Prognostic implications of LBBAP capture type

Background Left bundle branch area pacing (LBBAP) enables physiological ventricular activation, but whether different capture types translate into distinct clinical outcomes remains uncertain. Purpose To compare outcomes according to LBBAP capture type (confirmed, probable, left-septal, and deep-septal). Methods This single-center, retrospective study included consecutive patients who underwent LBBAP implantation between September 2022 and July 2024. Confirmed LBB capture was defined by QRS transition with output reduction (from non-selective to left ventricular septal pacing or to selective LBB capture), paced V6 RWPT<75 ms (<80 ms in LBBB), V6–V1 interpeak interval >44 ms, or selective capture on programmed stimulation. The primary outcome includes a composite of heart failure(HF)-related emergency department (ED) admissions or HF hospitalization. Results A total of 294 patients were included: 222 (75.6%) had confirmed, 36 (12.2%) probable, 20 (6.8%) left-septal, and 16 (5.4%) deep-septal capture. Mean age was 74.5±9.7 years, 68.4% were male, and median follow-up was 19.5 months. The most frequent indication was third-degree atrioventricular (AV) block (35.7%), followed by second-degree AV block (19.4%) and HF with reduced ejection fraction (14.6%). Paced QRS duration was significantly shorter with confirmed LBB capture (115.7±13.8ms) compared with left-septal (126.5±17.5ms; p=0.016) and deep-septal capture (139.6±15.7ms; p < 0.001), with no difference versus probable capture (121.1±14.3ms; p=0.224). Paced left ventricular activation time (LVAT) was also significantly shorter in the confirmed group (74ms, IQR 12) compared with probable (85ms, IQR 12), left-septal (96ms, IQR 8), and deep-septal capture (92ms, IQR 12; p<0.001). Baseline LVEF was 49.6±12.7%, slightly lower in the left-septal group (42.1±12.7%; p=0.035). Post-pacing LVEF (53.1±11.2%) and LVEF improvement (4.7±9.4%) were not significantly different across capture types (p=0.417 and p=0.802, respectively). In the multivariable Cox model adjusted for age, sex, and pacing percentage, and using confirmed LBB capture as the reference, no significant differences in clinical events were observed for probable (HR 1.326, 95% CI 0.155–11.353; p=0.797) or left-septal capture (HR 2.653, 95% CI 0.533–13.218; p=0.234). In contrast, deep-septal capture was associated with a significantly higher risk of the primary outcome (HR 5.204, 95% CI 1.213–22.321; p=0.026). Conclusion Confirmed LBB capture resulted in more physiological ventricular activation, with shorter paced QRS and LVAT. Deep-septal capture was associated with a higher risk of adverse events, suggesting that achieving true conduction-system capture may be essential for optimizing the clinical benefit of LBBAP.Primary outcome