Where to draw the mitral isthmus line in catheter ablation of atrial fibrillation: histological analysis

Transcription

1 European Heart Journal (2005) 26, doi: /eurheartj/ehi095 Clinical research Where to draw the mitral isthmus line in catheter ablation of atrial fibrillation: histological analysis Fred H.M. Wittkampf 1 *, Matthijs F. van Oosterhout 2, Peter Loh 1, Richard Derksen 1, Evert-jan Vonken 3, Piet J. Slootweg 1, and Siew Yen Ho 4 1 Heart Lung Center Utrecht, University Medical Center E03-406, PO Box 85500, 3508 GA, Utrecht, The Netherlands 2 University Medical Center Utrecht, Department of Pathology, Utrecht, The Netherlands 3 University Medical Center Utrecht, Department of Radiology, Utrecht, The Netherlands 4 Department of Paediatrics, National Heart and Lung Institute, and Royal Brompton Hospital, London, UK Received 6 October 2004; revised 6 October 2004; accepted 25 November 2004; online publish-ahead-of-print 6 January 2005 KEYWORDS Atrial fibrillation; Catheter ablation Introduction Catheter ablation of atrial fibrillation has evolved from ablation of arrhythmogenic foci within the pulmonary veins (PV) to complete electrical isolation of the PV, often guided by perimetric mapping with a multi-polar loop catheter positioned at the rim of each of the PV ostia. 1 4 Adding an ablation line connecting the inferior margin of the ostium of the left inferior PV (LIPV) to the mitral annulus appears to increase the success rate of treating atrial fibrillation by catheter ablation. 5 9 Another approach is to encircle both right and left pairs * Corresponding author. address: fredwittkampf@mac.com Aims A linear lesion between the left inferior pulmonary vein orifice and mitral annulus, the so-called mitral isthmus, may improve the success of catheter ablation for atrial fibrillation. Gaps in the lesion line, however, may facilitate left atrial flutter. The aim of the study was to determine the optimal location of the lesion line by serial sectioning of the isthmus area. Methods and results In a post-mortem study of 16 patients with normal left atria, serial sections of the isthmus area from 10 mm superior to and 30 mm inferior to the isthmus were studied by light microscopy. The length of the isthmus was mm. On average, the muscle sleeve around the coronary sinus ended 10 mm inferior to the isthmus. The prevalence of a ramus circumflexus,5 mm from the endocardial surface, decreased from 60% in the most superior section to 0% in the most inferior section. Atrial arteries were frequently present in all sections. Conclusions The thickness of atrial myocardium, the ramus circumflexus sometimes very close to the endocardium, a myocardial sleeve around the coronary sinus, and local cooling by atrial arteries and veins may complicate the creation of conduction block in the mitral isthmus. of PV ostia by multiple lesions followed by a linear lesion connecting both areas and another linear lesion from the inferior margin of the left-sided encircling lesion to the mitral annulus. 3,10,11 In their most extensive form, both techniques thus include a linear lesion in the posteroinferior wall of the left atrium through an area described by electrophysiologists as the mitral isthmus. In practice, however, the creation of complete isthmus block is not always successful. Gaps in the lesion line may result in conduction delay and facilitate left atrial flutter. 4,9,12 14 Recently Becker 15 analysed the anatomy of the isthmus section in 20 hearts. 16 The aim of the present study was to analyse a broader part of that isthmus to assess the optimal location for such a lesion line. & The European Society of Cardiology All rights reserved. For Permissions, please journals.permissions@oupjournals.org

