Optimal Pace Timing for LBBAP With or Without an Additional LV Lead: Results From the CSPOT Study

Authors, Journal, Affiliations, Type, DOI

• Authors: Robert D. Schaller, Paul Foley, Badrinathan Chandrasekaran, Jonathan Lyne, Marek Jastrzębski, Pugazhendhi Vijayaraman, Gaurav A. Upadhyay, Rafał Gardas, Travis Richardson, D'Anne Kudlik, Robert W. Stadler, Patrick Zimmerman, James Burrell, Robert Waxman, Richard N. Cornelussen, Bengt Herweg, Zachary Whinnett
• Journal: Europace 2026;28:euag103
• Affiliations: University of Pennsylvania; Great Western Hospital (Swindon); Beacon Hospital (Dublin); Jagiellonian University (Kraków); Geisinger Heart Institute; University of Chicago; Medical University of Silesia (Katowice); Vanderbilt University Medical Center; Medtronic (Mounds View and Bakken Research Center); University of South Florida and Tampa General Hospital; Imperial College London
• Type: Prospective, multicentre, acute haemodynamic and electrical study — secondary analysis of CSPOT (NCT04905290)
• DOI: 10.1093/europace/euag103

Overview

This CSPOT analysis examined how sensitive LBBAP, LOT-CRT and BVP are to atrioventricular (AV) and interventricular (VV) timing in 48 CRT-indicated patients with preserved AV conduction. All three modalities were similarly sensitive to AV delay: a 60 ms deviation from the optimum reduced the haemodynamic benefit by about 30%. The optimal paced AV delay differed by modality when judged by LV dP/dtmax — LBBAP 152 ms, LOT-CRT 171 ms, BVP 167 ms — but was similar when judged by QRS duration (164, 160 and 165 ms). Optimal AV delay was linearly related to the patient's intrinsic AV conduction: PAV ≈ 0.54 × Ap–LBBs + 3.8 ms for LBBAP, 0.56 × Ap–LBBs + 7.7 ms for LOT-CRT, and 0.48 × Ap–RVs + 28.0 ms for BVP. Patient-specific AV delays derived from these equations outperformed both nominal and best-fixed delays. Simultaneous biventricular activation (VV = 0) was optimal for LOT-CRT and BVP.

Keywords

Cardiac resynchronization therapy; Left bundle branch area pacing; Left bundle branch area pacing-optimized cardiac resynchronization therapy; AV delay; VV delay

Key Takeaways

Design and population

• CSPOT: prospective, multicentre, global study; 48 patients completed the acute protocol (previously described in the primary CSPOT report).
• Inclusion: CRT indication with preserved AV conduction; complete heart block and AF excluded.
• Mean age 65.3 ± 10.3 years; 67% male; ischaemic cardiomyopathy 29%; LVEF 30.6% ± 11%; QRS 171 ± 20.8 ms; PR 200 ± 53 ms.
• LBBB 40%, IVCD 60%; successful LBBAP 56%, DSP 44%.
• Protocol: atrial overdrive pacing (~10 bpm above intrinsic); five AV delays tested (AV70%, ±30 ms, ±60 ms); for LOT-CRT four VV delays (0, LBBAP first 30 ms, LV first 30 and 60 ms); for BVP three VV delays (0, LV first 30 and 60 ms).
• Optimal AV delay for each patient and modality was the vertex of a parabolic fit to LV dP/dtmax and, separately, to QRS duration.

Sensitivity to AV delay

• A 60 ms deviation from optimal AV delay reduced LV dP/dtmax benefit by approximately 30%.
• The range of LV dP/dtmax and QRS duration across a 120 ms AV sweep did not differ significantly among BVP, LBBAP and LOT-CRT (P = .22 and P = .29), indicating similar sensitivity to AV timing.
• An example patient: LBBAP AV sweep spanned 16.5% LV dP/dtmax and 35.5 ms QRS; optimal AV was 178 ms by dP/dtmax and 187 ms by QRS duration.

Optimal AV delay

• By LV dP/dtmax: LBBAP 152 ms vs BVP 167 ms (P = .01) and LOT-CRT 171 ms (P < .01); BVP and LOT-CRT similar.
• By QRS duration: LBBAP 164 ms, LOT-CRT 160 ms, BVP 165 ms; no significant differences.
• For LBBAP, the dP/dtmax-optimal AV delay was shorter than the QRS-optimal delay (152 vs 164 ms; P = .03); no such difference for LOT-CRT (P = .09) or BVP (P = .79).
• Conduction disease type (LBBB vs IVCD), capture type (LBBAP vs DSP) and cardiomyopathy aetiology did not significantly affect optimal AV delay.

