Author + information
- Received February 12, 2016
- Revision received July 29, 2016
- Accepted August 2, 2016
- Published online November 1, 2016.
- Yalçın Gökoğlan, MDa,b,
- Sanghamitra Mohanty, MDa,c,
- Carola Gianni, MDa,d,
- Pasquale Santangeli, MDe,
- Chintan Trivedi, MD, MPHa,
- Mahmut F. Güneş, MDa,f,
- Rong Bai, MDa,g,
- Amin Al-Ahmad, MDa,
- G. Joseph Gallinghouse, MDa,
- Rodney Horton, MDa,
- Patrick M. Hranitzky, MDa,
- Javier E. Sanchez, MDa,
- Salwa Beheiry, RNh,
- Richard Hongo, MDh,
- Dhanunjaya Lakkireddy, MDi,
- Madhu Reddy, MDi,
- Robert A. Schweikert, MDj,
- Antonio Dello Russo, MDk,
- Michela Casella, MDk,
- Claudio Tondo, MDk,
- J. David Burkhardt, MDa,
- Sakis Themistoclakis, MDl,
- Luigi Di Biase, MD, PhDa,m and
- Andrea Natale, MDa,c,h,n,o,p,∗ ()
- aTexas Cardiac Arrhythmia Institute, St. David’s Medical Center, Austin, Texas
- bDepartment of Cardiology, Gülhane Military Academy of Medicine, Ankara, Turkey
- cDell Medical School, University of Texas, Austin, Texas
- dDepartment of Clinical Sciences and Community Health, University of Milan, Milan, Italy
- eElectrophysiology Section, Cardiovascular Division, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania
- fDepartment of Cardiology, Turgut Ozal University Faculty of Medicine, Alparslan, Turkey
- gBeijing Anzhen Hospital, Capital Medical University, Beijing, China
- hElectrophysiology and Arrhythmia Services, California Pacific Medical Center, San Francisco, California
- iUniversity of Kansas Hospital, Kansas City, Kansas
- jAkron General Hospital, Akron, Ohio
- kRCCS Monzino Hospital, Milan, Italy
- lOspedale dell’Angelo, Mestre/Venice, Italy
- mMontefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York
- nInterventional Electrophysiology, Scripps Clinic, La Jolla, California
- oMetro Health Medical Center, Case Western Reserve University School of Medicine, Cleveland, Ohio
- pDivision of Cardiology, Stanford University, Stanford, California
- ↵∗Reprint requests and correspondence:
Dr. Andrea Natale, Texas Cardiac Arrhythmia Institute, St. David’s Medical Center, 3000 N. IH-35, Suite 720, Austin, Texas 78705.
Background Scar homogenization improves long-term ventricular arrhythmia–free survival compared with standard limited-substrate ablation in patients with post-infarction ventricular tachycardia (VT). Whether such benefit extends to patients with nonischemic cardiomyopathy and scar-related VT is unclear.
Objectives The aim of this study was to assess the long-term efficacy of an endoepicardial scar homogenization approach compared with standard ablation in this population.
Methods Consecutive patients with dilated nonischemic cardiomyopathy (n = 93), scar-related VTs, and evidence of low-voltage regions on the basis of pre-defined criteria on electroanatomic mapping (i.e., bipolar voltage <1.5 mV) underwent either standard VT ablation (group 1 [n = 57]) or endoepicardial ablation of all abnormal potentials within the electroanatomic scar (group 2 [n = 36]). Acute procedural success was defined as noninducibility of any VT at the end of the procedure; long-term success was defined as freedom from any ventricular arrhythmia at follow-up.
Results Acute procedural success rates were 69.4% and 42.1% after scar homogenization and standard ablation, respectively (p = 0.01). During a mean follow-up period of 14 ± 2 months, single-procedure success rates were 63.9% after scar homogenization and 38.6% after standard ablation (p = 0.031). After multivariate analysis, scar homogenization and left ventricular ejection fraction were predictors of long-term success. During follow-up, the rehospitalization rate was significantly lower in the scar homogenization group (p = 0.035).
