Ventricular Tachycardia in Ischemic Cardiomyopathy: Are You Certain You Are Ablating the Clinical VT?

Case Description

The patient is a 52-year-old male with ischemic cardiomyopathy (ICM), severe mitral regurgitation, an estimated left ventricular ejection fraction of 15-20%, and New York Heart Association class I-II. He had several stents placed in the left anterior descending coronary artery in 2008, and underwent multivessel coronary artery bypass graft surgery (i.e., five grafts) in 2009. Subsequently, he received a single-chamber, single-coil implantable cardioverter-defibrillator (ICD) in 2010 for primary prevention of sudden cardiac death. A year after ICD implantation, the patient received seven appropriate ICD shocks for episodes of monomorphic ventricular tachycardia (VT). He was placed on amiodarone, and the dose of beta-blockers was increased. He remained asymptomatic until 2012, when he had recurrent ICD shocks. Ever since, he has had a total of 40 episodes of tachyarrhythmias labeled by his device as either supraventricular tachycardia (SVT) or VT, for which he has received 28 ICD shocks. His baseline 12-lead electrocardiogram (ECG) revealed normal sinus rhythm with a left bundle branch block (LBBB) with a duration of 132 ms (Figure 1). Unfortunately, there was no ECG of the clinical arrhythmia available for interpretation. Device interrogation showed rapid monomorphic VT ranging from 188 bpm to 210 bpm, and one episode of VT that degenerated into VF at 255 bpm. Several of these episodes were successfully terminated with antitachycardia pacing. 

The patient was taken to the electrophysiology laboratory for an EP study and possible VT radiofrequency ablation. The procedure was performed initially under conscious sedation, and ultimately, under general anesthesia. Prior to the procedure, the patient’s ICD was temporarily reprogrammed and tachycardia therapies were disabled. A 4 French (F) quadripolar catheter was advanced to the His position and then to the right ventricular apex, where it was used for pacing and recording. A 10F SOUNDSTAR intracardiac echocardiogram catheter (Biosense Webster, Inc., a Johnson & Johnson company) was advanced to the right atrium, right ventricular outflow tract, and right ventricle, and was used for creation of a 3D anatomic shell of the LV and RV. A transseptal atrial puncture was performed using fluoroscopic and intracardiac ultrasound guidance. An Agilis large curve sheath (Abbott) was dragged from the SVC along the septum until a sudden leftward displacement was observed. Engagement of the fossa ovalis was confirmed by interatrial tenting seen under intracardiac ultrasound guidance. Subsequently, a SafeSept Transseptal Guidewire (Pressure Products) was used to puncture the interatrial septum. A bolus injection of heparin 120 units/kg was administered prior to the transseptal puncture, and the activated clotting time maintained more than 350 sec with additional boluses every 30 minutes as necessary. Via the Agilis sheath, we introduced the 3.5 mm SMARTTOUCH SF (FJ curve) catheter (Biosense Webster, Inc., a Johnson & Johnson company) into the LV. Subsequently, substrate mapping was performed over the endocardium of the LV. In the LV, an extensive area of low voltage (less than 1.5 mv) was present in the anteroseptal and inferoseptal walls. This scar extended from base to apex (Figure 2). A unipolar voltage map also suggested midmyocardial and epicardial involvement in the same area, probably indicating transmural scar from prior myocardial infarctions (Figure 2). 

