haines DE, Haissaguerre M, Hindricks G, Iesaka Y, Jackman W, Jalife J, Jais P, Kalman J, Keane D, Kim YH, Kirchhof P, Klein G, Kottkamp H, Kumagai K, Lindsay BD, Mansour M, Marchlinski FE, McCarthy PM, Mont JL, Morady F, Nademanee K, Nakagawa H, Natale A, Nattel S, Packer DL, Pappone C, Prystowsky E, Raviele A, Reddy V, Ruskin JN, Shemin RJ, Tsao HM, Wilber D: 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: Recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design: A report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial Fibrillation. Developed in partnership with the European Heart Rhythm Association (EHRA), a registered branch of the European Society of Cardiology (ESC) and the European Cardiac Arrhythmia Society (ECAS); and in collaboration with the American College of Cardiology (ACC), American Heart Association (AHA), the Asia Pacific Heart Rhythm Society (APHRS), and the Society of Thoracic Surgeons (STS).Endorsed by the governing bodies of the American College of Cardiology Foundation, the American Heart Association, the European Cardiac Arrhythmia Society, the European Heart Rhythm Association, the Society of Thoracic Surgeons, the Asia Pacific Heart Rhythm Society, and the Heart Rhythm Society. Heart Rhythm 2012;9:632-696. 3. Sacher F, Monahan KH, Thomas SP, Davidson N, Adragao P, Sanders P, Hocini M, Takahashi Y, Rotter M, Rostock T, Hsu LF, Clementy J, Haissaguerre M, Ross DL, Packer DL, Jais P: Phrenic nerve injury after atrial fibrillation catheter ablation: Characterization and outcome in a multicenter study. J Am Coll Cardiol 2006;47:2498-2503. 4. Fan R, Cano O, Ho SY, Bala R, Callans DJ, Dixit S, Garcia F, Gerstenfeld EP, Hutchinson M, Lin D, Riley M, Marchlinski FE: Characterization of the phrenic nerve course within the epicardial substrate of patients with nonischemic cardiomyopathy and ventricular tachycardia. Heart Rhythm 2009;6:59-64. 5. Mansour M, Refaat M, Heist EK, Mela T, Cury R, Holmvang G, Ruskin JN: Three-dimensional anatomy of the left atrium by magnetic resonance angiography: Implications for catheter ablation for atrial fibrillation. J CardiovascElectrophysiol 2006;17:719-723. 6. Lachman N, Syed FF, Habib A, Kapa S, Bisco SE, Venkatachalam KL, Asirvatham SJ: Correlative anatomy for the electrophysiologist, part II:
1062
7.
8.
9. 10.
11.
Journal of Cardiovascular Electrophysiology
Vol. 26, No. 10, October 2015
Cardiac ganglia, phrenic nerve, coronary venous system. J Cardiovasc Electrophysiol 2011;22:104-110. Andrade JG, Dubuc M, Ferreira J, Guerra PG, Landry E, Coulombe N, Rivard L, Macle L, Thibault B, Talajic M, Roy D, Khairy P: Histopathology of cryoballoon ablation-induced phrenic nerve injury. J Cardiovasc Electrophysiol 2014;25:187-194. Bunch TJ, Bruce GK, Mahapatra S, Johnson SB, Miller DV, Sarabanda AV, Milton MA, Packer DL: Mechanisms of phrenic nerve injury during radiofrequency ablation at the pulmonary vein orifice. J Cardiovasc Electrophysiol 2005;16:1318-1325. Andrie RP, Schrickel JW, Nickenig G, Lickfett L: Left phrenic nerve injury during cryoballoon ablation of the left superior pulmonary vein. Pacing Clin Electrophysiol 2012;35:e334-e336. Packer DL, Kowal RC, Wheelan KR, Irwin JM, Champagne J, Guerra PG, Dubuc M, Reddy V, Nelson L, Holcomb RG, Lehmann JW, Ruskin JN, Investigators SAC: 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;61:17131723. Metzner A, Rausch P, Lemes C, Reissmann B, Bardyszewski A, Tilz R, Rillig A, Mathew S, Deiss S, Kamioka M, Toennis T, Lin T, Ouyang F, Kuck KH, Wissner E: The incidence of phrenic nerve injury during
12.
13.
14.
15.
pulmonary vein isolation using the second-generation 28 mm cryoballoon. J Cardiovasc Electrophysiol 2014;25:466-470. Mondesert B, Andrade JG, Khairy P, Guerra PG, Dyrda K, Macle L, Rivard L, Thibault B, Talajic M, Roy D, Dubuc M, Shohoudi A: Clinical experience with a novel electromyographic approach to preventing phrenic nerve injury during cryoballoon ablation in atrial fibrillation. Circ Arrhythm Electrophysiol 2014;7:605-611. Schmidt B, Chun KR, Ouyang F, Metzner A, Antz M, Kuck KH: Three-dimensional reconstruction of the anatomic course of the right phrenic nerve in humans by pace mapping. Heart Rhythm 2008;5:11201126. Fukumoto K, Takatsuki S, Jinzaki M, Yamada M, Tanimoto K, Nishiyama N, Aizawa Y, Hagiwara Y, Fukuda Y, Kimura T, Miyoshi S, Kuribayashi S, Fukuda K: Three-dimensional imaging and mapping of the right and left phrenic nerves: Relevance to interventional cardiovascular therapy. Europace 2013;15:937-943. Horton R, Di Biase L, Reddy V, Neuzil P, Mohanty P, Sanchez J, Nguyen T, Mohanty S, Gallinghouse GJ, Bailey SM, Zagrodzky JD, Burkhardt JD, Natale A: Locating the right phrenic nerve by imaging the right pericardiophrenic artery with computerized tomographic angiography: Implications for balloon-based procedures. Heart Rhythm 2010;7:937-941.
