Short- and Long-Term Success of Substrate-Based Mapping and Ablation of Ventricular Tachycardia in Arrhythmogenic Right Ventricular Dysplasia
Background— Multiple morphologies, hemodynamic instability, or noninducibility may limit ventricular tachycardia (VT) ablation in patients with arrhythmogenic right ventricular dysplasia (ARVD). Substrate-based mapping and ablation may overcome these limitations. We report the results and success of substrate-based VT ablation in ARVD.
Methods and Results— Twenty-two patients with ARVD were studied. Traditional mapping for VT was limited because of multiple/changing VT morphologies (n=14), nonsustained VT (n=10), or hemodynamic intolerance (n=5). Sinus rhythm CARTO mapping was performed to define areas of “scar” (<0.5 mV) and “abnormal” myocardium (0.5 to 1.5 mV). Ablation was performed in “abnormal” regions, targeting sites with good pace maps compared with the induced VT(s). Linear lesions were created in these areas to (1) connect the scar/abnormal region to a valve continuity or other scar or (2) encircle the scar/abnormal region. Eighteen patients had implanted cardioverter defibrillators, 15 had implanted cardioverter defibrillator therapies, and 7 had sustained VT (6 with syncope). VTs (3±2 per patient) were induced (cycle length, 339±94 ms), and scar was identified in all patients. Scar areas were related to the tricuspid annulus, proximal right ventricular outflow tract, and anterior/inferior-apical walls. Lesions connected abnormal regions to the annulus (n=12) or other scars (n=4) and/or encircled abnormal regions (n=13). Per patient, a mean of 38±22 radiofrequency lesions was applied. Short-term success was achieved in 18 patients (82%). VT recurred in 23%, 27%, and 47% of patients after 1, 2, and 3 years’ follow-up, respectively.
Conclusions— Substrate-based ablation of VT in ARVD can achieve a good short-term success rate. However, recurrences become increasingly common during long-term follow-up.
Received September 30, 2004; revision received February 18, 2005; accepted March 3, 2005.
Arrhythmogenic right ventricular dysplasia (ARVD) is a progressive disease of the right ventricle (RV), with fatty and/or fibrous replacement of normal myocardium.1 In ARVD, fibrofatty replacement of myocardium produces “islands” of scar region that may lead to reentrant ventricular tachycardias (VTs), and patients have an increased risk of sudden cardiac death, mostly secondary to VT.1–3 The use of implantable cardioverter/defibrillators (ICDs) is the treatment of choice for many ARVD patients with VT.2,4,5 However, VT ablation may be indicated for some patients either as a primary therapy or as an adjunct to ICD implantation.3,6
Although successful VT ablation is feasible in ARVD patients, long-term recurrence is common, probably because of the progressive nature of the disease.2 Furthermore, multiple morphologies, hemodynamic instability, or noninducibility may limit the success of VT ablation.7–9 Substrate-based mapping and ablation during sinus rhythm may overcome these limitations.7–9 It has been reported in ischemic and nonischemic cardiomyopathy that ablation of unmappable VTs is feasible by mapping in sinus rhythm and by creating ablation lesions between areas of scar and normal myocardium or anatomic boundaries.8 However, this approach has never been reported in a large series of ARVD patients.
Thus, we sought to report the feasibility and results of substrate-based VT ablation in a series of ARVD patients in whom traditional mapping during tachycardia was not possible.
Data from 26 patients meeting diagnostic criteria for ARVD and undergoing ablation for VT between 1999 and 2002 were reviewed retrospectively. ARVD was diagnosed in all patients according to criteria published previously.2 Patients were all referred for ablation of drug-refractory, recurrent VT. Of these 26 patients, only 22 were included in this series because they were patients in whom traditional mapping and ablation during VT were limited because of nonsustained VT at the time of electrophysiological study, hemodynamic instability during VT, or multiple morphologies of induced VTs. Every patient included in this study gave written, informed consent before the procedure, and all data were collected in accordance with institutional ethics board guidelines.
Patients were studied in the electrophysiology laboratory in the postabsorptive state under conscious sedation. Venous access was obtained from the femoral vein(s). After positioning of multipolar catheters in the RV and high right atrium, all patients had electrophysiological testing at the beginning of the procedure. Arrhythmia induction was performed according to standard protocols by using burst ventricular pacing and programmed stimulation, including up to 3 ventricular extrastimuli at 2 different RV sites.10 Intravenous isoproterenol was also used (maximum dose, 10 μg/min) when necessary.10
Induced VTs were compared with 12-lead ECGs or ICD stored electrograms of the clinical VT when available. However, in 5 of 22 patients, no 12-lead ECGs of the VT were available. Therefore, although we primarily targeted VTs matching or similar to the clinical VTs, we finally targeted all inducible, sustained, monomorphic VTs at the time of electrophysiology study.
