#left-bundle-area-pacing #cardiac-resynchronization-therapy #heart-failure #conduction-system-pacing

LBBAP Outcomes

Definition

This page covers the electrical, haemodynamic and clinical results of left bundle branch area pacing (LBBAP): how it compares with right ventricular pacing (RVP), biventricular pacing (BVP) and left ventricular septal pacing (LVSP); how long the lead performs; and what complications and follow-up it requires. LBBAP was developed because RV apical pacing causes dyssynchrony and because His bundle pacing fails at the site of block. (sources/csp-jaccep-2023, rating: very high) (sources/lbbap-jacc-2019, rating: high)

Key Concepts

Acute haemodynamics and resynchronisation

LBBAP vs BVP (Liang et al.): LBBAP produced greater QRS reduction (−11 ms; P = .003) and a 6% greater increase in LV dP/dtmax (P = .002). (sources/csp-jaccep-2023, rating: very high)
LBBP RESYNC RCT (n=40): LBBP was superior to BVP for LVEF improvement, LVESV reduction and NT-proBNP reduction at 6 months.
LEVEL-AT RCT (n=70): conduction-system pacing (predominantly LBBAP) was non-inferior to BVP on LVAT, with greater LVAT reduction in per-protocol analysis; 23% crossed over from CSP to BVP.
HBP-CRT vs LBBAP-CRT (Ali et al., n=19): His bundle pacing achieved faster total biventricular activation; LBBAP was non-inferior for LVAT reduction, and delayed RV activation did not adversely affect haemodynamic response.
In LBBB with heart failure (n=88): paced QRS fell 167 → 124 ms, and LVEF in baseline QRS ≥120 ms rose 48.8 → 58.1%. (sources/longterm-lbbap-circep-2021, rating: high)
No large powered RCT has reported clinical outcomes. Left vs Left (NCT05650658, n=2,136) is the definitive test still awaited. (sources/csp-jaccep-2023, rating: very high)

Body-surface mapping shows a larger electrical effect than the QRS change suggests. A 40-electrode belt in 20 patients (LVEF 33 ± 13%; CRT indication in 18) found:

Optimised LBBAP vs intrinsic: QRS 157 → 133 ms (−16%), but SDAT 38.6 → 20.5 ms (−47%), LVAT 61.4 → 44.4 ms (−28%), LV dispersion −61% and RV dispersion −50%.
Capture mechanism did not matter: selective, nonselective, anodal and LV septal capture were not significantly different in any metric, though the study was probably underpowered.
Anodal capture was worse, not better: adding it to nonselective LBBP lengthened LVAT (51.9 vs 47.5 ms) and increased LV dispersion (10.5 vs 8.7 ms) despite a shorter QRS.
LOT-CRT shortened LVAT (22.7 vs 43.9 ms, P = .003) significantly.
Response: LVEF 32.6 → 45.8%, NYHA 3.03 → 1.9, echocardiographic response 83%, clinical response 80% at 8.2 ± 2 months; a >10% SDAT reduction occurred in 90%.

(All: sources/crp-lbbap-jice-2023, rating: low)

Clinical outcome cohorts

Study	Design / n	Comparison	Composite outcome	HF hospitalisation	Mortality
LBBAP-CRT (JACC 2023)	Retrospective, n=1,778; LVEF ≤35%	LBBAP vs BVP	21% vs 28% (adjusted HR 1.495 for BVP)	12% vs 19%	Not significant
I-CLAS LVEF ≤50% (HRS 2025)	n=2,579; propensity-matched 1,560	LBBAP vs BVP	22.2% vs 30.8% (HR 0.81)	13.6% vs 21.8% (HR 0.63)	12.4% vs 18.2% (HR 0.82, P = .156)
I-CLAS HFmrEF (HRS 2025)	n=1,004	LBBAP vs BVP	22% vs 34% (HR 0.64)	13% vs 23% (HR 0.64)	Not significant
LBBP vs LVSP vs BIVP (JACC EP 2024)	n=415	Three-way	Freedom from HFH or death: 76.6% vs 48.4% vs 43.6%	—	—
LEFT-BUNDLE-CRT (EHJ 2026)	RCT, n=176, typical LBBB	LBBAP vs BVP	Response 94.6% vs 89.7% (RR 0.95, 95% CI 0.88–1.02) — non-inferiority not met	—	—
MADURAI (HRS 2025)	n=225, LBBB-associated NICM	Scar <10% vs ≥10%	6.9% vs 26.1% (HR 16.52)	—	—

