Endocardial catheter ablation of postinfarction ventricular

Transcription

1 Original Article Epicardial Ablation of Ventricular Tachycardia in Ischemic Heart Disease Andrea Sarkozy, MD; Michifumi Tokuda, MD; Usha B. Tedrow, MD; Juan Sieria, MD; Gregory F. Michaud, MD; Gregory S. Couper, MD; Roy John, MD; William G. Stevenson, MD Background Epicardial approaches have increased ablation success in nonischemic cardiomyopathy, but the use for postinfarction ventricular tachycardias (VT) is less clear. We report the findings for epicardial VT ablation in postinfarct patients. Methods and Results Records of 444 consecutive patients with VT because of prior infarction referred for 600 catheter ablation procedures were reviewed. Epicardial procedures were performed in the electrophysiology laboratory in 56 (13%) patients using percutaneous (43 patients) or surgical (13 patients) epicardial access. In 7 patients, epicardial ablation was performed surgically in the operating room. In the electrophysiology laboratory epicardial VT targets were identified in 38 (68%) patients and epicardial ablation abolished 1 VT in 27 patients (6% of the total study population); inducibility was not tested after ablation in 4, and VT remained inducible in 7 patients. No ablation was performed in 18 (32%) patients because of no accessible epicardial target or a complication. Major complications occurred in 8 (14%) of the 57 electrophysiology laboratory procedures. After the first procedure any VT recurred in 21 (54%) of 39 patients who had epicardial ablation compared with 164 (47%) of 347 endocardial-only ablation patients (P=0.35). Conclusions Epicardial ablation is potentially useful in 6% of the postinfarction VT population, but the number could be substantially greater because more than two thirds of patients selected for epicardial mapping after failed ablation had an epicardial VT target. Successful epicardial ablation of a VT was not predictable from infarct location or other patient characteristics. (Circ Arrhythm Electrophysiol. 2013;6: ) Key Words: catheter ablation epicardial mapping myocardial ischemia tachycardia, ventricular Endocardial catheter ablation of postinfarction ventricular tachycardia (VT) fails in 20 to 50% of patients. 1 Although myocardial ischemic lesion formation is from the subendocardium to the epicardium, and most VT circuits seem to be endocardial, epicardial VT substrate also occurs and likely contributes to failure of endocardial ablation. In the past decade percutaneous access to the pericardial space for epicardial mapping and ablation in the electrophysiology (EP) laboratory became available. 2 In nonischemic dilated cardiomyopathies epicardial VT substrate is frequently identified and epicardial ablation seems to have increased success rates. 3,4 The incidence of epicardial circuits in postinfarction VT is not clear. The purpose of our study is to investigate the frequency and characteristics of patients with VT requiring epicardial mapping and ablation in patients with ischemic heart disease referred for catheter ablation. Clinical Perspective on p 1122 Methods Patient Population Consecutive patients undergoing catheter ablation between January 1999 and June 2012 in our tertiary care center were screened retrospectively for inclusion. Patients were included if they had coronary artery disease and infarct-related sustained monomorphic VT. Data collection and analysis were done under a protocol approved by the Brigham and Women s Hospital/Partners Institutional Review Board. Epicardial Mapping and Ablation The decision to proceed with epicardial access and mapping was made individually by the treating physician. Percutaneous pericardial puncture was performed as described by Sosa et al 2 using the subxiphoid approach (online-only Data Supplement). When needed, a subxiphoid surgical window was performed as described previously by a cardiac surgeon and under general anesthesia. 5 Epicardial mapping and ablation was performed similar to endocardial mapping and ablation as previously described. 1 Briefly, first programmed electric stimulation was performed. The morphologies of the induced VTs were noted. Substrate mapping was performed over the epicardium during sinus or paced rhythm with particular Received March 29, 2013; accepted September 20, From the Cardiology Department, University Hospital of Antwerp, University of Antwerp, Antwerp, Belgium (A.S.); Cardiovascular Division, Brigham and Women s Hospital, Harvard Medical School, Boston, MA (M.T., U.B.T., G.F.M., R.J., W.G.S.); Fundacion Barrie, Galicia, Spain (J.S.); Beca de Postgrado en el extranjero, 2012, UZ Brussel, VUB, Brussels, Belgium (J.S.); and Department of Cardiac Surgery, Brigham and Women s Hospital, Harvard Medical School, Boston, MA (G.S.C.). Guest Editor for this article was Gerhard Hindricks, MD. The online-only Data Supplement is available at Correspondence to William G. Stevenson, MD, Cardiovascular Division, Brigham and Women s Hospital, 75 Francis St, Boston, MA wstevenson@partners.org 2013 American Heart Association, Inc. Circ Arrhythm Electrophysiol is available at DOI: /CIRCEP

