Left bundle branch area pacing (LBBAP)-optimized implantable cardiac defibrillator (ICD), also known as LOT-ICD implantation, has recently gained traction.¹ Herein, an additional ventricular pace-sense lead delivers LBBAP while an older generation right ventricular (RV) DF-1 ICD lead (with its own truncating IS-1 pace-sense lead pin capped) is utilized for defibrillation. While this new strategy of cardiac resynchronization therapy (CRT) is a promising cost-effective alternative to conventional biventricular ICD (additional coronary sinus lead), some of its limitations are noteworthy. Firstly, the redundant IS-1 pace-sense lead renders the system potentially incompatible for magnetic resonance imaging (MRI). Secondly, the patient is encumbered with additional hardware burden from the LBBAP pace-sense lead and procedural duration.

Given this background, direct implantation of an ICD lead in the LBBA is actively explored for the dual purpose of CRT and defibrillation. This is presumptively viewed as an attractive option to retain MRI conditionality and eliminate hardware related challenges. However, the bulkier profile of most ICD leads, and lack of dedicated delivery sheaths have thus far precluded widespread applicability of this novel paradigm.

Hitherto, only three in-human published case reports/series have addressed this issue. A proof-of-concept case series (n=5, successful in 3) demonstrated the feasibility of a temporary LBBA implant using 7.3 Fr DF-1 Reliance 4-Front 64 cm type 0693, ICD lead (Boston Scientific^TM, Marlborough, MA, USA) with different pre-shaped/modified sheaths.² Subsequently, in an isolated case, a 6.8 Fr Durata 7122Q (DF-4) ICD lead (Abbott, Chicago, IL, USA) was permanently implanted utilizing a 10.5 Fr (Inner diameter 7 Fr) deflectable AgilisHisPro^TM (Abbott, Chicago, IL, USA) mapping catheter.³ Recently, successful permanent LBBA-ICD implant predominantly using a DF-1 lead was reported in 5/8 (62.5%) attempted patients. However, there was limited information regarding procedural caveats.⁴

Currently, dedicated proprietary slender profiled ICD leads are already in early stages of clinical testing. Their requisite technical compatibility with existing thinner delivery systems can conceivably allow them to be combobulated as dedicated LBBA implants. Nevertheless, given the paucity of data as well as a few unknowns, this prototypical concept warrants a detailed scientific evaluation regarding its pros and cons. In this context, we have performed a single-center exploratory study of implanting permanent transvenous DF-4 ICD lead in the LBBA (LBBA-ICD) in patients with a primary/secondary prevention indication.⁵ We attempted LBBA ICD implant using in 12 patients (median age 67.5 years, male 91.7%, median LVEF 30%, median septal thickness 9 mm, median QRS duration 140 ms, class I CRT indication 58.3%, primary prevention ICD indication in 75%). A successful permanent LBBA ICD implant with adequate pacing and sensing parameters was achieved in 9 (75%) patients (8 males, ischemic cardiomyopathy in 5, primary prevention ICD in 7, class I CRT indication in 5). Our goal was to validate the implant only after successful defibrillation of induced VF. This was fulfilled in most patients (7/9). In a 69-year-old male (ischemic cardiomyopathy, secondary prevention indication, no pacing/CRT indication) with appropriate sensing and pacing parameters, VF could not be induced despite multiple attempts. In another 80-year-old female (ischemic cardiomyopathy, primary prevention/class I CRT indication), there was repeated self-termination of induced VF which precluded defibrillation.

With this in the background, we aim to test the clinical feasibility of DF-1/DF-4 LBBA-ICD using a including success rate, procedural characteristics/complications, effectiveness of defibrillation and other relevant parameters acutely and on a longer-term follow-up of at least 1 year from a multicenter investigator-initiated study.

