Regional disparities in ventricular repolarization are known to promote reentrant ventricular arrhythmias; recent work suggests that this may be assessed noninvasively using QT/JT dispersion on the surface ECG. However, very little data is available on its clinical use. We studied pts with unexplained syncope to determine if inducible VT could be predicted by QT dispersion (QTd, defined as QTmax-QTmin), QT standard deviation (QTstd), and equivalent JT intervals. Values were also corrected using Bazett's formula. We enrolled 38 pts (22 M, age = 59 ± 18 yrs, 29% prior MI) who underwent programmed electrical stimulation (PES). Standard ECGs were recorded prior to PES and analyzed by a computerized analyzer/digitizer. Of the 38 pts, 9 (24%) had inducible sustained VT and 29 pts did not. Old QTstd JTd JTstd VT 59 ± 18 18 ± 5 63 ± 21 19 ± 4 No VT 46 ± 14 14 ± 4 51 ± 12 15 ± 5 p value 0.03 0.02 0.08 0.03 Standard QRS, QT and QTc, and corrected dispersion intervals were not significantly different between pts with and without inducible VT. Confirming the relationship of dispersion of ventricular repolarization and ventricular reentry, several ECG parameters of dispersion were associated with inducible VT in patients with unexplained syncope. QT dispersion is a novel and noninvasive measure of risk of VT.