[Quantitative analysis of variability of electrocardiograms typical for polymorphic arrhythmias]

A new approach to the analysis of variability of electrocardiograms (ECGs) typical of polymorphic arrhythmias is developed. In these ECGs, separate QRS complexes can be often hardly identified. As a result, the mathematical methods that have been elaborated hitherto are not suitable for such arrhythmias. The approach presented here is based on the quantitative estimation of the variability of neighboring parts of the ECG. In this case, the necessity of the identification of separate QRS complexes ceases to be significant. Based on this approach, the analysis of normalized ECG variability is developed in the framework of which two indices that characterize the oscillation variability and its changes in time are related to a part of the ECG and/or the ECG as a whole. Variations of these indices allow both the polymorphism of a separate ECG to be estimated and different ECGs to be compared with each other. The method presented may be useful in studies of the mechanisms and in the diagnosis of polymorphic arrhythmias.