Author + information
- Received April 30, 2018
- Accepted May 3, 2018
- Published online October 15, 2018.
- Jeremy P. Moore, MD, MSa,∗ (, )
- Kevin M. Shannon, MDa,
- Roberto G. Gallotti, MDa,
- Christopher J. McLeod, MBBSb,
- Anca Chiriac, MDb,
- Edward P. Walsh, MDc,
- Narayanswami Sreeram, MD, PhDd,
- Akash R. Patel, MDe,
- Natasja M. De Groot, MD, PhDf,
- Johannes von Alvensleben, MDg,
- Seshadri Balaji, MBBS, PhDh,
- David S. Frankel, MDi,
- Christina Y. Miyake, MD, MSj,
- James C. Perry, MDk and
- Kalyanam Shivkumar, MD, PhDl
- aDivision of Cardiology, Ahmanson/UCLA Adult Congenital Heart Disease Center, UCLA Medical Center, University of California, Los Angeles, Los Angeles, California
- bDepartment of Cardiovascular Diseases, Mayo Clinic Florida, Jacksonville, Florida
- cDepartment of Cardiology, Boston Children's Hospital, Boston, Massachusetts
- dDepartment of Pediatrics, University Hospital of Cologne, Cologne, Germany
- eDepartment of Pediatrics, University of California San Francisco, Benioff Children's Hospital, San Francisco, California
- fDepartment of Cardiology, Erasmus University Center, Rotterdam, the Netherlands
- gDivision of Cardiology, Heart Institute, Children’s Hospital Colorado, University of Colorado, Aurora, Colorado
- hDivision of Cardiology, Department of Pediatrics, Oregon Health and Science University, Portland, Oregon
- iElectrophysiology Section, Cardiovascular Division, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
- jLillie Frank Abercrombie Section of Pediatric Cardiology, Texas Children's Hospital, Houston, Texas
- kDivision of Cardiology, Department of Pediatrics, Rady Children’s Hospital, University of California-San Diego, San Diego, California
- lArrhythmia Center, Department of Cardiology, University of California, Los Angeles, Los Angeles, California
- ↵∗Address for correspondence:
Dr. Jeremy P. Moore, Division of Cardiology, Ahmanson/UCLA Adult Congenital Heart Disease Center, University of California Los Angeles, UCLA Medical Center, 100 Medical Plaza Drive, Suite 770, Los Angeles, California 90095.
Objectives The purpose of this study was to determine the ventricular arrhythmia (VA) substrates in patients with unoperated and post-surgical Ebstein's Anomaly (EA).
Background EA is associated with variable atrialization of the right ventricle and a propensity for VA and sudden death. There are scant data on catheter ablation for VA in this population.
Methods This was a retrospective study involving 11 congenital heart disease centers.
Results A total of 24 patients (median age 17 [interquartile range (IQR): 11 to 37] years; age range 1 to 68 years; 42% men) with EA undergoing catheter ablation were identified. Prior tricuspid valve (TV) surgery had been performed in 12 (50%). Presenting symptoms were palpitations in 15, syncope in 4, aborted cardiac arrest in 4, and none in 1. At procedure, 28 VA substrates were encountered and 25 were completely characterized (median 1 per patient; cycle length 305 [IQR: 268 to 400] ms). In 3 cases, premature ventricular contraction (PVC) foci were targeted (1 with a history of PVC-induced ventricular fibrillation). VA mechanisms were focal in 15 and macro–re-entrant in 10, and did not differ significantly between those with and those without prior TV surgery (p = 0.7). Focal VAs predominantly localized to the atrialized right ventricle ARV in unoperated patients and to diseased myocardium or Purkinje tissue after TV surgery. Macro–re-entry was related to isolated scar or split potentials in the ARV in unoperated patients, and larger, more diffuse scar after TV surgery. Complete success was achieved in 22 (92%). There were 2 of 13 complications in patients <18 years of age and none in patients >18 years of age. There was a single recurrence over a median follow-up of 3.4 years.
Conclusions VA in EA may be either focal or macro–re-entrant. In the absence of surgery, substrates chiefly involve the ARV. After surgery, focal VA involves injured myocardium or Purkinje tissue and re-entrant ventricular tachycardia is related to post-surgical scar. Catheter ablation is a reasonable therapeutic approach for these patients.
- catheter ablation
- congenital heart disease
- Ebstein’s anomaly
- sudden cardiac death
- ventricular tachycardia
Catheter ablation for patients with congenital heart disease (CHD) has evolved over the past decade to provide more definitive therapy for ventricular arrhythmias (VA). To date, mapping and ablation techniques have mostly been applied to patients with tetralogy of Fallot (TOF) who exhibit defined anatomical conduction isthmuses, allowing for durable substrate-based ablation (1). Thus far, catheter ablation for other congenital anomalies are not well reported, despite the importance of VA in many CHD syndromes (2).