Electroanatomic mapping was performed in all patients with the nonfluoroscopic CARTO mapping system (Biosense Webster Inc). Mapping was performed with a 7F 4-mm-tip deflectable ablation catheter (Navistar, Biosense Webster Inc) with 1-7-4-mm spacing between electrodes. Bipolar electrogram signals were filtered at 30 to 400 Hz and were displayed on a real-time recording system (Prucka Inc). In all patients, detailed endocardial voltage mapping of the RV was performed during sinus rhythm (n=21) or atrially paced rhythm (n=1) to identify regions of scar and abnormal endocardium. Bipolar voltage definitions of abnormal and normal myocardium were based on values validated previously in both right and left ventricles.8 Endocardial regions with a bipolar electrogram amplitude >1.5 mV were defined as “normal,” and “dense scar” areas were defined as those areas with an amplitude <0.5 mV. “Abnormal” myocardium was defined as a region with a bipolar electrogram amplitude between 0.5 and 1.5 mV. As per the study design, circumstances prevented traditional mapping during VT for all patients in this study.
Ablation Technique and Location of Lesions
When available, 12-lead ECGs of the VTs were used to regionalize their sites of origin to help identify a particular region of interest. Specific sites demonstrating “abnormal” voltage (defined earlier) were marked as targets when pace mapping at the site approximated the QRS morphology of the clinical and/or induced VTs (⩾9 of 12 match). Under CARTO guidance, linear ablation lines were created over target regions by sequential point lesions and were designed to (1) connect scar/abnormal myocardium to a valve continuity, (2) connect scar/abnormal myocardium to another scar, or (3) encircle the scar/abnormal region, depending on the scar location and size.8 Ablation was typically performed with a 7F 4-mm-tip deflectable ablation catheter (Navistar, Biosense Webster Inc), although we used an 8-mm-tip catheter when we could not deliver adequate power through the 4-mm tip. At each ablation point, radiofrequency energy was applied for 60 to 120 seconds with a power output starting at 30 W and titrating up to 50 W with a maximum temperature limit of 60°C. If an impedence drop of >10 Ω was observed during ablation, the power output was titrated downward. In 3 cases, a 7F, 3.5-mm irrigated-tip ablation catheter was used (Navistar ThermoCool, Biosense Webster Inc) after conventional 4-mm- and 8-mm-tip ablation failed to abolish inducible, hemodynamically unstable VTs.
Programmed RV stimulation was repeated after completion of ablation. As before, up to 3 extrastimuli at 2 different RV sites were used to attempt to induce VT. If VT was still inducible, additional ablation was performed in the same regions to further extend the lines or create new linear lesions when necessary. If the ablated VT or other sustained monomorphic VT could not be induced, the procedure was considered successful. No further ablation was performed when the only inducible rhythms were nontargeted premature ventricular contractions or polymorphic VT/ventricular fibrillation.
Patients continued to use the same oral antiarrhythmic medications after ablation as they had before ablation. Patients were routinely followed up in the outpatient clinic at 3, 6, and 12 months and then every 6 months thereafter. Patients with an ICD underwent device interrogation at each clinic visit to assess arrhythmia recurrence. Holter monitor recording was routinely obtained at 3 and 6 months after the procedure and whenever the patient had symptoms. Long-term success was defined as a lack of recurrence of sustained VT and/or appropriate ICD therapy.
Results are expressed as mean±SD unless otherwise indicated. The survival analysis describing freedom from VT recurrence was calculated with the Kaplan-Meier method.
Baseline clinical characteristics of the patients are summarized in Table 1. Fifteen (68%) patients were male, and the mean patient age was 41±15 years. Eighteen patients had previous ICD implantation for ventricular arrhythmias. All patients were taking antiarrhythmic medications at the time of ablation because of resistant, recurrent ventricular arrhythmias. The major indications for VT ablation were frequent ICD therapy (n=15), documented sustained VT with syncope (n=6), and documented sustained VT without syncope (n=1). For patients with frequent ICD therapy, the mean number of shocks was 5±3 (range, 3 to 11) in the 30 days before ablation, with a mean of 9±5 (range, 5 to 20) VT episodes. There was no or minimal left ventricular involvement in the patients with a mean left ventricular ejection fraction of 55±8%.
