LBBAP Implantation Technique

Definition

LBBAP implantation is the transvenous, transseptal placement of a pacing lead through the right side of the interventricular septum to the left subendocardial surface, aiming to capture the left bundle branch or the adjacent septal myocardium. The procedure is defined less by a single target point than by a sequence of electrical checks performed while the lead is advanced. (sources/csp-ehra-2023, rating: high)

Key Concepts

Anatomy

The target is defined by the tricuspid valve and the His bundle, but the anatomy underneath varies enough to explain much of the procedural difficulty. (sources/lbb-anatomy-jacccase-2020, rating: medium)

• The His position itself moves: the AV node-to-His transition is superior in 58.5% and inferior within the triangle of Koch in 41.5%. The membranous septum is 4.6 ± 1.5 mm (1–9 mm), and its interventricular component is minimal or absent in ~60%.
• The LBB origin is variable: <1 to 14 mm in width; the non-branching component is only 1–3 mm in 75–85% before the fascicles arise; the main trunk extends 10–15 mm apically, so a transseptal lead usually lands in the midseptal area.
• Three fascicles, not two: anterior, septal/middle and posterior networks, consistent with Durrer's three synchronously activated LV endocardial islands within 5 ms.
• The tissue crossed is variable: RV endocardial thickness, fatty infiltration, connective tissue around the LBB and fibrous tissue from conduction disease. The LBB itself is 1–25 cell layers thick, and multiple deployments can injure the RBB, perforate the septum or damage a septal artery.
• ICE landmark: the proximal LBB lies 10–20 mm from the tricuspid annulus toward the apex, over an area of 4.5 ± 1.1 cm². (sources/ice-guided-plbb-circep-2023, rating: high)

Leads and delivery systems

Two lead families dominate. They demand different skills rather than differing in proven efficacy. (sources/csp-ehra-2023, rating: high)

(Table: sources/lumenless-vs-stylet-leads-jcm-2024, rating: medium)

Helix retraction is the SDL-specific skill. Three countermeasures exist: (1) pretension — extend the helix, rotate the stylet tool 10–15 turns clockwise, then push it over the silicone seal to store torque (Solia S); (2) integrated locking tools — Abbott's tool co-rotates inner and outer coils; (3) a home-made cap — a silicone-cut lead cap bridges stylet guide and outer body (Ingevity+). No manufacturer offers a universal locking system. (sources/lumenless-vs-stylet-leads-jcm-2024, rating: medium)

• The procedural tell: while rotating the body, no counter-torque should be felt. Counter-torque signals failed penetration, usually entanglement.
• Perpendicularity buys selectivity, not penetration: SDL can be screwed off-axis, but selective LBBAP is then rarely achieved; the lumenless lead's floppy tip demands perpendicularity to penetrate at all.
• The lumenless lead's cost is extraction: it accepts no locking stylet; high torquability and tensile strength complicate extraction, and the fixed isodiametric helix can avulse adhered myocardium.
• The multicentre SDL experience: 353 consecutive patients at 8 Belgian hospitals, Solia S60 via Selectra 3D, implant success 334/353 (94.6%), independent of centre volume (median 25 implants, range 5–162; success 97 ± 3%, P = .75). All operators were certified electrophysiologists with >5 years of device experience. (sources/stylet-lbbap-jce-2022, rating: medium)
• SDL electrical performance: unipolar threshold 0.6 ± 0.4 V at 0.4 ms, R-wave 10 ± 5 mV, paced QRS 126 ± 21 ms, stimulus-LVAT 74 ± 17 ms, depth 13 ± 2 mm, screw attempts 2.2 ± 1.9. Thresholds were stable to 12 months (0.7 ± 0.3 V, P = .291) but on only 67/353 patients (19%).
• SDL capture type: nonselective LBBP 57%, selective LBBP 16%, LVSP 27% — 73% achieved LBB capture; LBB potentials were recorded in just 14%.
• SDLs are stiffer and arguably more prone to perforation. The failure mode is entanglement — the helix caught in right-sided subendocardial tissue — which under excessive traction causes elongation or fracture, with one unretrievable helix fracture in a centre's early experience.