2 690 F.H.M. Wittkampf et al. Methods The investigational protocol was approved by the ethics committee of the University Medical Center. We examined the heart specimens from 16 autopsies carried out routinely. The left atrium was opened from the anterior aspect to expose the posterior and inferior parts of its wall. From each left atrium, we excised a piece of tissue that included the veno-atrial junctions between both upper PV and the left atrioventricular junction (Figure 1 ). The excised pieces were mounted on a flat support and fixed in formalin. After fixation, the length of the mitral isthmus, defined as the line from the inferior margin of the LIPV ostium and perpendicular to the mitral annulus, was measured along the endocardial surface. Each piece was then cut longitudinally through the isthmus and further divided by making parallel cuts at 10 mm intervals, and approximately equally as long as the isthmus section (Figure 1 ). Because electrophysiologists usually orient relative to a catheter in the coronary sinus, it is more convenient to refer to these sections as being distal or proximal relative to the index cut through the isthmus. Full-face histological sections were prepared from each piece of tissue and stained with Van Gieson s stain. All sections were photographed digitally and analysed using standard software (NIH-Image). Over the complete length of each section, we measured the myocardial depth by noting the distance between the endocardium and the most subepicardial muscle fibres at 0.5- to 1-mm intervals. This measurement did not include the myocardial sleeve around the vein. We further recorded the positions of the great cardiac vein or its continuation, the coronary sinus (CS), and all arteries and veins with a diameter.0.25 mm, relative to the endocardium and mitral annulus (Figure 1 ). Since cuts at 10 mm intervals did not allow for a precise localization of the transition from the CS to the great cardiac vein, we therefore used the term vein to refer to both structures, acknowledging that the CS is usually invested in a myocardial sleeve whereas the great cardiac vein is usually devoid of sleeve. Anatomically, the junction between the great cardiac vein and CS is usually taken to be at the entrance of the vein (or ligament) of Marshall, or at the valve of Vieussens (Figure 2 ). Neither structure is readily visible on fluoroscopy. Using the inner wall of the veins and the external elastic lamina of the arteries as boundaries, the vessel diameters were calculated by assuming a circular vessel perimeter. For vessels showing an oval cross-sectional shape and a thinner wall thickness along the short axis, we assumed a slanted course relative to the section plane and then the short axis was taken as vessel diameter. Data are given as mean + s.d. Results The mean age of patients was years. The cause of death was non-cardiac in all 16 patients. Left ventricular myocardial infarcts were found in four patients including a small sub-endocardial infarct in one patient; the other patients had none or only mild cardiac pathology. Macroscopically, the left atria were not dilated and none of the patients had a history of atrial fibrillation. The 12-lead electrocardiograms that were available for eight patients all demonstrated sinus rhythm with normal P-wave morphology. One patient had left ventricular hypertrophy related to hypertension. In all 16 patients, the tissue sample included the mitral isthmus section and at least two proximal sections (Figure 1 ). The piece 10 mm distal to the isthmus was available in 15 patients while the most proximal section at 30 mm from the isthmus and close to the atrial septum could be harvested in 13/16 patients. Section 10 mm distal to isthmus In the section 10 mm distal to the mitral isthmus, available in 15/16 patients, the vein was present in 14 patients. None had a muscular venous sleeve. The average maximum distance between endocardium and subepicardial muscle fibres (myocardial depth) was mm for the first 5 mm. Measured over the complete length of the 15 available sections, the average of the maximum myocardial depth was mm and Figure 1 Posterior wall of the left atrium. Serial sections were taken at 10 mm intervals perpendicular to the mitral annulus. The second section from the right is the shortest connecting line between the lower border of the LIPV ostium and mitral annulus that we termed the mitral isthmus. In that area, cavities in which the tip of an ablation catheter can easily get entrapped are present. RIPV: right inferior pulmonary vein. Figure 2 This view of the epicardial aspect of a human heart shows the relationship of the mitral isthmus (broken line) to the great cardiac vein/coronary sinus transition indicated by the blue arrow.

