Arrhythmia/Electrophysiology

Transcription

1 Arrhythmia/Electrophysiology Irrigated Radiofrequency Catheter Ablation Guided by Electroanatomic Mapping for Recurrent Ventricular Tachycardia After Myocardial Infarction The Multicenter Thermocool Ventricular Tachycardia Ablation Trial William G. Stevenson, MD; David J. Wilber, MD; Andrea Natale, MD; Warren M. Jackman, MD; Francis E. Marchlinski, MD; Timothy Talbert, MD; Mario D. Gonzalez, MD; Seth J. Worley, MD; Emile G. Daoud, MD; Chun Hwang, MD; Claudio Schuger, MD; Thomas E. Bump, MD; Mohammad Jazayeri, MD; Gery F. Tomassoni, MD; Harry A. Kopelman, MD; Kyoko Soejima, MD; Hiroshi Nakagawa, MD; for the Multicenter Thermocool VT Ablation Trial Investigators Background Recurrent ventricular tachycardia (VT) is an important cause of mortality and morbidity late after myocardial infarction. With frequent use of implantable cardioverter-defibrillators, these VTs are often poorly defined and not tolerated for mapping, factors previously viewed as relative contraindications to ablation. This observational multicenter study assessed the outcome of VT ablation with a saline-irrigated catheter combined with an electroanatomic mapping system. Methods and Results Two hundred thirty-one patients (median LV ejection fraction, 0.25; heart failure in 62%) with recurrent episodes of monomorphic VT (median, 11 in the preceding 6 months) caused by prior myocardial infarction were enrolled. All inducible monomorphic VTs with a rate approximating or slower than any spontaneous VTs were targeted for ablation guided by electroanatomic mapping during sinus rhythm and/or VT. Patients were not excluded for multiple VTs (median, 3 per patient) or unmappable VT (present in 69% of patients). Ablation abolished all inducible VTs in 49% of patients. The primary end point of freedom from recurrent incessant VT or intermittent VT after 6 months of follow-up was achieved for 123 patients (53%). In 142 patients with implantable cardioverter-defibrillators before and after ablation for intermittent VT who survived 6 months, VT episodes were reduced from a median of 11.5 to 0 (P ). The 1-year mortality rate was 18%, with 72.5% of deaths attributed to ventricular arrhythmias or heart failure. The procedure mortality rate was 3%, with no strokes. Conclusions Catheter ablation is a reasonable option to reduce episodes of recurrent VT in patients with prior myocardial infarction, even when multiple and/or unmappable VTs are present. This population remains at high risk for death, warranting surveillance and further study. (Circulation. 2008;118: ) Key Words: ablation electrophysiology mapping myocardial infarction tachyarrhythmias Sustained ventricular tachycardia (VT) is an important cause of morbidity and sudden death in patients with prior myocardial infarction. Implantable cardioverter-defibrillators (ICDs) effectively terminate VT episodes, reducing sudden death, but ICD shocks reduce quality of life, and episodes of VT predict increased risk of death and heart failure despite effective ICD functioning. 1,2 To reduce VT episodes, antiarrhythmic drug therapy is administered to 18% to 30% of ICD patients with spontaneous VT. 1 3 Therapy with amiodarone or sotalol reduces VT episodes but with disappointing efficacy and side effects. 1,4 Clinical Perspective p 2782 Catheter ablation has been shown to be useful for reducing VT episodes in several single-center reports, but several Received April 24, 2008; accepted August 22, From the Brigham and Women s Hospital, Boston, Mass (W.G.S., K.S.); Loyola University Medical Center, Maywood, Ill (D.J.W.); Texas Cardiac Arrhythmia Institute, St David s Medical Center, Austin (A.N.); University of Oklahoma, Oklahoma City (W.M.J., H.N.); Hospital of the University of Pennsylvania, Philadelphia (F.E.M.); Diagnostic Cardiology Group, Parkridge Medical Center, Chattanooga, Tenn (T.T.); Penn State Heart and Vascular Institute, Hershey (M.D.G.); Lancaster Heart Foundation, Lancaster, Pa (S.J.W.); Ohio State University Medical Center, Columbus (E.G.D.); Utah Valley Medical Center, Provo (C.H.); Henry Ford Hospital, Detroit, Mich (C.S.); Christ Hospital, Oak Lawn, Ill (T.E.B.); Bellin Memorial Hospital, Green Bay, Wis (M.J.); Central Baptist Hospital, Lexington, Ky (G.F.T.); and Piedmont Heart Institute, Atlanta, Ga (H.A.K.). Dr Soejima is currently at the University of Miami School of Medicine, Miami, Fla. Guest Editor for this article was Michael E. Cain, MD. The online Data Supplement is available with this article at Correspondence to William G. Stevenson, MD, Cardiovascular Division, Brigham and Women s Hospital, Boston, MA wstevenson@partners.org 2008 American Heart Association, Inc. Circulation is available at DOI: /CIRCULATIONAHA

2 2774 Circulation December 16/23, 2008 factors that contribute to the difficulty of the procedure are recognized Most patients have several different VTs inducible in the electrophysiology laboratory, indicating a complex arrhythmia substrate with multiple potential reentry circuits. In patients with ICDs, VT is often terminated before the QRS morphology is recorded, so the morphology of the spontaneous VT is not known. VTs are often unstable for mapping because of hemodynamic intolerance, frequent changes from 1 VT to another during mapping, or an inability to reliably induce VT. Approaches that define areas of ventricular scar and potential VT sources during sinus rhythm have been developed to facilitate targeting of unstable or unmappable VTs. 5,7 9,11,13,14 These substrate mapping approaches are facilitated by electroanatomic mapping systems that allow reconstruction of ventricular anatomy and plot electrogram amplitude (voltage maps) to define low-voltage areas of scar or infarction. The potential for relatively large circuits and reentry paths located deep to the endocardium also limits ablation. By cooling the electrode-tissue interface, irrigated electrodes are able to deliver greater radiofrequency power to the tissue, produce larger lesions than radiofrequency ablation with standard solid electrodes, and have been suggested to facilitate ablation. 