The V6–V1 Interpeak Interval: A Novel Criterion for the Diagnosis of Left Bundle Branch Capture

Authors, Journal, Affiliations, Type, DOI

• Authors: Marek Jastrzębski (corresponding); Haran Burri; Grzegorz Kiełbasa; Karol Curila; Paweł Moskal; Agnieszka Bednarek; Marek Rajzer; Pugazhendhi Vijayaraman
• Journal: Europace 2022;24(1):40–47. Received 19 March 2021; accepted 19 June 2021; published online 12 July 2021. Open access (CC BY)
• Affiliations: First Department of Cardiology, Interventional Electrocardiology and Hypertension, Jagiellonian University Medical College, Kraków; Cardiac Pacing Unit, University Hospital of Geneva; Third Faculty of Medicine, Charles University and University Hospital Královské Vinohrady, Prague; Geisinger Heart Institute, Wilkes-Barre
• Type: Two-centre diagnostic-accuracy study — 468 patients screened, 124 included (239 ECGs): 124 during ns-LBB capture, 69 with transition to s-LBB, 46 with transition to LVS
• DOI: 10.1093/europace/euab164
• Disclosures: Vijayaraman — research, fellowship, speaker and consultant fees from Medtronic; consultant fees from Abbott, Biotronik, Eaglepoint and Boston Scientific. Jastrzębski, Curila and Moskal — consultant fees from Medtronic
• ⚠ Cohort overlap: this paper and the already-ingested sources/lbb-capture-criteria-hrs-2021 analyse the same 124 patients drawn from the same 468 screened at the same two centres. They are companion analyses, not independent studies — which explains why the 2022 systematic review's Table 1 counts the cohort twice

Overview

Every ECG criterion for left bundle capture had been built on the absolute V6 R-wave peak time, measured from the pacing stimulus. That measure is contaminated by everything between stimulus and QRS onset — latency, slow conduction through a diseased His–Purkinje system, and LV dilatation — so it is specific but insensitive. This study proposes a relative measure instead: the difference between the R-wave peak in V6 and the R-wave peak in V1, which uses the patient's own RV activation as the reference time point and therefore cancels any common delay. Across 124 patients with directly demonstrated capture, the interval was longest in selective LBB capture (62.3 ± 21.4 ms), intermediate in non-selective LBB capture (41.3 ± 14.0 ms) and shortest in LV septal capture (26.5 ± 8.6 ms). A cut-off of >33 ms separated ns-LBB from LVS capture with sensitivity 71.8% and specificity 90.0%, and >44 ms was 100% specific for LBB capture. The paper also supplies the mechanistic explanation — loss of septal myocardial capture delays only RV activation; loss of LBB capture delays only LV activation — and turns that into two transition criteria: a V1RWPT rise ≥15 ms identifies transition to selective capture (SN 59.4%, SP 95.6%), and a V6RWPT rise ≥15 ms identifies transition to LVS capture (SN 82.6%, SP 100%).

Keywords

Left bundle branch pacing; Left bundle branch capture; Left ventricular septal capture; Conduction system pacing; Electrocardiogram

Key Takeaways

What's new (the authors' own four points)

1. V6–V1 interpeak >33 ms differentiates ns-LBB from LVS capture — accurate, easy to measure, reproducible
2. Using the V1 R-wave peak as the intra-QRS reference rather than the pacing stimulus circumvents pacing latency and intraventricular conduction delay
3. Loss of LVS myocardial capture raises V1RWPT by ≥15 ms but does not affect V6RWPT — implying RV activation during ns-LBB pacing proceeds mainly by the direct transseptal route, not retrograde RBB conduction
4. Loss of LBB capture raises V6RWPT by ≥15 ms with 100% specificity and barely changes V1RWPT

The problem with V6RWPT

• The previously proposed fixed criteria — V6RWPT <75 ms for narrow QRS/RBBB and <80 ms for LBBB/IVCD/escape/asystole — are nearly 100% specific but overlap substantially between ns-LBB and septal myocardial capture, with LBB capture possible even at >100 ms
• The measure is influenced not only by capture but by initial latency and intraventricular conduction velocity
• And no QRS-based criterion existed for ns-LBB versus s-LBB at all

Methods — how "capture" was established

• Consecutive patients receiving LBB area pacing for bradycardia and/or heart failure were screened
• Inclusion required direct evidence of LBB capture during dynamic manoeuvres — a paced QR/rSR′ in V1 plus at least one of:
  1. ns-LBB → s-LBB transition as output falls
  2. ns-LBB → LVS transition as output falls
  3. ns-LBB → s-LBB during programmed stimulation or burst/incremental pacing
• Only unipolar pacing was used for diagnosis, to avoid anodal capture
• Measurements: all 12 leads simultaneously on a digital EP system (LabsystemPRO), digital calipers, 200 mm/s, signal augmentation, mean of ≥3 complexes
  • V1RWPT and V6RWPT measured from the pacing stimulus
  • V6–V1 interpeak = V6 R-wave peak minus V1 R-wave peak; for native and s-LBB beats, measured from QRS onset
• Statistics: Fisher exact, t-test/ANOVA, ROC, Bland–Altman interobserver analysis; R

