Viewpoint

What have we learned from mathematical models of defibrillation and postshock arrhythmogenesis? Application of bidomain simulations

Historically, the evidences of safety and efficacy that companies provide to regulatory agencies as support to the request for marketing authorization of a new medical product have been produced experimentally, either in vitro or in vivo. More recently, regulatory agencies started receiving and accepting evidences obtained in silico, i.e. through modelling and simulation. However, before any method (experimental or computational) can be acceptable for regulatory submission, the method itself must be considered “qualified” by the regulatory agency. This involves the assessment of the overall “credibility” that such a method has in providing specific evidence for a given regulatory procedure. In this paper, we describe a methodological framework for the credibility assessment of computational models built using mechanistic knowledge of physical and chemical phenomena, in addition to available biological and physiological knowledge; these are sometimes referred to as “biophysical” models. Using guiding examples, we explore the definition of the context of use, the risk analysis for the definition of the acceptability thresholds, and the various steps of a comprehensive verification, validation and uncertainty quantification process, to conclude with considerations on the credibility of a prediction for a specific context of use. While this paper does not provide a guideline for the formal qualification process, which only the regulatory agencies can provide, we expect it to help researchers to better appreciate the extent of scrutiny required, which should be considered early on in the development/use of any (new) in silico evidence.

Computational modelling, combined with experimental investigations, is a powerful method for investigating complex cardiac electrophysiological behaviour. The use of rabbit-specific models, due to the similarities of cardiac electrophysiology in this species with human, is especially prevalent. In this paper, we first briefly review rabbit-specific computational modelling of ventricular cell electrophysiology, multi-cellular simulations including cellular heterogeneity, and acute ischemia. This mini-review is followed by an original computational investigation of variability in the electrophysiological response of two experimentally-calibrated populations of rabbit-specific ventricular myocyte action potential models to acute ischemia. We performed a systematic exploration of the response of the model populations to varying degrees of ischemia and individual ischemic parameters, to investigate their individual and combined effects on action potential duration and refractoriness. This revealed complex interactions between model population variability and ischemic factors, which combined to enhance variability during ischemia. This represents an important step towards an improved understanding of the role that physiological variability may play in electrophysiological alterations during acute ischemia.

Detailed knowledge of tissue response to both systolic and diastolic shock is critical for understanding defibrillation. Diastolic field stimulation has been much less studied than systolic stimulation, particularly regarding transient virtual anodes. Here we investigated high-voltage-induced polarization and activation patterns in response to strong diastolic shocks of various durations and of both polarities, and tested the hypothesis that the activation versus shock duration curve contains a local minimum for moderate shock durations, and it grows for short and long durations. We found that 0.1–0.2-ms shocks produced slow and heterogeneous activation. During 0.8–1 ms shocks, the activation was very fast and homogeneous. Further shock extension to 8 ms delayed activation from 1.55 ± 0.27 ms and 1.63 ± 0.21 ms at 0.8 ms shock to 2.32 ± 0.41 ms and 2.37 ± 0.3 ms (N = 7) for normal and opposite polarities, respectively. The traces from hyperpolarized regions during 3–8 ms shocks exhibited four different phases: beginning negative polarization, fast depolarization, slow depolarization, and after-shock increase in upstroke velocity. Thus, the shocks of >3 ms in duration created strong hyperpolarization associated with significant delay (P < 0.05) in activation compared with moderate shocks of 0.8 and 1 ms. This effect appears as a dip in the activation-versus-shock-duration curve.

Cardiac tissue is always surrounded by conducting fluid, both in vivo (blood) and in experimental preparations (Tyrode's solution), which acts to increase conduction velocity (CV) close to the tissue-fluid interface, inducing transmural wavefront curvature. Despite its potential importance, computer modeling studies focused on arrhythmia mechanisms have previously not accounted for these bath-loading effects. Here, we investigate the increase in CV and concomitant change in transmural wavefront profiles upon both propagation and arrhythmia dynamics within models of differing anatomical complexity. In simplified slab models, in absence of transmural fiber rotation, bath-loading induced transmural wavefront curvature dominates, significantly increasing arrhythmia complexity compared to no bath. In the presence of fiber rotation, bath-loading effects are less striking and depend upon propagation direction: the bath accentuates natural concave curvature caused by transmurally rotating fibers, but attenuates convex curvature, which negates overall impact upon arrhythmia complexity. Finally, we demonstrate that the high degree of anatomical complexity within whole ventricular models modulates bath-loading induced transmural wavefront curvature. However, key is the increased surface CV that dramatically reduces both arrhythmia inducibility and resulting complexity by increasing wavelength and reducing the available excitable gap. Our findings highlight the importance of including bath-loading effects during arrhythmia mechanism investigations, which could have implications for interpreting and comparing simulation results with experimental data where such effects are inherently present.

Despite an impressive battery of preclinical cardiac electrophysiology experimental models and the assessment of QT during clinical trials, the risk of Torsades de Pointes (TdP), a potentially lethal ventricular arrhythmia, remains among the common reasons for drug market withdrawal or lack of approval. Due to the low prevalence of TdP, development of statistical evidence that other clinical markers could be better predictors of TdP has proven challenging. Preclinical studies have provided a deeper understanding of torsadogenic mechanisms and potential pro-arrhythmic markers to assess. Translating these preclinical insights into a quantitative clinical risk assessment remains challenging because of (i) species differences in cardiac electrophysiology and drug pharmacokinetics; and (ii) the inability to measure clinically specific cardiac electrophysiology metrics, and therefore ascertain the full predictive value of earlier preclinical components of the risk assessment process. The integrative capacity of cardiac electrophysiology modeling to span time and length scales may provide a quantitative and predictive framework, to complement expert-based preclinical- to-clinical cardiac risk assessment process. In this review, we present salient elements of this risk assessment process and describe essential components of cardiac electrophysiology modeling, to propose that a progressive integration of mechanistic components into a common quantitative framework may help improve the predictability of drug-induced TdP risk.