VA and sudden death are well described in patients with Ebstein’s anomaly (EA). The underlying mechanisms for this are unclear, and, although potentially multiple, fibrosis and myocardial stretch are common pathologic changes identified in the atrialized portion of the right ventricle (3–5). Recognizable on the surface electrocardiogram as a prolonged and fragmented QRS complex, the maladaptive properties of the atrialized right ventricle (ARV) of slowed conduction velocity and short effective refractory period (4,6–9) may conspire to provide a nidus for re-entrant and triggered arrhythmia (6,10–12). Recently, it has been demonstrated that the development of VA in patients with EA conveys a 6-fold increased risk for sudden death, emphasizing the importance of VA prevention and treatment in this population (13).
The present study aimed to clarify the role of catheter ablation for patients with EA. It was hypothesized that VA substrates would: 1) vary by operation status; and 2) be highly amenable to endocardial ablation with favorable long-term outcomes.
An international, multicenter retrospective study of all patients with EA and VA undergoing electrophysiology study with the intention of treating VA by catheter ablation was performed. After local ethics committee approval, all centers transmitted data securely to the coordinating center at the University of California at Los Angeles through the Research Electronic Data Capture web application. Clinical variables of interest included patient demographics, congenital diagnoses, EA severity, and surgical history. The age at onset of the first VA, nature of the tachycardia episodes, and treatments rendered before catheter ablation were noted. Particular attention was paid to the number and type of antiarrhythmic drugs used, cardioversions required, and presence and type of therapy delivered by an implantable cardioverter-defibrillator (ICD).
Catheter ablation was performed under a combination of sedation and analgesia or by general anesthesia at the discretion of the operating physician. Except for 3 procedures during the early part of the study period, all procedures were conducted with the assistance of 3-dimensional electroanatomical mapping. For hemodynamically tolerated monomorphic ventricular tachycardia (VT), a combination of entrainment and activation mapping as well as pace mapping was used to identify the tachycardia mechanism and either the critical isthmus for re-entrant VT or the site of origin for focal VT. For unstable VT, pace mapping or substrate modification were used as previously described (14). Whenever possible, the relationship between the VA and the underlying anatomical substrate was determined through a review of electrogram morphology and amplitude, annotation of epimyocardial or endomyocardial scar (15), correlation with previous surgical records, and assessment by intracardiac echocardiography. VT with different morphologies or cycle lengths in the same patient that were determined to be related to opposite circuits or alternate exit points from a single region of myocardial scar, and that could be ablated from 1 isthmus or focus, were considered to be 1 unique VT substrate for purposes of the study. Following successful catheter ablation, repeat testing was generally performed with programmed ventricular stimulation from 2 separate right ventricular sites with up to triple extrastimuli to a minimum coupling interval of 200 ms or ventricular refractoriness. Isoproterenol was used during the post-ablation phase at the discretion of the operating physician.
Acute procedural success was defined as elimination of all spontaneous and inducible VA following catheter ablation. Procedural complications were noted and specified in detail to the coordinating center for review. Following catheter ablation, all patients were routinely followed at the participating center at least annually. Symptoms possibly attributable to VA were evaluated by the treating physician. Recurrence was defined as sustained VA (i.e., lasting >30 s) demonstrated electrocardiographically or by cardiac electronic implantable device, VA requiring ICD therapy, or recurrence of the originally ablated target following an acutely successful catheter ablation procedure.
Data are presented as median (interquartile range [IQR]) for continuous variables and as frequencies and percentages for dichotomous variables. For comparison between surgical and nonsurgical groups (i.e. with or without a prior history of tricuspid valve surgery), the Wilcoxon rank sum method was used for continuous variables and Fisher exact test was used for categorical variables. A p value <0.05 is regarded as significant. Statistical analyses were performed with JMP software version 13.0 (SAS Institute, Cary, North Carolina).
From September 1995 to August 2017, a total of 24 patients (median age 17 [IQR: 11 to 37] years, age range 1 to 68 years; 42% men) from 11 participating centers undergoing electrophysiology study with intended catheter ablation for VA were identified (Table 1). Coexisting congenital anomalies were uncommon, with the exception of atrial septal defects in 20 (83%) patients. Sixteen patients had undergone prior surgical intervention, of which tricuspid valve (TV) surgery with or without an additional ventricular incision was performed in 12 (50%). For unoperated patients, the median TV displacement was 1.7 cm/m2. Results of cardiac magnetic resonance imaging with gadolinium administration were available in 5 patients. Delayed enhancement was visualized in 0 of 3 patients without prior TV surgery and in both patients with prior TV surgery.
