Left bundle branch pacing cardiac resynchronization therapy vs biventricular pacing cardiac resynchronization therapy–time to write a requiem for biventricular pacing-cardiac resynchronization therapy

Abstract

Cardiac resynchronization therapy (CRT) reduces heart failure (HF) hospitalizations and all-cause mortality in patients with HF with reduced ejection fraction with left bundle branch (LBB) block. Biventricular pacing (BVP) is considered the gold standard for achieving CRT; however, approximately 30%–40% of patients do not respond to BVP-CRT. Recent studies have demonstrated that LBB pacing (LBBP) produces remarkable results in CRT. In this meta-analysis, LBBP-CRT showed better outcomes than conventional BVP-CRT, including greater QRS duration reduction and left ventricular ejection fraction improvement, along with consistently lower pacing thresholds on follow-up. Additionally, there was a greater reduction in New York Heart Association class and brain natriuretic peptide levels. This study contributes to the growing body of encouraging data on LBBP-CRT from recent years. With ongoing technological advancements and increasing operator expertise, the day may not be far when LBBP-CRT becomes the standard of care rather than the exception.

Key Words: Heart failure; Left bundle branch block area pacing; Narrow QRS; New York Heart Association class; Left ventricular ejection fraction

Core Tip: Cardiac resynchronization therapy (CRT) reduces heart failure (HF) hospitalization and all-cause mortality in patients of HF with reduced ejection fraction with left bundle branch (LBB) block (LBBB). However approximately 30%–40% of patients do not respond to Biventricular pacing (BVP)-CRT. Some recent studies have shown that LBB pacing (LBBP) provides an alternative for CRT. In the current study, Yasmin et al demonstrate that LBBP-CRT is better than conventional BVP-CRT in terms of QRS duration reduction as well as left ventricular ejection fraction improvement with consistent lower pacing thresholds on follow-up. The present study adds to a body of emerging encouraging data for LBBB-CRT in the past few years. With evolving technological modification and operator experience, the day might not be far when LBBB-CRT becomes the convention rather than exception.

TO THE EDITOR

Cardiac resynchronization therapy (CRT) is a well-established, guideline-recommended intervention for patients with persistent heart failure (HF)-related symptoms, defined as New York Heart Association (NYHA) class II–IV, despite optimal medical therapy. It is indicated for those with a left ventricular (LV) ejection fraction (LVEF) ≤ 35% and left bundle branch (LBB) block (LBBB) or ≤ 40% in patients with an anticipated right ventricular (RV) pacing burden ≥ 40%[1].

Conventional CRT involves placing one lead over the LV epicardium via a cardiac vein and another in the RV apex, enabling biventricular pacing (BVP). Large multicenter clinical trials have demonstrated that CRT significantly reduces HF-related hospitalizations and all-cause mortality, particularly in patients with LVEF < 35%, sinus rhythm, LBBB morphology, and a QRS duration > 150 ms[2,3]. However, approximately 30%–40% of patients fail to exhibit any clinical or echocardiographic improvement with CRT, classifying them as "non-responders"[4]. This may be attributed to CRT's focus on improving electrical dyssynchrony by shortening QRS duration through earlier activation of the LV lateral wall, without correcting LBBB itself. BVP has other limitations, including procedural challenges, phrenic nerve stimulation (in nearly 40% of patients), rising pacing thresholds, or loss of capture over time. These issues may arise due to inadequate venous anatomy for lead placement, proximity of suitable venous branches to the phrenic nerve, or a lack of adequate branches in optimal lateral areas. Consequently, the LV lead is often implanted in suboptimal apical or anterior venous branches[5,6].

In the last decade, there has been growing interest in conduction system pacing (CSP). Although initial CSP studies primarily involved patients without HF, direct stimulation of the conduction system could play a pivotal role in addressing LV dyssynchrony in patients requiring CRT[7]. The earliest approach to CSP was His bundle pacing (HBP), which enables physiological conduction system activation. However, HBP has limitations, including a steep learning curve, longer procedural and fluoroscopy times, higher pacing thresholds, and the need for a backup RV lead in some cases[8-10]. In the His-SYNC trial, 48% of patients experienced failure in distal conduction system activation, necessitating crossover to BVP[10]. Similarly, the His-Alternative trial reported a 28% failure rate for achieving His bundle corrective pacing, with higher pacing thresholds in successful cases than BVP[11].

