- Email: [email protected]
Ablation of epicardial ventricular arrhythmias from nonepicardial sites
Ablation of epicardial ventricular arrhythmias from nonepicardial sites Miki Yokokawa, MD, Rakesh Latchamsetty, MD, Eric Good, DO, Aman Chugh, MD, Frank Pelosi Jr, MD Thomas Crawford, MD, Krit Jongnarangsin, MD, Hakan Oral, MD, Fred Morady, MD, Frank Bogun, MD From the Division of Cardiology, University of Michigan Health System, Ann Arbor, Michigan. BACKGROUND Idiopathic epicardial ventricular arrhythmias can be targeted from the coronary venous system or the pericardial space, the endocardium, or the aortic sinus cusps. OBJECTIVE The purpose of this study was to analyze systematically the contribution of ablation at sites other than the epicardium to eliminate an arrhythmia originating in the epicardium. METHODS In a consecutive patient series of 33 patients (14 women, age 51 ⫾ 14 years, ejection fraction 51% ⫾ 9%) with epicardial ventricular arrhythmias, mapping and ablation was performed via the cardiac venous system/pericardial space, the aortic sinus cusp, and the left ventricular endocardium. An arrhythmia was defined as epicardial if the earliest onset of activation and a matching pace-map (ⱖ10/12 leads) were identified in the epicardium. RESULTS In 12/33 patients (36%), either an endocardial approach alone (n ⫽ 3) or a combined endocardial/epicardial (n ⫽ 6), cusp/endocardial (n ⫽ 1), or cusp/epicardial (n ⫽ 2) approach
Introduction Idiopathic epicardial ventricular arrhythmias can be targeted from the coronary venous system,1,2 the epicardial space,3 or the endocardium or the aortic sinus cusps. The contribution of ablation at sites other than the epicardium has not been systematically analyzed.
Methods Patient characteristics The patients in this study consisted of 33 consecutive patients (14 women, age 51 ⫾ 14 years, ejection fraction 51% ⫾ 9%) with frequent premature ventricular complexes (PVCs; n ⫽ 22) or ventricular tachycardia (VT; n ⫽ 11, mean cycle length 400 ⫾ 72 ms) who were referred for catheter ablation and whose arrhythmia was mapped to the epicardium (Table 1). Among the 33 study patients, 22 patients (67%) had a history of palpitations, and 5 (15%) had syncope. Left ventricular function was impaired in 11 patients (33%), all of whom had a high PVC burden. Prior to the Drs. Bogun and Oral were supported by a grant from the Leducq Foundation. Address reprint requests and correspondence: Dr. Frank Bogun, Division of Cardiology, CVC Cardiovascular Medicine, 1500 East Medical Center Dr, SPC 5853, Ann Arbor, MI 48109-5853. E-mail address: [email protected]. (Received March 29, 2011; accepted June 14, 2011.)
was required to eliminate the ventricular arrhythmias. In 10 of 33 patients (30%), epicardial ablation alone was effective in eliminating epicardial ventricular arrhythmias. Ablation was ineffective due to failure to reach the site of origin with the ablation catheter in 5 of 33 patients (15%), the site of origin was too close to an epicardial artery or the phrenic nerve in 3 patients (6%), and power delivery was insufficient in 3 patients (9%). CONCLUSION About one-third of epicardial arrhythmias require ablation from sites other than the epicardium to eliminate the arrhythmia focus. KEYWORDS Ablation; Epicardial origin; Mapping; Ventricular arrhythmia ABBREVIATIONS PVC ⫽ premature ventricular complex; SOO ⫽ site of origin; VT ⫽ ventricular tachycardia (Heart Rhythm 2011;8:1525–1529) © 2011 Heart Rhythm Society. Published by Elsevier Inc. All rights reserved.
ablation procedure, all patients underwent echocardiography. If no contraindication was present, the patient underwent cardiac magnetic resonance imaging to assess left and right ventricular function, rule out arrhythmogenic right ventricular dysplasia, and assess for the presence of scar. The 33 subjects in this study were drawn from a series of 228 consecutive patients with idiopathic ventricular arrhythmias. Ninety-three originated in the right ventricular outflow tract, 34 originated in a papillary muscle, 17 originated in an aortic cusp, 13 originated from the left ventricular outflow tract, 11 originated at the mitral annulus, 3 originated in the pulmonary artery, 6 originated at the parahisian area, and 5 originated at the tricuspid annulus. Four were intramural, 6 came from the posterior or anterior fascicle, and the site of origin (SOO) of the ventricular arrhythmia was uncertain in 1.