Conclusions In patients with dilated nonischemic cardiomyopathy, scar-related VT, and evidence of low-voltage regions on electroanatomic mapping, endoepicardial homogenization of the scar significantly increased freedom from any recurrent ventricular arrhythmia compared with a standard limited-substrate ablation. However, the success rate with this approach appeared to be lower than previously reported with ischemic cardiomyopathy, presumably because of the septal and midmyocardial distribution of the scar in some patients.
The occurrence of sustained monomorphic scar-related ventricular tachycardia (VT) in patients with nonischemic cardiomyopathy (NICM) is associated with an increased risk for mortality (1,2). Arrhythmia management in this setting is quite challenging. Although implantable cardioverter-defibrillators (ICDs) are lifesaving, they neither modify the arrhythmogenic substrate nor prevent recurrences. Antiarrhythmic drugs, in contrast, have suboptimal efficacy and are associated with significant adverse effects (3,4). Currently, more patients with NICM are being referred for VT ablation, which may offer a better treatment option. Data on the efficacy of VT ablation in patients with NICM are limited, with VT-free survival rates ranging between 41% and 80% (5–12). Wide disparity among outcomes is likely because of differences in sample size, patient characteristics, ablation strategy, and follow-up periods. Additional epicardial mapping and ablation significantly reduced VT recurrence in patients with NICM (5,6,13). Moreover, in scar-related VT ablation, complete elimination of all abnormal potentials within the scar was associated with better outcomes (10,14–18).
In this context, our group previously demonstrated that endoepicardial scar homogenization improved long-term arrhythmia-free survival compared with standard ablation in patients with ischemic cardiomyopathy (ICM) (15,16). Whether such benefit extends also to patients with dilated nonischemic scar-related VT is unclear. Therefore, we sought to assess the long-term efficacy of an endoepicardial scar homogenization approach compared with standard ablation for the treatment of scar-related VT in patients with NICM.
This prospective study involved consecutive patients with dilated NICM, monomorphic scar-related VT, and evidence of low voltage (bipolar voltage <1.5 mV) on electroanatomic mapping who underwent radiofrequency (RF) ablation. NICM was defined as the presence of left ventricular systolic dysfunction with left ventricular ejection fraction (LVEF) <40% in the absence of significant coronary artery disease. All patients had ICDs before the ablation procedure. On the basis of operator preference, patients underwent either standard ablation (group 1) or a scar homogenization procedure (group 2).
In most patients, systemic anticoagulation was administered after epicardial access was obtained; patients received intravenous heparin with a minimum activation clotting time of 300 s. When epicardial ablation was performed after endocardial ablation, anticoagulation with heparin was reversed with protamine before attempting pericardial access. Anticoagulation for atrial fibrillation was discontinued before the procedure.
The strategy for standard ablation and scar homogenization has been described in detail in earlier publications from our group (15). Briefly, procedures were performed in the post-absorptive state under conscious sedation. Before epicardial ablation, coronary angiography was performed at the operator’s discretion to assess the location of coronary arteries.
Endocardial mapping was performed in all patients using a 3-dimensional electroanatomic mapping system (CARTO, Biosense Webster, Diamond Bar, California). Epicardial mapping and ablation were performed in all group 2 patients, whereas in group 1, it was done only when VT was inducible after endocardial ablation or when no endocardial scar was detected.
All patients underwent bipolar substrate mapping (scar defined as voltage <1.5 mV and dense scar <0.5 mV for the endocardium; voltage <1.0 mV for the epicardium, taking into account the fact that overlying fat may attenuate voltage). Areas of fractionated or late potentials were identified as abnormal potentials. RF energy was delivered using an open irrigated 3.5-mm catheter (ThermoCool, Biosense Webster) with power up to 50 W and a temperature limit of 40°C.
In group 1, in addition to substrate mapping, activation/entrainment and pace mapping were primarily used to determine the mechanism and circuit of the VT to identify potential sites for ablation. RF ablation was performed during VT, to obtain termination, and RF energy was applied until local electrogram abatement and/or loss of capture (high-output pacing up to 20 mA with a pulse duration of 10 ms) was achieved. If the VT was hemodynamically unstable, ablation target sites were identified on the basis of substrate mapping and pace mapping.
