Ethanol Ablation of Refractory Premature Ventricular Complex Originating from a Left Ventricular Summit Communicating Vein after Radiofrequency Catheter Ablation Failed in a Dilated Cardiomyopathy Patient

Abstract

During follow-up of a 60-year-old patient with dilated cardiomyopathy, a Holter electrocardiogram revealed monomorphic premature ventricular complexes (PVCs) accounting for 21-30% of total beats. Oral beta-blockers led to no improvement in PVC burden. The first radiofrequency catheter ablation attempt identified the PVC arising from the left ventricle summit communicating vein (CV) but failed to eliminate the PVC's origin. The second ablation attempt with selective infusions of 100% ethanol into the summit CV resulted in immediate termination of PVCs. The post-ablation course was uneventful. Echocardiography showed an improved ejection fraction, and a repeated Holter electrocardiogram showed no recurrence of PVCs during follow-up.

Introduction

Radiofrequency catheter ablation (RFCA) has become the standard treatment strategy for drug-refractory ventricular arrhythmias (VAs). However, identifying and treating VAs with deep intramural or epicardial origins remain challenging.

Yamada et al. suggested that a specific myocardial area called the left ventricular (LV) summit is the most common origin of idiopathic epicardial VAs (1). The LV summit is the highest epicardial region bounded by the left coronary arteries and great cardiac vein (GCV)-anterior interventricular vein (AIV) junction (2). Despite its importance, the LV summit is inaccessible to endocardial approaches, and its proximity to coronary vessels and overlying epicardial fat tissue precludes epicardial approaches, leading to a significant RFCA failure rate (3).

Trans-coronary ethanol ablation has been reported as an alternative approach for treating RFCA-refractory summit VAs. However, its use is limited owing to potential severe complications, such as coronary arterial dissection, thrombosis, pericarditis, and uncontrolled destruction of surrounding tissues (4). In recent years, instead of coronary instrumentation, retrograde coronary venous ethanol ablation (RCVEA) has emerged as a safe and promising treatment strategy (5,6). In addition, among the communicating veins (CVs) between the GCV and small cardiac venous system, a distinct vessel passing between the pulmonary and aortic annuli and draining to the GCV-AIV junction, ideally located in close association with the LV summit and called the summit CV, has been proposed as a novel approach to summit VAs (7).

Case Report

A 60-year-old man with a history of hypertension, dyslipidemia, obesity, and obstructive sleep apnea presented to our cardiology department for the evaluation of a reduced ejection fraction (EF) coincidently found on echocardiography. He had exertional dyspnea but denied any palpitations, chest pain, or syncope episodes.

A physical examination and laboratory analysis, including thyroid function tests and brain natriuretic peptide, were unremarkable. An electrocardiogram (ECG) revealed ventricular bigeminy and a monomorphic premature ventricular complex (PVC) with an inferior axis, left bundle branch block pattern, and a precordial transition zone in lead V4, suggesting an origin at the base of the heart (Fig. 1A). Coronary angiography and computed tomography (CT) coronary angiography showed no stenosis, and myocardial scintigraphy revealed no ischemia. Repeated echocardiography showed LV global systolic dysfunction with an EF of 40% and LV end-diastolic/systolic diameter of 61 mm/49 mm without significant valvular dysfunction. Cardiac magnetic resonance imaging (CMRI) indicated septal and basal intramural late gadolinium enhancement (LGE) (Fig. 1B), and dilated cardiomyopathy (DCM) was diagnosed.

Figure 1.

A: 12-lead electrocardiogram on admission showed frequent monomorphic premature ventricular complexes with bigeminy, an inferior axis, a left bundle block pattern, and a precordial transition zone in lead V4. B: CMR indicated septal and basal intramural late gadolinium enhancement. C: Pre-ablation high-resolution contrast-enhanced computed tomography confirmed the position of the communicating vein (CV) branching from the great cardiac vein-anterior interventricular vein (GCV-AIV) junction. D: A coronary sinus (CS) venogram showed the communicating vein branching from the GCV-AIV junction. E: Through the lumen of the 6-Fr CS catheter, a 2-Fr microcatheter was introduced into the CV.

During follow-up, repeated 24-h Holter monitoring indicated monomorphic PVCs accounting for 21-30% of the total beats and occasional non-sustained ventricular tachycardia of 3-4 consecutive beats. An oral beta-blocker (bisoprolol) was administered, but no improvement was observed during subsequent monitoring.

