Quantifying local cardiac substrate heterogeneity from high density recordings: an experimental study

• Published in: Europace (London, England) Vol. 26; no. Supplement_1
• Main Authors: Ruiperez-Campillo, S, Pancorbo, L, Ramirez, E, Chorro, F J, Merino, J L, Casado-Arroyo, R, Castells, F, Millet, J
• Format: Journal Article
• Language: English
• Published: 24.05.2024
• ISSN: 1099-5129; 1532-2092
• DOI: 10.1093/europace/euae102.650

Abstract Background and Purpose High-density (HD) grid catheters offer enhanced local electrical conduction characterization, owing to their unique grid arrangement. This arrangement enables the generation of orientation-independent EGMs, termed omnipolar EGMs (oEGMs), overcoming limitations associated with traditional unipolar and bipolar EGMs. This innovation holds potential for differentiating healthy tissue from scarred or arrhythmogenic areas. In this study, we introduce a novel metric, the Vector Field Heterogeneity (VFH), designed to quantify the heterogeneity of propagation wavefronts in local vector maps derived from oEGMs. Methods We developed and validated the VFH metric through a combination of simulations and experiment conducted on Langendorff-perfused rabbit hearts. Simulations involved creating a model that generated 100,000 simulated maps exhibiting various levels of disorganization. For experimental data, 68 propagation maps were utilized, including 29 from basal recordings, 39 from recordings with stimulation (4 or 6Hz), from five Langendorff-perfused rabbit hearts at 37ºC, using a multielectrode mapping catheter with 128 electrodes spaced 1mm apart. The VFH metric, based on vector field theory (see figure 2), was computed and compared to the widely adopted the Spatial Inhomogeneity Index (SI) from Statistical analysis was then performed to assess the distribution of heterogeneity values. Results The analysis of the VFH metric through simulations revealed a linear increase in VFH values with disorganization up to a 45º angle, followed by a plateau. Different catheter grid sizes demonstrated that larger grids reduced the standard deviation, with the VFH metric converging to a lower value of approximately 0.62 for larger grids. Experimental data indicated that for a 4x4 grid, VFH values averaged 0.35 without stimulation and decreased to 0.11 with stimulation. The VFH metric consistently outperformed the SI index in binary classification of basal vs stimulated conditions, as evidenced by higher Area Under Curve (AUC) values in both Receiver Operating Characteristic (ROC) and Precision-Recall (PR) analyses across various catheter sizes. For example, in the 4x4 grid, the AUROC for the VFH metric was 0.9752 compared to 0.8311 for the SI index, showcasing the VFH metric’s superior accuracy in assessing cardiac wavefront heterogeneity (see figure 2). Conclusion Vector Field Heterogeneity (VFH) emerges as an effective tool for characterizing cardiac tissue. It successfully differentiates between stimulated and non-stimulated tissue, accurately representing progressive disorganization in simulated maps. The VFH metric demonstrates superiority over the established Spatial Inhomogeneity (SI) index, making it a reliable parameter for assessing cardiac heterogeneity, especially on small high-density grids. It holds promise for clinical use in studying and understanding the cardiac substrate during arrhythmic conditions.Figure 1.Abstract Figure.Figure 2.Methods and Results.