The European Myocardial Infarct Amiodarone Trial (EMIAT) investigated the effects of amiodarone versus placebo in patients after myocardial infarction who had left ventricular ejection fraction ≤40% and were ≤75 years of age. The present substudy examined whether ventricular repolarization (VR) dynamicity could differentiate patients who died from cardiac death from their matched survivors in this EMIAT population. In addition, we assessed whether VR dynamicity could differentiate patients who died from arrhythmic cardiac death (ACD) and from non-ACD. VR dynamicity (determined from Holter's recordings at baseline) was compared before antiarrhythmic therapy in 118 patients who had cardiac death and 118 matched survivors according to age, gender, left ventricular ejection fraction, and subsequent administration or nonadministration of amiodarone. VR dynamicity was compared within the cardiac death group between the 59 patients who died from ACD and the 59 who died from non-ACD. VR dynamicity was expressed as the slope of the linear regression between QTo (measured automatically) and stable RR intervals. Patients who died were found to have a significant steeper rate dependence of QTo intervals during the 3 periods than their matched survivors. In multivariate analysis, the QTo/RR nocturnal interval appeared to be the best independent predictor of cardiac death. In addition, patients who died from ACD were found to have a significant steeper rate dependence of QTo intervals during the morning period than those who died from non-ACD. In the multivariate analysis, the QTo/RR morning interval remained the best independent predictor of ACD. Thus, in the EMIAT trial, evaluation of QT dynamicity is a strong predictor of cardiac death. In addition, QT dynamicity could predict the occurrence of ACD in cases of cardiac death.