Failed RFCA attempt

Initial RFCA was performed as the first-line treatment for drug-refractory PVC. A 6-Fr Electrode catheter (Epstar Fix CS LUMEN; Japan Lifeline, Tokyo, Japan) was advanced into the coronary sinus (CS) from the right internal jugular vein. A retrograde CS venogram showed multiple CS branches, including the summit CV, as previously detected on contrast-enhanced CT (Fig. 1C, D). Through the 6-Fr catheter previously inserted via the lumen, we introduced a 2.7-Fr microcatheter (Epstar Fix AIV; Japan Lifeline) into the summit CV to the pacemap and record electric signals (Fig. 1E). The subsequent activation map showed the earliest activation site recorded at the right ventricle outflow tract (RVOT) posteroseptum with a PaSo (CARTO 3 pace mapping software program) score (CARTO system; Biosense Webster, Diamond Bar, USA) of 0.944 (Fig. 2A).

Figure 2.

A: Activation and 3D mapping showed the earliest activation time of 25 ms at the right ventricular outflow tract posteroseptum. Several radiofrequency (RF) ablation procedures were performed but failed to eliminate the premature ventricular complexes (PVCs). B: Further mapping and generation of a pace map via the retrograde aortic approach at the right coronary cusp (RCC) yielded a high PaSo score and earliest activation time of 30 ms. Additional RF ablation was delivered at the RCC, resulting in the a temporary disappearance of the PVCs. C: PVCs recurred with a new QRS pattern, different from the initial PVC morphology but less frequently and with prolonged coupling intervals (436 to 450 ms). D: Activation mapping showed the earliest activation time of 32 ms and a perfect pace map from the microcatheter inserted in the communicating vein. E: Additional RF ablation from the LVOT using the microcatheter as a target failed to terminate the PVC.

The PVCs disappeared with radiofrequency ablation (5 times via the RVOT, ablation power: 30 W, duration: 60-90 s, mean contact force: 10-14 g) but soon recurred. A retrograde aortic approach was then performed, and a PaSo score of 0.969 was obtained. Ablation of the earliest site of the right coronary cusp (RCC; 3 times, ablation power: 30 W; duration: 60 s, mean contact force: 11 g) (Fig. 2B) resulted in the disappearance of the PVCs.

Shortly thereafter, a different form of PVC with distinct morphology appeared but had a less frequent and prolonged coupling interval (436-450 ms) (Fig. 2C), which suggests RFA-induced exit modification of an intramural origin of PVCs. Since the local ventricular activation time recorded at the summit CV preceded the onset of the QRS complex by 32 ms, the PaSo score was 0.979 (Fig. 2D). RF ablations from the LVOT were delivered again using the microcatheter tip as a landmark (Fig. 2E). However, the PVCs could not be completely terminated and gradually became more prominent (46.7% of total beats on Holter monitoring) 1 month after RFCA.

Successful ethanol ablation

RFCA was performed a second time. We introduced a 6-Fr and 2.7-Fr electrode catheter into the CS and summit CV, respectively, as in the first session. The earliest pre-QRS activation time of 38 ms was obtained using the microcatheter. A retrograde aortic approach was adopted again, and ablation of the earliest site of the left coronary cusp preceded the onset of the QRS complex by 20 ms (Fig. 3A). The PVCs disappeared but soon recurred.

Figure 3.

A: At the beginning of the second ablation procedure, several radiofrequency ablation attempts from the left coronary cusp targeting the microcatheter were performed but failed to terminate the premature ventricular complexes (PVCs). B: A 2.0×8-mm balloon was cannulated into the communicating vein (CV) and inflated to block retrograde flow from the CV. Injection of contrast confirmed the complete occlusion of the vein. C: Upper electrocardiogram (ECG): a cold saline infusion test was performed twice and resulted in temporarily slowing of the rate but not complete termination of PVCs. Lower ECG: After 1 cc of 100% ethanol was injected into the occluded CV from the tip of the balloon catheter, the PVCs were terminated.

We then performed ethanol ablation. The microcatheter was removed, and a 5-Fr angiographic catheter (JR Judkins Right; Terumo, Tokyo, Japan) was introduced into the GCV via a bidirectional guiding sheath (VIZIGO^Ⓡ; Biosense Webster) inserted from the right femoral vein. The summit CV diameter measured on the preprocedural CT images was 1.4 mm. Through the angiographic catheter, a 1.5×8-mm balloon (Emerge™, Boston Scientific, Marlborough, USA) was cannulated into the summit CV in an over-the-wire fashion through a 0.014″ guidewire (Cruise^Ⓡ; Asahi Intecc, Tokyo, Japan) and inflated. Since the 1.5×8-mm balloon failed to completely occlude the vein, we switched to a 2.0×8-mm balloon.

