Permanent Left Bundle Branch Pacing: What Is the Mechanism of Divergent Responses During Programmed Stimulation?

Authors, Journal, Affiliations, Type, DOI

• Author: Marek Jastrzębski, MD, PhD (sole author). ORCID 0000-0002-3318-6601
• Journal: J Cardiovasc Electrophysiol 2020;31:1222–1225. © 2020 Wiley Periodicals, Inc. Received 22 January 2020; revised 29 February 2020; accepted 13 March 2020
• Affiliation: First Department of Cardiology, Interventional Electrocardiology and Hypertension, Jagiellonian University Medical College, Kraków, Poland
• Type: "Device Rounds" — a single-case teaching/mechanism article, not original research. It presents one patient, offers no cohort, no control, no systematic comparison of protocols, and its central practical claim ("provides selective LBB QRS response very often") is stated explicitly as the author's experience rather than a measured result. The schema has no "case report" value, so source_type: other is used
• DOI: 10.1111/jce.14449
• Disclosures: moderate consultant and lecture fees from Medtronic and Abbott — note the lead used here (Medtronic 3830) and the C315His sheath are Medtronic products
• Relationship to an existing wiki source: this paper's reference 2 is sources/programmed-septal-stim-lbbp-jce-2020 — the same author's 143-patient study of programmed deep septal stimulation (JCE 2020;31:485–93). This is the mechanistic follow-up to it and is best read as a pair with it

Overview

A patient undergoing left bundle branch pacing had an ambiguous result: reducing pacing output produced no change in QRS morphology, only loss of capture at 0.5 V at 0.5 ms — a finding that could mean either that only myocardium was captured, or that the left bundle and the septal myocardium had equal capture thresholds so both were lost together. Programmed stimulation was used to settle it, and produced a striking result: at the same pacing output, same lead position and the same 280 ms coupling interval, the extrastimulus QRS was completely different depending only on the sequence of preceding cycle lengths — one sequence giving myocardial-only capture, the other giving selective LBB capture. The explanation is the Denker dissociation: myocardial effective refractory period reflects the cumulative effect of several preceding cycle lengths, whereas conduction-system ERP is governed predominantly by the immediately preceding one. Exploiting this drives the LBB-versus-myocardium ERP gap from the 55 ms measured in the companion 143-patient study to 150 ms, and can reverse its sign on demand. The paper therefore proposes that the original method's ~20% nondiagnostic rate was a property of the 600 ms basic drive train protocol rather than of programmed stimulation itself. Its evidence for that is one case plus the author's experience, so it opens the question rather than closing it.

Keywords

left bundle branch pacing, programmed stimulation, refractoriness

Key Takeaways

1 | Introduction — the case

• Patient: an 18-year-old girl with congenital complete atrioventricular block and an epicardial VVI pacemaker implanted at 4 months of age, admitted for endocardial DDD pacemaker implantation because of pacing-induced cardiomyopathy (exercise intolerance, LVEF 44%)
• Rationale for His bundle pacing: given the narrow QRS of the ventricular escape rhythm, permanent HBP was chosen as potentially optimal to prevent progression of, or reverse, the cardiomyopathy
• HBP failed: despite activation- and pace-mapping with the pacing lead (lead model 3830, Medtronic), the His bundle was not localised — probably beyond the reach of the fixed-curve delivery sheath used (C315His; Medtronic)
• LBBP was then attempted using the same lead/sheath setup, per the group's standard strategy for failed or suboptimal HBP, with fluoroscopic and ECG-based guidance as described elsewhere
• The ambiguity that prompted the maneuver: at the final lead position, during threshold assessment by pacing output reduction, no change in QRS morphology was observed — only loss of capture at 0.5 V at 0.5 ms. This "pointed either to myocardial capture only or to equal capture thresholds of LBB and myocardium"
• To distinguish these, programmed stimulation was performed
• The observation: in both panels of Figure 1, programmed stimulation was delivered from the pacing lead already in the final deep septal position, close to the LBB as evidenced by a recorded LBB potential. The paced QRS during the basic drive was relatively narrow with fast depolarisation over the left ventricular leads and with r′ in lead V1. But the final extrastimulus (S3), in both panels — despite the same pacing output, lead position and coupling interval (280 ms) — showed vastly different QRS morphology, both from the basic drive QRS and from each other

2 | Discussion — the mechanism

The physiological basis: diverging refractoriness with cycle length

• The clinical requirement: during permanent conduction system pacing it is important to confirm that direct capture of the His bundle or the LBB was achieved
• The existing approach: programmed stimulation exploits differences in refractoriness between the conduction system and the myocardium to diagnose conduction system capture, based on visualising one or both components of the fused nonselective paced QRS — the selective myocardial capture QRS and/or the selective conduction system capture QRS
• The known weakness, stated plainly: the method "works very well during His bundle pacing," but its performance during LBB pacing was suboptimal, "probably due to a smaller difference between effective refractory period (ERP) of the local septal myocardium and LBB when using standard pacing protocols" — note that the diagnosis here is attributed to the protocol, not the method
• The Denker principle this paper exploits (Denker et al., Circulation 1983;68:1212–21): abrupt changes in cycle length impact the refractoriness of myocardium and conduction system divergently
  • Myocardial ERP demonstrates a cumulative effect of several preceding cycle lengths — i.e. mainly the basic drive cycle length
  • Conduction system ERP is predominantly influenced by the immediately preceding cycle length
• This is the whole lever: these phenomena and pacing protocols can be used to augment and to reverse the differences in ERP between septal myocardium and LBB

