Conduction system pacing Where the lead goes, and how you know it’s working.

Part 1

What counts as conduction system pacing

His bundle pacing (HBP)

Capture of the His bundle itself, with the lead near the tricuspid valve summit on either the atrial or the ventricular side of the annulus. The His-to-QRS interval at the pacing site is at least 35 ms. CSP §Definitions

Left bundle branch area pacing (LBBAP)

Capture of the subendocardium on the left side of the septum, with or without conduction system capture. It is an umbrella term for left bundle branch pacing, left fascicular pacing and LV septal pacing, used when they can’t be told apart. A terminal R wave in V1 is usual but not universal. CSP §Definitions

LV septal pacing (LVSP)

The lead is deep in the basal to mid septum with a terminal R in V1, but there are no criteria for conduction system capture. It made up 21.5% of patients in the MELOS registry. CSP §Definitions

Deep septal pacing

The lead is deep in the septum but hasn’t reached the LV subendocardium: no terminal R in V1 and no features of left conduction system capture. It usually means the lead couldn’t be advanced further. CSP §Definitions

Selective vs non-selective

Whether the neighbouring myocardium is captured as well. Non-selective capture starts straight after the stimulus (a pseudo-delta wave); selective capture has an isoelectric gap first. Non-selective HBP is considered safer because there is back-up myocardial capture. At working output (2.5 V/0.4 ms), LBBAP is almost always non-selective. CSP §Definitions

Part 2

Where the evidence stands (October 2026)

What the 2026 ESC heart failure guideline recommends ESC HF 2026 §6.2.2

• Class I: CRT for symptomatic HFrEF with LVEF ≤35% despite optimal medical therapy, in sinus rhythm, with LBBB and QRS ≥150 ms.
• Class IIa: CRT rather than RV pacing for HFrEF with an indication for ventricular pacing for high-degree AV block, whatever the NYHA class or QRS width. An upgrade to CRT for LVEF ≤35% with worsening heart failure and a significant RV pacing burden. CRT for non-LBBB with QRS ≥150 ms, and for LBBB with QRS 130–149 ms.
• Class III: CRT for QRS <130 ms without a pacing indication for high-degree AV block.

Conduction system pacing in heart failure

• The guideline says LBBAP may prove a feasible alternative when CRT implantation is difficult. But it states that no randomised trial with patient-centred outcomes has tested CSP in HFrEF, so it makes no recommendation. ESC HF 2026 §6.2.3
• HeartSync-LBBP randomised 200 patients with LBBB and LVEF ≤35% to left bundle branch pacing or biventricular pacing at 6 centres in China. Over a median 36 months, death or heart failure hospitalisation was 8% vs 28% (HR 0.26), driven by fewer hospitalisations (7% vs 28%). All-cause death did not differ significantly (2% vs 5%). Implant success was 98% vs 94%. HeartSync
• The authors list its limits: all participants were Chinese, most had non-ischaemic cardiomyopathy (82.5%), and all centres were experienced, so the results may not apply to ischaemic cardiomyopathy or less experienced centres. HeartSync

HeartSync was published online in March 2026, but the guideline’s CSP section does not cite it. Read the two together: the guideline is the current ESC position, and HeartSync is randomised evidence that its CSP section does not discuss.

Part 3

His bundle pacing

Finding the His CSP §Locating the His

• The lead is usually delivered through a sheath with a posterior curve toward the septum, and the His is mapped with the lead itself in unipolar mode. A dedicated EP catheter is usually unnecessary.
• Map in RAO 20–30° for a sharp, near-field His signal with a small or absent atrial signal and a larger ventricular one.
• If no His signal can be found, pace-map at high output (5 V at 1 ms) looking for a narrow paced QRS. This helps in AV block without an escape rhythm.
• A proximal position on the atrial side of the annulus more often gives selective capture, sometimes with sensing problems. A distal position on the ventricular side more often gives non-selective capture, with better electrical parameters.
• In atrial fibrillation, fast atrial signals can mimic a His signal, and sites with large atrial signals should be avoided in case AV-junction ablation is planned later.

