Cardiac magnetic resonance-derived myocardial scar is associated with echocardiographic response and clinical prognosis of left bundle branch area pacing for cardiac resynchronization therapy.

Abstract

Left bundle branch area pacing (LBBAP) is a novel approach for cardiac resynchronization therapy (CRT), but the impact of myocardial substrate on its effect is poorly understood. This study aims to assess the association of cardiac magnetic resonance (CMR)-derived scar burden and the response of CRT via LBBAP. Consecutive patients with CRT indications who underwent CMR examination and successful LBBAP-CRT were retrospectively analysed. Cardiac magnetic resonance late gadolinium enhancement was used for scar assessment. Echocardiographic reverse remodelling and composite outcomes (defined as all-cause death or heart failure hospitalization) were evaluated. The echocardiographic response was defined as a ≥15% reduction of left ventricular end-systolic volume. Among the 54 patients included, LBBAP-CRT resulted in a 74.1% response rate. The non-responders had higher global, septal, and lateral scar burden (all P < 0.001). Global, septal, and lateral scar percentage all predicted echocardiographic response [area under the curve (AUC): 0.857, 0.864, and 0.822; positive likelihood ratio (+LR): 9.859, 5.594, and 3.059; and negative likelihood ratio (-LR): 0.323, 0.233, and 0.175 respectively], which was superior to QRS morphology criteria (Strauss left bundle branch abnormality: AUC: 0.696, +LR 2.101, and -LR 0.389). After a median follow-up time of 20.3 (11.5-38.7) months, higher global, lateral and septal scar burdens were all predictive of the composite outcome (hazard ratios: 4.996, 7.019, and 4.741, respectively; P's < 0.05). Lower scar burden was associated with higher response rate of LBBAP-CRT. The pre-procedure CMR scar evaluation provides further useful information to identify potential responders and clinical outcomes.