2 1116 Circ Arrhythm Electrophysiol December 2013 focus on the infarct scar region. Areas of low voltage (<1.5 mv) were identified. Late potentials in the scar were tagged. Pace-mapping was performed and areas of long S-QRS delays and where pace-mapping matched QRS morphology of an induced VT were tagged and targeted. Areas of electrically unexcitable scar (pacing threshold >10 ma at 2-ms pulse width) were tagged. If tolerated, VT was reinduced and activation and entrainment mapping were performed. Before epicardial ablation pacing confirmed that the phrenic nerve was not in close proximity. Coronary angiography was performed before initial ablation at sites where there was the possibility of an adjacent artery supplying viable ventricular myocardium. VT was thought to originate deep to a region if earliest activation for that chamber or epicardial space showed activation at the beginning of the QRS complex with focal spread of activation away, and entrainment (if performed) indicated an outer loop, exit, or bystander site, and ablation (if performed) failed to abolish VT. With external irrigation epicardial ablation was applied in the power control mode to a maximum power of 50 W or 45 C for 60 to 90 seconds. Irrigation flow was set to 1 ml/min during mapping and between 10 and 30 ml/min during ablation. Radiofrequency current applications were applied at the target region site until it was rendered nonexcitable (output 10 ma/2 ms). In the past 2 years pericardial fluid accumulation was followed with intracardiac echocardiography. Aspiration was regularly performed via the side port of the pericardial sheath. Surgical epicardial ablation in the operating room was performed under direct vision using a cryoablation probe (CryoCath, Montreal) or radiofrequency current ablation (COBRA RF device). After ablation programmed stimulation was performed with 3 extrastimuli from the right ventricle. Isoproterenol was administered if required for initiation previously. Programmed stimulation was not performed or was limited if there was concern that it may worsen a patient s tenuous hemodynamic status. Acute success was defined as absence of inducibility of any sustained monomorphic VT with programmed electric stimulation with 3 extrastimulus. Statistics Continuous variables are expressed as mean±sd or median (with first and third quartiles). Two-sided unpaired Student t test was used to compare continuous variables. Categorical and binary variables are presented as frequencies (percentages). χ 2 and Fisher exact tests were used to compare proportions and Kaplan Meier survival analysis was used to analyze the time to VT/ventricular fibrillation recurrence, redo procedures, and death, and the log-rank test was used to detect significant differences between groups. The baseline clinical characteristics, recurrence, and mortality were analyzed by including each patient only once in either the endocardial or the epicardial group (per patient analysis). In the analysis of procedural characteristics the analysis was performed including all first and redo procedures (per procedure analysis). Procedural parameters were analyzed using generalized estimated equations methods to account for potential correlation introduced by repeated procedures in the same patient. A P value 0.05 was considered statistically significant. The IBM SPSS Statistics version 20 software was used for the statistical analysis. Results A total of 444 patients undergoing 600 procedures were included (Figure 1). For 381 patients (86%) only endocardial procedures were performed (a total of 484 procedures; Figure 1). In 63 patients (14%) 1 epicardial mapping and ablation procedure was performed. Epicardial mapping and ablation was performed in the EP laboratory in 56 patients and surgically in the operating room in 7 patients (Table 1). The 7 surgical procedures had more limited mapping, and less extensive postablation testing than for procedures in the EP laboratory, and outcomes for these patients are therefore reported separately below. Figure 1. Flow diagram of the study patients. EP indicates electrophysiology; OR, operating room; and VT, ventricular tachycardia.

