Although both intravenous dipyridamole and adenosine have been used to produce coronary vasodilation during cardiac imaging, the relative potency of the commonly administered doses of these agents has not been evaluated. Accordingly, the coronary and systemic hemodynamic effects of intravenous adenosine (140 micrograms/kg per min) and intravenous dipyridamole (0.56 mg/kg over 4 min) were compared with a maximally dilating dose of intracoronary papaverine in 15 patients. Coronary blood flow responses were assessed using a Doppler catheter in a nonstenotic coronary artery. The protocol was discontinued in two patients because of transient asymptomatic atrioventricular (AV) block during adenosine infusion. The mean heart rate increased more with adenosine (11 +/- 9 beats/min) and dipyridamole (11 +/- 7 beats/min) than with papaverine (4 +/- 3 beats/min, p less than 0.05 vs. adenosine and papaverine). The mean arterial pressure decreased less with dipyridamole (-10 +/- 3 mm Hg) and papaverine (-9 +/- 4 mm Hg) than with adenosine (-16 +/- 5 mm Hg, p less than 0.01 vs. dipyridamole and papaverine). The peak/rest coronary blood flow velocity ratio was greater with papaverine (3.9 +/- 1.1) than with adenosine (3.4 +/- 1.2, p less than or equal to 0.05 vs. papaverine) or dipyridamole (3.1 +/- 1.2, p less than 0.01 vs. papaverine). A larger decrease in coronary resistance as measured by the coronary vascular resistance index occurred with papaverine (0.25 +/- 0.06) and adenosine (0.26 +/- 0.09) than with dipyridamole (0.31 +/- 0.10, p less than 0.01 vs. papaverine, p less than 0.05 vs. adenosine).(ABSTRACT TRUNCATED AT 250 WORDS)