Author + information
- Received April 17, 2017
- Revision received November 27, 2017
- Accepted November 28, 2017
- Published online February 2, 2018.
- Rajeev K. Pathak, MBBS, PhDa,
- Joe Fahed, MDa,
- Pasquale Santangeli, MD, PhDa,
- Matthew C. Hyman, MD, PhDa,
- Jackson J. Liang, DOa,
- Maciej Kubala, MDa,
- Tatsuya Hayashi, MDa,
- Daniele Muser, MDa,
- Manina Pathak, MBBS, MPHb,
- Arshneel Kochar, MDa,
- Simon A. Castro, MDa,
- Fermin C. Garcia, MDa,
- David S. Frankel, MDa,
- Gregory E. Supple, MDa,
- Robert D. Schaller, DOa,
- David Lin, MDa,
- Michael P. Riley, MDa,
- Rajat Deo, MDa,
- Andrew E. Epstein, MDa,
- Erica S. Zado, PA-Ca,
- Sanjay Dixit, MDa,
- David J. Callans, MDa and
- Francis E. Marchlinski, MDa,∗ ()
- aCardiovascular Division, Electrophysiology Section, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania
- bCardiovascular Division, University of Adelaide, Adelaide, Australia
- ↵∗Address for correspondence:
Dr. Francis E. Marchlinski, Hospital of the University of Pennsylvania, 9 Founders Pavilion – Cardiology, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.
Objectives This study reports the long-term outcome of patients with bundle branch re-entrant tachycardia (BBRT) who underwent catheter ablation for ventricular tachycardia (VT).
Background BBRT is an uncommon mechanism of VT. Data on long-term outcomes of patients with BBRT treated with catheter ablation are insufficient.
Methods Between 2005 and 2016, 32 patients had a sustained VT due to a bundle branch re-entrant mechanism. Diagnosis of BBRT was established per standard published criteria.
Results The mode of presentation was syncope in 17 patients (53%) and palpitations in 15 (47%). BBRT was inducible in all subjects, and successful ablation of the right bundle branch in 19 patients (59%) or the left bundle branch in 13 patients (41%) was performed. During follow-up of 95 ± 36 months, 6 patients (19%) died, 3 of progressive heart failure and 3 of noncardiac causes. Recurrent VT due to BBRT did not occur in any patient. At baseline, 25 patients (78%) had a prolonged HV interval (>55 ms) and 7 (22%) had a normal HV interval (≤55 ms). In patients with a normal HV interval, there was only 1 death (due to malignancy), and no one developed heart block during 90 ± 36 months of follow-up. Ten patients (31%) had normal left ventricular (LV) function (LV ejection fraction ≥50%), and 22 (69%) had depressed LV function (LV ejection fraction <50%). No deaths were recorded in patients with normal LV function (5 with no implantable cardioverter-defibrillator) compared with 6 deaths among patients with depressed LV function (n = 22; p = 0.07).
Conclusions Radiofrequency ablation of the bundle branch is an effective therapy for treatment of BBRT. Sustained BBRT can be seen in patients with normal LV systolic function and HV interval with excellent long-term outcomes after ablation.
- bundle branch re-entrant tachycardia
- catheter ablation
- implantable cardioverter-defibrillator
- ventricular tachycardia
Bundle branch re-entrant tachycardia (BBRT) accounts for up to 6% of the sustained ventricular tachycardia (VT) seen in practice (1,2). The clinical presentation of BBRT is varied, but pre-syncope or syncope is common. BBRT is amenable to radiofrequency ablation, and this approach has been shown to be highly effective (2–7).
BBRT is usually seen in patients with a structural heart disease and conduction system disease. These patients can have other myocardial VT in addition to BBRT (8,9). Implantable cardioverter-defibrillators (ICDs) are, therefore, often implanted in patients with BBRT as a secondary prevention strategy; however, BBRT can be seen in patients with normal left ventricular (LV) function (3,10). The long-term risk of further arrhythmic events and sudden cardiac death remains undefined in these patients. Therefore, uncertainty exists as to whether patients with normal LV function and BBRT should also undergo insertion of an ICD. Additionally, although conduction system disease with a prolonged His-ventricular (HV) interval during sinus rhythm (SR) is found in most patients with BBRT, there are reports of BBRT patients with normal HV intervals during SR (11). It has been unclear whether the conduction properties in these patients with BBRT deteriorate over time (either spontaneously or because of catheter ablation in proximity to the conduction system) and to what extent they are at risk for the development of advanced heart block.