Since the VT morphology was unknown and the question about SVT had been raised before, we decided to perform EPS before performing empirical substrate modification of the scar. Baseline rhythm at baseline was sinus rhythm with LBBB 130 ms. There was evidence of conduction delay in the His-Purkinje system as revealed by a prolonged HV interval at baseline or during atrial pacing (i.e., 75 ms) (Figure 3). There was no evidence of dual AV nodal physiology. The arrhythmia could be induced by programmed ventricular stimulation from the right ventricular apex. This was a wide complex tachycardia at 210 bpm with LBBB morphology transition in V3 and left inferior axis. The arrhythmia was induced with a single extrastimulus (500/280 ms). Precordial lead morphology was very similar to sinus rhythm (Figure 4). Figure 5 illustrates clear evidence of VA dissociation, strongly suggesting VT instead of SVT with aberrancy, in particular, AV reentrant tachycardia. Interestingly, the His electrogram clearly preceded QRS during VT, and the HV interval during VT was longer than the HV interval during sinus rhythm (84 ms vs 75 ms). Both findings are typically observed in bundle branch reentry VT (BBRVT). This prolongation of the HV interval during tachycardia is speculated to be caused by anisotropic conduction seen in the distal His bundle at the upper turnaround point of the tachycardia circuit. Spontaneous variations in V-V intervals were also preceded by similar changes in H-H intervals (Figure 6). Entrainment of the VT by rapid pacing at the RV apex showed manifest fusion, suggesting that the mechanism is likely to be BBRVT rather than myocardial reentry. Post-pacing interval (PPI) minus tachycardia cycle length (TCL) was 29 ms. The first PPI after entrainment from the RV apex has been correlated with the distance from the pacing site to the reentrant circuit. This can differentiate BBRVT, where the circuit is close to the RV apex, from ventricular myocardial reentrant tachycardia or AV nodal reentrant tachycardia with aberrancy, where the circuit is away from the RV apex. A value of PPI-TCL <30 ms after entrainment by RV apical stimulation makes BBRVT a likely possibility. Conversely, a PPI-TCL >30 ms makes BBRVT unlikely (Figure 7). We tried to entrain the tachycardia by atrial pacing to demonstrate concealed fusion, but AV conduction limited this maneuver. All the above findings proved that the patient has BBRVT, so the decision was made to localize and ablate the left bundle branch. Initially, the LB was localized (LB-QRS 48 ms) (Figure 8), and three radiofrequency lesions were delivered at this location, obtaining an impedance drop of 14 ohms during the third application. QRS duration increased from 132 to 180 ms after ablation, and the arrhythmia was no longer inducible, even with triple extrastimuli from the right and left ventricles (Figure 9). 

Subsequently, based on the large septal scar revealed by the endocardial voltage map and several signals within the low-voltage area showing late potentials and local abnormal ventricular activity (LAVA) (Figure 10), the decision was made to perform substrate modification of the scar (Figure 11). Areas of late potentials in the scar and areas of electrically unexcitable scar (pacing threshold greater than 10 mA at 2 ms pulse width) were tagged (Figure 10). Much of the region of scar surrounding the mid septum was not electrically excitable to 10 mA at 2 ms pulse width stimuli. Some sites captured with long delays and QRS morphologies, suggesting propagation away through protected channels. Ablation mapping of the LV scar was performed during sinus rhythm. This was achieved using a 3.5 mm, externally irrigated THERMOCOOL SMARTTOUCH catheter (Biosense Webster, Inc., a Johnson & Johnson company). RF ablation lesions were titrated up to a maximum of 50 W over 60-90 seconds, for an impedance drop of up to 18 ohms. Where possible, a contact force of greater than 10 grams was achieved. Lesion creation was ensured by a minimum impedance fall of 10 ohms with RF application. A total of 72 lesions in the interventricular septum were delivered for substrate modification (scar dechanneling/core isolation). RF ablation performed at likely reentry circuit sites based on substrate mapping rendered the region electrically unexcitable (Figure 11). After ablation, programmed stimulation with up to three extrastimuli during pacing at a cycle length of 500 ms was performed, and no arrhythmias were induced. The patient was discharged the next day off antiarrhythmic medications. Anticoagulation was administered for six weeks. At six-month follow-up, the patient has remained asymptomatic and device interrogation has revealed no arrhythmia recurrences.

