Author + information
- Received March 19, 2018
- Revision received May 23, 2018
- Accepted June 13, 2018
- Published online September 17, 2018.
- Taihei Itoh, MD and
- Takumi Yamada, MD, PhD∗ ()
- ↵∗Address for correspondence:
Dr. Takumi Yamada, Division of Cardiovascular Disease, University of Alabama at Birmingham, FOT 930A, 510 20th Street South, Birmingham, Alabama 35294-0019.
Objectives This study sought to reveal the characteristics and radiofrequency catheter ablation (RFCA) outcomes of multifocal His-Purkinje system (HPS) ventricular arrhythmias (VAs).
Background The details of those VAs, especially the safety and efficacy of their RFCA treatment, remain unclear.
Methods Thirty consecutive patients who underwent RFCA of focal HPS VAs between 2010 and 2016 (unifocal = 24, multifocal = 6) were studied by measuring the electrophysiological variables within the HPS.
Results Multifocal premature ventricular contractions (n = 1) and ventricular fibrillation (VF) (n = 5) were identified in the left posterior (n = 6), anterior (n = 4), and septal fascicles (n = 1), as well as the basal left bundle branch (LBB) (n = 2) and right bundle branch (RBB) (n = 2). In 2 patients with unifocal VAs and 4 patients with multifocal VAs, preferential conduction from an origin within the proximal fascicle (n = 4) or LBB (n = 2) to ≤3 breakout sites in the distal fascicles occurred with split or fractionated Purkinje potentials and/or conduction block at the site of origin. Among the multifocal VAs, 11 fascicle VAs, 1 RBB VA, and 1 LBB VA were successfully ablated with fascicular and/or bundle branch block, and complete atrioventricular block (CAVB), respectively. In the remaining LBB VAs and RBB VAs, RFCA was abandoned to avoid CAVB. Recurrence of ablated VAs or the incidence of VF did not differ between the unifocal and multifocal HPS VAs. Freedom from any HPS VA after RFCA was significantly higher in the patients with unifocal VAs than in the patients with multifocal VAs (92% vs. 33%; p = 0.001).
Conclusions Multifocal HPS VAs could occur and often present with preferential conduction from proximal origins to distal breakout sites within the HPS with abnormal Purkinje potentials and/or conduction properties. RFCA was effective but was limited by the risk of HPS impairment.
Focal ventricular arrhythmias (VAs) can originate from the His-Purkinje system (HPS) and trigger ventricular fibrillation (VF) with a short coupling interval during the vulnerable period of cardiac repolarization (1–8). These VAs most often originate from foci located within the distal HPS, including the left anterior (LAF), septal (LSF), and posterior fascicles (LPF), as well as the peripheral right bundle branch (RBB) (1–8), but they rarely originate from the basal left bundle branch (LBB) or proximal RBB (9,10). These focal VAs can be eliminated by radiofrequency catheter ablation (RFCA) without any significant impairment of the HPS (1–11). During the HPS VAs, polymorphic QRS morphologies are often observed as a result of variable conduction from a single VA origin to multiple breakout sites within the His-Purkinje network (1–3,9–11). However, multifocal HPS VAs have not been well characterized, and the efficacy and safety of their RFCA treatment remain unclear. The purpose of this study was to investigate the incidence, the clinical, electrocardiographic, and electrophysiological characteristics of multifocal HPS VAs, as well as the outcomes of RFCA on these VAs.
The study population consisted of 30 consecutive patients (15 men; age 55 ± 16 years) who underwent RFCA of symptomatic focal HPS VAs between April 2010 and December 2016. The baseline characteristics, including age, sex, left ventricular function, nature of the clinical arrhythmias, and 12-lead electrocardiograms during the VAs, were recorded. The institutional review board approved the study protocol, and all patients provided written informed consent for the procedure. All antiarrhythmic drugs, except amiodarone, were discontinued for at least 5 half-lives before the study.
For mapping and pacing, multipolar catheters were positioned in the coronary sinus and His bundle region via the right femoral vein. A 3.5-mm tip irrigated ablation catheter (Navistar ThermoCool, Biosense Webster, Diamond Bar, California) was introduced into the left or right ventricle through the right femoral artery or vein, respectively. During the procedures in the left ventricle, intravenous heparin was administered to maintain an activated clotting time of >300 s. When few premature ventricular contractions (PVCs) were observed at baseline, induction of VAs was attempted by burst ventricular pacing, with the addition of isoproterenol infusion at 2 to 4 μg/min. A mechanism of sustained ventricular tachycardia was diagnosed as focal by the absence of any entrainment phenomenon during overdrive pacing (Figure 1).
