Earliest activation time is a good predictor of successful ablation of idiopathic outflow tract ventricular arrhythmias

Abstract Background In idiopathic outflow tract ventricular arrhythmias (OT‐VAs), identifying the site with the earliest activation time (EAT) using activation mapping is critical to eliminating the arrhythmogenic focus. However, the optimal EAT for predicting successful radiofrequency catheter ablation (RFCA) has not been established. Hypothesis To evaluate the association between EAT and successful RFCA in idiopathic OT‐VAs and to determine the optimal cut‐off value of EAT for successful ablation. Methods We retrospectively analyzed patients undergoing RFCA for idiopathic OT‐VAs at a single center from January 2015 to December 2019. Results Acute procedural success was achieved in 168 patients (87.0%). Among these patients, 158 patients (81.9%) were classified in the clinical success group according to the recurrence of clinical VAs during median (Q1, Q3) follow‐up (330 days [182, 808]). EAT was significantly earlier in the clinical success group compared with the recurrence (p = .006) and initial failure (p < .0001) groups. The optimal EAT cut‐off value predicting clinical success was −30 ms in the right ventricular outflow tract (RVOT) with 77.4% sensitivity and 96.4% specificity. In all cases of successful ablation in the left ventricular outflow tract (LVOT), EAT in the RVOT was not earlier than −29 ms. Conclusions EAT in patients with successful catheter ablation was significantly earlier than that in patients with recurrence and initial failure. EAT earlier than −30 ms could be used as a key predictor of successful catheter ablation as well as an indicator of the need to shift focus from the RVOT to the LVOT.


| INTRODUCTION
Idiopathic outflow tract ventricular arrhythmias (OT-VAs) are the most common anatomical subtype of idiopathic VA and typically occur in healthy patients without clinically apparent structural heart disease.
The main mechanism of idiopathic OT-VAs is cyclic AMP-mediated triggered activity and arrhythmogenic focus is usually presented as a focal source. 1 Thus, finding the site of origin largely depends on activation mapping rather than pace mapping. In addition, identifying the site with the earliest activation time (EAT) is pivotal for eliminating the arrhythmogenic substrate. Generally, the site with presystolic activation over 20 ms and QS unipolar morphology is considered the possible site of origin of VAs. 2,3 However, the optimal EAT to predict successful radiofrequency catheter ablation (RFCA) has not been determined. Therefore, the purpose of this study was to investigate the association between successful RFCA and EAT and to identify the precise cut-off value predicting successful ablation in patients undergoing catheter ablation for OT-VAs.

| Electrophysiologic study and RFCA
All electrophysiologic study (EPS) were conducted in a fasting state without general anesthesia or sedation in patients who discontinued antiarrhythmic drugs for at least five half-lives before the study. A surface 12-lead electrocardiogram (ECG) and intracardiac electrogram (EGM) were simultaneously displayed and recorded using the Cardiolab™ electrophysiology recording system (GE Healthcare, Chicago, IL). After local anesthesia, a 6-Fr His-RV electrode catheter was advanced through a right femoral vein to the RV apex and His bundle region. Mapping and RFCA were performed using a 8-Fr quadripolar deflectable catheter with a 3.5-mm open-irrigated-tip (Thermocool SF NAV mainly in the early period, and Thermocool SmartTouch-SF catheter after 2018, Biosense Webster, Diamond Bar, CA) introduced through a long sheath via a right femoral vein. In the case of left ventricular outflow tract (LVOT), we used a retrograde aortic approach via the right femoral artery.
To identify the earliest site of endocardial activation, a 3-dimensional electroanatomic mapping system (Carto3, Biosense Webster) was used for activation mapping. Pace mapping was also performed at the site of earliest activation and the matching score was presented as a percentage using an automated ECG matching program (PASO, Biosense Webster). EAT was measured from the initial deflection on distal bipolar EGM to the earliest onset of PVC-QRS on the 12-lead surface ECG ( Figure S1). Except for cases with right bundle branch block (RBBB) morphology of clinical VAs, we usually began by mapping the right ventricular outflow tract (RVOT). If a site with EAT earlier than −20-−25 ms could not be identified in the RVOT, the LVOT was mapped.
Radiofrequency (RF) applications were delivered initially at the earliest site of endocardial activation with a maximal power of 25-40 W using an RF generator (Stockert, Biosense Webster). RF output was upward titrated independently from 25 to 40 W according to impedance and a temperature limit under 50 C. If clinical VAs were suppressed within 20-30 seconds, RF application was maintained for 60-90 seconds. Booster burn was always applied at the successful ablation site for 60 seconds to completely eliminate the arrhythmogenic substrate.

