Heterogeneous distribution of substrates between the endocardium and epicardium promotes ventricular fibrillation in arrhythmogenic right ventricular dysplasia/cardiomyopathy

Heterogeneous distribution of substrates between the endocardium and epicardium promotes... Abstract Aims Whether the distribution of scar in arrhythmogenic right ventricular cardiomyopathy (ARVC) plays a role in predicting different types of ventricular arrhythmias is unknown. This study aimed to investigate the prognostic value of scar distribution in patients with ARVC. Methods and results We studied 80 consecutive ARVC patients (46 men, mean age 47 ± 15 years) who underwent an electrophysiological study with ablation. Thirty-four patients receive both endocardial and epicardial mapping. Abnormal endocardial substrates and epicardial substrates were characterized. Three groups were defined according to the epicardial and endocardial scar gradient (<10%: transmural, 10–20%: intermediate, >20%: horizontal, as groups 1, 2, and 3, respectively). Sinus rhythm electrograms underwent a Hilbert–Huang spectral analysis and were displayed as 3D Simultaneous Amplitude Frequency Electrogram Transformation (SAFE-T) maps, which represented the arrhythmogenic potentials. The baseline characteristics were similar between the three groups. Group 3 patients had a higher incidence of fatal ventricular arrhythmias requiring defibrillation and cardiac arrest during the initial presentation despite having fewer premature ventricular complexes. A larger area of arrhythmogenic potentials in the epicardium was observed in patients with horizontal scar. The epicardial-endocardial scar gradient was independently associated with the occurrence of fatal ventricular arrhythmias after a multivariate adjustment. The total, ventricular tachycardia, and VF recurrent rates were higher in Group 3 during 38 ± 21 months of follow-up. Conclusion For ARVC, the epicardial substrate that extended in the horizontal plane rather than transmurally provided the arrhythmogenic substrate for a fatal ventricular arrhythmia circuit. Electroanatomic mapping , Hilbert–Huang transform , Sudden cardiac death , Ventricular fibrillation , Ventricular tachycardia What’s new? The epicardial-endocardial scar gradient was independently associated with the occurrence of fatal ventricular arrhythmias in arrhythmogenic right ventricular cardiomyopathy (ARVC) patients A high epicardial-endocardial scar gradient was associated with a higher ventricular tachycardia/ventricular fibrillation recurrence rate after catheter ablation in ARVC patients. Introduction Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a genetically and clinically heterogeneous disorder of cardiac muscles that is associated with ventricular arrhythmias, right ventricular dysfunction, and the risk of sudden cardiac death (SCD).1,2 The hallmark pathological lesion of ARVC is the transmural loss of the myocardium of the right ventricular (RV) free wall with replacement by fibrofatty tissue.3–5 The RV is predominantly affected, and both pathological and electroanatomic studies show a disproportionate progression of disease on the epicardium compared with the endocardium6–8. The scarring process modifies the transmural activation with a delayed epicardial activation and may predispose to the development of ventricular tachycardia (VT) circuits contained entirely within the epicardium in ARVC.1 Furthermore, previous studies have shown that the rapid progression of macroscopic endocardial scars occurs in only a subset of ARVC patients.9 It is unknown if the distribution of scar either extending horizontally or transmurally plays a role in fatal VAs. We hypothesized that the three-dimensional scar extension/distribution of the epicardium/endocardium in ARVC may play a role in fatal VAs. In this study, we studied the prognostic value of electroanatomic mapping of ARVC patients in terms of voltage mapping, delayed activation areas, and arrhythmogenic potentials through signal analyses. Methods Study population Between 2008 and 2015, 80 patients (46 men [58%] with a mean age of 47 ± 15 years) confirmed to have ARVC based on the 2010 revised Task Force Criteria10 were enrolled consecutively. All patients were admitted for ablation of drug-refractory VAs. We obtained the Institutional Review Board approval to conduct the retrospective study (VGH-IRB No.2014-30 10-004B). We performed 12-lead resting electrocardiography (ECG), signal-averaged ECGs (SAECGs), 24-h Holter monitoring, 2D echocardiography, coronary arteriography, and electrophysiological (EP) studies on all patients before the procedure. Fatal VAs were defined as ventricular fibrillation (VF), irregular tachycardias with regard to their polarity, amplitude, morphology, and sequence of the intracardiac electrograms, with a mean cycle length of <240 ms requiring defibrillation, or SCD receiving cardiopulmonary resuscitation. VT was defined as a regular tachycardia with a mean cycle length of >240 ms.11 EP study A standardized EP study was performed in the fasting state using either conscious sedation or general anesthesia. Anti-arrhythmic drugs were discontinued for a minimum of five half-lives before the radiofrequency catheter ablation. In the absence of spontaneous VT/VF, rapid ventricular pacing and programmed stimulation up to three extrastimuli were performed from the RV apex and/or RV outflow tract. Three basal cycle lengths (500-400-300 ms) were used for programmed stimulation. If VT/VF was still not inducible, intravenous isoprenaline (1–5 μg/min) was infused to achieve at least a 20% heart rate increment. If VT/VF was not inducible during pharmacological provocation, the induction protocol was repeated. The VT QRS morphologies were compared with those of the documented VTs. Pericardial access was obtained by a subxiphoid puncture if a previous endocardial ablation had failed, an epicardial substrate was suspected (based on surface ECG VT morphology), or minimal or no endocardial scar was present. Electroanatomic mapping was performed during sinus rhythm using the CARTO 3 system MEM version (Biosense Webster, Ltd., Haifa, Israel, user defined map [UDM] module). Mapping and catheter ablation was performed with an open-irrigated ablation catheter (Thermocool, Biosense Webster Inc., Diamond Bar, CA, USA). The unipolar filtering was set at 2–240 Hz, and bipolar filtering at 16–500 Hz. Wilson’s central terminal was assigned as the unipolar reference electrode. VT mapping and catheter ablation Induced VTs were analyzed for the cycle length and morphology using an EP recording system (BARD, Inc. LabSystemTM PRO EP Recording System). If an induced VT was hemodynamically tolerated, pacing from the mapping catheter at a cycle-length 20–30 ms faster than the tachycardia cycle-length was performed, while observing the entrainment response. A site was considered a VT mid-isthmus only if it demonstrated (1) concealed fusion on all 12 ECG leads during entrainment, (2) the post-pacing interval was within 30 ms of the VT cycle-length, (3) the stimulus-electrogram interval was within 30 ms of the electrogram-QRS interval following entrainment, and (4) the local electrogram to QRS interval was between 30 and 70% of the VT cycle-length. Confirmed VT isthmus sites were annotated on the map after VT termination into sinus rhythm. If the induced VT was unstable, pace mapping and an arrhythmogenic potential elimination was performed. If the VAs were isolated premature ventricular complexes (PVCs) or non-sustained VT, activation mapping, pacemapping, and an arrhythmogenic potential elimination was performed. RFCA was performed with power settings of 30–50 W endocardially and 25–30 W epicardially. Three-dimensional electroanatomic mapping Endocardial electroanatomic mapping The color display of the electroanatomic mapping included the bipolar electrogram voltage (scar amplitude, <0.5 mV; normal tissue amplitude, >1.5 mV),12 a late potential latency (the interval from the end of the QRS complex to the end of the local electrogram), and an arrhythmogenic potential map to depict normal and abnormal substrates. Normal RV free wall unipolar electrograms were defined as having peak-to-peak amplitudes of ≥ 5.5 mV.13 Epicardial electroanatomic mapping Percutaneous pericardial access was achieved using a Tuohy needle via the subxiphoid approach, as described by Sosa et al.14 An open irrigated catheter was used to perform mapping of the epicardium, with a focus on the RV epicardium. We defined epicardial scar zones as those areas with a bipolar signal amplitude of <1.0 mV. The correlation of the RV endocardial unipolar/epicardial bipolar scar Among patients with epicardial procedures, the correlation of an RV endocardial unipolar low-voltage zone (LVZ, %) and RV epicardial bipolar scar area (%) was studied. Pearson’s correlation index was applied for the correlation between RV epicardial bipolar scar region and RV endocardial unipolar LVZ. Intraclass correlation coefficient and Scatter plots of the Bland–Altman plot were used to evaluate the consistence. The intraclass correlation coefficient and Bland–Altman plot showed a high level of agreement (please see the result section) and the expected epicardial scar area by RV endocardial unipolar LVZ were applied to the rest of patients without epicardial mapping in this study. Definition of transmural scar and horizontal scar We defined the cut-off points of three groups on the basis of the percentiles of the difference (<33rd, 33rd to 66th, >66th percentile) between epicardial scar (bipolar signal amplitude of <1.0 mV from the epicardium or unipolar signal amplitude of < 5.5 mV from the endocardium)15 and the endocardial scar. (<10%: transmural scar, 10–20%: intermediate heterogeneous scar, >20%: horizontal scar, as groups 1, 2, and 3, respectively; Table 1 and Figure 1). Table 1 Baseline characteristics and outcome of the study patients in the three groups Groups  Overall (N = 80)  Group 1(N = 22)  Group 2 (N = 33)  Group 3(N = 25)  P value  Basic data             Age  46±14  50±13  43.3±13  50±14  0.077   Sex (M, %)  46(57.5%)  10(45.5%)  20(60.6%)  16(64.0%)  0.393   Smoking  22(27.5%)  5(25.0%)  10(31.2%)  7(29.2%)  0.889   Syncope  44(55.0%)  11(50.0%)  14(42.4%)  19(76.0%)  0.021   Palpitation  69(86.3%)  20(90.9%)  28(84.8%)  21(84.0%)  0.754   Short of breathe  32(40.0%)  7(31.8%)  15(45.5%)  10(40.0%)  0.600   Hypertension  25(31.3%)  8(36.4%)  7(21.2%)  10(40.0%)  0.258   Diabetes mellitus  6(7.5%)  3(13.6%)  1(3.0%)  2(8.0%)  0.341   HF NYHA I/II  7(8.8%)  1(4.5%)  3(9.1%)  3(12.0%)  0.663   HF NYHA III/IV  2(2.5%)  0(0.0%)  0(0.0%)  2(8.0%)  0.105  Implantable cardioverter defibrillator  56(70.0%)  14(63.6%)  21(63.6%)  21(84.0%)  0.183  Medication before Admission             Amiodarone  33(41.3%)  6(27.3%)  15(45.5%)  12(48.0%)  0.289   Beta-blocker  45(56.3%)  12(54.5%)  21(63.6%)  12(48.0%)  0.485   Mexitil  20(25.0%)  5(22.7%)  7(21.2%)  8(32.0%)  0.617   Propafenone  2(2.5%)  1(4.5%)  1(3.0%)  0(0.0%)  0.590  Medication at discharge             Amiodarone  18(22.5%)  4(18.2%)  9(27.3%)  5(20.0%)  0.685   Beta-blocker  36(45.0%)  9(40.9%)  15(45.5%)  12(48.0%)  0.886   Mexitil  11(13.8%)  2 (9.1%)  3(9.1%)  6(24.0%)  0.200  Structural assessment             LVEDd (mm)  47±5  47±5  46±5  49±6  0.169   LVEF (%)  56±9  54±8  58±9  56±9  0.469   RVEF (%)  42±13  43±12  43±13  40±13  0.687  Characteristic of VAs             Clinical PVC > 1000/day  46(57.5%)  20(90.9%)  15(45.5%)  11(44.0%)  0.001   Clinical PVC > 5000/day  32(40.0%)  16(72.7%)  14(42.4%)  2(8.0%)  <0.001   PVC/Day  9480±6962  17428±14943  9677±14001  2235±5705  <0.001   Clinical non-sustained VT  33(41.3%)  11(50%)  12(36.4%)  8(32.0%)  0.421   Clinical sustained VT  33(41.3%)  4 (18.2%)  16 (48.5%)  13 (52.0%)  0.034   Fatal ventricular arrhythmia  24(30.0%)  2(9.1%)  5 (15.2%)  17(68.0%)  <0.001  Task force criteria            Structural abnormalitiesa  24(30.0%)/41(51.3%)  8(36.4%)/11(50%)  8(24.2%)/18(54.6%)  8(32.0%)/12(48.0%)  0.869  Fibro-fatty replacementa  9(11.3%)/16(20.0%)  3(13.6%)/4(18.2%)  3(9.1%)/3(9.1%)  3(12.0%)/9(36.0%)  0.163  Depolarization changesa  7(8.8%)/65(81.3%)  2(9.1%)/17(77.3%)  2(6.1%)/26(78.7%)  3(12.0%)/22(88.0%)  0.262  Repolarization changea  11(13.8%)/32(40.0%)  4(18.2%)/7(31.8%)  2(6.1%)/14(42.4%)  5(20.0%)/11(44.0%)  0.435  Ventricular arrhythmiasa  11(13.8%)/48(60.0%)  5(22.7%)/17(77.3%)  1(3.0%)/11(33.3%)  5(20.0%)/20(80.0%)  0.533  Family historya  17(21.3%)/7(8.8%)  9(40.9%)/1(4.5%)  2(24.2%)/3(9.1%)  6(24.0%)/3(12.0%)  0.621  Follow-up             Recurrence of VT  18(22.5%)  1(4.5%)  6(18.2%)  11(44.0%)  0.004   Recurrence of VF  9(11.3%)  0(0.0%)  1(3.0%)  8(32.0%)  <0.001   Mortality  2(2.5%)  0(0.0%)  0(0.0%)  2(2.5%)  0.105  Groups  Overall (N = 80)  Group 1(N = 22)  Group 2 (N = 33)  Group 3(N = 25)  P value  Basic data             Age  46±14  50±13  43.3±13  50±14  0.077   Sex (M, %)  46(57.5%)  10(45.5%)  20(60.6%)  16(64.0%)  0.393   Smoking  22(27.5%)  5(25.0%)  10(31.2%)  7(29.2%)  0.889   Syncope  44(55.0%)  11(50.0%)  14(42.4%)  19(76.0%)  0.021   Palpitation  69(86.3%)  20(90.9%)  28(84.8%)  21(84.0%)  0.754   Short of breathe  32(40.0%)  7(31.8%)  15(45.5%)  10(40.0%)  0.600   Hypertension  25(31.3%)  8(36.4%)  7(21.2%)  10(40.0%)  0.258   Diabetes mellitus  6(7.5%)  3(13.6%)  1(3.0%)  2(8.0%)  0.341   HF NYHA I/II  7(8.8%)  1(4.5%)  3(9.1%)  3(12.0%)  0.663   HF NYHA III/IV  2(2.5%)  0(0.0%)  0(0.0%)  2(8.0%)  0.105  Implantable cardioverter defibrillator  56(70.0%)  14(63.6%)  21(63.6%)  21(84.0%)  0.183  Medication before Admission             Amiodarone  33(41.3%)  6(27.3%)  15(45.5%)  12(48.0%)  0.289   Beta-blocker  45(56.3%)  12(54.5%)  