Institutional Review Board (IRB) Approval: Our study protocol on adults (³18 years old) will seek approval by the local IRB and informed consent will be obtained from all patients. Proprietary leads and delivery systems already in clinical testing will be employed. Given the established institutional track record in conduction system pacing using such hardware in all participating centers, the implanting electrophysiologists have concurred that we were well poised in our quaternary care center for safe and effective implementation of this research protocol.⁶ This project is deemed to incur no additional procedural risks compared to a standard LBBAP implant with minor acceptable modifications to our clinical technique.

            After each implant, there was to be a systematic internal review by the electrophysiology section and any adverse outcomes were to be reported to the hierarchy. A three-month outpatient evaluation was mandated for all subjects to review any short-term concerns with subsequent ongoing follow-up.

            The implants will be geared towards patients with guideline-approved pacing/CRT indications.⁵ If there are no safety or efficacy concerns, those without such indications can also be included. The participating subjects will be counseled in accordance with the Belmont principles and provided a full disclosure of the study rationale. They will have the authority to opt in or out.

We shall include all consecutive consenting adult patients (³18 years of age) without a prior cardiac device who met primary or secondary prevention ICD indication. As noted previously, those without an additional pacing indication will also be eligible. Patients who need an ICD upgrade will also be excluded. Patient enrollment shall commence from August 1^st, 2024.

Our implant protocol will utilize the 7 Fr Durata 7122Q (DF-4)/ 7120 (DF-1) ICD lead (Abbott, Chicago, IL, USA) with a non-deflectable dual curve 9 Fr (Inner diameter 7.3 Fr) CPS Locator 3D– Medium DS2C200 (Abbott, Chicago, IL, USA) delivery sheath. A detailed pictorial depiction of the hardware with tabulated specifications is showcased in Figure 1. Here, we shall closely our previously published technique of implanting stylet-driven pace-sense leads at the LBBA.⁶

Briefly, the RV septal endocardial aspect will be pace-mapped in unipolar configuration with the distal tip electrode for an acceptable target site (‘W’ QRS configuration in V1 with a precordial transition no later than V5) in right anterior oblique (RAO) 20° fluoroscopic projection. The decision on an anterior/superior versus posterior/inferior location will be based on ease of reach and presence of local viable myocardium (gauged by sensing and pacing parameters). The ICD lead will not be prepared ex-vivo. After achieving a presumed perpendicular orientation at the implant site in left anterior oblique (LAO) 30° view, we shall proceed further.  The stylet will be fully inserted and customary 12-20 clockwise turns given to the DF-4 connector pin. This will be simultaneously visualized on fluoroscopy to ensure full exposure of the helix. For the DF-1 counterpart, the helix-locking tool can be connected to its IS-1 pin.

Pursuant to this, the stylet will be gently withdrawn by 2-3 cm and cycles of 4-5 rotations on the lead body alternating with the same number will be applied to the DF-4 connector pin to avoid retraction of helix. Such maneuvers will not be required for a DF-1 lead and lead body rotations can be delivered in toto. Lead fixation would be guided by continuous monitoring of pacing parameters (unipolar tip pacing impedance, paced QRS morphology, and injury current) as per published literature.⁷ Lead related parameters will be recorded by achieving a unipolar configuration and under magnified fluoroscopic visualization. Periodic assessments of left ventricular activation time (Lead V6 QRS), visualization of a left bundle/fascicular potential, and spontaneous ‘fixation’/’template’ beats will also guide lead fixation. A gradual transition from ‘W’ to an incomplete right bundle branch block QRS morphology (qR/Qr/rsR’/R) in lead V1 will be analyzed. The depth of penetration will be confirmed with unipolar ring capture in all patients. An RV angiogram is optional in selected patients to further adjudicate the above using a 6 Fr pigtail catheter (via femoral access in single and axillary access in dual chamber ICD, respectively).