In this paper, we present the follow-up results from 32 consecutive patients with BBRT treated with radiofrequency ablation and define their long-term outcomes related to conduction disease, ventricular arrhythmias, and death. In addition, we report the long-term outcomes of distinct subgroups of BBRT with normal LV function and those with normal HV intervals during SR at baseline.
Of the 1,279 patients who presented with a history of VT between 2005 and 2016 at the Hospital of the University of Pennsylvania, VT was inducible during ablation in 1,119 patients. Thirty-six patients (3.2%) were identified as having sustained monomorphic VT due to a bundle branch re-entrant mechanism (11 before 2010, 18 between 2010 and 2015, and 7 between 2015 and 2016). Four patients were excluded because they did not have at least 12 months of follow-up data. The remaining 32 patients were included in the analysis. All patients provided written informed consent for both the ablation procedure and inclusion in a registry as approved by the University of Pennsylvania Health System’s Institutional Review Board.
Patients underwent the procedure in the fasting state. The procedure was performed under conscious sedation whenever possible. General anesthesia was used when deemed necessary by the operator. When possible, antiarrhythmic drugs were discontinued ≥5 half-lives before the procedure. Catheters were placed into position in the heart under fluoroscopic guidance. Standard transvenous 6-F quadripolar catheters with 5-mm interelectrode distance (Bard Inc., Delran, New Jersey) were placed across the tricuspid valve for recording His and right bundle electrograms (when BBRT was suspected as the mechanism of VT) and in the right ventricular apex.
The programmed ventricular stimulation protocol to induce VT involved 2 drive-cycle lengths with up to 3 extrastimuli. Electrical stimulation was performed with a Prucka CardioLab recording system (GE, Houston, Texas). The 12-lead electrocardiogram (ECG) morphology of all spontaneous VTs, when available, and the intracardiac near-field and far-field electrograms of the ICDs were collected and compared with the VT(s) induced during the procedure and their electrograms. Induced VT(s) were identified as clinical if they matched the cycle length and morphology of stored ICD electrograms or the 12-lead ECG when available. If VT was not inducible, intravenous isoproterenol infusion was used to facilitate the induction.
A bundle branch re-entrant mechanism was determined based on the previously published criteria: 1) a His bundle electrogram preceded each QRS complex with initiation and maintenance of VT; 2) whenever there was cycle length variation during tachycardia, H-H interval variations preceded V-V variation during tachycardia; 3) the duration of the HV interval was the same or longer during tachycardia as compared with sinus beats; 4) tachycardia was not inducible after ablation of either the left bundle branch (LBB) or right bundle branch (RBB); and 5) the QRS morphology during tachycardia on the surface ECG, as well as the activation sequence of intracardiac electrograms, was consistent with depolarization of the ventricle via one of the bundle branches (4,9,12,13).
In 27 (84%) of 32 procedures, a deflectable 3.5-mm open irrigated-tip catheter (ThermoCool, Biosense Webster, Diamond Bar, California) was used for mapping and ablation, and in the remaining 5 (16%) procedures, a nonirrigated 4-mm-tip ablation catheter (NaviStar, Biosense Webster) was used. The mapping/ablation catheter was advanced to the right ventricle (transvenous approach) and LV (retrograde aortic or trans-septal approach) depending on the specific bundle branch being targeted for ablation. The primary ablation endpoint was creation of a durable bundle branch block. Patients underwent repeat ventricular stimulation, and all other induced VTs with a cycle length >250 ms were considered potentially relevant and routinely targeted for ablation. After that, programmed ventricular extrastimulation was repeated to confirm the noninducibility of BBRT. Finally, both decremental and programmed premature atrial stimulation were used to evaluate the conduction properties of the His-Purkinje system and to assess the propensity to develop infra-Hisian block.