3 Mitral isthmus mm in all cases (Table 1 ). In most sections, the maximum myocardial depth was found approximately in the middle of the section just above the vein. Arteries were often present. Based on their size and location in successive sections, nine arteries (in nine patients) were identified as the ramus circumflexus (RCx) or a branch of it nourishing ventricular myocardium. In six patients, its distance to the endocardium was,5 mm and one of these arteries was in direct contact with the atrial myocardium. Isthmus section The length of the isthmus was mm, range mm (Table 1 ). Again, the atrial myocardium was relatively thin within the first 5 mm from the annulus with an average maximum myocardial depth of only mm, range mm (Figure 3A C). The maximum myocardial depth was most often found approximately in the middle of the section just above the vein. The average of the maximum myocardial depth was mm, range mm (Table 1 ). A muscular sleeve around the vein was present in three patients. In two of these three patients, this sleeve was incomplete; very thin muscle fibres only partly (,30%) covered the perimeter of the vein and these muscle fibres were not connected to atrial myocardium at that level (Figure 3A ). In one patient, the sleeve was Table 1 Data of all sections complete and continuous with the atrial myocardium (Tables 1 2 ). A RCx was identified in seven sections in seven patients (Figures 3 and 4, Table 1 ). In five patients, the RCx was closer than 5 mm from the endocardium of which three were closer than 2 mm from the endocardium and in direct contact with the atrial myocardium. Of these five RCx, one was located above the level of the vein; two were sandwiched between vein and endocardium, and another two were located between the vein and mitral annulus. Section 10 mm proximal to isthmus The muscular sleeve around the vein was present in 11/16 patients. It was complete and connected to atrial myocardium in eight patients (Figure 3C ), but incomplete and not connected to atrial myocardium at this level in the three other patients. The average maximum myocardial depth was mm. Here too, the myocardial wall was thin near the annulus: the average maximum myocardial depth was 1.6 mm for the first 5 mm near the annulus. The maximum myocardial depth was most often found approximately in the middle of the section just above the vein. In five patients, the RCx was present in this section and three of these arteries were,5 mm from the endocardium. Proximal Mitral isthmus! Distal 230 mm 220 mm 210 mm 0 mm 10 mm Sections a n Length, mm Maximum myocardial depth, mm b Maximum myocardial range, mm Vein Present, n (%) 13 (100) 16 (100) 16 (100) 16 (100) 14 (88) Diameter, mm Distance to annulus, mm Muscular sleeve, n (%) 12 (92) 15 (94) 11 (69) 3 (19) 0 Connected sleeve, n (%) 12 (92) 14 (88) 8 (50) 1(6) 0 RCx n (%) 0 1(6) 5 (31) 7 (44) 9 (60) Diameter, mm Depth, mm n with depth,5mm n embedded c Atrial arteries.0.25 mm n n with depth,5mm Atrial veins.0.25 mm n n with depth,5mm a The length of the isthmus is measured along the endocardial surface from the inferior margin of the left inferior pulmonary vein ostium to the base of the mitral leaflet. The other sections were collected with approximately the same length as the isthmus section. b Maximum myocardial depth, maximum distance between endocardium and most subepicardial muscle fibres. c Embedded, in direct contact with atrial myocardium.

4 692 F.H.M. Wittkampf et al. Table 2 Obstacles in individual patients Figure 3 Three representative sections. (A) Isthmus section with an occluded RCx in close contact with atrial myocardium and at only 1.4 mm from the endocardium. This section also shows a very thin and incomplete muscular sleeve around the vein (V). (B) Isthmus section with an RCx and branch (RCx-b) and two atrial arteries. RCx-b is located below the level of the annulus and therefore not included in the data. All three arteries above the annulus are completely embedded within atrial myocardium. This section also happened to go through the middle of one of the isthmus crevices and shows the extremely thin myocardial wall at this location. (C) A section 10 mm proximal to the isthmus with a complete muscular sleeve around the vein. The creation of complete conduction block will only be possible when ablations include