3 Sarkozy et al Postinfarction VT Ablation 1117 Table 1. Baseline Clinical Characteristics Only Endocardial EP Laboratory Patients With Epicardial Access in the EP Laboratory Considering all 593 EP laboratory procedures in 443 patients (1 epicardial surgical patient had a failed endocardial ablation elsewhere), epicardial mapping was performed in 57 procedures in 56 patients. Of the 56 patients undergoing 57 epicardial access and mapping procedures in the EP laboratory (Table 1), 54 had previously failed a prior endocardial ablation procedure, 30 had failed an endocardial procedure at our center, and 24 had failed endocardial ablation at another center. In 2 patients the initial approach was epicardial rather than left ventricular (LV) endocardial because of the presence of an LV mobile thrombus in 1 and findings of coronary sinus mapping in another patient (Table I in the online-only Data Supplement). Percutaneous puncture was attempted in 10 patients with a history of coronary artery bypass (CABG) and was successful in 3. Epicardial access was achieved via percutaneous Epicardial Access EP Laboratory P Value* Surgical Ablation OR Total No. of patients Age, y 68±10 64± ±8 66±10 Men 330 (87%) 51 (91%) (100%) 388 (87%) Hypertension 196 (58%) 35 (73%) (50%) 233 (59%) Time from last MI, mo 142± ± ± ±113 NYHA class III or more 84 (25%) 8 (17%) (100%) 94 (24%) Number of vessels 2.3± ± ± ±0.9 diseased ICD in place 348 (92%) 55 (98%) (86%) 409 (92%) CRT in place 91 (24%) 16 (29%) (29%) 109 (25%) PCI in past 131 (36%) 29 (52%) (43%) 163 (38%) CABG in past 222 (58%) 16 (29%) < (43%) 241 (54%) LVEF, % 28±11 30± ±15 29±13 LVDs, mm 52±11 52± ±7** 51±12 LVEDd, mm 62±10 63± ±9 62±10 Previous VT ablation 102 (27%) 54 (96%) <0.01 7(100%) 163 (37%) 2 Previous VT ablation 16 (4%) 28 (50%) < (43%) 47 (11%) No. of AADs 2.5± ± ±1 2.6±1.1 Amiodarone 281 (81%) 43 (86%) (100%) 328(82%) VT/ICD shock in last the 313 (95%) 54 (96%) 1 6 (86%) 373 (95%) month before ablation Electrical storm 116 (30%) 14 (25%) (14%) 131 (30%) Inferior scar 243 (65%) 38 (68%) (67%) 285 (66%) Apical scar 120 (32%) 22 (39%) (50%) 145 (33%) Anterior scar 117 (32%) 16 (29%) (50%) 136 (31%) Lateral scar 99 (27%) 16 (29%) (50%) 118 (27%) Aneurysm 66 (18%) 4 (7%) (33%) 72 (16%) AAD indicates antiarrhythmic drugs; CABG, coronary artery bypass; CRT, cardiac resynchonization threrapy device; EP, electrophysiology; ICD, implanted cardioverterdefibrillator; LVDs, left ventricular dimension at end systole; LVEDd, left ventricular end diastolic dimension; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association Functional Classification; OR, operating room; PCI, percutaneous coronary intervention; and VT, ventricular tachycardia. *The depicted P value represents the statistical comparison between the patient groups only endocardial access EP laboratory vs epicardial access EP laboratory. Given the low number of patients in the surgical ablation group this group was not included in the comparison. In case a patient underwent repeated VT ablation in our hospital only the index hospitalization was included in the analysis. Not all parameters were available in each patient; the percentages express the ratio of patients to the total number of patients in which the given parameter was known. **indicates data from 3 patients. subxiphoid puncture in 42 patients in 43 procedures, 3 (7%) of whom had prior CABG surgery (Table II in the online-only Data Supplement). In 2 of these patients mapping of the anterior LV was not possible because of adhesions. One patient with a history of prior CABG surgery had successful epicardial ablation from the middle cardiac vein, thus pericardial access was not attempted. Epicardial access was obtained via a surgical subxiphoid window in the EP laboratory in 13 (23%) patients. In 6 of these patients prior attempted percutaneous access was not successful because of adhesions (5 of whom had prior CABG). In 7 patients percutaneous epicardial access was not attempted because of a history of prior CABG. The characteristics of the procedures with epicardial access are compared with the procedures with only endocardial access in Table 2. The only significant difference comparing the procedural characteristics of the EP laboratory epicardial procedures between patients with a history of CABG with