Transthoracic echocardiography was performed with a 3.5-MHz probe at baseline and at least once 1 year after the ablation procedure, with measures performed according to American Society of Echocardiography guidelines (14). Normal LV ejection fraction (LVEF) was defined as ≥50%, and low LVEF was defined as <50%.
Patients were evaluated at 4 to 8 weeks after ablation and then at 12-month intervals. For patients not followed up at our institution, the referring cardiologists were contacted and ICD interrogations reviewed to determine VT recurrence. The Social Security Death Index database was also queried for mortality information. Primary outcome was incidence of BBRT post ablation. Secondary outcome was incidence of any VT (defined as any sustained VT on ICD interrogation or 12-lead ECG) post ablation during the follow-up period. The acute efficacy was assessed based on inducibility of VT at the end of the ablation procedure with a consistent stimulation protocol.
Demographic, electrocardiographic, and echocardiographic characteristics are provided as continuous or categorical variables where appropriate. Descriptive statistics are reported as mean ± SD. Continuous variables are expressed as mean ± SD if normally distributed or median (interquartile range: 25th to 75th percentile) if not normally distributed. Categorical data are expressed as counts and percentages.
Continuous variables were compared with independent-sample parametric (unpaired Student t) or nonparametric (Mann-Whitney U) tests. Paired variables among the same patients were compared with paired-sample parametric (paired Student t) or nonparametric (paired Wilcoxon signed rank) tests. Categorical variables were compared with chi-square test or Fisher exact test when appropriate. For reporting herein, all decimal points were rounded to the closest exact numbers. All analysis was performed with STATA version 13.1 software (StataCorp, College Station, Texas).
Baseline characteristics of the study population are summarized in Table 1. Thirty-two consecutive patients with BBRT (age 69 ± 13 years, 81% males) underwent catheter ablation. Five patients (15%) had a history of valve replacement, 18 (56%) had significant coronary artery disease, and 8 (25%) had nonischemic cardiomyopathy. The patients presented with palpitations, with documented VT in 15 (47%), syncope with documented sustained VT in 6 (18%), and syncope in 11 (35%). Ten patients (31%) had normal LVEF (mean 61.0 ± 7.7%), and 22 (69%) had low LVEF (mean 28.5 ± 9.0%). At baseline, 4 patients (13%) had normal LV function, QRS duration (QRSd), and HV interval during electrophysiological study.
At the time of electrophysiological study, 26 patients (81%) presented in SR, with a mean PR interval of 151 ± 32 ms, and 6 (19%) presented with first-degree atrioventricular (AV) block with a PR interval of 256 ± 32 ms. The mean PR interval did not change significantly in either group after ablation (normal PR interval group: 151 ± 32 ms vs. 157 ± 23 ms [p = 0.80]; first-degree block: 256 ± 32 ms vs. 250 ± 32 ms [p = 0.90]). QRS morphology on the baseline 12-lead ECG showed nonspecific intraventricular conduction delay in 8 patients (25%), an LBB block (LBBB) pattern in 18 patients (56%), and an RBB block pattern in 6 patients (19%). QRSd was ≤120 ms in 6 patients (19%) and >120 ms in 26 (81%). The mean QRSd was 141 ± 29 ms before ablation and increased to 180 ± 25 ms after ablation (p < 0.0001). During electrophysiological study, 17 patients (53%) had only BBRT versus 15 (47%) with at least 1 other myocardial VT (mean 2.8 ± 1.0, range 2 to 5). At final follow-up, deterioration in conduction properties was not seen in any patient, with a mean PR interval 161 ± 23 ms (p = 0.80) and mean QRSd of 186 ± 28 ms (p = 0.60).
At baseline, the mean AH interval was 197 ± 32 ms and the mean HV interval was 73 ± 15 ms (7 patients [22%] had a normal HV interval [≤55 ms; mean 53 ± 2 ms]). BBRT occurred spontaneously in 7 patients (22%) or with ventricular extrastimulation in 25 (78%). For VT induction, an intravenous infusion of isoproterenol was required in 6 patients (19%). The tachycardia had LBBB morphology in 25 patients (78%; mean QRSd 179 ± 43 ms [range 115 to 251 ms], cycle length 354 ± 65 ms [range 240 to 470 ms]) and RBB block morphology in 7 patients (22%; mean QRSd 156 ± 31 ms [range 131 to 220 ms], cycle length 377 ± 63 ms [range 310 to 480 ms]). The frontal plane axis of tachycardia was left superior in 20 cases (63%), left inferior in 8 cases (25%), right superior in 2 cases (6%), and right inferior in 2 cases (6%).