Electrophysiological characteristics and indications for substrate-based mapping are summarized in Table 2. A mean of 3±2 VTs was induced per patient, with a mean cycle length of 339±94 ms. Conventional mapping of the VTs was limited in these patients because of multiple/changing VT morphologies (n=7), nonsustained VT (n=5), hemodynamic intolerance (n=3), multiple/changing VT morphologies and nonsustained VT (n=5), and multiple/changing VT morphologies and hemodynamic intolerance (n=2).
Mapping and Ablation
CARTO maps of the RV were successfully acquired during sinus rhythm (n=21) or atrial pacing (n=1) in all patients. An average of 230±65 points was acquired per voltage map. Scar was identified in all patients. Distributions of scar areas are detailed in Table 2 but were predominantly in the following areas, in descending order of frequency of occurrence: lteral tricuspid annulus, proximal outflow tract, anterior-anteroapical wall, and inferior-inferoapical wall (Figure 1).
Linear lesions were created in all patients (Table 2). In 16 patients, linear lesions connected abnormal regions to the tricuspid annulus (n=12) and to other scars (n=4). In 13 patients, lesions completely encircling the scar/abnormal regions were performed, with (n=7) and without (n=6) other linear lesions. Figures 2 through 4⇓⇓ depict sample CARTO voltage maps from 2 patients, showing the lesion sets created. Procedural characteristics are summarized in Table 3. A mean of 38±22 radiofrequency lesions was applied per patient. Mean linear lesion length was 5.1±2.0 cm (range, 2.1 to 9.4). After initial ablation, VT was still inducible in 8 patients. In all of these patients, additional ablation was performed to extend the ablation lines or to seek new regions where good pace maps could be found. However, VT remained inducible in 4 patients, yielding a short-term procedural success rate of 82% (18/22 patients). Three of the 4 patients with failed ablation underwent a second procedure with an irrigated-tip ablation catheter instead. This catheter allowed us to deliver more effective lesions underneath the tricuspid valve, and in all 3 cases, the VT became noninducible. Irrigated-tip ablation was never performed in the body of the RV or at the RV apex because of the risk of perforation.
Serious complications occurred in 1 patient. In this case, acute pericardial tamponade developed after ablation. The patient became hypotensive, and the diagnosis was confirmed by emergent transthoracic echocardiography. The effusion was drained and the patient was transferred to the cardiac intensive care unit, where he was observed overnight. The patient survived and no operative intervention was required. Two patients suffered femoral hematomas, but neither required transfusion or fluid administration. No cases of deep venous thrombosis, pulmonary embolism, or cerebral events occurred.
Patients were followed up for a median of 37 months (range, 25 to 44). Patients were discharged with the same oral antiarrhythmic medications as they had been before ablation: amiodarone (n=6), sotalol (n=11), mexiletine (n=3), and flecainide (n=4). The majority of patients were also discharged on β-blocker therapy (n=16). Mean maintenance dose after discharge was 333±81 mg/d for amiodarone and 269±54 mg/d for sotalol. Patients remained on their antiarrhythmic medication regimen for the duration of follow-up. Only patients with VT recurrence had additions and/or changes made to their regimen (see following paragraphs). None of the patients without VT recurrence was withdrawn from antiarrhythmic medication. All 4 of the patients who did not have ICDs implanted before ablation subsequently underwent ICD implantation after the ablation procedure. Therefore, all 22 patients had ICDs in place during the follow-up period.
VT recurrence occurred in 8 patients. Freedom from VT recurrence according to Kaplan-Meier survival analysis is depicted in Figure 5. Recurrence occurred in 23%, 27%, and 47% of patients after 1, 2, and 3 years’ follow-up, respectively. Appropriate ICD therapies for VT (shocks or antitachycardia pacing) accounted for all 8 recurrences. The mean cycle length of recurrent VTs was slower compared with the cycle length of VTs induced in the electrophysiology laboratory: 397±78 versus 339±94 ms. Complete 12-lead morphology of these recurrent VTs was not available in 6 patients, and therefore, their morphologies could not be compared with the laboratory-induced VTs. In 2 patients, however, complete 12-lead morphology of the recurrent VT was available. In one of these patients, the VT had a different morphology compared with VTs targeted during the initial ablation. In the other patient, a similar but not identical VT morphology was observed compared with the VT targeted during ablation.
