Successful ventricular tachycardia ablation in a patient with limited vascular and left ventricular endocardial access due to multiple mechanical cardiac support devices

Abstract

A 48-year-old male with ischemic cardiomyopathy was admitted because of cardiogenic shock due to incessant ventricular tachycardia (VT). Despite the introduction of multiple mechanical cardiac support (MCS) devices, his hemodynamics were unstable; therefore, VT ablation was performed. Due to the limited vascular and left ventricle (LV) access with the multiple MCS devices, only the left femoral artery and vein were accessible. In addition, the transseptal approach as an LV access was difficult because of the MitraClip (Abbott Park, IL, USA). Posterior papillary muscle-origin VT was successfully treated via a transaortic approach under the MCS. After the VT ablation, the patient recovered from the MCS.

Learning objective

Mechanical cardiac support (MCS) devices are critical during ventricular tachycardia (VT) ablation in patients with left ventricular (LV) dysfunction. However, the ablation strategy for cases with limited access due to multiple MCS devices has not been established. In this case, VT was successfully treated via a transaortic approach using intracardiac echocardiography. The number of cases requiring VT ablation for patients with limited vascular and LV endocardial access would increase; therefore, the accumulation of those cases is required to find a better strategy.

Introduction

It has been reported that mechanical cardiac support (MCS) devices are effective during ventricular tachycardia (VT) catheter ablation in patients with severe left ventricular (LV) dysfunction. However, vascular and endocardial LV access would be critical under multiple MCS devices. Further, crossing the catheters through the mitral valve after an edge-to-edge repair device (MitraClip®, Abbott Park, IL, USA) has been inserted has the risk of mechanical complications. We report a case who underwent a successful endocardial VT ablation via a transaortic approach with limited vascular and LV access due to the use of multiple MCS devices [percutaneous ventricular assist device (Impella 5.0®, Abiomed, Danvers, MA, USA), veno-arterial extracorporeal oxygenation (VA-ECMO), MitraClip, cardiac resynchronization therapy with defibrillator (CRT-D), and a pulmonary artery catheter].

Case report

A 48-year-old man with ischemic cardiomyopathy presented due to cardiogenic shock and incessant non-sustained ventricular tachycardia (NSVT) (Fig. 1A). He underwent MitraClip implantation for severe functional mitral regurgitation (MR) due to ischemic cardiomyopathy three months previously and CRT-D implantation 10 years prior. After admission, he did not recover from cardiogenic shock despite the administration of inotropic medications, so MCS devices were introduced. A percutaneous ventricular assist device (Impella 5.0) was inserted from the right subclavian artery, and a VA-ECMO device was inserted from the right femoral artery and vein. However, an incessant NSVT could not be suppressed by any antiarrhythmic drugs or deep sedation; therefore, we decided to perform an urgent VT ablation. The procedure was performed under general anesthesia with midazolam, dexmedetomidine, and fentanyl. The VT morphology exhibited a superior axis and right branch block pattern, suggesting that the origin of the arrhythmia was from the infero-posterior region of the LV. The patient had limited vascular access due to the use of multiple MCS devices and monitoring catheters (Fig. 1B and C) as follows: VA-ECMO device was inserted from the right femoral artery and vein, Impella 5.0 from the right subclavian artery, pulmonary artery catheter from the right jugular vein, and CRT-D implanted from the left subclavian vein, and the innominate vein was unfortunately occluded due to the CRT-D leads. Therefore, only the left femoral artery and vein were accessible sites. In addition, the transseptal access as an LV approach was difficult due to the presence of a MitraClip. As a result, we chose the transaortic approach for the LV endocardial access. First, the LV and right ventricular (RV) geometries, including the papillary muscles, were created using an intracardiac echocardiography catheter (SOUNDSTAR®; Biosense Webster, Inc., Diamond Bar, CA, USA) from the right atrium (RA) and RV and was integrated into the CARTO mapping system (CARTO3®, Biosense Webster Inc.) (Fig. 2). An LV endocardial voltage map was also created by a DECANAV® catheter (Biosense Webster, Inc.), revealing an extensive dense scar in the anteroseptal LV region. On the other hand, the voltage on the infero-posterior wall, including the posterior papillary muscle (PPM), was relatively preserved (Fig. 3A).

Fig. 1.

12‑lead electrocardiogram of the NSVT and fluoroscopic images during the procedure. The 12‑lead electrocardiogram (A) shows ventricular tachycardia with a superior axis and right bundle branch block type QRS morphology. The fluoroscopic images (B and C) show multiple devices such as the CRT-D leads, Swan-Ganz catheter, MitraClip, VA-ECMO, and Impella catheter. The ablation catheter was retrogradely inserted into the LV via the transaortic approach.