| 12-lead ECG analysis of the origin of OT-VA
Five 12-lead ECG algorithms including R-wave duration index, R/S-wave amplitude index, R-wave deflection interval combined with R-wave amplitude index, V 2 S/V 3 R index, and V 2 transition ratio were used to predict the origin of OT-VA. The following ECG parameters were measured during clinical VA (PVC/VT) using a digital caliper on the Cardiolab™ recording system: (1) R-wave duration in leads V 1 -V 2 ; (2) R-and S-wave amplitude in leads V 1 -V 3 ; (3) R-and S-wave amplitude in lead V 2 during sinus rhythm; (4) R-wave deflection interval in lead V 3 , and (4) Total QRS duration. We evaluated the predictive accuracy of these ECG algorithms for differentiating right from left OT-VA origin in patients with V 3 precordial transition of clinical OT-VA.

| Definitions
Clinical VA was defined as a PVC and/or VT with the morphology most frequently observed in preprocedural Holter monitoring and 12-lead ECG. Acute procedural success was defined as no detection of any clinical VA during a waiting period (≥30 minutes) after catheter ablation under a standardized stimulation protocol. Clinical success was defined as the complete elimination of clinical VT during the follow-up period and/or ≥80% reduction of clinical PVC on postprocedural Holter monitoring and regular 12-lead ECG. Reappearance of clinical VT and/or a <80% reduction in clinical PVC after acute procedural success was classified as recurrence. "Initial failure" indicated a case that did not achieve acute procedural success, while "no clinical success" was used to describe both initial failure and recurrence.

| Follow-up
All patients were scheduled to visit the outpatient clinic 1 month after the procedure and at 3-4-month intervals thereafter for 12-lead ECG.
Holter monitoring was performed 3-6 months after the procedure.
Patients with recurrent symptoms were evaluated immediately using Holter or event monitoring regardless of follow-up interval. In the case of follow-up loss, we contacted patients over the telephone to determine whether symptoms or VAs had recurred and encouraged patients to resume follow-up.

| Statistical analysis
Continuous variables are expressed as mean with standard deviation (SD) or median with 25th (Q1) and 75th (Q3) percentiles according to normality test, and categorical variables as the number with a percentage. To compare two groups, unpaired Student's t-test and Mann-Whitney test were performed for continuous variables and Pearson's χ 2 and Fisher's exact test for categorical variables. A comparison of continuous variables among the three groups was analyzed with Kruskal-Wallis test, and post hoc analysis was conducted with Dunn's multiple comparisons test. To evaluate sensitivity, specificity, the area under the curve (AUC), and the optimal cut-off value for predicting clinical success, receiver operating characteristic (ROC) analysis was used with DeLong's test to compare two ROC curves. Statistical significance was defined as a two-tailed p < .05. All statistical analyses were performed using SPSS statistical software version 25.0 (IBM, Armonk, NY).

| Baseline characteristics
Over the 5-year study period, a total of 219 patients underwent catheter ablation for OT-VAs. Twenty-six patients were excluded according to the exclusion criteria, and a total of 193 patients (mean age 48 ± 13 years; female 63%) were included in the study group ( Figure S2). Patients were divided into two groups based on clinical success. The clinical characteristics of the study cohort are summarized in   (Table S1). Nine patients with RBBB morphology PVC/VT had a successful ablation in the LVOT. One in five patients (20%) with PVC/VT transition in lead V 2 had a successful ablation at the RVOT, and five in 104 patients (4.8%) with PVC/VT transition ≥lead V 4 had a successful ablation at the LVOT ( Figure S3). There was no major complication after RFCA, except for one case with arteriovenous fistula at vascular access.