4 1118 Circ Arrhythm Electrophysiol December 2013 Table 2. Characteristics of Procedures in the EP Laboratory With Epicardial Access Compared With Only Endocardial Access Only Endocardial Access (%)* Epicardial Access (%)* P Value No. of procedures No. of VTs induced 2.8± ±1.4 <0.01 No. of VTs targeted 2.5± ±1.3 <0.01 VT with CL >400 ms 319 (60%) 39 (70%) 0.44 VT with CL <300 ms 161 (30%) 15 (27%) 0.71 Only mappable VT 108 (20%) 14 (25%) 0.15 Both mappable and not-mappable VTs 236 (44%) 23 (41%) 0.03 Procedure time, min/procedure 241±73 284± Fluoroscopy time, min/procedure 43±27 45± RF application time, min/procedure 38±22 27±20 <0.01 Acute success (no monomorphic VT 303 (56%) 28 (49%) 0.26 inducible) Major complications 51 (10%) 8 (14%) 0.09 Tamponade or major bleeding 4 3 Incessant therapy-refractory VT 1 2 Fatal pulmonary embolization 0 1 Retroperitoneal hematoma 2 0 Vascular access complications 19 2 Stroke/transient ischemic attack 5 0 Minor pericardial bleeding (<200 ml) NA 4 CL indicates cycle length; EP, electrophysiology; RF, radiofrequency current; and VT, ventricular tachycardia. *Not all parameters were available in each patient; the percentages express the ratio of procedures to the total number of procedures in which the given parameter was known. patients without a history of CABG was shorter fluoroscopy time (Table III in the online-only Data Supplement). A re-entry circuit was identified and rendered no longer inducible by epicardial ablation in 27 patients (6% of the total study population; Figure 2). A total of 39 epicardial VTs were successfully ablated in these patients (Table IV in the online-only Data Supplement). In 13 patients entrainment mapping established the ablation target. In 14 patients ablation was guided by substrate mapping with pace mapping (Figure I in the online-only Data Supplement). After successful ablation of epicardial VT, endocardial VTs with new morphologies remained inducible and were successfully ablated from the endocardium in 8 (30%) of 27 patients. In 5 patients the results of the epicardial ablation were considered indeterminate because VT was not inducible before ablation and ablation was guided by pace-mapping and sinus rhythm electrograms (4 patients) or programmed stimulation was not performed after ablation (1 patient). One patient underwent 2 epicardial procedures. At the first procedure VT was not inducible and substrate-guided ablation was performed. VT recurred 1 week later and at a second combined endocardial and epicardial procedure an epicardial VT was successfully ablated. Of the 39 patients who received epicardial ablation lesions, programmed stimulation was performed at the end of the procedure in 38. No sustained monomorphic VT of any morphology was inducible in 23 (59%) patients. In 25 patients (45%) the epicardial procedure failed. In 7 patients epicardial ablation at the best target region failed to abolish VT, including 1 with adjacent phrenic nerve capture limiting the ablation region, and 1 with premature termination of the procedure because of loss of epicardial access. In 2 of these patients repeated endocardial mapping during the same procedure, informed by the epicardial mapping, identified a region where endocardial ablation was successful. In 1 additional patient a repeat endocardial ablation 5 days later was successful. In 2 patients additional endocardial ablation attempts failed. One of these patients refused open heart surgery and died 2 days later because of refractory VT. Another patient continued antiarrhythmic drugs with intermittent episodes of VT. In the remaining 18 patients (32%) epicardial ablation was not performed after epicardial mapping. Two patients had a coronary artery over the target site; in 15 (27%) patients an adequate epicardial ablation target was not identified (Figure II in the online-only Data Supplement); and in 1 patient a right ventricular laceration required urgent surgical repair. Considering the 57 epicardial procedures performed in the EP laboratory major complications occurred in 8 (14%) patients (Table 2). Two patients experienced recurrent uncontrollable VT during the procedure. Both had poor ventricular function, electrical storm, or incessant VT that became uncontrollable with hemodynamic deterioration requiring percutaneous LV assist device. One died the following day, the other stabilized and died 2 years later. Two patients had large femoral hematomas, 1 requiring transfusion, and 1 patient had a fatal pulmonary embolism the day after the procedure. Major pericardial bleeding >200 ml occurred in 3 cases, 1 requiring emergent surgical repair, as noted above. Pericardial bleeding <200 ml that stopped spontaneously occurred in an additional 3 patients (5%).