NSVT, non-sustained ventricular tachycardia; CRT-D, cardiac resynchronization therapy with defibrillator; RV, right ventricle; LV; left ventricle; ECMO, extracorporeal oxygenation; ICD, implantable cardioverter defibrillator; ABL, ablation catheter; SGC, Swan-Ganz catheter; RAO, right anterior oblique; LAO, left anterior oblique.

Fig. 2.

LV and RV geometries on the CARTO system. The LV and RV geometries, including the papillary muscles, were created using an intracardiac echocardiographic catheter (SOUNDSTAR®) in the RAO (A and C) and LAO (B and D) views are shown. The Impella catheter was also drawn on the CARTO system.

RAO, right anterior oblique; LAO, left anterior oblique; Ao, aorta; RV, right ventricle; LV, left ventricle; AV, aortic valve; APM, anterior papillary muscle; PPM, posterior papillary muscle.

Fig. 3.

Voltage map during sinus rhythm and activation map during VT. The voltage maps during sinus rhythm (A) revealed an extensive dense scar in the anteroseptal LV region, and the voltage on the infero-posterior wall, including the PPM, was relatively preserved. An activation map of the target NSVT (B) exhibited a centrifugal activation pattern from the PPM. The blue dot indicates the successful ablation site of (B). The local potentials at that site recorded by the ablation catheter preceded the QRS by 31 msec during the VT in (C).

AP, antero-posterior; RAO, right anterior oblique; NSVT, non-sustained ventricular tachycardia; PPM, posterior papillary muscle.

An activation map of the target NSVT revealed a centrifugal activation pattern from the PPM (Fig. 3B). Fortunately, the electromagnetic interference was mild and did not require specific intervention. Although a detailed electrophysiological study could not be performed because the VT did not sustain, triggered activity or abnormal automaticity was speculated. The local potentials at the earliest activation site proceeded the QRS by 31 msec during the VT (Fig. 3C), and the NSVT disappeared after repeated radiofrequency applications at that point using a 3.5-mm-tip open-irrigated ablation catheter (THERMOCOOL SMARTTOUCH®; Biosense Webster, Inc.) with a radiofrequency application power of 35 watts while confirming the position of the PPM and ablation catheter by CARTOSOUND. Total procedure time was 345 min, and radiation time and doses were 42.2 min and 251 mGy, respectively. The patient's hemodynamics were stabilized with MCS during the procedure. After the ablation, his hemodynamics gradually improved, and the MCS devices were successfully withdrawn.

Discussion

Herein, we report a case of a successful VT ablation in a patient with limited vascular and LV endocardial access due to the use of multiple MCS devices. There have been few reports of VT ablation in patients with a MitraClip and MCS, and no established strategy exists. Haegeli et al. reported catheter ablation of VT by a transseptal approach in 5 patients with MitraClip devices [1]. In that report, the ablation catheter was crossed over the mitral valve with a MitraClip using a steerable sheath with the guidance of transesophageal 3-dimensional echocardiography and fluoroscopic images. As a result, no worsening of MR was documented after the procedure. Although this approach might be feasible, it should be limited to patients who are not eligible for a transaortic approach. In this case, we performed the procedure by a transaortic approach considering the risk of damaging the MitraClip.

In addition, catheter ablation in patients with severe LV dysfunction is challenging because the induced VTs can cause a hemodynamic collapse and sometimes require MCS, such as Impella and VA-ECMO. Aryana et al. reported that Impella support during catheter ablation of unstable VT could shorten the ablation time and hospital stay [2]. However, vascular access would be a critical issue under the use of multiple MCS devices. In this case, only the left femoral artery and vein were accessible. Creating the LV geometry with CARTOSOUND from the RA and RV was helpful to recognize the LV anatomical structures, especially the position of the PPM. The transaortic approach was successfully performed by adjusting the Impella and VA-ECMO flow when the ablation catheter was passed through the aortic valve. In addition, the MCS for stabilizing the hemodynamics contributed to performing an electrophysiological evaluation, resulting in the diagnosis of the PPM origin-NSVT in this case. On the other hand, an epicardial approach is also considered when an endovascular approach is difficult; however, this PPM-origin VT could probably not be treated by an epicardial approach. Recently, the effectiveness of VT ablation has been reported [3]. It is expected that the number of cases requiring VT ablation in patients with limited vascular and LV endocardial access will increase; therefore, the accumulation of those cases is required to find a better strategy.

Funding

None.

Declaration of competing interest

The authors declare no conflict of interest related to this paper.