Electrophysiologic study and mapping After informed consent was obtained, three multipolar electrode catheters were introduced into the right ventricle, right atrium, and His position. Programmed ventricular stimulation was performed from the right ventricle using up to four extrastimuli to assess for inducible VT. Electrograms were filtered at 50 to 500 Hz. The intracardiac electrograms and leads V1, I, II, and III were displayed on an oscilloscope and displayed at a speed of 100
1547-5271/$ -see front matter © 2011 Heart Rhythm Society. Published by Elsevier Inc. All rights reserved.
doi:10.1016/j.hrthm.2011.06.020
1526 Table 1
Heart Rhythm, Vol 8, No 10, October 2011 Characteristics of patients with ventricular arrhythmias with an epicardial origin
Variables
Epicardial ablation alone
Ablation from sites other than epicardium alone
Ineffective (RF catheter reaches SOO)
Number Onset of activation (ms) SOO Aortic sinus cusp LV endocardium Pace-map score SOO Aortic sinus cusp LV endocardium
10
12
3
–30 ⫾ 6 –19 ⫾ 15 –12 ⫾ 10
–29 ⫾ 7 –21 ⫾ 10 –26 ⫾ 6*†
–28 ⫾ 4 –16 ⫾ 1 –7 ⫾ 9
11 ⫾ 1 4⫾4 3⫾3
11 ⫾ 1 5⫾4 6⫾3
11 ⫾ 1 4⫾5 4⫾4
Onset of activation was measured from the earliest bipolar electrogram relative to the earliest onset of the QRS complex of the 12-lead ECG. LV ⫽ left ventricle; RF ⫽ radiofrequency; SOO ⫽ sight of origin. *P ⬍.05 group with epicardial ablation alone vs group with ablation from sites other than the epicardium alone. †P ⬍.05 group with ablation from sites other than the epicardium alone vs ineffective group.
mm/s. The recordings were stored on optical disk (EPMed Systems, West Berlin, NJ, USA). An electroanatomic mapping system (CARTO, Biosense Webster, Diamond Bar, CA, USA) was used to guide mapping. Activation mapping was performed during ventricular ectopy or VT. Pace-mapping was used at the earliest site of activation. Pace-maps were classified as matching (ⱖ10/12 leads) or not matching (Figure 1). A site was defined as epicardial if the earliest onset of activation was in the epicardium and if there was a matching pace-map at this site. The epicardium was accessed either via the coronary venous system (n ⫽ 33) or via both the coronary venous system and the pericardial space (accessed via percutaneous subxyphoid puncture as described by Sosa et al4; n ⫽ 4). Once the SOO was identified in the epicardium, the aortic sinus cusps and the left ventricular endocardium closest to the epicardial SOO were mapped with activation mapping (Figure 2) and pacemapping. Radiofrequency energy first was delivered at
the site of earliest activation, followed by sites with later activation if necessary.
Radiofrequency ablation Radiofrequency energy was delivered via a 3.5-mm irrigatedtip catheter (ThermoCool, NaviStar, Biosense Webster) at sites with earliest epicardial activation and/or matching pace-maps. If the epicardial SOO could not be reached with the ablation catheter or if the ventricular arrhythmia persisted after epicardial ablation, the earliest site in the endocardium or the aortic cusp was targeted with radiofrequency ablation. Radiofrequency energy was delivered at a power of 15 to 50 W and maximal temperature of 45°C targeting an impedance drop of 10 ⍀. Programmed ventricular stimulation with and without isoproterenol infusion was repeated at the end of the procedure. The onset of the rapid component of the bipolar electrogram was used to measure the local activation time at the SOO relative to the earliest onset of the QRS complex. If the earliest onset of activation was recorded in the epicardium, radiofre-
Figure 1 Left: Pace-map of an adjacent endocardial site where onset of activation was within 10 ms of the epicardial activation time and where radiofrequency ablation was performed. Middle: QRS morphology of a ventricular tachycardia (VT). Right: Pace-map obtained from the site of origin in the mid great cardiac vein. There is a 11/12 pace-map compared to the VT (middle panel).
Yokokawa et al
Epicardial Mapping and Ablation
1527
Figure 2 Left: Twelve-lead ECG tracings of an epicardial premature ventricular complex with an onset of activation of ⫺30 ms within the great cardiac vein. Right: Twelve-lead ECG tracings from the left ventricular endocardium where the onset of activation was ⫺20 ms.
quency energy was delivered in the epicardium via the coronary venous system or via the pericardial space after percutaneous subxyphoid access was obtained. Prior to the ablation, the coronary arteries were injected to assess the distance of the ablation catheter from the epicardial coronary arteries. If the SOO could not be reached with the ablation catheter, an attempt of radiofrequency energy delivery was made at sites other than the epicardium. This was the case for three patients in all of whom the earliest activity was within 10 ms of epicardial activation time.