In this cohort, the RBB was ablated in 19 patients (59%) and the LBB in 13 (41%). Of the 13 patients in whom the LBB was targeted, 11 had evidence of significant antegrade LBB delay at baseline, and 2 had sustained LBBB during catheter manipulation in the LV, in which case an empirical ablation at the basal septal region was performed. In 24 patients (75%), the ablation was performed in SR. Eight patients (25%) were ablated in VT. Successful ablation was performed in 31 patients without significant deterioration of AV conduction (AH interval 197 ± 32 ms to 205 ± 11 ms [p = 0.20] and HV interval 73 ± 15 ms to 80 ± 43 ms [p = 0.40]). One patient (3%) with a pre-existing biventricular device and baseline LBBB developed complete heart block after RB ablation. Of the 18 patients (56%) with incomplete LBBB at baseline, the RBB was targeted in 10 (31%) because they already had a pacing device (biventricular device in 8 and single-chamber ICD in 2 patients), whereas the LBB was targeted in 8 patients (25%) with no pacing device to avoid complete heart block. Of the 15 patients with other myocardial VT (mean 3 ± 1 VTs, range 2 to 5 VTs), the interventricular septum was identified as a source of VT and targeted for ablation in 5 patients (33%). At the end of the procedure, BBRT was not inducible in any patient, other clinical myocardial VT was still inducible in 2 patients (13%), and an additional 3 patients (9%) had at least 1 nonclinical VT that was still inducible.
Long-term clinical follow-up
Mean duration of follow-up was 95 ± 36 months. A total of 6 patients died during follow-up: 3 of progressive heart failure (4 to 60 months after the ablation) without any recurrent ventricular arrhythmia, 1 of sepsis (at 54 months), 1 of pulmonary embolism (at 2 months), and 1 due to malignancy (at 65 months). The remaining 26 patients (81%) were followed up for 88 ± 36 months, with no documented BBRT recurrence over this interval. Four patients (15%) had other documented myocardial VTs during follow-up. Notably, each of them had myocardial VT along with BBRT during the ablation procedure; 2 had recurrence of a previous VT, and the other 2 had new VTs.
Significance of narrow QRS (≤120 ms) and normal HV interval (≤55 ms) at baseline
Table 2 shows characteristics of these patients. At baseline, 6 patients (19%) had QRSd ≤120 ms (mean 104 ± 15 ms; range 82 to 120 ms) and 26 (81%) had QRSd >120 ms (mean 151 ± 23 ms; range 125 to 236 ms). There was no difference in follow-up duration between the 2 groups (97 ± 32 months for QRSd <120 ms vs. 95 ± 34 months for QRSd >120 ms; p = 0.90). A further prolongation of QRSd after ablation was seen in both groups, with a mean of 141 ± 10 ms in those with QRSd <120 ms (p = 0.0005) versus a mean of 180 ± 25 ms for those with QRSd >120 ms (p = 0.0001). Successful ablation of the RBB or LBB was performed in all patients without significant deterioration of AV conduction, except for 1 patient who had a pre-existing biventricular device and baseline LBBB and who developed complete heart block after right bundle ablation. Among patients with normal QRSd, there was no documented death or further VT recurrence during follow-up. At final follow-up, of the 6 patients with normal QRSd, 2 had ICDs, 1 had a biventricular ICD, and 3 had no device. The biventricular ICD was implanted at another hospital, and both of the regular ICDs were implanted for secondary prevention (syncope with documented sustained VT) before our procedure. Device interrogation at last follow-up showed that only 1 of these patients was pacemaker dependent.