5 Sarkozy et al Postinfarction VT Ablation 1119 Figure 2. Epicardial ablation in a patient with old inferior wall infarction, coronary artery bypass, and failed endocardial ablation. A, Twelve lead ECGs of 2 morphologies of induced ventricular tachycardia (VT) are shown at left; VT-1 at the top (cycle length 370 ms) was consistent with the clinically documented VT. Bipolar endocardial voltage map at the right viewed from the inferior aspect shows a large lowvoltage scar (voltage <1.5 mv is color coded from red to blue). Endocardial radiofrequency current (RF) lesions were placed at the exit site (inferoseptal, arrow) as well as the entrance site (apicolateral aspect of the scar, arrow) of VT-1 but failed to terminate VT. B, Shown are voltage map (left) and activation sequence map (right) obtained during epicardial mapping through a subxiphoid surgical window. Mapping during VT-1 revealed a short-pacing interval at the inferior apical region of the left ventricle 3 cm from the apex and slightly left of the septum. In addition, pace mapping from this site produced QRS morphology similar to the VT, and pacing did not produce diaphragmatic stimulation. Bipolar voltage map of this region (left) shows a small area of low-voltage (<1.5 mv) epicardial scar. Salineirrigated RF ablation at this region (arrow on right ECG) terminated VT and rendered it no longer inducible. The patient remained free of VT during the following 4 years. LAT indicates local activation time. Surgical Intraoperative Ablation Procedures In 7 patients (66±8 years; 7 men) surgical epicardial cryoablation was performed in the operating room after median sternotomy (5 patients) or lateral thoracotomy (2 patients). All had previous failed endocardial ablation and 3 had failed attempted percutaneous epicardial access. Concomitant cardiac surgery was performed in 5 patients; CABG, CABG and aneurysmectomy, CABG with aortic root and valve replacement, ventricular assist device implantation and redo mitral valve replacement, and aneurysmectomy because of endocarditis each in a single patient, respectively. Epicardial ablation was performed under direct vision, empirically targeting the area of epicardial scar identified visually and corresponding to the previously identified infarct region. Programmed stimulation was performed after rewarming in 3 patients with no inducible VT in all. There were 2 major complications: 1 transient

6 1120 Circ Arrhythm Electrophysiol December 2013 coronary occlusion with ST elevation infarction but spontaneous resolution of the thrombus after abciximab administration and 1 postoperative sepsis. Follow-Up Follow-up for the 45 patients who had epicardial ablation (38 patients who had epicardial ablation lesions placed in the EP laboratory and 7 patients who underwent epicardial ablation in the operating room) was compared to that for the 399 patients in whom only endocardial ablation was performed. During a median follow-up of 26.5 months (25th 75th percentile; months) death occurred in 159 (40%) of the endocardial ablation group and 16 (36%) of the epicardial ablation group (P=0.5; Figure 3). Repeat ablation was performed after 8 (17%) of the 46 procedures in the epicardial group and 147 (27%) of 554 procedures in the endocardial ablation group. Kaplan Meier analysis showed no significant difference in redo procedure free survival between patients with epicardial versus endocardial ablation (P=0.49). In the epicardial ablation group the median time to the repeat procedure was 200 days (25th 75th percentile; days). A combined endocardial and epicardial ablation procedure was performed in 1 and an endocardial ablation in 7, all achieving no inducible VT after ablation. However, 2 patients had early recurrence of VT requiring another endocardial procedure 5 and 61 days later, again with complete success acutely. In 58 patients, 6 (13%) of 45 patients in the epicardial and 52 (13%) of 399 endocardialonly group patients detailed information on VT recurrence from device interrogations after the first procedure was not available. Of those with detailed follow-up, any VT recurred after the first ablation procedure in 21 (54%) epicardial and in 164 (47%) endocardial-only ablation patients. Kaplan Meier analysis showed no significant difference in ventricular Figure 3. Kaplan Meier analysis of survival comparing patients after endocardial-only ablation with patients after epicardial