Results

Population

• 468 screened → 124 included; the rest were excluded because capture had been confirmed only by non-direct criteria (LBB potential, V6RWPT, QRS duration)
• Age 74.7 ± 11.4; male 71 (57.3%); LVEF 46.2 ± 15.5%; LVEDD 53.0 ± 8.7 mm; native QRS 133.6 ± 34.8 ms
• Indications: sick sinus 25.8%, AV block 33.1%, AF with bradycardia 10.5%, heart failure 30.6%
• Native QRS types: narrow 31.6%, LAFB 4.0%, RBBB 10.5%, RBBB + fascicular block 12.1%, NIVCD 12.1%, LBBB 14.5%, escape with RBBB-type 10.5%, escape with LBBB-type 1.6%
• 239 ECGs: 124 ns-LBB, 69 transition to s-LBB, 46 transition to LVS
• Capture diagnosis: transition to s-LBB at threshold test 55.6%, transition to LVS 37.1%, selective response at programmed stimulation 62.1%; LBB potential observed 57.2%
• Acute LBB capture threshold 0.81 ± 0.4 V; ventricular sensing 8.6 ± 4.7 mV; fluoroscopy 11.4 ± 9.3 min
• Paced QRS: ns-LBB 154.5 ± 21.2, s-LBB 144.5 ± 24.4, LVS 159.3 ± 20.2 ms

The V6–V1 interval by capture type

• >33 ms for ns-LBB vs LVS: SN 71.8%, SP 90.0%, AUC 84.7% (optimal cut-off 32.5 ms)
• >44 ms for LBB capture: SP 100% — at the cost of low sensitivity
• Why long values are diagnostic even though short values are not: a short interval occurs in all three capture types, but a long interval is possible only with LBB capture, because it requires delayed RV activation

The transition criteria, and the physiology behind them

• ns → s-LBB: V1RWPT rises 17.8 ± 10.0 ms (120.7 ± 16.7 → 138.5 ± 21.5 ms, P = 0.001) while V6RWPT is unchanged (77.2 ± 13.6 vs 76.6 ± 14.1 ms, P = 0.36) — because septal myocardial capture is lost, delaying RV activation only
  • V1RWPT rise ≥15 ms: SN 59.4%, SP 95.6% for transition to selective capture (met in 41/69 vs 2/46)
• ns → LVS: V6RWPT rises 19.9 ± 6.7 ms (78.4 ± 10.8 → 98.4 ± 13.9 ms) while V1RWPT rises only 6.2 ± 6.3 ms (119.3 ± 14.5 → 125.6 ± 13.8 ms, P < 0.001) — because LBB capture is lost, delaying LV activation only
  • V6RWPT rise ≥15 ms: SN 82.6%, SP 100% for transition to LVS capture
• The two wavefronts explain the LVS QRS: in LVS capture, RV and LV activate in opposite directions from the same septal site, producing substantial cancellation in V1 — a relatively narrow QRS without pronounced RBBB or LBBB features, and V1 and V6 R-wave peaks occurring at almost the same time

The clinical payoff — rescuing false negatives

• V6–V1 is independent of V6RWPT; the V6RWPT distribution was similar in patients with diagnostic and non-diagnostic intervals
• It correctly reclassified 44/69 (63.8%) cases falsely negative by V6RWPT >75 ms
• Combined criterion (V6RWPT <75 ms OR V6–V1 ≥33 ms): SN 78.2%, SP 94.2%
• V6–V1 is not affected by native QRS type (P = 0.47), whereas V1RWPT and V6RWPT both lengthen with conduction disease (P < 0.001) — the property that makes it a relative measure
• Latency was present in every case and correlated with V6RWPT in both capture types, but did not correlate with V6–V1
• Interobserver agreement: V6–V1 mean difference 0.97 ms (95% CI −0.5 to 2.44), limits of agreement −11.45 to 13.40 ms; both observers agreed on LBB capture in 88% of cases

Where the criterion fails

• In 14/69 (20.3%) the transition to s-LBB did not prolong V1RWPT (difference ≤10 ms). 4 of these 14 (28.7%) had an LBB potential-to-QRS interval ≥30 ms, versus 1/55 (1.8%) of those in whom V1RWPT did rise (P = 0.005) — i.e. very proximal leads activate the RV retrogradely via the right bundle, masking the loss of septal capture
• The authors concede that in some LVS cases near-instantaneous secondary activation of conduction tissue may be indistinguishable from ns-LBB capture — though clinically equivalent

