Left Bundle Branch Area Pacing Outcomes: The Multicentre European MELOS Study

Authors, Journal, Affiliations, Type, DOI

• Authors: Marek Jastrzębski (corresponding), Grzegorz Kiełbasa, Oscar Cano, Karol Curila, Luuk Heckman, Jan De Pooter, Milan Chovanec, et al. — 14 European centres
• Journal: European Heart Journal 2022;43(40):4161–73. © The Author(s) 2022, published by Oxford University Press on behalf of the European Society of Cardiology. Open access, CC BY-NC. Received 25 January 2022; revised 21 July 2022; accepted 28 July 2022; online ahead-of-print 18 August 2022
• Affiliations: First Department of Cardiology, Interventional Electrocardiology and Hypertension, Jagiellonian University, Kraków (lead centre); Hospital Universitari i Politècnic La Fe, Valencia; and 12 further European centres
• Type: Registry-based observational study — 2,533 patients at 14 European centres, all patients in whom LBBAP device implantation was attempted, for any indication. No comparator arm, no randomisation, no control group
• DOI: 10.1093/eurheartj/ehac445
• Filename year is correct — the raw file is MELOS-LBBAP-EHJ-2022.md and the journal of record is 2022. The only file in this batch whose year matches

Overview

MELOS is the study that turned left bundle branch area pacing from a collection of single-centre series into a characterised technique, and it did so by sheer scale: 2,533 patients at 14 European centres, against 30–143 patients in the single-centre studies that preceded it. Its value is that it measures the things small series cannot — a real-world success rate (92.4% for bradyarrhythmia but only 82.2% for heart failure), a learning curve (steepest over the first 110 cases, plateauing after 250), a complication rate (11.7% overall, 8.3% specific to the transseptal route), and above all a capture-type distribution (69.5% left bundle fascicular, 21.5% LV septal, 9% proximal LBB). That last figure is the most consequential: it means that in more than a fifth of "LBBAP" implants, the left bundle was not captured at all. The authors' own conclusion is notably uncelebratory for a large positive registry — the technique is feasible, but success in heart failure and safety need to be improved, and randomised trials are necessary.

Keywords

Not provided as an author list in the extracted text; the article's indexing terms would be: left bundle branch area pacing; conduction system pacing; left bundle fascicular pacing; left ventricular septal pacing; implantation success; complications; learning curve.

Key Takeaways

Background and the Origin of "LBBAP" as a Term

• The limitations of biventricular pacing for CRT prompted development of more physiological options
• The transseptal route to permanent LV septal pacing was first demonstrated in humans in 2016 by Mafi-Rad et al.; the technique was then modified by Huang et al., who showed that direct proximal left bundle branch pacing could be achieved via the same approach
• Why the umbrella term LBBAP is used: the paced QRS complexes of LVSP and LBBP differ only slightly, intermediate capture types exist (notably left bundle fascicular pacing, LBFP), there is a lack of standard and precise differentiating criteria, and clinical-outcome data are scarce. The authors therefore treat LBBAP as the common descriptor for the family of modalities
• Within four years of those two landmark papers, several small and medium-sized, mainly single-centre studies had shown feasibility — but valid questions about safety and in-depth characterisation in real-world practice remained, which is what MELOS addresses

Methods

• Registry-based observational study including patients in whom LBBAP device implantation was attempted at 14 European centres, for any indication
• 2,533 patients — mean age 73.9 years, 57.6% female, 27.5% with heart failure

Results — Success and Learning Curve

• LBBAP lead implantation success rate: 92.4% for bradyarrhythmia indications and 82.2% for heart failure indications — a 10-point gap between the two
• The learning curve was steepest for the initial 110 cases and plateaued after 250 cases — meaning a centre or operator does not reach stable competence until roughly 250 implants
• Independent predictors of LBBAP lead implantation failure: heart failure, broad baseline QRS, and left ventricular end-diastolic diameter

Results — Capture Types (the most consequential finding)

• More than one implant in five labelled "LBBAP" did not capture the left bundle at all. This is the primary-source quantification of the confound that sources/lvsp-vs-lbbp-cepc-2022 described qualitatively as "LBB capture rates vary from 60% to 90%"
• The distribution also shows that the majority capture type is fascicular, not proximal LBB — a third entity that neither of the two named modalities describes

Results — Electrical Parameters and Complications

• Capture threshold 0.77 V and sensing 10.6 mV, stable over mean follow-up of 6.4 months
• Complication rate 11.7%
• Complications specific to the ventricular transseptal route of the pacing lead occurred in 209 patients (8.3%) — so roughly two-thirds of all complications were attributable to the route itself rather than to pacing generally