21(63.6%)  12(48.0%)  0.485   Mexitil  20(25.0%)  5(22.7%)  7(21.2%)  8(32.0%)  0.617   Propafenone  2(2.5%)  1(4.5%)  1(3.0%)  0(0.0%)  0.590  Medication at discharge             Amiodarone  18(22.5%)  4(18.2%)  9(27.3%)  5(20.0%)  0.685   Beta-blocker  36(45.0%)  9(40.9%)  15(45.5%)  12(48.0%)  0.886   Mexitil  11(13.8%)  2 (9.1%)  3(9.1%)  6(24.0%)  0.200  Structural assessment             LVEDd (mm)  47±5  47±5  46±5  49±6  0.169   LVEF (%)  56±9  54±8  58±9  56±9  0.469   RVEF (%)  42±13  43±12  43±13  40±13  0.687  Characteristic of VAs             Clinical PVC > 1000/day  46(57.5%)  20(90.9%)  15(45.5%)  11(44.0%)  0.001   Clinical PVC > 5000/day  32(40.0%)  16(72.7%)  14(42.4%)  2(8.0%)  <0.001   PVC/Day  9480±6962  17428±14943  9677±14001  2235±5705  <0.001   Clinical non-sustained VT  33(41.3%)  11(50%)  12(36.4%)  8(32.0%)  0.421   Clinical sustained VT  33(41.3%)  4 (18.2%)  16 (48.5%)  13 (52.0%)  0.034   Fatal ventricular arrhythmia  24(30.0%)  2(9.1%)  5 (15.2%)  17(68.0%)  <0.001  Task force criteria            Structural abnormalitiesa  24(30.0%)/41(51.3%)  8(36.4%)/11(50%)  8(24.2%)/18(54.6%)  8(32.0%)/12(48.0%)  0.869  Fibro-fatty replacementa  9(11.3%)/16(20.0%)  3(13.6%)/4(18.2%)  3(9.1%)/3(9.1%)  3(12.0%)/9(36.0%)  0.163  Depolarization changesa  7(8.8%)/65(81.3%)  2(9.1%)/17(77.3%)  2(6.1%)/26(78.7%)  3(12.0%)/22(88.0%)  0.262  Repolarization changea  11(13.8%)/32(40.0%)  4(18.2%)/7(31.8%)  2(6.1%)/14(42.4%)  5(20.0%)/11(44.0%)  0.435  Ventricular arrhythmiasa  11(13.8%)/48(60.0%)  5(22.7%)/17(77.3%)  1(3.0%)/11(33.3%)  5(20.0%)/20(80.0%)  0.533  Family historya  17(21.3%)/7(8.8%)  9(40.9%)/1(4.5%)  2(24.2%)/3(9.1%)  6(24.0%)/3(12.0%)  0.621  Follow-up             Recurrence of VT  18(22.5%)  1(4.5%)  6(18.2%)  11(44.0%)  0.004   Recurrence of VF  9(11.3%)  0(0.0%)  1(3.0%)  8(32.0%)  <0.001   Mortality  2(2.5%)  0(0.0%)  0(0.0%)  2(2.5%)  0.105  a According to the 2010 Revised Task Force Criteria (Major/minor).10 HF indicates heart failure; LVEDd, left ventricular end diastolic diameter; LVEF, Left ventricular ejection fraction; NYHA, New York Heart Association; RVEF, Right ventricular ejection fraction; VF, ventricular fibrillation; VT, ventricular tachycardia; PVC, premature ventricular contraction. Figure 1 View largeDownload slide Substrate analyses according to the epicardium/endocardiun gradient. The ARVC patients were separated into three groups based on the transmural gradient of the scar distribution. ARVC indicates arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle. Figure 1 View largeDownload slide Substrate analyses according to the epicardium/endocardiun gradient. The ARVC patients were separated into three groups based on the transmural gradient of the scar distribution. ARVC indicates arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle. Late potentials were defined as local ventricular potentials occurring after the terminal portion of the surface QRS. Arrhythmogenic potentials were defined as abnormal potentials, including late potentials and abnormal early potentials inscribed within the QRS. The arrhythmogenic potential was visualized by Simultaneous Amplitude Frequency Electrogram Transformation (SAFE-T) mapping.16 SAFE-T mapping was used to identify objective high frequency components in substrate mapping. The arrhythmogenic potentials identified by SAFE-T mapping was associated with the VT circuit in the previous study. Regions of scar, LVZs, and late potentials were measured using the standard surface area measurement tool on the CARTO 3 system MEM Version (UDM Module). When multiple areas of confluent low voltages were present, the aggregate area from individual regions of interest was calculated. Follow-up Patients were followed-up for 1, 2, 3 months, and every 3 months after catheter ablation. Patients with an implantable cardioverter defibrillator (ICD) were interrogated at each visit. For symptomatic paients without an ICD, a detailed history was taken, and ECG, and 24-h Holter monitoring or Event Recorder were performed at each visit. The recurrence of VF was defined as a VA with a cycle length of 240 ms or less. Recurrence of VT was defined as a regular (monomorphic) or irregular (polymorphic) VA with a cycle length of more than 240 ms. The appropriateness of ICD therapies (shock or antitachycardia pacing) was deemed appropriate or inappropriate on the basis of standard criteria.17 The composite of recurrent VF/VT included the recurrence of VF, VT, and appropriate ICD therapies.3,11 The recurrence of PVCs of more than 5000 beats per day was defined as more than 5000 PVCs with single or multiple morphologies in the 24-h ECG recordings. The recurrence of PVCs of more than 1000 beats per day was defined as more than 1000 PVCs with single or multiple morphologies in the 24-h ECG recordings. The total recurrence was a composite of the PVC recurrence, VF/ICD shocks, and VT/ICD antitachycardia pacing. Statistical analysis Continuous variables are expressed as the mean ± standard deviation and were compared using a one-way analysis of variance. Nominal variables were compared using a chi-square test for linear trends. A P value of <0.05 was considered as statistically significant. All statistical analyses were performed using commercially available statistical SPSS version 17.0 software (SPSS, Chicago, IL, USA). Results Baseline characteristics The study was comprised of 80 patients (58% men, 47 ± 13 years) who fulfilled the diagnosis of definite ARVC based on the 2010 Revised Task Force Criteria. Of them, 69 (86.3%) patients experienced palpitations, 44 (55.0%) experienced syncopal episodes, 32 (40.0%) experienced shortness of breath episodes, 25 (31.1%) had hypertension, and 6 (0.1%) had diabetes mellitus. Electrocardiographic studies demonstrated depolarization abnormalities in 72 patients (90.0%), including the presence of epsilon waves in 7 (8.8%) and positive SAECGs in 60 (81.2%), while 43 (53.8%) had inverted T waves in the right precordial leads (V1,V2, and V3) or beyond in the absence of complete right bundle-branch block. Holter examinations documented episodes of sustained VT in 32 patients (40.0%), and PVCs of more than 5000 beats per day in 32 (40.0%), and non-sustained VT in 57 (83.8%). Clinical documented VF, episodes of SCD status post cardiopulmonary resuscitation, or appropriate ICD therapies were recorded in 24 patients (30.0%). Genetic mutation screening was performed in 58 (72.5%) patients and pathogenic mutations10 were identified in 22 patients (38.9%). There were 2 patients (3.4%) with multiple mutations. All patients underwent RV angiography and 62 (77.5%) received magnetic resonance imaging to assess the RV function. From the RV angiographic findings, 65 patients (81.3%) demonstrated regional RV akinesia, dyskinesia, or an aneurysm formation. Histologically, fibro-fatty replacement of the RV was identified in 25 patients (31.3%). EP study In the EP study, 19 patients (23.8%) had inducible VF, 29 (36.3%) had inducible sustained VTs, and 68 (85.0%) had inducible non-sustained VT. The non-inducibility was achieved after ablation with programed extra-stimulation and isoprenaline in all the patients. All 80 patients underwent detailed RV endocardial mapping. Twenty-two (27.5%), 33 (41.3%), and 25 (31.3%) patients exhibited a transmural scar (group 1), intermediate heterogeneous scar (group 2), and horizontal scar (group 3), respectively. In between, epicardial access was attempted and successfully achieved in 34 (42.5%) patients. Among the patients with available epicardial maps, 10 (29.4%), 10 (29.4%), and 14 (41.1%) were classified as group 1, group 2, and group 3, respectively. The consistency of the RV endocardial unipolar/epicardial bipolar scar The correlation of an RV endocardial unipolar low-voltage zone (LVZ, %) and RV epicardial bipolar scar area (%) was studied in 34 patients with epicardial procedures. The RV epicardial bipolar scar region was equivalent to the unipolar LVZ with an intraclass correlation coefficient of 0.956 (P <0.001) and Cronbach’s Alpha value of 0.977. There was no significant difference in the mean value (P = 0.863) with a significant correlation (Pearson’s 0.969, P = 0.001). Scatter plots of the Bland–Altman plot in see Supplementary material online, figure S1B shows that the mean of the difference between the RV endocardial Unipolar LVZ area (%) and RV epicardial bipolar scar area (%) falls close to the zero line (0% outside the deviation). Comparison between groups with different scar extensions Table 1 shows the comparison of the baseline characteristics in the patients with ARVC between the different groups. There was no significant difference in the age, gender, and presentations except for syncope, hypertension, diabetes, LV ejection fraction, Task Force Criteria, and LV diameter between the groups. More horizontal scar patients (group 3) with ARVC experienced syncope (50.0% vs. 42.4% vs. 76.0%, P = 0.021). Additionally, more horizontal scar patients (group 3) presented with sustained VT, and fatal VAs (18.2% vs. 48.5% vs. 52.0%, P = 0.034; 9.1% vs. 15.2% vs. 68.0%, P < 0.001) in spite of a lesser PVC burden of more than 1000/5000beats per day (90.9% vs. 45.5% vs. 44.0%, P = 0.001; 72.7% vs. 42.4% vs. 8.0%, P < 0.001, respectively). The data from these substrate maps are summarized in Table 2. No statistically significant differences in the endocardial mean unipolar/bipolar voltage or late potential area were found between the groups. The horizontal scar patients had a greater endocardial unipolar LVZ (34 ± 16 vs. 48 ± 23 vs. 77 ± 35 cm2, P < 0.001), and smaller RV endocardial bipolar scar (22 ± 25 vs. 30 ± 23 vs. 22 ± 22 cm2, P = 0.018). Additionally, horizontal scar patients had more epicardial bipolar LVZs by percentage (12 ± 8 vs. 18 ± 10 vs. 45 ± 23%, P < 0.001), arrhythmogenic potential areas (39 ± 8 vs. 31 ± 6 vs. 78 ± 22 cm2, P = 0.006), and a greater epicardial-endocardial scar gradient between the groups (7 ± 3 vs. 15 ± 4 vs. 28 ± 6%, P < 0.001). Figures 2 and 3 show an example of the Group 1 and Group 3 patients with transmural scar and horizontal scar, respectively. Table 2 EP study and substrate characteristics Groups  Overall  Group 1  Group 2  Group 3  P value  Inducibility  N = 80  N = 22  N = 33  N = 25     Inducible non-sustained VT  68(85.0%)  19(86.4%)  29(87.9%)  20(80.0%)  0.889   Inducible sustained VT  29(36.3%)  5 (22.7%)  10 (30.3%)  14 (56.0%)  0.039   Inducible VF  19(23.8%)  1 (4.5%)  4 (12.1%)  14 (56.0%)  <0.001  Right ventricular volume (mL)  150±41  139±36  155±40  153±41  0.396  Substrate characteristics            RV endocardium             Mean unipolar voltage (mV)  5.2±1.5  5.6±2.4  5.4±1.5  4.8±0.9  0.315   Mean bipolar voltage (mV)  2.6±1.7  2.7±1.5  2.9±1.9  2.2±1.5  0.397   Total activation time (ms)  158±33  157±38  153±29  164±16  0.378   Total area (cm2)  237±54  220±61  239±36  249±61  0.317   Unipolar LVZ(cm2)  53±31  34±16  48±23  77±35  <0.001   Unipolar LVZ (%)  21±8  16±8  21±9  30±7  <0.001   Bipolar LVZ (cm2)  25±22  19±16  14±16  7±9  0.485   Bipolar LVZ (%)  11±9  10±8  6±7  3±4  0.498   Bipolar scar (cm2)  13±14  22±25  30±23  22±22  0.018   Bipolar scar (%)  6±6  10±14  12±16  8±4  0.004   Late potentials area (cm2)  13±20  12±25  11±11  17±23  0.624   Late potentials area (%)  5±6  4±7  4±4  6±7  0.591   Arrhythmogenic potentials area (cm2)  8±6  7±5  6±3  11±7  0.127   Arrhythmogenic potentials area (%)  5±5  5±2  5±1  7±2  0.296  RV epicardium  N = 34  N = 10  N = 10  N = 14     Mean bipolar voltage (mV)  1.5±0.8  1.7±0.4  1.3±0.6  1.7±1.0  0.552   Total activation time (ms)  192±81  182±58  188±70  197±78  0.635   Total area (cm2)  435±130  402±132  492±169  404±88  0.234   Bipolar scar (cm2)  41±34  45±20  91±64  116±106  0.157   Bipolar scar (%)  14±12  12±8  28±10  45±23  <0.001   Late potentials area (cm2)  39±26  32±20  26±19  49±28  0.127   Late potentials area (%)  10±6  8±7  8±6  12±7  0.296   Arrhythmogenic potentials area (cm2)  37±20  39±8  31±6  78±22  0.006   Arrhythmogenic potentials area (%)  14±5  10±2  9±2  20±5  0.002  Epi-endo scar gradient (%)a  16±11  7±3  15±4  28±6  <0.001  Groups  Overall  Group 1  Group 2  Group 3  P value  Inducibility  N = 80  N = 22  N = 33  N = 25     Inducible non-sustained VT  68(85.0%)  19(86.4%)  29(87.9%)  20(80.0%)  0.889   Inducible sustained VT  29(36.3%)  5 (22.7%)  10 (30.3%)  14 (56.0%)  0.039   Inducible VF  19(23.8%)  1 (4.5%)  4 (12.1%)  14 (56.0%)  <0.001  Right ventricular volume (mL)  150±41  139±36  155±40  153±41  0.396  Substrate characteristics            RV endocardium             Mean unipolar voltage (mV)  5.2±1.5  5.6±2.4  5.4±1.5  4.8±0.9  0.315   Mean bipolar voltage (mV)  2.6±1.7  2.7±1.5  2.9±1.9  2.2±1.5  0.397   Total activation time (ms)  158±33  157±38  153±29  164±16  0.378   Total area (cm2)  237±54  220±61  239±36  249±61  0.317   Unipolar LVZ(cm2)  53±31  34±16  48±23  77±35  <0.001   Unipolar LVZ (%)  21±8  16±8  21±9  30±7  <0.001   Bipolar LVZ (cm2)  25±22  19±16  14±16  7±9  0.485   Bipolar LVZ (%)  11±9  10±8  6±7  3±4  0.498   Bipolar scar (cm2)  13±14  22±25  30±23  22±22  0.018   Bipolar scar (%)  6±6  10±14  12±16  8±4  0.004   Late potentials area (cm2)  13±20  12±25  11±11  17±23  0.624   Late potentials area (%)  5±6  4±7  4±4  6±7  0.591   Arrhythmogenic potentials area (cm2)  8±6  7±5  6±3  11±7  0.127   Arrhythmogenic potentials area (%)  5±5  5±2  5±1  7±2  0.296  RV epicardium  N = 34  N = 10  N = 10  N = 14     Mean bipolar voltage (mV)  1.5±0.8  1.7±0.4  1.3±0.6  1.7±1.0  0.552   Total activation time (ms)  192±81  182±58  188±70  197±78  0.635   Total area (cm2)  435±130  402±132  492±169  404±88  0.234   Bipolar scar (cm2)  41±34  45±20  91±64  