Increased dispersion of the QT interval has been observed during pacing or exercise stress testing in patients with coronary artery disease (CAD). It has not been established whether this phenomenon is a consequence of ischemia. Therefore, we sought to evaluate whether dipyridamole-induced myocardial ischemia, as directly detected by echocardiographic monitoring of regional contractile function, would affect QT dispersion. Twenty-four patients with nonsignificant and 34 patients with significant CAD but no previous myocardial infarction underwent dipyridamole stress echocardiography while not taking medications. QT dispersion was measured on a 12-lead electrocardiogram at baseline and at various times after dipyridamole infusion. Dipyridamole infusion did not influence QT dispersion in patients without CAD. QT dispersion was similarly unaffected in patients with CAD in whom dipyridamole did not induce wall motion abnormalities. In contrast, in patients with positive dipyridamole stress test findings, QT dispersion increased from 60 +/- 17 ms at baseline to 94 +/- 25 ms during peak infusion (p <0.0001), with a time course mirroring that of development of contractile abnormalities. QT dispersion returned to 63 +/- 25 ms upon relief of ischemia by administration of aminophylline. The increase in QT dispersion was significantly related to the extent of contractile dysfunction induced by dipyridamole. Although ST-segment depression occurred in only 40% of patients with positive dipyridamole stress test findings, 88% of such patients had an increase in QT dispersion. Analysis of the receiver-operating characteristic curve showed that a QT dispersion increase of > or =20 ms identified positive findings for dipyridamole stress echocardiography with 68% sensitivity and 91% specificity. Thus, QT dispersion is acutely affected by myocardial ischemia induced by the administration of dipyridamole. Measurement of QT dispersion may improve detection of stress-induced ischemia on surface electrocardiograms.