1. Prolongation of the QT interval is associated with a risk of life-threatening cardiac arrhythmias. In the present study, we examined whether reversible blockade of preganglionic sympathetic fibres of the heart, achieved by thoracic epidural anaesthesia, affects cardiac repolarization and thus confers an anti-arrhythmic effect. 2. Fifty-two men, aged between 28 and 65 years, were included in the study: 28 were patients scheduled for thoracic epidural anaesthesia (Group T) and 24 were patients scheduled for lumbar epidural anaesthesia (Group L). Epidural blockade was achieved with 0.5% isobaric bupivacaine solution. Measurements were taken from electrocardiogram fragments obtained before epidural anaesthesia and after detection of blockade (T1 or T8 segment sensory block in Groups T and L, respectively). Correction of the QT interval was calculated using Bazett's formula (QTcb), Fridericia's correction (QTcf) and the Framingham formula (QTcF). Transmural dispersion of repolarization (TDR) was determined using the Tpeak--Tend time, where Tpeak is the peak of the T wave and Tend is the end of the T wave. 3. Significant shortening of the QTcb interval and TDR was detected after blockade in Group T. However, there were no changes in the ECG parameters in Group L. 4. In conclusion, reversible blockade of preganglionic sympathetic fibres to the heart, achieved by thoracic epidural anaesthesia, results in a reduction in QTcb, as well as the TDR. These changes may explain the anti-arrhythmic action seen with central blockade.