More recently, LBB area (LBBA) pacing (LBBAP) has emerged as a viable alternative to traditional CRT. By stimulating the conduction system distal to the His bundle, LBBAP results in more homogeneous LV contraction and relaxation, potentially improving outcomes compared with conventional CRT strategies[12,13]. LBBAP can be performed using the Select Secure 3830 fixed-helix lead with a C315 HIS fixed-curve sheath (Medtronic Inc.) or an extendable/retractable helix lead such as the Solia S60 Lead with the Selectra 3D sheath (Biotronik SE and Co.) or the Tendril STS lead with the Agilis HisPro catheter (Abbott Inc.). In the 30^º right anterior oblique view, the ideal implant site is along an imaginary line one-third of the way from the aortic valve to the RV apex, approximately 1.5 cm distal to the tricuspid annulus. The sheath is advanced and rotated clockwise to cross tricuspid valve and enter RV. Then counterclockwise rotations provide advancements towards the RV basal septum. Unipolar pacing is used to identify the ideal lead fixation site, which is characterized by a paced QS complex with a notch in the nadir of lead V₁ ("W" pattern) and/or aVR/aVL discordance (e.g., taller R wave in lead II than lead III, negative aVR, and positive aVL). Once identified, the lead is secured to the RV septum with 1–2 rotations, followed by 4–5 additional clockwise rotations using a "pill-rolling" technique under fluoroscopy. As the lead penetrates the septum, impedance initially rises and then falls by approximately 100 Ω as it reaches the LBBA. Successful LBBAP is confirmed using established criteria[14].

LBBAP achieves successful LBBB capture in > 80% of cases, with shorter procedural and fluoroscopy times than HBP[15]. Various criteria for successful selective LBB are listed in Figure 1. Acceptable pacing thresholds are typically < 1.5 mV at a 0.5 ms pulse width.

Figure 1 The criteria for determining selective left bundle branch area capture during left bundle branch area pacing. The blue dotted line represents a hypothetical left bundle branch area pacing lead. LV: Left ventricular; LVEF: Left ventricular ejection fraction; NYHA: New York Heart Association.

Non-randomized studies have shown greater clinical and echocardiographic responses with LBB pacing (LBBP)-CRT than BVP[15]. However, comparisons of clinical efficacy between LBBP-CRT and optimized BVP (BVP-CRT) have not been well established. Two small randomized clinical trials (RCTs), LBB-RESYNC and LEVEL-AT, have compared LBBP-CRT and BVP-CRT[16,17]. In the LBB-RESYNC trial, 40 patients with non-ischemic cardiomyopathy and LBBB were randomized to LBBP-CRT or BVP-CRT, with a 6-month follow-up[16]. A crossover was allowed if anyone of them was unsuccessful. Crossovers occurred in 10% of patients receiving LBBP-CRT and 20% of those receiving BVP-CRT. LBBP-CRT demonstrated significantly greater improvement in LVEF (mean difference: 5.6%; 95%CI: 0.3–10.9; P = 0.039) and reductions in LV end-systolic volume (24.97 mL; 95%CI: 49.58 to 0.36) and NT-pro brain natriuretic peptide (BNP) levels (1071.80 pg/mL; 95%CI: 2099.40-44.20) than BVP-CRT, with comparable changes in NYHA class, 6-minute walk distance, QRS duration, and rates of CRT response.

LEVEL-AT (LV Activation Time Shortening with CSP vs Biventricular Resynchronization Therapy) is another RCT that was conducted to test noninferiority[17]. It included 70 patients with indication for CRT randomized in 1:1 ratio to BVP or CSP, who were followed up for 6 months. Crossover was allowed if the primary allocation procedure failed. LV activation time (LVAT), measured using electrocardiographic imaging, was the primary endpoint, while the secondary endpoints included LV reverse remodeling and the combined endpoint of HF hospitalization (HFH) or death at the 6-month follow-up. There was a similar decrease in LVAT between CSP and BVP (28 ms ± 26 ms vs 21 ms ± 20 ms, respectively; mean difference: 6.8 ms; 95%CI: –18.3 to 4.6; P < 0.001 for noninferiority). Similarly, there was a comparable change in LV end-systolic volume (37 mL ± 59 mL for CSP vs 30 mL ± 41 mL for BVP; mean difference: 8 mL; 95%CI: –33 to 17; P = 0.04 for noninferiority) and similar rates of mortality or HFH (2.9% vs 11.4%, respectively; P = 0.002 for noninferiority). Thus, in these RCTs, the outcomes of BVP and CSP were comparable, and LBBP-CRT was found to be superior to BVP-CRT.

