Can we rely on remote electrogram analysis to monitor long-term conduction system capture in left bundle branch area pacing?

Left bundle branch area pacing (LBBAP) is increasingly used to preserve physiological ventricular activation and prevent pacing-induced dyssynchrony.1 However, long-term maintenance of conduction system capture (CSC) is not always guaranteed. Lead micro- or macro-dislodgement or local tissue remodelling may result in a transition from conduction system to myocardial-only capture, with consequent widening of the paced QRS (pQRS).2 At present, CSC verification relies mainly on in-hospital 12-lead ECG assessment, which provides only intermittent evaluation and may fail to detect changes occurring between scheduled visits.3,4 Remote monitoring (RM) systems routinely transmit intracardiac electrograms (EGMs), allowing continuous observation of pacing-related electrical activity under real-world conditions,5 thereby providing the opportunity to detect subtle changes in pacing morphology. The purpose of this preliminary data analysis was to evaluate whether changes in NF-EGM morphology and duration, transmitted through RM, can closely mirror pQRS modifications, potentially reflecting CSC status in LBBAP patients. We retrospectively analysed 153 patients who underwent LBBAP with confirmed CSC at implantation and immediate RM activation. Patients were selected from a larger LBBAP population implanted at Santa Maria della Misericordia Hospital, Rovigo, Italy, based on the availability of matched RM transmission with 12-lead ECGs performed at implantation, 1-year follow-up (1y-FU) visits, or urgent visits. All implantations were performed according to the European Heart Rhythm Association Consensus criteria for conduction system pacing.6,7 Both lumenless leads (LLL) and stylet-driven leads (SDL) were utilized based on operator preference. Twelve-lead ECGs were analysed at baseline, immediately post-implant, and at 1y-FU. Paced QRS duration (pQRSd) and stimulus-to–R-wave peak time in lead V6 (V6RWPT) were measured. Based on pQRSd variations at 1y-FU, patients were classified into two groups: PQRS-STABLE (ΔpQRSd