Left bundle branch area pacing in mildly reduced heart failure: A systematic literature review and meta‐analysis

Abstract Cardiac resynchronization therapy (CRT) strategy for heart failure with mildly reduced ejection fraction (HFmrEF) is controversial. Left bundle branch area pacing (LBBAP) is an emerging pacing modality and an alternative option to CRT. This analysis aimed to perform a systematic review of the literature and meta‐analysis on the impact of the LBBAP strategy in HFmrEF, with left ventricular ejection fraction (LVEF) between 35% and 50%. PubMed, Embase, and Cochrane Library were searched for full‐text articles on LBBAP from inception to July 17, 2022. The outcomes of interest were QRS duration and LVEF at baseline and follow‐up in mid‐range heart failure. Data were extracted and summarized. A random‐effect model incorporating the potential heterogeneity was used to synthesize the results. Out of 1065 articles, 8 met the inclusion criteria for 211 mid‐range heart failure patients with an implant LBBAP across the 16 centers. The average implant success rate with lumenless pacing lead use was 91.3%, and 19 complications were reported among all 211 enrolled patients. During the average follow‐up of 9.1 months, the average LVEF was 39.8% at baseline and 50.5% at follow‐up (MD: 10.90%, 95% CI: 6.56−15.23, p < .01). Average QRS duration was 152.6 ms at baseline and 119.3 ms at follow‐up (MD: −34.51 ms, 95% CI: −60.00 to −9.02, p < .01). LBBAP could significantly reduce QRS duration and improve systolic function in a patient with LVEF between 35% and 50%. Application of LBBAP as a CRT strategy for HFmrEF may be a viable option.


| INTRODUCTION
The advantageous effect of cardiac resynchronization therapy (CRT) is well known in heart failure (HF) patients with reduced left ventricular ejection fraction (LVEF) and prolonged QRS duration. [1][2][3] Current global guidelines suggest CRT for class I or IIA indications in patients with LVEF ≤35% and symptomatic HF despite receiving optimal drug therapy. [4][5][6] However, an LVEF cut-off of ≤35% was determined by the patient enrollment criteria in major CRT trials, and of note, was adopted from cut-off values in prior major implantable cardiac defibrillator trials rather than from a prospective risk-benefit analysis of CRT for all LVEF ranges.
Although resynchronization therapy for HF patients with LVEF ≥35% is controversial and has not been clearly established, HF with LVEF >35% shows disease features similar to LVEF <35%, and treatment patterns are similar. 7,8 Additionally, there are reports that long-term clinical outcomes were poor when HF with mildly reduced ejection fraction (HFmrEF) was accompanied by left bundle branch block (LBBB) versus without LBBB. 9 Moreover, in a retrospective analysis of the PROSPECT trial database, CRT demonstrated significant clinical benefit among patients with an LVEF >35%. 10 The BLOCK-HF trial proved that CRT provided a clinical benefit over right ventricular pacing in patients with LVEF ≤50% and atrioventricular block who require ventricular pacing, and almost 70% (483/691) of the study population had an LVEF of 36%−50%. 11 Conduction system pacing (CSP) that directly activates the specialized conduction system was developed, and left bundle branch area pacing (LBBAP)-which overcomes the limitations of the previously used His bundle pacing (HBP)-is emerging and widely used. [12][13][14][15][16] Furthermore, the effectiveness and safety of LBBAP in patients with HF have also been reported. [17][18][19] To date, LBBAP has been used as an alternative to CRT in indicated patients, as well as a first option for patients indicated for CRT or pacemaker implantation. [17][18][19] LBBAP is more simple, convenient and cheaper than biventricular CRT; therefore, there is increasing clinical interest in adopting wider LVEF ranges for LBBAP-CRT among patients with HF and a long QRS duration, especially in patients with an LVEF of 36%−50%. Therefore, we aimed to systematically review the literature and meta-analysis on the impact of the LBBAP strategy in patients with HFmrEF with an LVEF between 35% and 50%.

| Data sources and searches
This systematic review and meta-analysis were carried out following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. 20 Searches were independently performed by two investigators (G. Y. and T. K.) who searched Embase, Medline (PubMed), and Cochrane's Library databases for related articles, with the keywords: "left bundle branch pacing" OR "left bundle branch area pacing." At the same time, references of the relevant original and review articles were manually sorted to undergo a comprehensive search. The search was restricted to research in humans published in English, and was completed on July 17, 2022.