15,16 The Thermocool VT trial was conducted to evaluate the safety and efficacy of a radiofrequency ablation catheter with external irrigation combined with an electroanatomic mapping system for ablation of recurrent VT caused by prior myocardial infarction. Methods Patients were recruited from 18 centers (see the online Data Supplement) between February 1999 and December Entry criteria were sustained monomorphic VT requiring termination by cardioversion or antiarrhythmic drug administration with 4 episodes in the previous 6 months despite an ICD or antiarrhythmic drug therapy. Patients without ICDs were eligible after 2 episodes of sustained VT. Exclusion criteria included serum creatinine 2.5 mg/dl, left ventricular (LV) ejection fraction 0.10, mobile LV thrombus on echocardiography, absence of vascular access to the LV, disease process likely to limit survival to 12 months, New York Heart Association class IV heart failure, cardiac surgery within the past 2 months (unless VT was incessant), unstable angina, severe aortic stenosis or mitral regurgitation with a flail leaflet, pregnancy, and age 18 years. In contrast to many prior reports, patients with multiple VTs, unmappable VT, and a history of prior failed VT ablation were not excluded. Electrophysiological Study and Ablation Mapping was performed during VT or sinus rhythm with the CARTO electroanatomic mapping system (Biosense Webster, Inc, Diamond Bar, Calif) and an irrigated radiofrequency ablation catheter (NaviStar Thermocool, Biosense Webster). This 7F catheter has a 3.5-mm, 8F distal electrode with 6 pores near the tip. Normal saline is infused through a lumen and exits the pores into the vasculature, cooling the electrode tip and electrode-tissue interface. During mapping, saline is infused at 2 ml/min. Before the onset of radiofrequency application, the infusion rate was increased to 30 ml/min and continued for the duration of the radiofrequency application (up to 120 seconds). Radiofrequency power was applied at a maximum of 50 W provided that temperature recorded from the electrode remained 50 C. Starting at a lower power (eg, 30 W) and titrating upward was allowed at the investigator s discretion. Energy application was terminated immediately if temperature exceeded 50 C or an impedance increase 10 occurred. Programmed stimulation for initiation of VT used 1, 2, and then 3 extrastimuli during 2 paced cycle lengths from 2 RV sites. Systemic anticoagulation with heparin was required for LV mapping; the specific regimen was at the discretion of the investigator. Ablation targeted all clinically relevant VTs, which were defined as all sustained monomorphic VTs that had a cycle length within 20 ms of a documented spontaneous VT when the cycle length of VT was known. VTs of shorter cycle lengths were targeted at the discretion of the investigator. Baseline mapping during sinus or paced rhythm could be performed to create a voltage map and to mark areas of interest on the basis of pace mapping or electrogram characteristics. The precise selection of target sites for ablation was left to the investigator with the following guidelines. It was recommended that stable VTs be mapped and ablated during VT. Ablation sites required at least 1 of the following criteria during VT: isolated middiastolic potential during VT, entrainment with concealed fusion or a postpacing interval within 30 ms of the VT cycle length, or catheter pressure that terminates and prevents reinitiation of VT. For unmappable VTs (see below), ablation sites were required to have abnormal low-amplitude electrograms and any of the following additional criteria: any of the criteria for mappable VT; electrograms with double potentials, wide fractionated potentials, or isolated late potentials during sinus or paced rhythm; paced QRS morphology similar to target VT morphology; stimulus to QRS interval 40 ms during pace mapping; or anatomic continuity with other lesions. The end point of the ablation procedure was the absence of inducible, clinically relevant, sustained monomorphic VT with the complete stimulation protocol or absence of adequate target sites as defined above. Because recordings of all VTs before and after ablation were not reliably obtainable to establish the definition of clinically relevant VTs, data are presented for all spontaneous and inducible monomorphic VTs. Patient safety was of paramount importance, and investigators were allowed to terminate the procedure without further programmed stimulation if, in their judgment, it was in the best interest of the patient. A second procedure could be performed before hospital discharge with additional ablation for inducible or spontaneous VT. Additional Evaluation and Follow-Up A transthoracic echocardiogram was required before and after ablation and was interpreted at the individual centers. A neurologist performed an examination before and after ablation to look for signs of stroke. After ablation, it was recommended that the previously ineffective antiarrhythmic drug regimen be continued for the first 6 months, after which time drug therapy was left to the discretion of the investigator. Antiarrhythmic drug therapy could be terminated for toxicity or intolerance at any time. Anticoagulation was recommended for 3 months after ablation with either 325 mg/d aspirin or warfarin, which was recommended if ablation had been performed over an area with 3 cm between ablation sites. Patients were seen by the investigator or referring physician at 2 and 6 months after discharge with ICD interrogation (when applicable) to assess recurrences. Patients without ICDs were supplied with transtelephonic ECG monitors for the following 6 months and were instructed to transmit ECG strips once a month, in addition to symptomatic events. Vital status at 1 year was assessed by telephone call. Definitions and End Points Mappable VT is sustained, hemodynamically tolerated, and reproducibly inducible. Unmappable VT changes to another VT during attempted mapping or cannot be reproducibly initiated, or it requires termination because of hemodynamic compromise. Incessant VT is continued VT despite attempted electric or pharmacological cardioversion, so that VT is present 50% of the time for a period of 12 hours. The prespecified primary end point was defined at 6 months of follow-up as no recurrence of sustained monomorphic VT. For patients with incessant VT, success was defined as no recurrence of