Data analysis ECG characteristics were assessed and the maximal deflection index was determined.1 Endocardial bipolar onset of activation and pace-maps were assessed at the SOO. The SOO was defined as the site where the earliest onset of activation and a matching pace-map were present. If ablation at this location was ineffective or was not feasible, mapping was performed in the aortic cusps and the left ventricular endocardium. Pace-mapping was performed at the site of earliest endocardial or cusp activation. The distance between the epicardial SOO and the second earliest activation site that was not epicardial was measured (Figure 3).
Follow-up The patients were seen 3 months postablation and followed for 12 to 48 months. Echocardiography was repeated within 3 months postablation if the ejection fraction was abnormal at baseline. A 24-hour Holter monitor was repeated 3 months after ablation.
Statistical analysis Continuous variables were expressed as mean ⫾ 1 SD and were compared with the Student t test. Discrete
variables were compared by 2 analysis or with Fisher exact test. P ⬍.05 was considered significant.
Results Patient characteristics Patients with epicardial arrhythmias had a mean PVC burden of 23% ⫾ 14%. In addition to the PVCs, 11 patients (33%) had nonsustained VT, and 5 (15%) had sustained VT (mean cycle length 388 ⫾ 77 ms). Eleven patients (33%) had impaired left ventricular function prior to the procedure. Their mean ejection fraction was 41% ⫾ 6%.
ECG characteristics The PVCs/VT had a right bundle branch block morphology in 24 patients (73%) and a left bundle branch block morphology in 9 (27%). Thirty-two patients (97%) had an inferior axis, and 1 (3%) with the SOO in the middle cardiac vein had a superior axis morphology. The precordial maximum deflection index was 0.52 ⫾ 0.06 (range 0.43–0.69).
Mapping and ablation The SOO was identified with activation mapping in 33 of 33 patients (Figure 4). In 12 o f33 patients (36%) with ablation from sites other than the epicardium alone, a site distant from the coronary venous system was within the 10-ms isochrone of the epicardial onset of activation that was determined within the coronary venous system. This site was located in the left ventricular endocardium (n ⫽ 10) or the aortic sinus cusp (n ⫽ 2). The onset of activation was similarly early at the SOO compared to the second earliest site that was not located in the epicardium (⫺29 ⫾ 7 vs ⫺24 ⫾ 7 ms; P ⫽ .06). The mean voltage at the effective
1528
Heart Rhythm, Vol 8, No 10, October 2011
Figure 3 Electroanatomic map showing the epicardial site of origin (SOO) of a ventricular arrhythmia (arrow) and displaying early activation of the left ventricular endocardium. GCV ⫽ great cardiac vein; SOO ⫽ site of origin.
endocardial sites was 1.29 ⫾ 0.46 mV. Radiofrequency energy was applied in the epicardium in 9 of 12 patients. However, this failed to eliminate the ventricular arrhythmias, so radiofrequency energy was delivered also at the endocardium (n ⫽ 7) or in an aortic sinus cusp (n ⫽ 2). This resulted in elimination of the ventricular arrhythmias. Radiofrequency energy was applied in the endocardium after the ablation catheter failed to reach the SOO in the epicardium in 3 of 12 patients that eliminated the ventricular arrhythmias. The mean distance between the SOO and the site of earliest activation other than the epicardium was 10 ⫾ 2 mm. In 10 of 33 patients (30%), the ventricular arrhythmias were eliminated by delivery of radiofrequency energy in the epicardium (via the coronary venous system in all 10 patients).
In 11 patients, ablation of the epicardial focus was ineffective. In 3 patients in whom the SOO could be reached with the mapping catheter, there was inadequate power delivery at the SOO. The endocardial or aortic sinus cusp site closest to the SOO was at a distance of 26 ⫾ 6 mm. In 5 patients, the SOO could not be reached from within the coronary venous system, and in 3 patients it was too close to a coronary artery (n ⫽ 2) or too close to the phrenic nerve for safe ablation (n ⫽ 1). In 4 patients the pericardial space was accessed by a subxyphoid puncture. The SOO was ⬍5 mm of the left anterior descending artery in 1 patient and below a layer of epicardial fat of ⱖ1 cm in 2 patients. Both patients had no evidence of scar tissue by magnetic resonance imaging. In 1 patient, ablation in the epicardium and the aortic cusp was required to eliminate the VT.
Figure 4 Three-dimensional reconstruction of a cardiac computed tomogram indicating the coronary venous system, including the middle cardiac vein (MCV), the coronary sinus (CS), proximal and distal great cardiac vein (GCVp and GCVd), and the coronary arteries—left anterior descending artery (LAD) and circumflex coronary artery (CX). Open circles indicate the site of origin of epicardial ventricular arrhythmias. The sites targeted via a pericardial approach with subxyphoid puncture are indicated by filled circles. Effective sites were encircled in black, and ineffective sites are encircled in white.