| Study selection and data extraction
Titles and abstracts were retrieved from articles searched using the keywords "left bundle branch pacing" or "left bundle branch area pacing." Studies were included if the article was in English and met the following criteria: patients aged >18 years; included patients with an LVEF from 35% to 50%; LVEF or QRS duration follow-up was performed after the procedure; and for LBBAP CRT, both patients with pacing indications (de novo or upgrade), and with HF without pacing indications (de novo CRT), were included. Among studies satisfying the above criteria, a pooled analysis of patients with a baseline LVEF of 35%−50% was used. Editorials, review articles, case reports/letters, and abstracts were excluded. The quality of the literature was evaluated by the above two researchers using the Newcastle−Ottawa scale criteria. 21

| Statistical analyses
To evaluate the change in LVEF and QRS duration over time, means, standard deviations, and sample sizes were extracted from articles to estimate the overall average values and confidence intervals (CIs).
Statistical heterogeneity was conducted by calculating Higgins I 2 ; I 2 > 50% was considered to indicate significant heterogeneity. 22 After confirming heterogeneity, we used a random effects model because the populations included in the individual studies were from different populations. 23 Potential publication bias was assessed by visual inspection of the funnel plot and the Egger regression asymmetry test.
If the p for bias was >.05, it was judged that there was no publication bias. 24 All tests with p < .05 were considered to be statistically significant. Statistical analyses were performed using R programming version 4.0.3 (The R Foundation for Statistical Computing).

| Studies selection and evaluation
In total, 694 articles were retrieved after excluding duplicates. The title and abstract of the articles were screened, and 566 were excluded due to being editorials/review articles, case reports/letters, having no relevant outcome, and not being in English. The remaining 128 articles underwent full-text review, and 120 were excluded for not meeting the inclusion criteria; finally, eight studies 25-32 were retrieved ( Figure 1 and Table 1 Table S1).

| Definition of LBBAP
The definition of LBBAP was described in seven studies and omitted in one. Confirmation of LBBAP was made by comprehensively reviewing the following findings: (1) the only criterion used in all seven studies was paced QRS morphology, presented with a right bundle branch block morphology pattern in lead V1; (2) abrupt shortening of Stim-LVAT (stimulus to a peak of the R wave) in leads V5 and V6 was used in four studies; (3) observed left bundle branch potential in pacing leads was used in four studies; and (4) demonstration of the transition from nonselective LBBP to selective LBBP was used in two studies. Only one study used contrast to ensure the location of the lead in the interventricular septum.

| Publication bias
According to Egger regression, the p for bias in QRS duration and LVEF were confirmed to be 0.126 and 0.433, respectively, indicating that there was no obvious publication bias. The funnel plots of the studies included in each analysis are presented in Supporting Information: Figure S1.

| Feasibility of LBBAP in HFmrEF
All studies included in this meta-analysis performed LBBAP using lumenless pacing leads, with an average acute success rate of 91.3%.

In 1 case series wherein LBBAP was performed in patients with
HFmrEF using stylet-driven leads, the success rate was 100% (n = 4). 38 Although procedure-related complications of septal perforation, pneumothorax, pocket infection, pocket hematoma, and lead dislodgements occurred during the follow-up period, no major implantation-related complications were observed.

| LBBAP as a resynchronization strategy for HFmrEF
In this meta-analysis, during the average follow-up of 9.

| CONCLUSION
In all studies included in this meta-analysis, LBBAP in patients with HFmrEF was feasible and safe. In the pooled analysis, LBBAP was found to significantly shorten the QRS duration and improve systolic cardiac function in patients with an LVEF of 35%−50%. This suggests that the application of LBBAP as a resynchronization strategy for patients with HFmrEF could be an acceptable option, especially in patients with both HFmrEF and dyssynchrony, where a decrease in LVEF is anticipated. A randomized, prospective study is warranted to evaluate the effect of LBBAP-CRT on patients with HF and an LVEF >35% and ≤35%.

AUTHOR CONTRIBUTIONS
Ga-In Yu and Tae-Hoon Kim contributed to the conception of the study, searched the articles, independently reviewed all identifed articles for eligibility, performed the data analyses, and wrote the manuscript. Yun-Ho Cho, Jae-Seok Bae, Jong-Hwa Ahn, and Jeong Yoon Jang assisted in data acquisition and review literature. Choong Hwan Kwak helped perform the analysis with constructive discussions. All authors reviewed the manuscript.

CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT
The data underlying this article will be shared on reasonable request to the corresponding author.