Digital MEDIcal TWIN for the Prediction of Arrhythmic Sudden Cardiac Death After a Myocardial Infarction

Overview

TWIN-SCD STEMI is a multicentric prospective, non-randomized pilot study designed to establish a multiparametric model for predicting life-threatening ventricular arrhythmias in patients experiencing their first myocardial infarction. The primary objective of the study is to develop an algorithm with superior predictive performance (sensitivity and specificity) compared to the currently used criterion for the implantation of a cardiac defibrillator, which is a left ventricular ejection fraction (LVEF) of less than 35%.

Sudden cardiac death (SCD) accounts for 10% of deaths in the general population, with 80% of these cases caused by malignant ventricular arrhythmias. These arrhythmias predominantly occur in individuals with cardiomyopathies, especially those with a history of myocardial infarction (MI). Currently, SCD prevention after a first MI relies on risk stratification based on myocardial contractility, specifically indicating prophylactic cardiac defibrillator implantation for patients with a left ventricular ejection fraction (LVEF) ≤ 35%. However, this approach is insufficient due to two main reasons: 1. Among patients with LVEF ≤ 35%, only a minority (2-5% per year) will experience arrhythmias and benefit from defibrillator implantation, while all face potential serious complications from the device. 2. Most SCD cases occur in patients with LVEF \> 35%. Although these patients have a lower individual arrhythmia risk (1-2% per year), they represent a substantially larger population at risk, with no stratification within this group. Emerging evidence suggests that additional parameters (biological markers, cardiac imaging, and electrophysiologic data) can better characterize the arrhythmogenic substrate in patients after their first MI, enabling improved identification of those at risk for arrhythmic SCD. The primary objective of this study is to develop a multiparametric model that outperforms the current criterion (LVEF \< 35%) for predicting the need for cardiac defibrillator implantation in patients presenting with a first myocardial infarction. Clinical data, biological data, imaging data of the scar using CT scan and MRI, and electrophysiological data from 12-lead ECG, 24-hour Holter ECG will be used to establish this multiparametric model. The model will be compared to the current criteria for predicting the occurrence of malignant ventricular arrhythmias (spontaneous during the one-year follow-up or induced by an electrophysiology study at 12 months and 24 months) in this cohort.