The Feasibility and Safety of Left Bundle Branch Pacing vs. Right Ventricular Pacing After Mid-Long-Term Follow-Up: A Single-Centre Experience

Authors, Journal, Affiliations, Type, DOI

• Authors: Xueying Chen and Qinchun Jin (joint first authors); Jin Bai; Wei Wang; Shengmei Qin; Jingfeng Wang; Yixiu Liang; Yangang Su (corresponding); Junbo Ge
• Journal: Europace 2020;22(Suppl 2):ii36–ii44. Received 18 May 2020; accepted 2 September 2020
• Affiliation: Department of Cardiology, Shanghai Institute of Cardiovascular Disease, Zhongshan Hospital, Fudan University, Shanghai
• Type: Prospective, consecutive, non-randomised observational comparison — 554 patients enrolled January 2018 – January 2019: LBBP 237 (of 250 attempted, 94.8% success) vs RVP 317; follow-up 18.1–18.4 months
• DOI: 10.1093/europace/euaa294
• Disclosures: none declared

Overview

This is one of the few studies to compare LBBP directly with right ventricular pacing, and it is candid about the result: LBBP produced a much narrower paced QRS (132 vs 178 ms) and stable, slightly better pacing parameters over 18 months, but it did not improve complications or cardiac outcomes in a population with normal LV function and a low event rate. The price is procedural: LBBP took more attempts (2.41 vs 1.49) and more fluoroscopy (9.15 vs 3.92 min), and produced a larger transient troponin rise (0.13 vs 0.04 ng/mL at day 1) — most plausibly from the more extensive septal trauma. LBBP also recorded an LBB potential in 89.9% and selective capture in 57.8%, with Sti-LVAT equalling the intrinsic PoLBB-LVAT (65.07 vs 65.32 ms), supporting genuine conduction-system capture. The study's honest conclusion is that safety and feasibility are established, but clinical superiority over RVP is not — and with 18 months and almost no events, it could not have been.

Keywords

Left bundle branch pacing; Right ventricular pacing; Feasibility; Follow-up

Key Takeaways

Design and population

• 554 consecutive patients indicated for pacemaker implantation, non-randomised to LBBP or RVP according to patient selection and clinical practice
• Exclusions: non-specific intraventricular conduction delay; heart failure with LVEF <50%; any CRT indication — so the cohort is a preserved-LV-function, non-CRT population
• LBBP attempted in 250, successful in 237 (94.8%)
  • 13 failures: 5 leads advanced easily and perforated to the LV at multiple sites when the sheath was removed; 8 could not be screwed deep enough to produce an RBBB-pattern paced QRS. All 13 received conventional RVP
• Baseline imbalance by indication: LBBP had more AV block (53.6% vs 38.2%) and less sinus node dysfunction (31.7% vs 52.1%) (P < .0001), and a wider intrinsic QRS (117.09 ± 25.82 vs 105.04 ± 12.18 ms, P < .0001) — i.e. more conduction disease
• Age, sex, hypertension, diabetes, AF, coronary disease, valvular disease, congenital disease, CKD and stroke history were similar
• Pacemaker types were similar (dual-chamber 82.7% vs 76.3%)

Procedural characteristics of LBBP

• PoLBB recorded in 213/237 (89.9%); PoLBB–V 21.35 ± 4.89 ms; amplitude 0.20 ± 0.21 mV
• PoLBB–LVAT 65.32 ± 10.55 ms vs Sti–LVAT 65.07 ± 8.58 ms (P = .895) — the close agreement is offered as evidence of direct LBB capture
• Selective LBBP 57.8%
• Paced QRS axis: normal 17.3%, left-axis deviation 79.75%, right-axis deviation 2.11%
• Lead depth 12.85 ± 2.20 mm vs echocardiographic septal thickness 9.84 ± 1.56 mm (P < .0001) — the lead tip lay beyond the measured septum on angiography
• Fluoroscopy 9.15 ± 3.52 min; attempts 2.41 ± 0.78

Paced QRS — the headline difference

• LBBP 132.02 ± 7.93 ms vs RVP 177.68 ± 15.58 ms (P < .0001) in the Results text and Abstract
• ⚠ Table 1 reports 122.02 ± 7.93 vs 157.68 ± 15.58 ms for the same comparison. The text and table differ by 10 ms in LBBP and 20 ms in RVP; the direction is the same but the absolute values are not reconcilable

Biomarkers

• cTnT was higher in LBBP at every time point:
  • baseline 0.03 ± 0.03 vs 0.02 ± 0.03 ng/mL (P = .002)
  • day 1 0.13 ± 0.09 vs 0.04 ± 0.03 ng/mL (P < .001)
  • day 2 0.10 ± 0.08 vs 0.03 ± 0.08 ng/mL (P < .001)
  • LBBP rose significantly from baseline (P < .001); RVP showed only a trend (P = .557 and .238)
• NT-proBNP was higher in LBBP at baseline (1097.55 ± 1709.16 vs 520.73 ± 731.31 pg/mL, P < .001) and at day 1 (1198.60 ± 2162.96 vs 521.95 ± 753.43 pg/mL, P = .002), but not at day 2 (P = .531), and fell from baseline in LBBP by day 2 (P = .039)
• The authors attribute the troponin difference to the procedure — more attempts, longer fluoroscopy, more myocardial injury — rather than to a specific hazard of septal pacing; no patient had ACS or pulmonary embolism peri-procedurally