LBBAP will be stratified as LBBP, left anterior/left posterior fascicular (LAF/LPF) pacing and left ventricular septal pacing (LVSP) as per established criteria.⁸ Additional challenges encountered during actual implant and modifications adopted to optimize outcomes are well explicated below. We will use established criteria to monitor for complications such as transeptal perforation.^7,8

The length of the defibrillator coil in the right atrium will be based on RAO 20° fluoroscopic measurement on a frozen frame timed to the electrocardiographic gated R wave peak corresponding to end-diastole. The plane of the tricuspid valve annulus is referenced along a line perpendicular to the long axis of the heart and running superiorly from the radiolucent fad-pad landmark correlating with the coronary sinus ostium. Our technique replete with requisite images is previously published.⁷

VF induction will be attempted in all cases after implant with device in the pocket, but prior to closure of incision. Low energy shock delivered on the T-wave shall be the default strategy. After a drive train of 6 ventricular paced beats at 600 ms, a 2 joule (J) shock is timed at 300 ms from the last paced impulse. If this is unsuccessful in inducing VF, direct current fibber protocol (burst pacing at S1-S1 30 ms; 7.5 V amplitude; 1.5 ms pulse width) will be attempted.

We will collate demographic, clinical, echocardiographic, and relevant procedure/device-related data in accordance with our previous publications.^6,7 Pertinent parameters will be recorded at baseline, implant, discharge, and at 3-month outpatient visit. The echocardiographic criteria including left ventricular ejection fraction (LVEF), interventricular septal thickness, severity of tricuspid valve regurgitation, tricuspid annular plane systolic excursion (TAPSE), used in the manuscript are based on standard guidelines.⁹

We shall use IBM SPSS Ver 23.0 software (SPSS Inc, Chicago, Illinois, USA) for statistical analysis. Categorical variables will be expressed as frequencies/percentages while continuous variables will be represented as median with 25^th-75^th interquartile range (IQR). Paired T test will be used to compare continuous variables involving the same subset of patients. Categorical data will be analyzed using Fisher’s exact test. A p value £ 0.05 was deemed to be statistically significant.

Procedural Caveats and Considerations from our Initial Exploratory Study

Pertinent procedural caveats concordant with our learning curve merit a thorough deliberation. In the absence of dedicated hardware, we improvised by utilizing a ‘make-shift’ arrangement of available proprietary tools. Practical difficulties were encountered that required modifications in our implant techniques compared to conventional LBBAP.

Firstly, the inner diameter of the CPS locater 3D sheath (2.44 mm) adept for the corresponding proprietary pace-sense lead could barely accommodate the thicker 7 Fr Durata lead (maximum diameter of 2.34 mm). These specifications are illustrated. Consequently, lead body rotations were rendered slower, and a continuous note of helix retraction was mandated. The helix locking tool specific for the conventional IS-1 pace-sense lead did not fit the DF-4 ICD pin.

Secondly, the thicker coiled portion of the ICD lead precluded its easy retraction once exposed beyond the distal end of the sheath. This was particularly relevant if the sheath with the lead accidently retracted into the right atrium during the initial attempt and required swift repositioning. Herein, we were severely limited in our ability to repeatedly push/pull the body of the ICD coil back and forth across the tricuspid annulus.  In one such case, the lead could not be easily pulled back into the distal end of the sheath resulting in its significant deformation from a grazing effect. Thus, the entire lead-sheath complex had to be withdrawn in toto.

Thirdly, the longer (65 cm) version of the DF-1/DF-4 Durata lead was chosen to facilitate easy slitting. Given the fact that even minor distortions of the sheath penalized the smooth passage of the lead through it, we constantly strived to avoid any bends or kinks in the former. The lead-sheath profile mismatch rendered the irrigation side-port rather redundant. Hence, we were unable to perform our two routine contrast injections to check the pre-screw septal orientation of the lead and its post-implant fixation depth.⁶ Therefore, we performed a single surrogate RV hand-injection angiogram with a pigtail catheter proximate to the implanted lead using a separate venous access (femoral vs. axillary for single and dual chamber ICDs, respectively). In lieu of contrast injection adjudication, we relied on our prior experience to gauge the septal orientation prior to its fixation.