Mapping and RFCA
Electroanatomical mapping was performed using a 3-dimensional mapping system (CARTO, Biosense Webster). When VAs exhibited multiple different QRS morphologies, activation mapping was performed during the VAs with the most dominant QRS morphology. When Purkinje potentials preceding the QRS onset were recorded during both the VAs and sinus rhythm, those VAs were considered to originate from the HPS (Figure 2). The 3 fascicles in the left ventricle were delineated with an anatomic consideration on the 3-dimensional map by mapping pursuing the Purkinje potentials (Figure 3). Activation mapping along the fascicles was performed with a combination of 3-dimensional mapping and conventional mapping that compared the activations recorded from the distal and proximal electrode pairs of the mapping catheter. With this mapping technique, even a few mapping points could reveal an activation pattern of the Purkinje potentials and could identify a site of an HPS VA origin. When a QRS complex did not follow the Purkinje potentials, it was considered that a conduction block between the VA origin and ventricular myocardium (P-Vpvc block) occurred. When 2 separated Purkinje potentials with a short interval were followed by a QRS complex, those Purkinje potentials were defined as split Purkinje potentials. Pace mapping was performed using the distal bipolar electrodes at a pacing cycle length of 500 ms and at the minimum stimulus amplitude required for consistent capture, up to an output of ≤20 mA and pulsewidth of 2.0 ms. The pace map score was determined as the number of leads with identical heights of the R wave/depth of the S wave (R/S) ratio match (12 represented a perfect R/S ratio match in all 12 leads) and the number of leads with fine notching matches in the 12-lead electrocardiogram (perfect pace mapping was equal to 24 points). A VA-matched pace map was defined as a pace map with a score of ≥21.
Irrigated radiofrequency energy was delivered at sites that exhibited the earliest bipolar activation and/or a local unipolar QS pattern preceding the QRS onset during the VAs. Irrigated radiofrequency current was delivered in the power-control mode, starting at 30 W, with an irrigation flow rate of 30 ml/min. The radiofrequency power was titrated up to 50 W, with the goal of being able to achieve a decrease in the impedance of 8 to 10 ohms, and with care taken to limit the temperature to <40°C. When an acceleration or reduction in the incidence of VAs was observed during the first 10 s of the application, the radiofrequency delivery was continued for 60 to 120 s. Otherwise, the radiofrequency delivery was terminated, and the catheter was repositioned. When VAs with different QRS morphologies remained after the successful ablation of the first VAs, mapping and RFCA were performed with the same protocol. The endpoint of the RFCA was the elimination and noninducibility of VAs during an isoproterenol infusion of 2 to 4 μg/min and burst ventricular pacing (to a cycle length of ≥240 ms).
Definition of multiple HPS VA origins and a single HPS VA origin with preferential conduction to multiple breakout sites
When the Purkinje potentials preceding the QRS onset during VAs or VA-matched pace maps were recorded at a fascicle, it was considered that a breakout site should be located at that fascicle. When the activation of a Purkinje potential recorded during VAs was earlier at a distal site than a proximal site in the same fascicle, it was considered that the HPS VA origin should be located at that fascicle. When earlier Purkinje potentials during VAs were recorded or successful RFCA of VAs was achieved at a fascicle distinct from that with a breakout or proximal LBB, it was suggested that there should be preferential conduction from the origin of the VA to the breakout site within the HPS. Based on these definitions, when multiple HPS VA origins were identified in a patient, the patient was determined to have multifocal HPS VAs, which were differentiated from VAs with preferential conduction from a single origin to multiple breakout sites. Those definitions could rule out an induction of an exit block of the first HPS VA origin by mapping and RFCA, with a shift of the breakout site to a distant location.
In this study, the intrinsic QRS morphologies recorded at baseline were mapped first. However, different QRS morphologies from those at baseline could occur during mapping and after RFCA for 2 possible reasons. First, suppression or disappearance of the major VAs by HPS impairment or elimination of the origins by RFCA could allow VAs originating from a different origin to emerge. Second, HPS impairment could change the conduction pattern within the HPS, resulting in different QRS morphologies. In this study, QRS morphologies newly developed after the instrumentation in the ventricles or RFCA were also studied when it was assured that those QRS morphologies occurred with the first mechanism.
Unifocal and multifocal HPS VAs were classified into subgroups according to the presence of preferential conduction within the HPS. Unifocal HPS VAs without and with preferential conduction were grouped as UF-1 and UF-2, respectively. Multifocal HPS VAs without and with preferential conduction to multiple different fascicles or a single fascicle were grouped as MF-1, MF-2, or MF-3, respectively.