Relationship to intrinsic AV conduction

• Mean Ap–RVs (279.0 ± 49.2 ms) was nearly identical to Ap–LBBs (278.4 ± 48.1 ms; P = .85).
• Linear regression for optimal paced AV delay:
  • BVP: PAV = 0.48 × (Ap–RVs) + 28.0 ms (P < .0001; R² = 0.66).
  • LBBAP: PAV = 0.54 × (Ap–LBBs) + 3.8 ms (P < .0001; R² = 0.57).
  • LOT-CRT: PAV = 0.56 × (Ap–LBBs) + 7.7 ms (P < .0001; R² = 0.69).

Fixed versus patient-specific AV delay

• LBBAP: best fixed PAV 155 ms (mean error 34.6 ms); nominal 130 ms (error 38.7 ms); regression-based patient-specific PAV (error 26.6 ms).
• LOT-CRT: best fixed 163 ms (error 29.2 ms); nominal 130 ms (error 37.8 ms); regression-based (error 19.3 ms).
• BVP: best fixed 163 ms (error 33.0 ms); nominal 130 ms (error 40.7 ms); regression-based (error 22.2 ms).
• The nominal 130 ms PAV was >30 ms from optimal in 55.3% of LBBAP patients by dP/dtmax and 46.9% by QRS; in 55.9% and 40.0% of LOT-CRT patients, respectively.

Optimal VV delay

• LOT-CRT: simultaneous pacing (VV = 0) gave the best QRS reduction; pacing LBBAP first by 30 ms significantly reduced dP/dtmax versus simultaneous (P = .01) and versus LV-first 30 ms (P < .01).
• BVP: no significant dP/dtmax difference among VV delays, but simultaneous pacing gave the greatest QRS reduction.
• The authors suggest the optimal VV delay depends on the mechanism: LBBAP-first for conduction-system delay, LV-first for intramyocardial delay (citing simulation work), and their mixed LBBB/IVCD cohort favoured VV = 0.

Interpretation

• AV optimisation is as important for LBBAP and LOT-CRT as for BVP; empirical nominal settings are frequently suboptimal.
• A simple patient-specific rule based on the intrinsic AV interval can be implemented at implant or follow-up.
• Simultaneous LV and conduction-system activation is a reasonable default for LOT-CRT when the mechanism is mixed.

Limitations of the document

• Acute study only — LV dP/dtmax and QRS duration are surrogates, not clinical outcomes.
• Small sample (n = 48) and subgroup analyses underpowered.
• Exclusion of complete heart block and AF — the regression equations apply only to patients with intact AV conduction in sinus rhythm.
• Atrial overdrive pacing prolongs the AV interval; the relationship to sensed AV delay was not formally validated.
• Parabolic fitting excluded patients with flat or poorly fitted AV responses, which may limit generalisability; patients insensitive to AV timing will not benefit from any optimisation method.
• Pacing configurations tested in a fixed order, not randomised; measurement was by a single unblinded reviewer.
• Cohort biased toward IVCD over LBBB, and the study was not powered to compare conduction subtypes.
• Industry sponsorship: Medtronic funded the study and employs several authors; multiple investigators report device-industry honoraria.

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the pacing modality
• concepts/LOT-CRT — LBBAP plus coronary venous pacing
• concepts/Electrical-Resynchronization-Metrics — the main subject
• concepts/Electrical-Resynchronization-Metrics — VV timing
• concepts/Cardiac-Resynchronization-Therapy — the clinical context
• concepts/Electrical-Resynchronization-Metrics — QRS duration and LV dP/dtmax

Key Entities Mentioned

• CSPOT (Conduction System Pacing Optimized Therapy) study — the trial
• Medtronic — sponsor and device supplier
• University of Pennsylvania — lead site

Wiki Pages Updated

• Created: wiki/sources/lot-europace-2026.md
• Updated: wiki/concepts/LOT-CRT.md
• Updated: wiki/concepts/Electrical-Resynchronization-Metrics.md
• Updated: wiki/sourceindex.md
• Updated: log.md