We have performed an experimental study aiming at characterizing the specific electrical signatures of localized heterogeneities and their role in arrhythmogenesis. For this, local repolarization, conduction and microstructural heterogeneities were created in ex vivo right ventricles. We found that microstructural substrates have the most pronounced impact on electrograms, especially when combined with sodium channel blockers, whereas local action potential duration (APD) shortening does not lead to electrogram fragmentation even though it is associated with the highest prevalence of spontaneous arrhythmias. We subsequently performed a dedicated study on the spontaneous arrhythmias occurring in the latter model (APD heterogeneity). Sixteen ventricular fibrillation (VF) initiations were optically mapped and showed triggers originating in all cases from the border zone between altered and normal repolarization areas. Importantly, our study in a model of localized repolarization heterogeneity demonstrates that VF triggers originate consistently from the border zone of the repolarization dispersion.
We have been able to produce computer models of hearts with small structural anomalies which can sustain re-entries. Using tools developed in our group, we could remove a portion of a coarse heart mesh, place the re-entrant structure within it, and then remesh the space in between the two. The heart was then paced as in sinus rhythm and body surface potentials calculated along with electrograms as recorded from clinical omnipole catheters. We also investigated the morphology of electrograms near the entrance sites of structural abnormalities which create channels. Using a simple sheet model, we compared our results with clinical electrograms recorded near the entrance of re-entrant channel. We were able to reproduce and explain the intaQRS activation alternans behaviour seen in the clinical recordings, relating it to the specific tissue conditions needed.
Building upon the challenges of accurately localizing ECG electrodes in clinical environments, two AI- and camera-based methods for automating electrode localization in body surface potential maps (BSPMs) were developed, eliminating the need for MRI or CT scans. These approaches provide safer alternatives to traditional imaging techniques, making advanced ECG mapping more practical in clinical environments.
Based on our prior work on noninvasive stimulation, we have developed a dedicated system for intercostal sonication, which has been validated in a series of pre-clinical studies. We showed that with such a transducer, intercostal sonication could be carried out without significant losses when the intercostal distance was 1 cm or more, which is within the range of values observed in vivo in man. This was quantified using dedicated MRI methods developed in the laboratory.
We conducted a clinical study to evaluate the characteristics of recurrent VF recorded on implantable defibrillator electrograms, associated with idiopathic VF. We found three variables to be the most discriminant of the underlying mechanisms: Sinus tachycardia was more frequent in microstructural substrates, whereas short-coupled triggers were most frequent in Purkinje-related VF, which also had shorter VF cycle lengths. The multivariable combination provided the highest prediction, discriminating 81% of VF substrates with a high probability (>80%). Interestingly, ectopy were inconsistently present before VF. Characteristics of arrhythmia recurrences on implantable cardioverter-defibrillators can thus provide phenotypic markers of the distinct and hidden substrates underlying IVF.