Pacing parameters over 18 months

• Thresholds were lower in LBBP at implant and remained marginally lower until they converged by 18 months (0.73 vs 0.76 V, P = .195)
• Sensed R waves were consistently higher in LBBP and rose after implantation before plateauing, while RVP sensing stayed flat
• Impedance fell progressively in LBBP (533 → 402 Ω) but remained stable in RVP (~492 Ω) — the expected consequence of steroid-eluting deep septal fixation and tissue remodelling

Complications and cardiac outcomes

• No significant difference in complications (P = .179) or cardiac outcomes (P = .673)
• The single LBBP perforation was detected at 1 month in an SSS patient: higher unipolar than bipolar threshold and lower impedance (5 V/0.4 ms vs 2.5 V/0.4 ms; 402 vs 276 Ω), with CT showing the tip ~10 mm into the LV cavity and the ring still in the septum; the lead was repositioned
• The two dislodgements occurred at 1 month in an AF/low-ventricular-rate patient with a single-chamber device and an AVB patient with a dual-chamber device
• One VT (left ventricular summit/outflow-tract tachycardia) at 1 month required ablation; one ACS at 3 months required stenting of the LAD and RCA

Limitations of the document

Authors' own stated limitations

• Single-centre, non-randomised observational study; patients underwent LBBP or RVP according to individual choice and operator practice
• The lack of randomisation produced a heterogeneous population, which may bias the estimates
• Long-term lead performance and the efficacy of LBB capture remain unknown
• Long-term, multicentre, randomised trials are necessary

Additional limitations evident in the study

• The comparison is confounded by indication. LBBP patients had more AV block and wider baseline QRS; RVP patients had more sinus node disease. These are different populations with different pacing burdens and different expected outcomes, and no adjustment is performed
• The cohort was deliberately low-risk and non-CRT (LVEF ≥50%, no NIVCD, no CRT indication), so the absence of a clinical benefit is uninformative — an event rate of 0–1% over 18 months cannot distinguish the strategies. The study is a safety and feasibility comparison, not an efficacy trial
• ⚠ The paced QRS values are internally inconsistent: the text/abstract report 132.02 vs 177.68 ms, Table 1 reports 122.02 vs 157.68 ms. Both favour LBBP, but the discrepancy is large enough that the headline effect size cannot be quoted precisely
• The troponin difference is confounded by procedural burden. LBBP had more attempts (2.41 vs 1.49) and longer fluoroscopy (9.15 vs 3.92 min), so the higher cTnT may reflect the learning-curve technique rather than a property of septal pacing. No adjustment or mediation analysis is presented
• NT-proBNP was higher in LBBP at baseline, indicating residual confounding by comorbidity, and the day-2 comparison is underpowered
• The 94.8% success rate includes a specific failure mode — five leads that advanced too easily and perforated at multiple sites when the sheath was removed — which is a technical hazard the paper describes but does not analyse
• No data on ventricular pacing percentage, medications, or echocardiographic remodelling, so the electrical advantages cannot be linked to structural outcomes
• Complications were ascertained clinically, not by systematic imaging, so asymptomatic perforation or lead displacement may be undercounted (the one detected perforation was found because of electrical changes)
• Eighteen months is mid-term, and the title's "mid-long-term" overstates it; late threshold rise and lead failure are not captured
• No blinding, no core lab, single experienced centre — the same group that developed the technique
• The impedance fall in LBBP to ~400 Ω is within the range that overlaps with perforation criteria in other series; the paper does not discuss whether this creates diagnostic ambiguity at follow-up
• Rating rationale: rated medium. It is a reasonably large, consecutive, prospective comparison and one of the few direct LBBP-versus-RVP datasets, with a clear electrical result and honest reporting of the null clinical outcome. It is not high because it is non-randomised and confounded by indication, underpowered for outcomes by design, internally inconsistent on the primary ECG endpoint, and single-centre

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the technique under comparison
• entities/CRT — excluded by design, which limits generalisability
• Right ventricular pacing — the conventional comparator and the source of pacing-induced dyssynchrony
• Paced QRS duration — 132 vs 178 ms (text) or 122 vs 158 ms (table); the primary electrical endpoint
• Sti-LVAT and PoLBB–LVAT — 65.07 vs 65.32 ms, their equality supporting direct LBB capture
• LBB potential — recorded in 89.9%
• Selective LBBP — 57.8%
• cTnT and NT-proBNP — post-procedural biomarker release and its procedural correlates
• Lead depth vs septal thickness — 12.85 vs 9.84 mm, implying the tip lies beyond the angiographic septal border
• Pacing threshold, sensing and impedance trajectories — LBBP thresholds converge with RVP by 18 months while sensing remains higher and impedance falls

Key Entities Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — primary concept page
• Zhongshan Hospital, Fudan University, Shanghai — the investigating centre
• Xueying Chen; Yangang Su; Junbo Ge — the investigators
• Medtronic 3830 lead and C315 His sheath; Medtronic 2290 programmer — the LBBP system
• GE CardioLab 2000 — the EP recording system
• Oxford University Press / European Society of Cardiology / Europace — publisher

Wiki Pages Updated

• Created: wiki/sources/lbbp-vs-rvp-europace-2020.md
• Updated: wiki/concepts/Left-Bundle-Branch-Area-Pacing.md
• Updated: wiki/concepts/Conduction-System-Pacing.md
• Updated: wiki/entities/CRT.md
• Updated: wiki/sourceindex.md
• Updated: wiki/wikiindex.md