Noninvasive stratification of ventricular substrate by electrocardiographic imaging

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Abnormalities of cardiac activation and conduction velocity (CV) are known pro-arrhythmogenic factors. Electrocardiographic imaging (ECGI) is presented as a non-invasive tool for arrhythmogenic substrate analysis. However, electrocardiographic features such propagation activation patterns, and CV are unknown in healthy subjects. In this study, ECGI technology has been evaluated for ventricular tissue stratification in patients with and without structural heart disease. Methods Nineteen patients were included in the study. The healthy ventricles group (Panel A) included 10 patients (8 males, 57 ± 6 years, NYHA = I and LVEF 61 ± 5 %) with normal baseline QRS and normal values of ventricular morphology, function and perfusion assessed by late gadolinium enhancement cardiac magnetic resonance (LGE-CMR). The structural heart disease group (Panel B) included 9 patients (9 males, 66±6 years, NYHA 2±1 and LVEF 26±5%) with ischemic cardiomyopathy (ICM) confirmed by LGE-CMR and indication for ablation due to ventricular tachycardia. For each patient, fibrotic areas were identified by ventricular segmentation of the LGE-CMR and ventricles were regionalized in 16 regions. Epicardial electrograms were noninvasively reconstructed by ECGI, late activation time (LAT) and CV maps were calculated to study electrocardiographic features. Results – Ventricles with fibrosis secondary to ICM revealed heterogenous activation patters and higher activation time distribution compared to healthy ventricles (120 ms vs 80 ms; p<0.01). Density distribution of CV was evaluated in the 0-200 cm/s interval, patients with healthy ventricular tissue showed homogenous CV distribution whereas ICM patients presented higher concentration in the slow CV region with a maximum peak around 75 cm/s. Moreover, CV values were significantly lower in fibrotic myocardium present in patients with structural heart disease compared to healthy myocardium (77 ± 40 cm/s vs 114 ± 46 cm/s; p<0.01), Panel C anterior view. Non-fibrotic regions in the ischemic ventricle group, Panel C inferior view, presented similar CV distributions to healthy ventricular tissue (114 ± 46 cm/s vs 120 ± 48 cm/s; p = 0.14). Conclusions ECGI can noninvasively identify fibrotic tissue by analysing ventricular activation and CV patterns. ECGI could be a potential early tool to identify patients at risk of sudden cardiac death.