Simultaneous 12-lead electrocardiograms during VAs and pace mapping were recorded digitally at sweep speeds of 200 mm/s in all patients for an offline analysis. The QRS morphologies, including a bundle branch block pattern, the axis, and the configuration in leads I and V6, were examined. The QRS duration and coupling interval of VAs were measured with electronic calipers by 2 experienced investigators blinded to the site of the origin. For analysis of the coupling intervals, the first 10 consecutive beats of the clinical VAs that were recorded during the procedure were selected.
Follow-up after the procedure included clinic visits with 12-lead electrocardiography and 24-hour Holter recording, and telephone calls to all patients and their referring physicians. All patients who reported symptoms underwent 24-hour Holter recording or event monitoring to document the cause of the symptoms. Successful ablation was defined as no recurrence of any VAs during >6 months of follow-up.
Continuous variables were expressed as mean ± SD. Comparisons of the continuous variables between the 2 groups were analyzed with the 2-tailed paired t-test or Mann-Whitney U test. Comparisons of the continuous variables among the 3 groups were analyzed with 1-way analysis of variance. When appropriate, a further post hoc pairwise comparison was performed by Tukey’s test. The categorical variables, which were expressed as numbers and percentages in the different groups, were compared with a chi-square test and Yates correction, if necessary. Freedom from recurrent HPS VAs or VF was determined by Kaplan-Meier analyses with a log-rank test. Statistical significance was selected as p < 0.05.
In 19 of the 30 patients (9 men; age: 52 ± 15 years) with focal HPS VAs, there was no evidence of structural heart disease. In the remaining 11 patients (6 men; age 61 ± 16 years), 3 had a history of coronary artery disease with a normal left ventricular ejection fraction. Eight patients had ischemic (n = 5) or nonischemic cardiomyopathy (n = 3). One patient had PVC-induced cardiomyopathy. The clinical VAs presented as sustained ventricular tachycardia in 1 patient, nonsustained ventricular tachycardia in 4 patients, and PVCs in 25 patients. Fourteen of the 30 patients (7 men; age 55 ± 16 years) had a history of documented VF, and 7 of those patients had no structural heart disease. All 14 patients received an implantable cardioverter-defibrillator for secondary prevention of VF before (n = 9) or after (n = 5) RFCA. The remaining 16 patients had no history of VF or unexplained syncope. All 30 patients had no family history of sudden cardiac death.
Mapping and RFCA
Activation mapping identified all VA origins at the sites where Purkinje potentials preceded the QRS onset during VAs. Twenty-four of the 30 patients (13 men; age 56 ± 15 years) only had a single VA origin, which was successfully ablated at the LPF (n = 17), LAF (n = 4), or RBB (n = 3). At baseline, the VAs exhibited a single QRS morphology in 22 of these 24 patients (group UF-1) and multiple different QRS morphologies in the remaining 2 patients, with no evidence of structural heart disease (group UF-2). In 5 of the 24 patients, the VA origins were located in the proximal fascicle, or basal RBB, and the RFCA was complicated by an LAF block in 1 patient, complete RBB block in 2 patients, and interventricular conduction delay in 2 patients. In the remaining 6 patients (2 men; age 52 ± 21 years), multiple VA origins were identified within the HPS. There were no significant differences in any clinical characteristics or baseline electrophysiological measures between the patients with unifocal and multifocal HPS VAs, although the incidence of VF and a conduction disturbance at baseline tended to be higher in the patients with multifocal HPS VAs than in the patients with unifocal HPS VAs (Table 1).
