A Continuous Pacing and Recording Technique for Differentiating LBBP From LV Septal Pacing: Electrophysiologic Evidence From an Intrapatient-Controlled Study

Authors, Journal, Affiliations, Type, DOI

• Authors: Jiabo Shen, MD; Longfu Jiang, MD; Hao Wu, MD; Xiaojie Cai, MD; Shanshan Zhuo, MD; Lifang Pan, MD
• Journal: Canadian Journal of Cardiology 2023;39:1e10 (as printed in the extraction). © 2022 Canadian Cardiovascular Society, published by Elsevier. DOI issued September 2022
• Affiliations: Department of Cardiology, Hwa Mei Hospital, University of Chinese Academy of Sciences, Ningbo, Zhejiang, China; and Department of Global Health, Ningbo Institute of Life and Health Industry, University of Chinese Academy of Sciences
• Type: Retrospective, single-centre, intrapatient-controlled observational study — 97 consecutive patients, each serving as their own comparison across six pacing modalities
• DOI: 10.1016/j.cjca.2022.09.008
• Disclosures: "Dr Jiang owns the patent for John Jiang's connecting cable" — the specific device that makes the continuous-pacing technique possible, and the technique this paper validates. The other authors report no conflicts. This is a material relationship to weigh when reading the paper's advocacy for the technique
• Funding: Zhejiang Provincial Public Service and Application Research Foundation (LGF22H020009); Ningbo Health Branding Subject Fund (PPXK2018-01)
• Why it matters here: it delivers an intrapatient comparison across the entire pacing spectrum — from right ventricular septal pacing through to selective LBB capture — which no other source in this wiki does. And its conclusion is the most direct statement in the corpus against the field's habit of lumping these modalities together (see Contradictions)

Overview

This paper's method is its argument: by pacing continuously and uninterrupted while the lead is being screwed in — using a rotatable connector (the "John Jiang connecting cable") that keeps current flowing through the rotating lead — the operator can watch the paced ECG and intracardiac EGM change beat by beat as the electrode crosses the septum. That turns insertion into a natural experiment inside each patient, and the authors use it to characterise six discrete pacing modalities encountered in sequence: right ventricular septal pacing (RVSP), two intermediate intraventricular septal stages (IVSP1, IVSP2), LV septal endocardial pacing (LVS(e)P), nonselective LBBP, and selective LBBP. In 97 patients, 87 (89.7%) reached nonselective LBB capture and 82 (84.5%) selective capture; 9 proved to have only LVS(e)P and 1 could not penetrate the septum at all. The key measurements are the Stim-V6RWPT, which falls stepwise across the modalities (91.0 → 86.9 → 82.1 → 67.1 ms) with an abrupt drop at the LBB capture threshold, and the Stim-QRSend, which does not have its minimum at LBB capture but at IVSP2 (136.6 ms) — so selective LBB capture produced the widest QRS-end of all (150.6 ms). The authors conclude that these modalities should not be united under the "LBBAP" umbrella until clinical outcome data show which is superior.

Keywords

(No author keyword list printed. Abbreviations: RVSP, IVSP1/2, LVS(e)P, NSLBBP, SLBBP; Stim-V6RWPT; Stim-QRSend; TVA; COI.)

Key Takeaways

Abstract

• Background: LBBP is "a promising approach for achieving near-physiologic pacing. However, differentiating LBBP from left ventricular septal endocardial pacing (LVS(e)P)" is the problem
• Methods: LBBP using continuous uninterrupted pacing and real-time monitoring of ECGs with intracardiac EGMs in 97 consecutive patients; electrophysiologic characteristics evaluated across six modalities — RVSP, IVSP1 and IVSP2, LVS(e)P, nonselective LBBP (NSLBBP), and selective LBBP (SLBBP)
• Results:
  • 87 of 97 (89.7%) met the criteria for nonselective LBB capture — "abrupt change in paced QRS morphology with a transition from Qr to QR/qR in lead V1 and shortening of Stim-V6RWPT of ≥10 ms with constant output while rather than after lead screwing"
  • Selective LBB capture in 82 (84.5%)
  • Stim-V6RWPT of NSLBBP and SLBBP were significantly shorter than LVS(e)P — 67.1 ± 8.7 ms, 67.0 ± 9.3 ms, and 82.1 ± 10.9 ms respectively
  • "Stim-QRSend was the narrowest in IVSP2 (136.6 ± 15.2 ms) instead of NSLBBP (140.0 ± 17.1 ms)"
• Conclusions: the uninterrupted pacing technique for differentiating LBBP from LVS(e)P within the same patients is feasible, and "LBBP and LVS(e)P differ in ventricular electrical synchronization"

