Aims: QTV has been shown to be predictive of cardiac mortality post MI. The aim of this study was to decompose QTV into physiological sources and assess their individual role for cardiac risk stratification in post MI patients to shed some light on the possible mechanisms underlying QTV’s predictive power. Methods: ECG and respiratory signals of 941 patients post MI were – recorded at the hospital of the Technische Universität München, and the Klinikum Rechts der Isar, both in Munich, Germany - analysed retrospectively. Signals were recorded at 1.6 kHz for 30 minutes in each patient within 2 weeks of index MI. The main outcome measure was total mortality during a follow-up period of 5 years. Patients were excluded from the analysis if they had atrial fibrillation or more than 10% ectopic beats. Beat to beat RR, QT and respiration were extracted from the recorded signals. Parametric spectral estimation using autoregressive modelling was employed to decompose QTV into contributions by heart rate, respiration and unexplained sources. Results: Multivariate cox regression analysis showed that total QTV and the part independent of RR and respiration (alongside QTc and mean QT) predicted 5-year mortality rate with a hazard ratio of 2.7 [2.0-3.6] (p <0.001). A combined QTV risk score (constructed from these variables) added significant predictive value when combined with traditional risk markers (HR=1.76 (1.2-2.5), p=0.002). In patients with LVEF > 35%, and in those with GRACE score ≥120, the QTV risk score was associated with a significantly higher 5-year mortality rate (19% vs. 4.1%, p < 0.0001 and 38% vs. 9.6%, p <0.0001, respectively). Conclusion: Mechanism driving QTV independent of heart rate and respiration play a dominant role in the predictive power of QTV post MI. The QTV risk score may help identify high risk patients which are traditionally considered low risk.