Non-Invasive Assessment of Ventricular Electrical Heterogeneity to Optimize Left Bundle Branch Area Pacing

Authors, Journal, Affiliations, Type, DOI

• Authors: Pugazhendhi Vijayaraman, Grace Hughes, Marilee Manganiello, Alicia Johns, Subham Ghosh
• Journal: Journal of Interventional Cardiac Electrophysiology 2023;66(5):1103–1112 (published online 30 July 2022)
• Affiliations: Cardiac Electrophysiology, Geisinger Heart Institute, Geisinger Commonwealth School of Medicine, Wilkes Barre, PA; Clinical Trials Unit and Department of Population Health Sciences, Geisinger Health System; Medtronic Inc, Mounds View, MN
• Type: Prospective single-centre observational study (ClinicalTrials.gov NCT04583709)
• DOI: 10.1007/s10840-022-01315-9

Overview

This study used a 40-electrode body-surface ECG belt to characterise ventricular electrical heterogeneity (VEH) during different forms of left bundle branch area pacing (LBBAP) in 20 patients, most with a CRT indication. It measured global heterogeneity as the standard deviation of activation times (SDAT) across all belt electrodes, plus average LV activation time (LVAT) and separate LV and RV dispersion. Optimised LBBAP reduced every metric substantially, with a 47% fall in SDAT but only a 16% fall in QRS duration, and produced ~80% clinical and echocardiographic response at ~8 months. The study's most distinctive findings are that the electrical benefit was independent of capture mechanism — selective, nonselective, anodal and LV septal capture could not be separated — and that anodal capture worsened LVAT and LV dispersion despite shortening QRS.

Keywords

Cardiac resynchronization therapy; ECG belt; Left bundle branch pacing; LV septal pacing; Anodal capture

Key Takeaways

Background and rationale

• CRT improves symptoms, function and LV function in LBBB with prolonged QRS, but a significant proportion of patients do not respond; proposed reasons include suboptimal electrical resynchronisation and mixed conduction disease with peripheral conduction delay.
• A 53-electrode ECG belt had previously shown that a >10% reduction in SDAT correlated with improvement in LVEF and LV end-systolic volume.
• LBBAP is used as an alternative means of achieving CRT, but the electrical differences between its capture forms were unclear.

Definitions used

• LBBAP: pacing in the region of the LBB.
• LBBP: LBB capture is confirmed.
• Nonselective LBBP: at programmed output, capture of both the LBB and LV septal myocardium.
• Selective LBBP: at near-threshold output, capture of the LBB alone.
• LV septal pacing: LV myocardial capture without evidence of LBB capture.
• Anodal capture: RV septal myocardial capture from the ring electrode, which can further narrow QRS.

Methods

• 20 patients (15 male), mean LVEF 33 ± 13%, NYHA 3.05 ± 0.6; CRT indication in 18, AV node ablation in one patient with HCM and one with Emery-Dreifuss muscular dystrophy.
• ECG belt: 40 electrodes around the upper torso, unipolar recordings, 5-second acquisition, automatic representative-beat selection and activation mapping.
• Metrics: SDAT (global electrical heterogeneity), average LVAT (mean activation time on left-sided electrodes), LV dispersion and RV dispersion (SDAT of left- and right-sided electrodes respectively).
• Optimisation rule: maximal SDAT reduction, then maximal LVAT reduction, then QRS duration reduction.
• Configurations tested: LBBAP 20, nonselective LBBP 18, anodal capture 16, selective LBBP 5, LV septal pacing 9, LOT-CRT 7.
• LBBB definition: Strauss criteria (QRS >130 ms women, >140 ms men, with ≥2 mid-QRS notches/slurs in I, aVL, V1, V2, V5 or V6). Mixed conduction disease was defined as persistent QRS >120 ms after LBBB correction by selective HBP.

Baseline characteristics

• Age 65 ± 18 years; hypertension 80%; CAD 60%; AF 40%; CKD 35%.
• Cardiomyopathy in 19/20: ischaemic 50%, non-ischaemic 40%, hypertrophic 5%, muscular dystrophy 5%.
• Conduction disease: complete LBBB 9 (45%), mixed conduction disease 4 (20%), non-specific IVCD 3 (15%), RBBB 1, complete AV block with RV pacing 3.
• Baseline QRS 156 ± 24 ms, narrowing to 124 ± 18 ms during optimised LBBAP.

Procedural characteristics

• Devices: CRT-D 14 (70%), CRT-P 4 (20%), DDD pacemaker 2 (10%).
• LBBAP alone 13 (65%); LOT-CRT 7 (35%).
• LBB capture confirmed in 18; LV septal-only pacing in 2.
• Of 18 nonselective LBBP patients, stable selective LBBP (n=5) or LV septal pacing (n=7) could be produced at near-threshold output in 12; in the other 6, output-dependent transitions seen at implant were not reproducible at follow-up.

