Introduction: Several electrocardiography predictors of ventricular arrhythmia and sudden cardiac death risk have been described, including QT and TpeakTend (Tpe) interval duration. It was the aim of the present study to fi nd the lead with the maximal TpeakTend interval and to assess its relationship to the QT interval. Methods: A total of 65 patients (included in a heart failure group and a control group), underwent standard 12 lead ECG. QT and Tpe interval were measured in each lead and compared with the values obtained in other leads. QT interval was corrected for heart rate using the Bazett formula (QTc). Results: Maximal Tpe was measured most frequent in lead V2 (31%), V3 (28%) and DI (22%). Comparing Tpe intervals in the 12 leads, the highest values were obtained in V3 (86 ± 30 ms), V2 (85 ± 29 ms) and V4 (77 ± 22 ms). QT and Tpe positively correlated in each lead, with the best correlation in V1 (r = 0.481). A significant correlation was also found between maximal Tpe and QT intervals (QTmax) (r = 0.297). A prolonged QTmax (> 450 ms) was a specifi c predictor of Tpe > 100 ms (specifi city = 0.811, 95% CI: 0.658 – 0.905). A prolonged QTc (> 450 ms) was a sensitive predictor of Tpe > 100 ms (sensitivity = 0.607, 95% CI: 0.31 – 0.616). Conclusions: The most reliable leads for Tpe measurement in patients with heart failure and in healthy controls, are the precordial leads: V2, V3, V4. Prolonged QTc and QTmax are specific and sensitive predictors of prolonged Tpe.