Dr. David J. Callans, University of Pennsylvania School of Medicine, discussed the uses and techniques of epicardial ablation. Indications for epicardial ablation include Chagas disease, VT associated with dilated cardiomyopathy (DCM), post-MI (rarely), cardiac sarcoid, myocarditis, right ventricular (RV) cardiomyopathy, left ventricular (LV) non-compaction, and some forms of idiopathic VT.

Electrocardiographic (ECG) clues to the epicardial location of VT include appearance of a pseudo-delta wave or a precordial maximum deflection index ≥0.55.  Callans has focused on the unexpected presence of Q waves in areas of the heart that are subtended by the ECG lead. If there is an epicardial focus on the basal inferior wall, a QS complex will be recorded in the inferior leads. In patients with VT just anterior to the aorta ECG shows a pattern break where V1 is positive, V2 is initially positive but has a deep S wave at the end of the QRS, and V3 is positive. This suggests that the VT is just underneath lead V2, which is anterior to the aorta.

Figure 1. Manifest entrainment at earliest endocardial site 【Click to enlarge】

Figure 2. Imaging clues to epicardial location 【Click to enlarge】

Array mapping can also indicate VT located in the epicardium. This technique is important for locating VT for epicardial ablation in patients with DCM and RV cardiomyopathy. Additionally, when pacing is done at the site of earliest activation, manifest entrainment is seen, most easily noted in lead V2, suggesting an epicardial site (Figure 1).

MRI with delayed enhancement or intercardiac echocardiography can provide clues to epicardial location. In a patient with sarcoid and uniform VT, the epicardial surface looked normal but the subepicardium had a dense hyperechoic region that correlated with the scar observed on voltage mapping (Figure 2). Epicardial ablation was performed successfully in this patient.

The Cleveland Clinic did a study of epicardial ablation in 20 patients with idiopathic VT who failed endocardial ablation. The VT was mapped to epicardial sites in 9 of the patients and epicardial ablation was successful in 8 of these.

Unique problems associated with epicardial ablation include the need for general anesthesia and a potential for coronary artery occlusion, phrenic nerve damage, and persistent pericardial bleeding. Additionally, insulating epicardial fat can cause false low voltage areas on mapping and cause radio frequency ablation to fail. Both problems are ameliorated with cold ablation.

Dr. Callans concluded that epicardial origin of VT is a possible cause of failed endocardial ablation. The epicardial approach may be particularly important in idiopathic VT, VT in DCM, and VT in RV cardiomyopathy.

Successful ablation of ventricular tachycardia (VT) in patients with arrhythmogenic right ventricular cardiomyopathy (ARVC) using substrate-based mapping as a guide has been reported; however, the endpoint for ablation has not been determined. The aim of this study, presented by Dr. Akihiko Nogami, Yokohama Rosai Hospital, was to assess the usefulness of a change in the isolated delayed component (IDC) as an endpoint of catheter ablation.

An IDC was defined as a distinct ventricular electrogram after the QRS end separated by ≥40 msec. The study included 18 patients with ARVC. The endpoints were disappearance of all IDCs and VT non-inducibility.

Figure 1. Before and After Ablation #1 in Patient #16 【Click to enlarge】

One patient was a 56 year old male with early VT recurrences after ablations. Although the VT became non-inducible after previous ablations, IDCs were still present. Radio frequency (RF) ablation was applied to the last successful ablation site, eliminating the IDC that had occurred during sinus rhythm (Figure 1). Additional IDCs were detected and eliminated with a total of 12 RF applications.

IDCs were recorded in 16 patients. The mean interval from QRS to the latest IDC was 185 msec. The endpoint was achieved in 56% of patients. After ablation, the IDCs were dissociated in 1, eliminated in 5, exhibited second degree block in 1, significantly delayed in 3, and unchanged in 6 patients.

While VT was suppressed in all 10 patients with a changed IDC, it was suppressed in only 3 of the 8 patients with no or an unchanged IDC after ablation. During the mean follow-up of 53 months, 3 patients died but there were no sudden deaths. Sustained VT occurred in 6 patients.

Figure 2. Hazard Ratio and 95% Confidence Intervals for the Risk of VT Recurrence 【Click to enlarge】

There was no significant difference in the VT-free curves between the clinical VT non-inducible and inducible groups. There was a significantly lower probability of VT recurrence in the patients with changed IDC versus those with unchanged or no IDC (p<0.02). A change in IDC after ablation significantly reduced the risk of VT recurrence (p<0.05) (Figure 2). An IDC at the ablation site was the strongest predictor for no recurrence (p<0.02).

Dr. Nogami concluded that in patients with ARVC, IDCs occurring during sinus rhythm are related to clinical VT and can be a target for ablation. A change in the IDC can be a useful endpoint. If no IDC is recorded from the endocardium, epicardial mapping and ablation might be needed. Qualitative analysis of the serial signal-averaged ECGs might be useful for long-term follow-up.