(All rows: sources/lbbap-crt-jacc-2023, rating: high; sources/iclas-lbbap-crt-hrs-2025, rating: high; sources/iclas-hfmref-hrs-2025, rating: high; sources/lbbp-lvsp-crt-jaccep-2024, rating: high; sources/left-bundle-crt-ehj-2026, rating: high; sources/mri-lbbap-hrs-2025, rating: high)

The pattern is consistent: heart failure hospitalisation falls, mortality does not differ after adjustment, and the effect is largest in patients with LBBB and less myocardial scar. In the propensity-matched I-CLAS cohort, LVEF rose +11.5% with LBBAP vs +8.9% with BVP. In MADURAI, scar burden was the dominant modifier: patients with <10% scar had a 6.9% vs 26.1% composite of death, HFH or VT/VF, and LVEF normalised in 82.2% vs 25% at 2 years. (sources/mri-lbbap-hrs-2025, rating: high)

The first dedicated randomised trial in typical LBBB did not confirm non-inferiority in its intention-to-treat analysis. On-treatment response (92.8% vs 91.3%; RR 0.98, 95% CI 0.91–1.06) suggested comparable efficacy but was exploratory; crossovers were 20.7% vs 9.8%. (sources/left-bundle-crt-ehj-2026, rating: high)

LBBAP vs RVP

Evidence	Design / n	Comparison	Key finding
Direct cohort (Europace 2020)	Non-randomised; 237 LBBP vs 317 RVP; LVEF ≥50%	LBBP vs RVP	QRS 132 vs 178 ms; better thresholds and sensing; HFH 0 vs 1 (0.4%); CV mortality 0 vs 0; no difference in cardiac outcomes
OptimPacing (NCT04624763)	Ongoing RCT; n=683	LBBP vs RVP	Primary: death + HFH + pacing-induced cardiomyopathy
PROTECT-HF	Ongoing RCT; n=2,600	CSP vs RVP	Primary: CV death + HFH + QoL + upgrade; 48-month follow-up
LEAP (NCT04595487)	Ongoing RCT; n=470	LVSP vs RVP	Tests whether conduction-system capture is required

(Table: sources/lbbp-vs-rvp-europace-2020, rating: medium; sources/csp-jaccep-2023, rating: very high)

The direct cohort (237 LBBP vs 317 RVP; LVEF ≥50%) had lower thresholds at implant (0.58 vs 0.71 V) that converged by 18 months (0.73 vs 0.76 V), higher sensing (13.9 vs 11.6 mV) and lower impedance (~402 Ω). LBBP also caused a larger transient troponin rise (0.13 vs 0.04 ng/mL at day 1), required more attempts (2.41 vs 1.49) and longer fluoroscopy (9.15 vs 3.92 min). (sources/lbbp-vs-rvp-europace-2020, rating: medium)

Hard-outcome evidence is pending. OptimPacing (NCT04624763, n=683) is testing LBBP vs RVP for death, HFH and pacing-induced cardiomyopathy; PROTECT-HF (n=2,600) tests CSP vs RVP; and LEAP (n=470) tests LVSP vs RVP. (sources/csp-jaccep-2023, rating: very high)