After retrograde contrast injection to confirm complete occlusion (Fig. 3B), the guidewire was removed through the balloon catheter lumen. Subsequently, a cold saline infusion test was performed twice, resulting in temporary slowing of the rate but not complete termination of PVCs. However, a subsequent infusion of 1 cc of 100% ethanol over 90 s successfully resulted in the immediate termination of t PVCs (Fig. 3C). We repeated the extra ethanol infusions twice with 1 cc each. A total dose of 3 cc was delivered without any acute complications. The procedure time was 181 min, the fluoroscopy time was 34.5 min, and 34 cc of contrast was used.

Post-ablation ultrasound showed no pericardial effusion. During three months of post-ablation follow-up, repeated 24-h Holter monitoring showed a sinus rhythm of 60-70/min and no recurrence of PVCs. Subsequent echocardiography showed an improved EF (53%) and an LV end-diastolic/systolic diameter of 59 mm/42 mm.

Discussion

A significant burden of PVCs worsens LV dysfunction due to irregular rhythm and uncoordinated contractility of the LV segments, which eventually leads to decompensated heart failure. Frequent PVCs of ＞24% may cause PVC-induced cardiomyopathy (8), an acquired, potentially reversible form of cardiomyopathy. Taieb et al. reported that eliminating frequent isolated PVC in DCM patients with RFCA can normalize the clinical status and LV function (9). In this case, DCM was diagnosed based on clinical presentation, echocardiographic findings, and the presence of LGE on CMR (10). Ethanol ablation of frequent refractory PVC improved the patient's cardiac function, and recurrence was not observed during a long follow-up period.

RCVEA should be considered when conventional endocardial RFCA has failed, a deep intramural origin (especially LV summit) is suspected, and an ideal earliest activation is obtained from within the targeted vein. Other potential options include sequential/simultaneous unipolar RFA from both the endocardial and epicardial sides, percutaneous epicardial ablation, bipolar RFA between two catheters from both the left and right sides of the heart, and RFA using half-normal saline-irrigated electrodes.

Of note, the efficacy and safety of some of these therapies have not been fully established, and their indications should be carefully evaluated. We decided to perform RCVEA because it is the least invasive and relatively simple procedure, and no serious complications have been reported. Nevertheless, several potential complications, including venous thrombosis and rupture, pericardial effusion, atrioventricular block, or damage to adjacent coronary arteries, especially proximal left coronary arteries, have been reported to occur when performing ethanol ablation in the LV summit area, as in this case.

The RCVEA procedure seems straightforward but poses some technical challenges. First, the anatomy of the coronary venous system is highly variable among individuals (11). In our case, a relatively large and solitary summit CV ideally drained directly from the LV summit. However, in many cases, the lumen of the target vein may be too small, making device cannulation challenging, or the veins may be too distant from the intramural origin of the VAs. In addition, collateral venous flows might drain the injected ethanol away from the targeted myocardium, reducing the therapeutic effects. To solve these problems, several advanced methods, such as the “multiple balloons, multiple veins approach”, have been proposed and proven highly successful in treating complex summit VA cases (12).

A careful pre-ablation assessment of the venous anatomy is crucial for the success of RCVEA. In this regard, high-resolution CT with contrast was performed to evaluate the patient's cardiac venous system and the surrounding cardiac structures. This helped us identify the summit CV suitable for ethanol ablation and determine the appropriate balloon size. We also suggest routine detailed mapping of coronary veins using a microcatheter when performing RFCA of VAs with a suspected LV summit origin. Furthermore, the microcatheter tip can be used as an anatomical landmark for precise RFCA delivery. In the present case, instrumentation into the target coronary veins was also challenging due to a steep angle at the GCV-AIV transition. We utilized a JR 5-Fr angioplasty guiding catheter with backup support from a bi-directional guiding sheath to achieve smooth and stable cannulation into the summit CV.

Written informed consent was obtained from the patient for the publication of this case report and accompanying images.

The authors state that they have no Conflict of Interest (COI).

Financial Support

This work was supported by the Takagi Hospital Cardiology Research Grant.