Earliest ECG Indicators of Impending Loss of Left Bundle Branch Capture in LBBAP

The short answer: the bundle-capture threshold and the myocardial (LV septal) capture threshold are dissociable, and the bundle threshold almost always climbs first. So the earliest harbinger is not a single morphology change at a fixed output — it is the divergence of those two thresholds on serial decremental testing, accompanied by a handful of beat-to-beat morphology cues that appear at the programmed output once the safety margin narrows. The markers below are ranked by how early they declare themselves.

Why two thresholds matter

On decremental output testing there are two transitions: an upper transition where nonselective or selective LBB capture (NS-LBBP / S-LBBP) collapses to LV septal capture only (LVSP), marked by an abrupt V6RWPT jump; and a lower transition where myocardial capture is lost entirely. The upper transition defines the LBB-capture threshold. Because the LBB threshold can climb independently of — and usually ahead of — the myocardial threshold, comfortable myocardial capture can persist at the programmed output while conduction-system capture is silently being lost.

1. Rising LBB-capture threshold (divergence) — the earliest quantitative marker

Track, at each interrogation, the output at which the V6RWPT step occurs. A creeping rise in that value — even while the myocardial threshold remains flat at under 1 V — is the earliest objective evidence of interface deterioration (maturation reversal, microdislodgement, peri-lead fibrosis, or developing exit block at the Purkinje–myocardial junction). Because symptoms appear only once the LBB threshold crosses the programmed output, serial measurement typically affords weeks of warning.

2. V6 R-wave peak time (LVAT): abrupt jump, and pre-loss instability

Frank loss of LBB capture produces a discrete, abrupt V6RWPT prolongation — commonly 10–15 ms or more across a single output step — distinct from the smooth, output-independent V6RWPT of true bundle capture. The impending sign, before that step is reached, is loss of the stable, output-independent character of V6RWPT: beat-to-beat alternans, intermittent single-beat prolongation, or fusion morphologies appearing at a previously clean output. That instability signals operation at the LBB strength–duration margin.

3. Stimulus-to-V6 R-wave peak prolongation (exit delay)

Distinct from the QRS-onset-to-peak measurement, a creeping increase in the total stimulus-to-V6-peak interval — particularly an emerging or lengthening isoelectric latency between the stimulus and QRS onset — indicates developing exit delay at the electrode–bundle interface. This can precede an outright threshold rise and is relatively specific for interface degradation rather than generic output drift.

4. Narrowing V6–V1 interpeak interval

Genuine LBB capture preserves a physiologic interventricular delay — rapid LV activation via the conduction system, relatively delayed RV — producing a wide V6–V1 interpeak interval. Loss of bundle capture collapses that delay as LV activation becomes myocardial and slower. A serial decline in the interpeak interval is a sensitive marker of waning conduction-system contribution, even when individual beats still appear captured.

5. Nonselective→selective transition at unchanged output

A beat pattern that was reliably nonselective becoming selective at the same programmed output is a specific red flag for electrode retraction out of the septum: the local myocardial component is lost while the tip still abuts conducting tissue. It is easily misread as benign because the bundle is still captured — but it is a mechanical warning, not a reassurance.

6. Attenuation of the V1 terminal R wave

Lower in sensitivity and more subjective, progressive loss of the RBBB-like terminal R in V1 tracks with diminishing relative RV-activation delay and is worth noting on serial tracings.

The device blind spot

Device-based capture-management algorithms confirm a ventricular evoked response — that is, myocardial capture — and cannot distinguish LBB capture from pure LV septal capture. Such algorithms can therefore report stable capture and a stable threshold while conduction-system capture is being lost. This makes serial 12-lead ECG surveillance — fixed-output tracings plus periodic decremental morphology testing — the operative monitoring instrument for this particular transition.

A note on limb-lead-only recorders

The validated criteria above (V6RWPT, V6–V1 interpeak, V1 morphology) require precordial leads. Single- or six-lead recorders that capture only the frontal-plane limb leads cannot assess these directly; frontal-plane QRS axis and limb-lead QRS duration serve only as crude surrogates. Reliable surveillance of impending bundle loss requires a true precordial recording with reproducible lead placement.