Accuracy of noninvasive electrocardiographic imaging using isopotential versus isochronal map for identifying the site of origin of ventricular arrhythmias

Funding Acknowledgements Type of funding sources: Private hospital(s). Main funding source(s): Learning Health Background and aim Previous studies reporting the results of noninvasive electrocardiographic imaging (ECGI) used both isopotential map and isochronal map to assess the site of origin (SOO) of premature ventricular contractions (PVCs). The aim of this study was to evaluate the accuracy of both methods globally, and according to the location of the arrhythmia: from the right ventricular outflow tract (RVOT) or other locations (non-RVOT). Methods We studied with ECGI 35 consecutive patients with frequent (>10.000/24 h) PVCs. Patients were excluded if the ablation was not performed (7 patients) or was unsuccessful (7 patients). The study group consisted of 21 patients, 11 male, median age of 56 (44-71) years. The ECGI was performed with the epi-endocardial system Amycard. Two noninvasive maps were obtained: isopotential map based on the analysis of the shape and amplitude of the unipolar electrogram (FND method) and an isochronal map obtained using the activation direction method (ADM) (Figure). The invasive activation map was obtained with the Carto or the Ensite system and radiofrequency was applied at the earliest activation site with QS morphology on the unipolar electrogram and a pace match of at least 11/12. The SOO of the PVCs was considered the site where the PVCs were abolished. We assessed the accuracy of the ECGI to identify the SOO of the PVCs using both methods. A perfect match was defined as a predicted location by the ECGI within the same anatomic segment of the actual SOO of the PVCs, whereas a near match as a predicted location within the same segment or a contiguous one. Values are presented as median (Q1-Q3) Results PVCs originated in the RVOT in 11 (52%) patients. The percentage of near matches was not significantly different between FND and ADM methods (95% vs 86%, p=0.50), however the percentage of perfect matches was significantly higher with the FND than with the ADM technique (95% vs 67 %, p=0.031). We found no significant differences in accuracy according to the location of the PVCs, in the RVOT or outside (Table). Conclusions We found a good agreement between ECGI and invasive maps, however the FND technique showed a better accuracy regardless the site of the PVCs.