The aim of the study was to assess whether parameters based on the T-wave loop and QRS loop predict mortality, and cardiac mortality in particular, during follow-up of consecutive survivors of acute myocardial infarction (AMI). Patients with AMI (n = 437), treated according to contemporary guidelines, underwent digital high-resolution electrocardiography in orthogonal Frank leads (X, Y, Z) 5 to 14 days after AMI. Several T-wave and QRS loop parameters, such as the width and height of the loops and their ratio, T-wave loop dispersion (TWLD), QRS loop dispersion, and co-sine of the angle between the main vectors of the T-wave and QRS loops (TCRT), were calculated using a custom-made software package. During an average follow-up period of 43 +/- 14 months, 53 patients (12%) died. Of these 53 deaths, 35 were cardiac. TWLD and TCRT were the T-wave loop/QRS loop parameters that best predicted for cardiac mortality on univariate comparison (35.4 +/- 5.62 vs 32.8 +/- 2.87 for TWLD, p < 0.001 and -0.135 +/- 0.665 vs -0.657 +/- 0.518 for TCRT, p < 0.001, alive vs cardiac death, respectively). After adjustment for clinical risk markers in the Cox regression analysis, TWLD still significantly predicted for cardiac mortality (p < 0.05); however, TCRT had lost its predictive power. TWLD did not have significant univariate or multivariate association with noncardiac mortality. In conclusion, TWLD that describes the shape of the T-wave loop is a specific predictor of cardiac death and independent of the clinical risk markers in the current treatment era of patients with AMI.