Clinical Prediction Model for Unsuccessful Left Bundle Branch Area Pacing.

INTRODUCTION: Left bundle branch area pacing (LBBAP) has rapidly become a preferred form of cardiac physiologic pacing; however, predictors of unsuccessful implantation remain unclear. METHODS AND RESULTS: Between April 2020 and March 2024, LBBAP was attempted in 145 patients. Procedural success was defined by ≥ 1 of the following: output-dependent QRS morphology changes, V6 R-wave peak time (RWPT) < 75 ms, V6-V1 interpeak interval ≥ 44 ms, or concordance between intrinsic and paced V6 RWPT derived from left bundle potentials. Baseline clinical, electrocardiographic, and echocardiographic variables were compared between success (n = 126, 86.9%) and failure (n = 19, 13.1%) groups. Multivariable logistic regression using Firth's penalized likelihood identified nonspecific intraventricular conduction delay (NIVCD) (odds ratio [OR] = 4.34; p = 0.02), interventricular septal thickness (IVSd) (OR = 1.39 per mm; p = 0.03), and right atrial area (RAA) (OR = 1.12 per cm2; p = 0.01) as independent predictors of unsuccessful LBBAP. A model integrating these variables demonstrated strong discrimination (area under the curve [AUC] = 0.84), which was confirmed by bootstrap validation (AUC = 0.85; 95% confidence interval = 0.76-0.93). A nomogram was constructed to enable individualized risk estimation. CONCLUSION: NIVCD, septal hypertrophy, and right atrial enlargement are independent predictors of LBBAP failure. The resulting validated prediction model and nomogram provide individualized pre-procedural risk estimation and may assist in optimizing patient selection and procedural strategy.