Normal HV interval
Table 2 shows the characteristics of these patients. At baseline, 25 patients (78%) had a prolonged HV interval (>55 ms; mean 78 ± 11 ms; range 60 to 100 ms) and 7 (22%) had a normal HV interval (≤55 ms; mean 53 ± 2 ms; range 48 to 54 ms). There was no difference in follow-up duration between the 2 groups (90 ± 36 months in the normal HV group vs. 94 ± 31 months in the group with prolonged HV). In patients with a normal HV interval during SR, the mean HV interval during tachycardia was significantly longer (53 ± 2 ms vs. 76 ± 4 ms; p < 0.0001). A further prolongation of HV interval during tachycardia was also seen in the group with prolonged HV (mean 78 ± 11 ms vs. 89 ± 13 ms; p = 0.009). Successful ablation of the RBB or LBB was performed in all patients without significant deterioration of AV conduction. In patients with a normal HV interval, there was only 1 death (due to malignancy) and no VT recurrence during follow-up. At final follow-up, 3 patients had an ICD, 1 had a biventricular ICD, and 2 had no device. Device interrogation at last follow-up showed none of these patients were pacemaker dependent.
Significance of normal LV function at baseline
Table 3 shows electrocardiographic characteristics of these patients. At baseline, 10 patients (31%) had normal LV function (LVEF ≥50%) and 22 (69%) had low LV function (LVEF <50%). The mean follow-up duration for the 2 groups was similar (90 ± 33 months in the normal LVEF group vs. 88 ± 32 months in the low LVEF group; p = 0.07). Of the 10 patients with a normal LVEF, 9 (90%) had BBRT only and 1 (10%) had multiple VT morphologies (3 VTs). All patients had successful ablation, with no recurrence of VT after ablation, and there were no deaths in this group. At final follow-up, 5 patients with normal LVEF had an ICD, and none were pacemaker dependent or required any ICD therapy. Of note, at long-term follow-up, deterioration in LV function was not seen in any patient with a preserved LVEF at baseline, even after iatrogenic bundle branch block (mean LVEF 61 ± 8% to 59 ± 5% after 90 ± 33 months of follow-up; p = 0.80).
This study reports the long-term outcome of patients with BBRT who underwent VT catheter ablation. Tailored ablation strategies of either the right or left bundle can cure BBRT without exacerbating progression of the conduction system disease. In the presence of low LVEF in patients with BBRT, other myocardial VTs are commonly seen (47%). When these VTs are targeted at the time of BBRT ablation, it can result in long-term VT-free survival; in the present study, only 2 patients (13%) had VT recurrence, and 2 (6%) had new VT. Importantly, and less commonly, BBRT can be found in patients with normal LVEF, and these patients are at much lower risk of additional myocardial VTs (10%; p = 0.005). To the best of our knowledge, this study describes the largest cohort of BBRT patients to date with the longest follow-up. Of note, patients with a normal HV interval and normal LV function have excellent long-term prognosis, with VT-free survival and no need for permanent pacing after ablation.
In prior studies, BBRT has been found to be associated with valvular disease, both with and without concomitant ventricular dysfunction (15). In our study, we found that LV dysfunction was present in 22 cases (69%), whereas 10 patients (31%) had normal ejection fraction. Pooled results of prior studies (Table 4) show that approximately 20% of BBRT patients can have normal LV function (1,4,8). Additionally, it was previously thought that apparent conduction system disease was necessary for the development of sustained BBRT, but this might not be the case (8). In our study, 22% of the patients had a normal HV interval, and significant prolongation of the HV interval only developed during tachycardia. This suggests either a functional or fixed conduction block in the His-Purkinje system could be sufficient to maintain a BBRT mechanism (3). Also, during ablation, if the morphology suggests BBRT, a catheter should be placed at the His location for quick and accurate diagnosis of the VT.
The long-term impact of iatrogenic bundle branch block in a patient with normal LVEF or conduction was previously unclear. There is a paucity of data in the literature regarding such patients (Table 4) (1,9,16). In this study, only 1 patient showed progressive worsening of the conduction properties in the long-term.