Choosing the entry site

• Use the His potential or the tricuspid valve summit as the reference in RAO 20–30°, advance the sheath ~15–20 mm towards the RV apex, then counter-clockwise torque to reach the basal-to-mid septum. (sources/csp-ehra-2023, rating: high)
• Good initial-site features: a "W" pattern (nadir notch in V1) and discordant QRS in II and III (II slightly positive, III negative or equiphasic).
• Liu et al.'s sector technique: target 15–35 mm from the tricuspid annulus at −10° to 30° in RAO 30°.
• Keep the entry in the upper half of the conventional zone when the tricuspid valve is normal: a lead entry ≤16.1 mm from the annulus independently predicted TR worsening. (sources/tricuspid-regurgitation-lbbap-hrs-2023, rating: medium)

The fluoroscopic 9-partition method supplies the only coordinate system derived specifically because LBBP had no standard fluoroscopic methodology. (sources/lbbap-9-partition-hrs-2020, rating: medium)

• The method: in RAO 30°, identify the ventricular contraction ring (CT confirms it is composed of cardiac valves); divide the IVS from ring to apex into high, medium and low, then cross with three further divisions → 9 partitions.
• Validation: all 11 His-bundle leads fell in the second partition, while 94.3% (33/35) of LBB leads fell at the junction of the second and fifth. Lead-tip-to-noncoronary/right-coronary-cusp distance was 3.8 ± 0.6 cm for LBB versus 1.9 ± 0.2 cm for His.
• Caveat: validated on the same 45-patient cohort used to derive it; only 11 His leads underpin the partition-2 claim; the grid is a 2-D RAO projection and cannot resolve depth within the septum — the dimension that separates LVSP from LBBP.

Contrast-guided entry replaces the search for a His potential or a "W" pattern with an opacified landmark. (sources/contrast-guided-lbbap-hrs-2021, rating: medium)

• Technique: opacify the tricuspid valve annulus with 10–15 mL of contrast via a second C315 sheath in RAO 30°, marking the annulus summit; place the lead in a fan-shaped target 15–35 mm from the summit spanning 110–230°.
• Randomised result (n=60): procedure 66.8 vs 85.5 min, fluoroscopy 7.8 vs 11.1 min, attempts 2.03 vs 2.96, LBB potential 79.3% vs 46.4% — all significant. Pacing parameters, LVAT and paced QRS were identical.
• The gain is procedural, not electrophysiological. It formalises two rules: avoid placing the lead above the TVA summit, and enter 10–20 mm distal to the annulus, where the endocardium is thinner.

Intracardiac echocardiography addresses the repeated attempts needed to reach the target. (sources/ice-guided-plbb-circep-2023, rating: high)

• Randomised, n=101: screw-in attempts 1.43 ± 0.62 vs 1.98 ± 0.75 (P = 0.0002), procedure 26 ± 8 vs 43 ± 9 min, fluoroscopy 7.4 ± 1.8 vs 10.7 ± 2.4 min. LBB trunk pacing was 46.8% vs 25% (P = 0.031) and PLBB pacing 91.5% vs 72.7% (P = 0.0267), with shorter QRS in the ICE group.
• Caveats: single-centre, n≈50 per arm, procedural surrogates with no clinical outcomes, and all operators were experienced — which matters because MELOS found the learning curve plateaus only after 250 cases. (sources/melos-lbbap-ehj-2022, rating: high)

Crossing the septum

• Lead orientation: 10–40° (usually 20–30°) superior to horizontal in LAO 30–40°, confirming perpendicular septal entry. (sources/csp-ehra-2023, rating: high)
• RAO checks: a posterior course shows no lead progression; an anterior course shows no terminal R in V1 despite apparent progression.
• Three behaviours during rotation:
  • Screwdriver effect (desired): smooth, steady progression into the septum.
  • Entanglement: torque build-up with no penetration → reposition, do not force.
  • Drill effect: multiple rotations without progression → a tunnel forms without secure fixation; change position or increase push/support.

The Fuwai procedural framework

The primary "how to implant" reference from the group that developed the transventricular-septal approach. (sources/lbbap-implant-jce-2019, rating: medium)

