Clinical and electrophysiological characteristics of ventricular arrhythmias arising from pulmonary cusps

Abstract Introduction Ventricular arrhythmias (VAs) have been successfully ablated from the pulmonary sinus cusps establishing pulmonary artery (PA) as a distinct site of arrhythmic foci. The aim of the present study was to determine the clinical presentation, electrocardiographic, and ablation characteristics of PA‐VAs. Methods Thirty consecutive patients with right ventricular outflow tract (RVOT)‐type VAs were included in this retrospective study. Three‐dimensional electroanatomic mapping was performed in all patients. Mapping was performed initially in RVOT, and later within the PA. Mapping was performed in the PA if there was no early activation, unsatisfactory pace‐map, or ablation in RVOT were unsuccessful. All PA‐VAs were mapped and ablated by looping the catheter in a reverse U fashion. Results Among 30 patients, 8 (26.6%) patients VAs were successfully ablated within PA. Electrocardiography (ECG) revealed that the QRS duration was significantly wider in the PA‐VAs group compared to the RVOT‐VAs group (155 ± 14.14 vs 142.40 ± 8.12 ms, P < .01). Mapping by reversed U method of PA‐VAs revealed earlier activation (55 ± 9.66 vs 12.00 ± 8.61 ms, P < .01) in PA compared to RVOT. An isolated discrete prepotential was present at the successful site in 50% (n = 4). Conclusion Pulmonary artery‐VAs are an important subset of VA originating from the outflow tract. They have a wider baseline QRS duration compared to RVOT‐VAs. Presence of a prepotential aids in the identification of a successful ablation site. Mapping utilizing the reversed U method can help in localization and successful ablation of PA‐VAs.


| INTRODUC TI ON
Idiopathic ventricular arrhythmias (VAs) account for approximately 10% of patients referred for VA evaluation. [1][2][3][4] Majority of these VAs originate from the outflow tract. The success rate of catheter ablation of these arrhythmias is approximately 80%-95%. 5 Ablation of outflow tract VAs originating from the pulmonary artery (PA) was first described by Timmermans et al 6 . Few studies have demonstrated successful ablation of VAs above the pulmonary sinus cusp (PSC) establishing the PA as a distinct site of right ventricular outflow tract (RVOT)-VAs. [7][8][9] However, there is no uniform approach or ablation strategy for this subset of VAs. The objectives of this study were to describe the clinical presentation, electrocardiographic, and ablation characteristics of PA-VAs in comparison with RVOT-VA.

| Electrocardiographic analysis
A 12-lead electrocardiography (ECG) was recorded during the VAs. The following characteristics were analyzed: (a) QRS duration, (b) precordial R-wave transition (earliest lead with R/S ratio >1), (c) R-wave morphology in Lead I, (d) R-wave amplitude and morphology in leads II, III, and aVF, (e) Q-wave amplitude in leads aVR and aVL.

| Radiofrequency ablation
All anti-arrhythmic drugs were discontinued for at least five half-lives before the procedure. Three-dimensional (3D) electromagnetic mapping was performed using EnSite NavX™ (St. Jude Medical, Inc). A 7Fr irrigated tip ablation catheter with a 3.5-mm tip was used for mapping and ablation. PA was mapped using a reversed U method in all patients. Preface® guiding sheath curve (8 F) (Biosense Webster Inc) was used to stabilize and manipulate the mapping catheter to access all three individual PSCs. The reverse U curve was made by curving the ablation catheter in the PA to position the tip into the PSC (Figure 1).
Activation mapping was used to identify the origin of VAs ( Figure 2). If the clinical arrhythmia did not occur spontaneously or with programmed ventricular stimulation, intravenous isoproterenol (1-4 μg min) was administered to provoke the arrhythmia. Initial activation mapping was performed in the RVOT region and if no early activation sites were identified, test burns were unsuccessful, or pace mapping was not good then the PSC was mapped. Activation time was based on the earliest bipolar signals and rapid down stroke of the unipolar signal correlating with the first sharp peak of the bipolar electrogram. In every patient, pace-mapping was also performed. A pace-map was considered "good" when score is ≥20.

| Statistical analysis
Continuous variables are expressed as mean ± SD, and categorical variables as numbers and percentages. Categorical variables were compared using chi-square analysis or Fisher exact test. The parameters in different groups were compared by t test. A P value < .05 was considered significant.

| Clinical characteristics
A total of 30 outflow tract-VAs originating from the RVOT and PA were included in the study. Among these, the sites of successful ablation were RVOT-septum (n = 15), RVOT-free wall (n = 7), and PA (n = 8). The comparison of baseline characteristics of PA-VAs with RVOT group is shown in Table 1. Their mean age was 39.25 ± 10.16 years. The PVC burden was 15.25%.
The mean LVEF of patients with PA-VAs was 51.75 ± 7.28%.
Age of patients with PA-VAs was significantly lower compared to RVOT-VAs (39.25 ± 10.16 vs 51.82 ± 11.07 years, P < .01). Majority of PA-VAs group had VT and NSVT at presentation compared to RVOT group (61.5% vs 38.5% P = .01). There was no significant difference in LVEF between these two groups (51.75 ± 7.28 vs 53.32 ± 11.7 P = .72). In PA-VA group 25% (n = 2) had LVD. CAG and CMR were normal in patients with LVD. Medical therapy consisted of beta blockers, calcium channel blockers, amiodarone, or sotalol.

