Objectives: Mental stress can significantly affect ventricular repolarization, which could potentially trigger arrhythmias. We compared the effect of mental stress on repolarization indexed by the amplitude and area of the T wave in patients with coronary artery disease (CAD) and healthy subjects. Methods: Fourteen healthy controls (11 M, mean age 42 years) and 14 patients with stable CAD (12 M, mean age 64) underwent a mental stress protocol consisting of mental arithmetic followed by a speech (5 minutes each), which was performed on two occasions following either nitroglycerine (NTG) or placebo. Multiple 12-lead electrocardiograms were acquired and repolarization was analyzed using automatically measured T wave amplitude (Tamp) and area (Tarea). Results: When preceded by placebo the overall effect of mental stress, whether induced by arithmetic or speech, was significantly different in CAD patients compared with controls, with a decrease in Tamp and Tarea in controls and an increase in patients; e.g., change in Tamp during arithmetic −20 ± 3 &mgr;V in controls versus 4 ± 2 &mgr;V in patients, p < .001, and during speech −9 ± 3 &mgr;V in controls versus 7 ± 1 &mgr;V in patients, p < .001. Following NTG, the effect of stress on repolarization was similar in the 2 groups, with a reversed effect, i.e., decrease instead of increase in Tamp and Tarea in CAD patients. Conclusions: The effect of mental stress on ventricular repolarization is significantly different in CAD patients compared with healthy controls. These differences are considerably reduced by NTG. CAD = coronary artery disease; ECG = electrocardiogram; NTG = nitroglycerine; Tamp = maximal amplitude of the T wave; Tarea = total area under the T wave.