Electrocardiographic imaging predicts response to left bundle branch area pacing in patients undergoing cardiac resynchronization therapy

Background Left bundle branch area pacing (LBBAP) has emerged as an alternative to conventional biventricular pacing for cardiac resynchronization therapy (CRT). However, response to LBBAP remains heterogeneous, and identifying reliable predictors of CRT response remains challenging. Electrocardiographic imaging (ECGI) provides a non-invasive approach to characterize ventricular activation and may help to identify patients most likely to benefit from LBBAP. Purpose To explore whether ECGI-derived metrics and clinical variables measured at baseline, after implantation, and their changes can predict response to LBBAP-based CRT. Methods Two-centre study including 32 patients with heart failure and guideline-based indication for CRT who underwent LBBAP. Baseline clinical, echocardiographic, electrocardiographic and ECGI data were collected. ECGI metrics assessed included Total Activation Time (TAT), 95th percentile of Left Ventricular Activation Time (LVAT95), Left Ventricular Dyssynchrony Index (LVDI), and Ventricular Electrical Uncoupling (VEU). CRT response was defined as an improvement in left ventricular ejection fraction (LVEF) >5 points without death or heart transplantation at 6 months. Multivariate logistic regression analyses were performed to identify independent predictors of CRT response. Results During follow-up, 64% of the patients were classified as responders. The percentage of ventricular pacing was similar between groups (96% ± 6 in responders vs. 99% ± 0.4 in non-responders). In univariate analysis, ischemic etiology, final TAT, final LVAT95 and final LVDI were associated with CRT response. Among these, only final LVAT95 remained independently associated with CRT response in multivariate analysis (OR 1.09, 95% CI 1.03–1.18; p = 0.007). Final LVAT95 also demonstrated the strongest univariate association and the best discrimination between responders and non-responders (AUC = 0.80). The optimal cutoff value of 58ms provided a sensitivity of 0.83, specificity of 0.85, PPV of 0.77, and NPV of 0.89. Conclusion ECGI enables detailed, non-invasive assessment of ventricular activation in patients with heart failure undergoing LBBAP for CRT. Among all predictors tested, final LVAT95 was the strongest predictor of CRT response, suggesting a role for ECGI in optimizing pacing configuration and improving CRT effectiveness.Final LVAT95Univariate and multivariate analysis