ECG belt results

• Optimised LBBAP vs intrinsic rhythm (all significant):

• Anodal capture added to nonselective LBBP did not improve QRS duration, SDAT or RV dispersion, and significantly worsened LVAT (51.9 ± 11 vs 47.5 ± 11 ms, p = 0.01) and LV dispersion (10.5 ± 7.4 vs 8.7 ± 7.1 ms, p = 0.01).
• Selective vs nonselective LBBP: no significant difference in any ECG parameter.
• LV septal pacing vs nonselective LBBP: non-significant worsening of all parameters.
• LOT-CRT vs LBBAP: significant shortening of LVAT (22.7 ± 14.4 vs 43.9 ± 15.2 ms, p = 0.003); other parameters improved non-significantly.

Echocardiographic and clinical outcomes

• Mean follow-up 8.2 ± 2 months.
• LVEF 32.6 ± 13.3% → 45.8 ± 10.3% (p = 0.001); LV end-diastolic diameter 60.7 → 52.2 mm; end-diastolic volume 206.7 → 151.7 mL; end-systolic volume 140.6 → 87.9 mL.
• Echocardiographic response (≥5% LVEF improvement) in 15/18 (83%) with reduced LVEF; improvement or preservation of LV function in 17/20 (85%); hyper-response in 8/18 (44%).
• NYHA 3.03 ± 0.6 → 1.9 ± 0.6 (p < 0.0001); clinical response in 16/20 (80%).

Correlation between SDAT reduction and response

• SDAT reduction >10% in 18/20 (90%); >20% in 15/20 (75%).
• Of the 18 with >10% SDAT reduction, 16 (89%) had echocardiographic response or preserved normal EF; of the 15 with >20% reduction, 13 (87%) did.
• SDAT reduction appeared independent of paced QRS duration.

Discussion points

• ECG belt metrics had not previously been studied during permanent LBBAP.
• Salden et al. had shown temporary LV septal pacing produced larger SDAT reductions than biventricular pacing, and that adding RV apical pacing to LV septal pacing worsened acute haemodynamics — consistent with this study's finding that anodal capture worsens LVAT and LV dispersion.
• The authors argue that the true impact of LBB capture versus LV septal-only pacing must be understood before LBBAP can be first-line CRT; their experience suggests LBBP activates the LV free wall 20–30 ms earlier than LV septal pacing, but whether this translates into outcomes is unknown.
• They cite Jastrzębski et al. for the observation that, in LOT-CRT, patients with evidence of LBB capture had greater clinical and echocardiographic response than those without.

Limitations of the document

• Very small (n=20), single-centre, observational study with a heterogeneous cohort of varying conduction disease; the authors state it is underpowered and should be considered hypothesis-generating.
• A 40-electrode belt may lack the spatial resolution to detect real differences between selective, nonselective, anodal and LV septal capture — so the null results between capture forms may be a measurement floor rather than a true equivalence.
• No control group and no randomisation; echocardiographic and clinical outcomes are uncontrolled comparisons against baseline.
• Follow-up is short (8.2 ± 2 months) and the response definition is an arbitrary ≥5% LVEF threshold.
• The optimisation rule (SDAT first, then LVAT, then QRS) was not validated against outcomes — patients were optimised by it and then found to respond, which is circular.
• Industry involvement is direct: the study was funded by a Medtronic External Research Program, one author is a Medtronic employee, and the senior author reports extensive Medtronic consultancy, speaking and research support; the ECG belt is a Medtronic device.
• No hard clinical endpoints (mortality, heart failure hospitalisation).
• The quantitative claims in the text and figures do not always agree — e.g. the abstract reports p < 0.0001 for LVAT while the results text reports p = 0.008, and the abstract's LVEF result is given as 46 ± 11% against 45.8 ± 10.3% in the results.

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the intervention being optimised
• concepts/Cardiac-Resynchronization-Therapy — the clinical context and comparison standard
• concepts/Electrical-Resynchronization-Metrics — SDAT, LVAT and dispersion as measures of electrical dyssynchrony
• concepts/Conduction-System-Pacing — LBBAP as a form of physiological pacing

Key Entities Mentioned

• Medtronic — ECG belt system, study funding, and an author employment relationship
• Geisinger Heart Institute — coordinating centre
• ECG Belt Clinical Research system — the 40-electrode body-surface mapping device
• LOT-CRT — LBBAP combined with coronary-venous LV pacing

Wiki Pages Updated

• Created: wiki/sources/crp-lbbap-jice-2023.md
• Created: wiki/concepts/Electrical-Resynchronization-Metrics.md
• Updated: wiki/concepts/Left-Bundle-Branch-Area-Pacing.md
• Updated: wiki/concepts/Cardiac-Resynchronization-Therapy.md
• Updated: wiki/sourceindex.md
• Updated: log.md