• Site identification has two strategies.
  • Single-lead method (SLM): place the sheath in the His bundle region near the tricuspid annulus under RAO 30°, record the His potential and save the fluoroscopic image as a marker, then advance inferior to the septal leaflet, targeting 10–20 mm from the His region. Alternatively the tricuspid annulus itself is the marker.
  • Dual-lead method (DLM): two 3830 leads and two C315 sheaths. The first lead marks the His region or assesses the block site in LBBB, or a second lead searches for a better site while the first acts as an anatomical reference. Reserved for enlarged right atrium and/or RV.
• The stopping rule: cease advancement as soon as right bundle branch delay appears in V1/V2; monitor the transition of paced QRS morphology and the notch shift in V1 continuously.
• The contact signal: a myocardial injury current, generally after four to five clockwise turns. An absent or low-amplitude injury current means the helix is not properly engaged → screw back and relocate.
• Depth is bounded by three numbers: normal septal thickness 6–11 mm, the 3830 helix 1.8 mm long, and the helix-to-ring space 9 mm — so ring electrode capture at 5 V/0.5 ms implies the helix is at or near the left septal surface.
• But ring capture is not necessary: septal thickness varies between patients and sites, and the lead is not completely perpendicular. Ring capture "is helpful to confirm the depth of the lead, but not necessary, especially for patients whose thickness of the ventricular septum is thin"; assess septal thickness before implantation instead.
• In LBBB, place a temporary backup pacing lead before LBBP — the manoeuvre can cause RBBB through temporary mechanical injury, risking inadvertent complete heart block.
• Slit the C315 sheath after implanting the other leads to reduce the risk of LBBP lead dislodgement.
• Perforation's electrical signature and direction: low threshold with high impedance suggests the lead is inside the septum; a sudden decrease in impedance, sensed R-wave amplitude and/or loss of capture indicates the helix has entered the LV cavity.
• An additional depth cue: the morphology of PVCs evoked during advancement may represent the depth of the pacing lead in the ventricular septum.
• Programming: because the lead sits in left-sided septal tissue, sensing is high and threshold low, so far-field atrial oversensing and ventricular undersensing are avoidable. Ventricular safety pacing and autocapture can be enabled; in LBBB the AV delay is titrated from 40 ms to the intrinsic PR interval, programming the value giving the narrowest QRS.

Real-time depth monitoring

Five intermittent methods require stopping rotation and reconnecting the pacemaker. (sources/csp-ehra-2023, rating: high)

Two signals can be read without interrupting the manoeuvre. (sources/fixation-beats-hrs-2020, rating: high)

• Fixation beats are the ectopic beats generated mechanically as the helix crosses the septum. LBB-area fixation beats — Qr/qR/rsR′ in V1 and/or QRS ≤130 ms — appeared at 327/339 final positions versus 9/939 intermediate positions (SN 96.4%, SP 97.3%, PPV 97.3%, NPV 96.5%). Continuing rotation until they appeared gave one rotation episode in 15/21 patients (71.4%) versus a median of 3, with unchanged pacing parameters.
• The stopping rule is the point: every observed perforation sequence involved continuing past the first fixation beat. Perforation beats differ — dominant R in V1 with a notch on the descending limb (Marriott's sign), QRS ~190 ms.
• The M-beat — rsR′ in V1, slurred S in V6 (>50% amplitude increase) and a rounded aVR terminal deflection — mimics the selective LBB-paced beat. It occurred in 69/210 (32.8%) and predicted selective capture with SP 96.8%, PPV 98.6%, SN only 58.6%: a rule-in marker, not a rule-out. (sources/m-beat-jce-2022, rating: medium)
• Current of injury (COI) on the unipolar ventricular EGM: distinct tip COI (ST elevation ≥25% of the intrinsic R wave) was present in 475/513 attempts (92.6%). Its presence predicted improvement of an initially unacceptable threshold within 10–20 minutes (33/47 with COI vs 0/8 without), and its absence identified all 11 perforations — SN 100%, SP 94.6%, NPV 100%, PPV only 29%. (sources/coi-lead-depth-hrs-2022, rating: medium)
• The tip/ring COI ratio is a depth index: it correlates negatively with angiographic depth (r = −0.624) and falls to 0.07 ± 0.04 at microperforation; a ratio <1 flagged microperforation with SN 100%, SP 98.2%, NPV 100%.
• The LBB COI — measured on the LBB potential–ventricle segment — is a different signal: present in 77/115 narrow-QRS patients (67.0%), in three morphologies, and associated with low-threshold capture in 100% vs 76.3% and selective capture in 54.5% vs 0%. Its disappearance, together with ventricular COI, marked perforation. (sources/lbb-current-of-injury-jce-2020, rating: medium)

Continuous pacing during rotation removes the blind interval. Two implementations exist: a rotatable connector, and pacing from the lead stylet. (sources/coi-lbbap-hrs-2024, rating: medium) (sources/impedance-lbbap-jce-2022, rating: medium)