Symptoms included palpitations alone in 15 patients, syncope in 4, and aborted cardiac arrest in 4. One patient was asymptomatic, with inducible VT at the time of routine preoperative electrophysiology study. An ICD was placed before electrophysiology study in 4 patients, including 2 patients who presented with aborted cardiac arrest. Pre-procedural therapies included antiarrhythmic drugs in 19 patients, synchronized cardioversion in 10 (median 2 [IQR: 1 to 5] per patient), ICD delivered antitachycardia pacing in 2, and ICD shocks in 2. Patients with prior TV surgery were more likely to present with right ventricular systolic dysfunction and aborted cardiac arrest than were those without prior surgery (p = 0.03 and p < 0.001, respectively).
Electrophysiology study and catheter ablation
Procedural characteristics are shown in Table 2. The median time from onset of the first clinical VA to electrophysiology study was 1.0 (IQR: 0.1 to 2.9) year. Thirty VT morphologies and 3 premature ventricular contraction (PVC) foci were studied (median 1 per patient; median cycle length 305 [IQR: 268 to 400] ms). Of the combined 33 VAs, 28 were unique and 25 were completely characterized (incomplete mapping in 2, mapping deferred in 1). Including PVCs, VA was of focal origin in 15 (60%) patients and macro–re-entrant in 10 (40%). Activation mapping (79%) was the most common VA mapping method (p = 0.037) and irrigated radiofrequency energy was used in 50% of cases. Acute success was achieved in 22 (92%) patients. There was 1 failed attempt, because of persistent accelerated ventricular rhythm after partially successful catheter ablation. There were 2 complications (patients of 4.5 and 6.0 years of age), both related to dissection of the ARV. In 1 case, dissection of the free wall resulted in obstruction to right ventricular inflow and persistent desaturation, ultimately resulting in successful surgical repair. The second patient developed an intramural hematoma, which resolved spontaneously with conservative management.
VA substrates in relation to prior surgical repair
For patients without prior TV repair, 12 unique VTs were observed and could be characterized in all patients (Figure 1). Tachycardia was of focal origin in 8 patients and macro–re-entrant in 4. Focal VT originated from localized scar or fractionated potentials in the ARV in 6 patients and from apparently healthy myocardium outside of the ARV in 2. Macro–re-entrant VT was related to discrete basal inferior or basolateral scar or split potentials within the ARV in 3 cases (Figure 2). In the final case, bundle branch re-entrant VT was observed.
For patients with prior TV surgery, 16 unique VAs were observed (15 targeted for catheter ablation) and 13 were completely characterized. These were focal VT in 4 patients, PVC foci in 3, and macro–re-entrant VT in 6. Focal VT was related to localized fractionated potentials, and involved components of Purkinje tissue in 2 (left posterior fascicle in 1, right bundle branch in 1) patients and the anterior septum near resected leaflet tissue in 2 others. Similarly, PVC foci were related to right bundle potentials at the distal moderator band in 1 patient and from 2 distinct regions of fractionated potentials at resected septal leaflet tissue in a separate patient. PVC-induced VF was targeted during 1 case. PVC foci were frequent enough to allow for activation and pace mapping in all cases, and required the use of isoproterenol for maintenance in 2 patients. Although macro–re-entry related to ventricular scar could occur regardless of prior cardiac surgery, regions of scar were larger and more confluent, involving the anterior or inferior right ventricular free walls in those patients who had undergone previous TV surgery.
Over a median follow-up of 3.4 years, there was 1 clinical recurrence in a 68-year-old patient with complete atrioventricular block following TV replacement, who had undergone pre-procedure ICD placement. The episode was terminated with a successful ICD shock. There were no deaths or cardiac transplantation at follow-up. The single patient with a failed ablation because of accelerated ventricular rhythm did not demonstrate recurrence during the follow-up period.
Despite an elevated risk for sudden death and a known tendency to develop VA, catheter ablation for patients with EA has only rarely been reported (16–19). The present investigation represents the collective experience of several large CHD centers involved in the management of patients with EA and VA. This detailed analysis of ablation procedures reveals several important findings. First, it identifies that the location of VA in patients with this anomaly is primarily determined by whether a prior TV operation has been performed. For those patients with no history of TV operation, the origins of VA were found to be predominantly within the ARV. For patients who had undergone operative intervention, focal VA was related to sites of surgical resection or diseased Purkinje tissue; with macro–re-entrant VA most commonly related to surgical scars. Both focal and macro–re-entrant VA can occur in this population regardless of prior surgery, and catheter ablation in these centers was found to be highly effective both acutely and over midterm follow-up.