In group 2, detailed substrate mapping was performed with identification of all abnormal electrograms within the electroanatomically defined scar. Activation mapping was eventually performed to confirm that the clinical VT was associated with the identified scar. In this group, ablation was empirically performed throughout the entire scar endocardially (scar homogenization). “Normal” electrograms were defined as electrograms with 3 or fewer sharp and discrete deflections from baseline, amplitude >1.5 mV, duration <70 ms (and/or amplitude/duration ratio >0.046) (15). Any electrograms not fitting this definition were categorized as “abnormal” and targeted for ablation (15). Epicardially, a scar was ablated if it contained at least 3 abnormal electrograms. When epicardial ablation was performed after endocardial ablation, anticoagulation with heparin was reversed with protamine before attempting the pericardial access. Figure 1 illustrates a representative sample of ablation lesions placed in patients in groups 1 and 2.
In both groups, the acute procedural endpoint was noninducibility of any monomorphic VT (except for nonclinical VTs with cycle length <200 ms, polymorphic VT, and ventricular fibrillation). At the end of the ablation, programmed stimulation (2 drive cycle length, 3 extra stimuli ≥200 ms, ≥2 sites) with and without isoproterenol (up to 5 μg/min) was performed in all but 3 patients, in whom the procedure was terminated prematurely because of hemodynamic instability, to test the inducibility of any ventricular arrhythmias (VAs).
Patients with hemodynamically unstable VT received cardiopulmonary support with percutaneous left ventricular assist devices.
All patients were hospitalized overnight with continuous rhythm monitoring and usually discharged the day after the procedure. Previously ineffective antiarrhythmic drugs were prescribed and discontinued 3 months later.
All ICDs were programmed to detect VTs slower than clinical VTs. Patients were followed with remote monitoring, as well as ICD interrogations and office visits every 3 months thereafter.
The primary outcome endpoint of the study was freedom from any VA recurrence, defined as any recurrent sustained VT or ventricular fibrillation receiving device-based treatment (antitachycardia pacing or shock).
Renal function was assessed by serum creatinine level; post-ablation increase in the baseline serum creatinine level beyond the physiological range was considered an adverse event (19). Rehospitalization was defined as a hospital admission for heart failure or arrhythmia-related causes during the follow-up period after the index ablation.
Continuous data are described as mean ± SD and categorical data as numbers and percentages. Student t tests and chi-square or Fisher exact tests were used to compare differences across groups. Long-term VA-free survival was assessed by Kaplan-Meier analysis, and the log-rank test was used to compare VA-free survival between the 2 groups. Multivariate Cox regression analysis was used to identify significant predictors of recurrence; all potential confounders were entered into the model if significant association was observed in the univariate analysis. All tests were 2 sided, and a value of p < 0.05 was considered to indicate statistical significance. Statistical analyses were performed using IBM SPSS Statistics 22.0 (IBM Corporation, Armonk, New York).
We screened 108 consecutive patients, of whom 93 (57 in group 1 and 36 in group 2) with NICM who were undergoing RF ablation for scar-related VTs and presented with low-voltage regions (scar) on electroanatomic voltage mapping were included in the study. The remaining 16 patients were excluded for not having endocardial or epicardial scar, defined with standard bipolar voltage cutoffs. The mean age of our study population was 55.6 ± 9.4 years; 63 (67.7%) were men. In terms of baseline characteristics, there were no differences in age, sex, hypertension, diabetes, and New York Heart Association functional class between the 2 groups; LVEFs were significantly lower in group 2 compared with group 1 (29 ± 5% vs. 32 ± 6%, p = 0.028) (Table 1). The scar locations in both groups are provided in Table 2.
RF time was longer in group 2 compared with group 1 (Table 3). Epicardial mapping was performed in all 36 patients (100%) in group 2 and 30 of 57 patients (52.6%) in group 1 (p < 0.001) in whom either VT was inducible after endocardial ablation or there was no endocardial scar. Fourteen of the 66 patients (21.2%) undergoing epicardial ablation required phrenic nerve protection.
Acute procedural outcomes
More patients in group 2 were noninducible for any VT compared with group 1 (25 of 36 [69.4%] vs. 24 of 57 [42.1%], p = 0.01). However, there was no significant difference between the groups regarding noninducibility of clinical VTs (28 of 36 [77.8%] vs. 36 of 57 [63.2%], p = 0.138) (Figure 2). During the ablation procedure, cardiopulmonary support was used in 7 patients in the standard ablation group because of hemodynamic instability, whereas it was not required for any patient in the scar homogenization group (p = 0.041).