• Rotatable connector, 105 patients: ventricular COI amplitude peaked within the interventricular septum at 26.3 ± 11.3 mV; lead depth 15.1 ± 2.6 mm; perforation 3/105 (3.4%); final threshold 0.61 ± 0.44 V from an initial 0.76 ± 0.56 V.
• COI amplitude does not separate RVSP from selective LBBP (10.4 ± 5.8 vs 11.7 ± 6.3 mV), so the proposed marker is qualitative morphology, not amplitude.
• Stylet pacing is a valid substitute for connector-pin pacing: impedance correlated R² = 0.9481, bias 12 ± 37 Ω, with comparable electrograms and paced morphologies.
• The impedance trajectory becomes a real-time landmark: a small initial increase, then a progressive decrease of 143 ± 87 Ω from RV septum to LV sub-endocardium, then a discrete increase immediately before the final position.
• Feasibility (n=20 uninterrupted vs n=10 interrupted): selective capture 70% in both arms (P = .584), no QRS difference, first-attempt success 18/20 (90%) vs 7/10 (70%) — descriptive only; the study was not randomised and was underpowered.
• A precise perforation-surveillance rule: "If the impedance dropped to 600 Ω, the current of injury was monitored... A sudden drop in the COI magnitude indicates a possible perforation, and further rotations need to be stopped." (sources/continuous-pacing-lvsp-cjc-2023, rating: medium)

Continuous pacing also reveals a six-station continuum during screwing, observed in 97 consecutive patients. (sources/continuous-pacing-lvsp-cjc-2023, rating: medium)

Yields were NSLBBP 87/97 (89.7%), SLBBP 82/97 (84.5%), LVS(e)P 9, RVSP 1. Stim-V6RWPT fell stepwise — IVSP1 91.0 ± 13.7, IVSP2 86.9 ± 12.3, LVS(e)P 82.1 ± 10.9, NSLBBP 67.1 ± 8.7, SLBBP 67.0 ± 9.3 ms (each step P < 0.05).

Watching the QRS during rotation gives the direct read-out when an EP recording system is available. (sources/csp-jaccep-2023, rating: very high)

• Signs of capture: sudden V6RWPT shortening ≥10 ms, R′ amplitude increase in V1, S waves appearing in I, V5, V6, and repolarisation normalising in V5–V6.
• The requirement is a rotational adapter; real-time recognition is difficult, so consecutive complexes are often reviewed retrospectively.

Perforation — recognition and management

Perforation is the defining risk of the transseptal route, and its reported incidence depends entirely on how it is sought. (sources/lumenless-vs-stylet-leads-jcm-2024, rating: medium)

• The electrical signature is a triad: a marked COI fall with QS morphology and impedance <450 Ω.
• A dedicated 212-patient series using electrical criteria reported 14.1%, with a unipolar impedance <450 Ω rule that was 100% sensitive but only 83% predictive, and no imaging confirmation — the diagnosis was defined by the same parameters being evaluated. (sources/septal-perforation-hrs-2022, rating: medium)
• The multicentre SDL registry reported 2% (8/353). All were asymptomatic; each lead was retracted and reimplanted in a different deep septal position; no VSD, septal haematoma or LBBAP-lead tamponade occurred. The authors attribute the higher rate than single-centre lumenless reports to including the learning curve. (sources/stylet-lbbap-jce-2022, rating: medium)
• Three intraprocedural signs used to suspect it: impedance fall >200 Ω, unipolar threshold >3 V, or contrast leak into the LV.
• Management: overt perforation → reposition (do not simply withdraw); late perforation requires anticoagulation. Micro-perforation → observe if electrical parameters are stable; no anticoagulation needed, endothelialisation is protective. (sources/csp-ehra-2023, rating: high)

Electrical targets and programming

• Acceptable capture threshold: <1.5 V at 0.5 ms, ideally <1 V; bipolar sensing ideally >4 mV. (sources/csp-ehra-2023, rating: high)
• Check bipolar versus unipolar QRS width — anodal capture from the ring electrode may further narrow QRS.
• Program ventricular safety pacing and autocapture where available, and titrate the AV delay to the narrowest paced QRS in LBBB.
• CRT-port wiring depends on the rhythm: in chronic AF, the LBBP lead can be connected to the atrial port; in sinus rhythm it goes to the LV port with the AV delay adjusted to allow native right bundle conduction; in IVCD the LBBP lead can be used together with the LV lead. (sources/lbbap-beginners-guide-hrs-2019, rating: high)
• Anodal capture should be measured, not assumed: document the anodal capture threshold and the nonselective-to-selective transition threshold. Bipolar pacing at ~3.5 V/0.4 ms is suggested to achieve anodal capture for RV-delay compensation, but programming above anodal capture is not mandatory and trades battery drain for a marginal electrical gain.