Discussion

BBRVT is a form of macroreentrant tachycardia involving the bundle of His, both bundle branches, and the ventricular myocardium in the circuit. It generally occurs in the background of dilated cardiomyopathy, prior valve surgery, or other cardiac disease such as myotonic dystrophy, and rarely in patients with normal ventricular function.¹ The vast majority is related to non-ischemic substrates; BBRVT can be the mechanism of tachycardia in 30-50% of patients with non-ischemic dilated cardiomyopathy who have inducible sustained monomorphic ventricular tachycardia.² On the contrary, the incidence of this tachycardia is no more than 5-6% in patients who have ischemic heart disease, mostly occurring in patients with large anterior myocardial infarction and right bundle branch block with left anterior fascicular block or left posterior fascicular block.³ However, taking into consideration that BBRVT characteristically displays a QRS that is identical to the QRS in sinus rhythm, ICD detection algorithms may mislabel it as SVT, leading to underdiagnosis, as seen in our case. Clinically, BBRVT usually results in marked hemodynamic compromise and often presents with syncope, presyncope, or sudden cardiac arrest. When a ventricular tachycardia is induced, the presence of His deflections preceding every ventricular deflection should alert the possibility of this entity. It is important to show that oscillations in the H-H cycle length result in variations in V-V cycle length. Entrainment of the tachycardia from the atrium and right ventricular apex and characteristics of PPI can be used to differentiate this arrhythmia from intramyocardial reentry and SVT with aberrancy. Right or left bundle branch ablation usually cures the tachycardia, and recurrence is uncommon. In patients with nonischemic dilated cardiomyopathy and in patients with prior valve surgery, approximately one-third of inducible sustained VTs are BBRVTs. This arrhythmia is seen in patients with underlying disease of the His-Purkinje system. A resting ECG of these patients often reveals conduction abnormalities in the form of a prolonged PR interval, nonspecific intraventricular conduction delay, incomplete or complete BBB, and occasionally, complete heart block. During VT, the surface QRS morphology is of a typical BBB pattern, with a LBBB pattern being more common than a RBBB pattern. 

Criteria based on the recording of the His bundle potentials have several limitations, such as failure to record the His bundle electrogram during VT either because of catheter instability or because of severe His-Purkinje disease. Oscillations in the V-V interval may precede those of the H-H interval because of conduction variations in the antegrade rather than the retrograde conducting bundle branch.⁴ The HV interval during tachycardia may occasionally be shorter than the HV interval during sinus rhythm if the His recording is proximal to the turnaround point of the reentrant circuit. The demonstration of concealed fusion during entrainment of a wide QRS complex tachycardia by atrial stimulation rules out pure myocardial reentry as the mechanism and favors BBRVT. However, entrainment of BBRVT by atrial pacing frequently requires isoproterenol infusion to improve AV nodal conduction that may be poorly tolerated because of acceleration of the tachycardia. 

As stated above, right bundle branch ablation usually cures the tachycardia, and recurrence is uncommon. Even though it is technically more challenging in patients with a complete LBBB pattern during sinus rhythm (such as our patient), ablation of the left bundle may be more appropriate for the treatment of BBRVT. This can mitigate the risk of complete heart block by avoiding right bundle ablation, as antegrade ventricular activation occurs through the right bundle.⁵

The optimal ablation approach in patients with BBRVT and large ischemic substrates, such as in this case, is unknown. It seems reasonable to perform substrate modification in addition to bundle branch ablation, particularly in patients with extensive substrates where the clinical arrhythmia has not been well documented. The risks and benefits of a longer procedure, taking into consideration longer anesthesia as well as longer fluoroscopy and RF times, needs to be entertained for clinical decision making. Large studies with long-term follow-up are needed to determine the best ablation strategy for these patients.

Disclosure: Dr. Di Biase is a consultant for Stereotaxis, Biosense Webster, Boston Scientific and Abbott; he has received speaker honoraria/travel from Medtronic, Janssen, Pfizer, EpiEP, and BIOTRONIK. The other authors have no conflicts of interest to report regarding the content herein.

References

1. Balasundaram R, Rao HB, Kalavakolanu S, Narasimhan C. Catheter ablation of bundle branch reentrant ventricular tachycardia. Heart Rhythm. 2008;5:S68-S72.
2. Tchou P, Mehdirad AA. Bundle branch reentry ventricular tachycardia. Pacing Clin Electrophysiol. 1995;18:1427-1437.
3. Benito B, Josephson ME. Ventricular tachycardia in coronary artery disease. Rev Esp Cardiol. (Engl Ed) 2012;65:939-955.
4. Merino JL, Peinado R, Fernandez-Lozano I, Sobrino N, Sobrino JA. Transient entrainment of bundle-branch reentry by atrial and ventricular stimulation: elucidation of the tachycardia mechanism through analysis of the surface ECG. Circulation. 1999;100:1784-1790.
5. Blanck Z, Deshpande S, Jazayeri MR, Akhtar M. Catheter ablation of the left bundle branch for the treatment of sustained bundle branch reentrant ventricular tachycardia. J Cardiovasc Electrophysiol. 1995;6:40-43.