Confirming His capture CSP §His capture

• The key test is a sudden change in QRS morphology as the output is reduced.
• A difference of under 12 ms between the His-to-V6 R-wave peak time in intrinsic rhythm and the stimulus-to-V6 time during pacing indicates His capture.
• Don’t accept a paced QRS that still looks like LBBB (notches or slurs in I and V4–V6, stimulus to V6 R-wave peak over 110 ms).
• Thresholds: 1.5 V at 0.5 ms or less is optimal; up to 2.0–2.5 V may be acceptable in some situations. Bipolar sensing should ideally be over 2 mV. Check for atrial or His oversensing, which can be fatal in complete heart block.
• Citing the 2021 ESC pacing guideline (class IIa), the consensus advises considering a back-up ventricular lead in pacemaker dependency, high-grade or infra-nodal AV block, high thresholds, or planned AV-junction ablation.

Part 4

LBBAP: finding the site and crossing the septum

Choosing the entry point CSP §Insertion site

1. Mark a reference in RAO 20–30°: either the His recording or, more simply, the summit of the tricuspid annulus.
2. Advance the sheath about 15–20 mm toward the RV apex, then turn it counter-clockwise onto the basal to mid septum. Most leads have about 15 mm between the screw tip and the back of the ring, which helps judge distance.
3. Pace from the right septal surface. A good site often shows a “W” (a notch at the bottom of the QRS) in V1, and discordant II and III: II positive or equiphasic, III equiphasic or negative.

• Sector methods target an area 15–35 mm from the tricuspid summit in the RAO 30° image, at an angle of −10° to 30°. These sites are further from the His and may capture the fascicles rather than the left bundle itself.
• Lead positions more than 16–19 mm from the tricuspid annulus were associated with less tricuspid regurgitation.

Crossing the septum CSP §Septal penetration

• Point the sheath perpendicular to the septum. In LAO 30–40°, the lead should point 10–40° above horizontal, most often 20–30°. A higher entry site sits more horizontal; a lower one more vertical.
• Check in RAO if the lead won’t advance (it may be heading posteriorly) or if V1 never develops a terminal r or R (it may be heading anteriorly toward the apex).
• Watch how the lead behaves as it is turned. The screwdriver effect is smooth progress into the septum. The drill effect is turning without advancing, which bores a tunnel and risks dislodgement. Strong torque early on usually means the screw is caught in the endocardium.

In the explorer, the LBBAP lead enters the septum 18 mm from the His landmark. In LAO 30–40° its septal segment points about 15–16° above horizontal in this atlas. That is inside the 10–40° range but below the usual 20–30°. The septum is 9 mm thick at that point in this one atlas anatomy; real septa vary. Methods are on the References page.

Part 5

LBBAP: confirming conduction system capture

None of the cut-offs below is perfect, and values depend on the patient. A V6 R-wave peak time of 75 ms could be septal capture in a small heart, while 100 ms could be conduction system capture in a large dilated heart with bundle branch block. CSP §LBB capture

• Acceptable thresholds are under 1.5 V at 0.5 ms (ideally under 1 V), and bipolar sensing should ideally be over 4 mV. Thresholds can start high from the current of injury, so retest after a few minutes. CSP §LBB capture
• For comparison, HeartSync counted LBB pacing as successful when the unipolar paced QRS had an RBBB pattern and threshold testing showed a transition, either to selective LBB capture or to LV septal capture. HeartSync

Part 6

Complications and follow-up

• HBP: the most frequent problem is a rise in capture threshold with loss of His capture (up to 17%), or lead revision (up to 11%). Under- and oversensing also occur. CSP §Complications
• Septal perforation during LBBAP has been reported in 0–14.1% of patients. A low unipolar current of injury is a key warning sign; a sudden fall in pacing impedance is another. If perforation is found during the procedure, reposition the lead rather than just pulling it back. Late perforation is reported in 0.1–0.3%. CSP §Complications
• Other effects: worsening tricuspid regurgitation in up to a third of patients, more often with basal leads, and a troponin rise to over three times normal in about half (49.4%). CSP §Complications
• Follow-up: loss of the terminal R in V1 was seen in 4% in MELOS, so a 12-lead ECG at follow-up matters. CSP §Complications
• Discharge: the 2026 EHRA same-day discharge consensus notes that data on conduction system pacing are still limited. Observation times for other device procedures are on the workflow page, step 8. SDD Advice 1

Part 7

Standard leads: wrong positions and common mistakes

This part covers conventional right ventricular and right atrial leads, from the 2021 EHRA practical guide on implantation technique. Test yourself with Implanting leads (mistakes).

