Outcomes of Percutaneous Trans–Right Atrial Access to the Left Ventricle for Catheter Ablation of Ventricular Tachycardia in Patients With Mechanical Aortic and Mitral Valves

Key Points

Question  Is percutaneous catheter ablation of recurrent ventricular tachycardia associated with a left ventricular (LV) endocardial substrate in patients with combined mitral and aortic mechanical valves possible?

Findings  This case series is a report of the feasibility and safety of an entirely percutaneous approach to obtaining LV access in patients with aortic and mitral mechanical valves, which uses a conventional transfemoral venous route and involves the creation of an iatrogenic Gerbode defect by direct puncture and balloon dilatation of the inferior-septal process of the LV from the adjacent inferior and medial aspects of the right atrium. This approach was successfully performed in 4 consecutive patients and eliminated drug-refractory recurrent ventricular tachycardia associated with an LV substrate, with no periprocedural complications.

Meaning  This novel technique of obtaining percutaneous LV endocardial access in patients with mechanical valves in the aortic and mitral positions may allow for catheter ablation of recurrent ventricular tachycardia in patients in whom conventional LV access via retrograde aortic or atrial transseptal routes is not possible.

Importance  In patients with mechanical valves in the aortic and mitral positions, percutaneous access to the left ventricle (LV) via a transfemoral approach for catheter ablation of ventricular tachycardia (VT) has been considered infeasible.

Objective  To describe the outcomes of a novel percutaneous trans–right atrial (RA) access to the LV via a femoral venous approach for catheter ablation of VT in patients with mechanical aortic and mitral valves.

Design, Setting, and Participants  This observational study included consecutive patients with mechanical valves in the aortic and mitral positions and recurrent monomorphic drug-refractory VT associated with an LV substrate. Percutaneous LV access was performed from a transfemoral venous route with the aid of a deflectable sheath and a radiofrequency wire by creating an iatrogenic Gerbode defect with direct puncture of the inferior and medial aspect of the RA, adjacent to the inferior-septal process of the LV (ISP-LV), under intracardiac echography guidance. Once the wire crossed to the LV, balloon dilatation of the ventriculotomy site (with a noncompliant balloon; diameter, 8 to 10 mm) was performed to facilitate passage of the sheath within the LV.

Exposures  Percutaneous trans-RA access to the LV via puncture of the ISP-LV to perform catheter ablation of VT in patients with mechanical aortic and mitral valves.

Main Outcomes and Measures  Feasibility and safety of a trans-RA access to the LV for catheter ablation of VT.

Results  A total of 4 patients (mean [SD] age, 60 [7] years; mean [SD] LV ejection fraction, 31% [9%]) with recurrent VT associated with an LV substrate (ischemic cardiomyopathy, 3 patients; nonischemic cardiomyopathy, 1 patient) and mechanical valves in the aortic and mitral position underwent trans-RA access through the ISP-LV for catheter ablation of VT. The time to obtain LV access ranged from 60 minutes (first case) to 22 minutes (last case) (mean [SD], 36 [15] minutes). No complications associated with the access occurred. In particular, in the 3 patients with preserved atrioventricular conduction at baseline, no new conduction abnormalities were observed after the access. Complete VT noninducibility at programmed ventricular stimulation was achieved in 3 cases, and no patient had VT recurrence at a median follow-up of 14 months (range, 6-21 months).

Conclusions and Relevance  A percutaneous trans-RA access to the LV via a femoral venous approach for catheter ablation of VT in patients with mechanical aortic and mitral valves is feasible and appears safe. This novel technique may allow for catheter ablation of VT in a population of patients in whom conventional LV access via retrograde aortic or atrial transseptal routes is not possible.

In patients with combined mitral and aortic mechanical valves and recurrent ventricular tachycardia (VT) associated with a left ventricular (LV) substrate, traditional retrograde aortic or antegrade transmitral LV access approaches are not possible.^1-4 We describe a novel technique to obtain LV access in patients with mechanical aortic and mitral valves, which involves direct puncture of the inferior-septal process of the LV (ISP-LV) from the adjacent right atrium (RA).

Consecutive patients with recurrent VT associated with an LV substrate and mechanical valves in the mitral and aortic position underwent planned trans-RA access to the LV at the University of Pennsylvania between February 2018 and May 2019. Patients who underwent access to the LV via puncture of the muscular interventricular septum, as previously described,^5,6 were not included. All patients signed an informed consent form, and this study was approved by the University of Pennsylvania Health System. Data were entered in an institutional review board–approved registry.