ablation. arrhythmia free survival rates after the first ablation between patients undergoing only endocardial versus epicardial ablation (P=0.35; Figure III in the online-only Data Supplement). In the epicardial group the median time to first recurrence of VT was 8 days (25th 75th percentile; days). After the last procedure VT/ventricular fibrillation occurred in 15 (42%) of the 36 patients with detailed follow-up in the epicardial ablation and in 114 (34%) of the 334 patients in the endocardial-only ablation group. Kaplan Meier analysis showed no significant difference in ventricular arrhythmia free survival rates after the last ablation between patients undergoing only endocardial versus epicardial ablation (P=0.2; Figure IV in the online-only Data Supplement). In the group of patients undergoing only endocardial ablation, the absence of any inducible VT after ablation was correlated with VT-free survival (P<0.01). However, in the epicardial ablation group, the absence of inducible VT was not correlated with VT-free survival (P=0.66). The presence of only mappable VT, VT with a cycle length >400 ms, inferior scar location, hypertension, VT termination with epicardial ablation, or ejection fraction ( 25%) was not correlated with outcome. Considering only the 27 patients for whom an epicardial VT was identified and ablated in the EP laboratory, any VT recurred in 13 (52%) and 2 were lost to follow-up. Redo endocardial ablation was performed in 5 of the 27 patients during follow-up all achieving no inducible VT after ablation. Characteristics of Patients With Epicardial VTs Other than less prior cardiac surgery, and more prior failed ablations, patients who had an epicardial VT identified were similar to those who had an acutely successful endocardial procedure consistent with an endocardial VT (Table 3). Infarct locations were not statistically different between the epicardial and endocardial ablation group. Discussion The incidence of VTs that can be more easily ablated from the epicardium rather than the endocardium in patients with coronary artery disease is unclear. Two decades ago Svenson et al 6 reported a series of 30 patients with recurrent ischemic VT referred for surgical ablation, in whom 33% of patients had 1 epicardial VT that was ablated with laser photocoagulation. An LV aneurysm was less frequent among patients with epicardial circuits and 90% of patients had inferior or posterior infarcts. In 2000, Sosa et al 7 reported percutaneous epicardial mapping in 14 patients with recurrent VT after inferior infarction. An epicardial circuit was identified in 7 of 30 VTs (23%) in 7 of the 14 patients. Di Biase et al 8 performed epicardial and endocardial mapping with extensive endocardial ablation over low-voltage regions in 43 patients with coronary artery disease, without prior CABG, who presented with electrical storm. Additional epicardial ablation was performed in the 14 (33%) patients who had delayed, fragmented, or low-voltage potentials in the epicardium, but the contribution of epicardial ablation to outcome was not clear. In a multicenter study of 218 patients undergoing epicardial ablation 85 patients had VT because of ischemic heart disease, 9 67% of whom had failed endocardial ablation. Epicardial ablation was performed in

7 Sarkozy et al Postinfarction VT Ablation 1121 Table 3. Comparison of Patients Whose Clinical VT Was Terminated or Rendered Noninducible After Ablation From Epicardium vs From Only Endocardium combination with endocardial ablation in 85% of the patients and was successful in 75% of the patients. In the current study, epicardial mapping was performed in 13% of a consecutive series of patients referred for catheter ablation for VT because of coronary artery disease. Ablation was performed in 68% of these patients, but in 13% of patients ablation at the best target site failed to abolish inducible VT. In 32% ablation was not performed usually because an epicardial target was not identified, despite prior failed endocardial ablation. Considering the entire population, successful epicardial ablation of a VT circuit was performed only in 6% of patients. These findings suggest that VT circuits that can be interrupted from the epicardium remain substantially less common than endocardial circuits. However, there are several limitations and sources of bias that influence these observations, including referral bias, preferential