The CPS locator 3D sheath has a stiffer profile than the C315His sheath (Medtronic Inc, Minneapolis, MN, USA). Despite this, we still encountered some degree of flaying of its primary and secondary curves when the Durata ICD lead (with stylet fully inserted) was introduced through it. Consequently, maneuvering the sheath to an anterior septal location was mostly impossible which contributed to frequent lead fixations in the posterior fascicular region. In this context, partial retraction of the stylet provided a marginal benefit for maneuvering anteriorly. For similar reasons, a robust perpendicular orientation to the septum as is conventional with a pace-sense lead could not be replicated here. This is reflected in a more negative median post-fixation angle q (-5 degrees) compared to our published experience with LBBAP (median q of +5 degrees).⁶

Pursuant to our learning curve, we re-configured our technique to provide additional slack to the implanted lead only after slitting the sheath. The snugly fit hardware (lead-sheath interaction) rendered our usual practice of slack with sheath in place rather defunct due to associated risk of micro-dislodgement. The dedicated slitting tool for the CPS locator 3D sheath is equipped with a long-grooved extension for its insertion alongside lead in the proximal hub. However, this tool was not useful in the setting of a tight insertion space between the thicker ICD lead and sheath. The proprietary slitting tool for coronary sinus sheath was found to be a better fit in our scenario. Nevertheless, the groove for housing the lead during slitting had to be still manually enlarged. 

Finally, the more proximal location of the implanted LBBA ICD lead on the septum contributed to a significant redundancy of the defibrillation coil within the right atrium. This fact was thoroughly deliberated with attempted induction of VF and successful defibrillation as articulated previously.

Our limited yet exploratory case series (n=12) with short-term follow-up has previously demonstrated the feasibility, reasonable success rate (75%), and adequate safety of a DF-4 LBBA-ICD implant in an experienced setting while concurrently acknowledging pertinent procedural limitations. Importantly, far-field atrial oversensing was not noted on the RV channel (programmed true bipolar with tip to ring sensing). The results here are comparable to a recently published smaller study (n=8 with 62.5% success) using mostly a DF-1 lead (Durata 7 Fr, Abbott, Chicago, IL, USA). The authors reported a mean short-term follow-up of 3.8 ± 2.2 months with less emphasis on procedural caveats and without parameters of RV function.⁴

We, in our case series of 12 patients, encountered a significant learning curve replete with unique challenges which we circum-navigated. This was despite using a slender profile ICD lead (7 Fr Durata; Abbott, Chicago, IL, USA) with the current proprietary tools indicated for similar caliber IS-1 pace-sense leads adept for LBBAP. The proverbial elephant in the room is the thickened coiled portion of the ICD lead which proved to be a very snug fit with the sheath. There was neither any allowance for easy maneuverability nor any space to inject contrast in the side port rendering it obsolete. Salient procedural caveats explicated previously are summarized in brevity as follows.

            There was limited margin for error with regards to redeployment of the ICD lead in a different area once the coil was prolapsed through the sheath. Any subsequent retraction into the sheath was not only arduous but also perilous to the hardware integrity leading to its excess wastage. Consequently, procedural/fluoroscopic times were longer than anticipated. Separate vascular access was required for performing RV angiograms to adjudicate the lead orientation in septum because of defunct side port. There was a significant bias towards a more posterior implant owing to sheath-lead alignment issues as noted previously. The physiognomy of the lead in this setting enforced an obligatory redundancy of the defibrillation coil into the right atrium. Fortunately, this did not adversely impact successful defibrillation or echocardiographic parameters such as tricuspid valve regurgitation or TAPSE at least in the short term. 

            Thus far, our field has progressed from CRT-D through conduction system pacing optimized ICD implants in an effort towards calibrating the conjoint need for cardiac resynchronization and defibrillation in selected patients. Therefore, in a broad sense, the paradigm of LBBA ICD is envisioned as the logical continuum in this quest. This multicenter collaborative study is construed as a nascent attempt to innovate as well as cultivate a more systematic and broader approach towards fine-tuning this technology.