Methods

Population and definitions

• Consecutive patients with indications for permanent pacemaker implantation, April 2021 to March 2022, at Hwa Mei Hospital, Ningbo
• Ethics committee approval; informed consent from every patient
• LBB capture was defined dynamically, by paced QRS morphology during continuous monitoring:
  1. LVS(e)P → NSLBBP: "the paced QRS morphology changed with a transition from Qr to QR/qR in lead V1 and the Stim-V6RWPT for 2 adjacent paced QRS complexes was abruptly decreased by ≥10 ms with a constant output (2 V/0.5 ms) during the process of lead screwing and remained shortest and constant at low and high outputs"
  2. NSLBBP → SLBBP: "the paced QRS morphology changed with a transition from QR/qR to rsR/rSR/rSr in lead V1 and a distinct isoelectric interval and discrete component appeared on intracardiac EGM with a fixed Stim-V6RWPT"

The six modalities, defined by the paced morphology at successive depths

The technique

• Transventricular septal implantation via left axillary or left subclavian vein, C315 His sheath (Medtronic), lumenless 3830 lead (Medtronic)
• Angiography first, to visualise the tricuspid valve annulus under RAO 30°, with the image saved as a reference marker to locate the LBBP target zone
• ⚠ The enabling device: the "John Jiang connecting cable" (Xinwell Medical Technology Co, Ningbo) — "similar to a double-layer metal ring", with a "mechanism of action... similar to that of bearings, allowing components to move with respect to each other", so that "the electric current is fully conducted during the rotation for continuous pacing and recording". The senior author holds the patent (see Disclosures)
• Continuous uninterrupted pacing at 2 V/0.5 ms, with the implanter watching fluoroscopy while a second operator assessed the ECG and EGM. Stim-V6RWPT was measured live; other measurements were made offline
• Distinction from the interrupted method: "In contrast to the interrupted pacing technique, in which the alligator clip is disconnected to rotate the lead, this technique enables beat-to-beat monitoring of the change in paced QRS morphology (precordial transition, notch in V1, Stim-V6RWPT) and impedance as the lead advances"
• Pacing protocol at >10 beats/min above native sinus rhythm, to exclude spontaneous heart-rate change confounding the QRS measurement
• Real-time QRS guidance: as the lead advanced from right to left, "the notch on the paced QRS complex in lead V1 moved from the nadir to the end of the QRS (resultant RBBB pattern) and the Stim-V6RWPT gradually shortened"
• When to stop and reposition: "When the paced QRS changes (such as when the R/S transition zone cannot advance to the V1 lead and/or the notch in V1 lead cannot move to the end of the QRS-wave) are unsatisfactory, lead screwing should be stopped, and the screw-in site should be replaced"
• Output was intermittently raised to 5 V/0.5 ms to compare Stim-V6RWPT with the 2 V value; if RWPT was prolonged at 2 V relative to 5 V, more turns were given to obtain the shortest RWPT
• ⚠ Impedance and perforation surveillance, with a specific threshold: "If the impedance dropped to 600 Ω, the current of injury (COI) was monitored to determine whether its amplitude dropped suddenly. A sudden drop in the COI magnitude indicates a possible perforation, and further rotations need to be stopped"
• Abandonment rule: "LBBP was abandoned if LBB capture was not obtained at 3 or 4 sites" — and those cases were then counted in the LVS(e)P or RVSP group
• Lead depth assessed with sheath angiography

Results

Attainment of each modality (n=97)

• 87 (89.7%) confirmed NSLBBP by abrupt Stim-V6RWPT shortening ≥10 ms
• 82 (84.5%) achieved SLBBP with isoelectric intervals and discrete components on threshold testing
• 9 patients confirmed to have LVS(e)P — absence of the ≥10 ms abrupt shortening and no isoelectric interval
• 1 patient had RVSP only, because of inability to penetrate the ventricular septum for deep septal placement
• Paced parameter changes were analysed in the 82 patients meeting selective LBB criteria
• Pacing indications: AV block 49; sick sinus syndrome 28; AF with slow ventricular rate 7; CRT 3

Stim-V6RWPT — a clean stepwise fall, with an abrupt step at capture

• IVSP1 > IVSP2 (P < 0.05); IVSP2 > LVS(e)P (P < 0.05); LVS(e)P → NSLBBP the significant abrupt shortening (P < 0.05); no significant difference between NSLBBP and SLBBP
• So within-patient, the V6 activation time shortens progressively through the septum and then drops by ~15 ms at the moment of LBB capture

Stim-QRSend — the counterintuitive result

• IVSP1 137.1 ± 15.2 ms vs IVSP2 136.6 ± 15.2 ms — no significant difference
• IVSP2 136.6 ± 15.2 ms vs LVS(e)P 141.3 ± 15.6 ms (P < 0.05)
• LVS(e)P vs NSLBBP (140.0 ± 17.1 ms) — no significant difference
• ⭐ NSLBBP 140.0 ± 17.1 ms vs SLBBP 150.6 ± 19.3 ms (P < 0.05)
• The narrowest Stim-QRSend of all six modalities was IVSP2 (136.6 ± 15.2 ms), not any form of LBB capture — the authors' Figure 4 interpretation: "The Stim-QRSend of IVSP2 was the shortest among all of the pacing modalities. This suggests that IVSP2 preserves better interventricular synchrony"