3 Stevenson et al Ventricular Tachycardia Ablation 2775 incessant VT. ICD interrogations were interpreted by the investigators. One patient in whom no ablation lesions were placed withdrew and had no follow-up after hospital discharge. The study was approved by the institutional review boards of the respective participating centers. Written informed consent was obtained from all subjects. All adverse events that occurred within 7 days of ablation were evaluated by an independent Adverse Events Committee. Statistical Analysis This study was sponsored by Biosense Webster, Inc, to obtain data for Food and Drug Administration submission. It was estimated that a chronic success rate in this patient population of 50% would be acceptable and that a sample of 158 patients would provide an 80% power to detect a chronic success rate 40%. It was calculated that with a 22% rate of adverse events, 198 patients would provide 80% power to detect a rate of all major adverse events 30%. A total of 240 patients were planned to allow for additional attrition. A total of 240 patients were enrolled, but 9 were withdrawn before ablation (exclusion criteria identified after enrollment in 5, decision of the investigator not to pursue ablation because of electrophysiology laboratory equipment failure or absence of provocable VT in 4 patients), leaving 231 patients who were included in the study. Continuous variables are summarized as medians and upper and lower quartiles. Ordinal and continuous variables were compared between independent groups by use of the Wilcoxon rank-sum test. Dichotomous variables were compared by the 2 test. The significance of the differences between before and after values was assessed with the signed-rank test. Survival curves were plotted with the Kaplan Meier method. Multivariable analyses used stepwise logistic regression and Cox proportional-hazards models to assess predictors of the primary outcome and mortality rates, respectively (a 0.10 level of significance was used for variable entry and removal from the stepwise models). The following variables were included as candidates for entry into the stepwise models: age, prevalent heart failure, prevalent diabetes, prior coronary artery bypass surgery, history of atrial fibrillation, anterior myocardial infarction, LV ejection fraction, incessant VT, -blocker therapy, failed amiodarone, total number of inducible monomorphic VTs, maximum cycle length of inducible VT, presence of any mappable VT, presence of any inducible VT after ablation, and amiodarone therapy at the time of ablation. For analysis of the primary outcome, a history of prior ablation was also included. Analyses were performed with SAS version 9.1 (SAS Institute Inc, Cary, NC) by the Harvard Cardiovascular Research Institute. The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written. Results The majority of the 231 patients were male with characteristics of advanced disease, including markedly depressed ventricular function (median LV ejection fraction, 0.25), a history of congestive heart failure in 62%, and a history of atrial fibrillation in 29% (Table 1). They had a median of 11 episodes of VT in the preceding 6 months. VT was incessant in 16% of patients. Therapy with amiodarone had failed in 70%. The majority (63%) had prior inferior wall infarctions. An ICD was present in 94% of patients before ablation. A prior ablation procedure had been performed in 37% of patients. Electrophysiological Characteristics A total of 864 different VTs (median, 3 per patient) were induced (Table 2). The median cycle length of the slowest VT induced in each patient was 440 ms (upper and lower quartiles, 370 and 526 ms). The investigators designated 238 VTs as clinical VTs, 278 as nonclinical, and the remaining 348 as unknown in relation to spontaneous VT. A clinical VT was identified in 149 patients (65%). At least 1 VT was deemed mappable by the investigator in 154 patients (67%), but just 72 (31%) had only mappable VTs. The majority of patients had unmappable VTs, with 31% having only unmappable VTs and 38% having both mappable and unmappable VTs. Of the 462 VTs that were unmappable, 359 (78%) in 143 patients were not tolerated hemodynamically, and 103 were not reliably inducible or sustained without shifting to another VT morphology. Ablation lesions were not applied in 5 of the 231 patients (2%). In 2 patients, VT became incessant and uncontrollable, resulting in death during the procedure. In 1 patient, VT stopped and was not inducible. In 2 patients, the investigator chose not to ablate VT originating close to the His bundle and in the right ventricle, respectively. All 5 of these patients are included as acute and chronic failures for analysis of outcomes. Of the 226 patients in whom ablation was attempted, a single procedure was performed in 207 patients, 2 procedures were performed in 18 patients, and 3 procedures were performed in 1 patient before hospital discharge. A median of 24 radiofrequency ablation lesions was applied per procedure at a median power of 45 W with a median duration of 91 seconds per lesion (Table 2). The investigator described the ablation lesion pattern as a line along the scar border in 39%, a focal ablation in 30%, and multiple lesions through the target in 28% of patients. The 30-mL/min irrigation rate during ablation resulted in administration of 1342 ml saline (median). Median procedure duration was 315 minutes, with a fluoroscopy time of 45 minutes. Procedural Outcome After the last ablation procedure, at least 1 monomorphic VT of any morphology remained inducible in 99 patients (43%) (Table 2). No monomorphic VT was inducible in 113 patients (49%) with a complete stimulation protocol. Programmed stimulation was not done or consisted of 3 extrastimuli without inducing VT in 19 patients (8%). After ablation, inducible VT was usually faster than the initial VTs (Table 2). Overall, the median cycle length of the slowest inducible VT shortened from 440 ms (lower quartile, 364 ms; upper quartile, 550 ms) before ablation to 313 ms (lower quartile, 270 ms; upper quartile, 389 ms) after ablation (P ). Thus, immediately after ablation, either VT was not inducible or the substrate was modified to faster inducible VTs in 188 patients (81%). VT During Follow-Up The status regarding recurrent VT at 6 months was known for all patients except the 5 who did not undergo ablation after study entry (all 5 were counted as failures). The primary end point of freedom from VT or incessant VT during 6 months of follow-up was achieved for 123 patients (53%). Of the 108 ablation failures, 7 patients died within 7 days of the procedure (see below), 5 did not receive ablation lesions (see above), and the remaining 96 experienced recurrent incessant or intermittent VT. Of the 24 patients who died within 6 months, 19 had recurrent VT, and 5 died without recurrent VT.