Of the 8 patients with VT recurrence, 4 had VTs that were slow enough with medication to be amenable to antitachycardia pacing and not necessitating repeated ablation. Two of these patients had additional antiarrhythmic therapy added to their regimen (mexiletine to amiodarone and mexiletine to sotalol). One patient had an ICD shock but has been controlled by adding mexiletine to amiodarone, and the patient has elected not to undergo repeated ablation. The other 3 patients underwent repeated ablation for VTs causing multiple shocks. In all 3 cases, further ablation in abnormal regions along the tricuspid valve with an 8-mm-tip ablation catheter was required to eliminate the inducible VT(s) in the short term. Although the follow-up duration after the second ablation is limited to these 3 patients, to date (median, 8 months; range, 6 to 11), 2 have been stable with medication (amiodarone and sotalol+mexiletine), and 1 has had recurrent slow VT, which was recently successfully ablated with an irrigation-tip catheter by extending the lesions underneath the tricuspid valve.
None of the patients died during the follow-up period. One of the patients was listed for cardiac transplantation for heart failure but had not yet received it at the end of the follow-up period.
This study demonstrates that in a subset of ARVD patients, mapping of VT(s) may be limited by changing/multiple morphologies, nonsustainability, or hemodynamic instability. By using substrate-based voltage mapping to identify regions of scar and abnormal myocardium, we were able to create linear ablation lesions guided by pace mapping to connect or encircle the abnormal regions. The study shows that this technique can achieve a high rate of short-term success, but recurrence becomes increasingly common over time. To our knowledge, this is the largest report to examine the efficacy of substrate-based mapping and ablation of VT in ARVD patients.
Substrate in ARVD
ARVD is a disease in which normal myocardium is replaced by fatty and then fibrous tissue, causing thinning and scarring, predominantly in the RV.11 Indeed, in our series, we found regions of abnormal voltage suggestive of scar in all patients. These abnormal regions were mostly located around the lateral tricuspid annulus, proximal outflow tract, anterior-anteroapical wall, and inferior-inferoapical wall. This pattern is consistent with previous pathological studies demonstrating a predilection for scar to occur in these particular areas in ARVD.1,2 It is unclear why these specific regions develop scar, although local apoptosis or inflammation may play a role.1,11 Furthermore, the use of electroanatomic voltage mapping has been previously validated to correlate with areas of scar and fatty infiltration detected by magnetic resonance imaging in ARVD patients.12 Therefore, our definition of the substrate seems to be consistent with prior published data.
Regions of abnormal myocardium did not always follow the pattern of dense scar surrounded by a ring of abnormal myocardium, often referred to as the scar border zone. Sometimes abnormal voltages alone without dense scar defined the regions. This is because ARVD is quite a different process from scarring caused by myocardial infarction. Infarction causes dense scar surrounded by a border zone because of the ischemic penumbra that surrounds the infarcted territory. In ARVD, however, the process is patchy and may cause inhomogeneous scarring in anatomically disparate areas. Regardless, previous data have shown that it is still possible to identify well-demarcated borders around these abnormal regions,12 and our present study corroborates this finding.
Recently, Marchlinski et al13 published results of electroanatomic mapping in patients with RV cardiomyopathy, likely due to postinflammatory fibrosis. Their patient population was different from ours, because none of their patients had a family history of ARVD or sudden death, whereas the majority of our patients had this history. We did find some similarities in the distribution of scar, with a predominance of perivalvular voltage abnormalities, particularly around the tricuspid valve, and abnormalities in the septum and free wall. However, although the other series reported relative sparing of the apex, we found that apical involvement with extension into the anterior and inferior walls was not uncommon.
Mechanisms of Substrate-Based Ablation in ARVD
Intraoperative studies of patients with ischemic cardiomyopathy have shown that isolating ventricular scar by circumferential ablation can reduce or even cure VT.14,15 That is because the exit sites of VT reentrant circuits are often located in the border zone around the scar, so circumferential ablation may eliminate many of these critical exits.16 Furthermore, ablation connecting scar to an anatomic boundary has been shown to work by eliminating potential narrow isthmuses, which may be part of a larger circuit.8,17 Thus, these ablative techniques may be successful in ARVD, provided that the VTs in this disease follow the reentrant circuit paradigm described in the postinfarction setting. Ellison et al3 have shown that like the postmyocardial infarction setting, most VTs in ARVD can be entrained and that critical isthmuses can be identified and ablated to successfully terminate tachycardia. However, those authors also found that compared with patients after infarction, there was a higher rate of successful ablation in areas not defined as being central isthmus sites, such as outer or inner loop circuit locations. Furthermore, the short-term success rate was lower than in postinfarction patients, especially in RV sites away from the tricuspid or pulmonic annulus. One of the reasons for this may be the diffuse nature of ARVD, creating a large number of potential reentrant circuits within the same region of interest.