Periprocedural therapeutic warfarin was continued, and unfractionated heparin with a target-activated clotting time of more than 350 seconds was administered after vascular access. The optimal access point was defined by intracardiac echography (ICE) as the inferoseptal RA adjacent to the most inferior portion of the ISP-LV (at the level of the inferior third of the coronary sinus ostium) (Video 1). The Video details how access was obtained, which used the following stepwise approach:

1. The sheath was deflected to obtain contact with the inferomedial aspect of the RA, adjacent to the inferior portion of the ISP-LV (Video 2).

2. Puncture of the ISP-LV from the RA side was performed with a dedicated 0.035-in radiofrequency (RF) wire (PowerWire RF Wire; Baylis Medical), with RF delivered for 4 to 5 seconds at 15 to 25 W (Video 3).

3. The RF wire was exchanged with a stiffer wire (Amplatz Super Stiff; Boston Scientific) using a hydrophilic 4F microcatheter (Terumo Glidecath; Terumo) (Video 4).

4. Balloon dilatation of the access site with a noncompliant balloon (width, 8-10 mm; length, 2-4 cm) was performed over the Amplatz wire. Once the ISP-LV access site was appropriately dilated, the balloon was deflated and the deflectable sheath advanced over the balloon in the LV (Video 5).

Figure 1 and Figure 2 show sequential fluoroscopic and ICE images, respectively, demonstrating the access technique. Details of the catheter ablation approach are reported in the eMethods and eFigure 1 in the Supplement. After completion of the ablation procedure, the deflectable sheath and catheter were withdrawn to the RA and the residual ISP-LV to RA shunt size (width) was measured with ICE color Doppler analysis. A second transthoracic echocardiogram (TTE) was performed 24 hours postprocedure and between 4 and 8 weeks postdischarge to evaluate for residual shunts. Further images are in eFigure 2 in the Supplement.

A total of 4 patients (mean [SD] age, 60 [7] years; 100% men; mean [SD] LV ejection fraction, 31% [9%]) were included (Table). The median international normalized ratio on the day of the procedure was 2.7 (range, 2.2-3.1). After selecting the site of access with ICE (step 1), periprocedural coronary angiogram was performed in the first 2 cases, and this confirmed adequate distance from coronary branches visible on angiography. In both instances, a clear delineation of the atrioventricular nodal (AVN) artery was not possible. These 2 patients also underwent preprocedural, contrast-enhanced computed tomography (CT), which did not demonstrate any major coronary artery branch in proximity to the planned access site. The course of the AVN artery also could not be defined by preprocedural, contrast-enhanced CT. Based on the experience from these first 2 patients, preprocedural CT or intraprocedural coronary angiography were not performed in the most recent cases.

The mean (SD) time to obtain LV access (from the beginning of step 1 to the end of step 4) was 36 (15) minutes and ranged from 60 minutes (in the first case) to 22 minutes (in the last case). Radiofrequency application over the wire induced sustained ventricular arrhythmias in 2 cases (ventricular fibrillation in patient 2 and VT in patient 3) requiring a single, external DC shock (step 2). In the remaining 2 cases, nonsustained runs of VT were observed. In the 3 patients with baseline intact AVN conduction, no new conduction abnormalities were observed after completion of LV access.

The number of distinct VTs that were recorded and targeted for ablation was a median of 3 per patient (range, 2 to 6 per patient). After ablation, complete VT noninducibility was achieved in 3 patients (75%). The residual iatrogenic Gerbode shunt visualized at ICE postprocedure measured a mean (SD) of 4.0 (0.7) mm (range, 3-5 mm) (eFigure 3 in the Supplement; Video 6). A TTE was repeated 24 hours postprocedure and made visualization of the residual Gerbode defect feasible in 2 of 4 cases. In all patients, the defect was no longer detected on repeated TTE 4 to 8 weeks postprocedure.

After a median follow-up of 14 months (range, 6-21 months), no patient experienced VT recurrence. No patient developed new or worsening heart failure symptoms or had heart failure hospitalizations during follow-up, and no patient developed new conduction abnormalities. One patient (patient 3) died 2 months postprocedure from an unrelated illness.

This is, to our knowledge, the first description of an entirely percutaneous approach to obtain LV access in patients with aortic and mitral mechanical valves that involves the creation of an iatrogenic Gerbode defect by direct puncture and balloon dilatation of the ISP-LV from the adjacent inferior and medial aspect of the RA. This approach was successfully performed in 4 consecutive cases, and it appeared safe.

