Cerclage parahisian septal pacing through the septal perforator branch of the great cardiac vein: Bedside-to-bench development of a novel technique and lead

Introduction

Physiological biventricular pacing is an attractive alternative to conventional right ventricular (RV) pacing in the management of conduction system disease.¹ RV septal pacing is effective but dogged by lead instability^2,3 and occasional serious tricuspid valve injury.⁴ Permanent His-bundle pacing has been achieved^5–11 but challenges remain, including high pacing threshold, low ventricular electrogram amplitude, lead instability, and procedural complexity.¹² We recently described an alternative pacemaker lead trajectory to achieve physiological pacing, by lead implantation into the interventricular septum through the septal perforator branch of the great cardiac vein, the terminal branch of the coronary sinus.¹³ We first exploited this catheter trajectory to achieve transcatheter mitral cerclage annuloplasty,^14,15 therefore we term our technique cerclage parahisian septal pacing. In our initial case description, we achieved physiological cerclage parahisian septal pacing with a low pacing threshold using a conventional bipolar coronary sinus pacing electrode. In this study, we evaluated cerclage parahisian septal pacing further in a prospective cohort series and introduced a new type of pacing lead.

Methods

Technique of cerclage parahisian septal pacing

The implantation technique is shown in Figure 1. A fluoroscopic His-bundle target was created using a transfemoral 6F CRD curve quadripolar electrophysiology catheter. The axillary vein was accessed through a subclavicular pocket, and the coronary sinus was engaged using a 9F CPS Aim SL catheter (Abbott, Minneapolis, MN) and a polymer-jacketed nitinol guidewire. A septal perforator branch of the anterior interventricular vein was identified on coronary sinus angiography and engaged with a 0.014-inch Choice PT coronary guidewire (Boston Scientific, Natick, MA) through a dual-lumen microcatheter (Crusade, Kaneka Medix, Japan). Through the side hole of the dual-lumen microcatheter, a rigid coronary guidewire designed to penetrate chronic total occlusion (MiracleBros3 or 6, Asahi-Intecc, Seto, Japan) was advanced through the septal perforator vein across the interventricular septal myocardium toward the His catheter target into the RV chamber.

Figure 1.

Representative clinical cerclage pacing procedures in 2 patients with a prosthetic tricuspid valve. A, B: Septal perforator branch of anterior interventricular vein was identified on preprocedural cardiac computed tomography. C: Coronary sinus venogram identified a suitable septal branch. Through a double-lumen microcatheter (D), a guidewire penetrated the interventricular septum (E) to deliver a bipolar lead into the right ventricle (F). G: In 2 cases, a snare provided countertraction (H) to deliver the bipolar lead into the septum.

A bipolar venous pacing lead (patient 1: Easytrak 2 IS-1 4543, Boston Scientific; patients 2–5: Attain Ability 4196 lead, Medtronic, Minneapolis, MN) was then advanced via the coronary sinus into the interventricular septum over the septal 0.014-inch guidewire. In 2 cases, this required countertraction on the guidewire through a transfemoral RV snare (Ensnare, Merit Medical, South Jordan, UT). The lead was advanced and retracted to accomplish the narrowest paced QRS complex before removal of the guidewire and delivery sheath. An atrial lead was placed as usual. No other complications occurred.

Clinical study

The institutional review boards of Pusan National University Yangsan Hospital and of Asan Medical Center approved a prospective evaluation of 10 subjects (NCT03438591). Patients in whom conventional pacing was not feasible for previous tricuspid valve surgery were recruited. In patient 5, cerclage pacing was attempted in lieu of conventional cardiac resynchronization treatment. All subjects provided written informed consent. Subjects were examined 1 and 6 months after the procedure by device interrogation and chest radiography.

Custom cerclage pacing lead for preclinical investigation

Our experience with human subjects informed the need for a pacing lead for cerclage implantation that addresses shortcomings in deliverability and nonselective stimulation. Therefore, we developed a purpose-built quadripolar cerclage pacing lead (Tau-PNU Co, Yangsan, Korea) for preclinical testing (Figure 2). The lead has a tapering shaft, from the 2.4F distal part with a tapered tip design to 4F near the septal vein, and has a monorail guidewire lumen at the distal end (otherwise known as single-operator exchange) to accommodate a stiffening stylet to enhance pushability. In addition, the cerclage pacing lead has a narrow 3-mm interelectrode spacing instead of 20 mm to enhance the selectivity of His-bundle stimulation vs surrounding nonspecific myocardium.The 6 pacing vectors selectable among the 4 electrodes are expected to allow pacing optimization without lead repositioning.