application(s) from inside the vein. Sections shown in panels B and C were bent to fit in the embedding cassette. More proximal sections The prevalence of the muscular sleeve around the vein increased while the prevalence of the RCx decreased. (Table 1, Figure 5 ) At 30 mm proximal to the isthmus, the RCx was absent in all sections while the myocardial sleeve was present in 12/13 cases (Tables 1 and 2 ). All sections Proximal Mitral!Distal isthmus Patient 230 mm 220 mm 210 mm 0 mm 10 mm 1 n.a. S S 2 S S, M S, M 3 S, M S RCx RCx 4 S, M S, M S 5 S S 6 n.a S, M M RCx RCx 7 S RCx RCx RCx 8 S S RCx 9 S S S 10 S S RCx RCx RCx 11 S, M S, M, RCx S, M, RCx M, RCx n.a. 12 M 13 S S S S 14 n.a. S 15 S S S RCx 16 S, M S, M S, M RCx, RCx, 5 mm from the endocardial surface. M, myocardial depth. 5 mm. S, the presence of a complete muscular sleeve around the vein. n.a., not available. There was some increase in myocardial depth from distal to proximal sections (Table 1 ). Overlap of atrial myocardium onto the endocardial aspect of the mitral valve leaflet was only found in one patient in each of the four most distal section groups, but this overlap was always,0.5 mm. In all sections where an incomplete myocardial sleeve was present around the vein, it was also not connected to atrial myocardium at that level. Atrial arteries and atrial veins.0.25 mm in diameter, often embedded in atrial myocardium, were frequently present at all section levels (Figure 4, Table 1 ). A notable observation relevant for catheter ablation is the presence of crevices in approximately the middle area of the mitral isthmus, close to the base of the left atrial appendage, in 15/16 patients (Figure 1 ). These depressions, presumed to be the valleys between remnants of pectinate muscles that extended from the left atrial appendage, were present in an otherwise very smooth area. Importantly, many of them appeared large enough to engage the tip of an ablation electrode. The remaining wall thickness at the bottom of the depressions could be extremely thin (Figure 3B ).

5 Mitral isthmus 693 isthmus in only 1/16 patient, and 30 mm proximal to the isthmus in 0/13 patients (Table 2 ). Discussion Figure 4 Prevalence of arteries in the 16 isthmus sections classified according to their diameter. Prevalence (%) was calculated by dividing the total number of arteries within the specified diameter range by the number of isthmus sections. 16 Only arteries.0.25 mm in diameter and,5 mm from the endocardium are included. Embedded: in direct contact with atrial myocardium. Figure 5 Major obstacles in the various sections. The prevalence (% scale, left abscissa) of a muscular sleeve around the vein decreases whereas that of the RCx increases from proximal to distal sections. Also plotted (mm scale, right abscissa) is the maximum myocardial depth, defined as the maximum distance between the endocardium and the most subepicardial muscle fibres (not including the myocardial venous sleeve) measured over the complete length of each section, with its standard deviation (vertical bars). The histological data allowed for selecting the optimal location(s) for an ablation line through the isthmus area in the individual patients using arbitrary limits for the distance between endocardium and RCx (5 mm), maximum myocardial depth (5 mm), and the presence or absence of a muscular sleeve around the vein that was connected to atrial myocardium. With these criteria, isthmus block could have been created safely and successfully 10 mm distal to the isthmus in 9/15 patients, at the isthmus in 10/16 patients, 10 mm proximal to the isthmus in 5/16 patients, 20 mm proximal to the Ablation for atrial fibrillation is mainly carried out in the left atrium and an understanding of left atrial anatomy and pulmonary veno-atrial junctions is helpful to electrophysiologists. Anatomically, the left atrium comprises four components: septum, appendage, vestibule, and venous component. 