The above findings suggest 2 distinct subgroups of patients with BBRT: a group with structural heart disease or conduction system disease (or both) and another with normal heart structure and function. In either group, BBRT is curable by catheter ablation. Other VTs may be present, particularly in those patients with depressed LV function and apparent conduction system disease. Our data suggest that insertion of an ICD may be appropriate in patients who have other inducible VT or who have significant LV dysfunction (11). Although not powered to make this conclusion, the outcome of our study cohort could suggest that patients without impaired LV systolic function and with no other inducible VT might not require ICD insertion. This subgroup had excellent long-term outcomes for VT-free survival without deterioration of conduction properties during follow-up. However, larger cohorts of this unique subset of BBRT patients will be required before any conclusions can be drawn.
These findings emphasize the need for better risk-stratification strategies to identify patients who would most benefit, or who would be unlikely to benefit, from ICDs (17). This is particularly important given the comorbidities associated with ICD insertion, such as infection, inappropriate shocks, and lead- and device-related problems (1,8,11). Identification of subgroups of patients at higher or lower risk for sudden cardiac death will improve the clinical efficacy and cost-effectiveness of such therapies and help avoid complications (18).
This is an observational registry data and has the usual limitations inherent to observational studies. Detection and ascertainment biases are possible in our cohort, but the collection of data and follow-up were similar in all patients, and these influences should be reduced because the collection of outcome data was done via well-defined data sources. The incidence of BBRT observed in the current study is lower than the percentage seen in other observational studies. This could be because of the referral pattern and patient cohort seen in our tertiary referral center.
BBRT is curable by radiofrequency catheter ablation. Although BBRT is more often seen in patients with baseline conduction disease and cardiomyopathy, the absence of a prolonged HV interval or intraventricular conduction defects during SR should not exclude the consideration of BBRT. Patients with BBRT with no apparent conduction abnormalities and those with no structural heart disease have excellent long-term prognosis and VT-free survival.
COMPETENCY IN MEDICAL KNOWLEDGE: BBRT is an uncommon mechanism of VT. The clinical and electrophysiological characteristics can be variable and should be recognized given the serious clinical manifestations and potential for cure.
TRANSLATIONAL OUTLOOK 1: Catheter ablation is easily performed and is an effective treatment for BBRT. BBRT can be seen in patients with normal cardiac structure and function. Clinical and electrophysiological features should enable one to correctly diagnose and treat BBRT with good long-term VT elimination.
TRANSLATIONAL OUTLOOK 2: Patients with BBRT with no apparent conduction abnormalities and those with no structural heart disease have excellent long-term prognosis and VT-free survival. Identification of risk groups could improve clinical efficacy and cost-effectiveness of therapies provided and avoid complications.
Dr. Pathak is supported by an early career fellowship from the National Health and Medical Research Council of Australia. This work was supported in part by the Mark S. Marchlinski EP Research Fund at the University of Pennsylvania. All authors have reported that they have no relationships relevant to the contents of this paper to disclose. This study was presented as an abstract at the Heart Rhythm Society Meetings, Chicago, Illinois (May 2017).
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
- bundle branch re-entrant tachycardia
- implantable cardioverter-defibrillator
- left bundle branch
- left bundle branch block
- left ventricular
- left ventricular ejection fraction
- QRS duration
- right bundle branch
- sinus rhythm
- ventricular tachycardia
- Received April 17, 2017.
- Revision received November 27, 2017.
- Accepted November 28, 2017.
- 2018 American College of Cardiology Foundation
- Blanck Z.,
- Jazayeri M.,
- Dhala A.,
- Deshpande S.,
- Sra J.,
- Akhtar M.
- Roberts J.D.,
- Gollob M.H.,
- Young C.,
- et al.
- Narasimhan C.,
- Jazayeri M.R.,
- Sra J.,
- et al.
- Caceres J.,
- Jazayeri M.,
- McKinnie J.,
- et al.
- Tchou P.,
- Jazayeri M.,
- Denker S.,
- Dongas J.,
- Caceres J.,
- Akhtar M.
- Touboul P.,
- Kirkorian G.,
- Atallah G.,
- et al.
- Cohen T.J.,
- Chien W.W.,
- Lurie K.G.,
- et al.
- Kober L.,
- Thune J.J.,
- Nielsen J.C.,
- et al.
- Kusa S.,
- Taniguchi H.,
- Hachiya H.,
- et al.