Figure 2.

Conceptual image of cerclage pacing using the new quadripolar lead (A) and profiles of newly developed custom-made quadripolar lead (B).

Animal study

Animal experiments were approved by the Pusan National University Yangsan Hospital animal care and use committee. Seven naïve farm swine (weight 44.6 ±13.3 kg) underwent nonsurvival testing under general anesthesia. Bipolar cerclage lead prototypes with varying interpolar spacing were tested during repositioning along multiple cerclage-trajectory septal locations. Paced and sinus QRS duration were measured before the final quadripolar 3-mm paced lead was tested. Activation mapping was tested during sinus, cerclage, and RV apical pacing in 1 animal (CARTO3, Biosense Webster, Diamond Bar, CA). After euthanasia, the distance between the septal leads and the His bundle was assessed by a cardiac pathologist (JWS).

Statistical analysis

Data were analyzed “as treated” among subjects who underwent attempted cerclage pacing. Categorical variables are presented as frequency with percentage and compared using the x² or Fisher exact test. Continuous variables are given as median and interquartile range. Paired and unpaired comparisons between treatment groups were evaluated by the Wilcoxon rank sum test. All intervals were obtained from a median value of 10 consecutive measurements. All analyses were performed using R programming language version 3.3.1 (www.r-project.org).

Results

Clinical protocol

During the study period 2016–2018, 7 subjects were screened at 2 participating institutions. Two were excluded for inadequate septal vein visualization by computed tomographic (CT) or coronary sinus venography. Five subjects were enrolled and underwent cerclage pacing at tempts. Baseline and procedure characteristics are listed in Table 1.

Table 1.

Baseline patient and implant characteristics of the study participants

Characteristics	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5
Age (y)	52	63	67	70	60
Sex	Female	Female	Female	Female	Female
Pacemaker indication	CAVB	SSS	SSS	SSS	CRT
Comorbidities	MVR, TVR	TAP, maze	TVR	TVR, sutureless AVR	DCMP
Left ventricular ejection fraction (%)	37	62	67	62	25
Baseline QRS width (ms)	126	98	102	177 (LBBB)	212
During procedure
Procedure time (min)	177	185	267	230	227
Fluoroscopic time (min)	45	57	68	40	70
Contrast use (mL)	350	200	350	230	150
Radiation dose (μGy/m2)	20615	4957	20853	8000	18938
RV snare required	Yes	Yes	No	No	No
At baseline
Paced QRS width (ms)	179	124	122	161	156
QRSpaced/QRSnative ratio	1.42	1.27	1.20	0.91	0.74
Sensed R wave (V)	15 mV	>12.0 mV	9.7 mV	>12.0 mV	4.6 mV
Ventricular threshold (mV)	0.6	1.1	0.8	1.4	0.75
Pulse width (ms)	0.4	0.4	0.4	0.4	0.4
Lead impedance (Ω)	800	816	665	840	340
At 6 months
Paced QRS width (ms)	180	127	124	160	—
QRSpaced/QRSnative ratio	1.43	1.30	1.22	0.90	—
Sensed R wave (V)	19.1	7.5	11.9	>12.0 mV	—
Ventricular threshold (mV)	0.8	1.5	1.25	1.25	—
Pulse width (ms)	0.4	0.4	0.4	0.4	—
Lead impedance (Ω)	839	703	532	589	—

AVR = aortic valve replacement; CAVB = complete atrioventricular block; CRT = cardiac resynchronization therapy; DCMP = dilated cardiomyopathy; LBBB = left bundle branch block; MVR = mitral valve replacement; RV = right ventricle; SSS = sick sinus syndrome; TAP = tricuspid annuloplasty; TVR = tricuspid valve replacement.