By targeting our ablation lesions in abnormal, border regions guided by pace mapping and by extending linear lesions to scars or anatomic boundaries, our ablative approach targets the exit sites of many potential VT circuits. Furthermore, by extending lesions to within the scar region, we may also be ablating more proximal critical slow-conduction zones of multiple circuits.17,18 This may explain why our approach achieved a higher short-term success rate compared with a single-lesion/focal ablative approach. Because ARVD can be such a diffuse process, an ablation strategy that uses larger and longer lesions may be advantageous in addressing all potential circuits. Because we did not perform any entrainment mapping in these patients, however, we cannot make any definitive statement about the relation of our lesions to specific parts of the reentrant circuit(s). Furthermore, the short-term success rate in our series was a little higher than that reported for substrate-based mapping in other cardiomyopathies (especially in the left ventricle).8,19 This may be because ARVD RVs have thinner walls, increasing the likelihood of transmural ablation that eliminates subendocardial and epicardial circuits, obviating the need for epicardial ablation approaches.
Although short-term success was relatively high in our series, recurrence was very common after 3 years. This has also been the experience of other investigators.2,3,6 One reason is that ARVD is a progressive disease, with new regions of scar developing over time and creating new circuits or changes in current ones.20 Another explanation is that because ARVD may create a large number of potential circuits, an empiric approach may not eliminate all of the possible pathways. Achieving adequate energy delivery may also be a problem, particularly in regions difficult to access by catheter. In particular, we found that abnormal regions along the tricuspid annulus required reablation in all patients with recurrent VT who underwent repeated procedures. Furthermore, to achieve effective lesions underneath the tricuspid valve, we needed to use an 8-mm-tip or irrigation-tip ablation catheter in most patients. It must be emphasized that caution must be taken when using irrigation-tip ablation in patients with very thin ventricular walls. We did not use irrigation-tip ablation anywhere other than around the tricuspid annulus. The risk of creating deeper lesions in thin walls of the RV body or apex potentially increases the risk of perforation, although the magnitude of this risk is not well known.
Marchlinski et al13 recently reported a high long-term success rate with substrate-based VT ablation in RV myopathy patients (84% for 27 months). However, as mentioned earlier, the progressive nature of ARVD, in contrast to postinflammatory myopathy, is a possible explanation for why the long-term success rate in our series was much lower.
Although VT, especially in combination with RV dysfunction, is known to predict a high risk of mortality,21 there were no deaths in our series, largely because all of the patients eventually had ICDs implanted. This is consistent with the series by Hulot et al,21 who also observed no deaths in patients with ICDs.
ARVD patients are at risk of sudden cardiac death due to ventricular arrhythmias, and ICDs are frequently implanted in those thought to be at highest risk, based on individual clinical characteristics. However, these patients have high rates of appropriate ICD therapies over time.4 VT ablation may therefore be an important adjuvant therapy, having been previously reported to reduce the number of ICD therapies and improve quality of life.6,22 In particular, the substrate-based approach in this study can achieve a relatively high short-term success rate and may be an important palliative treatment for ARVD patients with numerous VT episodes and ICD therapies. The approach also obviates the need for having stable, sustained, well-tolerated VTs for mapping and ablation and can be applied in a wider spectrum of ARVD patients with VTs. However, given the high long-term recurrence rate, this technique is far from curative. It is interesting that the cycle length of the recurrent VTs was slower, and it is possible that although ablation does not cure the VT in the long term, slower recurrences may be more amenable to antitachycardia pacing or drug therapy. Perhaps repeated attempts at substrate modification may be required over time in ARVD patients as the disease evolves. However, it is important to emphasize that VT ablation cannot be considered a curative procedure in this patient population.