4 2776 Circulation December 16/23, 2008 Table 1. Patient Characteristics by Primary End Point Total Success Failure Median or % 25th 75th Percentile Median or % 25th to 75th Percentile Median or % 25th to 75th Percentile P n Age, y Male, % Hypertension, % Heart failure, % Diabetes, % Stroke, % COPD, % PCI, % CABG, % History of AF, % MI to ablation, mo LV ejection fraction, % MI location, % Anterior MI Inferior MI Posterior MI Multiple MIs Prior ablation VT episodes in preceding 6 mo, n Incessant VT, % ICD, % Drug therapy ACEI or ARB, %* Blocker, % Statins, %* Failed AADs, % Amiodarone Sotalol Dofetilide Class I COPD indicates chronic obstructive lung disease; PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; AF, atrial fibrillation; MI, myocardial infarction; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; and AAD, antiarrhythmic drug. *Known for 153 patients. Incessant VT was present in 37 patients before ablation. During follow-up, 9 (24%) had recurrence of incessant VT (4 died, and VT became controllable after repeat ablation in 3 patients and with drug therapy in 2 patients). Of the 28 patients free of incessant VT, 9 had recurrence of intermittent VT that was now terminated by an ICD. Thus, 113 patients (49%) were free of any VT recurrence during follow-up. The impact of ablation on VT frequency for the 6 months before ablation and after ablation for the 142 patients (excluding those with incessant VT) who had an ICD before ablation and who completed 6 months of follow-up is shown in Figure 1. VT was reduced from a median of 11.5 episodes to a median of 0 episodes per 6 months (lower quartile, 0; upper quartile, 7) (P ). The frequency of VT was reduced by 75% in 67% of patients. An increase in the number of VT episodes was observed in 20% of patients. Changes in antiarrhythmic drug therapy are shown in Table I of the online Data Supplement. After ablation, antiarrhythmic drugs were continued without change in 55% of patients, reduced in 26% of patients, and increased in 19% of patients. At last follow-up, 28% of patients were no longer taking antiarrhythmic drug therapy (other than -adrenergic blockers). Antiarrhythmic drug therapy was reduced more often in patients with a successful outcome, 35% of whom were not taking antiarrhythmic drugs at 6 months of follow-up. Predictors of Primary Outcome: Recurrent VT As shown in Tables 1 and 2, compared with patients with a successful outcome, patients who failed the primary outcome

5 Stevenson et al Ventricular Tachycardia Ablation 2777 Table 2. Electrophysiology Study Findings Total Success Failure n Median or % 25th 75th Percentile Median or % 25th 75th Percentile Median or % 25th 75th Percentile P n Induced VTs VTs, n/patient Longest VT CL, ms Shortest VT CL, ms Mappable VT only, % Unmappable VT only, % patients Mappable and un-mappable VTs, % patients Mean CL of mappable VTs, ms Ablation RF lesions, n RF power, W RF duration, s Maximum temperature, C Fluoroscopy time, min Procedure, min Irrigation saline, L Acute outcome Any Inducible VT, n Slowest VT CL, ms Slower VT after ablation, % patients Faster VT after ablation, % patients Unchanged VT CL after ablation, % patients CL indicates cycle length; RF, radiofrequency. were older; had more heart failure, more atrial fibrillation, multiple myocardial infarction locations, and more inducible VTs; received more radiofrequency lesions; and more often had VT inducible after ablation. LV ejection fraction, a prior ablation before this trial, unmappable VTs, and VT cycle length were not different between groups. The cycle length of inducible VT after ablation, the change in cycle length produced by ablation, and the pattern of ablation lesions specified by the investigator (not shown) also were not different between the 2 groups. In multivariable analysis, incessant VT was associated with better outcomes. Increasing number of inducible VTs, a history of heart failure, and a history of atrial fibrillation were predictors of worse outcomes (Table 3). Death The vital status at 1 year was known for 225 of 231 patients; 40 patients died, and 6 were lost before completing 1 year of follow-up (including the 5 who did not undergo ablation). The 1-year actuarial mortality rate was 18% (Figure 2). Of the 40 deaths, 14 (37.5%) were attributed to ventricular arrhythmias (in hospital in 9 [see below] and out of hospital in 5 patients), and 35% were attributed to heart failure (Table 4). Compared with survivors, patients who died had markers of more severe disease, including heart failure in 86%, multiple infarct locations, and lower LV ejection fraction (Table 5). They also had more markers of difficult-to-control arrhythmias, including failure of amiodarone, use of antiarrhythmic drug combinations and class I antiarrhythmic drugs after ablation, longer procedure and fluoroscopy times, and more recurrent arrhythmias during follow-up. In multivariable analysis (Table 6), LV ejection fraction, atrial fibrillation, the total number of radiofrequency lesions, the maximum cycle length of inducible VT, and recurrent VT during follow-up were associated with death. The presence of any mappable VT was associated with a lower risk. When LV ejection fraction and recurrence of any VT were excluded from the model, a history of heart failure was a predictor of death (hazard ratio, 3.2; P 0.02), but heart failure was