Cerclage-trajectory lead implantation and pacing was successful in all 5 subjects, but intraseptal pacing was achieved in only 4 of 5 subjects. Procedure and fluoroscopy times were lengthy: 227 minutes (177–267) and 57 minutes (40–70), respectively. Snare countertraction was required to deliver the lead over the guidewire in 3 subjects. In the first subject, the bipolar EASYTRAK lead did not advance beyond the mouth of the septal perforator vein into the septal myocardium but nevertheless was implanted abutting the septum, yielding a wide paced QRS complex. In the subsequent 4 subjects, an Attain Ability 4196 lead was used in the desired intraseptal position. Immediate ventricular sensed R wave, pacing threshold, and lead impedance were 15 mV (4.6–15), 0.8 V (0.6–1., at 0.4-ms pulse width), and 800 Ω (340–840), respectively. QRS duration was around the normal range (patients 2 and 3) or narrower than baseline widening (patients 4 and 5) in all but the first subject having the suboptimal lead implantation site (Figure 3). Follow-up echocardiography data were available for 3 patients with successful cerclage pacing, and all patients showed left ventricular ejection fraction preservation or improvement at 12-month follow-up (37% to 52% in patient 1; 62% to 58% in patient 2; 61% to 58% in patient 4).

Figure 3.

Successful cerclage pacing (A: patient 3) resulted in relatively narrow paced QRS complex (122 ms), which was close to the native QRS complex (102 ms). B: In patient 4 with intrinsic left bundle branch block, cerclage pacing resulted in only slight narrowing of QRS complex (171 to 161 ms) with remaining left bundle branch block configuration.

Lead dislodgment occurred in 1 subject (patient 5) on postprocedure day 1, which was attributed to inadequate lead redundancy. The patient was converted to conventional cardiac resynchronization therapy. Lead position and electrophysiological parameters were stable through 6 months in all other patients. Because of the lead dislodgment in 1 patient and the lengthy procedure time in all patients, the clinical protocol was suspended. We turned our attention to developing and testing a purpose-built cerclage pacing lead.

Animal protocol

We investigated the relationship between interelectrode distance and QRS duration when implanted in a cerclage trajectory (Figure 4). Commercial bipolar leads with 20-mm interelectrode spacing exhibited paced/native QRS ratio of 1.37 ± 0.17, compared with narrower (≤10 mm) interelectrode spacing exhibiting paced/native QRS ratio of 1.05 ± 0.11 (P <.001).

Figure 4.

Effect of interelectrode spacing on QRS morphology during cerclage pacing in swine. Commercial bipolar lead with 20-mm interelectrode distance (A) compared with a prototype 3-mm interelectrode distance (B) shown on fluoroscopy and surface electrocardiography. C: Comparison of paced/native QRS duration during cerclage pacing in swine using commercial 20-mm interelectrode spacing vs custom 3-mm interelectrode spacing, compared with right ventricular apical pacing.

We tested a purpose-built monorail + stylet quadripolar cerclage pacing lead with 3-mm interelectrode spacing. The low-profile (2.4F–4F) leads easily traversed interventricular septal myocardium over the guidewire without requiring snare countertraction. During pacing at multiple intraseptal locations adjacent to the His Bundle along the cerclage tractor, QRS duration was similar to sinus rhythm (Figure 5). Cerclage pacing exhibited similar activation patterns to sinus rhythm (Figure 6), in contrast to RV apical pacing. Necropsy examination of explanted hearts showed proximity of the purpose-built quadripolar electrode to the His bundle, when implanted along the cerclage trajectory (Figure 5).

Figure 5.

Anatomical and electrocardiographic relationship between cerclage pacing lead location with reference to the His bundle. The initial cerclage pacing lead deployed at wire 1 location (A) resulted in a narrow paced QRS complex (B). Pacing from adjacent branch (C: wire 2) also generated a narrow QRS (D). Explanted heart viewed from the right atrium showed close proximity of both wire 1 and wire 2 to the His bundle (F). All the vectors of quadripolar leads showed similar QRS width.

Figure 6.

Anteroposterior and posteroanterior projections of left ventricular (LV) activation mapping during sinus rhythm (A), cerclage pacing (B), and right ventricular apical pacing (C) in naïve swine. Cerclage pacing and sinus rhythm both exhibited superior to inferior axis and short LV activation time, in contrast to right ventricular apical pacing. Yellow dots denote region with His electrogram.