The more detailed clinical characteristics, and electrocardiographic and electrophysiological characteristics in the 6 patients with multifocal HPS VAs are shown in Tables 2 and 3⇓⇓. In these 6 patients, a total of 15 focal HPS VA origins were identified at the LAF in 4 patients, at the LPF in 6 patients, at the LSF in 1 patient, at the LBB in 2 patients, and at the RBB in 2 patients. The Purkinje potentials recorded at the successful ablation sites of those VA origins preceded the QRS onset by 36 ± 16 ms during the VAs. Three patients had 2 foci, and the remaining 3 patients had 3 foci. Multiple different VAs were observed at baseline in 3 patients (cases 2, 3, and 4), and different VAs subsequently occurred after the ablation of the baseline VAs in 5 patients (all, except case 2). RFCA was avoided in 2 of the 15 VA origins in the LBB and RBB (cases 3 and 6) because of the risk of complete atrioventricular block. In those cases, the baseline QRS was normal. The remaining 13 VA origins were successfully ablated at the LAF in 4 patients, the LPF in 6 patients, the LSF in 1 patient, the LBB in 1 patient, and the RBB in 1 patient. In 2 of the 6 patients, an HPS VA origin exhibited a single QRS morphology (cases 1 and 2; group MF-1) (Figure 2). In 3 of the remaining 4 patients (cases 3, 4, and 5; group MF-2) (Figures 3, 4, 5⇓⇓), activation from 1 VA origin preferentially conducted to multiple different breakout sites in different fascicles, which resulted in presentation of different QRS morphologies. In the remaining patient (case 6; group MF-3), activation from the VA origin in the basal LBB preferentially conducted to the LSF only. In 1 patient with a history of nonischemic cardiomyopathy and placement of a biventricular implantable cardioverter-defibrillator due to baseline LBB block, RFCA was performed at the basal LBB because it was necessary to treat a drug-refractory electrical storm. RFCA was successful, but was complicated by complete atrioventricular block. In the remaining 5 patients, the His-ventricular interval and QRS duration during sinus rhythm were prolonged after the ablation. In 1 of these 5 patients, the QRS morphology during sinus rhythm remained normal, but the other 4 patients developed a fascicular block (LAF block in 1, LPF block and RBB block in 1, and LBB block in 2). No other complications occurred.
The prevalence of VA origins identified in the proximal Purkinje system was significantly greater for the multifocal HPS VAs than unifocal HPS VAs (p = 0.013) (Table 1). The incidence of conduction disturbance complicated by RFCA was significantly greater in the multifocal HPS VAs than unifocal HPS VAs (p = 0.004) (Table 1).
Abnormal Purkinje potentials and the conduction property through the HPS
In groups UF-1 and MF-1, which were without preferential conduction, neither a pre-systolic split or fractionated Purkinje potential nor conduction block between the VA origin and ventricular myocardium (P-Vpvc block) was recorded at the site of VA origins. In groups UF-2, MF-2, and MF-3, an occurrence of preferential conduction from a VA origin to a breakout site was suggested; such abnormal Purkinje potentials and/or P-VPVC block were recorded at the site of all VA origins during the VAs (Figures 3C, 3D, 4B, and 5B, Table 3). The incidence of abnormal Purkinje potentials and P-Vpvc block was significantly greater in the patients with multifocal HPS VAs than in the patients with unifocal HPS VAs (p = 0.007).
During a follow-up of 55 ± 24 months after the last procedure, all 24 patients with unifocal HPS VAs remained free from any VF. Two of these patients had a recurrence of PVCs, which were successfully controlled by sotalol. During a follow-up of 38 ± 30 months after the last procedure, in 6 patients with multifocal HPS VAs, 3 of the 13 successfully ablated HPS VA origins (1 in the LAF and 2 in the LPF) recurred. The VAs originating from these fascicles were successfully controlled by antiarrhythmic drugs (flecainide in 1, quinidine in 1, and amiodarone in 1). The post-procedural PVC burden per Holter recording was noted to be <1% and 2.9 ± 1.2% in the patients with and without freedom from any focal HPS VAs. Five of the 6 patients with multifocal HPS VAs remain free from VF. VF recurred in the remaining patient, but it was successfully controlled by amiodarone. PVCs with origins that were not ablated because of the risk of complete atrioventricular block were successfully controlled by flecainide for the RBB PVC and amiodarone for the LBB PVC.
There were no significant differences in the recurrence rate of each ablated HPS VA between the single and multiple origins (p = 0.214). The rate of freedom from any focal HPS VAs after the last procedure was significantly higher in the patients with a single origin than those with multiple origins (p = 0.001).
Comparison of the VF incidence before and after RFCA between the patients with unifocal and multifocal HPS VAs
Among the 30 patients, VF tended to occur more often in the patients with multifocal HPS VAs than in patients with unifocal HPS VAs (p = 0.059). Among the 19 patients with no structural heart disease (16 patients with unifocal HPS VAs and 3 patients with multifocal HPS VAs), there were no significant differences in the incidence of VF between the 2 groups (p = 0.30). Among the 14 patients with a history of VF, there was no significant difference in VF recurrence after the last procedure between the 2 groups (p = 0.18).
This study revealed that 20% of the patients with focal HPS VAs had multiple VA origins, which were distributed all over the HPS. VF and some kind of baseline conduction disturbance in the HPS more often occurred in these patients than in the patients with unifocal HPS VAs. In addition, abnormal Purkinje potentials and conduction properties at the site of VA origins were more often observed in the patients with multifocal HPS VAs than in patients with unifocal HPS-VAs. These findings suggested that the underlying disease within the HPS might be associated with an occurrence of multifocal HPS VAs.