17 The vestibular component is the atrial wall leading to, and terminating in, the hinge line (annulus) of the mitral valve. In recent years, a linear lesion has been added in a part of the vestibule, the so-called mitral isthmus, as part of the catheter ablation strategy for atrial fibrillation. Although the vestibular component is generally smooth on the endocardial surface and appears simple in composition, our study shows that there are features that may affect the rate of success in creating a transmural lesion line through the isthmus or, conversely, result in complications should too much energy be deployed. Applying the nomenclature of Cosio et al. 18 in fluoroscopic left anterior-oblique projection, the isthmus is estimated to lie adjacent to the posterior sector of the left atrioventricular junction. With the index cut at the isthmus, the cut distal to the index is at 10 mm superior while the proximal cuts are inferior in attitudinal orientation. At all five section levels, the myocardium was relatively thin near the annulus. There, relatively low radio frequency (RF) power should theoretically suffice to create transmural lesions and problems in creating conduction block in that area are therefore most likely due to poor tissue contact or catheter instability. The maximum myocardial depth was most often found approximately in the middle of the sections, predominantly close to and above the vein. Besides relatively thick myocardium, other structures may complicate or impede the creation of conduction block by catheter ablation. 19,20 This includes (i) a myocardial sleeve around the vein, continuous with atrial myocardium that may bridge an endocardial lesion line, (ii) an RCx in intimate contact with atrial myocardium that may be damaged by RF ablation, (iii) local cooling of atrial myocardium by the above-mentioned vessels and by other atrial arteries and veins, and (iv) crevices in the isthmus area that may entrap the tip of the ablation catheter. While open flush, irrigated ablation electrodes may be safer than conventional electrodes by preventing protein aggregation and coagulum formation, the lack of temperature feedback may impose certain risks. When the tip of the ablation electrode is entrapped in one of the crevices in the isthmus (Figures 1 and 3B ), the delivery of nominal RF power levels without temperature feedback may lead to steam explosions within the thin atrial wall and pericardial effusion. 21,22 Our serial sectioning demonstrates that the muscular venous sleeve and relatively thick myocardium are the

6 694 F.H.M. Wittkampf et al. main obstacles with a too proximal position of the ablation line, while the RCx may be at risk with a more distal position of the line (Figure 5, Tables 1 and 2 ) The optimal location for the creation of isthmus conduction block varied between patients. Within the 40-mm wide area that was investigated, endocardially applied lesions of 5 mm depth would suffice to safely and successfully create an isthmus block in 60, 63, 31, 6, and 0% at 10 mm distal to the isthmus, in the isthmus, and at 10, 20, and 30 mm proximal to the isthmus, respectively (Table 2 ). In 3/16 patients, the creation of isthmus block by endocardial RF application may have been very difficult due to the presence of an obstacle throughout the complete isthmus area that was investigated. While a proximal location of the ablation line will be safer by avoiding the RCx, such a procedure will often require ablation from inside the vein due to the presence of a myocardial sleeve around this vein. Recently, Becker 15 studied the gross morphological features of the isthmus section in 20 patients. Contrary to his findings, our histological assessment revealed thickest atrial wall midway, with tapering at either end of the isthmus. This study also described the isthmus as a watershed area between right coronary artery and left circumflex coronary arteries. Our serial sectioning, however, demonstrates that the prevalence of major coronary arteries decreased from distal to proximal sections, and thus the likelihood of arterial damage would decrease with a more proximal location of the ablation line. In addition, the prevalence of atrial arteries did not differ between the various sections although it appeared to be somewhat lower in the section 10 mm proximal to the isthmus. During an ablation procedure, one is usually not informed about the extension of the myocardial sleeve around the vein, the thickness of atrial myocardium, or about the presence and position of important arteries. Preferably, the creation of isthmus block should be attempted as far as possible from the RCx, but just distal to the end of the complete sleeve. When the sleeve is incomplete at its distal