Résumé
the autonomic nervous system (ANS) is both a recognized trigger and modulator of arrhythmogenesis [1]. Thanks to a new method of electrocardiogram analysis, we can accurately model in vivo the effect of a compound on the ANS. This model, called HFAM (High Frequency Autonomous Modulation), is based on the study of high-frequency oscillations (>0.1Hz) of RR intervals and heart rate, and enables us to identify different states of the ANS over very short periods of 10seconds: sympathetic predominance (S3 oscillations), parasympathetic predominance (S1 oscillations) or co-activation (S2 oscillations) [2]. The latter state could not be estimated using the methods available to date.
Develop an in silico approach to understand some arrhythmias.
based on this work, we are developing an in silico prediction model, integrating the effects of the tested compounds on ANS, rhythm variability and ventricular repolarization. We have modified the M-cell model described by Ten Tusscher et al. (2006) [3] by integrating the effects on HF oscillations produced by different compounds, as well as the known effects of alpha1-, beta1-adrenergic and muscarinic stimulation on the different ionic conductances of the model.
with this bioinformatics model, we succeeded in identifying all pro-arrhythmic molecules, without any false negatives (11/11). Applied to a clinical case of recovered cardiac arrest, the model also enabled us to propose the mechanism of the cardiac arrhythmia.
This model therefore represents a breakthrough in the early prediction of iatrogenic arrhythmias and could lead to a better understanding of the mechanisms of certain clinically observed arrhythmias, in order to better prevent or treat them effectively.