Yasmin et al[18] published a meta-analysis in a journal comparing LBBP-CRT with conventional BVP-CRT. It included 389 participants (159 in the LBBP-CRT group and 230 in the BVP-CRT group) from six studies (one RCT and five observational studies) with a median follow-up of nine months. The primary outcome was QRS duration (QRSd), while secondary outcomes included pacing thresholds, improvement in LVEF, and reductions in LV systolic (LVESD) and diastolic (LVEDD) dimensions. The success rate of LBBP-CRT was 91.1%, and 50.3% of the population consisted of males, with a mean age of 64 years ± 4 years. The reduction in QRSd was greater with LBBP-CRT than with BVP-CRT (115.4 ms vs 138.0 ms, P < 0.00001). Similarly, LBBP-CRT resulted in a significantly higher improvement in LVEF (LBBP-CRT mean: 43.8 vs BVP-CRT mean: 37.6; 95%CI: 4.48-8.97, P < 0.00001). There was also a significant reduction in LVEDD (LBBP-CRT mean: 55.9 vs BVP-CRT mean: 60.6; 95%CI: -7.21 to -3.03, P < 0.00001) and LVESD (LBBP-CRT mean: 42.0 vs BVP-CRT mean: 47.3; 95%CI: -8.80 to -2.35, P < 0.00007) compared with BVP-CRT. In addition, LBBP-CRT led to greater reductions in NYHA class and BNP levels [LBBP-CRT mean: 1.3 vs BVP-CRT mean: 1.8; 95%CI: -0.73 to -0.21, P < 0.00003, and LBBP-CRT mean: 311.3 vs BVP-CRT mean: 1145.3; standardized mean difference (MD) = -0.66, 95%CI: -0.96 to -0.35, P < 0.00001, respectively]. The pacing thresholds were also significantly lower with LBBP-CRT (LBBP-CRT mean: 0.69 vs BVP-CRT mean: 1.24; MD = -0.56, 95%CI: -0.69 to -0.43, P < 0.00001). It is well established that narrower QRSd is associated with better electrical and mechanical resynchronization and improved clinical outcomes. The results of this meta-analysis clearly demonstrate that LBBP-CRT is an effective alternative to BVP-CRT, and it may even be superior. Figure 2 illustrates the expected benefits of LBBP-CRT compared with BVP-CRT based on the present meta-analysis. These benefits are expected to ultimately improve outcomes in HF patients, although hard endpoints were not formally tested in this meta-analysis. The study's small sample size of fewer than 400 participants is a limitation. As discussed by the authors, the short follow-up period and geographic restriction to the People's Republic of China are additional limitations.

Figure 2 The potential advantages of left bundle branch area pacing based cardiac resynchronization therapy over biventricular cardiac resynchronization therapy based on randomized clinical trial’s and meta-analyses. LBBP: Left bundle branch pacing; LVSP: Left ventricular septal pacing.

Previous meta-analyses have also shown the efficacy of LBBP-CRT over BVP-CRT. A meta-analysis by Tan et al[19] included 527 patients from eight nonrandomized studies. In studies with a BVP comparison group, patients with LBBP showed greater reductions in paced QRS duration (mean difference: 27.91 ms; 95%CI: 22.33–33.50) and NYHA class (MD = 0.59; 95%CI: 0.28–0.90), along with higher LVEF (MD = 6.77%; 95%CI: 3.84–9.71). Patients with underlying LBBB benefited more than those in the non-LBBB subset[19]. Another meta-analysis by Parlavecchio et al[20] demonstrated that LBBP-CRT significantly reduced HFH and pacing thresholds (MD = -0.60) at follow-up. This larger meta-analysis included more than 1000 patients from ten studies. There was also significant improvement in LVEF, NYHA class, and response rates to CRT (88.5% vs 72.5%; rate ratio = 1.19; 95%CI: 1.07–1.32) compared with BVP-CRT. Similarly, a meta-analysis by Yousaf et al[21] revealed greater reductions in QRS duration and NYHA class with LBBP-CRT. There was also greater improvement in LVEF and other echocardiographic parameters. Furthermore, LBBP-CRT led to better clinical outcomes, including fewer HFH and improved survival. This is likely the largest study to date, involving approximately 3000 patients from 12 studies.

BENEFITS OF LBBP-CRT OVER BIVENTRICULAR-CRT

The success rate of LBBP-CRT varies from 81.1% to 98.1% in individuals meeting CRT indications, according to various studies. The failure rate is higher in biventricular-CRT (BiV-CRT) due to several factors, such as unfavorable coronary sinus anatomy or stenosis. Additionally, some patients may experience phrenic nerve stimulation, and about 30%–40% of patients do not respond to BiV-CRT. HBP also has limitations, including poor sensing amplitude, a steep learning curve, increasing pacing thresholds over time, and the inability to implant leads beyond block sites. Consequently, LBBP-CRT offers a higher implant success rate than HBP-CRT and BiV-CRT. Moreover, it restores physiological electrical synchrony, unlike traditional BiV-CRT, which achieves synchronization through variable fusion of wavefronts propagating from the endocardium and epicardium[22].