end, ablation of such a sleeve does not appear to be necessary for obtaining isthmus block because the histological images suggest that such incomplete sleeves are not connected to the myocardial wall. Our study confirms the observation of Becker 15 is that the CS and great cardiac vein do not mark the annulus. In the area that was studied (5 cuts ¼ 40 mm), the vein runs at the atrial side of the annulus at a distance that increased from 6 mm in the most distal section to 11 mm in the most proximal section (Table 1 ). This slanted course corresponds to an angle of 78 relative to the annulus, and this angle appears small enough to allow for estimation of the direction of the shortest connection line between LIPV ostium and mitral annulus. Atrial arteries that are fully embedded in atrial myocardium are likely to be damaged by transmural ablations. Such ablations may then affect more atrial myocardium than the ablation line itself and one might speculate that this could contribute to the late curative effect of left atrial lesions in patients with atrial fibrillation. 27 Study limitations Although the heart specimens were not from patients with atrial fibrillation, the size of the left atrium was within normal range and comparable to patients with idiopathic atrial fibrillation, the patient group usually treated with catheter ablation. We acknowledge that the relevance of our observations to catheter ablation is necessarily speculative, but we think it is important to draw attention to the variability in the structure of the mitral isthmus and its vicinity. Conclusions The muscle sleeve around the CS and an RCx close to the endocardial surface are the two major limiting factors for the creation of mitral isthmus conduction block by catheter ablation. On average, the distal end of the sleeve terminates 10 mm proximal to the shortest connecting line between the LIPV and mitral annulus. The RCx, approaching superiorly, may reach the isthmus and be,5 mm from the endocardium in 30% of patients. The optimal location for the creation of mitral isthmus conduction block differs between patients. The highest success rates may be expected at the isthmus and 10 mm distal to the isthmus, but cooling by the RCx blood flow and/or damage to this artery may complicate such a procedure at either location. A more proximal location of the ablation line will lower the risk of arterial damage, but may require ablation from inside the vein in a high number of patients. Important atrial arteries are often present in the isthmus area. Isthmus ablation and arterial damage may thus affect more atrial myocardium than the ablation line itself. Small crevices, present in the isthmus area in almost all patients, can entrap the tip of the ablation catheter, which may lead to excessive tissue heating and tamponade. Acknowledgements The authors are grateful to Pieter Doevendans for his help with the preparation of this manuscript. References 1. Haïssaguerre M, Jaïs P, Shah DC, Takahashi A, Hocini M, Quiniou G, Garrigue S, Le Mouroux A, Le Metayer P, Clementy J. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998;339: Haïssaguerre M, Jaïs P, Shah DC, Garrigue S, Takahashi A, Lavergne T, Hocini M, Peng JT, Roudaut R, Clementy J. Electrophysiological end point for catheter ablation of atrial fibrillation initiated from multiple pulmonary venous foci. Circulation 2000;101: Oral H, Knight BP, Tada H, Ozaydin M, Chugh A, Hassan S, Scharf C, Lai SW, Greenstein R, Pelosi F Jr, Strickberger SA, Morady F. Pulmonary vein isolation for paroxysmal and persistent atrial fibrillation. Circulation 2002;105: Oral H, Scharf C, Chugh A, Hall B, Cheung P, Good E, Veerareddy S, Pelosi F Jr, Morady F. Catheter ablation for paroxysmal atrial