Discussion — Themes

• Technique evolution; success rate; learning curves; the variety of LBBAP capture types; left bundle fascicular pacing as a novel conduction system pacing modality; LVSP as a simple method for indirect left bundle activation; capture types in other studies; and complications related to the ventricular transseptal route

Conclusions

• LBBAP is feasible as a primary pacing technique for both bradyarrhythmia and heart failure indications
• Success rate in heart failure patients and safety need to be improved
• For wider use of LBBAP, randomized trials are necessary to assess clinical outcomes

Limitations of the document

Authors' own stated limitations

• Lack of homogeneity across the 14 centres. Multiple centres and operators were involved, so there was inconsistency in the implantation technique, the LBB capture criteria applied during implantation, the methods used for QRS duration and interval measurement, and the enrolment strategy — a substantial caveat when the primary endpoints are a success rate and a capture-type distribution
• No independent central adjudication committee, and partially retrospective data retrieval — which "might have resulted in underreporting of failures and complications". The authors counter that these were nonetheless reported at a higher percentage than in any other study, which is a fair point but does not restore the missing adjudication
• Follow-up was short and narrow: an average of only 6 months, limited to procedural outcomes, complications, echocardiographic response and electrical parameters. Follow-up 12-lead ECG was available for only 1,357 of 2,533 patients (53.6%), "potentially influencing the reported incidence of loss of LBBAP"
• Neither mortality nor heart failure episodes were analysed — the two outcomes a pacing technique ultimately has to answer for
• Results might be less applicable to non-European populations

Additional limitations evident in the study

• It reports process and safety, not benefit. Success rates, capture types, thresholds and complications are all measured; no clinical outcome is compared against any alternative. The registry cannot say whether LBBAP is better than biventricular pacing or conventional RV pacing — which the authors acknowledge by calling for randomised trials
• The 11.7% complication rate sits alongside an 8.3% route-specific rate, and the study is not powered to identify what drives either. With retrospective retrieval and no adjudication, the natural next question — which operators, which techniques, which patients — cannot be answered from these data
• The learning-curve finding cuts against early adoption. A plateau only after 250 cases means that outcomes reported from low-volume centres, or from the early experience of any centre, are systematically worse than the stable figure. MELOS is itself a composite of centres at different points on that curve, which may make its success rate a blend rather than a target
• More than a fifth of implants did not achieve LBB capture (21.5% LVSP), yet the study's own framing treats LBBAP as one family — so the headline success rate of 92.4% is a rate of implanting a lead in the area, not of capturing the left bundle
• No cost, procedure-time or fluoroscopy data are reported, which matters for a technique being proposed as a primary pacing modality
• The authors' conclusion is more restrained than the abstract's numbers suggest, and should be read as the paper's real position: feasible, but success in heart failure and safety need improvement before wider use
• Rating rationale: rated high — the largest LBBAP cohort in the wiki by two orders of magnitude (2,533 vs 30–143 in the single-centre studies), 14 centres, published in EHJ. Not very high: registry-based, observational, no adjudication, no comparator, no clinical outcomes, and the authors themselves call for randomised trials

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the umbrella concept this study characterises
• concepts/Conduction-System-Pacing — parent category
• LBBAP capture types — left bundle fascicular (69.5%), LV septal (21.5%), proximal LBB (9%); the study's central contribution and the primary-source resolution of the "how often is the LBB actually captured" question
• Left bundle fascicular pacing (LBFP) — the predominant capture type and a third entity beyond LVSP and LBBP
• Learning curve (110 / 250 cases) — the case volume at which competence stabilises
• Ventricular transseptal route complications (8.3%) — complications attributable to the lead's route rather than to pacing generally
• Implant success rate by indication — 92.4% bradyarrhythmia vs 82.2% heart failure

Key Entities Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — primary concept page
• Mafi-Rad et al. (2016) — first-in-human transseptal LVSP; Huang et al. — modified it to achieve proximal LBBP; the two landmark papers MELOS builds on
• Jagiellonian University, Kraków — lead centre; Hospital Universitari i Politècnic La Fe, Valencia; 14 European centres total
• European Society of Cardiology / Oxford University Press — society and publisher
• This is the primary source the wiki previously cited second-hand — several MELOS figures on the concept pages were attributed to sources/csp-ehra-2023 because no MELOS page existed

Wiki Pages Updated

• Created: wiki/sources/melos-lbbap-ehj-2022.md
• Updated: wiki/concepts/Left-Bundle-Branch-Area-Pacing.md — MELOS claims re-attributed to the primary source
• Updated: wiki/concepts/Conduction-System-Pacing.md
• Updated: wiki/sourceindex.md
• Updated: wiki/wikiindex.md