Previous studies have demonstrated the effectiveness of targeting areas for radiofrequency catheter ablation (RFCA) in cases of unmappable ventricular tachycardia (VT) on the basis of voltage mapping using the electroanatomical mapping system. An abnormal area in left ventricular (LV) endocardial mapping is defined as <1.5 mV. In patients with old myocardial infarction (OMI), this mapped area usually is too large to identify the target sites for RFCA. Wilber et al reported successful RFCA based on targeting the exit site of the slow conduction channel at the endocardial scar border. However, Verma et al reported that 18% of successfully ablated sites in patients with OMI-VT were located within the scar.

Dr. Yukio Sekiguchi, University of Tsukuba, evaluated the feasibility and efficacy of targeting RFCA to the marked low voltage areas (LVA) around scars detected by high-resolution substrate mapping. Substrate mapping was performed on 31 patients with OMI-VT. Normal ventricular voltage was defined as >0.6 mV, LVA as >0.1-≤0.6 mV, and dense scars as ≤0.1 mV. The basic linear ablation design criteria were: Lesions must cross through the LVA, where a good pace map was obtained; Lesions must combine between scar and scar within the LVA; and Lesions must extend from the scar to the normal voltage area. RF was not applied to nonarrhythmogenic areas.

Figure 1. High resolution substrate mapping, with the white lines indicating where RF was targeted. 【Click to enlarge】

Figure 2. VT could not be induced after RFCA. 【Click to enlarge】

The average number of mapping points was about 300, compared with other studies, which have reported a mean number of 70 to 210 mapped points. High resolution substrate mapping in one patient showed that the LVA was limited and several dense scars and the VT circuit isthmus were identified (Figure 1). After RFCA in this patient, VT could not be induced (Figure 2).

The first ablation was successful in eliminating VT in 20 patients and failed in 9 patients. A second ablation was successful in 7 patients and not attempted in 2. Two patients died, one of sepsis and the other of a second MI. At 38 ± 18 months follow-up, 93% of the patients were free of VT after one or more ablations.

Data from this study showed that there is a higher prevalence of isolated potentials and longer S-QRS intervals within the LVA. Critical isthmuses of VT may exist within the LVA, where the electrogram amplitude is less than 0.6 mV. Dr. Sekiguchi concluded that catheter ablation targeting LVAs with an amplitude ≤0.6 mV appears to be effective for the treatment of OMI-VT.

Dr. Yoshinori Kobayashi, Tokai University, discussed the mechanisms and ablation of Purkinje fiber (PF)-related ventricular tachyarrhythmias (VT) associated with ischemic heart disease. After coronary occlusion, changes occur in the action potential of PFs, including decreased Vmax and action potential interlude. Both monomorphic VT and polymorphic VT/ventricular fibrillation (VF) in which PFs play an important role can occur after MI.

Dr. Kobayashi described a case of incessant polymorphic VT/VF triggered by premature ventricular contractions (PVC) originating in the surviving PFs. A 67 year-old male had VT/VF, which emerged in the subacute phase of anterior MI, requiring multiple DC shocks. ECG revealed that the VT/VF was induced by PVCs with the same QRS morphology of relatively narrow configurations. Rescue catheter ablation was successfully applied, targeting the PVCs that triggered the VT/VF.

Figure 1. ECG and electrophysiological data of the 7 patients. 【Click to enlarge】

Seven males with severe LV dysfunction with incessant polymorphic VT/VF were successfully treated with rescue catheter ablation. Four patients were in the acute or sub-acute phase of MI, while the remaining 3 had severe coronary heart disease without evidence of MI. Catheter ablation was successful in all 7 patients, resulting in noninducible VT in 2 patients and PVC elimination in 5 patients (Figure 1). Four patients had polymorphic PVCs and 3 had monomorphic PVCs. All except one PVC showed right bundle branch block (RBBB) morphology with either a superior or inferior axis and relatively narrow QRS. At the successful ablation sites, the Purkinje potential preceded the QRS complex by a mean 66 ms during PVCs.

Figure 2. Activation sequences during VT. 【Click to enlarge】

Four patients presented with intrafascicular reentrant tachycardia mimicking idiopathic LVT. Intrafascicular VT occurred in the acute phase of MI in 2 patients and in the remote phase in the other 2. The QRS morphology was RBBB and superior axis in all 4 cases. Three patients had both diastolic (Pd) and presystolic (Ps) Purkinje potentials (Figure 2). The activation sequence of the Pd was from base to apex and of the Ps was from apex to base. The VTs were ablated in all 4 patients by RF applied to the left ventricular posterior septum. After eliminating the VT, Pd appeared after the QRS complex in 2 patients.

The results of these studies suggest that the PF network might be related to the occurrence and perpetuation of arrhythmia at least in the initial phase of the tachycardia. Detailed evaluations using advanced three dimensional mapping systems might further the mechanistic and anatomical understanding of these tachyarrhythmias.