Discussion

This prospective cohort study demonstrates the initial feasibility and safety of a novel technique of parahisian septal pacing. In this technique of cerclage pacing, the coronary sinus pacemaker lead is implanted in the interventricular septum via the first septal perforator branch of the anterior interventricular vein in proximity to the His bundle. In this cohort, cerclage pacing in the intended position yielded narrow QRS and acceptable pacing threshold, sensing, and lead impedance. The clinical protocol was terminated prematurely because of the unsuitability of the bulky and very poorly trackable commercial coronary vein leads for this implant position, leading to prolonged procedures, a case of malposition, and lead dislodgment. However, the clinical experience informed the development of a novel purpose-built cerclage pacing lead with 3 key characteristics: (1) low-profile (2.4F–4 F) monorail design with a tapered tip to traverse interventricular septal myocardium easily; (2) narrow (3-mm) interelectrode spacing; and (3) a quadripolar design to allow selectable stimulation of His or parahisian targets. A serpentine configuration is expected to prevent lead dislodgment. In animals, the purpose-built pacing lead achieved easy deliverability and narrow QRS.

Cerclage pacing leads are implanted deep into the interventricular septal myocardium rather than on the epicardial surface. This procedure targets pacing from the parahisian area and demonstrates its potential in patients with intact intrinsic conduction systems (patient 3 in Figure 3). However, whether cerclage pacing could recruit the His– Purkinje system (HPS) in patients with conduction system disorders, such as atrioventricular nodal block, intrahisian block, or bundle branch block, remains unknown. In patient 4, although QRS duration was only slightly shortened by cerclage pacing, left bundle branch block configuration remained, indicating partial recruitment of left-sided HPS (Figure 3). We speculate that diffuse distal conduction or intramyocardial disease might have prevented effective HPS capture with cerclage pacing in the proximal HPS.¹⁶ More data on cerclage pacing in patients with conduction system disease would be required to establish the advantage of the procedure.

Compared with coronary venous and epicardial leads, cerclage pacing avoids tricuspid valve traversal and its contribution to tricuspid valve regurgitation.^17–19 Compared with RV apical leads, cerclage positioning can achieve physiological pacing and avert pacing-induced ventricular dysfunction.^5,10 As observed by others, we could achieve narrow QRS pacing at low thresholds along a wide range of intraseptal parahisian tissue locations, even without direct His-bundle penetration.^20,21 Unlike the current Medtronic 3830 lead, which is implanted in the His region using screws, our lead system slides into the septal perforator branch of the great cardiac vein. Our lead system has unique advantages over the 3830 lead, as it has a lower pacing threshold and avoids tissue damage in the His–bundle region. Narrow interelectrode distances with a quadripolar design provides multiple options for electrode pair pacing to maximally recruit the His–Purkinje conduction system. However, as the body of the lead is located in the intramyocardial tissue under constant mechanical stress, long-term durability of the lead should be assured before clinical application. The possibility of lead dislodgment is another issue, as our new lead is narrower than previous leads, so additional mechanisms for lead stability should be developed. Recently, Huang et al²² demonstrated direct LBB pacing via an RV approach, achieving physiological pacing with an acceptable pacing threshold. This technique is based on the same concept as ours, but the approaches come from opposite directions. LBB pacing and cerclage pacing could serve complementary roles in permanent His-bundle pacing.

There are important limitations of cerclage pacing in its current stage. The procedure and fluoroscopy times are extremely long, and the procedures required snare countertraction when using commercial large-diameter and poorly trackable coronary venous leads. Use of contrast media for vein selection and radiation exposure are other drawbacks of the cerclage pacing procedure. Only 5 of 7 patients have adequate visualization of the septal vein with CT imaging. More imaging data on septal vein anatomy (such as angiography or CT) are required to make the procedure easier. Previous operation on the tricuspid valve in most cases has made use of the snare much more difficult or even impossible. Therefore, the clinical protocol was terminated prematurely. Nevertheless, cerclage pacing demonstrates potential as a viable option as a complement or alternative to His-bundle pacing or LBB pacing as the procedure continues to evolve. Our purpose-built cerclage pacing lead is designed to over-come this limitation. Naïve swine may not directly model humans with conduction system disease.

Conclusion

Cerclage parahisian septal pacing is a feasible approach to achieve physiological pacing. Although available commercial pacing leads are ill-suited for cerclage pacing, this early human experience informed the development of a purpose-built prototype that is easy to implant. Cerclage pacing, using a dedicated catheter, should be investigated as an alternative to direct right-sided approaches to His-bundle pacing.