Time-domain analysis of beat-to-beat variability of repolarization morphology in patients with ischemic cardiomyopathy

There is growing evidence that beat-to-beat changes in ventricular repolarization contribute to increased vulnerability to ventricular arrhythmias. Beat-to-beat repolarization variability is usually measured in the electrocardiogram (ECG) by tracking consecutive QT or RT intervals. However, these measurements strongly depend on the accurate identification of T-wave endpoints, and they do not reflect changes in repolarization morphology. In this article, we propose a new computerized time-domain method to measure beat-to-beat variability of repolarization morphology without the need to identify T-wave endpoints. The repolarization correlation index (RCI) is computed for each beat to determine the difference between the morphology of repolarization within a heart-rate dependent repolarization window compared to a template (median) repolarization morphology. The repolarization variability index (RVI) describes the mean value of repolarization correlation in a studied ECG recording. To validate our method, we analyzed repolarization variability in 128-beat segments from Holter ECG recordings of 42 ischemic cardiomyopathy (ICM) patients compared to 36 healthy subjects. The ICM patients had significantly higher values of RVI than healthy subjects (in lead X: 0.045 +/- 0.035 vs. 0.024 +/- 0.010, respectively; P < .001); 18 (43%) ICM patients had RVI values above the 97.5th percentile of healthy subjects (>0.044). No significant correlation was found between the RVI values and the magnitude of heart rate, heart rate variability, QTc interval duration, or ejection fraction in studied ICM patients. In conclusion, our time-domain method, based on computation of repolarization correlation indices for consecutive beats, provides a new approach to quantify beat-to-beat variability of repolarization morphology without the need to identify T-wave endpoints.