Yokokawa et al
Epicardial Mapping and Ablation
Mean ablation time was 16 ⫾ 12 minutes, and mean procedural time was 307 ⫾ 80 minutes. Coronary angiograms were performed before and after ablation in all patients in whom radiofrequency energy was applied in the coronary venous system or the aortic sinus cusp. All coronary arteries remained patent after the ablation procedure. No complications occurred.
Follow-up One patient with an effective epicardial-alone ablation procedure had recurrent palpitations due to PVCs 2 weeks after the procedure. In 2 of 12 patients in whom ablation elsewhere than the epicardium was performed had recurrent palpitations due to reappearance of PVCs 5 and 3 months after the initial procedure. A repeat ablation procedure was carried out from the endocardium in one patient and from the epicardium via the great cardiac vein in the other patient, resulting in elimination of the targeted arrhythmias. None of the other patients with effective ablations had recurrent ventricular arrhythmias. In the patients with a successful procedure, the mean PVC burden decreased from 22% ⫾ 13% to 0.4% ⫾ 0.9% (P ⬍.0001). All 5 sustained VTs were effectively ablated. Ejection fraction improved from 40% ⫾ 6% to 52% ⫾ 3% (P ⫽ .02) in the patients with an effective procedure and abnormal left ventricular function. In the 11 patients with failed procedures, the PVC burden changed from 24% ⫾ 15% to 11% ⫾ 8% (P ⫽ .04) with antiarrhythmic medications.
Discussion Main findings In this series of patients with epicardial ventricular arrhythmias without structural heart disease, about one-third of the patients had a successful outcome with ablation at a site other than the epicardium alone. Early timing at the left ventricular endocardium or in an aortic cusp indicated that ablation at sites other than the epicardium probably would be required to eliminate an epicardial focus.
Mapping of areas adjacent to the epicardial focus The presence of a similarly early onset of activation close to the epicardial SOO was associated with a closer proximity to the adjacent endocardium or aortic sinus cusp. However, the pace-maps at these sites were quite different, indicating that these sites were not the SOO of the targeted arrhythmia. These findings are suggestive of preferential activation5 from the epicardium to adjacent areas such as the endocardium or the aortic sinus cusps. Anatomic proximity appears to enable adjacent sites to be activated almost simultaneously to the SOO. Fiber orientation relative to adjacent sites may facilitate preferential activation to more remote sites.
1529
Origin of epicardial ventricular arrhythmias The majority of the arrhythmias mapped in this series were located in close vicinity to the coronary venous system from within which they often were effectively ablated. Therefore, a subxyphoid approach is not necessary in every patient with an arrhythmia originating from the epicardium. Although we postulated an epicardial focus to be present in the patients described in this study, a deeper SOO of the ventricular arrhythmia cannot be excluded. It is possible that this focus can be reached only by a combination of ablation from the epicardium and another site. Radiofrequency ablation with an irrigated-tip catheter from the endocardium or the aortic cusp may help to eliminate the focus if radiofrequency energy delivery at an epicardial SOO is unsuccessful. Unfortunately, because the epicardial fat is most pronounced in the AV groove, mapping there is likely to fail, especially if the thickness of epicardial fat is ⱖ1 cm.6
Study limitations It is possible that patients in whom an epicardial arrhythmia was targeted from sites other than the epicardium had an intramural focus that required ablation from adjacent locations. However, the fact that a matching pace-map and the earliest onset of activation were present at the epicardium strongly suggests that the SOO was epicardial in the patients in this study.
Conclusion In approximately one-third of patients with idiopathic ventricular arrhythmias, a successful outcome is achieved by ablation at a nearby endocardial site or by epicardial ablation in combination with ablation at the endocardium or in an aortic cusp.
Acknowledgment We thank Dr. Benoit Desjardins (University of Pennsylvania) for providing us with the computed tomographic images that are the basis for Figure 4.
References 1.
2.
3.
4.
5.
6.