LIMITATIONS AND CHALLENGES OF LBBP-CRT

Although LBBP-CRT demonstrates higher implantation success, it may not be suitable for all HF patients requiring CRT. Some challenges include difficulty in screwing the pacing lead into the interventricular septum (IVS). This could result from thickened IVS, myocardial fibrosis, interference by the septal tricuspid leaflet when implantation is too close to the tricuspid annulus, or non-coaxial alignment of the pacing lead and sheath. Another challenge could be failure to capture the LBB, which may occur due to myocardial fibrosis or when LBBB sites are distal to the LBB. Similarly, another challenge is the inability to correct electromechanical dyssynchrony in nonspecific intraventricular conduction delay (IVCD). In HF patients with nonspecific IVCD, the conduction abnormality occurs within the ventricular wall, with normal conduction in the His bundle and major bundle branches. LBBP is ineffective in such cases. Furthermore, complications like IVS perforation are particularly seen in patients with thin IVS, underscoring the need to measure septal thickness before the procedure[22]. BiV-CRT has been shown to be cost-effective in several analyses[23,24]. However, as LBBP-CRT is a novel concept with evolving techniques, cost-effectiveness data for this modality is yet to emerge.

LONG-TERM SUCCESS OF LBBP-CRT OVER BIV-CRT

A large, multicenter, international, observational retrospective study by Vijayaraman et al[25] demonstrated that LBBP-CRT significantly reduced the primary composite endpoint of all-cause mortality or HFH compared with BiV-CRT. It also resulted in greater narrowing of QRS duration and better echocardiographic responses in all patients, including those with LBBB[25].

Several ongoing studies aim to further investigate the efficacy of LBBP-CRT. The BATTLE study (NCT06061627) is a prospective, multicenter RCT enrolling patients with chronic HF and intraventricular conduction block. It compares LBBP-CRT to conventional BiV-CRT, with a follow-up of at least 6 months. The primary endpoint is improvement in LVEF. The LeCaRT trial (NCT05365568) is an RCT involving 170 patients comparing LBBP-CRT and BiV-CRT, focusing on clinical outcomes such as the composite of death and HFH or worsening HF. Left vs left study (NCT05650658) is a large RCT with over 2000 patients and more than 5 years of follow-up. It compares HBP/LBBP-CRT to BiV-CRT, assessing outcomes like all-cause mortality and HFH.

Not all HF patients with a wide QRS will benefit from LBBP-CRT. For instance, patients with nonspecific IVCD often have underlying myocardial disease (e.g., ischemic heart disease, hypertrophic cardiomyopathy, or post-myocarditis with extensive scar tissue), where LBBP alone may be insufficient. These cases often require BiV-CRT, with or without LBBP, for better outcomes. Findings from Vijayaraman et al[26] and observations at our center suggest that patient selection should prioritize these factors, as outlined in Table 1.

Table 1 Factors that patients prioritize.

Left bundle branch-CRT	Biventricular-CRT
Typical LBBB with NICMP	LBBB with ischemic cardiomyopathy with septal scar
Pacing induced cardiomyopathy	nonspecific interventricular conduction delay with heart failure with reduced ejection fraction
Masquerading bundle branch block with NICMP	Structural heart disease with septal scar e.g. cardiac sarcoidosis, myocardial tuberculosis, hypertrophic cardiomyopathy, inherited cardiomyopathy

CRT: Cardiac resynchronization therapy; LBBB: Left bundle branch block; NICMP: Nonischemic cardiomyopathy.

CONCLUSION

The present meta-analysis concludes that, compared with BVP-CRT, LBBP-CRT is both safe and effective. A narrower QRS duration leads to better mechanical synchronization of the ventricles, and LBBP achieves this with lower pacing thresholds. Consequently, LBBP-CRT is associated with improved LV function. The meta-analysis confirms that CRT via LBBP provides more effective electrical and mechanical resynchronization. It also highlights the significantly lower pacing thresholds achieved at implantation and follow-up with LBBP-CRT. All six studies analyzed reported better improvements in LVEF, reductions in LVESD and LVEDD, greater decreases in NYHA class and BNP levels during 6 months of follow-up.