A structured protocol for the medium-term follow-up of left bundle branch area pacing—what to track, why it matters, and when surveillance can transition from intensive to annual cadence.
Left bundle branch area pacing (LBBAP) has rapidly become the preferred physiologic pacing strategy for patients with anticipated high ventricular pacing burden, evolving pacing-induced cardiomyopathy (PICM), or upgrade from chronic RV pacing. The mid-term outcome data are favorable, but the long tail of LBBAP follow-up is shorter than for conventional RV pacing, and meaningful drift can occur in either electrical performance or structural response.
A structured surveillance frame matters because the variables that actually predict long-term outcome are not the ones that get most attention at routine device checks. Ejection fraction, for example, is a relatively insensitive endpoint when baseline EF is preserved; the early signal of remodeling drift lives elsewhere. The six-variable framework below is built around that asymmetry.
The single most informative bedside variable for tracking the integrity of conduction system capture is the paced 12-lead ECG. A clean post-implant tracing in all 12 leads, recorded under stable pacing conditions, becomes the reference standard for every subsequent comparison.
Three features warrant attention on serial paced tracings: prolongation of V6 R-wave peak time, loss of the typical RBBB-like terminal pattern in V1, and unexplained QRS widening relative to the immediate post-implant baseline. Any of these may indicate a drift from selective LBB capture toward non-selective or LV septal-only capture.
Selective LBB capture preserves the physiologic transseptal and apex-to-base activation sequence. Non-selective capture introduces a parallel LV septal myocardial component that modestly alters the activation pattern; LV septal-only capture represents further departure from true conduction system pacing. The clinical implications of these transitions are not catastrophic in the short term, but they are the early electrical signature that other variables in this framework will eventually reflect.
Capture threshold on the LBBAP lead is routinely measured at every device check and is therefore an effortless data point to track longitudinally.
The expected pattern is stable thresholds with preserved paced QRS morphology. A rising threshold in combination with paced QRS morphology change is the finding that warrants attention, because it suggests progressive loss of true conduction system capture. Either variable in isolation is less informative.
Threshold drift alone may reflect benign lead-tissue maturation, local fibrosis at the septal entry site, or measurement variability. Morphology change alone may reflect rate-dependent fusion patterns or transient programming differences. The combination—rising threshold plus paced QRS change—is the signature of a clinically meaningful shift in the capture substrate and may justify reprogramming, lead repositioning consideration, or simply heightened surveillance of the other framework variables.
In a DDD device with intact native AV conduction, ventricular pacing percentage (Vp%) is one of the strongest modifiers of long-term remodeling exposure. The lower the Vp%, the smaller the cumulative dose of any non-physiologic activation, regardless of pacing site.
A gradual upward drift in Vp% across remote and in-person checks. This pattern often indicates that the native AV node is conducting more slowly with aging or progressive nodal disease, encroaching on the programmed AV window and triggering more ventricular pacing than necessary.
Revisiting AV delays and AV search hysteresis can restore intrinsic conduction and minimize unnecessary ventricular pacing. In LBBAP this is less catastrophic than in RV pacing—because the pacing is physiologic—but the principle of minimizing pacing burden when reasonable still applies. Vp% drift is also a clinical signal in its own right, sometimes preceding overt AV block or symptomatic bradycardia.
Echocardiography at 6 months post-upgrade, again at 12 months, then annually thereafter. The goal is to capture the expected trajectory of reverse remodeling in patients upgraded from chronic RV pacing or with prior PICM substrate.
Regression of eccentric LV remodeling, normalization of LA volume, and improvement in diastolic indices—E/e prime ratio, deceleration time, LA reservoir strain. These endpoints are far more sensitive than ejection fraction when baseline EF is preserved.
When EF sits near 55% at baseline, it has limited dynamic range as a remodeling readout—a meaningful structural change can occur with EF varying only a few percentage points within measurement noise. LV geometry and LA volume, by contrast, track continuously with chronic hemodynamic and electrical load. LA volume in particular integrates years of filling pressure exposure and is one of the earliest variables to respond to restored synchrony. If LA volume is regressing while EF is unchanged, the remodeling response is favorable.
Modern device remote monitoring captures a near-continuous record of ventricular ectopy, non-sustained ventricular tachycardia (NSVT), and atrial arrhythmia episodes. These data are usually summarized at each transmission and accumulate into a longitudinal trend that is easy to read once a baseline is established.
Three findings deserve attention: a rising PVC burden out of proportion to the individual's own baseline, new NSVT runs that were not present in the early post-implant period, and atrial tachyarrhythmia episodes—particularly atrial fibrillation in patients with antecedent LA dilation.
Absolute thresholds for ectopy burden are less useful than trend deviation from each individual's established post-implant baseline. A patient whose PVC burden has been stable at 2% and rises to 6% over six months is a different clinical situation from a patient whose baseline has always been 8%. The first warrants attention; the second may not. Rising PVC burden also frequently precedes other early signals of remodeling drift and is one of the few variables in this framework that updates daily.
High-sensitivity cardiac troponin T (hs-TnT) is increasingly used as a continuous variable in chronic cardiac disease, not just as a binary marker for acute injury. In the LBBAP follow-up context, it adds an inexpensive biochemical signal of ongoing myocardial stress that complements the structural and electrical variables.
The trajectory matters more than any individual value. A downward trend from the pre-upgrade level toward or below the population reference range is consistent with resolution of pacing-mediated myocardial stress. A flat or rising trajectory in the absence of an alternative explanation warrants closer evaluation.
Measurement at 3 to 6 months post-upgrade provides a meaningful comparison point against the pre-upgrade baseline. Chronic mildly elevated hs-TnT in the setting of high-burden non-physiologic pacing has been described as a biochemical correlate of subclinical remodeling. Restored synchrony with LBBAP typically reduces this signal, and the trajectory is a useful continuous readout that does not depend on imaging cadence or device interrogation timing.
| Variable | Cadence | Primary Signal |
|---|---|---|
| Paced 12-lead ECG | Implant baseline, 1 mo, 6 mo, 12 mo, then annually | V6 R-wave peak time, V1 morphology, QRS width |
| LBBAP capture threshold | Every device check (remote + in-person) | Threshold trend with paced QRS morphology |
| Ventricular pacing % | Every remote and in-person check | Upward drift in Vp% |
| Echocardiogram | 6 mo, 12 mo, then annually | LV geometry, LA volume, diastolic indices |
| Arrhythmia / ectopy burden | Continuous via remote monitoring | PVC trend, new NSVT, AT/AF episodes |
| hs-TnT | Pre-upgrade baseline and 3–6 mo post | Continuous trajectory vs. baseline |
The variables that actually predict long-term outcome after LBBAP are not the ones that dominate routine device checks. Paced QRS morphology, capture threshold trends interpreted together with morphology, ventricular pacing percentage drift, LV geometry and LA volume on serial echo, ectopy and arrhythmia trends from remote monitoring, and hs-TnT trajectory together provide a six-channel readout of the system's ongoing performance.
Stability or improvement across all six channels at 12 months allows confident transition to annual surveillance. Deterioration in any single channel is a signal—not necessarily an emergency, but an invitation to investigate the underlying mechanism before the other channels follow.