BACKGROUND Late potentials represent an arrhythmogenic substrate in chronically infarcted myocardium. It is hypothesized that acute transient ischemia enhances anisotropic electrical ventricular activation and facilitates reentry mechanisms. Study aim was the prospective assessment of the impact of dipyridamole-induced myocardial ischemia on the signal-averaged ECG. METHODS Dipyridamole stress thallium-201 SPECT imaging was utilized to avoid noise contamination of the signal-averaged ECG from exercise and to document evidence and localization of myocardial ischemia or persistent perfusion defects in 68 patients with suspected coronary artery disease. Before and during dipyridamole-induced vasodilatation serial signal-averaged ECG was performed to evaluate the influence of transient ischemia on the occurrence of late potentials. RESULTS There was a significant difference between heart rate at rest and heart rate under dipyridamole influence in patients with inducible ischemia (70 +/- 13 vs. 87 +/- 13; P < 0.0001) in contrast to patients without dipyridamole-induced ischemia (74 +/- 20 vs. 80 +/- 16; n.s.). The number of averaged beats and achieved noise level was comparable between both groups. Thirty-three of 68 patients (49%) revealed dipyridamole-induced ischemia; however, no changes of the SAECG parameters, such as QRS, RMS, LAS at 25-250 and 40-250 Hz bandpass filtering in the leads X, Y, Z and vector magnitude, respectively, were observed as a result of ischemia. CONCLUSIONS These results suggest that transient myocardial ischemia does not affect the signal-averaged ECG. Clinically, the signal-averaged ECG analysis seems not to be helpful in identifying patients with silent ischemia.