7 Mitral isthmus 695 fibrillation: segmental pulmonary vein ostial ablation versus left atrial ablation. Circulation 2003;108: Jaïs P, Shah DC, Haïssaguerre M, Takahashi A, Lavergne T, Hocini M, Garrigue S, Barold SS, Le Metayer P, Clementy J. Efficacy and safety of septal and left-atrial linear ablation for atrial fibrillation. Am J Cardiol 1999;84:139R 146R. 6. Macle L, Jaïs P, Weerasooriya R, Hocini M, Shah DC, Choi KJ, Scavee C, Raybaud F, Clementy J, Haïssaguerre M. Irrigated-tip catheter ablation of pulmonary veins for treatment of atrial fibrillation. J Cardiovasc Electrophysiol 2002;13: Jaïs P, Weerasooriya R, Shah DC, Hocini M, Macle L, Choi KJ, Scavee C, Haïssaguerre M, Clementy J. Ablation therapy for atrial fibrillation; past present and future. Cardiovasc Res 2002;54: Scavee C, Weerasooriya R, Jais P, Hocini M, Haissaguerre M. Linear ablation in the left atrium using a nonfluoroscopic mapping system. J Cardiovasc Electrophysiol 2003;14: Ernst S, Ouyang F, Lober F, Antz M, Kuck KH. Catheter-induced linear lesions in the left atrium in patients with atrial fibrillation. J Am Coll Cardiol 2003; 42: Pappone C, Rosanio S, Oreto G, Tocchi M, Gugliotta F, Vicedomini G, Salvati A, Dicandia C, Mazzone P, Santinelli V, Gulletta S, Chierchia S. Circumferential radiofrequency ablation of pulmonary vein ostia: A new anatomic approach for curing atrial fibrillation. Circulation 2000;102: Pappone C, Oreto G, Rosanio S, Vicedomini G, Tocchi M, Gugliotta F, Salvati A, Dicandia C, Calabro MP, Mazzone P, Ficarra E, Di Gioia C, Gulletta S, Nardi S, Santinelli V, Benussi S, Alfieri O. Atrial electroanatomic remodeling after circumferential radiofrequency pulmonary vein ablation: efficacy of an anatomic approach in a large cohort of patients with atrial fibrillation. Circulation 2001;104: Jaïs P, Shah DC, Haïssaguerre M, Hocini M, Peng JT, Takahashi A, Garrigue S, Le Metayer P, Clementy J. Mapping and ablation of left atrial flutters. Circulation 2000;101: Ouyang F, Ernst S, Vogtmann T, Goya M, Volkmer M, Schaumann A, Bansch D, Antz M, Kuck KH. Characterization of reentrant circuits in left atrial macroreentrant tachycardia: critical isthmus block can prevent atrial tachycardia recurrence. Circulation 2002; 105: Villacastin J, Perez-Castellano N, Moreno J, Gonzalez R. Left atrial flutter after radiofrequency catheter ablation of focal atrial fibrillation. J Cardiovasc Electrophysiol 2003;14: Becker AE. Left atrial isthmus: anatomic aspects relevant for linear catheter ablation procedures in humans. J Cardiovasc Electrophysiol 2004;15: Jaïs P, Hocini M, Sanders P, Haïssaguerre M. The left atrial isthmus: from dissection bench to ablation lab. J Cardiovasc Electrophysiol 2004;15: Ho SY, Sanchez-Quintana D, Cabrera JA, Anderson RH. Anatomy of the left atrium: implications for radiofrequency ablation of atrial fibrillation. J Cardiovasc Electrophysiol 1999;10: Cosio FG, Anderson RH, Kuck KH, Becker A, Borggrefe M, Campbell RW, Gaita F, Guiraudon GM, Haissaguerre M, Rufilanchas JJ, Thiene G, Wellens HJ, Langberg J, Benditt DG, Bharati S, Klein G, Marchlinski F, Saksena S. Living anatomy of the atrioventricular junctions. A guide to electrophysiologic mapping. Circulation 1999; 100:e31 e Fuller IA, Wood MA. Intramural coronary vasculature prevents transmural radiofrequency lesion formation: implications for linear ablation. Circulation 2003;107: Igawa O, Adachi M, Hisatome I, Matsui Y. Histopathologic background for resistance to conventional catheter ablation of common atrial flutter. J Cardiovasc Electrophysiol 2004;15: Nakagawa H, Yamanashi WS, Pitha JV, Arruda M, Wang X, Ohtomo K, Beckman KJ, McClelland JH, Lazzara R, Jackman WM. Comparison of in vivo tissue temperature profile and lesion geometry for radiofrequency ablation with a saline-irrigated electrode versus temperature control in a canine thigh muscle preparation. Circulation 1995;91: Juneja R, O Callaghan P, Rowland E. Tissue rupture and bubble formation during radiofrequency catheter ablation. Circulation 2001; 103: Pons M, Beck L, Leclercq F, Ferriere M, Albat B, Davy JM. Chronic left main coronary artery occlusion: a complication of radiofrequency ablation of idiopathic left ventricular tachycardia. Pacing Clin Electrophysiol 1997;20: Ouali S, Anselme F, Savoure A, Cribier A. Acute coronary occlusion during radiofrequency catheter ablation of typical atrial flutter. J Cardiovasc Electrophysiol 2002;13: de Paola AA, Leite LR, Arfelli E. Mechanical reperfusion of acute right coronary artery occlusion after radiofrequency catheter ablation and long-term follow-up angiography. J Invasive Cardiol 2003; 15: Lüdinghausen VM, Ohmachi N, Boot C. Myocardial coverage of the coronary sinus and related veins. Clin Anat 1992; 5: Nath S, Whayne JG, Kaul S, Goodman NC, Jayaweera AR, Haines DE. Effects of radiofrequency catheter ablation on regional myocardial blood flow. Possible mechanism for late electrophysiological outcome. Circulation 1994;89: