Author + information
- Received April 15, 2015
- Revision received May 11, 2015
- Accepted May 21, 2015
- Published online October 1, 2015.
- Ahmed Karim Talib, MD, PhD∗,
- Akihiko Nogami, MD, PhD∗∗ (, )
- Suguru Nishiuchi, MD†,
- Shinya Kowase, MD‡,
- Kenji Kurosaki, MD‡,
- Yumie Matsui, MD§,
- Satoshi Kawada, MD‖,
- Atsuyuki Watanabe, MD‖,
- Masatsugu Nozoe, MD, PhD¶,
- Kikuya Uno, MD, PhD#,
- Atsuhiko Yagishita, MD∗∗,
- Yasuteru Yamauchi, MD∗∗,
- Yoshihide Takahashi, MD††,
- Taishi Kuwahara, MD††,
- Atsushi Takahashi, MD††,
- Koji Kumagai, MD, PhD†,
- Shigeto Naito, MD†,
- Tetsuya Asakawa, MD‡‡,
- Yukio Sekiguchi, MD∗ and
- Kazutaka Aonuma, MD, PhD∗
- ∗Cardiovascular Division, Faculty of Medicine, University of Tsukuba, Tsukuba, Japan
- †Division of Cardiology, Gunma Prefectural Cardiovascular Center, Maebashi, Japan
- ‡Department of Heart Rhythm Management, Yokohama Rosai Hospital, Yokohama, Japan
- §Department of Cardiology, Saiseikai Izuo Hospital, Osaka, Japan
- ‖Division of Cardiology, Fukuyama City Hospital, Fukuyama, Japan
- ¶Division of Cardiology, Cardiovascular and Aortic Center, Saiseikai Fukuoka General Hospital, Fukuoka, Japan
- #Sapporo Heart Center, Sapporo Cardiovascular Clinic, Sapporo, Japan
- ∗∗Department of Cardiology, Musashino Red Cross Hospital, Tokyo, Japan
- ††Cardiovascular Center, Yokosuka Kyosai Hospital, Yokosuka, Japan
- ‡‡Cardiology Division, Yamanashi Kosei Hospital, Yamanashi, Japan
- ↵∗Reprint requests and correspondence:
Dr. Akihiko Nogami, Cardiovascular Division, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8575, Japan.
Objectives This study sought to demonstrate the prevalence, mechanism, and electrocardiographic and electrophysiological characteristics of upper septal idiopathic left fascicular ventricular tachycardia (US-ILVT).
Background ILVT is classified into left anterior and posterior types with no clear data about US-ILVT.
Methods Among 193 ILVT patients, we identified 12 patients (6.2%; age 41 ± 22 years, 7 men) with US-ILVT.
Results Of 12 patients with US-ILVT, 6 patients (50%) had previous history of radiofrequency catheter ablation for common ILVT. Sustained VT (cycle length: 349 ± 53 ms) was seen in all patients with a QRS interval slightly wider (104 ± 18 ms) than that during sinus rhythm (90 ± 19 ms). The VT exhibited an identical QRS configuration as sinus rhythm in 6 (50%) and incomplete right bundle branch block configuration in another 6. His-ventricular interval during VT was always shorter than that during sinus rhythm (27 ± 5 ms vs. 47 ± 10 ms). Purkinje potentials were activated in a reverse direction to that of common ILVT; namely, the diastolic potential (P1) was activated retrogradely but the pre-systolic potential (P2) was activated antegradely. At the left upper-middle ventricular septum, P1 potential was recorded during VT, preceding the QRS by 54 ± 20 ms. Radiofrequency catheter ablation at that site eliminated the VT with no recurrence during a 58 ± 35 months of follow-up.
Conclusions US-ILVT is an identifiable VT that shares common criteria with ILVT and has a narrow QRS interval. Some US-ILVT cases appeared after common ILVT ablation. It is a reverse type of common ILVT (orthodromic form) with baseline morphological abnormalities that might provide a potential substrate for such VT.
Verapamil-sensitive fascicular tachycardia is the most common form of idiopathic left ventricular tachycardia (ILVT). It was first recognized as an electrocardiographic entity by Zipes et al. (1), who defined its morphology as right bundle branch block (RBBB) and left axis deviation. That VT was successfully suppressed by radiofrequency catheter ablation (RFCA) at the vicinity of the left posterior fascicle (LPF) (2). Another less common type of fascicular VT, characterized by RBBB morphology and right axis deviation, which has been described by Ohe et al. (3), can be suppressed by RFCA at the left anterior fascicle (LAF) area (3,4).
These types of VT have been studied almost extensively; however, little is known about the prevalence, mechanism, and surface electrocardiographic (ECG) and electrophysiological characteristics of upper septal (US) fascicular idiopathic left ventricular tachycardia (US-ILVT) (5–7). After analyzing data from 9 different experienced centers, the purpose of this study was to clarify the above-mentioned points along with the results of long-term follow-up after RFCA.
From February 2006 through September 2014, in a multicenter study analyzing data of 193 patients who underwent electrophysiological study of verapamil-sensitive fascicular VT, we identified 12 patients (7 men, mean age: 41 ± 22 years) who had distinct ECG and electrophysiological characteristics of US-ILVT. In each patient, after detailed medical history and examination, structural heart diseases were ruled out by a standard investigation protocol including 12-lead ECG, chest radiographs, echocardiography, cardiac computed tomography when appropriate, and coronary angiography when indicated. The study was approved by the local research ethics committees of the participating institutes, and all patients gave their written informed consent.
An electrophysiology study was performed after withdrawing all antiarrhythmic drugs for ≥5 half-lives. Standard multielectrode catheters were placed in the high right atrium, His-bundle region, and right ventricular septum. Programmed atrial and ventricular stimulation was performed using a maximum of 3 extrastimuli at 2 different driven cycle lengths from the right atrium and right ventricular septum. If sustained VT was not induced, the stimulation was repeated during isoproterenol infusion (0.5 to 2.0 μg/min).
Mapping and ablation
RFCA was performed in 10 patients. In the remaining 2 patients (#4 and #6) who had the same clinical and electrophysiological characteristics of US-ILVT, RFCA was not performed, but an electrophysiological study was done. Through a femoral arterial approach, a 7-F quadripolar steerable electrode catheter with a 4-mm tip and 2-mm interelectrode spacing between the distal 2 electrodes was positioned at the interventricular septum of the left ventricle to record the intracardiac electrograms, as well as to pace and ablate. In a few patients, an octapolar or decapolar steerable electrode catheter with 1.25-mm electrode length and 2-mm interelectrode spacing (Boston Scientific, Natick, Massachusetts; or St. Jude Medical, Minnetonka, Minnesota) was positioned at the left ventricular septum (LVS). A 3-dimensional electroanatomical system (CARTO, Biosense-Webster, Diamond Bar, California; or NavX system, EnSite, St. Jude Medical) was used in 9 patients.
In 8 of 10 patients who underwent RFCA, an irrigation catheter was used. In the majority of the cases, RF energy was delivered for 30 to 120 s during sinus rhythm to avoid any left bundle or atrioventricular block. Maximum power was 50 W. When using a nonirrigated catheter, RFCA was done in temperature-controlled mode at a maximal target temperature of 55°C and maximum power of 50 W. We performed RFCA in this region using a low power output (i.e., 10 W), which was increased gradually while carefully monitoring for the development of junctional rhythm or atrioventricular block. After ablation, programmed stimulation was repeated with and without isoproterenol infusion.
The patients were monitored for 1 to 3 days after the ablation. After discharge, the patients were followed up in an outpatient clinic at an interval of 3 months in the first year after ablation; after that, they were followed up at approximately yearly intervals. The follow-up period was for 58 ± 35 months (12 to 112 months) without using any antiarrhythmic medications.
The values are given as the mean ± SD. The significance of the differences between groups was assessed by the Student t test. Changes in electrophysiological parameters were analyzed by paired Student t test. A p value of <0.05 was considered statistically significant.
Prevalence and clinical characteristics
Among 193 patients who were referred for the therapy of ILVT, 12 patients (6.2%) had US type. The patient characteristics are shown in Table 1. There were 7 men and 5 women between 15 and 83 years of age. All had normal LV systolic function. Ten patients (83%) presented with palpitations and 2 presented with syncope associated with VT. Intravenous administration of verapamil was used in 9 patients, and was effective in VT termination in all. Six patients (50%) had a previous history of common ILVT, which was treated by RFCA. Among them, 4 patients had 2 ablation sessions for common ILVT.
Surface ECG findings
The baseline ECG exhibited sinus rhythm in all patients with QRS duration of <120 ms. A notable finding was the presence of a Q-wave in the inferior leads and/or an S-wave in lead I and/or aVL. No structural abnormalities could explain these findings in 6 patients, whereas in the remaining 6 patients, those morphological changes developed after previous RFCA session(s) at the LPF vicinity, which resulted in new or deepening Q waves in the inferior leads and/or appearance of or deepening S waves in lead I and/or aVL (Table 1, Figure 1).
As shown in Table 2, all patients presented with sustained monomorphic VT (cycle length: 349 ± 53 ms) with QRS duration slightly wider than that during sinus rhythm (104 ± 18 ms during VT vs. 90 ± 19 ms during the VT, p = 0.882). Figure 2 shows representative ECG during sinus rhythm and VT. The VT can be classified into 2 distinct QRS morphological types: 1) type 1: 6 patients (50%) had an identical precordial QRS configuration to that during sinus rhythm with right axis deviation in 4 patients (Figure 2A) and an identical QRS configuration and axis to that of sinus rhythm in the remaining 2 patients (Figure 2B); and 2) type 2: 6 patients (50%) had incomplete RBBB pattern with an identical QRS axis to that during sinus rhythm in 4 patients (Figure 2C) and normal QRS axis but slightly different from that during sinus rhythm in the remaining 2 patients (Figure 2D).
Electrophysiological characteristics, mapping, and ablation
The electrophysiological findings are summarized in Table 3. All the treated patients had normal His-ventricular (H-V) interval during sinus rhythm. The US-VT was induced spontaneously during catheter manipulation in 3 patients, by burst atrial pacing in 4 patients, burst ventricular pacing in 2 patients, and by programmed ventricular extrastimulus in 2 patients. During VT, retrograde activation of the His bundle was recorded before the onset of the QRS complex with an H-V interval that was always shorter during VT than that during sinus rhythm (27 ± 10 ms vs. 47 ± 5 ms, p < 0.001). The electrophysiological characteristics are summarized in Table 4. Overdrive pacing from the right atrium was attempted in 9 patients; however, ventricular capture was seen in only 4 patients because of atrioventricular block during atrial entrainment pacing in the rest of the patients. In these 4 patients, there was evidence of constant and progressive fusions consistent with transient entrainment and suggesting a reentrant mechanism of the VT.
Typical surface ECG and intracardiac electrograms of a patient with US-ILVT are shown in Figure 3 (Patient #9). During sinus rhythm, baseline ECG exhibited normal conduction intervals. The patient developed common ILVT in which his ECG showed RBBB configuration and left axis deviation. A few years later, after 2 ablation sessions for common ILVT, the patient developed US-ILVT. ECG during sinus rhythm showed an S-wave in lead I and a Q-wave in leads III and aVF. ECG during VT exhibited incomplete RBBB pattern and an identical QRS axis to that of sinus rhythm (Figure 3A). Figure 3B shows intracardiac recordings during sinus rhythm, common ILVT, and US-ILVT. During sinus rhythm, a decapolar electrode catheter on the LVS revealed that the conduction propagated antegradely generating a pre-systolic potential (P2), which was recorded after the His-bundle potential and before the onset of the QRS complex, suggesting LPF and the adjacent Purkinje potentials (Figure 3B, left panel). During common ILVT, a diastolic potential (P1) and pre-systolic potential (P2) were seen (Figure 3B, middle panel). Whereas P1 was recorded earlier from the proximal than from the distal electrodes, P2 was recorded earlier from the distal than from the proximal electrodes. His-bundle potential was recorded after the onset of the QRS complex (negative H-V interval). During US-ILVT (Figure 3B, right panel), P1 was recorded earlier from the distal than from the proximal electrodes, whereas P2 was recorded earlier from the proximal than from the distal electrodes, similar to that during sinus rhythm. His-bundle potential preceded the onset of the QRS complex by 32 ms, which was shorter than that during sinus rhythm (45 ms). Figure 3C shows the position of a decapolar catheter on the LVS. An ablation catheter was positioned at the upper-middle ventricular septum (Figure 4A) where the diastolic P1 preceded the onset of the QRS interval by 52 ms. Entrainment from this site resulted in concealed fusion and post-pacing interval–ventricular tachycardia cycle length (VTCL) difference of 5 ms and an interval between the pacing stimulus and QRS onset of 52 ms, equal to the P-QRS interval during VT (Figure 4B). VT was slowed and terminated by RF energy application at this site within a few seconds (Figure 4C) and VT became noninducible.