selection of patients with prior failed endocardial ablation, and without prior cardiac surgery for epicardial mapping. Only 2 patients had not failed endocardial ablation. It is possible that some circuits can be Pts Successfully Ablated With Only Endocardial Ablation Pts Successfully Ablated With Epicardial Ablation P Value No. of patients Age, y 68±10 67± Men 194 (86%) 24 (89%) 1 Hypertension* 131 (60%) 16 (80%) 0.09 Time from last MI, mo 148± ± NYHA class III or more* 52 (24%) 3 (14%) 0.42 No. of vessels diseased 2.3±0.8 2± ICD in place 204 (91%) 27 (100%) 0.14 CRT in place 51 (23%) 7 (26%) 0.81 PCI in past* 78 (36%) 12 (44%) 0.53 CABG in past 133 (59%) 10 (37%) 0.04 LVEF, % 29±12 27± LVDs, mm 51±11 56± LVEDd, mm 62±11 64± Previous VT ablation 51 (23%) 25 (93%) < Previous VT ablation 7 (3%) 14 (52%) <0.01 Amiodarone, n* 171 (76%) 19 (70%) 0.79 VT/ICD shock in last the month before 189 (84%) 25 (93%) 0.66 ablation* Electric storm* 72 (32%) 4 (15%) 0.17 Inferior scar* 142 (65%) 21 (78%) 0.2 Apical scar* 69 (32%) 9 (33%) 0.83 Anterior scar* 65 (30%) 6 (22%) 0.5 Lateral scar* 58 (27%) 8 (30%) 0.82 Aneurysm* 30 (14%) 1 (4%) 0.22 CABG indicates coronary artery bypass; CRT, cardiac resynchonization threrapy device; ICD, implanted cardioverterdefibrillator; LVDs, left ventricular dimension at end systole; LVEDd, left ventricular end diastolic dimension; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association Functional Classification; PCI, percutaneous coronary intervention; and VT, ventricular tachycardia. *Not all parameters were available in each patient; the percentages express the ratio of patients to the total number of patients in which the given parameter was known. interrupted from either the epicardium or the endocardium. Some patients have both endocardial and epicardial part of the circuits, but if endocardial ablation reduces the frequency of VT, further ablation attempts may not be considered. Prior bypass surgery had been performed in 54% of the total population and the anticipated difficulty with percutaneous puncture influences the decision to attempt epicardial mapping. Considering only patients without a history of prior surgery, we performed epicardial mapping in 15%. Other observations are of interest. Even if an epicardial VT is identified, additional ablation on the endocardium is often required for other VT circuits. Evidence for additional endocardial circuits was seen in 50% of our patients in whom a VT was ablated successfully from the epicardium. Although prior studies suggested that epicardial circuits are more likely for inferior wall infarcts, 6,10 22% of our patients with successful epicardial ablation had anterior wall infarcts. Therefore, epicardial mapping is a reasonable option for these patients and can be the initial approach when a mobile LV thrombus precludes endocardial mapping, as in one of our patients.

8 1122 Circ Arrhythm Electrophysiol December 2013 Major complications occurred in 14% of patients undergoing epicardial mapping in the EP laboratory, similar to the 13% rate reported in a multicenter study, but included a right ventricular laceration requiring emergent surgery. 11 The acute and long-term success were similar for patients undergoing epicardial or only endocardial ablation, suggesting that subepicardial VT substrate is likely not the main reason for recurrence of VT after catheter ablation. Inadequate permanent lesion formation, suboptimal recognition of ablation targets, evolution of the substrate, and deep intramural VT circuits likely contribute. We did not identify any clinical correlate of epicardial VTs. We have recently reported that surface ECG QRS morphology of the VT also fails to distinguish epicardial from endocardial origin of VTs in ischemic heart disease. 