6 2778 Circulation December 16/23, months pre - ablation 6 months post - ablation Median 11.5 N = 142 Median 0 P < Figure 1. Plot showing the frequency of VT during the 6 months before (blue lines) and after (red lines) ablation for 142 patients with ICDs before and after ablation who survived 6 months. Patients with incessant VT are not included. Each line represents an individual patient. Patients are arranged by recurrences during follow-up, ranging from greatest (bottom) to fewest (top). The frequency of VT episodes was reduced by 75% in 67% of patients. Upper boundaries are artificially truncated as indicated at 100 episodes. > >100 Number of VT Events correlated with recurrent VT and LV ejection fraction, so it was not included in the final model. Procedure-Related Complications Seven patients (3%) died within 7 days of the procedure (Table II of the online Data Supplement). In 6 patients, death was preceded by uncontrollable VT with progressive hypotension or cardiac arrest. Death occurred in the electrophysiology laboratory in 4 patients and following continued VT after the procedure in 2 patients. These patients had a median age of 68 years (range, 64 to 77 years), median LV ejection fraction of 0.23 (range, 0.15 to 0.29), and a median of 31 episodes (range, 4 to 88) of VT in the preceding 6 months. The remaining death occurred in a patient who suffered cardiac perforation and tamponade after sitting up abruptly during the procedure, followed by right coronary occlusion and cardiogenic shock. In addition, 1 elderly patient with heart failure and chronic lung disease developed frequent VT after ablation, had the ICD turned off, and died 45 days after ablation. There were 27 nonfatal, significant complications related to the procedure in 24 patients (7.3%) (see the Data Supplement). Heart failure occurred in 6 patients during or shortly after the procedure. These patients had a median age of 75 years (range, 58 to 78 years) and LV ejection fraction of 0.30 (range, 0.15 to 0.35); 4 had a history of heart failure. They received a median of 19 radiofrequency lesions (range, 12 to Table 3. Multivariable Analysis for Primary End Point OR 95% CI P Incessant VT Total monomorphic VTs Atrial fibrillation Heart failure OR indicates odds ratio. Variables included age, heart failure, diabetes mellitus, prior coronary bypass surgery, atrial fibrillation, anterior wall myocardial infarction, prior ablation, LV ejection fraction, incessant VT, therapy with -blocker, amiodarone therapy, number of inducible VTs, maximum cycle length of inducible VT, the presence of mappable VT, unmappable VT, and any VT induced after ablation. 44) and a median of 1500 ml (range, 877 to 2000 ml) saline through the irrigated ablation catheter. In 1 patient, echocardiography showed mild mitral regurgitation before the procedure, severe mitral regurgitation after the procedure (without clinical heart failure), and improved regurgitation on repeat imaging before hospital discharge; a change in pacing after ablation or ablation around a papillary muscle was speculated to be the cause. Complications related to vascular access (femoral hematomas or pseudoaneurysms) occurred in 4.7% of patients. No patient had a thromboembolic complication or stroke detected by neurological examination. Discussion This report is the largest study to date of catheter ablation for recurrent monomorphic VT caused by coronary artery disease. It was prospective, was analyzed according to intention to treat (including patients who did not receive ablation), and had external monitoring that may have improved the rigor for detecting end points and complications compared with some prior reports. Ablation was performed with a radiofrequency ablation electrode with open irrigation combined with an electroanatomic mapping system that facilitated substrate mapping (discussed below). Irrigated radiofrequency ablation electrodes create larger lesions than standard 4-mm electrodes, making them of particular interest for ablation of scar-related VTs, in which the reentry circuits are often relatively wide and can extend deep within the myocardium Despite a population of patients with severely depressed ventricular function and drug-refractory, frequent arrhythmias, recurrent VT was abolished in approximately half of the patients. For many of those in whom VT recurred, the frequency of episodes was substantially reduced. Successful ablation allowed the reduction or withdrawal of antiarrhythmic drugs for some patients. In contrast to many prior studies, patients with VTs that were considered unmappable were included, and 90% of patients had ICDs before ablation. This study also defines risks in this patient population. The procedure mortality rate was 3%, and the 1-year mortality rate was 18%,

7 Stevenson et al Ventricular Tachycardia Ablation 2779 Figure 2. Kaplan Meier survival curve. Dashed lines are 95% CI boundaries. with ventricular arrhythmias and heart failure accounting for 70% of deaths despite the presence of an ICD. Many studies of catheter ablation used selected patients with hemodynamically tolerated, stable VTs for which an ECG of spontaneous VT had been obtained and mapping could be performed during VT. 6,12 However, most patients with recurrent VT now have an ICD that promptly terminates VT, so its hemodynamic impact and ECG morphology are often not known. Therefore, we sought to ablate all monomorphic VTs, including unmappable VTs. The electroanatomic mapping system used allows re-creation of ventricular geometry and displays low-voltage areas of scar or infarction. 8 Mapping during stable sinus or paced rhythm to identify targets for VT has been referred to as substrate mapping. 8,11 Specific approaches targeting potential arrhythmogenic areas and channels with these systems were in evolution during the course of this study, so a standard approach could not be incorporated by protocol into this study. Investigators were allowed to target any abnormal area that exhibited 1 mapping criteria associated with reentry circuits and were requested to characterize the approach that they took. These designations did not differ between patients Table 4. Causes of Death n % Total deaths 40 Cardiac death Procedure death 7 d after procedure Late in-hospital arrhythmia Out-of-hospital cardiac arrest Heart failure Other cardiac (postsurgery, MI) Noncardiac death (pneumonia, n 2; cancer, n 2) Unknown cause MI indicates myocardial infarction. with successful and failed outcomes, and this study does not provide insight into the best method for substrate mapping. Smaller single-center series using 1 of these techniques that have evolved during the course of this study have reported lower VT recurrence rates. 