116±106  0.157   Bipolar scar (%)  14±12  12±8  28±10  45±23  <0.001   Late potentials area (cm2)  39±26  32±20  26±19  49±28  0.127   Late potentials area (%)  10±6  8±7  8±6  12±7  0.296   Arrhythmogenic potentials area (cm2)  37±20  39±8  31±6  78±22  0.006   Arrhythmogenic potentials area (%)  14±5  10±2  9±2  20±5  0.002  Epi-endo scar gradient (%)a  16±11  7±3  15±4  28±6  <0.001  a Epicardial scar (%) was measured by bipolar signal amplitude of <1.0 mV from the epicardium or unipolar signal amplitude of < 5.5 mV from the endocardium. LVZ, low-voltage zone; RV, right ventricle; VF, ventricular fibrillation; VT, ventricular tachycardia. Figure 2 View largeDownload slide An ARVC patient with a transmural scar (Group 1). A: (Left panel) The percentage of the scar area was 15.0% in the endocardium and 15.0% in the epicardium, respectively (Group 1). (Center panel) The latest RV activation was seen overlying the inferior right ventricular free wall on the endocardium and epicardium. (Right panel) The SAFE-T map showing the arrhythmogenic potentials located at the same location as the latest right ventricular activation in the inferior right ventricular free wall. 1B: Multiple sustained VTs other than VF were induced in the EP study. ARVC, arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle; SAFE-T, Simultaneous Amplitude Frequency Electrogram Transformation; VF, ventricular fibrillation. Figure 2 View largeDownload slide An ARVC patient with a transmural scar (Group 1). A: (Left panel) The percentage of the scar area was 15.0% in the endocardium and 15.0% in the epicardium, respectively (Group 1). (Center panel) The latest RV activation was seen overlying the inferior right ventricular free wall on the endocardium and epicardium. (Right panel) The SAFE-T map showing the arrhythmogenic potentials located at the same location as the latest right ventricular activation in the inferior right ventricular free wall. 1B: Multiple sustained VTs other than VF were induced in the EP study. ARVC, arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle; SAFE-T, Simultaneous Amplitude Frequency Electrogram Transformation; VF, ventricular fibrillation. Figure 3 View largeDownload slide An ARVC patient with a horizontal scar presenting with VF. (A): A time-frequency analysis of the bipolar electrograms with or without high frequency components in normal or low voltage zones (numbers 1, 2, 3, 4 correspond to the locations in Figure 4B); (B): (Left panel) The scar in the endocardial and epicardial bipolar voltage maps was 4.3% and 65%, respectively (Group 3). (Center panel) The latest epicardial RV activation was seen overlying the RV posterior epicardium. (Right panel) A SAFE-T map demonstrating the distribution of the arrhythmogenic potentials, which contained the isthmus (dot line) for one stable VT in Figure 4C. (D): The documented ECG in the same patient exhibiting ventricular fibrillation. Abbreviation as in Figure 2. Figure 3 View largeDownload slide An ARVC patient with a horizontal scar presenting with VF. (A): A time-frequency analysis of the bipolar electrograms with or without high frequency components in normal or low voltage zones (numbers 1, 2, 3, 4 correspond to the locations in Figure 4B); (B): (Left panel) The scar in the endocardial and epicardial bipolar voltage maps was 4.3% and 65%, respectively (Group 3). (Center panel) The latest epicardial RV activation was seen overlying the RV posterior epicardium. (Right panel) A SAFE-T map demonstrating the distribution of the arrhythmogenic potentials, which contained the isthmus (dot line) for one stable VT in Figure 4C. (D): The documented ECG in the same patient exhibiting ventricular fibrillation. Abbreviation as in Figure 2. Among the 34 patients with available epicardial maps, there were no statistically significant differences in the epicardial mean unipolar/bipolar voltage or late potential area between three groups. Additionally, horizontal scar patients had more epicardial bipolar scar by percentage (12 ± 8 vs. 18 ± 10 vs. 45 ± 23%, P < 0.001) and arrhythmogenic potential areas (39 ± 8 vs. 31 ± 6 vs. 78 ± 22 cm2, P = 0.006). EP study, VT mapping, and catheter ablation Of the 80 ARVC patients, a total of 182 VT were induced, including 75 sustained VTs in 29 patients, 107 non-sustained VTs in 68 patients, and VF in 19 patients. Of the 75 sustained VTs, 22 (29.3%) were mappable VTs and 41 (70.7%) were unmapable. Activation mapping and entrainment were achieved in 22 stable VTs, including 6 (27.3%) VTs in Group 1, 5 (22.7%) VTs in Group 2, and 11 (50%) in Groups 3 (P = 0.255). The center of the isthmuses were located within dense scar (bipolar voltage < 0.5 mV) in 6 (27.3%), at border zones (bipolar voltage: 0.5–1.5 mV) in 16 (72.7%), in areas with LPs in 15 (68.1%), and in areas with arrhythmogenic potentials in 22 (100.0%). The exits of the isthmuses were identified within dense scar in 18 (81.8%), at border zones in 4 (18.2%), in areas with LPs in 2 (9.1%), and in areas with arrhythmogenic potentials in 22 (100.0%). A total of 72 conducting channels were identified by activation maps during sinus rhythm, including 23 (19.4%) in Group 1, 23 (30.6%) in Group 2, and 26 in Group 3 (P = 0.543). All the conducting channels were located within dense scar or at border zones of scar. Arrhythmogenic potentials could be identified in all conducting isthmuses and LPs could be identified in 52 (72.2%) conducting isthmuses. Additionally, 107 triggers were identified, including 34 (31.7%) in Group 1, 41 (38.3%) in Group 2, and 42 (39.3%) in Group 2 (P = 0.337). Of these triggers, 81 (75.7%%) were located within border zones, 26 (24.3%) originated from areas nearby border zones, 3 (2.8%) were within areas with LPs, and 72 (67.3%) were within areas with arrhythmogenic potentials. Acute procedural success was achieved in all patients without any inducible VF or sustained VTs after the catheter ablation under a uniform induction protocol with and without an intravenous infusion of 1–5 μg/min isoprenaline to achieve at least a 20% heart rate increment. Residual LPs or arrhythmogenic potentials were not eliminated in 12 patients due to a prolonged procedure time in 3 patients (25.0%), patient intolerance in 4 (33.3%), and nearby coronary arteries or phrenic nerves in 5 (41.7%). Two patients were complicated with pulmonary edema with prolonged mechanical ventilation after the procedure and were smoothly discharged after 30 and 27 days of hospitalization, respectively. One patient was complicated with a pseudo aneurysm and recovered after a prolonged compression. Univariate and multivariate logistic regression analyses for odds ratios of fatal VAs were performed in patients with an epicardial-endocardial approach (Table 3) and the whole study population (Table 4). After multivariate adjustment, the epicardial-endocardial scar gradient was independently associated with fatal VAs in ARVC patients. Table 3 Uni/multivariate analysis of the fatal VA incidence in ARVC patients with epi/endo mapping (34 patients)   Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.278  1.03 (0.98–1.09)       Gender  0.239  2.86 (0.50–16.43)       Diabetes mellitus  0.518  2.00 (0.24–16.44)       Hypertension  0.442  1.75 (0.42–7.28)      RV endocardium           Mean bipolar voltage (mV)  0.227  1.81 (0.69–4.76)       Mean unipolar voltage (mV)  0.867  1.05 (0.57–1.93)       Total activation time (ms)  0.226  1.01 (0.99–1.02)       Bipolar scar (%)  0.019  0.81 (0.68–0.97)  0.702  1.09 (0.71–1.66)   Late potentials area (%)  0.368  1.05 (0.95–1.15)       Arrhythmogenic potentials area (cm2)  0.850  1.02 (0.82–1.27)      RV epicardium           Mean bipolar voltage (mV)  0.346  1.60 (0.60–1.21)       Mean unipolar voltage (mV)  0.900  1.02 (0.78–1.33)       Total activation time (ms)  0.485  1.00 (0.99–1.01)       Bipolar scar (%)  0.028  1.05 (1.01–1.09)  0.630  0.91 (0.63–1.33)   Late potentials area (%)  0.018  1.22 (1.04–1.44)  0.510  0.83 (0.47–1.45)   Arrhythmogenic potentials area (%)  0.004  1.73 (1.19–2.52)  0.196  3.30 (0.54–1.15)  Substrate characteristics           Epi/Endo Scar Gradient  <0.001  1.21 (1.11–1.31)  0.035  1.67 (1.04–2.68)    Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.278  1.03 (0.98–1.09)       Gender  0.239  2.86 (0.50–16.43)       Diabetes mellitus  0.518  2.00 (0.24–16.44)       Hypertension  0.442  1.75 (0.42–7.28)      RV endocardium           Mean bipolar voltage (mV)  0.227  1.81 (0.69–4.76)       Mean unipolar voltage (mV)  0.867  1.05 (0.57–1.93)       Total activation time (ms)  0.226  1.01 (0.99–1.02)       Bipolar scar (%)  0.019  0.81 (0.68–0.97)  0.702  1.09 (0.71–1.66)   Late potentials area (%)  0.368  1.05 (0.95–1.15)       Arrhythmogenic potentials area (cm2)  0.850  1.02 (0.82–1.27)      RV epicardium           Mean bipolar voltage (mV)  0.346  1.60 (0.60–1.21)       Mean unipolar voltage (mV)  0.900  1.02 (0.78–1.33)       Total activation time (ms)  0.485  1.00 (0.99–1.01)       Bipolar scar (%)  0.028  1.05 (1.01–1.09)  0.630  0.91 (0.63–1.33)   Late potentials area (%)  0.018  1.22 (1.04–1.44)  0.510  0.83 (0.47–1.45)   Arrhythmogenic potentials area (%)  0.004  1.73 (1.19–2.52)  0.196  3.30 (0.54–1.15)  Substrate characteristics           Epi/Endo Scar Gradient  <0.001  1.21 (1.11–1.31)  0.035  1.67 (1.04–2.68)  ARVC, arrhythmogenic right ventricular dysplasia; CI, confidence interval; OR, odds ratio; RV, right ventricle; VA, ventricular arrhythmia. Table 4 Uni/multivariate analysis of the fatal VA incidence in whole study (80 patients)   Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.111  1.03 (0.99–1.07)       Gender  0.280  1.73 (0.64–4.70)       Diabetes mellitus  0.205  2.60 (0.59–11.41)       Hypertension  0.431  1.50 (0.55–4.12)      RV endocardium           Mean bipolar voltage (mV)  0.347  0.81 (0.53–1.25)       Mean unipolar voltage (mV)  0.730  1.06 (0.78–1.44)       Total activation time (ms)  0.112  1.01 (0.99–1.02)       Bipolar scar (%)  0.094  0.92 (0.84–1.01)       Unipolar low-voltage area (%)  <0.001  1.15 (1.06–1.24)  0.058  1.09 (0.99–1.19)   Late potentials area (%)  0.128  1.08 (0.99–1.20)       Arrhythmogenic potentials area (cm2)  0.263  1.15 (0.90–1.48)      Substrate characteristics           Epi/endo scar gradienta  <0.001  1.21 (1.11–1.31)  0.001  5.47 (2.00–14.97)    Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.111  1.03 (0.99–1.07)       Gender  0.280  1.73 (0.64–4.70)       Diabetes mellitus  0.205  2.60 (0.59–11.41)       Hypertension  0.431  1.50 (0.55–4.12)      RV endocardium           Mean bipolar voltage (mV)  0.347  0.81 (0.53–1.25)       Mean unipolar voltage (mV)  0.730  1.06 (0.78–1.44)       Total activation time (ms)  0.112  1.01 (0.99–1.02)       Bipolar scar (%)  0.094  0.92 (0.84–1.01)       Unipolar low-voltage area (%)  <0.001  1.15 (1.06–1.24)  0.058  1.09 (0.99–1.19)   Late potentials area (%)  0.128  1.08 (0.99–1.20)       Arrhythmogenic potentials area (cm2)  0.263  1.15 (0.90–1.48)      Substrate characteristics           Epi/endo scar gradienta  <0.001  1.21 (1.11–1.31)  0.001  5.47 (2.00–14.97)  a Epicardial scar (%) was measured by a bipolar signal amplitude of <1.0 mV from the epicardium or unipolar signal amplitude of < 5.5 mV from the endocardium. ARVC, arrhythmogenic right ventricular dysplasia; CI, confidence interval; OR, odds ratio; RV, right ventricle; VA, ventricular arrhythmia. The recurrences of VAs after catheter ablation between the different groups during the follow-up During a mean follow-up period of 38 ± 11 months (9–93 months), a total of 39 patients (48.8%) developed recurrences with a mean duration of 35 ± 19 months after an acute successful ablation, including 10 patients (12.5%) with recurrences of VF/ICD shocks with a mean duration of 28 ± 13 months, 18 (22.5%) with recurrences of sustained VT/ICD antitachycardia pacing therapy with a mean duration of 25 ± 13 months, 13 (16.3%) with recurrences of PVCs > 5000 beats per day with a mean duration of 26 ± 17 months, 29 (36.3%) with recurrences of PVCs of >1000 beats per day with a mean duration of 21 ± 16 months, and 23 (%) that received repeated procedures at a mean duration 29 ± 19 months. Among the 23 patients with redo-procedures, 5 (21.7%) developed recurrence of ventricular arrhythmia in the mean follow-up 27 ± 15 months (3 non-sustained VT 2 PVCs > 5000/day, and 3 received third ablation). Two patients (2.5%) died of non-cardiac diseases (pneumonia). Of the 36 patients with PVC/VT recurrences, recurrent VAs with the same or similar ECG morphology as documented before were detected in 32 (88.9%) patients and different morphologies in 4 (11.1%). A Kaplan–Meier curve analysis demonstrated higher recurrences of VF and VT in the horizontal scar group than the other groups (Log-rank P < 0.001, Figure 4). Figure 4 View largeDownload slide Kaplan–Meier survival Curve according to the Scar Distribution Patterns. A–F) Kaplan–Meier curve of the re-ablation, total recurrence, 1000 PVC beats/day recurrence, 5000 PVC beats/day recurrence, VF recurrence, and VF recurrence free survival curve in the different scar distribution patterns. PVC, premature ventricular complex; VT, ventricular tachycardia; VF, ventricular fibrillation. Group 1: Epi-Endo <10%; Group 2: Epi-Endo >10% and < 20%; Group 3: Epi-Endo >20%. Figure 4 View largeDownload slide Kaplan–Meier survival Curve according to the Scar Distribution Patterns. A–F) Kaplan–Meier curve of the re-ablation, total recurrence, 1000 PVC beats/day recurrence, 5000 PVC beats/day recurrence, VF recurrence, and VF recurrence free survival curve in the different scar distribution patterns. PVC, premature ventricular complex; VT, ventricular tachycardia; VF, ventricular