Noncontact whole-chamber charge density mapping of the left ventricle: Preclinical evaluation in a sheep model.

BACKGROUND: Conventional contact-based electroanatomic mapping is poorly suited for rapid or dynamic ventricular arrhythmias. Whole-chamber charge density (CD) mapping could efficiently characterize complex ventricular tachyarrhythmias and yield insights into their underlying mechanisms. OBJECTIVE: The purpose of this study was to evaluate the feasibility and accuracy of noncontact whole-chamber left ventricular (LV) CD mapping and to characterize CD activation patterns during sinus rhythm, ventricular pacing, and ventricular fibrillation (VF). METHODS: Ischemic scar as defined by CD amplitude thresholds was compared to late gadolinium enhancement criteria on magnetic resonance imaging using an iterative closest point algorithm. Electrograms recorded at sites of tissue contact were compared to the nearest noncontact CD-derived electrograms to calculate signal morphology cross-correlations and time differences. Regions of consistently slow conduction were examined relative to areas of scar and to localized irregular activation (LIA) during VF. RESULTS: Areas under receiver operating characteristic curves (AUCs) of CD-defined dense and total LV scar were 0.92 ± 0.03 and 0.87 ± 0.06, with accuracies of 0.86 ± 0.03 and 0.80 ± 0.05, respectively. Morphology cross-correlation between 8677 contact and corresponding noncontact electrograms was 0.93 ± 0.10, with a mean time difference of 2.5 ± 5.6 ms. Areas of consistently slow conduction tended to occur at scar borders and exhibited spatial agreement with LIA during VF (AUC 0.90 ± 0.02). CONCLUSION: Noncontact LV CD mapping can accurately delineate ischemic scar. CD-derived ventricular electrograms correlate strongly with conventional contact-based electrograms. Regions with consistently slow conduction are often at scar borders and tend to harbor LIA during VF.