Successful Substrate Ablation for Ventricular Fibrillation in a Patient With Nonischemic Cardiomyopathy

Abstract

A 58-year-old woman with nonischemic cardiomyopathy underwent an ablation for refractory ventricular fibrillation (VF). In addition to elimination of the premature ventricular contraction triggering VF, substrate ablation entailed elimination of dormant Purkinje potentials, which were unmasked in regions where local pacing revealed a narrow QRS with stimulus QRS latency. VF trigger and abnormal Purkinje ablation completely eliminated the refractory VF. (Level of Difficulty: Intermediate.)

Central Illustration

Case Presentation

A 58-year-old woman with nonischemic cardiomyopathy (NICM) was referred to our hospital with multiple implantable cardioverter-defibrillator (ICD) shocks. The ICD record revealed appropriate shocks for ventricular fibrillation (VF) storm.

Learning Objectives

• •To recognize the efficacy of substrate ablation targeting an abnormal Purkinje network when premature ventricular contraction ablation is not sufficient for VF elimination in nonischemic cardiomyopathy patients.
• •To emphasize the importance of local pacing in unmasking the distribution of abnormal Purkinje network hidden inside the scar.

Past Medical History

Diagnosis of NICM was made in another institution 24 years earlier, and an ICD was implanted 16 years earlier owing to nonsustained ventricular tachycardia (VT) and low left ventricular ejection fraction (LVEF). Before ICD implantation, reassessment of her cardiac dysfunction was performed. No evidence of myocardial infarction was identified in either electrocardiography or echocardiography. Echocardiography revealed diffuse left ventricular hypokinesis (LVEF: 21%) with a dilated ventricle (LV end-diastolic dimension: 62 mm). Neither stenosis nor obstructions were observed with the use of coronary angiography, and acetylcholine provocation test was negative. Blood testing did not show any specific findings suggesting cardiomyopathies such as sarcoidosis or amyloidosis. Biopsy from the ventricular septum showed mild to moderate fibrosis without diagnostic findings for specific cardiomyopathies. Based on these findings, the patient was diagnosed with NICM.

Differential Diagnosis

The differential diagnosis for underlying structural heart disease involved myocardial infarction with nonobstructive coronary arteries caused by myocarditis or unknown etiology, and cardiac sarcoidosis, which was not confirmed with endomyocardial biopsy.

Investigations

In addition to the regular occurrence of 1–2 ICD shocks per month, the patient recently suffered from 5 repetitive ICD shocks in 24 h. No specific changes were observed in the electrocardiogram, echocardiogram (Figures 1A and 1B) (LVEF = 17%), or blood test (electrolyte status). Neither stenosis nor coronary artery obstructions were observed in the coronary angiogram. magnetic resonance imaging (MRI) for the assessment of arrhythmia substrate was unfortunately not performed because the ICD system was MRI noncompatible.

Figure 1.

Preprocedural Electrocardiogram and Echocardiogram

(A) 12-lead electrocardiogram showed a normal QRS interval (120 ms) without any ischemic change. (B) Echocardiography showed dilation of the left ventricle (end-diastolic dimension 67 mm and end-systolic dimension 62 mm) and reduced ejection fraction (17%). Thickness of the interventricular septum (9 mm) and the posterior wall (10 mm) was relatively preserved.

Management

An emergency ablation procedure for VF storm was performed to mitigate the condition. Because the ICD record suggested premature ventricular contraction (PVC) triggering VF, the clinical PVC was initially targeted.

Voltage mapping with use of the Pentaray catheter and Carto system (Biosense-Webster) was performed in sinus rhythm, revealing a scar extending from the basal to apical septum, where unipolar low-voltage area (<8.3 mV) was smaller than the bipolar low-voltage area (<1.5 mV) (Figures 2A and 2B).

Figure 2.

Voltage Maps and Intracardiac Electrograms During the Initial Procedure

Both bipolar and unipolar voltage mapping generally demonstrated a low-voltage area in the basal to apical septum (A), but a relatively preserved area at the mid-septum according to the unipolar voltage map (B). (C) The intracardiac electrogram of VF initiation showing the Purkinje potential preceding QRS onset with latency at the mid-septum, where a good pacemap was obtained (D). ABL = ablation; CS = coronary sinus; dis = distal; prox = proximal; PVC = premature ventricular contraction; RAO = right anterior oblique; RV = right ventricle; VF = ventricular fibrillation.

VF was easily induced during procedure, and intracardiac electrogram (EGM) of VF initiation on the mid-septum showed the Purkinje potential preceding QRS onset with latency (Figure 2C), where a good pace map was also observed (Figure 2D). Ablation on this site significantly reduced the VF, and additional applications targeting Purkinje potentials around this site eliminated the storm.

Although repetitive ICD shocks were not observed after the first procedure, a few VF episodes triggered by different PVCs were still observed within 2 weeks after catheter ablation. Therefore, a second catheter ablation was performed. Although mechanical PVCs triggering VF were observed, spontaneous PVCs initiating VF were not identified in the second procedure. Therefore, the septal Purkinje network was targeted for substrate ablation. Although typical Purkinje potentials preceding the QRS onset were not observed, abnormal fractionated EGMs hidden inside the QRS were instead distributed in the septum (Figures 3A and 3B). Interestingly, pacing on these sites resulted in several types of narrow QRS with a stimulus QRS latency (Figure 3B), suggesting that a conduction system and/or Purkinje network were exclusively captured, but with a conduction disturbance due to the original damaged septal scar and/or the prior ablation.

