Estimating the Probability of Cellular Arrhythmias with Simplified Statistical Models that Account for Experimentally Observed Uncertainty in Underlying Biophysical Mechanisms

Early after-depolarizations (EADs) are action potential (AP) repolarization abnormalities that can trigger lethal arrhythmias. Simulations using biophysically-detailed cardiac myocyte models can reveal how model parameters influence the probability of these cellular arrhythmias, however such analyses can pose a huge computational burden. We have previously developed a highly simplified approach in which logistic regression models (LRMs) map parameters of complex cell models to the probability of ectopic beats (EBs). Here, we extend this approach to predict the probability of early after-depolarizations (P(EAD)). We use the LRM to investigate how changes in parameters of the slow-activating delayed rectifier current (IKs) affect P(EAD) for 17 different Long QT syndrome type 1 (LQTS1) mutations. We compare P(EAD) for these 17 LQTS1 mutations with two other recently proposed model-based arrhythmia risk metrics. These three model-based risk metrics yield similar prediction performance; however, they all fail to predict relative clinical risk for a significant number of the 17 studied LQTS1 mutations. The consistent successes and failures of all three risk metrics suggest that important functional characteristics of LQTS1 mutations may not yet be fully known. Author summaryAn early after-depolarization (EAD) is an abnormal cellular electrical event which can trigger dangerous arrhythmias in the heart. We use our previously developed method to build a simple logistic regression model (LRM) that estimates the probability of EAD (P(EAD)) as a function of myocyte model parameters. Using this LRM along with two other recently published model-based arrhythmia risk predictors, we estimate risk of arrhythmia for 17 Long QT syndrome type 1 (LQTS1) mutations. Results show that all approaches have similar prediction performance in that there are a set of mutations whose relative clinical risk for arrhythmia are well estimated using these metrics, but that relative risk is consistently over- or under-estimated across all approaches for a significant number of other mutations. We believe this indicates that the functional characterization of the LQTS1 phenotype is incomplete.