7 9,11,13,18 VT recurs in 19% to 50% of patients, although the frequency is reduced in the majority. Multiple morphologies of VT and unstable VTs, as in our patient population, have been associated with increased risk of recurrence. 10,14 Major complications occurred in 10% of patients, a frequency consistent with single-center reports. The procedure mortality rate of 3% is also similar to a previous, smaller trial. 19 Early deaths (6 of 7 in the present study) are often due to uncontrollable VT. Fatal, uncontrollable VT may be the inevitable consequence of an unsuccessful procedure in high-risk patients, but a proarrhythmic effect of the procedure cannot be excluded. Reducing antiarrhythmic drugs before the procedure to facilitate initiation of VT, radiofrequency lesions, and sympathetic responses to induced VT and saline administration are factors that could potentially aggravate VT. Proarrhythmia may also have contributed to increased episodes of VT in 20% of patients during follow-up, although changes in antiarrhythmic drug therapy (implemented in 43% of these patients), disease progression, and ablation failure with spontaneous variability in episodes are also likely causes. Although strokes and other embolic events are well-known complications of ablation procedures, none were observed in this study. A neurological examination performed by a neurologist was required before and after the procedure. Cranial imaging was not performed, and it is possible that small strokes and transient events escaped detection. Nonetheless, this observation is reassuring. Cerebral or systemic embolism occurred in 2.7% of patients in a previous multicenter trial that used a closed irrigation system for radiofrequency ablation. 19 Better surface cooling and the flow provided by open irrigation compared with closed

8 2780 Circulation December 16/23, 2008 Table 5. Characteristics of Patients Surviving Versus Dying Within 1 Year Survived Died % or Median 25th 75th Percentile % or Median 25th 75th Percentile P n Age, y Male, % Hypertension, % Congestive failure, % Diabetes mellitus, % Stroke, % CABG, % Valve disease/replacement, % History of AF, % Anterior MI, % Multiple MI locations, n Prior ablation, % VT episodes in the preceding 6 mo, n LVEF, % Incessant VT, % Blocker, % Failed amiodarone, % Monomorphic VTs inducible, n Maximum CL of inducible VT, ms Any mappable VT, % RF lesions per procedure, n Fluoroscopy time, min Procedure time, min Irrigation saline, L Postablation Any inducible VT, % CL slowest inducible VT, ms Recurrent VT within 6 mo, % Warfarin, % Any AAD, % Class I AAD, % Class III AAD, % AAD combination, % Amiodarone, % Class III AAD are those excluding amiodarone. CABG indicates coronary artery bypass grafting; AF, atrial fibrillation; MI, myocardial infarction; LVEF, LV ejection fraction; CL, cycle length; and AAD, antiarrhythmic drug. irrigation have been suggested to reduce coagulum formation on the ablation catheter and tissue surface. 17 However, the frequency of embolic events in reported studies is low, and this trial does not establish that open irrigation reduces embolic events. During radiofrequency ablation, irrigated electrodes are prone to deep heating within the tissue, causing steam formation that can explode through the tissue ( steam pops ). 17,20 No cases of tamponade related to radiofrequency ablation occurred in this trial. Perforation seems to be rare during ablation in infarct regions of dense fibrosis. The risk is likely greater in structurally normal ventricles and the thin-walled atrium. External irrigation administers an intravascular volume load that can aggravate heart failure or precipitate pulmonary edema, particularly in this patient population that commonly has depressed ventricular function and heart failure. We took several precautions to manage this risk. A urinary catheter was recommended for careful attention to fluid balance and diuresis. Patients with marked renal insufficiency (serum creatinine 2.5 mg/dl) in whom the additional volume load may be difficult to manage were excluded. Despite these precautions, volume administration may have contributed to the 7 cases of pulmonary edema or respiratory insufficiency observed. The significant mortality rate of 18% at 1 year is a major concern. Annual death rates in previous series ranges from

9 Stevenson et al Ventricular Tachycardia Ablation 2781 Table 6. Multivariable Analysis for Death HR 95% CI P LV EF Any mappable VT History of AF Recurrent VT Total RF lesions Max VT CL HR indicates hazard ratio; EF, ejection fraction; AF, atrial fibrillation; RF, radiofrequency; and CL, cycle length. 5% to 20%, with death from progressive heart failure being a common cause. 19,21,22 This death rate is consistent with the severity of heart disease in our population, which had a median LV ejection fraction of 0.25, a history of heart failure in 62%, and atrial fibrillation in 29%, all markers of increased risk of death from heart failure In addition, spontaneous ventricular arrhythmias are associated with increased rates of death and heart failure even in patients with ICDs, suggesting that they are a marker for deteriorating ventricular function. 