The two divergent responses, and how to tell them apart

• Upper panel — short-long sequence preceding S3: LBB ERP is prolonged and myocardial ERP is shortened, so septal myocardial-only capture QRS was seen at S3. Confirmed by loss of r′ in V1 and by a retrograde LBB potential (dashed arrow) after the paced QRS
• Lower panel — long-short sequence preceding S3: myocardial ERP is prolonged and LBB ERP abruptly shortened, so S3 selectively captures the LBB. Evidenced by four findings together:
  1. Augmentation of the R wave in V1
  2. Lack of retrograde LBB potential
  3. Presence of a latency interval after the pacing stimulus
  4. Appearance of a discrete deflection on the ventricular intracardiac electrogram
• A subtlety worth retaining: the selectively captured QRS morphology differs from the selective QRS that forms part of the fused nonselective drive-train QRS, and the latency interval is longer than the LBB potential-to-QRS interval. The author attributes both to LBB capture occurring within the relative refractory period, leading to some conduction aberrancy / incomplete capture of the LBB

The quantitative result — the ERP gap is protocol-dependent

• Using Denker protocols, the ERP relationship was not merely widened but inverted in the same patient:

• The comparison the author draws: the recent report (the companion paper) found a difference of only 55 ms (263 vs 318 ms) using a 600 ms basic drive train, which produced ~20% of cases where a diagnostic response could not be evoked
• Consequently, myocardial paced QRS and selective LBB paced QRS were observed over a wide range of coupling intervals in this case
• Two further figures the author cites against the standard protocol:
  • With a 600 ms drive train, a selective LBB paced QRS response was observed in only 2 of 143 cases in that study
  • In contrast, "in the author's experience," a long-short-short protocol either with a slow basic drive train or using double extra stimuli during intrinsic rhythm "provides selective LBB QRS response very often"

The author's own stated limitations of the technique

• False positives are possible. A change in QRS morphology with short-coupled extrastimuli may reflect delay in myocardial conduction when the myocardium's relative refractory period is encroached upon. Therefore a small (i.e. <30 ms) difference in ERP between LBB and myocardium produces a response that is difficult to interpret categorically → a false-positive response. The Denker protocols limit this phenomenon but do not eliminate it
• Underestimation of LBB ERP. Significant conduction delay within the LBB with short-coupled extrastimuli may produce a shift to myocardial QRS morphology despite LBB capture. This would not cause a wrong diagnosis, but it underestimates the ERP of the LBB
• The general caution: "during the interpretation of programmed stimulation in patients with conduction system pacing, limitations of this technique have to be kept in mind"

Conclusion

• The programmed stimulation response in this case proved that direct LBB capture had been achieved, and that the lack of QRS morphology change with lowering pacing output was related to equal capture thresholds of both tissues — not to a lack of LBB capture
• The author's proposal: Denker programmed stimulation protocols might significantly increase the diagnostic yield of programmed stimulation in LBB pacing cases
• Framed as a belief ("This author believes"), not a demonstrated result

Limitations of the document

Authors' own stated limitations

• False-positive responses are possible when the ERP gap is small (<30 ms), because encroachment on the myocardium's relative refractory period delays myocardial conduction and changes QRS morphology independently of capture. The Denker protocols limit but do not eliminate this
• Intra-LBB conduction delay can shift the QRS to a myocardial morphology despite LBB capture, causing the LBB ERP to be underestimated — a mis-measurement rather than a mis-diagnosis, but one that degrades the quantitative output
• The technique's interpretation requires electrophysiological judgement, and the limitations "have to be kept in mind"