Daniels DV, Lu YY, Morton JB, et al. Idiopathic epicardial left ventricular tachycardia originating remote from the sinus of Valsalva: electrophysiological characteristics, catheter ablation, and identification from the 12-lead electrocardiogram. Circulation 2006;113:1659 –1666. Baman TS, Ilg KJ, Gupta SK, et al. Mapping and ablation of epicardial idiopathic ventricular arrhythmias from within the coronary venous system. Circ Arrhythm Electrophysiol 2010;3:274 –279. Schweikert RA, Saliba WI, Tomassoni G, et al. Percutaneous pericardial instrumentation for endo-epicardial mapping of previously failed ablations. Circulation 2003;108:1329 –1335. Sosa E, Scanavacca M, d’Avila A, Pilleggi F. A new technique to perform epicardial mapping in the electrophysiology laboratory. J Cardiovasc Electrophysiol 1996;7:531–536. Bogun F, Taj M, Ting M, et al. Spatial resolution of pace mapping of idiopathic ventricular tachycardia/ectopy originating in the right ventricular outflow tract. Heart Rhythm 2008;5:339 –544. Desjardins B, Morady F, Bogun F. Effect of epicardial fat on electroanatomical mapping and epicardial catheter ablation. J Am Coll Cardiol 2010;56:1320 –1327.
Introduction Idiopathic epicardial ventricular arrhythmias can be targeted from the coronary venous system,1,2 the epicardial space,3 or the endocardium or the aortic sinus cusps. The contribution of ablation at sites other than the epicardium has not been systematically analyzed.
Methods Patient characteristics The patients in this study consisted of 33 consecutive patients (14 women, age 51 ⫾ 14 years, ejection fraction 51% ⫾ 9%) with frequent premature ventricular complexes (PVCs; n ⫽ 22) or ventricular tachycardia (VT; n ⫽ 11, mean cycle length 400 ⫾ 72 ms) who were referred for catheter ablation and whose arrhythmia was mapped to the epicardium (Table 1). Among the 33 study patients, 22 patients (67%) had a history of palpitations, and 5 (15%) had syncope. Left ventricular function was impaired in 11 patients (33%), all of whom had a high PVC burden. Prior to the Drs. Bogun and Oral were supported by a grant from the Leducq Foundation. Address reprint requests and correspondence: Dr. Frank Bogun, Division of Cardiology, CVC Cardiovascular Medicine, 1500 East Medical Center Dr, SPC 5853, Ann Arbor, MI 48109-5853. E-mail address: [email protected]. (Received March 29, 2011; accepted June 14, 2011.)
was required to eliminate the ventricular arrhythmias. In 10 of 33 patients (30%), epicardial ablation alone was effective in eliminating epicardial ventricular arrhythmias. Ablation was ineffective due to failure to reach the site of origin with the ablation catheter in 5 of 33 patients (15%), the site of origin was too close to an epicardial artery or the phrenic nerve in 3 patients (6%), and power delivery was insufficient in 3 patients (9%). CONCLUSION About one-third of epicardial arrhythmias require ablation from sites other than the epicardium to eliminate the arrhythmia focus. KEYWORDS Ablation; Epicardial origin; Mapping; Ventricular arrhythmia ABBREVIATIONS PVC ⫽ premature ventricular complex; SOO ⫽ site of origin; VT ⫽ ventricular tachycardia (Heart Rhythm 2011;8:1525–1529) © 2011 Heart Rhythm Society. Published by Elsevier Inc. All rights reserved.
ablation procedure, all patients underwent echocardiography. If no contraindication was present, the patient underwent cardiac magnetic resonance imaging to assess left and right ventricular function, rule out arrhythmogenic right ventricular dysplasia, and assess for the presence of scar. The 33 subjects in this study were drawn from a series of 228 consecutive patients with idiopathic ventricular arrhythmias. Ninety-three originated in the right ventricular outflow tract, 34 originated in a papillary muscle, 17 originated in an aortic cusp, 13 originated from the left ventricular outflow tract, 11 originated at the mitral annulus, 3 originated in the pulmonary artery, 6 originated at the parahisian area, and 5 originated at the tricuspid annulus. Four were intramural, 6 came from the posterior or anterior fascicle, and the site of origin (SOO) of the ventricular arrhythmia was uncertain in 1.
Electrophysiologic study and mapping After informed consent was obtained, three multipolar electrode catheters were introduced into the right ventricle, right atrium, and His position. Programmed ventricular stimulation was performed from the right ventricle using up to four extrastimuli to assess for inducible VT. Electrograms were filtered at 50 to 500 Hz. The intracardiac electrograms and leads V1, I, II, and III were displayed on an oscilloscope and displayed at a speed of 100
1547-5271/$ -see front matter © 2011 Heart Rhythm Society. Published by Elsevier Inc. All rights reserved.
