The aim of this study was to determine whether the use of a magnetic resonance (MR) susceptibility contrast medium, dysprosium diethylenetriamine pentaacetic acid-bismethylamide (Dy DTPA-BMA; Sprodiamide), may characterize myocardial perfusion abnormalities in a dog model of 90 minutes of coronary occlusion followed by 24 hours of reperfusion (no-reflow phenomenon installed). First-pass MR imaging after an intravenous bolus administration of the contrast agent was performed at the end of reperfusion. Signal intensity analysis on MR imaging, planimetry of pathological data, and blood flow determination were obtained by reference methods for comparison. Dogs were separated into two groups according to the level of collateral blood flow level (group I, <22.5 % of the flow in the non-ischemic zone; group II, >22.5 % of the flow in the non-ischemic zone). Signal intensity-time curves in the ischemic and non-ischemic left ventricle walls were extracted. Mean collateral blood flow was lower during occlusion in group I (9.8 +/- 5.4%, n = 5) than in group II (38 +/- 12.5%, n = 7, P < 0.05). Mean infarct size (expressed as a percentage of the area at risk) was significantly larger in group I (low collateral blood flow; 25.3 +/- 14.6%) than in group II (high collateral blood flow; 5.8 +/- 1.1%, P < 0.05). After rapid injection, a transient decrease of signal intensity induced by Dy DTPA-BMA was observed in both remote and ischemic myocardium but more markedly in remote normally perfused myocardium. Hence, during the transit of a susceptibility-type contrast agent, ischemic myocardium after ischemia and reperfusion appeared as a relative high signal intensity area. First-pass MR imaging with susceptibility contrast agent demonstrated the no- or low-reflow phenomenon. However, the behavior of the myocardial signal intensity-time-related curves did not allow distinction between the two groups of dogs.