Additional limitations evident in the study

• The evidence is a single case (n=1) in a single patient, reported by a single author, with no control and no comparator protocol series. The entire quantitative claim of a 150 ms separation and its reversal comes from one patient's one set of tracings. Whether these ERP values or this responsiveness generalise to other patients, septal thicknesses, lead depths or substrates is not tested
• The central practical claim is explicitly the author's experience, not a measurement. "In the author's experience, a long-short-short protocol … provides selective LBB QRS response very often" — "very often" is unquantified, and there is no denominator, no consecutive series, and no comparison against the 2-of-143 figure it is contrasted with. The rhetorical contrast (2/143 vs "very often") is therefore not a like-for-like comparison, and no statistical claim is or can be made
• The comparison to the companion study is cross-patient and post hoc. The 55 ms figure comes from a different cohort studied with a different protocol; the 150 ms and 40 ms figures come from one patient in this paper. Presenting them in the same frame as though they measured the same thing is the paper's weakest inference
• The patient is an atypical substrate for the question. An 18-year-old with congenital complete AV block and a 14-year history of epicardial RV pacing is not representative of the typical LBBP population (who are elderly and often have acquired conduction disease) — and congenital AV block with a narrow escape rhythm is a specific, unusual physiological setting. The ERPs reported may reflect this substrate
• The mechanism is inherited, not discovered here. The divergent cycle-length dependence of myocardial versus conduction-system refractoriness is Denker et al. 1983 (sources/denker-hps-refractoriness-circ-1983, rating: high) — a human study of 30 patients, not a canine one. (The canine study routinely confused with it is Han and Moe, 1969, which found the opposite: parallel behaviour in both tissues, and is the principal contradicting evidence for the dissociation this paper relies on.) This paper's contribution is applying a known protocol to a new target (LBB capture), which is a genuine and useful idea but not new physiology
• The mechanistic explanation is open at BOTH ends of the chain. The 1983 source did not derive a mechanism either — it demonstrated the divergence and stated explicitly that "an exact mechanism for this divergence in behavior cannot be derived from our study", offering the rate-of-change account only as a consistency argument and acknowledging an unmeasurable Purkinje-muscle junction confound. So what this paper applies is a phenomenon (1983) as a protocol (2020), with the underlying explanation unresolved in both
• ⚠ Correction to an earlier version of this page. Denker et al. 1983 was described here as a "canine/clinical study". It is a human study (n=30, Milwaukee). Corrected on ingesting the primary source, sources/denker-hps-refractoriness-circ-1983
• No clinical consequence is demonstrated. Establishing that the LBB was captured does not show that knowing so changed management or outcome. The case's ambiguity was resolved intellectually; the paper does not report what was done differently as a result
• The ambiguity this technique resolves has an easier route that the paper does not discuss. A no-change-in-QRS-on-output-reduction finding is what the output-dependent QRS-transition and V6RWPT criteria are designed for; the paper does not compare programmed stimulation against those surface-ECG criteria in this case, nor address cases where the LBB potential is absent (it was present here, which is what made the deep septal position credible in the first place)
• Rating rationale: a single-case teaching article with no cohort, no control, no systematic protocol comparison, and whose key practical claim is stated as personal experience. Its value is mechanistic and hypothesis-generating — it supplies a concrete, testable explanation for the companion study's nondiagnostic rate and a specific protocol to test. But it sits below the medium floor that every other LBBAP source in this wiki occupies, including the 30-patient feasibility study and the 45-patient partition study — hence low

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the technique; this paper addresses how to confirm LBB capture within it
• concepts/Conduction-System-Pacing — the umbrella category
• concepts/His-Bundle-Pacing — the initially intended modality, abandoned when the His bundle could not be localised with a C315His sheath
• Effective refractory period (ERP) differential — myocardium vs LBB; the maneuver's entire basis, and here shown to be protocol-dependent (55 ms at a 600 ms drive train; 150 ms or reversed with Denker protocols)
• Denker protocols — programmed stimulation exploiting divergent cycle-length dependence of myocardial (cumulative) versus conduction-system (immediately preceding) refractoriness; the paper's proposed fix for the nondiagnostic rate
• Programmed extrastimulus technique — the maneuver; an extrastimulus encroaching on one tissue's ERP produces a sudden QRS morphology change identifying which tissue was captured
• "Myocardial" vs "selective LBB" response — the two diagnostic categories, with a nondiagnostic third at 20.3% in the companion study
• Retrograde LBB potential — the marker of myocardial-only capture here (present after the paced QRS); its absence plus V1 R-wave augmentation marks selective LBB capture
• Relative refractory period / latency interval — the proposed explanation for why the selectively captured QRS differs from the selective component of the drive-train QRS

Key Entities Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — primary concept page
• Medtronic SelectSecure 3830 — the pacing lead used (lumenless, 4.1 Fr); the same lead as in the companion study
• Medtronic C315His — the fixed-curve delivery sheath used; the stated reason His bundle localisation failed, and the reason the same setup was retained for LBBP
• Medtronic and Abbott — disclosed consultant and lecture fees to the author
• Jagiellonian University Medical College, Kraków — the single investigating centre; the same group as sources/lbb-capture-criteria-hrs-2021 and sources/programmed-septal-stim-lbbp-jce-2020
• Denker et al. (1983), Circulation 68:1212–21 — the source of the divergent-refractoriness principle the paper applies
• Wiley Periodicals — publisher (J Cardiovasc Electrophysiol)

Wiki Pages Updated

• Created: wiki/sources/denker-pes-lbbap-jce-2020.md
• Updated: wiki/concepts/Left-Bundle-Branch-Area-Pacing.md
• Updated: wiki/sourceindex.md
• Updated: wiki/wikiindex.md
• Later the same day (2026-10-16), revised during the ingest of sources/denker-hps-refractoriness-circ-1983: the Denker 1983 attribution was corrected from "canine/clinical" to a human study of 30 patients (the canine paper is Han and Moe, which found the opposite), the primary source was linked, and a note was added that the 1983 paper also did not derive a mechanism