What it does and does not show. The study establishes electrical superiority and procedural safety; it is underpowered and too short to show clinical benefit. No completed randomised trial in this evidence base has compared LBBAP with RVP for death, heart failure hospitalisation or pacing-induced cardiomyopathy. OptimPacing is designed to test this directly; PROTECT-HF and LEAP address related questions. Until they report, the claim that LBBAP prevents pacing-induced cardiomyopathy remains an extrapolation from electrical surrogates. (sources/csp-jaccep-2023, rating: very high)

The rationale is that RV apical pacing causes electrical and mechanical dyssynchrony and increases heart failure and AF risk, and alternative RV sites are not superior. (sources/lbbap-jacc-2019, rating: high)

Long-term lead performance

In 618 successful implants followed 18.6 ± 6.7 months (231 beyond 2 years), the unipolar LBB threshold was 0.65 ± 0.27 V at implant and 0.69 ± 0.24 V at 2 years — flat. Selective capture fell from 74.4% at implant to 30.9% at follow-up, attributed to a fall in the local myocardial threshold rather than loss of LBB capture. Complications were few — 2 intraprocedural and 1 late perforation, 2 revisions for dislodgement, no procedure-related death — but permanent RBB injury is reported as 39%, 55% and 20.4% in three different places in the same paper. (sources/longterm-lbbap-circep-2021, rating: high)

Tricuspid regurgitation

The long-term registry reported worsening in 11.1% and improvement in 31.4%. (sources/longterm-lbbap-circep-2021, rating: high)
A dedicated echo study found worsening by ≥1 grade in 29/89 (32.6%) over 19.0 ± 6.5 months, with 23 of the 29 having none/trivial or mild TR at baseline. (sources/tricuspid-regurgitation-lbbap-hrs-2023, rating: medium)
In the normal-valve subgroup (n=58), lead-to-tricuspid-annulus distance was 19.0 ± 7.6 vs 23.9 ± 5.4 mm (P = .006) in those who deteriorated; lead depth and septal thickness did not differ.
A distance ≤16.1 mm was an independent predictor (HR 0.20 for >16.1 mm, 95% CI 0.06–0.76), AUC 0.721 — suggesting the consensus 10–20 mm entry zone should be kept in its upper half when the tricuspid valve is normal.
The proposed mechanism is mechanical interference with the annulus or subvalvular apparatus.

Complications

Complication	Reported range
Septal perforation	0–14.1%, depending on detection method
Right bundle branch block	19.9% (6.3% permanent)
Complete heart block	9.4% acute (2.6% permanent)
Tricuspid regurgitation worsening	7.3–32.6%, higher with proximal lead position
Lead dislodgement	0.3–1.5%
Loss of LBB capture	0.3–11.5%
Threshold rise >1 V	0.3–1.8%
Coronary artery fistula	1.4–2.0%
Troponin rise >3× ULN	49.4% from mechanical septal trauma

(Table: sources/csp-ehra-2023, rating: high; sources/septal-perforation-hrs-2022, rating: medium; sources/tricuspid-regurgitation-lbbap-hrs-2023, rating: medium)

Follow-up

More spaced than HBP; remote monitoring is acceptable if the device captures thresholds reliably. (sources/csp-ehra-2023, rating: high)
In person: check the paced 12-lead ECG, verify V6RWPT, and assess the terminal R in V1 — its loss suggests micro-dislodgement.
Loss of terminal R in V1 at follow-up was reported in 4% in MELOS. (sources/melos-lbbap-ehj-2022, rating: high)
Register the potential-to-V6 interval in the implant report for future follow-up monitoring. (sources/lbb-capture-criteria-hrs-2021, rating: high)

LOT-CRT

LOT-CRT combines an LBBAP lead with coronary-vein LV pacing. The full evidence base is on concepts/LOT-CRT. In brief:

Multicentre series (n=112, 81% success): QRS 182 → 144 ms, versus 7% reduction with BVP and 11% with LBBAP alone; LVEF 28.5 → 37.2%; NYHA 2.9 → 1.9. (sources/lot-crt-hrs-2021, rating: medium)
In IVCD (n=85, 24 months): QRS reduction 42.7 vs 21.9 ms; LVEF 37% vs 30.5%; echocardiographic response 60% vs 31.6%; composite cardiovascular death or HFH 6.7% vs 25.5% (HR 0.33). (sources/lot-crt-circep-2023, rating: medium)
CSPOT acute haemodynamics (n=48): LOT-CRT gave the greatest QRS shortening (29.5 ms) but LV dP/dtmax similar to BVP (25.8% vs 26.4%), both above LBBAP alone (19.3%); the incremental benefit over LBBAP was concentrated in QRS ≥171 ms with deep septal pacing (+14.5% dP/dtmax, 20.8 ms QRS). (sources/lot-crt-circep-2024, rating: high)
CSPOT echocardiographic substudy (n=34 vs propensity-matched BVP): LVEF +16.1% vs +6.1%, LVESV −43.5% vs −20.9%; adjusted differences remained significant. (sources/lot-crt-hrs-2025, rating: medium)
Timing: AV optimisation matters as much as for BVP (a 60 ms deviation ≈ 30% loss of haemodynamic benefit); PAV ≈ 0.56 × Ap–LBBs + 7.7 ms; VV = 0 optimal on average. (sources/lot-europace-2026, rating: high)
Single-lead defibrillation option: the 4.7-F OmniaSecure lead in the LBBA position had stable electrical performance and 96.8% freedom from lead-related major complications at 12 months. (sources/leadr-lbbap-hrs-2025, rating: high)

Defibrillation, antitachycardia pacing and substrate

LBBA defibrillation lead (LEADR): 316 attempts with 302 successes (95.6%), stable thresholds through 12 months, and 97.7% / 96.8% freedom from lead-related major complications at 6 / 12 months. Appropriate therapy was delivered in 7.9% and inappropriate shocks in 2.1%. In the LOT-CRT subgroup, QRS fell 170.6 → 157.5 ms, LVEF rose 29.7 → 38.1%, LVESVi fell 53.8 → 40.5, KCCQ improved by 14.2, and CRT response was 82%. (sources/leadr-lbbap-hrs-2025, rating: high)
LBBA antitachycardia pacing (canine ischaemia-reperfusion model): LBBA ATP terminated 70.2% of 79 VT episodes versus 47.3% for RV apical ATP (RR 1.44, 95% CI 1.03–2.01, P = .040), without more acceleration or VF. Crossover to the paced activation sequence occurred one beat earlier (4.1 vs 5.1 beats), and Purkinje activation was earlier (18.8 vs 39.8 ms). (sources/lbbap-atp-jaccep-2026, rating: medium)
His-bundle ATP, the proximal comparator: in the same canine model, His ATP did not improve termination over RV ATP (23 vs 25 successful trials) but reduced VT acceleration (6 vs 14, P = .05) and degeneration into VF (1 vs 7, P = .02). The efficacy signal therefore appears specific to left-sided Purkinje engagement rather than conduction-system pacing in general. (sources/his-atp-vt-jce-2023, rating: medium)
Proarrhythmia is possible despite acute resynchronisation. A 78-year-old with pacing-induced cardiomyopathy upgraded to LOT-CRT had QRS narrowing from 185 to 132 ms, but developed QT prolongation, T-wave inversion and PVC-triggered VF on day 1 without a reversible cause. She was treated with propranolol and had no recurrence over 10 days; no ICD was implanted. (sources/lot-crt-vt-ane-2026, rating: low)
Scar and conduction velocity determine which strategy works. In an in-silico model of 24 ventricular meshes, severe slowing of the LV His–Purkinje system attenuated conduction-system pacing while CRT performed better and HOT/LOT-CRT added further benefit; slow myocardium favoured CSP. Septal scar made CSP ineffective but CRT effective; lateral scar did the reverse. (sources/scar-csp-jce-2023, rating: medium)