Anatomically, the ablation site was selected at the left upper-middle ventricular septum, mid-way between the His-bundle recording site and the LV apex. Ablation at the most basal portion of the LVS was not performed to avoid left bundle/atrioventricular block. Electrophysiologically, the ablation site was characterized by the presence of the diastolic Purkinje potential (P1) during VT and the absence of both His-bundle and atrial potentials during sinus rhythm. A diastolic Purkinje potential (P1) at the successful ablation site preceded the QRS onset (P1-QRS) by 55 ± 21 ms (Table 4). Entrainment pacing at the ablation site was performed in 9 patients and resulted in concealed fusion in 4 patients and a post-pacing interval–VTCL difference of 16 ± 5 ms. In the remaining 5 patients, selective capture of the diastolic Purkinje potential (P1) was not obtained.
Upper septal ILVT subtypes
Although almost similar electrophysiological findings of the successful ablation site were observed in all patients who underwent RFCA, the H-V interval during VT and the VTCL were relatively longer in morphological type 1 group (an identical precordial QRS configuration to that during sinus rhythm) than in type 2 group (incomplete RBBB pattern); H-V interval: 31 ± 11 ms versus 22 ± 5 ms; VTCL: 368 ± 61 ms versus 329 ± 40 ms, for type 1 and type 2 groups, respectively.
Complications and follow-up
No atrioventricular block was seen during or after the RFCA. In 1 patient (#12), transient left bundle branch block was seen, which recovered 8 s after the discontinuation of the RF energy delivery. One patient (#9) developed a mild worsening of an already existing right axis deviation. During 58 ± 35 months (12 to 229 months) of follow-up of ablated patients, no VT recurrence was seen without using any antiarrhythmic medications in any patient. In 2 patients (#4 and #6) who did not undergo RFCA, treatment with beta-blockers and verapamil was started. There have been no recurrences during a period of 11 and 13 months, respectively.
This is the first study to demonstrate that 6.2% of ILVT cases are due to US-ILVT, which is characterized as follows: 1) one-half of the patients had a previous history of RFCA for common ILVT, mostly multiple ablation sessions; 2) normal sinus ECG and intracardiac conduction intervals with the presence of minor morphological abnormalities including Q waves in the inferior limb leads and/or S waves in limb leads I and/or aVL; 3) VT exhibiting an identical precordial R-wave progression to that of sinus rhythm or incomplete RBBB morphology; 4) during VT, retrograde activation of the His-bundle before the QRS onset occurring with an H-V interval significantly shorter during VT than that during sinus rhythm; 5) the presence of common characteristics of re-entrant common ILVT such as inducibility with ventricular and/or atrial stimulation, entrainment, and verapamil-sensitivity; and 6) successful VT ablation at the left upper-middle ventricular septum, where the diastolic Purkinje potential (P1) was recorded during VT.
Although sporadic cases of left US-VT have been reported (5–7), to the best of our knowledge, this is the first study to fully demonstrate the prevalence, mechanism, and clinical and electrophysiological characteristics of this type of VT.