Prior studies have defined patients with aortic and mitral mechanical valves as having a no-entry LV and advocated for surgical epicardial access with epicardial-only ablation or percutaneous transapical access.^7-9 Noninvasive cardiac radioablation may also represent a suitable treatment alternative in these patients.^10,11 We have previously documented the feasibility of percutaneous access to the LV with puncture of the muscular interventricular septum,^5,6 which is only feasible via a right internal jugular vein approach to obtain a favorable angle of access to the right ventricular aspect of the septum. Therefore, although it is possible, LV access via interventricular septal puncture is cumbersome and associated with suboptimal ergonomics and increased radiation exposure to the operator.

In this study, the selection of the ISP-LV adjacent to the inferomedial RA as a suitable site for LV access was mostly meant to replicate an angle of access to the LV similar to a conventional atrial transseptal approach, to facilitate catheter manipulation. Mapping different regions of the LV was not difficult, with the exception of the basal-septal LV adjacent to the access point (in none of the cases did the abnormal substrate extend to the basal-septal LV adjacent to the access site). Particular attention was exercised to always maintain the sheath across the access site, because readvancement of the sheath within the LV over the mapping or ablation catheter may be more challenging than what is typically encountered with an atrial transseptal approach. The choice of the most inferior aspect of the ISP-LV (at the level of the inferior third of the coronary sinus ostium) was to minimize the chances of injuring the compact AVN, which is positioned more superiorly, as well as to prevent injury to the AVN artery, which typically originates from the right coronary artery and runs in the epicardial RA, connecting to the AVN at the most superior aspect of the coronary sinus ostium.¹² Three of the 4 patients included in this series had preserved baseline AVN conduction, and we did not observe any new conduction abnormalities following the LV access or over follow-up.

We used an insulated RF wire to perform the puncture, and the issue that was noted with RF-assisted access was pro-arrhythmia: 2 patients had sustained ventricular fibrillation or VT during RF applications and were managed with cardioversion. The time to obtain access varied considerably from the first to the last case of this series, which is attributable to an inherent learning curve with this novel technique.

Importantly, this technique heavily relies on intraprocedural ICE imaging. As such, our results may not be generalized to centers that do not routinely use ICE for complex ablation procedures or operators who are not proficient with ICE imaging.

At the end of the procedure, a residual Gerbode defect could be detected in all patients, although it appeared substantially smaller than expected based on the size of the balloon used for dilatation. This may be because of the spiral distribution of the interventricular septum muscular fibers, which may favor some degree of elastic recoil with partial closure once the access sheath has been withdrawn in the RA.¹³ The residual Gerbode defect could still be visualized at TTE 24 hours postprocedure in 3 of 4 cases, and there was no evidence of residual shunt at follow-up TTE between 4 and 8 weeks postprocedure. The use of TTE to image the postprocedural residual Gerbode defect may have been suboptimal to detect and characterize small residual defects. More advanced imaging modalities might better assess and confirm the resolution of the iatrogenic Gerbode defect.

An entirely percutaneous trans-RA approach that involves the creation of an iatrogenic Gerbode defect through puncture and balloon dilatation of the ISP-LV via a conventional femoral venous approach is feasible. It appears safe to obtain LV access for catheter ablation of VT in patients with aortic and mitral mechanical valves.

Accepted for Publication: July 22, 2020.

Corresponding Author: Pasquale Santangeli, MD, PhD, Electrophysiology Section, Cardiovascular Division, Hospital of the University of Pennsylvania, 9 Founders Pavilion–Cardiology, 3400 Spruce St, Philadelphia, PA 19104 (pasquale.santangeli@pennmedicine.upenn.edu).

Published Online: September 30, 2020. doi:10.1001/jamacardio.2020.4414

Author Contributions: Dr Santangeli had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Santangeli, Hyman, Muser, Shivkumar.

Acquisition, analysis, or interpretation of data: Santangeli, Hyman, Callans, Shivkumar, Marchlinski.

Drafting of the manuscript: Santangeli, Muser, Shivkumar.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Santangeli.

Administrative, technical, or material support: Hyman, Marchlinski.

Conflict of Interest Disclosures: Dr Santangeli grants and personal fees from Biosense Webster Biosense Webster and personal fees from Baylis Medical during the conduct of the study. Dr Marchlinski is a consultant for Biosense Webster and reported personal fees from Abbott Medical for participation in a scientific advisory board outside the submitted work. Dr Shivkumar reported receiving support from the National Institutes of Health (grants OT2OD023848 and OT2OD028201) and being a founder of NeuCures and EP Dynamics outside the submitted work; in addition, University of California, Los Angeles has a patent application (US 2015/0182255 A1) related to this work by Dr Shivkumar, assigned to the regents of the University of California. No other disclosures were reported.

Funding/Support: Drs Santangeli and Marchlinski are supported by the Winkelman Family Fund in Cardiovascular Innovation.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.