CMR characterisation of patients with heart failure and left bundle branch block

Abstract Introduction Cardiac magnetic resonance (CMR) is recommended internationally for tissue characterisation, volumetric assessment and assessment of prior myocardial infarction (1). Patients with left bundle branch block (LBBB) form a significant sub-group of individuals, however, there is limited data available to accurate determine whether LBBB is associated with any CMR-definable phenotypes. Aim We aimed to identify the distinctive cardiac magnetic resonance (CMR) features of patients with left bundle branch block (LBBB) and heart failure with reduced ejection fraction (HFrEF) of presumed non-ischaemic aetiology. Methods Participants from the MATCH (MyocArdial Tissue CHaracteristics in patients with heart failure according to glycaemic status) registry were prospectively enrolled. Exclusion criteria included: LVEF >40% at initial evaluation, established diagnosis of coronary artery disease, known presence of structural heart disease, suspected myocarditis, congenital heart disease and significantly impaired renal function. Following identification of HF, guideline-directed medical therapy was initiated based on the decision of the treating physician. LV recovery was defined as achieving ≥10% absolute improvement to a level of 40% or greater of LVEF from baseline evaluation upon the date of CMR quantification. Electrocardiograms were analysed to establish the presence or absence of LBBB. Results A total of 391 patients were recruited including 115 (29.4%) presenting with LBBB. Compared to controls, LBBB patients exhibited larger left ventricles and smaller right ventricles, but no differences were observed with respect to LVEF (35.8±12 vs. 38± 12 %, p=0.105) or right ventricular ejection fraction (RVEF). The prevalence of ischaemic scar (18.3% vs 21%, p=0.54), non-ischaemic fibrosis (25.2% vs 32.6%, p=0.15) and inducible ischaemia (6.1% vs. 7.6%, p=0.6) was comparable in patients with LBBB and those without. The overall rate of LV recovery was not significantly different between LBBB and non-LBBB patients (27.8% vs. 33%, p=0.32) from their baseline evaluation to the time of CMR assessment (70 [42 - 128] days). In the multivariate logistic regression, reduced LVEF remained an independent predictor of LV non-recovery in patients with LBBB, even after adjusting for gender and QRS duration (OR 0.93, 95%CI 0.87 – 0.99, p=0.03). In contrast, this association was not observed among patients without LBBB, whereas ischaemic scar was the primary predictor of LV non-recovery (OR 4.31, 95%CI 1.85 – 10.04, p=0.0007). Conclusion In this study patients presenting with a combination of HFrEF and LBBB had a larger LV cavity and smaller RV cavity than those without LBBB. However, there were no differences in the prevalence of ischaemic scar, non-ischaemic scar of inducible ischaemia according to the presence of absence of LBBB. Rates of LV recovery were similar between LBBB and non-LBBB patients.Clinical CMR ParametersSegmental Distributions