The present findings may facilitate the management of patients with EA and VA by allowing greater anticipation of VA substrates in pre- and postoperative subgroups and clarify the role of catheter-based intervention. Even though EA is one of the rarest forms of CHD (representing <1% of such cases), these patients are prone to sudden cardiac death (SCD), with data suggesting an incidence of SCD similar to that of the more familiar congenital heart defect of TOF (13). The present study findings may also have implications for the etiology of SCD in this under-represented population.
To date, the bulk of experience with VA in the CHD population has centered on patients with surgically repaired TOF, where macro–re-entry has been implicated as the dominant mechanism. Over the past decade, exploration of a limited number of anatomical conduction isthmuses in the TOF population has improved understanding of VA pathogenesis and has allowed for more focused catheter ablation (1,20–22). In contrast, the present report suggests that such well-defined and reproducible anatomical isthmuses infrequently play a role in the pathogenesis of VA for patients with EA. Indeed, such patients displayed a variety of VA substrates that varied with respect to their operative status.
For unoperated patients, monomorphic VT predominantly involved the basal ARV (Figure 1). Prior histologic evaluation of the ARV has demonstrated clusters of cardiomyocytes isolated by networks of fibrosis (4) that extend to involve the interventricular septum in patients with EA (5). These features likely explain the distribution of diseased myocardium and small regions of scar near the true tricuspid annulus that promoted VT. Indeed, re-entrant VT could often be ablated by linear lesions between the true annulus and nearby localized scar in this region (Figure 2). Re-entrant VT involving the basal right ventricle in unoperated patients appears to be a singular feature of EA not observed in the general CHD population.
For post-surgical patients, VA substrates were not restricted to the ARV and instead were usually observed in relation to surgical sequelae or damaged Purkinje tissue. It is likely that the presence of a dilated and dysfunctional right ventricle promoted the development of focal stretch–mediated Purkinje-related VA in many cases. In others, the underlying substrate for macro–re-entry was larger and involved more homogeneous regions of scar after TV surgery, consistent with obligatory features of repair that include resection of displaced leaflet tissue and imbrication of a dilated and aneurysmal ARV.
Efficacy of catheter ablation
In the present series, acute success was achieved for nearly all patients, with a single recurrence over a median follow-up of approximately 3.5 years. This compares favorably with VT ablation for other forms of CHD, where acute success has been reported to be in the range of 62% to 71% (1,22) and with higher rates of recurrence (1). This favorable experience may be due in part to several characteristic features of EA. First, activation mapping or entrainment maneuvers appear to frequently be possible for these patients, with substrate modification in the present series used as adjunctive therapy and rarely as a stand-alone approach. Second, many VTs in this group were caused by hemodynamically tolerated focal substrates that allowed for mapping and ablation of the site of origin. Last, it is possible that the anatomy of the right ventricle in EA may be naturally more amenable to catheter-based therapy, with transmural delivery of radiofrequency energy because of a relatively thin-walled chamber typically void of hypertrophy. Catheter ablation failure for other forms of CHD has often been attributed to VT circuits buried deep within the ventricular myocardium, especially after TOF repair where the isthmus may be protected by thick muscle bundles or a surgical bioprosthetic valve (1,21). Although the presence of surgical right ventricular plication could potentially render a heavily protected and unfavorable substrate in post-surgical patients with EA, this did not prove to be a significant challenge.
Implications for risk stratification
Patients with EA are prone to SCD, with an incidence of approximately 2 per 1,000 patient-years that is much higher than for other forms of CHD (13). Not surprisingly, a history of VA predicts future events in this patient group, particularly in the postoperative period (13,23). The present study points toward common features of postoperative EA that may be associated with SCD. In this series, surgically repaired patients were more likely to exhibit depressed right ventricular systolic function and aborted cardiac arrest before catheter ablation.
These study findings may have implications for the evaluation and risk management against SCD after surgical repair of EA. Given the observed responsiveness to programmed stimulation, it is possible that invasive electrophysiology testing may ultimately prove to be a useful maneuver to uncover latent VA substrates in this population. Such a strategy, in the appropriate clinical setting, could potentially improve the identification of high-risk postoperative patients that could benefit from primary prevention ICD placement. However, such a strategy would likely require validation in the form of a long-term prospective study.