| ECG characteristics
Electrocardiography characteristics of PA-VAs and RVOT-VAs are shown in Table 2. PVCs exhibited LBBB-type morphology with inferior-axis. Precordial QRS transition was noted in V3 in 25% (n = 2), V4 in 50% (n = 4), V5 in 12.5% (n = 1), and V6 in 12.5% (n = 1). The QRS F I G U R E 2 Pulmonary angiography in right anterior oblique view (RAO) and Left anterior oblique (LAO) view. APC, Anterior pulmonary cusp; LPC, Left pulmonary cusp F I G U R E 3 A, 3D activation mapping of pulmonary artery with cusps with relation of cusps in Right anterior oblique (RAO) and left anterior oblique view (LAO) view with earliest activation in the Left pulmonary cusp. B, Electrocardiogram with intracardiac signals from the map at the earliest site showing discrete isolated prepotential (with intervening isoelectric segment) (yellow arrow head). Discrete isolated prepotential precedes the premature ventricular complex QRS by 63 ms. In sinus rhythm, the prepotential follows the local bipolar electrogram. APC, Anterior pulmonary cusp; LPC, Left pulmonary cusp; RPC, Right pulmonary cusp duration was significantly wider in PA VA group compared to the RVOT group (155 ± 14.14 vs 142.40 ± 8.12 ms, P < .01). In addition, lead I positivity was more common in PA-VAs (62.5% vs 18.18%, P = .02). The amplitude of R in II, III, aVF, and amplitude of Q wave in aVL and aVF did not differ significantly between the groups. In addition, lead III/II R wave ratio and aVL/aVR Q wave ratio were not different.

| Mapping and ablation characteristics of PA-VA
The RVOT approach was successful in 22 (73.33%) patients but failed to eliminate/identify early site of arrhythmia in remaining eight patients. Among these eight patients, PA was mapped and successfully ablated in the PSC using a reversed U method. In  PA. The mean procedure and fluoroscopy time were 135 ± 32 and 18 ± 8.6 minutes, respectively.

| D ISCUSS I ON
The main finding of the study was that PA mapping should be considered when ablation fails to suppress VA in the RVOT. Clinical clues for PA-VA include a younger age at presentation and absence of good sites/signals on mapping below the pulmonary cusp. In addition, the presence of a discrete prepotential can help identify the successful site of arrhythmia ablation.

| Baseline clinical characteristics
The anatomical basis of PA-VAs lies in the myocardial extensions in the PA above the cuspal and intercuspal region. 10

| Electrophysiological characteristics of PA-VA
Bipolar mapping of the VAs in the PA revealed multicomponent electrograms in 50% consisting of sharp potential followed by a dull electrogram without an intervening isoelectric segment. While the rest 50% showed a discrete potential with isoelectric line. In these patients, the site of discrete prepotential aided in the identification of the successful ablation site. The sharp potential represents near field activation of the muscular remnants in PA preceding the dull far field ventricular activation electrogram from the RVOT. This relationship reversed in sinus rhythm.
Unique feature of the bipolar signal in this study was identification of a discrete prepotential at the arrhythmic focus. This discrete prepotential was first described by Miyazaki et al but subsequent studies did not show this finding in PA-VAs. 14 This is similar to the prepotential found in PVC originating from the aortic cusps.
Hachiya et al hypothesized the origin of LVOT prepotential were from the dead-end tract at the top of the ventricular septum. 15 Since both out flow tracts overlap each other, prepotential in PA-VAs may be due to a similar mechanism.
We also employed pace-mapping in all patients at site of earliest activation. In PA-VAs group, good pace-map (score ≥20) was obtained in 87.5%. This high percentage was possible due to improved catheter contact by reverse U method. Due to the smaller muscle mass in the PA, higher output was required to capture the myocardium (10 mA) in this study. On the contrary, Sekiguchi et al was able to perform satisfactory pace mapping only in 37% of patients with the conventional method. 9 A change in QRS morphology after initial ablation in RVOT did not occur in our study because when the early suppression of arrhythmia was not observed within 10 -15 seconds of energy delivery, we immediately switched over to PA for further mapping. In summary, activation map with early signals consisting of sharp potential preceding the ventricular electrogram along with good pace-map identify the foci in the PA.
In this study, both acute and long term procedural success are better compared to studies using RVOT ablation without PA mapping. In a multicenter outcome study of RFCA of PVCs, recurrence occurred in 18% of RVOT patients during a mean follow-up of 20.2 ± 21.7 months. 16 In that study, all PVCs were ablated only in the RVOT. This high failure rate can be explained by the nature of muscular connections between the RVOT and PA, which is characterized by a triangular muscular sleeve, with broader base in RVOT requiring more RFCA lesions over a wide area compared to tapered distal end in PA which has a smaller area.

| Limitations of the study
This is a single-center retrospective study with a small sample size. Smaller sample size prevented a comparative study of ECG characteristics among the three PA cuspal positions. Our approach of switching over to PA instead of upfront mapping of PA may not provide a true PA-VAs prevalence. ICE which is more precise in conforming and localizing the catheter tip was not utilized in this study.

| CON CLUS ION
In this study, approximately 25% of RVOT-VAs originated from the PA. Ablation by inverted U method is safe and efficacious with good long-term results. Discrete prepotential, good pace-map, and early activation time help in identification of the PA-VA foci.