In all patients, although the baseline ECG exhibited normal conduction intervals, minor morphological changes in the limb leads were seen including Q waves in the inferior leads and/or S waves in leads I and/or aVL. Interestingly, these ECG changes developed after common ILVT ablation in 50% of the patients, indicating that the local RF energy that was applied to the Purkinje fiber network of the LPF area resulted in conduction damage to the LPF and created those ECG changes. It is noteworthy that in 6 patients who had undergone RFCA for common ILVT, the original VT was completely cured. In the rest of the patients, these morphological abnormalities were of an unknown etiology. Whether iatrogenic or idiopathic, these morphological changes indicate local conduction damage that might have created a slow conduction area and provided a substrate for re-entry resulting in the US-ILVT.
ECG characteristics of upper septal ILVT
The QRS morphology during this VT was quite characteristic, with a narrow QRS complex and either an identical QRS configuration to that of sinus rhythm or an incomplete RBBB configuration, and the limb leads exhibited either normal axis or right axis deviation. It is known that ILVT arising from the intraventricular conduction system typically have RBBB morphology and left (common form) or right (uncommon form) axis depending on the Purkinje networks involved (1–4). In the US type, the presence of a narrow QRS interval, almost identical QRS morphology to that of sinus rhythm, and normal frontal plane axis indicate a near-normal activation using the US His-Purkinje network that is not significantly different from the normal ventricular depolarization. Some of the present cases had been misdiagnosed as supraventricular tachycardia in previous hospitals because of their QRS configuration and their verapamil-sensitivity. The present study calls attention to the observation that narrow-QRS tachycardia can have a ventricular origin and should be considered, especially in patients with a history of RFCA of Purkinje-related arrhythmias.
In addition to the presence of the general electrophysiological characteristics of ILVT, 2 important characteristics were seen in US-ILVT. First, during VT, retrograde activation of the His bundle was recorded before the onset of the QRS complex with an H-V interval that was significantly shorter than that during sinus rhythm, and this can be used to differentiate this tachycardia from supraventricular tachycardia. This is in contrast to the common form of ILVT where His potential is usually recorded after the QRS onset (8). Second, during the VT, diastolic Purkinje potential preceding the His potential was recorded from the left upper-middle ventricular septum, where the left bundle potential was recorded during sinus rhythm. Targeting that diastolic potential resulted in VT termination.
Mechanism of the tachycardias and circuit diagram
Overwhelming evidence suggests that ILVT is caused by a re-entrant circuit incorporating the Purkinje system with an excitable gap and a slow conduction area (8,9). To demonstrate the VT circuit in US-VT, the patient mentioned in Figures 3 and 4 is taken as an example. During sinus rhythm, the conduction propagated antegradely (proximal to distal) generating a pre-systolic Purkinje potential (P2) followed by the ventricular activation (Figure 3B, left panel). During common ILVT, the activation propagated antegradely from the basal to the apical site on the LVS (Figure 3B, middle panel). On the other hand, in US-ILVT, the activation sequence of P1 was from the distal to proximal septum, whereas the activation sequence of P2 was from the proximal to distal septum, similar to that during sinus rhythm (Figure 3B, right panel).
Taking these findings together, the hypothetical circuit of the US-ILVT is depicted in Figure 5. During sinus rhythm (Figure 5A), the sinus impulse propagates antegradely down the LPF giving rise to P2 then the activation goes from P2 to P1 at the point of fusion; therefore, P1 is buried in the local ventricular activation (8). During common ILVT (Figure 5B), P1 and P2 are activated in the reverse direction, which explains why the activation sequenced of P2 is reversed during sinus rhythm and VT. We suggest that the US-ILVT (Figure 5C) is a “reverse common” or “fast-slow” variant of ILVT, where both LAF and LPF are the antegrade limbs of the re-entrant circuit, whereas the retrograde activation occurs via the abnormal Purkinje fiber at middle fascicular area. Although His bundle and right bundle branch are bystanders, they were activated just after the activation of LAF and LPF. This explains why this VT exhibits a narrow QRS configuration with a near-normal activation sequence and inferior axis. P1 represents the common retrograde limb of the circuit during the VT and can be a suitable ablation target.