Six patients undergoing standard ablation (group 1) and 1 patient from the scar homogenization group (group 2) developed hematoma, pseudoaneurysm, or arteriovenous fistula. Femoral vein complications were observed to be associated with the use of larger introducers required to place the device (intra-aortic balloon pump, percutaneous Impella device [Abiomed, Danvers, Massachusetts], or extracorporeal membrane oxygenation system) for hemodynamic support. Pericardial effusions occurred in 1 patient in each group, who were both treated conservatively. The procedure was terminated prematurely for instability in 3 patients in group 1. In addition, 2 patients in group 1 developed transient post-ablation renal dysfunction. Altogether, procedural complications were significantly more common in the standard ablation group (12 of 57 [21%] vs. 2 of 36 [5.6%], p = 0.042).
After ablation, 33 (57.9%) and 22 (61%) patients were on beta-blockers, and 29 (50.8%) and 11 (30.5%) were on amiodarone in groups 1 and 2, respectively.
During a mean follow-up period of 14 ± 2 months, 45 of 93 patients (48.4%) were free of any VA recurrence after a single procedure. Freedom from any VAs in group 2 was significantly more common compared with group 1 (23 of 36 [64%] vs. 22 of 57 [38.6%], log-rank p = 0.031) (Central Illustration).
Of note, among the 13 patients experiencing recurrent VAs at follow-up in group 2, 9 (69.2%) had incomplete elimination of all abnormal potentials within the scar at the time of the index procedure because the scar was either septal/midmyocardial (n = 6) or close to the epicardial mitral annulus with overlying fat or coronary arteries (n = 3).
A total of 14 of 93 patients (15%) experienced atrial fibrillation at follow-up. During follow-up, 3 patients (8.3%) in the scar homogenization group and 15 (26.3%) in the standard ablation group required rehospitalization for VT and heart failure–related causes (p = 0.035).
Predictors of recurrence
In univariate analysis, low LVEF (p < 0.001), male (p = 0.043), acute procedural failure (p < 0.001), and standard ablation (p = 0.032) were predictors of VA recurrence. After multivariate analysis, scar homogenization (hazard ratio: 0.48; 95% confidence interval: 0.27 to 0.96; p = 0.027) and LVEF (hazard ratio: 0.93; 95% confidence interval: 0.91 to 0.98; p = 0.008) were the only predictors of long-term freedom from VA. More specifically, endoepicardial scar homogenization was independently associated with a 41% relative risk reduction of any VA recurrence at follow-up.
To the best of our knowledge, this is the first study showing that endoepicardial scar homogenization improves long-term VA-free survival compared with limited-substrate ablation in patients with NICM (Central Illustration).
Our major findings were as follows: 1) after a single procedure and a mean follow-up period of 14 ± 2 months, freedom from any VA in the scar homogenization group was 64%, compared with 39% in the standard ablation group; 2) scar homogenization and LVEF were predictors of long-term freedom from any VA; 3) post-ablation noninducibility of clinical VT was not correlated with long-term ablation outcome; and 4) location of the scar influenced ablation outcome by limiting access to catheter ablation. In addition, significant reduction in the rehospitalization rate following scar homogenization highlighted the clinical relevance of this ablation strategy in VT patients with NICM.
We found that the ablation strategy of endoepicardial scar homogenization was associated with 64% freedom from any VA at follow-up, significantly higher than that of limited-substrate ablation. As in ICM, scar-related re-entry is the most common underlying mechanism for sustained VT (5,20). A scar is a complex structure, in which channels of surviving myocytes (delayed, fractionated, split potentials) (21) play an important role for VT perpetuation. With activation/entrainment mapping, it is possible to identify the channels that are critical for a given clinical VT circuit. Furthermore, with high-density substrate mapping, other small islets of surviving myocytes within scar can be identified, unrelated to VT circuit (7,22). These “innocent islets” are a potential substrate to perpetuate other undocumented VTs, and in ICM, their complete elimination has been reported to be associated with a better long-term outcome after ablation (15). The same rationale can be applied in NICM, and homogenization of the entire scar eliminates all the potential arrhythmogenic sources, regardless of their participation in a re-entrant VT circuit, improving outcomes, as shown in this study. Moreover, nonischemic arrhythmogenic substrate involves more commonly basal perivalvular regions with a midmyocardial/epicardial predilection (5,6,20), making epicardial mapping and ablation a vital part of this ablation strategy.