fibrillation. Group 1: Epi-Endo <10%; Group 2: Epi-Endo >10% and < 20%; Group 3: Epi-Endo >20%. Discussion Main findings First, this study showed that ARVC patients with transmural scars (epicardial-endocardial scar gradient < 10%, extending vertically) had a greater number of clinical and EP lab induced occurrences of PVCs and fewer VF episodes. In contrast, ARVC patients with horizontal scar (Group 3) had fewer clinical isolated PVCs, but a higher incidence of VF episodes. Second, the scar discrepancy was independently correlated with the occurrence of VF. Third, the total, VT and VF recurrences were higher in the ARVC patients with a horizontal scar than in those with transmural scar. Last, the arrhythmogenic potentials were more prominent in the patients with a horizontal scar distribution. ARVC and fatal VAs The present study describes the importance of scar discrepancy in the clinical presentation of fatal VAs and the inducibility of VF in the EP lab. This novel finding might help clinical electrophysiologists predict VF occurrences and develop decision strategies for secondary or primary prevention. In a previous study, patients meeting the criteria for ICD implantations had a higher rate of rapid and potentially life-threatening arrhythmias.3 Environmental factors including higher temperatures, especially in the summer, and larger variations in the humidity within 3 days of the event, were independently associated with the development of certain events.10 In another study involving 136 ARVC patients who received ICD implantations and were followed up for 39 ± 25 months, SCD or hemodynamic unstable VAs and left ventricular involvement were independent predictors of VF/flutter.18 Generally, the acute success of catheter ablation was feasible and obtained in 71–100% of patients with ARVC19 with various ablation endpoints. Acute successful ablation was achieved in our current study population without inducible sustained VT or VF under isoprenaline after catheter ablation. In patients with frequent inducible unstable VT or VF, general anesthesia was applied for requirement of repeated direct current shock. Post-ablation inducibility was performed under sedative status without interruption in these patients. The sedative medication might bias the result of the final inducibility, which was higher than previous reports. Horizontal distribution scar and fatal VAs Data from observational clinical studies on ARVC have provided a number of clinical predictors, including VT/VF episodes and unexplained syncope, of adverse events and death. Other independent risk factors for arrhythmogenic events include dysfunction of the RV, left ventricle, or both, a male gender, gene mutations, inducibility, and the amount of electroanatomical scar.19 Currently, there is no consensus on the association of the development of fatal VAs in ARVC patients. In our study, we identified horizontal scar as an independent predictor of the clinical presentation of VF. The epicardial arrhythmogenic potential area is more prominent in patients with a horizontal scar. Local horizontal progression of scar might lead to aberrations in the action potentials and development of an ionic gradient between the layers with a consequent development of slow conduction. This would result in the formation of arrhythmogenic potentials in the epicardial area, which becomes a substrate for reentry in phase 2 and gives rise to polymorphic VT or VF.20 In the substrate mapping, the number of identified conducting channels did not significantly differ. However, the arrhythmogenic potentials might suggest functional intramural conducting channels, which were masked in the electroanatomic mapping. Intramural unstable scar might play an important role in the presentation of fatal VAs. Intramural channels for reentry have been previously demonstrated during VT or VF.21 However, whether stable intramural reentry occurs during VF and to what extent it is required to maintain VF are unanswered questions. It is possible that an intramural disease substrate providing functional channels could serve as a source of rapid activation during VF. These channels were usually spatiotemporally unstable and could not maintain monomorphic VT.21 The substrate characteristics of our present study revealed that scar with a horizontal distribution was a more important factor of the presentation of fatal VAs and inducibility of VF in EP studies. Patients in Group 3 might presented with more advanced stage of ARVC, which was associated with higher incidence of fatal arrhythmia. However, current literature suggests that advanced stage of ARVC was associated with progressive RV dilatation and limited scar progression.9 The similar RV volume in three groups didn’t support the difference in stage of ARVC. (Table 2) The fatal VAs were associated with electric instability in ARVC patients. Our present study showed that the RV scar endo-epicardial gradient in substrate maps might play a substantial role in the clinical VA characteristics and recurrence after catheter ablation. The cardiac magnetic resonance image might provide useful information in ARVC diagnosis but have limited value in the evaluation of the RV scar endo-epicardial gradient and in the risk stratification for VA recurrence before the ablation procedure. The correlation of the RV endocardial unipolar/epicardial bipolar scar The association between RV endocardial unipolar voltage and RV epicardial scar was firstly report by Hutchinson MD et al.14 In our retrospective study, epicardial mapping and ablation was only performed in 42.5% (34 of 80) patients refractory to endocardial ablation. Statistical analysis for correlation and agreements showed high correlation between RV epicardial bipolar scar and RV endocardial unipolar voltage in these 34 patients. Based on the statistical results, we could predict the epi-endocardial scar discrepancy from RV endocardial substrate. Further prospective trials with both epicardial/endocardial mapping are needed to prove the hypotheses that this finding is clinically useful, reproducible, and generalizable to all ARVC patients. Study limitations First, only patients with a high arrhythomogenic burden and who were refractory to antiarrhythmic medications were included in the substrate analysis. The data may not be extrapolated to all patients with ARVC patients. Second, epicardial mapping was performed only in patients refractory to endocardial ablation because of the ablation protocol in our hospital. Third, epicardial scar predicted by endocardial unipolar mapping remains an indirect predictor. Further prospective trials with both epicardial/endocardial mapping are required. Conclusions In ARVC, the epicardial substrate that extended in the horizontal plane rather than transmural plane provided the arrhythmogenic substrate for confined epicardial VT circuits. A higher epicardial arrhythmogenic potential density promoted arrhythmogenicity and was more likely to predispose to electrical instability during the first presentation despite a low PVC load. The heterogeneous distribution of scar between the epicardium and endocardium seems to play a role in the arrhythmogenicity of ARVC. Acknowledgements This work was supported by Ministry of Science and Technology of Taiwan for National Yang-Ming University, Taipei Veterans General Hospital (MOST102-2314-B-010-056-MY2, 103-2314-B-010-048-MY3, 104-2314-B-010 -055-MY3); Grant of Taipei Veterans General Hospital (V104E7-001, V105C-122); Clinical trial in Taipei Veterans General Hospital (C13-092) supported by Biosense Webster (IIS 290); Taipei Veterans General Hospital-National Yang-Ming University Excellent Physicians Scientists Cultivation Program (105-V-B-021). Conflicts of interest: none declared. References 1 Haqqani HM, Tschabrunn CM, Betensky BP, Lavi N, Tzou WS, Zado ES et al.   Layered activation of epicardial scar in arrhythmogenic right ventricular dysplasia: possible substrate for confined epicardial circuits. Circ Arrhythm Electrophysiol  2012; 5: 796– 803. Google Scholar CrossRef Search ADS PubMed  2 Basso C, Bauce B, Corrado D, Thiene G. Pathophysiology of arrhythmogenic cardiomyopathy. Nat Rev Cardiol  2012; 9: 223– 33. Google Scholar CrossRef Search ADS   3 Li CH, Lin YJ, Huang JL, Wu TJ, Cheng CC, Lin WS et al.   Long-term follow-up in patients with arrhythmogenic right ventricular cardiomyopathy. J Cardiovasc Electrophysiol  2012; 23: 750– 6. Google Scholar CrossRef Search ADS PubMed  4 Basso C, Corrado D, Marcus FI, Nava A, Thiene G. Arrhythmogenic right ventricular cardiomyopathy. Lancet  2009; 373: 1289– 300. Google Scholar CrossRef Search ADS PubMed  5 Corrado D, Basso C, Thiene G, McKenna WJ, Davies MJ, Fontaliran F et al.   Spectrum of clinicopathologic manifestations of arrhythmogenic right ventricular cardiomyopathy/dysplasia: a multicenter study. J Am Coll Cardiol  1997; 30: 1512– 20. Google Scholar CrossRef Search ADS PubMed  6 Basso C, Thiene G, Corrado D, Angelini A, Nava A, Valente M. Arrhythmogenic right ventricular cardiomyopathy. Dysplasia, dystrophy, or myocarditis? Circulation  1996; 94: 983– 91. Google Scholar CrossRef Search ADS PubMed  7 Garcia FC, Bazan V, Zado ES, Ren JF, Marchlinski FE. Epicardial substrate and outcome with epicardial ablation of ventricular tachycardia in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Circulation  2009; 120: 366– 75. Google Scholar CrossRef Search ADS PubMed  8 Berruezo A, Acosta J, Fernández-Armenta J, Pedrote A, Barrera A, Arana-Rueda E et al.   Safety, long-term outcomes and predictors of recurrence after first-line combined endoepicardial ventricular tachycardia substrate ablation in arrhythmogenic cardiomyopathy. Impact of arrhythmic substrate distribution pattern. A prospective multicentre study. Europace  2017; 19: 607– 16. Google Scholar PubMed  9 Riley MP, Zado E, Bala R, Cooper J, Dixit S, Garcia F et al.   Lack of uniform progression of endocardial scar in patients with arrhythmogenic right ventricular dysplasia/cardiomyopathy and ventricular tachycardia. Cir Arrhythm Electrophysiol  2010; 3: 332– 8. Google Scholar CrossRef Search ADS   10 Marcus FI, McKenna WJ, Sherrill D, Basso C, Bauce B, Bluemke DA et al.   Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the Task Force Criteria. Eur Heart J  2010; 31: 806– 14. Google Scholar CrossRef Search ADS PubMed  11 Chung FP, Li HR, Chong E, Pan CH, Lin YJ, Chang SL et al.   Seasonal variation in the frequency of sudden cardiac death and ventricular tachyarrhythmia in patients with arrhythmogenic right ventricular dysplasia/cardiomyopathy: the effect of meteorological factors. Heart Rhythm  2013; 10: 1859– 66. Google Scholar CrossRef Search ADS PubMed  12 Marchlinski FE, Callans DJ, Gottlieb CD, Zado E. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation  2000; 101: 1288– 96. Google Scholar CrossRef Search ADS PubMed  13 Polin GM, Haqqani H, Tzou W, Hutchinson MD, Garcia FC, Callans DJ et al.   Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Heart Rhythm  2011; 8: 76– 83. Google Scholar CrossRef Search ADS PubMed  14 Sosa E, Scanavacca M, d'Avila A, Pilleggi F. A new technique to perform epicardial mapping in the electrophysiology laboratory. J Cardiovasc Electrophysiol  1996; 7: 531– 6. Google Scholar CrossRef Search ADS PubMed  15 Hutchinson MD, Gerstenfeld EP, Desjardins B, Bala R, Riley MP, Garcia FC et al.   Endocardial unipolar voltage mapping to detect epicardial ventricular tachycardia substrate in patients with nonischemic left ventricular cardiomyopathy. Cir Arrhythm Electrophysiol.  2011; 4: 49– 55. Google Scholar CrossRef Search ADS   16 Lin CY, Silberbauer J, Lin YJ, Lo MT, Lin C, Chang HC et al.   Simultaneous amplitude frequency electrogram transformation (SAFE-T) mapping to identify ventricular tachycardia arrhythmogenic potentials in sinus rhythm. JACC: Clin Electrophysiol  2016; 2: 459– 70. Google Scholar CrossRef Search ADS   17 Thiene G, Nava A, Corrado D, Rossi L, Pennelli N. Right ventricular cardiomyopathy and sudden death in young people. N Engl J Med  1988; 318: 129– 33. Google Scholar CrossRef Search ADS PubMed  18 Corrado D, Leoni L, Link MS, Della Bella P, Gaita F, Curnis A et al.   Implantable cardioverter-defibrillator therapy for prevention of sudden death in patients with arrhythmogenic right ventricular cardiomyopathy/dysplasia. Circulation  2003; 108: 3084– 91. Google Scholar CrossRef Search ADS PubMed  19 Corrado D, Wichter T, Link MS, Hauer RN, Marchlinski FE, Anastasakis A et al.   Treatment of arrhythmogenic right ventricular cardiomyopathy/dysplasia: an international task force consensus statement. Circulation  2015; 132: 441– 53. Google Scholar CrossRef Search ADS PubMed  20 Gussak I, Antzelevitch C, Bjerregaard P, Towbin JA, Chaitman BR. The Brugada syndrome: clinical, electrophysiologic and genetic aspects. J Am Coll Cardiol  1999; 33: 5– 15. Google Scholar CrossRef Search ADS PubMed  21 Valderrabano M, Lee MH, Ohara T, Lai AC, Fishbein MC, Lin SF et al.   Dynamics of intramural and transmural reentry during ventricular fibrillation in isolated swine ventricles. Circ Res  2001; 88: 839– 48. Google Scholar CrossRef Search ADS PubMed  Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions, please email: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Europace Oxford University Press

Heterogeneous distribution of substrates between the endocardium and epicardium promotes ventricular fibrillation in arrhythmogenic right ventricular dysplasia/cardiomyopathy

Loading next page...
 
/lp/ou_press/heterogeneous-distribution-of-substrates-between-the-endocardium-and-xEHmjK1FSq
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions, please email: journals.permissions@oup.com.