Complications

• LVS perforation in 3 patients
• "Apart from LVS perforation in 3 patients, no other procedure-related complications, such as electrode displacement, pericardial effusion, infection, or pocket hematoma, were observed"

Discussion

The synchrony argument

• "LBBP maintains or restores the physiologic LV activation sequence and provides a shorter V6RWPT and better ventricular synchrony compared with LVS(e)P"
• Heckman et al. found LBBP gave "a small but significant optimization in ventricular synchronization" versus LVS(e)P; Hou et al. found LBB capture may generate "a superior advantage" over LVS myocardial capture
• ⚠ But the counter-evidence is acknowledged: "Curila et al. reported that LBBP preserved physiologic LV depolarisation but resulted in GREATER interventricular electrical dyssynchrony than LVSP. Additional clinical trials are needed to confirm these findings." — the intraventricular gain and the interventricular cost are being traded against each other
• "As the debate continues over which pacing modality is preferred, clinicians need an accurate way to differentiate between LBBP and LVS(e)P"

Why the modalities are not interchangeable

• "Both pacing techniques involve placement of the lead in the deep septum, resulting in a paced RBBB pattern in lead V1 and a narrow Stim-QRSend after the procedure. Therefore, the collective term LBB area pacing (LBBAP) has emerged... However, LBBAP indicates LVS(e)P or LBBP without clear evidence of LBB engagement"
• "The definition of LBBP is the capturing of the left bundle or its branches along with the adjacent LVS myocardium. If only the LVS myocardium is captured... then it is referred to as LVS(e)P or deep septal pacing. Therefore, LVS(e)P and LBBP are different pacing methods"
• The mechanistic basis for the abrupt step: "The LBB and working myocardium are distinct tissues with different electrophysiologic features, such as propagation speed, refractory period, capture thresholds, and action potential duration... As a result of the different conduction velocities between the LBB and local myocardium, the change in pacing modalities results in an abrupt change" in the ECG

Limitations (the authors' own)

• "A single-centre retrospective study with a relatively small sample size"
• Operator dependence, stated candidly: the study "was conducted at a centre with experienced implanters with the ability to carefully identify paced ECG and EGM morphologic changes. This may be the reason for the relatively high success rate of the LBB capture. Therefore, the generalisation of the continuous pacing and recording technique needs to be further validated in multicentre studies"
• "Randomised controlled trials are urgently needed to confirm the results of this study and provide guidance for clinicians"
• "Physiologic measurements were not used to confirm the physiologic outcomes of the different pacing patterns" — the synchrony claims rest on ECG intervals, not on measured haemodynamics or acute LV function
• Long-term comparison needed of both electrophysiologic characteristics and clinical effects

Conclusion — and the paper's strongest claim

• The uninterrupted pacing technique is feasible and highlights "the need for real-time monitoring of the paced QRS morphologic changes to distinguish between different pacing modalities during transseptal lead implantation"
• ⭐ "Selective and nonselective LBBP, LVS(e)P, or even IVSP should NOT be brought under the LBBAP umbrella until there is clinical evidence that the optimal pacing mode is superior in terms of left ventricular ejection fraction, quality of life, and other clinical outcomes."
• The technique is therefore proposed as the means to generate the clinical data that would settle which modality is preferable

Limitations of the document

Authors' own stated limitations

• Single-centre, retrospective, relatively small sample; generalisation explicitly requires multicentre validation
• Operator dependence — the authors concede the high success rate may reflect their own expertise in reading paced ECG/EGM morphology, which is a fair and unusually candid admission
• No physiological or haemodynamic confirmation of the synchrony differences; every conclusion rests on ECG intervals
• No clinical outcomes and no long-term follow-up — which is precisely the evidence the conclusion demands before the modalities are unified

