Intravenous dipyridamole-thallium imaging unmasks ischemia in patients unable to exercise adequately. However, some of these patients can perform limited exercise, which, if added, may provide useful information. Treadmill exercise combined with dipyridamole-thallium imaging was performed in 100 patients and results compared with those of 100 other blindly age- and sex-matched patients who received dipyridamole alone. Exercise began after completion of the dipyridamole infusion. Mean +/- 1 standard deviation peak heart rate (109 +/- 19 vs 83 +/- 12 beats/min, p less than 0.0001) and peak systolic and diastolic blood pressure (146 +/- 28/77 +/- 14 vs 125 +/- 24/68 +/- 11 mm Hg, p less than 0.0001) were higher in the exercise group compared with the nonexercise group. There was no difference in the occurrence of chest pain, but more patients in the exercise group developed ST-segment depression (26 vs 12%, p less than 0.0001). The exercise group had fewer noncardiac side effects (4 vs 12%, p less than 0.01) and a higher target (heart) to background (liver) count ratio (2.1 +/- 0.7 vs 1.2 +/- 0.3; p less than 0.01), due to fewer liver counts. There were no deaths, myocardial infarctions or sustained arrhythmias in either group. Combined treadmill exercise and dipyridamole testing is safe, associated with fewer noncardiac side effects, a higher target to background ratio and a higher incidence of clinical electrocardiographic ischemia than dipyridamole alone. Therefore, it is recommended whenever possible.