12 Conclusions In patients referred for catheter ablation of recurrent VT because of ischemic heart disease, evidence of epicardial VT circuits was observed in 6% of patients overall, although the incidence could be substantially greater. A substantial number of patients have had prior heart surgery, increasing the difficulty of epicardial procedures although surgical approaches are options. With current techniques, efficacy and risks are acceptable. The only correlate for the need for epicardial ablation was previous failed ablation, therefore a staged approach with epicardial mapping after failed endocardial ablation is reasonable. Acknowledgments Dr Sarkozy was supported financially, in part, by the UZ Brussel VUB, Brussels, Belgium. Disclosures Dr Michaud received research support from Boston Scientific and Medtronic Inc honoraria and from Boston Scientific and St. Jude Medical Inc. The other authors report no conflicts. References 1. Stevenson WG, Wilber DJ, Natale A, Jackman WM, Marchlinski FE, Talbert T, Gonzalez MD, Worley SJ, Daoud EG, Hwang C, Schuger C, Bump TE, Jazayeri M, Tomassoni GF, Kopelman HA, Soejima K, Nakagawa H; Multicenter Thermocool VT Ablation Trial Investigators. Irrigated radiofrequency catheter ablation guided by electroanatomic mapping for recurrent ventricular tachycardia after myocardial infarction: the multicenter thermocool ventricular tachycardia ablation trial. Circulation. 2008;118: 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: Soejima K, Stevenson WG, Sapp JL, Selwyn AP, Couper Gregory, Epstein LM. Endocardial and epicardial radiofrequency ablation of ventricular tachycardia associated with dilated cardiomyopathy. J Am Coll Cardiol. 2004;43: Cano O, Hutchinson M, Lin D, Garcia F, Zado E, Bala R, Riley M, Cooper J, Dixit S, Gerstenfeld E, Callans D, Marchlinski FE. Electroanatomic substrate and ablation outcome for suspected epicardial ventricular tachycardia in left ventricular nonischemic cardiomyopathy. J Am Coll Cardiol. 2009;54: Soejima K, Couper G, Cooper JM, Sapp JL, Epstein LM, Stevenson WG. Subxiphoid surgical approach for epicardial catheter-based mapping and ablation in patients with prior cardiac surgery or difficult pericardial access. Circulation. 2004;110: Svenson RH, Littmann L, Gallagher JJ, Selle JG, Zimmern SH, Fedor JM, Colavita PG. Termination of ventricular tachycardia with epicardial laser photocoagulation: a clinical comparison with patients undergoing successful endocardial photocoagulation alone. J Am Coll Cardiol. 1990;15: Sosa E, Scanavacca M, d Avila A, Oliveira F, Ramires JA. Nonsurgical transthoracic epicardial catheter ablation to treat recurrent ventricular tachycardia occurring late after myocardial infarction. J Am Coll Cardiol. 2000;35: Di Biase L, Santangeli P, Burkhardt DJ, Bai R, Mohanty P, Carbucicchio C, Dello Russo A, Casella M, Mohanty S, Pump A, Hongo R, Beheiry S, Pelargonio G, Santarelli P, Zucchetti M, Horton R, Sanchez JE, Elayi CS, Lakkireddy D, Tondo C, Natale A. Endo-epicardial homogenization of the scar versus limited substrate ablation for the treatment of electrical storms in patients with ischemic cardiomyopathy. J Am Coll Cardiol. 2012;60: Della Bella P, Brugada J, Zeppenfeld K, Merino J, Neuzil P, Maury P, Maccabelli G, Vergara P, Baratto F, Berruezo A, Wijnmaalen AP. Epicardial ablation for ventricular tachycardia: a European multicenter study. Circ Arrhythm Electrophysiol. 2011;4: Yoshiga Y, Mathew S, Wissner E, Tilz R, Fuernkranz A, Metzner A, Rillig A, Konstantinidou M, Igarashi M, Kuck KH, Ouyang F. Correlation between substrate location and ablation strategy in patients with ventricular tachycardia late after myocardial infarction. Heart Rhythm. 2012;9: Sacher F, Roberts-Thomson K, Maury P, Tedrow U, Nault I, Steven D, Hocini M, Koplan B, Leroux L, Derval N, Seiler J, Wright MJ, Epstein L, Haissaguerre M, Jais P, Stevenson WG. Epicardial ventricular tachycardia ablation a multicenter safety study. J Am Coll Cardiol. 2010;55: Martinek M, Stevenson WG, Inada K, Tokuda M, Tedrow UB. QRS characteristics fail to reliably identify ventricular tachycardias that require epicardial ablation in ischemic heart disease. J Cardiovasc Electrophysiol. 2012;23: CLINICAL PERSPECTIVE In the past decade percutaneous access has proven useful for ablation of ventricular tachycardia in nonischemic cardiomyopathy, but its use for ischemic heart disease is not well defined. In this single center retrospective study, epicardial mapping was performed when endocardial ablation failed or was not feasible in 13% of 444 patients and >70% of them had an epicardial target for ablation. Successful ablation of a ventricular tachycardia was performed in 6% of the total population. The presence of an epicardial ventricular tachycardia was not predicted by clinical factors. Although epicardial mapping and ablation are important in 6% of patients, it is probably not the main reason for failure of endocardial catheter ablation. A staged approach with epicardial mapping after failed endocardial ablation remains a reasonable strategy.