S8 - An in-silico coupled electro-mechanical model to identify arrhythmogenic patterns in mitral valve prolapse

Introduction: : Mitral valve prolapse (MVP) is a common valvular disorder, and an important subset of patients exhibits ventricular arrhythmias and even sudden cardiac death despite minimal regurgitation [1]. Arrhythmogenic MVP (A-MVP) phenotype often presents with distinctive electromechanical (EM) features that benign MVP (NA-MVP) lacks. Clinically, A-MVP patients frequently show inferior-lead T-wave inversions on the ECG, prolonged ventricular repolarization (QT interval), and complex ventricular ectopy. Structurally, A-MVP is associated with subtle fibrosis in the papillary muscles or infero-basal left ventricle, which can create a substrate for electrical heterogeneity. A hallmark of A-MVP is increased mechanical dispersion and patients with a significantly higher standard deviation of time-to-peak have a more chance of diagnosis as A-MVP [2]. These distinctive signatures underscore the need for personalized tools to differentiate A-MVP from NA-MVP for risk stratification.

Materials and

Methods: : A patient-specific electrophysiological (EP) framework was developed to determine whether the arrhythmogenic signatures could be reproduced and explained in silico. Ventricular geometry and regional strain were derived from cardiac MRI (CMR) acquired in MVP patients (Fig1,a). A finite-element model was constructed from the CMR data, and myocardial fiber architecture was assigned by means of a rule-based helix-angle algorithm (Fig1,b-c). Fiber orientations were arranged in the natural counter-wound helix from endocardium to epicardium, a configuration considered critical for realistic ventricular mechanics and electrical anisotropy. A simplified His–Purkinje network was subsequently integrated along the endocardium to represent the specialized conduction system, facilitating rapid activation propagation in vivo. Electrical activation and recovery were simulated with an anisotropic monodomain model that solved the cardiac reaction–diffusion equation on the ventricular mesh. The electrical model was coupled to a biomechanical model of active tension development in the myocardium, enabling computation of regional deformation and strain. This EP–EM coupling permitted the generation of synthetic 12-lead ECG signals and mechanical contraction metrics within a single simulation. Model parameters were calibrated so that the NA-MVP baseline simulation reproduced the subject’s CMR-derived strain and yielded a normal ECG. A-MVP-specific changes, including regional fibrosis that affected local conduction and shortening, were then introduced to create an A-MVP simulation for the same anatomy. Electrical outputs (QRS/T-wave morphology, QT interval, activation maps) and mechanical outputs (regional strain timing, displacement patterns, mechanical dispersion) were compared between the A-MVP and NA-MVP scenarios.

Results, Conclusions, and Discussions:: Preliminary simulations were conducted in which regional prolongation of action-potential duration and delayed mechanical activation were introduced in inferolateral segments to explore their potential impact on whole-heart signals. Under these imposed conditions, inferior-lead T-wave inversion and a modest lengthening of repolarization emerged, whereas the baseline (unaltered) model preserved normal repolarization timing and T-wave orientation. When Purkinje activation timing was locally slowed near papillary-muscle insertion sites (Fig 2,b), occasional premature activations appeared in the waveform output; such ectopy was rarely produced when uniform conduction was maintained. On the mechanical side, the imposed delays led to later peak strain in the affected segments, broadening the overall distribution of contraction timings, while the baseline model retained a more synchronous shortening pattern. The co-occurrence of localized electrical delay and regional mechanical dyssynchrony suggests a possible electromechanical coupling mechanism, though these effects remain modest and will require further refinement and validation before firm conclusions can be drawn.

Conclusion: The preliminary simulations indicate that a patient-specific ventricular EM framework may help differentiate A-MVP from NA-MVP and could offer a non-invasive route to risk stratification once further validated.

Acknowledgements and/or References (Optional):: RA was supported by the NIH, under grant R00HL138288 and R56HL172052 .