2 It is interesting that of the 24 patients who died within 6 months, 19 (79%) had failed ablation and had recurrent episodes of VT. The potential for ablation to adversely affect LV function, however, remains concerning. The association of number of radiofrequency lesions and death is of concern, although it seems likely that the number of radiofrequency lesions will be greater in patients with a larger number of VTs and in whom ablation is failing, so they are likely to have recurrent VT associated with death. Confining ablation lesions to regions of low-voltage scar that is less likely to contain contracting myocardium seems prudent. Although we obtained echocardiograms before and after ablation for the detection of gross changes, interpretation was left to the individual centers and was not blinded. An increase in mitral regurgitation was noted in 1 patient as described. Prior studies assessing LV ejection fraction after ablation have not shown deterioration. 8,26 The present findings further emphasize the importance of appropriate therapy for ventricular dysfunction in this population. Several study limitations require emphasis. The patient population is selected by referral for ablation, resulting in a group of patients with advanced heart disease. The follow-up period was short, only 6 months for assessment of recurrent arrhythmias and 12 months for death. Although continued antiarrhythmic drug therapy was recommended, drugs were often reduced after a successful ablation, as allowed for side effects. In some cases, removal of a proarrhythmic drug effect may have contributed to a beneficial outcome; in other cases, emergence of a VT that was suppressed by the drug at the time of ablation may have contributed to procedure failure. We were not able to clearly define whether a recurrent VT was related to a VT that was targeted for ablation because recurrences were usually terminated by an ICD. Conclusions Patients with recurrent VT who are failing antiarrhythmic drug therapy are a high-risk population with substantial mortality rates warranting attention to therapies for ventricular dysfunction and arrhythmias. Catheter ablation is a reasonable option to reduce episodes of recurrent VT in patients with prior myocardial infarction. Multiple and unmappable VTs can be targeted with ablation combined with electroanatomic substrate mapping with acceptable risks and outcomes. Acknowledgments We wish to express our appreciation to Brenda Aker, Robert Stagg, Chandan Vinekar, Christine Liu, Philippa Hill, Marcia Yaross, and their team at Biosense Webster, Inc, for their support of the trial. We also wish to thank Joseph Massaro, PhD, of the Harvard Cardiovascular Research Institute for statistical analysis. Source of Funding Funding for this study was provided by Biosense Webster, Inc. Disclosures Drs Stevenson, Wilber, Natale, Jackman, Marchlinski, and Nakagawa have received research grants and speaking honoraria from and served as consultants to Biosense Webster, Inc. Dr Gonzalez has served on the speakers bureau for St Jude Medical. Dr Worley has received research grants from and served as a consultant to Medtronic and Pressure Products; received honoraria from Medtronic, St Jude, Guidant, and ELA; and served on the advisory board for St Jude Medical. Drs Daoud, Tomassoni, and Kopelman have received honoraria from and served as consultants to Biosense Webster, Inc. Dr Schuger has served as a consultant to Biosense Webster. Dr Nakagawa has received research grants and speaking honoraria from and served as a consultant to St Jude Medical and received honoraria from St Jude Medical and Boston Scientific. The other authors report no conflicts. References 1. Connolly SJ, Dorian P, Roberts RS, Gent M, Bailin S, Fain ES, Thorpe K, Champagne J, Talajic M, Coutu B, Gronefeld GC, Hohnloser SH. Comparison of beta-blockers, amiodarone plus beta-blockers, or sotalol for prevention of shocks from implantable cardioverter defibrillators: the OPTIC Study: a randomized trial. JAMA. 2006;295: Moss AJ, Greenberg H, Case RB, Zareba W, Hall WJ, Brown MW, Daubert JP, McNitt S, Andrews ML, Elkin AD. Long-term clinical course of patients after termination of ventricular tachyarrhythmia by an implanted defibrillator. Circulation. 2004;110: Steinberg JS, Martins J, Sadanandan S, Goldner B, Menchavez E, Domanski M, Russo A, Tullo N, Hallstrom A. Antiarrhythmic drug use in the implantable defibrillator arm of the Antiarrhythmics Versus Implantable Defibrillators (AVID) Study. Am Heart J. 2001;142: Pacifico A, Hohnloser SH, Williams JH, Tao B, Saksena S, Henry PD, Prystowsky EN. Prevention of implantable-defibrillator shocks by treatment with sotalol: d,l-sotalol Implantable Cardioverter-Defibrillator Study Group. N Engl J Med. 1999;340: Stevenson WG, Soejima K. Catheter ablation for ventricular tachycardia. Circulation. 2007;115: Bogun F, Kim HM, Han J, Tamirissa K, Tschopp D, Reich S, Elmouchi D, Igic P, Lemola K, Good E, Oral H, Chugh A, Pelosi F, Morady F. Comparison of mapping criteria for hemodynamically tolerated, postinfarction ventricular tachycardia. Heart Rhythm. 2006;3: Kottkamp H, Wetzel U, Schirdewahn P, Dorszewski A, Gerds-Li JH, Carbucicchio C, Kobza R, Hindricks G. Catheter ablation of ventricular tachycardia in remote myocardial infarction: substrate description guiding placement of individual linear lesions targeting noninducibility. J Cardiovasc Electrophysiol. 2003;14: Marchlinski FE, Callans DJ, Gottlieb CD, Zado E. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation. 2000; 101: Arenal A, del Castillo S, Gonzalez-Torrecilla E, Atienza F, Ortiz M, Jimenez J, Puchol A, Garcia J, Almendral J. Tachycardia-related channel