Which views to check

• Check the final lead position in RAO 30° and LAO 40–60°. EHRA 2021
• RAO 30° shows the long axis of the heart and outlines the apex better than PA. LAO 40–60° rules out a lead in a coronary sinus tributary or in the left ventricle (through a septal defect or an arterial puncture), which can be missed in PA. EHRA 2021

Right ventricular lead

• Coronary sinus. Inadvertent cannulation of the coronary sinus or its branches is suggested by the lead “tracking” and by the absence of ventricular premature beats; confirm it in LAO. EHRA 2021
• Apex. For active-fixation leads, perforation was independently associated with an apical position (odds ratio 3.37). The lead may sit slightly proximal to the apex, pointing slightly inferiorly. EHRA 2021
• Septum drifting anterior. Leads meant for the septum are often found on the anterior free wall or in the anteroseptal groove. That site may cause LV dysfunction and perforation, and rarely infarction when the helix is screwed into the left anterior descending artery. A true septal position is the target (septomarginal trabeculation or moderator band). EHRA 2021
• Why LAO alone misleads. Studies using CT or echocardiography found that LAO cannot confirm a septal position; an RAO view is also needed, and the angle of the lead in LAO does not indicate septal position because slack changes it. A landmark: in RAO 20–30° draw a 3 × 3 grid from the lateral border of the spine to the apex and aim for the middle square (the landmark can fail with a dilated or hypertrophied LV). ECG criteria for septal pacing were unreliable in every study that validated them against imaging. EHRA 2021
• Outflow tract. The RVOT septum is a small target and thin-walled (3–5 mm tapering to 1–2 mm below the valve), which raises concern for perforation. EHRA 2021
• Block with LBBB. Transient traumatic AV block can occur while positioning the RV lead in patients with underlying LBBB; have rescue pacing ready, even before fixation. EHRA 2021

Right atrial lead

• Lateral wall and tip. The appendage wall is 1–2 mm thick and thinnest laterally, so avoid the lateral wall. The tip of the appendage lies over the aortic root, which active-fixation leads can lacerate; the anterior appendage is probably the safer target. EHRA 2021
• Far-field R wave. The medial appendage lies near the RVOT, so a far-field R wave may be sensed and cause inappropriate mode switching. Keep its amplitude under 20% of the near-field atrial electrogram, check it before deploying the helix, and look at the waveform, not just the number. EHRA 2021
• Pain. Parasternal pain during implantation may mean perforation even when the electrical parameters are normal; it is confirmed if the pain stops after repositioning. EHRA 2021
• Lead type. Passive J-shaped leads carried a lower risk of pericardial complications than active ones but may dislodge slightly more often. EHRA 2021

Fixation, slack and testing

• Confirm helix extension with the fluoroscopic markers, not the number of turns. Over-rotation turns the whole lead body, which reduces stability and raises the risk of perforation. Do not let the pinch-on tool spin back after fixation, as that partly unscrews the helix. EHRA 2021
• Test stability by withdrawing the stylet to the inlet and buckling the lead by pushing, leaving enough slack for a “heel”; do not pull on the lead. At the apex, pace at maximum output while watching or palpating the left hemidiaphragm to exclude phrenic capture. EHRA 2021
• Acute targets: threshold ≤1.5 V at 0.5 ms, sensing ≥1.5 mV (atrium) and ≥4 mV (ventricle), impedance within the lead’s limits (usually 400–1200 Ω). For active leads confirm a current of injury that grows after the helix is out. A high threshold just after fixation often falls within minutes, so recheck before moving the lead. A negative electrogram can mean reversed clips or perforation. EHRA 2021

When it goes wrong

• Perforation during the procedure can cause vagal symptoms, chest pain, high thresholds, diaphragmatic capture and, rarely, an inverted current of injury. Tamponade is suggested by haemodynamic compromise and an immobile cardiac silhouette on fluoroscopy; confirm with echocardiography and treat with emergent pericardiocentesis. EHRA 2021
• Venous access can mislead too: a J-wire in the azygos vein runs above the right main bronchus in PA and plunges posteriorly in LAO. With a persistent left superior vena cava, leads run left of the spine and posteriorly within the coronary sinus. EHRA 2021