doi:10.1016/j.hrthm.2011.06.020
1526 Table 1
Heart Rhythm, Vol 8, No 10, October 2011 Characteristics of patients with ventricular arrhythmias with an epicardial origin
Variables
Epicardial ablation alone
Ablation from sites other than epicardium alone
Ineffective (RF catheter reaches SOO)
Number Onset of activation (ms) SOO Aortic sinus cusp LV endocardium Pace-map score SOO Aortic sinus cusp LV endocardium
10
12
3
–30 ⫾ 6 –19 ⫾ 15 –12 ⫾ 10
–29 ⫾ 7 –21 ⫾ 10 –26 ⫾ 6*†
–28 ⫾ 4 –16 ⫾ 1 –7 ⫾ 9
11 ⫾ 1 4⫾4 3⫾3
11 ⫾ 1 5⫾4 6⫾3
11 ⫾ 1 4⫾5 4⫾4
Onset of activation was measured from the earliest bipolar electrogram relative to the earliest onset of the QRS complex of the 12-lead ECG. LV ⫽ left ventricle; RF ⫽ radiofrequency; SOO ⫽ sight of origin. *P ⬍.05 group with epicardial ablation alone vs group with ablation from sites other than the epicardium alone. †P ⬍.05 group with ablation from sites other than the epicardium alone vs ineffective group.
mm/s. The recordings were stored on optical disk (EPMed Systems, West Berlin, NJ, USA). An electroanatomic mapping system (CARTO, Biosense Webster, Diamond Bar, CA, USA) was used to guide mapping. Activation mapping was performed during ventricular ectopy or VT. Pace-mapping was used at the earliest site of activation. Pace-maps were classified as matching (ⱖ10/12 leads) or not matching (Figure 1). A site was defined as epicardial if the earliest onset of activation was in the epicardium and if there was a matching pace-map at this site. The epicardium was accessed either via the coronary venous system (n ⫽ 33) or via both the coronary venous system and the pericardial space (accessed via percutaneous subxyphoid puncture as described by Sosa et al4; n ⫽ 4). Once the SOO was identified in the epicardium, the aortic sinus cusps and the left ventricular endocardium closest to the epicardial SOO were mapped with activation mapping (Figure 2) and pacemapping. Radiofrequency energy first was delivered at
the site of earliest activation, followed by sites with later activation if necessary.
Radiofrequency ablation Radiofrequency energy was delivered via a 3.5-mm irrigatedtip catheter (ThermoCool, NaviStar, Biosense Webster) at sites with earliest epicardial activation and/or matching pace-maps. If the epicardial SOO could not be reached with the ablation catheter or if the ventricular arrhythmia persisted after epicardial ablation, the earliest site in the endocardium or the aortic cusp was targeted with radiofrequency ablation. Radiofrequency energy was delivered at a power of 15 to 50 W and maximal temperature of 45°C targeting an impedance drop of 10 ⍀. Programmed ventricular stimulation with and without isoproterenol infusion was repeated at the end of the procedure. The onset of the rapid component of the bipolar electrogram was used to measure the local activation time at the SOO relative to the earliest onset of the QRS complex. If the earliest onset of activation was recorded in the epicardium, radiofre-
Figure 1 Left: Pace-map of an adjacent endocardial site where onset of activation was within 10 ms of the epicardial activation time and where radiofrequency ablation was performed. Middle: QRS morphology of a ventricular tachycardia (VT). Right: Pace-map obtained from the site of origin in the mid great cardiac vein. There is a 11/12 pace-map compared to the VT (middle panel).
Yokokawa et al
Epicardial Mapping and Ablation
1527
Figure 2 Left: Twelve-lead ECG tracings of an epicardial premature ventricular complex with an onset of activation of ⫺30 ms within the great cardiac vein. Right: Twelve-lead ECG tracings from the left ventricular endocardium where the onset of activation was ⫺20 ms.
quency energy was delivered in the epicardium via the coronary venous system or via the pericardial space after percutaneous subxyphoid access was obtained. Prior to the ablation, the coronary arteries were injected to assess the distance of the ablation catheter from the epicardial coronary arteries. If the SOO could not be reached with the ablation catheter, an attempt of radiofrequency energy delivery was made at sites other than the epicardium. This was the case for three patients in all of whom the earliest activity was within 10 ms of epicardial activation time.
Data analysis ECG characteristics were assessed and the maximal deflection index was determined.1 Endocardial bipolar onset of activation and pace-maps were assessed at the SOO. The SOO was defined as the site where the earliest onset of activation and a matching pace-map were present. If ablation at this location was ineffective or was not feasible, mapping was performed in the aortic cusps and the left ventricular endocardium. Pace-mapping was performed at the site of earliest endocardial or cusp activation. The distance between the epicardial SOO and the second earliest activation site that was not epicardial was measured (Figure 3).
Follow-up The patients were seen 3 months postablation and followed for 12 to 48 months. Echocardiography was repeated within 3 months postablation if the ejection fraction was abnormal at baseline. A 24-hour Holter monitor was repeated 3 months after ablation.