Although associated with better outcomes, the benefit of such an approach appeared inferior to that observed for ischemic substrates. We previously demonstrated that in ICM patients, endoepicardial scar homogenization was associated with an 81% long-term VA-free survival rate (15), whereas in the present study, the success rate was 64% after scar homogenization in an NICM population. Fundamental differences in the arrhythmogenic substrate of these 2 cardiomyopathies might partly explain these results. In both, fibrosis and scar were heterogeneous and complex, but ischemia led to a predominantly endocardial scar. Early reperfusion therapies can change the pathophysiology of post-infarction scars and increase their complexity with the presence of multiple slow conduction channels with exit sites located at different areas, even in the subepicardium (23). However, in NICM, the peculiar location of the scar (mostly septal and midmyocardial) may influence long-term arrhythmia recurrence and could explain the difference between the 2 populations. Thus, the lower success rate in the NICM population was likely due to the distribution of the scars compared with patients with ICM.
Of note, among the 13 patients who had recurrence in the scar homogenization group, 9 (69.2%) had incomplete elimination of their arrhythmogenic substrate for either a septal/midmyocardial (n = 6) or perivalvular (n = 3) location of the scar. This strongly suggested that the ablation outcome was influenced by the scar distribution and the subsequent poor accessibility for ablation in this subpopulation. In fact, in a study conducted by Bogun et al. (24), at 19 ± 11 months of follow-up, post-ablation recurrence was reported in all patients who had intramural scars. Although substrate for VT in NICM characteristically affects the left ventricular basal-lateral region, involvement of the interventricular septum, either isolated or in combination, has been reported (9,20). Both basal and septal regions pose a challenge to access. A perivalvular location prevents RF ablation because of the overlying fat and coronary arteries, and the septal lesions usually lie deep in the myocardium (25). Multiple ablation procedures, ablation along both sides of ventricles, or alcohol ablation may be required to successfully eliminate these septal arrhythmogenic circuits (25,26). We did not assess scar progression in this study. However, in our experience, change in scar size, as defined with standard bipolar criteria, is typically not observed during redo procedures.
We found endoepicardial scar homogenization to be a predictor of long-term success in patients with NICM. It is important to identify factors that predict the efficacy of an ablation procedure. Traditionally, in VT ablation, noninducibility of VT at the end of the procedure has been used as an endpoint and predictor of long-term success; yet data supporting this are derived mainly from patients with ICM (27–29). In NICM, this is still controversial: Dinov et al. (12) and Piers et al. (30) showed that noninducibility was associated with better long-term success, whereas Tokuda et al. (31) did not. We found that noninducibility of any VT at the end of the procedure did not predict long-term success. However, endo-epicardial scar homogenization was independently associated with a 59% relative risk reduction of any VT recurrence over follow-up. This was in line with the work of Jaïs et al. (17), showing that complete elimination of all abnormal potentials was a predictor of better outcomes in both ICM and NICM. The other independent predictor of long-term success was higher LVEF, which likely reflects a less complex substrate. This finding was consistent with the fact that more diseased myocardium may harbor a more potential arrhythmogenic substrate responsible for recurrences.
One limitation of this study was lack of randomization. However, our comparison between 2 groups was done using a prospective design, and the study sample was big enough to allow us to assess the impact on recurrences of the 2 ablation strategies. Additionally, we recognize that a potential consequence of this strategy could be greater tissue destruction, leading to limitation in contractile function of the heart. However, minimal change in post-ablation LVEF indicates that the scar homogenization approach was not detrimental. All the patients included in this series had evidence of scar on electroanatomic voltage mapping, as defined by standard bipolar cutoff criteria. As such, our findings cannot be generalized to patients with NICM and no evidence of low voltage (<1.5 mV) on bipolar electroanatomic voltage mapping, who were excluded from this study.