It has been reported that a focal HPS VA could exhibit multiple QRS morphologies when a breakout site shifted from 1 fascicle to another at baseline or after ablation at 1 breakout site (1–3,9–11). When a breakout site shifts spontaneously during HPS VAs, preferential conduction is suggested to occur within the HPS. In this study, during the HPS VAs, preferential conduction spontaneously occurred through the left HPS from an origin within the proximal HPS to multiple breakout sites in the distal HPS, which resulted in multiple QRS morphologies. This kind of preferential conduction more often occurred in the patients with multifocal HPS VAs than in the patients with unifocal VAs. These findings also might suggest the presence of underlying disease in the HPS in patients with multifocal HPS VAs.
Mapping of HPS VAs in this study was challenging, and it was difficult to complete the activation maps because multiple QRS morphologies occurred irregularly in a setting of multiple VA origins and preferential conduction from a single origin to multiple breakout sites. Despite these challenges, RFCA of those VAs was highly successful with our systematic mapping technique, which had a combination of 3-dimensional mapping and conventional mapping that compared the activations recorded from the distal and proximal electrode pairs of the mapping catheter.
In this study, there were no significant differences in either the recurrence rate of the ablated VAs or incidence of VF after RFCA between the patients with unifocal and multifocal HPS VAs. Therefore, RFCA of multifocal HPS VAs was as effective as that of unifocal HPS VAs. In contrast, successful ablation of each VA resulted in progressive deterioration of the HPS conduction in the patients with multifocal HPS VAs. In addition, in the patients with multifocal HPS VAs, the VA origins were often located in the proximal portion of the LBB or RBB, and RFCA of these VA origins might predispose those patients to complete atrioventricular block. Therefore, RFCA of the multifocal HPS VAs was often limited by those risks.
In this study, the statistical analysis demonstrated no significant differences among some of the clinical, electrocardiographic, and electrophysiological characteristics and outcomes of the RFCA between the patients with multifocal HPS-VAs and unifocal HPS-VAs. Because the number of study patients was small, the comparisons between the 2 groups might be considered as a reference, and further study would be expected in the future.
This study demonstrated that 20% of patients with focal HPS VAs had multiple origins that were distributed all over the HPS. Electrocardiographic and electrophysiological conduction disturbances were often observed in these patients. Preferential conduction from a VA origin to multiple breakout sites within the HPS that resulted in multiple QRS morphologies more often occurred in the patients with multifocal HPS VAs than in the patients with unifocal HPS VAs. These findings suggested that the underlying disease within the HPS might be associated with an occurrence of multifocal HPS VAs. RFCA of multifocal HPS VAs was as effective as that of unifocal HPS VAs. Although the mapping of the HPS VAs was challenging due to multiple VA origins and preferential conduction, catheter ablation of those VAs was highly successful with the systematic mapping technique. However, RFCA of multifocal HPS VAs was often limited by the risk of complete atrioventricular block.
COMPETENCY IN MEDICAL KNOWLEDGE: Multifocal ventricular arrhythmias originating from the HPS can occur and often present with preferential conduction through the HPS from an origin to multiple breakout sites with abnormal Purkinje potentials and/or conduction properties through the HPS. RFCA of these VAs is effective but is limited by the risk of HPS impairment.
TRANSLATIONAL OUTLOOK: Although the mapping of the HPS VAs is challenging due to multiple VA origins and preferential conduction, catheter ablation of these VAs is highly successful with a systematic mapping technique. Nonetheless, the development of a better mapping system is expected to facilitate catheter ablation of these VAs.
Both 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
- His-Purkinje system
- left anterior fascicle
- left bundle branch
- left posterior fascicle
- left septal fascicle
- premature ventricular contractions
- right bundle branch
- radiofrequency catheter ablation
- heights of the R-wave/depth of the S-wave ratio
- ventricular arrhythmia
- ventricular fibrillation
- Received March 19, 2018.
- Revision received May 23, 2018.
- Accepted June 13, 2018.
- 2018 American College of Cardiology Foundation
- Haïssaguerre M.,
- Shoda M.,
- Jaïs P.,
- et al.
- Talib A.K.,
- Nogami A.,
- Morishima I.,
- et al.
- Chen M.,
- Gu K.,
- Yang B.,
- et al.
- Pathak R.K.,
- Betensky B.P.,
- Santangeli P.,
- Dixit S.