10 2782 Circulation December 16/23, 2008 in the scar tissue in patients with sustained monomorphic ventricular tachycardias: influence of the voltage scar definition. Circulation. 2004; 110: Della Bella P, Riva S, Fassini G, Giraldi F, Berti M, Klersy C, Trevisi N. Incidence and significance of pleomorphism in patients with postmyocardial infarction ventricular tachycardia: acute and long-term outcome of radiofrequency catheter ablation. Eur Heart J. 2004;25: Reddy VY, Neuzil P, Taborsky M, Ruskin JN. Short-term results of substrate mapping and radiofrequency ablation of ischemic ventricular tachycardia using a saline-irrigated catheter. J Am Coll Cardiol. 2003; 41: de Chillou C, Lacroix D, Klug D, Magnin-Poull I, Marquie C, Messier M, Andronache M, Kouakam C, Sadoul N, Chen J, Aliot E, Kacet S. Isthmus characteristics of reentrant ventricular tachycardia after myocardial infarction. Circulation. 2002;105: Soejima K, Stevenson WG, Maisel WH, Sapp JL, Epstein LM. Electrically unexcitable scar mapping based on pacing threshold for identification of the reentry circuit isthmus. Circulation. 2002;106: Soejima K, Suzuki M, Maisel WH, Brunckhorst CB, Delacretaz E, Blier L, Tung S, Khan H, Stevenson WG. Catheter ablation in patients with multiple and unstable ventricular tachycardias after myocardial infarction: short ablation lines guided by reentry circuit isthmuses and sinus rhythm mapping. Circulation. 2001;104: Soejima K, Delacretaz E, Suzuki M, Brunckhorst CB, Maisel WH, Friedman PL, Stevenson WG. Saline-cooled versus standard radiofrequency catheter ablation for infarct-related ventricular tachycardias. Circulation. 2001;103: 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: Yokoyama K, Nakagawa H, Wittkampf FH, Pitha JV, Lazzara R, Jackman WM. Comparison of electrode cooling between internal and open irrigation in radiofrequency ablation lesion depth and incidence of thrombus and steam pop. Circulation. 2006;113: Arenal A, Glez-Torrecilla E, Ortiz M, Villacastin J, Fdez-Portales J, Sousa E, del Castillo S, Perez de Isla L, Jimenez J, Almendral J. Ablation of electrograms with an isolated, delayed component as treatment of unmappable monomorphic ventricular tachycardias in patients with structural heart disease. J Am Coll Cardiol. 2003;41: Calkins H, Epstein A, Packer D, Arria AM, Hummel J, Gilligan DM, Trusso J, Carlson M, Luceri R, Kopelman H, Wilber D, Wharton JM, Stevenson W. Catheter ablation of ventricular tachycardia in patients with structural heart disease using cooled radiofrequency energy: results of a prospective multicenter study. J Am Coll Cardiol. 2000;35: Cooper JM, Sapp JL, Tedrow U, Pellegrini CP, Robinson D, Epstein LM, Stevenson WG. Ablation with an internally irrigated radiofrequency catheter: learning how to avoid steam pops. Heart Rhythm. 2004;1: Nabar A, Rodriguez LM, Batra RK, Timmermans C, Cheriex E, Wellens HJ. Echocardiographic predictors of survival in patients undergoing radiofrequency ablation of postinfarct clinical ventricular tachycardia. J Cardiovasc Electrophysiol. 2002;13:S118 S Della Bella P, De Ponti R, Uriarte JA, Tondo C, Klersy C, Carbucicchio C, Storti C, Riva S, Longobardi M. Catheter ablation and antiarrhythmic drugs for haemodynamically tolerated post-infarction ventricular tachycardia: long-term outcome in relation to acute electrophysiological findings. Eur Heart J. 2002;23: Zareba W, Steinberg JS, McNitt S, Daubert JP, Piotrowicz K, Moss AJ. Implantable cardioverter-defibrillator therapy and risk of congestive heart failure or death in MADIT II patients with atrial fibrillation. Heart Rhythm. 2006;3: Solomon SD, Anavekar N, Skali H, McMurray JJ, Swedberg K, Yusuf S, Granger CB, Michelson EL, Wang D, Pocock S, Pfeffer MA. Influence of ejection fraction on cardiovascular outcomes in a broad spectrum of heart failure patients. Circulation. 2005;112: Pinski SL, Yao Q, Epstein AE, Lancaster S, Greene HL, Pacifico A, Cook JR, Jadonath R, Marinchak RA. Determinants of outcome in patients with sustained ventricular tachyarrhythmias: the Antiarrhythmics Versus Implantable Defibrillators (AVID) study registry. Am Heart J. 2000;139: Khan HH, Maisel WH, Ho C, Suzuki M, Soejima K, Solomon S, Stevenson WG. Effect of radiofrequency catheter ablation of ventricular tachycardia on left ventricular function in patients with prior myocardial infarction. J Interv Card Electrophysiol. 2002;7: CLINICAL PERSPECTIVE Although implantable cardioverter-defibrillators reduce death resulting from ventricular tachycardia (VT), episodes of VT decrease quality of life and predict increased rates of death. Antiarrhythmic drugs are often used to suppress VT but have potential adverse effects and relatively poor efficacy. In the largest study to date of catheter ablation for recurrent monomorphic VT caused by coronary artery disease, we prospectively evaluated radiofrequency catheter ablation using an irrigated catheter combined with an electroanatomic mapping system to facilitate substrate mapping during sinus rhythm. In contrast to prior studies, patients with hemodynamically unstable, unmappable VTs and multiple VTs were included because these VTs are often present in patients with ICDs. Despite a population with severely depressed ventricular function and drug-refractory, frequent VT, ablation abolished recurrent VT in approximately half of the patients. Of those in whom VT recurred, the frequency of episodes was substantially reduced for many, allowing reduction or withdrawal of antiarrhythmic drugs for some patients. The procedure mortality rate was 3%, and there were no strokes. The 1-year mortality rate was 18%, with ventricular arrhythmias and heart failure accounting for 70% of deaths. The present study demonstrates that patients with recurrent sustained VT and coronary artery disease are a high-risk population with substantial death risk despite implantable cardioverter-defibrillators. Catheter ablation is a reasonable option to reduce VT episodes, even if multiple and unmappable VTs are present.