Statistical analysis Continuous variables were expressed as mean ⫾ 1 SD and were compared with the Student t test. Discrete
variables were compared by 2 analysis or with Fisher exact test. P ⬍.05 was considered significant.
Results Patient characteristics Patients with epicardial arrhythmias had a mean PVC burden of 23% ⫾ 14%. In addition to the PVCs, 11 patients (33%) had nonsustained VT, and 5 (15%) had sustained VT (mean cycle length 388 ⫾ 77 ms). Eleven patients (33%) had impaired left ventricular function prior to the procedure. Their mean ejection fraction was 41% ⫾ 6%.
ECG characteristics The PVCs/VT had a right bundle branch block morphology in 24 patients (73%) and a left bundle branch block morphology in 9 (27%). Thirty-two patients (97%) had an inferior axis, and 1 (3%) with the SOO in the middle cardiac vein had a superior axis morphology. The precordial maximum deflection index was 0.52 ⫾ 0.06 (range 0.43–0.69).
Mapping and ablation The SOO was identified with activation mapping in 33 of 33 patients (Figure 4). In 12 o f33 patients (36%) with ablation from sites other than the epicardium alone, a site distant from the coronary venous system was within the 10-ms isochrone of the epicardial onset of activation that was determined within the coronary venous system. This site was located in the left ventricular endocardium (n ⫽ 10) or the aortic sinus cusp (n ⫽ 2). The onset of activation was similarly early at the SOO compared to the second earliest site that was not located in the epicardium (⫺29 ⫾ 7 vs ⫺24 ⫾ 7 ms; P ⫽ .06). The mean voltage at the effective
1528
Heart Rhythm, Vol 8, No 10, October 2011
Figure 3 Electroanatomic map showing the epicardial site of origin (SOO) of a ventricular arrhythmia (arrow) and displaying early activation of the left ventricular endocardium. GCV ⫽ great cardiac vein; SOO ⫽ site of origin.
endocardial sites was 1.29 ⫾ 0.46 mV. Radiofrequency energy was applied in the epicardium in 9 of 12 patients. However, this failed to eliminate the ventricular arrhythmias, so radiofrequency energy was delivered also at the endocardium (n ⫽ 7) or in an aortic sinus cusp (n ⫽ 2). This resulted in elimination of the ventricular arrhythmias. Radiofrequency energy was applied in the endocardium after the ablation catheter failed to reach the SOO in the epicardium in 3 of 12 patients that eliminated the ventricular arrhythmias. The mean distance between the SOO and the site of earliest activation other than the epicardium was 10 ⫾ 2 mm. In 10 of 33 patients (30%), the ventricular arrhythmias were eliminated by delivery of radiofrequency energy in the epicardium (via the coronary venous system in all 10 patients).
In 11 patients, ablation of the epicardial focus was ineffective. In 3 patients in whom the SOO could be reached with the mapping catheter, there was inadequate power delivery at the SOO. The endocardial or aortic sinus cusp site closest to the SOO was at a distance of 26 ⫾ 6 mm. In 5 patients, the SOO could not be reached from within the coronary venous system, and in 3 patients it was too close to a coronary artery (n ⫽ 2) or too close to the phrenic nerve for safe ablation (n ⫽ 1). In 4 patients the pericardial space was accessed by a subxyphoid puncture. The SOO was ⬍5 mm of the left anterior descending artery in 1 patient and below a layer of epicardial fat of ⱖ1 cm in 2 patients. Both patients had no evidence of scar tissue by magnetic resonance imaging. In 1 patient, ablation in the epicardium and the aortic cusp was required to eliminate the VT.
Figure 4 Three-dimensional reconstruction of a cardiac computed tomogram indicating the coronary venous system, including the middle cardiac vein (MCV), the coronary sinus (CS), proximal and distal great cardiac vein (GCVp and GCVd), and the coronary arteries—left anterior descending artery (LAD) and circumflex coronary artery (CX). Open circles indicate the site of origin of epicardial ventricular arrhythmias. The sites targeted via a pericardial approach with subxyphoid puncture are indicated by filled circles. Effective sites were encircled in black, and ineffective sites are encircled in white.
Yokokawa et al
Epicardial Mapping and Ablation
Mean ablation time was 16 ⫾ 12 minutes, and mean procedural time was 307 ⫾ 80 minutes. Coronary angiograms were performed before and after ablation in all patients in whom radiofrequency energy was applied in the coronary venous system or the aortic sinus cusp. All coronary arteries remained patent after the ablation procedure. No complications occurred.