ISSN
1099-5129
eISSN
1532-2092
D.O.I.
10.1093/europace/euw393
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims Whether the distribution of scar in arrhythmogenic right ventricular cardiomyopathy (ARVC) plays a role in predicting different types of ventricular arrhythmias is unknown. This study aimed to investigate the prognostic value of scar distribution in patients with ARVC. Methods and results We studied 80 consecutive ARVC patients (46 men, mean age 47 ± 15 years) who underwent an electrophysiological study with ablation. Thirty-four patients receive both endocardial and epicardial mapping. Abnormal endocardial substrates and epicardial substrates were characterized. Three groups were defined according to the epicardial and endocardial scar gradient (<10%: transmural, 10–20%: intermediate, >20%: horizontal, as groups 1, 2, and 3, respectively). Sinus rhythm electrograms underwent a Hilbert–Huang spectral analysis and were displayed as 3D Simultaneous Amplitude Frequency Electrogram Transformation (SAFE-T) maps, which represented the arrhythmogenic potentials. The baseline characteristics were similar between the three groups. Group 3 patients had a higher incidence of fatal ventricular arrhythmias requiring defibrillation and cardiac arrest during the initial presentation despite having fewer premature ventricular complexes. A larger area of arrhythmogenic potentials in the epicardium was observed in patients with horizontal scar. The epicardial-endocardial scar gradient was independently associated with the occurrence of fatal ventricular arrhythmias after a multivariate adjustment. The total, ventricular tachycardia, and VF recurrent rates were higher in Group 3 during 38 ± 21 months of follow-up. Conclusion For ARVC, the epicardial substrate that extended in the horizontal plane rather than transmurally provided the arrhythmogenic substrate for a fatal ventricular arrhythmia circuit. Electroanatomic mapping , Hilbert–Huang transform , Sudden cardiac death , Ventricular fibrillation , Ventricular tachycardia What’s new? The epicardial-endocardial scar gradient was independently associated with the occurrence of fatal ventricular arrhythmias in arrhythmogenic right ventricular cardiomyopathy (ARVC) patients A high epicardial-endocardial scar gradient was associated with a higher ventricular tachycardia/ventricular fibrillation recurrence rate after catheter ablation in ARVC patients. Introduction Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a genetically and clinically heterogeneous disorder of cardiac muscles that is associated with ventricular arrhythmias, right ventricular dysfunction, and the risk of sudden cardiac death (SCD).1,2 The hallmark pathological lesion of ARVC is the transmural loss of the myocardium of the right ventricular (RV) free wall with replacement by fibrofatty tissue.3–5 The RV is predominantly affected, and both pathological and electroanatomic studies show a disproportionate progression of disease on the epicardium compared with the endocardium6–8. The scarring process modifies the transmural activation with a delayed epicardial activation and may predispose to the development of ventricular tachycardia (VT) circuits contained entirely within the epicardium in ARVC.1 Furthermore, previous studies have shown that the rapid progression of macroscopic endocardial scars occurs in only a subset of ARVC patients.9 It is unknown if the distribution of scar either extending horizontally or transmurally plays a role in fatal VAs. We hypothesized that the three-dimensional scar extension/distribution of the epicardium/endocardium in ARVC may play a role in fatal VAs. In this study, we studied the prognostic value of electroanatomic mapping of ARVC patients in terms of voltage mapping, delayed activation areas, and arrhythmogenic potentials through signal analyses. Methods Study population Between 2008 and 2015, 80 patients (46 men [58%] with a mean age of 47 ± 15 years) confirmed to have ARVC based on the 2010 revised Task Force Criteria10 were enrolled consecutively. All patients were admitted for ablation of drug-refractory VAs. We obtained the Institutional Review Board approval to conduct the retrospective study (VGH-IRB No.2014-30 10-004B). We performed 12-lead resting electrocardiography (ECG), signal-averaged ECGs (SAECGs), 24-h Holter monitoring, 2D echocardiography, coronary arteriography, and electrophysiological (EP) studies on all patients before the procedure. Fatal VAs were defined as ventricular fibrillation (VF), irregular tachycardias with regard to their polarity, amplitude, morphology, and sequence of the intracardiac electrograms, with a mean cycle length of <240 ms requiring defibrillation, or SCD receiving cardiopulmonary resuscitation. VT was defined as a regular tachycardia with a mean cycle length of >240 ms.11 EP study A standardized EP study was performed in the fasting state using either conscious sedation or general anesthesia. Anti-arrhythmic drugs were discontinued for a minimum of five half-lives before the radiofrequency catheter ablation. In the absence of spontaneous VT/VF, rapid ventricular pacing and programmed stimulation up to three extrastimuli were performed from the RV apex and/or RV outflow tract. Three basal cycle lengths (500-400-300 ms) were used for programmed stimulation. If VT/VF was still not inducible, intravenous isoprenaline (1–5 μg/min) was infused to achieve at least a 20% heart rate increment. If VT/VF was not inducible during pharmacological provocation, the induction protocol was repeated. The VT QRS morphologies were compared with those of the documented VTs. Pericardial access was obtained by a subxiphoid puncture if a previous endocardial ablation had failed, an epicardial substrate was suspected (based on surface ECG VT morphology), or minimal or no endocardial scar was present. Electroanatomic mapping was performed during sinus rhythm using the CARTO 3 system MEM version (Biosense Webster, Ltd., Haifa, Israel, user defined map [UDM] module). Mapping and catheter ablation was performed with an open-irrigated ablation catheter (Thermocool, Biosense Webster Inc., Diamond Bar, CA, USA). The unipolar filtering was set at 2–240 Hz, and bipolar filtering at 16–500 Hz. Wilson’s central terminal was assigned as the unipolar reference electrode. VT mapping and catheter ablation Induced VTs were analyzed for the cycle length and morphology using an EP recording system (BARD, Inc. LabSystemTM PRO EP Recording System). If an induced VT was hemodynamically tolerated, pacing from the mapping catheter at a cycle-length 20–30 ms faster than the tachycardia cycle-length was performed, while observing the entrainment response. A site was considered a VT mid-isthmus only if it demonstrated (1) concealed fusion on all 12 ECG leads during entrainment, (2) the post-pacing interval was within 30 ms of the VT cycle-length, (3) the stimulus-electrogram interval was within 30 ms of the electrogram-QRS interval following entrainment, and (4) the local electrogram to QRS interval was between 30 and 70% of the VT cycle-length. Confirmed VT isthmus sites were annotated on the map after VT termination into sinus rhythm. If the induced VT was unstable, pace mapping and an arrhythmogenic potential elimination was performed. If the VAs were isolated premature ventricular complexes (PVCs) or non-sustained VT, activation mapping, pacemapping, and an arrhythmogenic potential elimination was performed. RFCA was performed with power settings of 30–50 W endocardially and 25–30 W epicardially. Three-dimensional electroanatomic mapping Endocardial electroanatomic mapping The color display of the electroanatomic mapping included the bipolar electrogram voltage (scar amplitude, <0.5 mV; normal tissue amplitude, >1.5 mV),12 a late potential latency (the interval from the end of the QRS complex to the end of the local electrogram), and an arrhythmogenic potential map to depict normal and abnormal substrates. Normal RV free wall unipolar electrograms were defined as having peak-to-peak amplitudes of ≥ 5.5 mV.13 Epicardial electroanatomic mapping Percutaneous pericardial access was achieved using a Tuohy needle via the subxiphoid approach, as described by Sosa et al.14 An open irrigated catheter was used to perform mapping of the epicardium, with a focus on the RV epicardium. We defined epicardial scar zones as those areas with a bipolar signal amplitude of <1.0 mV. The correlation of the RV endocardial unipolar/epicardial bipolar scar Among patients with epicardial procedures, the correlation of an RV endocardial unipolar low-voltage zone (LVZ, %) and RV epicardial bipolar scar area (%) was studied. Pearson’s correlation index was applied for the correlation between RV epicardial bipolar scar region and RV endocardial unipolar LVZ. Intraclass correlation coefficient and Scatter plots of the Bland–Altman plot were used to evaluate the consistence. The intraclass correlation coefficient and Bland–Altman plot showed a high level of agreement (please see the result section) and the expected epicardial scar area by RV endocardial unipolar LVZ were applied to the rest of patients without epicardial mapping in this study. Definition of transmural scar and horizontal scar We defined the cut-off points of three groups on the basis of the percentiles of the difference (<33rd, 33rd to 66th, >66th percentile) between epicardial scar (bipolar signal amplitude of <1.0 mV from the epicardium or unipolar signal amplitude of < 5.5 mV from the endocardium)15 and the endocardial scar. (<10%: transmural scar, 10–20%: intermediate heterogeneous scar, >20%: horizontal scar, as groups 1, 2, and 3, respectively; Table 1 and Figure 1). Table 1 Baseline characteristics and outcome of the study patients in the three groups Groups  Overall (N = 80)  Group 1(N = 22)  Group 2 (N = 33)  Group 3(N = 25)  P value  Basic data             Age  46±14  50±13  43.3±13  50±14  0.077   Sex (M, %)  46(57.5%)  10(45.5%)  20(60.6%)  16(64.0%)  0.393   Smoking  22(27.5%)  5(25.0%)  10(31.2%)  7(29.2%)  0.889   Syncope  44(55.0%)  11(50.0%)  14(42.4%)  19(76.0%)  0.021   Palpitation  69(86.3%)  20(90.9%)  28(84.8%)  21(84.0%)  0.754   Short of breathe  32(40.0%)  7(31.8%)  15(45.5%)  10(40.0%)  0.600   Hypertension  25(31.3%)  8(36.4%)  7(21.2%)  10(40.0%)  0.258   Diabetes mellitus  6(7.5%)  3(13.6%)  1(3.0%)  2(8.0%)  0.341   HF NYHA I/II  7(8.8%)  1(4.5%)  3(9.1%)  3(12.0%)  0.663   HF NYHA III/IV  2(2.5%)  0(0.0%)  0(0.0%)  2(8.0%)  0.105  Implantable cardioverter defibrillator  56(70.0%)  14(63.6%)  21(63.6%)  21(84.0%)  0.183  Medication before Admission             Amiodarone  33(41.3%)  6(27.3%)  15(45.5%)  12(48.0%)  0.289   Beta-blocker  45(56.3%)  12(54.5%)  21(63.6%)  12(48.0%)  0.485   Mexitil  20(25.0%)  5(22.7%)  7(21.2%)  8(32.0%)  0.617   Propafenone  2(2.5%)  1(4.5%)  1(3.0%)  0(0.0%)  0.590  Medication at discharge             Amiodarone  18(22.5%)  4(18.2%)  9(27.3%)  5(20.0%)  0.685   Beta-blocker  36(45.0%)  9(40.9%)  15(45.5%)  12(48.0%)  0.886   Mexitil  11(13.8%)  2 (9.1%)  3(9.1%)  6(24.0%)  0.200  Structural assessment             LVEDd (mm)  47±5  47±5  46±5  49±6  0.169   LVEF (%)  56±9  54±8  58±9  56±9  0.469   RVEF (%)  42±13  43±12  43±13  40±13  0.687  Characteristic of VAs             Clinical PVC > 1000/day  46(57.5%)  20(90.9%)  15(45.5%)  11(44.0%)  0.001   Clinical PVC > 5000/day  32(40.0%)  16(72.7%)  14(42.4%)  2(8.0%)  <0.001   PVC/Day  9480±6962  17428±14943  9677±14001  2235±5705  <0.001   Clinical non-sustained VT  33(41.3%)  11(50%)  12(36.4%)  8(32.0%)  0.421   Clinical sustained VT  33(41.3%)  4 (18.2%)  16 (48.5%)  13 (52.0%)  0.034   Fatal ventricular arrhythmia  24(30.0%)  2(9.1%)  5 (15.2%)  17(68.0%)  <0.001  Task force criteria            Structural abnormalitiesa  24(30.0%)/41(51.3%)  8(36.4%)/11(50%)  8(24.2%)/18(54.6%)  8(32.0%)/12(48.0%)  0.869  Fibro-fatty replacementa  9(11.3%)/16(20.0%)  3(13.6%)/4(18.2%)  3(9.1%)/3(9.1%)  3(12.0%)/9(36.0%)  0.163  Depolarization changesa  7(8.8%)/65(81.3%)  2(9.1%)/17(77.3%)  2(6.1%)/26(78.7%)  3(12.0%)/22(88.0%)  0.262  Repolarization changea  11(13.8%)/32(40.0%)  4(18.2%)/7(31.8%)  2(6.1%)/14(42.4%)  5(20.0%)/11(44.0%)  0.435  Ventricular arrhythmiasa  11(13.8%)/48(60.0%)  5(22.7%)/17(77.3%)  1(3.0%)/11(33.3%)  5(20.0%)/20(80.0%)  0.533  Family historya  17(21.3%)/7(8.8%)  9(40.9%)/1(4.5%)  2(24.2%)/3(9.1%)  6(24.0%)/3(12.0%)  0.621  Follow-up             Recurrence of VT  18(22.5%)  1(4.5%)  6(18.2%)  11(44.0%)  0.004   Recurrence of VF  9(11.3%)  0(0.0%)  1(3.0%)  8(32.0%)  <0.001   Mortality  2(2.5%)  0(0.0%)  0(0.0%)  2(2.5%)  0.105  Groups  Overall (N = 80)  Group 1(N = 22)  Group 2 (N = 33)  Group 3(N = 25)  P value  Basic data             Age  46±14  50±13  43.3±13  50±14  0.077   Sex (M, %)  46(57.5%)  10(45.5%)  20(60.6%)  16(64.0%)  0.393   Smoking  22(27.5%)  5(25.0%)  10(31.2%)  7(29.2%)  0.889   Syncope  44(55.0%)  11(50.0%)  14(42.4%)  19(76.0%)  0.021   Palpitation  69(86.3%)  20(90.9%)  28(84.8%)  21(84.0%)  0.754   Short of breathe  32(40.0%)  7(31.8%)  15(45.5%)  10(40.0%)  0.600   Hypertension  25(31.3%)  8(36.4%)  7(21.2%)  10(40.0%)  0.258   Diabetes mellitus  6(7.5%)  3(13.6%)  1(3.0%)  2(8.0%)  0.341   HF NYHA I/II  7(8.8%)  1(4.5%)  3(9.1%)  3(12.0%)  0.663   HF NYHA III/IV  2(2.5%)  0(0.0%)  0(0.0%)  2(8.0%)  0.105  Implantable cardioverter defibrillator  56(70.0%)  14(63.6%)  21(63.6%)  21(84.0%)  0.183  Medication before Admission             Amiodarone  33(41.3%)  6(27.3%)  15(45.5%)  12(48.0%)  0.289   Beta-blocker  45(56.3%)  12(54.5%)  21(63.6%)  12(48.0%)  0.485   Mexitil  20(25.0%)  5(22.7%)  7(21.2%)  8(32.0%)  0.617   