Limitations of the document

Authors' own stated limitations

• Retrograde right bundle branch activation is a potential confounder of the proposed RV/LV activation patterns. Because 64.4% had RBBB, LBBB or complete AV block — conditions likely to preclude retrograde RBB conduction — the findings may not apply to patients with preserved conduction
  • Their counter-evidence: narrow-QRS patients had a minimal 3 ms difference in V1RWPT versus RBBB/LBBB patients; narrow-QRS patients still prolonged V1RWPT on transition to selective capture; and in intrinsic RBBB, V1RWPT during s-LBB pacing equalled the native value while ns-LBB pacing shortened it
• Measurement precision: measurements required an EP system at 200 mm/s with digital calipers; a compass and 25 mm/s printouts "may be less accurate", and borderline cases may be hard to categorise
• Some LVS cases with instantaneous secondary conduction-tissue activation may not be distinguishable from ns-LBB by this criterion

Additional limitations evident in the study

• The criterion is derived in, and from, the minority it is meant to replace. Only 124 of 468 (26.5%) were included, because direct capture evidence was required. The V6–V1 interval is therefore validated in patients who had a diagnostic transition, while its clinical value would lie in the patients who do not — and it is never tested there. This is the same structural criticism levelled at the 2025 morphology-only criteria
• The reference standard is circular to a degree. Capture was defined by the very transition phenomena the paper then uses to validate a new marker of those transitions. The 100% specificity of a ≥15 ms V6RWPT rise is a description of the derivation cohort, not an independent validation
• No external validation and no confidence intervals for the diagnostic metrics. The cut-offs (33 and 44 ms) were optimised on the same 124 patients, so SN/SP are optimistic; no hold-out or second cohort is reported
• The >44 ms rule is a rule-in test only. Specificity 100% with "low" sensitivity (not formally reported for this cut-off) means the criterion cannot exclude capture — and the more useful >33 ms threshold has SN 71.8%, so nearly three in ten true captures are missed
• Interobserver limits of agreement (±12–13 ms) are wide relative to the 33 ms threshold — roughly a third of the decision boundary — so borderline measurements are genuinely ambiguous between observers
• The 20.3% failure rate on transition to selective capture is not random: it concentrates in very proximal leads (LBB potential–QRS ≥30 ms), which are exactly the implants where the operator most needs to know whether septal capture was lost. The criterion is weakest where the anatomy is most proximal
• Only unipolar pacing was used for diagnosis (correctly, to avoid anodal capture), but the criteria's performance under bipolar pacing — routine at follow-up — is not reported
• No outcomes, no follow-up, no comparison with a clinical endpoint. The paper establishes that the interval tracks capture type acutely; whether using it improves implant success, reduces perforation, or predicts remodelling is untested
• Cohort composition is elderly and highly comorbid (79.8% hypertension, 41.1% coronary disease, 39.5% diabetes), with LVEF 46% — not a general pacing population, and not the CRT population where capture adjudication matters most
• Rating rationale: rated high. It is a two-centre study with an unusually strict capture definition, a clean physiological mechanism, formal diagnostic metrics, and a criterion that has since been adopted into consensus practice. It is not very high because the cut-offs are internally derived without external validation, the criterion is validated only in the 26% with a diagnostic transition, the >44 ms rule is insensitive, interobserver agreement is modest at the decision boundary, and there are no outcomes

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the technique; this paper supplies its relative ECG criterion
• concepts/Conduction-System-Pacing — the umbrella category
• V6–V1 interpeak interval — RV-to-LV R-wave peak difference; longest in selective LBB, intermediate in non-selective LBB, shortest in LV septal capture
• V6RWPT / V1RWPT — absolute R-wave peak times measured from the pacing stimulus; both contaminated by latency and conduction disease
• Non-selective (ns), selective (s) LBB capture, and LVS myocardial capture — the three capture states the paper separates
• Pacing latency — the stimulus-to-QRS interval present in all cases; the confounder the relative measure cancels
• Anodal capture — avoided by using unipolar pacing for diagnosis
• Retrograde right bundle branch conduction — the alternative RV activation route that can mask loss of septal capture in very proximal leads
• LBB potential-to-QRS interval — a marker of lead proximity to the His–LBB junction; ≥30 ms predicted non-prolongation of V1RWPT
• Cancellation effect in V1 — how LVS capture produces a narrow QRS without a pronounced RBBB pattern

Key Entities Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — primary concept page
• Jagiellonian University Medical College, Kraków — coordinating centre
• University Hospital of Geneva; Charles University, Prague; Geisinger Heart Institute — collaborating centres
• Marek Jastrzębski; Haran Burri; Karol Curila; Pugazhendhi Vijayaraman — the investigators who developed the physiology-based ECG criteria for LBB capture
• Boston Scientific LabsystemPRO — the EP recording system used for measurement
• Medtronic, Abbott, Biotronik, Eaglepoint, Boston Scientific — disclosed industry relationships of the senior authors
• Oxford University Press / European Society of Cardiology / Europace — publisher

Wiki Pages Updated

• Created: wiki/sources/v6-v1-interpeak-europace-2022.md
• Updated: wiki/concepts/Left-Bundle-Branch-Area-Pacing.md
• Updated: wiki/concepts/Conduction-System-Pacing.md
• Updated: wiki/sourceindex.md
• Updated: wiki/wikiindex.md