In patients with NICM, scar-related VT, and evidence of low-voltage regions on bipolar electroanatomic voltage mapping, endoepicardial scar homogenization significantly increased freedom from any recurrent VA compared with standard ablation. This approach might be considered an initial strategy for VT ablation in patients with NICM. However, the success rate with this approach appeared to be lower than the previously reported study with ICM (16), presumably because of the septal and midmyocardial distribution of the scar in some cases.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In patients with NICM, sustained monomorphic VT is usually due to scar-related re-entry in low-voltage zones. A hybrid approach involving scar homogenization across the endocardial-to-epicardial continuum seems more effective than limited ablation of the endocardial substrate and may be safer, but success is limited when intramural scar tissue is inaccessible.
TRANSLATIONAL OUTLOOK: Further studies are needed to identify patients with NICM and VT most likely to benefit from bipolar ablative scar homogenization and define the procedural techniques that optimize clinical outcomes.
Dr. Di Biase is a consultant for Biosense Webster, Boston Scientific, Stereotaxis, and St Jude Medical; and has received speaking honoraria and travel reimbursement from Medtronic, Atricure, EPiEP, and Biotronik. Dr. Natale has received speaking honoraria from Boston Scientific, Biosense Webster, St. Jude Medical, Biotronik, and Medtronic; and is a consultant for Biosense Webster, St. Jude Medical, and Janssen. Dr. Burkhardt is a consultant for Biosense Webster and Stereotaxis. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Gökoğlan and Mohanty contributed equally to this work.
- Abbreviations and Acronyms
- implantable cardioverter-defibrillator
- ischemic cardiomyopathy
- left ventricular ejection fraction
- nonischemic cardiomyopathy
- ventricular arrhythmia
- ventricular tachycardia
- Received February 12, 2016.
- Revision received July 29, 2016.
- Accepted August 2, 2016.
- American College of Cardiology Foundation
- Strickberger S.A.,
- Hummel J.D.,
- Bartlett T.G.,
- et al.,
- for the AMIOVIRT Investigators
- Zipes D.P.,
- Camm A.J.,
- Borggrefe M.,
- et al.,
- for the American College of Cardiology,
- American Heart Association Task Force,
- European Society of Cardiology Committee for Practice Guidelines
- Soejima K.,
- Stevenson W.G.,
- Sapp J.L.,
- et al.
- Cano O.,
- Hutchinson M.,
- Lin D.,
- et al.
- Nakahara S.,
- Tung R.,
- Ramirez R.J.,
- et al.
- Dinov B.,
- Fiedler L.,
- Schönbauer R.,
- et al.
- Dinov B.,
- Arya A.,
- Schratter A.,
- et al.
- Arenal A.,
- Glez-Torrecilla E.,
- Ortiz M.,
- et al.
- Di Biase L.,
- Santangeli P.,
- Burkhardt D.J.,
- et al.
- Di Biase L.,
- Burkhardt J.D.,
- Lakkireddy D.,
- et al.
- Jaïs P.,
- Maury P.,
- Khairy P.,
- et al.
- Bai R.,
- Di Biase L.,
- Shivkumar K.,
- et al.
- Hsia H.H.,
- Callans D.J.,
- Marchlinski F.E.
- de Bakker J.M.,
- van Capelle F.J.,
- Janse M.J.,
- et al.
- Nayyar S.,
- Wilson L.,
- Ganesan A.N.,
- et al.
- Wijnmaalen A.P.,
- Schalij M.J.,
- Von der Thusen J.,
- Klautz R.J.M.,
- Zeppenfeld K.
- Bogun F.,
- Desjardins B.,
- Good E.,
- et al.
- Oloriz T.,
- Silberbauer J.,
- Maccabelli G.,
- et al.
- Sivagangabalan G.,
- Pouliopoulos J.,
- Huang K.,
- et al.
- Carbucicchio C.,
- Santamaria M.,
- Trevisi N.,
- et al.
- Della Bella P.,
- De Ponti R.,
- Uriarte J.A.,
- et al.
- Rothman S.A.,
- Hsia H.H.,
- Cossú S.F.,
- et al.
- Piers S.R.,
- Leong D.P.,
- van Huls van Taxis C.F.,
- et al.
- Tokuda M.,
- Tedrow U.B.,
- Kojodjojo P.,
- et al.