Follow-up One patient with an effective epicardial-alone ablation procedure had recurrent palpitations due to PVCs 2 weeks after the procedure. In 2 of 12 patients in whom ablation elsewhere than the epicardium was performed had recurrent palpitations due to reappearance of PVCs 5 and 3 months after the initial procedure. A repeat ablation procedure was carried out from the endocardium in one patient and from the epicardium via the great cardiac vein in the other patient, resulting in elimination of the targeted arrhythmias. None of the other patients with effective ablations had recurrent ventricular arrhythmias. In the patients with a successful procedure, the mean PVC burden decreased from 22% ⫾ 13% to 0.4% ⫾ 0.9% (P ⬍.0001). All 5 sustained VTs were effectively ablated. Ejection fraction improved from 40% ⫾ 6% to 52% ⫾ 3% (P ⫽ .02) in the patients with an effective procedure and abnormal left ventricular function. In the 11 patients with failed procedures, the PVC burden changed from 24% ⫾ 15% to 11% ⫾ 8% (P ⫽ .04) with antiarrhythmic medications.
Discussion Main findings In this series of patients with epicardial ventricular arrhythmias without structural heart disease, about one-third of the patients had a successful outcome with ablation at a site other than the epicardium alone. Early timing at the left ventricular endocardium or in an aortic cusp indicated that ablation at sites other than the epicardium probably would be required to eliminate an epicardial focus.
Mapping of areas adjacent to the epicardial focus The presence of a similarly early onset of activation close to the epicardial SOO was associated with a closer proximity to the adjacent endocardium or aortic sinus cusp. However, the pace-maps at these sites were quite different, indicating that these sites were not the SOO of the targeted arrhythmia. These findings are suggestive of preferential activation5 from the epicardium to adjacent areas such as the endocardium or the aortic sinus cusps. Anatomic proximity appears to enable adjacent sites to be activated almost simultaneously to the SOO. Fiber orientation relative to adjacent sites may facilitate preferential activation to more remote sites.
1529
Origin of epicardial ventricular arrhythmias The majority of the arrhythmias mapped in this series were located in close vicinity to the coronary venous system from within which they often were effectively ablated. Therefore, a subxyphoid approach is not necessary in every patient with an arrhythmia originating from the epicardium. Although we postulated an epicardial focus to be present in the patients described in this study, a deeper SOO of the ventricular arrhythmia cannot be excluded. It is possible that this focus can be reached only by a combination of ablation from the epicardium and another site. Radiofrequency ablation with an irrigated-tip catheter from the endocardium or the aortic cusp may help to eliminate the focus if radiofrequency energy delivery at an epicardial SOO is unsuccessful. Unfortunately, because the epicardial fat is most pronounced in the AV groove, mapping there is likely to fail, especially if the thickness of epicardial fat is ⱖ1 cm.6
Study limitations It is possible that patients in whom an epicardial arrhythmia was targeted from sites other than the epicardium had an intramural focus that required ablation from adjacent locations. However, the fact that a matching pace-map and the earliest onset of activation were present at the epicardium strongly suggests that the SOO was epicardial in the patients in this study.
Conclusion In approximately one-third of patients with idiopathic ventricular arrhythmias, a successful outcome is achieved by ablation at a nearby endocardial site or by epicardial ablation in combination with ablation at the endocardium or in an aortic cusp.
Acknowledgment We thank Dr. Benoit Desjardins (University of Pennsylvania) for providing us with the computed tomographic images that are the basis for Figure 4.
References 1.
2.
3.
4.
5.
6.
Daniels DV, Lu YY, Morton JB, et al. Idiopathic epicardial left ventricular tachycardia originating remote from the sinus of Valsalva: electrophysiological characteristics, catheter ablation, and identification from the 12-lead electrocardiogram. Circulation 2006;113:1659 –1666. Baman TS, Ilg KJ, Gupta SK, et al. Mapping and ablation of epicardial idiopathic ventricular arrhythmias from within the coronary venous system. Circ Arrhythm Electrophysiol 2010;3:274 –279. Schweikert RA, Saliba WI, Tomassoni G, et al. Percutaneous pericardial instrumentation for endo-epicardial mapping of previously failed ablations. Circulation 2003;108:1329 –1335. Sosa E, Scanavacca M, d’Avila A, Pilleggi F. A new technique to perform epicardial mapping in the electrophysiology laboratory. J Cardiovasc Electrophysiol 1996;7:531–536. Bogun F, Taj M, Ting M, et al. Spatial resolution of pace mapping of idiopathic ventricular tachycardia/ectopy originating in the right ventricular outflow tract. Heart Rhythm 2008;5:339 –544. Desjardins B, Morady F, Bogun F. Effect of epicardial fat on electroanatomical mapping and epicardial catheter ablation. J Am Coll Cardiol 2010;56:1320 –1327.