Propafenone  2(2.5%)  1(4.5%)  1(3.0%)  0(0.0%)  0.590  Medication at discharge             Amiodarone  18(22.5%)  4(18.2%)  9(27.3%)  5(20.0%)  0.685   Beta-blocker  36(45.0%)  9(40.9%)  15(45.5%)  12(48.0%)  0.886   Mexitil  11(13.8%)  2 (9.1%)  3(9.1%)  6(24.0%)  0.200  Structural assessment             LVEDd (mm)  47±5  47±5  46±5  49±6  0.169   LVEF (%)  56±9  54±8  58±9  56±9  0.469   RVEF (%)  42±13  43±12  43±13  40±13  0.687  Characteristic of VAs             Clinical PVC > 1000/day  46(57.5%)  20(90.9%)  15(45.5%)  11(44.0%)  0.001   Clinical PVC > 5000/day  32(40.0%)  16(72.7%)  14(42.4%)  2(8.0%)  <0.001   PVC/Day  9480±6962  17428±14943  9677±14001  2235±5705  <0.001   Clinical non-sustained VT  33(41.3%)  11(50%)  12(36.4%)  8(32.0%)  0.421   Clinical sustained VT  33(41.3%)  4 (18.2%)  16 (48.5%)  13 (52.0%)  0.034   Fatal ventricular arrhythmia  24(30.0%)  2(9.1%)  5 (15.2%)  17(68.0%)  <0.001  Task force criteria            Structural abnormalitiesa  24(30.0%)/41(51.3%)  8(36.4%)/11(50%)  8(24.2%)/18(54.6%)  8(32.0%)/12(48.0%)  0.869  Fibro-fatty replacementa  9(11.3%)/16(20.0%)  3(13.6%)/4(18.2%)  3(9.1%)/3(9.1%)  3(12.0%)/9(36.0%)  0.163  Depolarization changesa  7(8.8%)/65(81.3%)  2(9.1%)/17(77.3%)  2(6.1%)/26(78.7%)  3(12.0%)/22(88.0%)  0.262  Repolarization changea  11(13.8%)/32(40.0%)  4(18.2%)/7(31.8%)  2(6.1%)/14(42.4%)  5(20.0%)/11(44.0%)  0.435  Ventricular arrhythmiasa  11(13.8%)/48(60.0%)  5(22.7%)/17(77.3%)  1(3.0%)/11(33.3%)  5(20.0%)/20(80.0%)  0.533  Family historya  17(21.3%)/7(8.8%)  9(40.9%)/1(4.5%)  2(24.2%)/3(9.1%)  6(24.0%)/3(12.0%)  0.621  Follow-up             Recurrence of VT  18(22.5%)  1(4.5%)  6(18.2%)  11(44.0%)  0.004   Recurrence of VF  9(11.3%)  0(0.0%)  1(3.0%)  8(32.0%)  <0.001   Mortality  2(2.5%)  0(0.0%)  0(0.0%)  2(2.5%)  0.105  a According to the 2010 Revised Task Force Criteria (Major/minor).10 HF indicates heart failure; LVEDd, left ventricular end diastolic diameter; LVEF, Left ventricular ejection fraction; NYHA, New York Heart Association; RVEF, Right ventricular ejection fraction; VF, ventricular fibrillation; VT, ventricular tachycardia; PVC, premature ventricular contraction. Figure 1 View largeDownload slide Substrate analyses according to the epicardium/endocardiun gradient. The ARVC patients were separated into three groups based on the transmural gradient of the scar distribution. ARVC indicates arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle. Figure 1 View largeDownload slide Substrate analyses according to the epicardium/endocardiun gradient. The ARVC patients were separated into three groups based on the transmural gradient of the scar distribution. ARVC indicates arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle. Late potentials were defined as local ventricular potentials occurring after the terminal portion of the surface QRS. Arrhythmogenic potentials were defined as abnormal potentials, including late potentials and abnormal early potentials inscribed within the QRS. The arrhythmogenic potential was visualized by Simultaneous Amplitude Frequency Electrogram Transformation (SAFE-T) mapping.16 SAFE-T mapping was used to identify objective high frequency components in substrate mapping. The arrhythmogenic potentials identified by SAFE-T mapping was associated with the VT circuit in the previous study. Regions of scar, LVZs, and late potentials were measured using the standard surface area measurement tool on the CARTO 3 system MEM Version (UDM Module). When multiple areas of confluent low voltages were present, the aggregate area from individual regions of interest was calculated. Follow-up Patients were followed-up for 1, 2, 3 months, and every 3 months after catheter ablation. Patients with an implantable cardioverter defibrillator (ICD) were interrogated at each visit. For symptomatic paients without an ICD, a detailed history was taken, and ECG, and 24-h Holter monitoring or Event Recorder were performed at each visit. The recurrence of VF was defined as a VA with a cycle length of 240 ms or less. Recurrence of VT was defined as a regular (monomorphic) or irregular (polymorphic) VA with a cycle length of more than 240 ms. The appropriateness of ICD therapies (shock or antitachycardia pacing) was deemed appropriate or inappropriate on the basis of standard criteria.17 The composite of recurrent VF/VT included the recurrence of VF, VT, and appropriate ICD therapies.3,11 The recurrence of PVCs of more than 5000 beats per day was defined as more than 5000 PVCs with single or multiple morphologies in the 24-h ECG recordings. The recurrence of PVCs of more than 1000 beats per day was defined as more than 1000 PVCs with single or multiple morphologies in the 24-h ECG recordings. The total recurrence was a composite of the PVC recurrence, VF/ICD shocks, and VT/ICD antitachycardia pacing. Statistical analysis Continuous variables are expressed as the mean ± standard deviation and were compared using a one-way analysis of variance. Nominal variables were compared using a chi-square test for linear trends. A P value of <0.05 was considered as statistically significant. All statistical analyses were performed using commercially available statistical SPSS version 17.0 software (SPSS, Chicago, IL, USA). Results Baseline characteristics The study was comprised of 80 patients (58% men, 47 ± 13 years) who fulfilled the diagnosis of definite ARVC based on the 2010 Revised Task Force Criteria. Of them, 69 (86.3%) patients experienced palpitations, 44 (55.0%) experienced syncopal episodes, 32 (40.0%) experienced shortness of breath episodes, 25 (31.1%) had hypertension, and 6 (0.1%) had diabetes mellitus. Electrocardiographic studies demonstrated depolarization abnormalities in 72 patients (90.0%), including the presence of epsilon waves in 7 (8.8%) and positive SAECGs in 60 (81.2%), while 43 (53.8%) had inverted T waves in the right precordial leads (V1,V2, and V3) or beyond in the absence of complete right bundle-branch block. Holter examinations documented episodes of sustained VT in 32 patients (40.0%), and PVCs of more than 5000 beats per day in 32 (40.0%), and non-sustained VT in 57 (83.8%). Clinical documented VF, episodes of SCD status post cardiopulmonary resuscitation, or appropriate ICD therapies were recorded in 24 patients (30.0%). Genetic mutation screening was performed in 58 (72.5%) patients and pathogenic mutations10 were identified in 22 patients (38.9%). There were 2 patients (3.4%) with multiple mutations. All patients underwent RV angiography and 62 (77.5%) received magnetic resonance imaging to assess the RV function. From the RV angiographic findings, 65 patients (81.3%) demonstrated regional RV akinesia, dyskinesia, or an aneurysm formation. Histologically, fibro-fatty replacement of the RV was identified in 25 patients (31.3%). EP study In the EP study, 19 patients (23.8%) had inducible VF, 29 (36.3%) had inducible sustained VTs, and 68 (85.0%) had inducible non-sustained VT. The non-inducibility was achieved after ablation with programed extra-stimulation and isoprenaline in all the patients. All 80 patients underwent detailed RV endocardial mapping. Twenty-two (27.5%), 33 (41.3%), and 25 (31.3%) patients exhibited a transmural scar (group 1), intermediate heterogeneous scar (group 2), and horizontal scar (group 3), respectively. In between, epicardial access was attempted and successfully achieved in 34 (42.5%) patients. Among the patients with available epicardial maps, 10 (29.4%), 10 (29.4%), and 14 (41.1%) were classified as group 1, group 2, and group 3, respectively. The consistency of the RV endocardial unipolar/epicardial bipolar scar The correlation of an RV endocardial unipolar low-voltage zone (LVZ, %) and RV epicardial bipolar scar area (%) was studied in 34 patients with epicardial procedures. The RV epicardial bipolar scar region was equivalent to the unipolar LVZ with an intraclass correlation coefficient of 0.956 (P <0.001) and Cronbach’s Alpha value of 0.977. There was no significant difference in the mean value (P = 0.863) with a significant correlation (Pearson’s 0.969, P = 0.001). Scatter plots of the Bland–Altman plot in see Supplementary material online, figure S1B shows that the mean of the difference between the RV endocardial Unipolar LVZ area (%) and RV epicardial bipolar scar area (%) falls close to the zero line (0% outside the deviation). Comparison between groups with different scar extensions Table 1 shows the comparison of the baseline characteristics in the patients with ARVC between the different groups. There was no significant difference in the age, gender, and presentations except for syncope, hypertension, diabetes, LV ejection fraction, Task Force Criteria, and LV diameter between the groups. More horizontal scar patients (group 3) with ARVC experienced syncope (50.0% vs. 42.4% vs. 76.0%, P = 0.021). Additionally, more horizontal scar patients (group 3) presented with sustained VT, and fatal VAs (18.2% vs. 48.5% vs. 52.0%, P = 0.034; 9.1% vs. 15.2% vs. 68.0%, P < 0.001) in spite of a lesser PVC burden of more than 1000/5000beats per day (90.9% vs. 45.5% vs. 44.0%, P = 0.001; 72.7% vs. 42.4% vs. 8.0%, P < 0.001, respectively). The data from these substrate maps are summarized in Table 2. No statistically significant differences in the endocardial mean unipolar/bipolar voltage or late potential area were found between the groups. The horizontal scar patients had a greater endocardial unipolar LVZ (34 ± 16 vs. 48 ± 23 vs. 77 ± 35 cm2, P < 0.001), and smaller RV endocardial bipolar scar (22 ± 25 vs. 30 ± 23 vs. 22 ± 22 cm2, P = 0.018). Additionally, horizontal scar patients had more epicardial bipolar LVZs by percentage (12 ± 8 vs. 18 ± 10 vs. 45 ± 23%, P < 0.001), arrhythmogenic potential areas (39 ± 8 vs. 31 ± 6 vs. 78 ± 22 cm2, P = 0.006), and a greater epicardial-endocardial scar gradient between the groups (7 ± 3 vs. 15 ± 4 vs. 28 ± 6%, P < 0.001). Figures 2 and 3 show an example of the Group 1 and Group 3 patients with transmural scar and horizontal scar, respectively. Table 2 EP study and substrate characteristics Groups  Overall  Group 1  Group 2  Group 3  P value  Inducibility  N = 80  N = 22  N = 33  N = 25     Inducible non-sustained VT  68(85.0%)  19(86.4%)  29(87.9%)  20(80.0%)  0.889   Inducible sustained VT  29(36.3%)  5 (22.7%)  10 (30.3%)  14 (56.0%)  0.039   Inducible VF  19(23.8%)  1 (4.5%)  4 (12.1%)  14 (56.0%)  <0.001  Right ventricular volume (mL)  150±41  139±36  155±40  153±41  0.396  Substrate characteristics            RV endocardium             Mean unipolar voltage (mV)  5.2±1.5  5.6±2.4  5.4±1.5  4.8±0.9  0.315   Mean bipolar voltage (mV)  2.6±1.7  2.7±1.5  2.9±1.9  2.2±1.5  0.397   Total activation time (ms)  158±33  157±38  153±29  164±16  0.378   Total area (cm2)  237±54  220±61  239±36  249±61  0.317   Unipolar LVZ(cm2)  53±31  34±16  48±23  77±35  <0.001   Unipolar LVZ (%)  21±8  16±8  21±9  30±7  <0.001   Bipolar LVZ (cm2)  25±22  19±16  14±16  7±9  0.485   Bipolar LVZ (%)  11±9  10±8  6±7  3±4  0.498   Bipolar scar (cm2)  13±14  22±25  30±23  22±22  0.018   Bipolar scar (%)  6±6  10±14  12±16  8±4  0.004   Late potentials area (cm2)  13±20  12±25  11±11  17±23  0.624   Late potentials area (%)  5±6  4±7  4±4  6±7  0.591   Arrhythmogenic potentials area (cm2)  8±6  7±5  6±3  11±7  0.127   Arrhythmogenic potentials area (%)  5±5  5±2  5±1  7±2  0.296  RV epicardium  N = 34  N = 10  N = 10  N = 14     Mean bipolar voltage (mV)  1.5±0.8  1.7±0.4  1.3±0.6  1.7±1.0  0.552   Total activation time (ms)  192±81  182±58  188±70  197±78  0.635   Total area (cm2)  435±130  402±132  492±169  404±88  0.234   Bipolar scar (cm2)  41±34  45±20  91±64  116±106  0.157   Bipolar scar (%)  14±12  12±8  28±10  45±23  <0.001   Late potentials area (cm2)  39±26  32±20  26±19  49±28  0.127   Late potentials area (%)  10±6  8±7  8±6  12±7  0.296   Arrhythmogenic potentials area (cm2)  37±20  39±8  31±6  78±22  0.006   Arrhythmogenic potentials area (%)  14±5  10±2  9±2  20±5  0.002  Epi-endo scar gradient (%)a  16±11  7±3  15±4  28±6  <0.001  Groups  Overall  Group 1  Group 2  Group 3  P value  Inducibility  N = 80  N = 22  N = 33  N = 25     Inducible non-sustained VT  68(85.0%)  19(86.4%)  29(87.9%)  20(80.0%)  0.889   Inducible sustained VT  29(36.3%)  5 (22.7%)  10 (30.3%)  14 (56.0%)  0.039   Inducible VF  19(23.8%)  1 (4.5%)  4 (12.1%)  14 (56.0%)  <0.001  Right ventricular volume (mL)  150±41  139±36  155±40  153±41  0.396  Substrate characteristics            RV endocardium             Mean unipolar voltage (mV)  5.2±1.5  5.6±2.4  5.4±1.5  4.8±0.9  0.315   Mean bipolar voltage (mV)  2.6±1.7  2.7±1.5  2.9±1.9  2.2±1.5  0.397   Total activation time (ms)  158±33  157±38  153±29  164±16  0.378   Total area (cm2)  237±54  220±61  239±36  249±61  0.317   Unipolar LVZ(cm2)  53±31  34±16  48±23  77±35  <0.001   Unipolar LVZ (%)  21±8  16±8  21±9  30±7  <0.001   Bipolar LVZ (cm2)  25±22  19±16  14±16  7±9  0.485   Bipolar LVZ (%)  11±9  10±8  6±7  3±4  0.498   Bipolar scar (cm2)  13±14  22±25  30±23  22±22  0.018   Bipolar scar (%)  6±6  10±14  12±16  8±4  0.004   Late potentials area (cm2)  13±20  12±25  11±11  17±23  0.624   Late potentials area (%)  5±6  4±7  4±4  6±7  0.591   Arrhythmogenic potentials area (cm2)  8±6  7±5  6±3  11±7  0.127   Arrhythmogenic potentials area (%)  5±5  5±2  5±1  7±2  0.296  RV epicardium  N = 34  N = 10  N = 10  N = 14     Mean bipolar voltage (mV)  1.5±0.8  1.7±0.4  1.3±0.6  1.7±1.0  0.552   Total activation time (ms)  192±81  182±58  188±70  197±78  0.635   Total area (cm2)  435±130  402±132  492±169  404±88  0.234   Bipolar scar (cm2)  41±34  45±20  91±64  116±106  0.157   