The short term reproducibility of exercise testing in 25 patients who had exercise induced ST segment elevation without baseline regional asynergy or a previous myocardial infarction, who had different responses to the dipyridamole test, was assessed. The patients performed a dipyridamole echocardiography test and a second exercise stress test. All underwent coronary arteriography. Seventeen patients had transient regional asynergy after dipyridamole (group 1) and either ST segment elevation (14 patients) or depression (three patients); a second group of eight had no asynergy and no electrocardiographic changes (group 2). The repeated exercise stress test was positive in 16 of the 17 patients of group 1 (11 with ST elevation and five with ST depression) and in two patients of group 2 (both had ST depression and one had coronary artery disease). The dipyridamole echocardiography test was positive in 17 of the 19 patients with coronary artery disease and was negative in all six patients without coronary artery disease. The repeated exercise stress test was positive in 17 of the 19 patients with coronary artery disease and in one patient without. The dipyridamole echocardiography test and a repeated exercise stress test, but not a single exercise stress test, identified coronary artery disease causing exercise induced ST segment elevation.