Abstract. Patients with dilated and hypertrophic cardiomyopathies are at risk for arrhythmias and have increased QT variability. However, there has been controversy about whether increased QT variability predicts the risk for ventricular arrhythmias. The increased QT variability index that includes the prognostic power of decreased heart rate variability is shown to identify patients with sudden cardiac death in a mixed patient population. Patients with congenital long QT syndrome may be at risk for arrhythmic events even without prolongation of repolarization on a standard electrocardiogram. This observation suggests that other factors such as distrurbances in temporal dynamics of repolarization may be involved in arrhythmogenesis in these patients. Recent data show that patients with congenital long QT syndrome have increased temporal variability of repolarization duration, increased spatial complexity of repolarization, and increased temporal variability of spatial complexity of repolarization compared to unaffected family members. These observations indicate that repolarization is a very dynamic phenomenon. However, there is no direct evidence showing that these changes in repolarization dynamics are related to the risk of arrhythmic events. Gene mutations cause disturbances in cellular ionic channels leading to the prolongation of ventricular repolarization in patients with congenital long QT syndrome. It is also well known that QT interval heart rate dependency is altered in these patients. However, the exact mechanism underlying the changes in repolarization dynamics in patients with congenital long QT syndrome is not known. Recent observations show that increased temporal complexity of repolarization, measured by using approximate entropy, independently predicts mortality but not the combined endpoint of death/appropriate implantable cardioverter-defibrillator shock in high risk patients with decreased left ventricular function, a finding suggesting that these changes in nonlinear dynamics of repolarization are not specifically related to the risk for malignant ventricular tachyarrhythmias. None of the repolarization variability measures determined by conventional summary statistics were associated with mortality in these patients. These observations support the concept that such as in a case of heart rate variability nonlinear meseasures used in the analysis of beat-to-beat variability of repolarization duration may reveal more delicate changes in repolarization dynamics than conventional measures of variability. In healthy subjects repolarization and heart rate are coupled. Changes in heart rate seem to be a dominant determinant of changes in repolarization duration based on broadly observed heart rate dependency of repolarization. However, it has been suggested that the influence of autonomic nervous system on repolarization duration and heart rate may be qualitively dissimilar in patients with cardiac diseases. Different kind of changes in nonlinear dynamics of repolarization than in dynamics of heart rate were observed to be related to risk of mortality in high risk patients with decreased left ventricular function and implantable cardioverter-defibrillators. This observation suggests that autonomic modulation of repolarization dynamics and heart rate dynamics may be particularly divergent in patients with the highest risk for mortality. Changes in neurohumoral influences and ionic channels and currents are also potential factors that may explain the observed relationship between increased temporal complexity of repolarization and mortality in patients with reduced left ventricular function. However, the exact mechanism underlying this association remains to be resolved.