Bipolar scar (%)  14±12  12±8  28±10  45±23  <0.001   Late potentials area (cm2)  39±26  32±20  26±19  49±28  0.127   Late potentials area (%)  10±6  8±7  8±6  12±7  0.296   Arrhythmogenic potentials area (cm2)  37±20  39±8  31±6  78±22  0.006   Arrhythmogenic potentials area (%)  14±5  10±2  9±2  20±5  0.002  Epi-endo scar gradient (%)a  16±11  7±3  15±4  28±6  <0.001  a Epicardial scar (%) was measured by bipolar signal amplitude of <1.0 mV from the epicardium or unipolar signal amplitude of < 5.5 mV from the endocardium. LVZ, low-voltage zone; RV, right ventricle; VF, ventricular fibrillation; VT, ventricular tachycardia. Figure 2 View largeDownload slide An ARVC patient with a transmural scar (Group 1). A: (Left panel) The percentage of the scar area was 15.0% in the endocardium and 15.0% in the epicardium, respectively (Group 1). (Center panel) The latest RV activation was seen overlying the inferior right ventricular free wall on the endocardium and epicardium. (Right panel) The SAFE-T map showing the arrhythmogenic potentials located at the same location as the latest right ventricular activation in the inferior right ventricular free wall. 1B: Multiple sustained VTs other than VF were induced in the EP study. ARVC, arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle; SAFE-T, Simultaneous Amplitude Frequency Electrogram Transformation; VF, ventricular fibrillation. Figure 2 View largeDownload slide An ARVC patient with a transmural scar (Group 1). A: (Left panel) The percentage of the scar area was 15.0% in the endocardium and 15.0% in the epicardium, respectively (Group 1). (Center panel) The latest RV activation was seen overlying the inferior right ventricular free wall on the endocardium and epicardium. (Right panel) The SAFE-T map showing the arrhythmogenic potentials located at the same location as the latest right ventricular activation in the inferior right ventricular free wall. 1B: Multiple sustained VTs other than VF were induced in the EP study. ARVC, arrhythmogenic right ventricular cardiomyopathy; RV, right ventricle; SAFE-T, Simultaneous Amplitude Frequency Electrogram Transformation; VF, ventricular fibrillation. Figure 3 View largeDownload slide An ARVC patient with a horizontal scar presenting with VF. (A): A time-frequency analysis of the bipolar electrograms with or without high frequency components in normal or low voltage zones (numbers 1, 2, 3, 4 correspond to the locations in Figure 4B); (B): (Left panel) The scar in the endocardial and epicardial bipolar voltage maps was 4.3% and 65%, respectively (Group 3). (Center panel) The latest epicardial RV activation was seen overlying the RV posterior epicardium. (Right panel) A SAFE-T map demonstrating the distribution of the arrhythmogenic potentials, which contained the isthmus (dot line) for one stable VT in Figure 4C. (D): The documented ECG in the same patient exhibiting ventricular fibrillation. Abbreviation as in Figure 2. Figure 3 View largeDownload slide An ARVC patient with a horizontal scar presenting with VF. (A): A time-frequency analysis of the bipolar electrograms with or without high frequency components in normal or low voltage zones (numbers 1, 2, 3, 4 correspond to the locations in Figure 4B); (B): (Left panel) The scar in the endocardial and epicardial bipolar voltage maps was 4.3% and 65%, respectively (Group 3). (Center panel) The latest epicardial RV activation was seen overlying the RV posterior epicardium. (Right panel) A SAFE-T map demonstrating the distribution of the arrhythmogenic potentials, which contained the isthmus (dot line) for one stable VT in Figure 4C. (D): The documented ECG in the same patient exhibiting ventricular fibrillation. Abbreviation as in Figure 2. Among the 34 patients with available epicardial maps, there were no statistically significant differences in the epicardial mean unipolar/bipolar voltage or late potential area between three groups. Additionally, horizontal scar patients had more epicardial bipolar scar by percentage (12 ± 8 vs. 18 ± 10 vs. 45 ± 23%, P < 0.001) and arrhythmogenic potential areas (39 ± 8 vs. 31 ± 6 vs. 78 ± 22 cm2, P = 0.006). EP study, VT mapping, and catheter ablation Of the 80 ARVC patients, a total of 182 VT were induced, including 75 sustained VTs in 29 patients, 107 non-sustained VTs in 68 patients, and VF in 19 patients. Of the 75 sustained VTs, 22 (29.3%) were mappable VTs and 41 (70.7%) were unmapable. Activation mapping and entrainment were achieved in 22 stable VTs, including 6 (27.3%) VTs in Group 1, 5 (22.7%) VTs in Group 2, and 11 (50%) in Groups 3 (P = 0.255). The center of the isthmuses were located within dense scar (bipolar voltage < 0.5 mV) in 6 (27.3%), at border zones (bipolar voltage: 0.5–1.5 mV) in 16 (72.7%), in areas with LPs in 15 (68.1%), and in areas with arrhythmogenic potentials in 22 (100.0%). The exits of the isthmuses were identified within dense scar in 18 (81.8%), at border zones in 4 (18.2%), in areas with LPs in 2 (9.1%), and in areas with arrhythmogenic potentials in 22 (100.0%). A total of 72 conducting channels were identified by activation maps during sinus rhythm, including 23 (19.4%) in Group 1, 23 (30.6%) in Group 2, and 26 in Group 3 (P = 0.543). All the conducting channels were located within dense scar or at border zones of scar. Arrhythmogenic potentials could be identified in all conducting isthmuses and LPs could be identified in 52 (72.2%) conducting isthmuses. Additionally, 107 triggers were identified, including 34 (31.7%) in Group 1, 41 (38.3%) in Group 2, and 42 (39.3%) in Group 2 (P = 0.337). Of these triggers, 81 (75.7%%) were located within border zones, 26 (24.3%) originated from areas nearby border zones, 3 (2.8%) were within areas with LPs, and 72 (67.3%) were within areas with arrhythmogenic potentials. Acute procedural success was achieved in all patients without any inducible VF or sustained VTs after the catheter ablation under a uniform induction protocol with and without an intravenous infusion of 1–5 μg/min isoprenaline to achieve at least a 20% heart rate increment. Residual LPs or arrhythmogenic potentials were not eliminated in 12 patients due to a prolonged procedure time in 3 patients (25.0%), patient intolerance in 4 (33.3%), and nearby coronary arteries or phrenic nerves in 5 (41.7%). Two patients were complicated with pulmonary edema with prolonged mechanical ventilation after the procedure and were smoothly discharged after 30 and 27 days of hospitalization, respectively. One patient was complicated with a pseudo aneurysm and recovered after a prolonged compression. Univariate and multivariate logistic regression analyses for odds ratios of fatal VAs were performed in patients with an epicardial-endocardial approach (Table 3) and the whole study population (Table 4). After multivariate adjustment, the epicardial-endocardial scar gradient was independently associated with fatal VAs in ARVC patients. Table 3 Uni/multivariate analysis of the fatal VA incidence in ARVC patients with epi/endo mapping (34 patients)   Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.278  1.03 (0.98–1.09)       Gender  0.239  2.86 (0.50–16.43)       Diabetes mellitus  0.518  2.00 (0.24–16.44)       Hypertension  0.442  1.75 (0.42–7.28)      RV endocardium           Mean bipolar voltage (mV)  0.227  1.81 (0.69–4.76)       Mean unipolar voltage (mV)  0.867  1.05 (0.57–1.93)       Total activation time (ms)  0.226  1.01 (0.99–1.02)       Bipolar scar (%)  0.019  0.81 (0.68–0.97)  0.702  1.09 (0.71–1.66)   Late potentials area (%)  0.368  1.05 (0.95–1.15)       Arrhythmogenic potentials area (cm2)  0.850  1.02 (0.82–1.27)      RV epicardium           Mean bipolar voltage (mV)  0.346  1.60 (0.60–1.21)       Mean unipolar voltage (mV)  0.900  1.02 (0.78–1.33)       Total activation time (ms)  0.485  1.00 (0.99–1.01)       Bipolar scar (%)  0.028  1.05 (1.01–1.09)  0.630  0.91 (0.63–1.33)   Late potentials area (%)  0.018  1.22 (1.04–1.44)  0.510  0.83 (0.47–1.45)   Arrhythmogenic potentials area (%)  0.004  1.73 (1.19–2.52)  0.196  3.30 (0.54–1.15)  Substrate characteristics           Epi/Endo Scar Gradient  <0.001  1.21 (1.11–1.31)  0.035  1.67 (1.04–2.68)    Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.278  1.03 (0.98–1.09)       Gender  0.239  2.86 (0.50–16.43)       Diabetes mellitus  0.518  2.00 (0.24–16.44)       Hypertension  0.442  1.75 (0.42–7.28)      RV endocardium           Mean bipolar voltage (mV)  0.227  1.81 (0.69–4.76)       Mean unipolar voltage (mV)  0.867  1.05 (0.57–1.93)       Total activation time (ms)  0.226  1.01 (0.99–1.02)       Bipolar scar (%)  0.019  0.81 (0.68–0.97)  0.702  1.09 (0.71–1.66)   Late potentials area (%)  0.368  1.05 (0.95–1.15)       Arrhythmogenic potentials area (cm2)  0.850  1.02 (0.82–1.27)      RV epicardium           Mean bipolar voltage (mV)  0.346  1.60 (0.60–1.21)       Mean unipolar voltage (mV)  0.900  1.02 (0.78–1.33)       Total activation time (ms)  0.485  1.00 (0.99–1.01)       Bipolar scar (%)  0.028  1.05 (1.01–1.09)  0.630  0.91 (0.63–1.33)   Late potentials area (%)  0.018  1.22 (1.04–1.44)  0.510  0.83 (0.47–1.45)   Arrhythmogenic potentials area (%)  0.004  1.73 (1.19–2.52)  0.196  3.30 (0.54–1.15)  Substrate characteristics           Epi/Endo Scar Gradient  <0.001  1.21 (1.11–1.31)  0.035  1.67 (1.04–2.68)  ARVC, arrhythmogenic right ventricular dysplasia; CI, confidence interval; OR, odds ratio; RV, right ventricle; VA, ventricular arrhythmia. Table 4 Uni/multivariate analysis of the fatal VA incidence in whole study (80 patients)   Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.111  1.03 (0.99–1.07)       Gender  0.280  1.73 (0.64–4.70)       Diabetes mellitus  0.205  2.60 (0.59–11.41)       Hypertension  0.431  1.50 (0.55–4.12)      RV endocardium           Mean bipolar voltage (mV)  0.347  0.81 (0.53–1.25)       Mean unipolar voltage (mV)  0.730  1.06 (0.78–1.44)       Total activation time (ms)  0.112  1.01 (0.99–1.02)       Bipolar scar (%)  0.094  0.92 (0.84–1.01)       Unipolar low-voltage area (%)  <0.001  1.15 (1.06–1.24)  0.058  1.09 (0.99–1.19)   Late potentials area (%)  0.128  1.08 (0.99–1.20)       Arrhythmogenic potentials area (cm2)  0.263  1.15 (0.90–1.48)      Substrate characteristics           Epi/endo scar gradienta  <0.001  1.21 (1.11–1.31)  0.001  5.47 (2.00–14.97)    Univariate analysis   Multivariate analysis     P value  OR (95%CI)  P value  OR (95%CI)  Baseline characteristics           Age  0.111  1.03 (0.99–1.07)       Gender  0.280  1.73 (0.64–4.70)       Diabetes mellitus  0.205  2.60 (0.59–11.41)       Hypertension  0.431  1.50 (0.55–4.12)      RV endocardium           Mean bipolar voltage (mV)  0.347  0.81 (0.53–1.25)       Mean unipolar voltage (mV)  0.730  1.06 (0.78–1.44)       Total activation time (ms)  0.112  1.01 (0.99–1.02)       Bipolar scar (%)  0.094  0.92 (0.84–1.01)       Unipolar low-voltage area (%)  <0.001  1.15 (1.06–1.24)  0.058  1.09 (0.99–1.19)   Late potentials area (%)  0.128  1.08 (0.99–1.20)       Arrhythmogenic potentials area (cm2)  0.263  1.15 (0.90–1.48)      Substrate characteristics           Epi/endo scar gradienta  <0.001  1.21 (1.11–1.31)  0.001  5.47 (2.00–14.97)  a Epicardial scar (%) was measured by a bipolar signal amplitude of <1.0 mV from the epicardium or unipolar signal amplitude of < 5.5 mV from the endocardium. ARVC, arrhythmogenic right ventricular dysplasia; CI, confidence interval; OR, odds ratio; RV, right ventricle; VA, ventricular arrhythmia. The recurrences of VAs after catheter ablation between the different groups during the follow-up During a mean follow-up period of 38 ± 11 months (9–93 months), a total of 39 patients (48.8%) developed recurrences with a mean duration of 35 ± 19 months after an acute successful ablation, including 10 patients (12.5%) with recurrences of VF/ICD shocks with a mean duration of 28 ± 13 months, 18 (22.5%) with recurrences of sustained VT/ICD antitachycardia pacing therapy with a mean duration of 25 ± 13 months, 13 (16.3%) with recurrences of PVCs > 5000 beats per day with a mean duration of 26 ± 17 months, 29 (36.3%) with recurrences of PVCs of >1000 beats per day with a mean duration of 21 ± 16 months, and 23 (%) that received repeated procedures at a mean duration 29 ± 19 months. Among the 23 patients with redo-procedures, 5 (21.7%) developed recurrence of ventricular arrhythmia in the mean follow-up 27 ± 15 months (3 non-sustained VT 2 PVCs > 5000/day, and 3 received third ablation). Two patients (2.5%) died of non-cardiac diseases (pneumonia). Of the 36 patients with PVC/VT recurrences, recurrent VAs with the same or similar ECG morphology as documented before were detected in 32 (88.9%) patients and different morphologies in 4 (11.1%). A Kaplan–Meier curve analysis demonstrated higher recurrences of VF and VT in the horizontal scar group than the other groups (Log-rank P < 0.001, Figure 4). Figure 4 View largeDownload slide Kaplan–Meier survival Curve according to the Scar Distribution Patterns. A–F) Kaplan–Meier curve of the re-ablation, total recurrence, 1000 PVC beats/day recurrence, 5000 PVC beats/day recurrence, VF recurrence, and VF recurrence free survival curve in the different scar distribution patterns. PVC, premature ventricular complex; VT, ventricular tachycardia; VF, ventricular fibrillation. Group 1: Epi-Endo <10%; Group 2: Epi-Endo >10% and < 20%; Group 3: Epi-Endo >20%. Figure 4 View largeDownload slide Kaplan–Meier survival Curve according to the Scar Distribution Patterns. A–F) Kaplan–Meier curve of the re-ablation, total recurrence, 1000 PVC beats/day recurrence, 5000 PVC beats/day recurrence, VF recurrence, and VF recurrence free