An important limitation of this study was that it only included patients with more advanced disease, as evidenced by the presence of scar found in all patients during electroanatomic mapping. Some ARVD patients may have VT or ventricular fibrillation without evident scar,20 and for such patients, a substrate-based approach would not be applicable. Another limitation is that we did not confirm the presence of block across our ablation lines. This was a purely empiric approach to ablation and was not based on the demonstration of electrical block. We also did not perform any entrainment mapping, so we cannot correlate the location of our lesions to specific portions of the VT reentrant circuit(s) as defined by entrainment criteria. Pace mapping to target VTs is used to identify possible circuit exit site locations, but the technique is known to have limitations.23
Another limitation is that our definition of recurrence included any sustained VT and did not specify recurrence of VTs that were documented at the time of the initial procedure. This was because recurrence data were primarily obtained from stored ICD electrograms, which did not always allow us to identify whether the VT was one that was previously ablated or whether it was a de novo presentation. Using ICD therapy as a surrogate for VT recurrence is a known limitation.
The total number of patients was also small, and whether results would be similar in a larger cohort is unknown. This was also a very select population with definable scar, and all were referred specifically for VT ablation. However, this study had one of the largest cohorts of ARVD patients reported to date and is the largest that has specifically investigated ablation therapy. The low prevalence of this disease makes it very difficult to report any therapeutic outcome in very large numbers of patients.
Finally, we cannot rule out that in this series of ablation patients, arrhythmia may have subsided as part of the natural history of electrical storm rather than because of ablation. This is unlikely, though, given the frequent nature of the VT episodes; it would have been very coincidental that the electrical storm subsided immediately after ablation in all 22 patients. However, a randomized trial with a control arm would be required to definitively assess the impact of ablation on VT abatement.
Substrate-based mapping and ablation of VT are feasible in ARVD patients in whom conventional mapping during tachycardia is not possible. This method may achieve a high short-term success rate. However, recurrences become increasingly common over the long term, possibly as a result of the progressive nature of this disease.⇓
Atul Verma is supported by a fellowship award from the Heart and Stroke Foundation of Canada.
Basso C, Thiene G, Corrado D, Angelini A, Nava A, Valente M. Arrhythmogenic right ventricular cardiomyopathy: dysplasia, dystrophy, or myocarditis? Circulation. 1996; 94: 983–991.
Corrado D, Basso C, Thiene G. Arrhythmogenic right ventricular cardiomyopathy: diagnosis, prognosis, and treatment. Heart. 2000; 83: 588–595.
Roguin A, Bomma CS, Nasir K, Tandri H, Tichnell C, James C, Rutberg J, Crosson J, Spevak PJ, Berger RD, Halperin HR, Calkins H. Implantable cardioverter-defibrillators in patients with arrhythmogenic right ventricular dysplasia/cardiomyopathy. J Am Coll Cardiol. 2004; 43: 1843–1852.
Corrado D, Leoni L, Link MS, Della Bella P, Gaita F, Curnis A, Salerno JU, Igidbashian D, Raviele A, Disertori M, Zanotto G, Verlato R, Vergara G, Delise P, Turrini P, Basso C, Naccarella F, Maddalena F, Estes NA 3rd, Buja G, Thiene G. Implantable cardioverter-defibrillator therapy for prevention of sudden death in patients with arrhythmogenic right ventricular cardiomyopathy/dysplasia. Circulation. 2003; 108: 3084–3091.
Reithmann C, Hahnefeld A, Remp T, Dorwarth U, Dugas M, Steinbeck G, Hoffmann E. Electroanatomic mapping of endocardial right ventricular activation as a guide for catheter ablation in patients with arrhythmogenic right ventricular dysplasia. Pacing Clin Electrophysiol. 2003; 26: 1308–1316.
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: 1288–1296.
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: 664–669.
Stevenson WG, Khan H, Sager P, Saxon LA, Middlekauff HR, Natterson PD, Wiener I. Identification of reentry circuit sites during catheter mapping and radiofrequency ablation of ventricular tachycardia late after myocardial infarction. Circulation. 1993; 88: 1647–1670.
Marchlinski FE, Zado E, Dixit S, Gerstenfeld E, Callans DJ, Hsia H, Lin D, Nayak H, Russo A, Pulliam W. Electroanatomic substrate and outcome of catheter ablative therapy for ventricular tachycardia in setting of right ventricular cardiomyopathy. Circulation. 2004; 110: 2293–2298.
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: 675–681.
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: feasibility for guiding ventricular tachycardia ablation. Circulation. 2002; 106: 1678–1683.
Hulot JS, Jouven X, Empana JP, Frank R, Fontaine G. Natural history and risk stratification of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Circulation. 2004; 110: 1879–1884.
Stevenson WG, Friedman PL, Sweeney MO. Catheter ablation as an adjunct to ICD therapy. Circulation. 1997; 96: 1378–1380.