Although the present collection of patients was small and validation of the results is required in the context of larger studies, this experience suggests that VA may be treatable by catheter ablation with excellent efficacy and low rate of recurrence. Therefore, tangible benefits in terms of reduced incidence of SCD might be anticipated; and similar to the situation for TOF, successful catheter ablation may ultimately decrease the need for ICD placement in select cases.
The 2 complications in this series were both related to intramural dissection of the ARV in children. Historically, concern for free-wall perforation has led some to avoid catheter ablation of VA altogether for patients with EA (24). As the ARV may be quite thin, this is a valid concern and careful consideration should be given to other potential treatment modalities. Unexpectedly, both cases of catheter trauma in the present series involved dissection of the free wall in the ARV, rather than true perforation. Nevertheless, given the potential severity of this complication it is crucial to exercise extreme caution, especially for young, unoperated patients with EA.
The present study was limited primarily by its retrospective design and inclusion of various mapping eras. Consequently, electroanatomical maps were not available for some of the early study cases and VT characteristics and substrate description was often derived from procedural reports. Similarly, the illustration of successful ablation sites was based on procedural reports as well as personal communication with site investigators. As there is a known spectrum of anatomic severity inherent to EA, these illustrations were ultimately imperfect representations of the individual anatomies and were simply intended to summarize the data in a readily understandable manner. In addition, the generalizability of these results may also be limited, given that all procedures were performed at referral centers specializing in complex congenital ablation procedures. Finally, despite the inclusion of multiple high-volume CHD centers, the present cohort of EA patients was relatively small. Given evidence of SCD that is similar to that of TOF (13), a greater understanding of VA mechanisms and treatment outcomes in this unusual CHD population is of clinical importance. The present study, which is the largest to date, is unique and provides novel insights into the mechanisms and implications of VA in patients with EA.
VA in patients with EA is amenable to catheter ablation with low risk for recurrence over medium-term follow-up. Unlike other forms of CHD, substrates associated with EA appear to be both focal and macro–re-entrant. VA most often involves the ARV in unoperated patients and post-surgical scar or Purkinje tissue after surgery. Catheter ablation appears to be a very effective therapeutic approach in this cohort.
COMPETENCY IN MEDICAL KNOWLEDGE: The targets for VA ablation in patients with EA appear to be dictated primarily by prior surgical status. For those without prior surgery, targets generally arise from the ARV. For those after TV surgery, a variety of sources such as operative scar, imbricated muscle, or damaged Purkinje tissue may play a role. This information may be useful to guide catheter ablation strategies for this unique population.
TRANSLATIONAL OUTLOOK: Larger multicenter, prospective studies are needed to further investigate the characteristics and outcome of catheter ablation for EA as well as other underrepresented forms of CHD. In addition, studies evaluating risk for SCD in light of the present findings may also be warranted.
The authors thank David Factor for his contribution to manuscript artwork.
Dr. Shivkumar is supported by National Institutes of Health Grant Nos. HL084261 and OT2OD023848. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrialized right ventricle
- congenital heart disease
- Ebstein’s anomaly
- implantable cardioverter defibrillator
- interquartile range
- premature ventricular contraction
- sudden cardiac death
- tetralogy of Fallot
- tricuspid valve
- ventricular arrhythmia
- ventricular tachycardia
- Received April 30, 2018.
- Accepted May 3, 2018.
- 2018 American College of Cardiology Foundation
- Kapel G.F.,
- Reichlin T.,
- Wijnmaalen A.P.,
- et al.
- Koyak Z.,
- Harris L.,
- de Groot J.R.,
- et al.
- Kastor J.A.,
- Goldreyer B.N.,
- Josephson M.E.,
- et al.
- Park S.J.,
- Chung S.,
- On Y.K.,
- et al.
- Watson H.
- Attenhoffer C.H.,
- Tan N.Y.,
- Hassan A.,
- et al.
- Jais P.,
- Maury P.,
- Khairy P.,
- et al.
- Marchlinski F.E.,
- Callans D.J.,
- Gottlieb C.D.,
- Zado E.
- Zeppenfeld K.,
- Schalij M.J.,
- Bartelings M.M.,
- et al.
- Kapel G.F.,
- Reichlin T.,
- Wijnmaalen A.P.,
- et al.
- Kapel G.F.,
- Sacher F.,
- Dekkers O.M.,
- et al.
- Oh J.K.,
- Holmes D.R. Jr..,
- Hayes D.L.,
- Porter C.B.,
- Danielson G.K.
- Cipolletta L.,
- Luzi M.,
- Piangerelli L.,
- Guerra F.,
- Capucci A.