survival curve in the different scar distribution patterns. PVC, premature ventricular complex; VT, ventricular tachycardia; VF, ventricular fibrillation. Group 1: Epi-Endo <10%; Group 2: Epi-Endo >10% and < 20%; Group 3: Epi-Endo >20%. Discussion Main findings First, this study showed that ARVC patients with transmural scars (epicardial-endocardial scar gradient < 10%, extending vertically) had a greater number of clinical and EP lab induced occurrences of PVCs and fewer VF episodes. In contrast, ARVC patients with horizontal scar (Group 3) had fewer clinical isolated PVCs, but a higher incidence of VF episodes. Second, the scar discrepancy was independently correlated with the occurrence of VF. Third, the total, VT and VF recurrences were higher in the ARVC patients with a horizontal scar than in those with transmural scar. Last, the arrhythmogenic potentials were more prominent in the patients with a horizontal scar distribution. ARVC and fatal VAs The present study describes the importance of scar discrepancy in the clinical presentation of fatal VAs and the inducibility of VF in the EP lab. This novel finding might help clinical electrophysiologists predict VF occurrences and develop decision strategies for secondary or primary prevention. In a previous study, patients meeting the criteria for ICD implantations had a higher rate of rapid and potentially life-threatening arrhythmias.3 Environmental factors including higher temperatures, especially in the summer, and larger variations in the humidity within 3 days of the event, were independently associated with the development of certain events.10 In another study involving 136 ARVC patients who received ICD implantations and were followed up for 39 ± 25 months, SCD or hemodynamic unstable VAs and left ventricular involvement were independent predictors of VF/flutter.18 Generally, the acute success of catheter ablation was feasible and obtained in 71–100% of patients with ARVC19 with various ablation endpoints. Acute successful ablation was achieved in our current study population without inducible sustained VT or VF under isoprenaline after catheter ablation. In patients with frequent inducible unstable VT or VF, general anesthesia was applied for requirement of repeated direct current shock. Post-ablation inducibility was performed under sedative status without interruption in these patients. The sedative medication might bias the result of the final inducibility, which was higher than previous reports. Horizontal distribution scar and fatal VAs Data from observational clinical studies on ARVC have provided a number of clinical predictors, including VT/VF episodes and unexplained syncope, of adverse events and death. Other independent risk factors for arrhythmogenic events include dysfunction of the RV, left ventricle, or both, a male gender, gene mutations, inducibility, and the amount of electroanatomical scar.19 Currently, there is no consensus on the association of the development of fatal VAs in ARVC patients. In our study, we identified horizontal scar as an independent predictor of the clinical presentation of VF. The epicardial arrhythmogenic potential area is more prominent in patients with a horizontal scar. Local horizontal progression of scar might lead to aberrations in the action potentials and development of an ionic gradient between the layers with a consequent development of slow conduction. This would result in the formation of arrhythmogenic potentials in the epicardial area, which becomes a substrate for reentry in phase 2 and gives rise to polymorphic VT or VF.20 In the substrate mapping, the number of identified conducting channels did not significantly differ. However, the arrhythmogenic potentials might suggest functional intramural conducting channels, which were masked in the electroanatomic mapping. Intramural unstable scar might play an important role in the presentation of fatal VAs. Intramural channels for reentry have been previously demonstrated during VT or VF.21 However, whether stable intramural reentry occurs during VF and to what extent it is required to maintain VF are unanswered questions. It is possible that an intramural disease substrate providing functional channels could serve as a source of rapid activation during VF. These channels were usually spatiotemporally unstable and could not maintain monomorphic VT.21 The substrate characteristics of our present study revealed that scar with a horizontal distribution was a more important factor of the presentation of fatal VAs and inducibility of VF in EP studies. Patients in Group 3 might presented with more advanced stage of ARVC, which was associated with higher incidence of fatal arrhythmia. However, current literature suggests that advanced stage of ARVC was associated with progressive RV dilatation and limited scar progression.9 The similar RV volume in three groups didn’t support the difference in stage of ARVC. (Table 2) The fatal VAs were associated with electric instability in ARVC patients. Our present study showed that the RV scar endo-epicardial gradient in substrate maps might play a substantial role in the clinical VA characteristics and recurrence after catheter ablation. The cardiac magnetic resonance image might provide useful information in ARVC diagnosis but have limited value in the evaluation of the RV scar endo-epicardial gradient and in the risk stratification for VA recurrence before the ablation procedure. The correlation of the RV endocardial unipolar/epicardial bipolar scar The association between RV endocardial unipolar voltage and RV epicardial scar was firstly report by Hutchinson MD et al.14 In our retrospective study, epicardial mapping and ablation was only performed in 42.5% (34 of 80) patients refractory to endocardial ablation. Statistical analysis for correlation and agreements showed high correlation between RV epicardial bipolar scar and RV endocardial unipolar voltage in these 34 patients. Based on the statistical results, we could predict the epi-endocardial scar discrepancy from RV endocardial substrate. Further prospective trials with both epicardial/endocardial mapping are needed to prove the hypotheses that this finding is clinically useful, reproducible, and generalizable to all ARVC patients. Study limitations First, only patients with a high arrhythomogenic burden and who were refractory to antiarrhythmic medications were included in the substrate analysis. The data may not be extrapolated to all patients with ARVC patients. Second, epicardial mapping was performed only in patients refractory to endocardial ablation because of the ablation protocol in our hospital. Third, epicardial scar predicted by endocardial unipolar mapping remains an indirect predictor. Further prospective trials with both epicardial/endocardial mapping are required. Conclusions In ARVC, the epicardial substrate that extended in the horizontal plane rather than transmural plane provided the arrhythmogenic substrate for confined epicardial VT circuits. A higher epicardial arrhythmogenic potential density promoted arrhythmogenicity and was more likely to predispose to electrical instability during the first presentation despite a low PVC load. The heterogeneous distribution of scar between the epicardium and endocardium seems to play a role in the arrhythmogenicity of ARVC. Acknowledgements This work was supported by Ministry of Science and Technology of Taiwan for National Yang-Ming University, Taipei Veterans General Hospital (MOST102-2314-B-010-056-MY2, 103-2314-B-010-048-MY3, 104-2314-B-010 -055-MY3); Grant of Taipei Veterans General Hospital (V104E7-001, V105C-122); Clinical trial in Taipei Veterans General Hospital (C13-092) supported by Biosense Webster (IIS 290); Taipei Veterans General Hospital-National Yang-Ming University Excellent Physicians Scientists Cultivation Program (105-V-B-021). Conflicts of interest: none declared. References 1 Haqqani HM, Tschabrunn CM, Betensky BP, Lavi N, Tzou WS, Zado ES et al.   Layered activation of epicardial scar in arrhythmogenic right ventricular dysplasia: possible substrate for confined epicardial circuits. Circ Arrhythm Electrophysiol  2012; 5: 796– 803. Google Scholar CrossRef Search ADS PubMed  2 Basso C, Bauce B, Corrado D, Thiene G. Pathophysiology of arrhythmogenic cardiomyopathy. Nat Rev Cardiol  2012; 9: 223– 33. Google Scholar CrossRef Search ADS   3 Li CH, Lin YJ, Huang JL, Wu TJ, Cheng CC, Lin WS et al.   Long-term follow-up in patients with arrhythmogenic right ventricular cardiomyopathy. J Cardiovasc Electrophysiol  2012; 23: 750– 6. Google Scholar CrossRef Search ADS PubMed  4 Basso C, Corrado D, Marcus FI, Nava A, Thiene G. Arrhythmogenic right ventricular cardiomyopathy. Lancet  2009; 373: 1289– 300. Google Scholar CrossRef Search ADS PubMed  5 Corrado D, Basso C, Thiene G, McKenna WJ, Davies MJ, Fontaliran F et al.   Spectrum of clinicopathologic manifestations of arrhythmogenic right ventricular cardiomyopathy/dysplasia: a multicenter study. J Am Coll Cardiol  1997; 30: 1512– 20. Google Scholar CrossRef Search ADS PubMed  6 Basso C, Thiene G, Corrado D, Angelini A, Nava A, Valente M. Arrhythmogenic right ventricular cardiomyopathy. Dysplasia, dystrophy, or myocarditis? Circulation  1996; 94: 983– 91. Google Scholar CrossRef Search ADS PubMed  7 Garcia FC, Bazan V, Zado ES, Ren JF, Marchlinski FE. Epicardial substrate and outcome with epicardial ablation of ventricular tachycardia in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Circulation  2009; 120: 366– 75. Google Scholar CrossRef Search ADS PubMed  8 Berruezo A, Acosta J, Fernández-Armenta J, Pedrote A, Barrera A, Arana-Rueda E et al.   Safety, long-term outcomes and predictors of recurrence after first-line combined endoepicardial ventricular tachycardia substrate ablation in arrhythmogenic cardiomyopathy. Impact of arrhythmic substrate distribution pattern. A prospective multicentre study. Europace  2017; 19: 607– 16. Google Scholar PubMed  9 Riley MP, Zado E, Bala R, Cooper J, Dixit S, Garcia F et al.   Lack of uniform progression of endocardial scar in patients with arrhythmogenic right ventricular dysplasia/cardiomyopathy and ventricular tachycardia. Cir Arrhythm Electrophysiol  2010; 3: 332– 8. Google Scholar CrossRef Search ADS   10 Marcus FI, McKenna WJ, Sherrill D, Basso C, Bauce B, Bluemke DA et al.   Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the Task Force Criteria. Eur Heart J  2010; 31: 806– 14. Google Scholar CrossRef Search ADS PubMed  11 Chung FP, Li HR, Chong E, Pan CH, Lin YJ, Chang SL et al.   Seasonal variation in the frequency of sudden cardiac death and ventricular tachyarrhythmia in patients with arrhythmogenic right ventricular dysplasia/cardiomyopathy: the effect of meteorological factors. Heart Rhythm  2013; 10: 1859– 66. Google Scholar CrossRef Search ADS PubMed  12 Marchlinski FE, Callans DJ, Gottlieb CD, Zado E. Linear ablation lesions for control of unmappable ventricular tachycardia in patients with ischemic and nonischemic cardiomyopathy. Circulation  2000; 101: 1288– 96. Google Scholar CrossRef Search ADS PubMed  13 Polin GM, Haqqani H, Tzou W, Hutchinson MD, Garcia FC, Callans DJ et al.   Endocardial unipolar voltage mapping to identify epicardial substrate in arrhythmogenic right ventricular cardiomyopathy/dysplasia. Heart Rhythm  2011; 8: 76– 83. Google Scholar CrossRef Search ADS PubMed  14 Sosa E, Scanavacca M, d'Avila A, Pilleggi F. A new technique to perform epicardial mapping in the electrophysiology laboratory. J Cardiovasc Electrophysiol  1996; 7: 531– 6. Google Scholar CrossRef Search ADS PubMed  15 Hutchinson MD, Gerstenfeld EP, Desjardins B, Bala R, Riley MP, Garcia FC et al.   Endocardial unipolar voltage mapping to detect epicardial ventricular tachycardia substrate in patients with nonischemic left ventricular cardiomyopathy. Cir Arrhythm Electrophysiol.  2011; 4: 49– 55. Google Scholar CrossRef Search ADS   16 Lin CY, Silberbauer J, Lin YJ, Lo MT, Lin C, Chang HC et al.   Simultaneous amplitude frequency electrogram transformation (SAFE-T) mapping to identify ventricular tachycardia arrhythmogenic potentials in sinus rhythm. JACC: Clin Electrophysiol  2016; 2: 459– 70. Google Scholar CrossRef Search ADS   17 Thiene G, Nava A, Corrado D, Rossi L, Pennelli N. Right ventricular cardiomyopathy and sudden death in young people. N Engl J Med  1988; 318: 129– 33. Google Scholar CrossRef Search ADS PubMed  18 Corrado D, Leoni L, Link MS, Della Bella P, Gaita F, Curnis A et al.   Implantable cardioverter-defibrillator therapy for prevention of sudden death in patients with arrhythmogenic right ventricular cardiomyopathy/dysplasia. Circulation  2003; 108: 3084– 91. Google Scholar CrossRef Search ADS PubMed  19 Corrado D, Wichter T, Link MS, Hauer RN, Marchlinski FE, Anastasakis A et al.   Treatment of arrhythmogenic right ventricular cardiomyopathy/dysplasia: an international task force consensus statement. Circulation  2015; 132: 441– 53. Google Scholar CrossRef Search ADS PubMed  20 Gussak I, Antzelevitch C, Bjerregaard P, Towbin JA, Chaitman BR. The Brugada syndrome: clinical, electrophysiologic and genetic aspects. J Am Coll Cardiol  1999; 33: 5– 15. Google Scholar CrossRef Search ADS PubMed  21 Valderrabano M, Lee MH, Ohara T, Lai AC, Fishbein MC, Lin SF et al.   Dynamics of intramural and transmural reentry during ventricular fibrillation in isolated swine ventricles. Circ Res  2001; 88: 839– 48. Google Scholar CrossRef Search ADS PubMed  Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions, please email: journals.permissions@oup.com.

Journal

EuropaceOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off