| Main findings
The aim of this study was to evaluate the association between EAT and successful RFCA in idiopathic OT-VAs and to find the optimal cut-off value of EAT for predicting successful RFCA. First, we found that EAT was significantly earlier in the clinical success group compared with the recurrence and initial failure groups. Second, the optimal cut-off value for EAT could predict successful RFCA in the RVOT and LVOT with high accuracy. The results of this study have important clinical implications, as these findings will provide useful information to establish the optimal RFCA strategy during the procedure.

| EAT as a predictor for successful ablation
In the past two decades, an improved understanding of the mechanism and anatomy of OT-VAs and advances in electroanatomic mapping systems have made RFCA a standard therapy in drug-refractory patients and those with a high PVC burden associated with reduced LV systolic function. 4,5 Given the complexity of outflow tract anatomy and its variations, precise electroanatomical mapping is crucial for successful RFCA.
Pace mapping could be useful when spontaneous ventricular ectopies are not frequently induced enough to create detailed electroanatomic mapping. 6 However, Yamada et al. reported that for VAs originating from the aortic root, pace mapping may not be accurate due to preferential conduction from the aortic root to the RVOT. 7 In addition, Bogun et al. showed that the spatial resolution of pace mapping is inferior to that of activation mapping, and, more interestingly, pace mapping was inaccurate in nearly 20% of patients (three of 16 cases). 8  vs. −28 ± 11 ms; p < .01) and suggested that a local activation time with −30 ms was the best cut-off value for predicting successful ablation. 9 In 2012, Yamada et al. reported that there was no difference in EAT between an initial success group and a failure group, but EAT predicted recurrence after successful ablation in a retrospective study with idiopathic RVOT-VAs. 10 These conflicting findings may be due to

| EAT as a key element of ablation strategy
In patients without structural heart disease, the electrical activity caused by PVC is conducted through the normal myocardium; thus, it exhibits a characteristic ECG shape depending on the location of its occurrence. 11 Several studies revealed relatively high accuracy for differentiating right from left OT-VA origin on 12-lead ECG algorithm. [12][13][14][15] However, the accuracy of prediction varies according to the patient population because there is considerable variability in lead positions, shape of the chest wall, heart rotation, and conduction properties from person to person. 16,17 In general, since the LVOT is posteriorly positioned compared with the RVOT, PVC originating from the LVOT tends to show a larger and wider R-wave at lead V 1 and V 2 .
Therefore, when precordial transition is ≤lead V 2 , PVC is likely to originate from the LVOT; in contrast, when precordial transition is ≥lead V 4 , PVC is likely to originate from the RVOT. 15 However, in cases with precordial transition in lead V 3 , it is difficult to discriminate between the RVOT and LVOT, and the accuracy of various ECG algorithms is relatively low. 11 In our study, among patients who underwent successful ablation, 20% of those who had a successful ablation in the RVOT showed an early transition of PVC/VT in lead V 2 , while 4.8% of patients who had a successful ablation in the LVOT showed a late transition of PVC/VT (≥lead V 4 ) ( Figure S3). Moreover, in cases with PVC/VT transition in V 3 , various ECG algorithms showed low predictive performance for differentiating between the RVOT and LVOT (Table 3). Therefore, it is essential to establish an optimal ablation strategy because electroanatomic mapping is the only way to identify the precise site of origin. We always mapped the RVOT first except for cases with an RBBB PVC/VT morphology. In all cases of successful ablation in the LVOT, EAT in the RVOT was not earlier than −29 ms.

| CONCLUSIONS
EAT in patients with successful catheter ablation was significantly earlier than that in patients with recurrence and initial failure.