To evaluate the vasodilator effect of adenosine triphosphate (ATP) compared with adenosine in stress perfusion cardiac magnetic resonance (CMR) examinations. A total of thirty-three patients underwent clinically indicated stress/rest perfusion CMR examination following intravenous injection of a total dose of 0.2 mmol/kg of gadobutrol. Individuals were randomly assigned to ATP (160 mcg/kg/min) or adenosine (140 mcg/kg/min). The vasodilator effect of both drugs was analyzed by comparing differences in heart rate, symptoms during stress, and semiquantitative myocardial and splenic perfusion parameters, including time, time to peak, upslope, myocardial perfusion reserve index, tissue perfusion values, splenic and myocardial signal intensity ratios, and splenic-to-myocardial signal intensity ratios. No significant difference was found in heart rate variation between the stressors (26.1¿±¿19.1 bpm for ATP vs. 21.7¿±¿17.3 bpm for adenosine, p¿=¿0.52). Patients receiving ATP referred less pronounced clinical symptoms. Semiquantitative myocardial perfusion parameters were comparable, and patients in the adenosine and ATP groups showed similar myocardial perfusion reserve index values (2.34 [1.62-2.73] vs 1.63 [1.29-2.10], p¿=¿0.07). Splenic switch off was visually confirmed in all patients and estimated spleen to myocardium ratio was similar (0.92 [0.53-1.09] vs 0.81 [0.53-0.86] with ATP and adenosine, respectively, p¿=¿0.12). Both ATP and adenosine are potent coronary vasodilators that can be safely employed in stress-CMR. Both stressor cause similar hyperemic response. Splenic switch-off can be used to assess stress adequacy in patients undergoing stress-CMR with either adenosine or ATP.