Time-efficient three-dimensional transmural scar assessment provides relevant substrate characterization for ventricular tachycardia features and long-term recurrences in ischemic cardiomyopathy

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): - Ministerio de Ciencia e Innovación and the ProCNIC Foundation - Grant TEC2017-82408-R Background We aimed to validate a 3D methodology for transmural scar assessment using time-efficient upsampled models from 2D delayed gadolinium-enhanced cardiac magnetic resonance (LGE-CMR) sequences and determine the clinical implications in ischemic cardiomyopathy-related ventricular tachycardia (VT) episodes. Methods Translational study including 10 pigs with myocardial infarction and 15 patients with spontaneous VT episodes and underlying infarct-related scar. Pigs underwent 3D LGE-CMR sequences to test 3D transmural-based scar assessment using several downsampled resolutions from the original images. Validation in patients included 3D and 2D LGE-CMR studies performed within the same protocol. Transmural-based scar areas were used for correlation analyses with the cycle length (CL) of VT episodes. Scar volumes were used as a benchmark comparison. Results In pigs, 3D transmural-based scar areas showed interclass correlation coefficients >0.94 between original 3D high-resolution- and downsampled-derived models. In patients, scar area quantification in myocardial regions with 3D transmurality <0.2 (for upsampled 2D-derived models) and <0.1 (for 3D-derived models) showed direct correlation with the CL of spontaneous VT episodes (r = 0.69; p < 0.01 and r = 0.79; p < 0.01, respectively) (See Abstract Figure). After ablation, patients with VT recurrences showed lower scar areas on 3D and upsampled 2D models using <0.1 and <0.2 transmurality criteria, respectively, than patients without VT recurrences (48.5 [37.7, 67.5] vs. 91.5 [84.3, 117.3] cm2, respectively, for 3D-derived models; p = 0.004, and 64.0 [43.5, 69.5] vs. 83.0 [70.2, 99.7] cm2, respectively, for upsampled 2D models; p = 0.04). Conversely, time-consuming scar volume reconstructions from 2D or 3D LGE-CMR sequences did not show significant differences between patients with and without VT recurrences. Conclusions Three-dimensional transmural scar assessment in upsampled 2D-LGE-CMR-derived models provides time-efficient characterization of infarct-related myocardial substrates associated with the CL of spontaneous VT episodes and VT recurrences after ablation. Abstract Figure