Additional limitations evident in the study

• ⚠ A declared conflict sits at the centre of the method. The rotating connector that makes continuous pacing possible is patented by the senior author (Dr Jiang) and manufactured by a local company (Xinwell). The paper's central recommendation — that this technique should be used to distinguish modalities — is therefore also a recommendation to adopt the author's device. That does not make the finding wrong, but it belongs with the evidence, and the paper states it plainly enough that the reader can weigh it
• The "LBB capture" definition is the same ECG surrogate other sources question. NSLBBP here is defined by abrupt Stim-V6RWPT shortening ≥10 ms with a Qrs→QR/qR transition — the criterion that the direct-evidence study finds 100% specific but only applies once capture exists (sources/lbbp-vs-lvsp-criteria-jaccep-2021). The present study has no independent electrophysiological standard, so its 89.7% "capture" rate is a rate of ECG transition, not of proven LBB engagement — and, as the authors themselves concede, "LBBAP indicates LVS(e)P or LBBP without clear evidence of LBB engagement"
• The abandonment rule shapes the denominator. "LBBP was abandoned if LBB capture was not obtained at 3 or 4 sites", and those patients were then reclassified into the LVS(e)P or RVSP group. So the 9 LVS(e)P cases are not simply a different anatomy — they are patients the operator stopped pursuing, at a threshold chosen by the operator. The 89.7% success figure is therefore contingent on that rule
• The counterintuitive QRSend result is reported but not pursued. IVSP2 had the narrowest Stim-QRSend of all six modalities, and SLBBP the widest (150.6 ms) — meaning the best conduction-system capture produced the worst interventricular synchrony measure. The authors note it in the figure legend ("IVSP2 preserves better interventricular synchrony") but do not reconcile it with their own claim that LBBP gives better synchrony than LVS(e)P, nor with the Curila finding they cite. This is the same inversion the direct-evidence study documents, arrived at independently, and it deserves more than a legend
• "Synchrony" is inferred from intervals, and the two intervals point in opposite directions. LBBP wins on Stim-V6RWPT (intraventricular activation, 67 vs 82 ms) and loses on Stim-QRSend (interventricular/total, 150.6 vs 140.0 ms). Calling LBBP the more synchronous modality therefore requires choosing an intraventricular over an interventricular metric — a choice the paper makes implicitly rather than defending
• The sample denominator shifts across analyses. 97 enrolled, 87 reached NSLBBP, 82 met selective criteria and were used for the parameter analysis, 9 were LVS(e)P and 1 RVSP — so the group statistics are computed on 82, while the success claims are stated against 97
• No blinding, no independent adjudication of the six morphologies. The modality labels are applied by the operators themselves during a live case — the same people whose skill the authors identify as the confounder. There is no kappa for morphologic classification and no core-lab review, unlike the JACC:EP 2025 criteria paper
• Pacing indications are a mixed bradycardia population (AVB 49, SSS 28, AF-slow 7, CRT 3), and the six-modality spectrum is characterised in aggregate, with no analysis by substrate — so we do not learn whether LBB capture is harder to distinguish in LBBB, wide QRS, or cardiomyopathy
• Rating rationale: rated medium. In its favour: n=97, a genuinely intrapatient-controlled design that no other source here achieves, six clearly operationally-defined modalities, and a candid conclusion that challenges its own field's terminology. Against: retrospective, single-centre, operator-dependent, no physiological confirmation, no independent adjudication, a sample that shifts between analyses, an abandonment rule that defines the comparator group, and a patented device at the centre of the method. It is more informative than a small feasibility series and less rigorous than a blinded validation study — hence medium

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the technique, and the umbrella term this paper argues against
• concepts/Conduction-System-Pacing — the wider category
• The six pacing modalities — RVSP, IVSP1, IVSP2, LVS(e)P, NSLBBP, SLBBP — the paper's taxonomy for the whole traversal of the septum, unique in this corpus
• Stim-V6RWPT — stimulus to V6 R-wave peak; falls stepwise across modalities and drops abruptly at LBB capture
• Stim-QRSend — stimulus to end of QRS across all 12 leads; the interventricular measure, whose minimum is not at LBB capture
• Continuous uninterrupted pacing — the technique; a rotatable connector keeps current flowing while the lead rotates
• Current of injury (COI) — monitored with impedance to detect perforation; a sudden COI drop is the stop signal
• Intrapatient control — the design strength: each patient is compared against their own six in-sequence states
• Interventricular vs intraventricular synchrony — the two metrics that point in opposite directions here

Key Entities Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — primary concept page
• "John Jiang connecting cable" (Xinwell Medical Technology Co, Ningbo) — the rotatable connector enabling continuous pacing; patented by the senior author
• Medtronic C315 His sheath and lumenless 3830 lead — the standard lumenless setup
• Hwa Mei Hospital, University of Chinese Academy of Sciences, Ningbo — the single investigating centre
• Curila et al. — cited for the finding that LBBP, despite preserving LV depolarisation, causes greater interventricular dyssynchrony than LVSP
• Heckman et al. and Hou et al. — cited for the smaller synchrony advantage of LBBP over LVS(e)P
• Canadian Cardiovascular Society / Elsevier — publisher (Canadian Journal of Cardiology)

Wiki Pages Updated

• Created: wiki/sources/continuous-pacing-lvsp-cjc-2023.md
• Updated: wiki/concepts/Left-Bundle-Branch-Area-Pacing.md
• Updated: wiki/sourceindex.md
• Updated: wiki/wikiindex.md