Furthermore, the US-ILVT cases with identical precordial QRS configuration to that of sinus rhythm (type 1) had a longer H-V interval than did those with incomplete RBBB configuration (type 2); hence, the upper turn around site of the former is more basal than that of the latter. Longer VTCL in type 1 group further supports such a hypothesis and indicates a larger re-entry circuit in morphological type 1 than that in type 2 group (Figure 6).
Arrhythmogenesis of upper septal ILVT
Although the presence of a trifascicular (anterior, posterior, and middle) left Purkinje system was proved by many histological studies mainly on the basis of animal or post-mortem examinations (10,11), the exact cause of the rarity of US-ILVT is unknown. Two possibilities may explain this. 1) A modification of the conduction properties of the Purkinje network without blocking the conduction of the LPF. Because 50% of our patients had previous RFCA session(s) in the vicinity of the LPF, it is possible that, after ablating the distal LPF for the treatment of common ILVT, the new VT occurred more proximally in the His-Purkinje system, rendering the QRS complex narrower. It is noteworthy that our ILVT cases were ablated in experienced centers and the VT was successfully ablated without any further recurrence of the original VT. For instance, Figure 1 shows the baseline and post-ablation ECG of Patient #9 who had ablation resulting in loss of the S waves with the appearance of new Q waves in the inferior limb leads and deeper S waves in leads I and aVL, indicating local damage to the distal LPF. Although such a finding has been proposed as one of effective endpoints for ablation of ILVT in some studies (12), it pre-disposed the patient to US-ILVT. And, 2) recently, in an in vivo study of the isolated conduction within the left His-Purkinje system in 25 patients, Long et al. (13) found that the LBB bifurcated into 2 divisions conforming to the LAF and LPF in 23 patients, and into 3 divisions conforming to LAF, left middle, and LPF in only 2 patients, suggesting a low incidence of septal fascicle in human hearts that may partially explain such a very low incidence of US-VT.
The most significant limitation was the sample size. However, to the best of our knowledge, this was the first study to reflect the disease rarity and open a new window for mapping and ablation of US-ILVT, for which, catheter ablation was avoided for fear of creating an atrioventricular block. Furthermore, depending on the operator’s preference, good entrainment mapping was not obtained in all cases.
US-ILVT is a rare but identifiable VT that shares the common criteria of ILVT such as inducibility with ventricular and atrial stimulation, entrainment, and verapamil-sensitivity with a narrow QRS interval. It is a reverse type of common ILVT (orthodromic form), using the LPF (or LAF) as the antegrade limb of the re-entry circuit, with baseline morphological abnormalities that might provide a potential substrate for such type of VT.
COMPETENCY IN MEDICAL KNOWLEDGE: US-ILVT is a rare but identifiable VT and is a reverse type of common ILVT (orthodromic form) with baseline morphological abnormalities that might provide a potential substrate for such type of VT.
TRANSLATIONAL OUTLOOK: Whole circuits of common (posterior type), uncommon (anterior type), ILVT and US-ILVT are still unclear, especially the slowest and longest part in the circuit. Further studies are needed to explore its biomedical mechanism.
Dr. Nogami has received lecture honoraria from St. Jude Medical and Boston Scientific; and an endowment from Medtronic and Johnson & Johnson. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- idiopathic left ventricular tachycardia
- left anterior fascicle
- left posterior fascicle
- left ventricle
- left ventricular septum
- right bundle branch block
- radiofrequency catheter ablation
- upper septal
- upper septal idiopathic left ventricular tachycardia
- ventricular tachycardia
- ventricular tachycardia cycle length
- Received April 15, 2015.
- Revision received May 11, 2015.
- Accepted May 21, 2015.
- American College of Cardiology Foundation
- Nakagawa H.,
- Beckman K.J.,
- McClelland J.H.,
- et al.
- Ohe T.,
- Shimomura K.,
- Aihara N.,
- et al.
- Nogami A.,
- Naito S.,
- Tada H.,
- et al.
- Tsuchiya T.,
- Okumura K.,
- Honda T.,
- Isawa A.,
- Ashikaga K.
- Long D.Y.,
- Dong J.Z.,
- Sang C.H.,
- et al.