Contradictions / Open Questions

Heart failure hospitalisation falls, but mortality does not. Three large cohorts agree on HFH (13–14% vs 19–23%) and none shows a significant mortality difference after adjustment. The I-CLAS LVEF ≤50% cohort came closest (12.4% vs 18.2%, HR 0.82, P = .156). A therapy that reduces hospitalisation without reducing death is useful but not disease-modifying, and the absence of a mortality signal may reflect confounding by indication rather than a true null. (sources/lbbap-crt-jacc-2023, rating: high) (sources/iclas-lbbap-crt-hrs-2025, rating: high)
The first randomised trial did not meet its non-inferiority margin. LEFT-BUNDLE-CRT reported ITT response 94.6% BiVP vs 89.7% LBBAP (RR 0.95, 95% CI 0.88–1.02); the on-treatment analysis met the margin (RR 0.98, 0.91–1.06) but was exploratory, and 20.7% of the BiVP arm crossed over to LBBAP. The trial therefore neither establishes LBBAP superiority nor excludes it; a large powered trial remains necessary. (sources/left-bundle-crt-ehj-2026, rating: high)
LBBAP is electrically superior to RVP, but clinical superiority has not been shown. In the only completed direct comparison, LBBP produced a much narrower paced QRS, lower thresholds and higher sensing, yet complications and cardiac outcomes were statistically identical — in a cohort with preserved LVEF, no CRT indication, 18 months of follow-up and 0–1% event rates. The study could not have detected a difference. OptimPacing (n=683) is the dedicated hard-outcome test. (sources/lbbp-vs-rvp-europace-2020, rating: medium) (sources/csp-jaccep-2023, rating: very high)
The tricuspid regurgitation rate differs threefold between studies, with no control arm. The long-term registry reported worsening in 11.1%; the dedicated echo study reported 32.6%. The authors attribute the difference to longer follow-up and more baseline normal valves, but TR grading, case mix and adjudication also differ, and neither study used a core laboratory or a non-LBBP control. The positional signal (lead closer to the annulus → more TR) is the more robust finding. (sources/longterm-lbbap-circep-2021, rating: high) (sources/tricuspid-regurgitation-lbbap-hrs-2023, rating: medium)
Which electrical metric defines a better configuration? The metrics disagree. Anodal capture shortens the surface QRS while lengthening LVAT and increasing LV dispersion, and earlier work found no haemodynamic benefit from anodal capture despite the narrower QRS. Conversely, selective LBBP is the most convincing capture state but produces the longest paced QRS. QRS duration, activation-time dispersion and proof of capture rank the same configurations differently, and no study has tested which ranking predicts outcome. (sources/crp-lbbap-jice-2023, rating: low) (sources/csp-jaccep-2023, rating: very high) (sources/lbbp-vs-lvsp-criteria-jaccep-2021, rating: medium)
LOT-CRT gives the best QRS but not always the best acute haemodynamics. In CSPOT, LOT-CRT shortened QRS more than BVP (29.5 vs 18.5 ms) yet produced nearly identical LV dP/dtmax (25.8% vs 26.4%); the incremental benefit over LBBAP appeared only in wide-QRS patients with deep septal pacing. The echocardiographic substudy, by contrast, showed a large LVEF and LVESV advantage. Acute and chronic endpoints therefore point in different directions. (sources/lot-crt-circep-2024, rating: high) (sources/lot-crt-hrs-2025, rating: medium)

Connections

Part of concepts/Left-Bundle-Branch-Area-Pacing — the parent overview
Related to concepts/LBBAP-Capture-Criteria — how capture was confirmed in these cohorts
Related to concepts/LBBAP-Implantation-Technique — procedural determinants of outcome
Related to concepts/Cardiac-Resynchronization-Therapy — the comparator strategy
Related to concepts/LOT-CRT — the hybrid alternative in wide QRS and IVCD
Related to concepts/Conduction-System-Pacing — the wider family
Related to entities/Heart-Failure — the clinical syndrome being treated
Related to concepts/Electrical-Resynchronization-Metrics — why QRS and activation metrics diverge
Related to concepts/Pacing-Induced-Cardiomyopathy — an important indication