Figure 3.

12-Lead and Intracardiac Electrograms in SR and Local Pacing

(A) In the basal septum, low-voltage fractionated signals were recorded during SR. Pacing at this site resulted in a stimulus QRS latency and narrow QRS. (B) 12-lead electrocardiograms of pacing and local electrograms during SR in the septum. Local pacing demonstrates multiple narrow QRS morphologies with a stimulus QRS latency. EGM = electrogram; SR = sinus rhythm; other abbreviations as in Figure 2.

After unmasking these substrates with the use of pacing, these abnormal EGMs with stimulus QRS latency were entirely eliminated by means of additional ablation (30 to 50 W, 31 min). Loss of local capture was confirmed in each ablation site. Elimination of the abnormal Purkinje substrate in the septum resulted in a complete left bundle branch block. Because ICD battery exhaustion due to the repetitive ICD shocks was observed in this low-LVEF patient, the ICD was upgraded to a cardiac resynchronization therapy defibrillator during hospitalization. After 1-year follow-up, VF has never recurred and the patient has not developed heart failure.

Discussion

This case provides 2 key messages: 1) Purkinje fibers might serve not only as a trigger but also as a substrate for VF in NICM; and 2) local pacing targeting narrow QRS and long stimulus QRS latency is an efficient means to unmask abnormal Purkinje potentials (substrate of VF).

Berenfeld and Jalife demonstrated in a computerized 3-dimensional model that: 1) focal activity may originate at the Purkinje-muscle junctions; 2) Purkinje-muscle reentry is the mechanism of VF initiation; and 3) the Purkinje system becomes irrelevant after intramyocardial reentry is established (1). Despite their report, the effect of substrate ablation for VF has not been systematically studied, although substrate ablation for VT has already been well established. One report retrospectively demonstrated the effect of both triggering PVC ablation and substrate ablation in recurrent polymorphic VT or VF with structural heart disease, but with a limited number of patients (2).

In the present case, we clearly demonstrate that targeting the PVC triggering VF was not sufficient to eliminate VF, but eliminating the abnormal Purkinje network finally removed the refractory VF in NICM. The initial ablation session targeting the PVC triggering VF was moderately effective to mitigate the VF storm, but did not completely eliminate VF. Furthermore, substrate ablation targeting EGMs with stimulus QRS latency and narrow QRS, suggesting abnormal Purkinje potentials, finally eliminated the VF. These findings suggest that VF was possibly eliminated by ablating not only the focal activity site of the Purkinje-muscle junction (first ablation session), but also the Purkinje-muscle reentry sites, unmasked by means of local pacing (second ablation session).

Imaging is a useful tool in detecting the arrhythmia substrate in advance of the procedure. Both contrast-enhanced cardiac MRI and computed tomography are known to depict ventricular arrhythmia substrates (3,4). However, neither imaging was obtained just before the procedure owing to the MRI noncompatibility of the ICD system and unavailability of software such as the Music system (4).

Typically, Purkinje EGMs are observed just before QRS onset. However, in the present study, instead of typical Purkinje EGMs, only discrete potentials inside the QRS were observed in the septum during sinus rhythm. Local pacing was required to “unmask” the Purkinje with the combination of narrow QRS and latency. We theorize that this may be because antegrade activation to this site through the proximal Purkinje fibers is blocked and the site can be activated only retrogradely over the distal Purkinje fibers (Figure 4A). The stimulus QRS latency observed during local pacing may be because of conduction disturbance due to the original damaged scar and prior ablation on the septum (Figure 4B). In addition, this conduction disturbance in the Purkinje network might facilitate the perpetuation of reentry, resulting in refractory VF. This case indicates that local pacing may be useful to unmask the distribution of Purkinje network hidden inside the scar.

Figure 4.

Schematic Representation During Sinus Rhythm and Local Pacing

Local abnormal fractionated signals (probably damaged Purkinje-related electrograms) during sinus rhythm and narrow QRS with a stimulus QRS latency by means of locally pacing this signal may be explained by the following mechanism: (A) Antegrade activation to this site (B) through the proximal Purkinje fibers may be blocked (red line) and this site is retrogradely activated (blue dotted arrow) over the distal Purkinje fibers. (B) During local pacing (red circle), the ventricle is retrogradely activated (i) through the distal Purkinje system (blue dotted arrows), or (ii) through both the proximal and the distal Purkinje system. LAF = left anterior fascicule; LPF = left posterior fascicule; RBB = right bundle branch.

Follow-Up

The patient underwent ICD check 3, 6, and 12 months after ablation and showed no ventricular arrhythmia events.

Conclusions

When PVC ablation is not sufficient for VF elimination, substrate ablation targeting an abnormal Purkinje network may be an option in a patient with NICM. Local pacing might be useful to unmask the distribution of an abnormal Purkinje network hidden inside the scar.

Funding Support and Author Disclosures

This work was supported by Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research no. JP20K17074. Dr Takahashi has received honoraria from Johnson & Johnson. Dr Takahashi and Takigawa received endowments from Medtronic Japan, Boston Scientific, Japan Lifeline, and WIN International. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.