Contradictions / Open Questions

• Reposition after a failed attempt, or stop? The SDL registry treats repositioning as routine and low-risk (mean 2.2 ± 1.9 attempts), and the ICE trial averaged 1.43–1.98 attempts. The Fuwai guide takes the opposite position: "We do not recommend multiple attempts for LBBP lead placement because repeated attempts... will increase potential risks of local myocardial injury and left ventricular perforation." Two other studies hide the question inside an operator-chosen stopping rule — failure after 5 attempts or 20 minutes, or after 3–4 sites — so every published success rate is contingent on a threshold the operator selected. (sources/stylet-lbbap-jce-2022, rating: medium) (sources/lbbap-implant-jce-2019, rating: medium) (sources/lbbap-avb-hrs-2019, rating: medium)
• How common is septal perforation, really? The published range spans an order of magnitude: 0–14.1% overall, 2% (8/353) in the SDL registry, 14.1% in a series using electrical criteria, and up to 10.2% (lumenless) / 11.6% (SDL) in a technical review. These are not the same event: the registry counts perforations that changed management, while the higher figures include echocardiographic micro-perforation and asymptomatic contrast leak. A single percentage carries no comparable meaning without its detection method. (sources/lumenless-vs-stylet-leads-jcm-2024, rating: medium) (sources/septal-perforation-hrs-2022, rating: medium)
• The lead-type choice has no outcome evidence, only a risk trade-off. Lumenless leads are stable but hard to extract; stylet-driven leads allow pacing during rotation but carry helix retraction and early fracture. No study randomises lead type, and no study reports extraction success or complication rates by lead type in a comparable cohort. The choice is made on procedural feel, not evidence. (sources/lumenless-vs-stylet-leads-jcm-2024, rating: medium)
• Does the stylet-driven lead widen who can perform LBBAP? The SDL registry reports success independent of centre volume, which reads as an argument for broader uptake. But perforation (2%) and coronary fistula (1.4%) were higher than single-centre lumenless reports, the comparison is cross-study, and all operators were experienced electrophysiologists. Volume-independence describes experts, not novices. (sources/stylet-lbbap-jce-2022, rating: medium)
• The "success" definition is not standardised. This registry accepts LBBP or LVSP as success, while other reports accept only proven conduction-system capture. MELOS reports 92.4% success while finding 21.5% septal-only; the SDL registry reports 94.6% while finding 27% septal-only. Neither publishes a success rate restricted to confirmed LBB capture, so the headline numbers are not measuring the same thing. (sources/stylet-lbbap-jce-2022, rating: medium) (sources/melos-lbbap-ehj-2022, rating: high)

Connections

• Part of concepts/Left-Bundle-Branch-Area-Pacing — the parent overview
• Related to concepts/LBBAP-Capture-Criteria — how capture is confirmed during and after implantation
• Related to concepts/Conduction-System-Pacing — the wider family of techniques
• Related to concepts/His-Bundle-Pacing — the proximal alternative and its anatomical reference points
• Related to concepts/LBBAP-Outcomes — what the procedure achieves
• Relevant to concepts/Pacing-Induced-Cardiomyopathy — a common indication for upgrade

Sources

• sources/csp-ehra-2023
• sources/csp-jaccep-2023
• sources/lbbap-beginners-guide-hrs-2019
• sources/lbbap-implant-jce-2019
• sources/lbbap-avb-hrs-2019
• sources/lbb-anatomy-jacccase-2020
• sources/lbbap-9-partition-hrs-2020
• sources/contrast-guided-lbbap-hrs-2021
• sources/ice-guided-plbb-circep-2023
• sources/lumenless-vs-stylet-leads-jcm-2024
• sources/stylet-lbbap-jce-2022
• sources/melos-lbbap-ehj-2022
• sources/fixation-beats-hrs-2020
• sources/m-beat-jce-2022
• sources/coi-lead-depth-hrs-2022
• sources/lbb-current-of-injury-jce-2020
• sources/coi-lbbap-hrs-2024
• sources/impedance-lbbap-jce-2022
• sources/continuous-pacing-lvsp-cjc-2023
• sources/septal-perforation-hrs-2022
• sources/tricuspid-regurgitation-lbbap-hrs-2023