We tested the hypothesis that when acute coronary occlusion is caused by thrombus, part of the no-reflow phenomenon may result from spontaneous or coronary angioplasty-induced microthromboemboli, and that this phenomenon may be partly or wholly reversible. Accordingly, a thrombus was created in the left anterior descending coronary artery of 6 dogs and was labeled in vivo with (99m)Tc-DMP-444 that binds to the IIb/IIIa platelet receptor. Angioplasty was then performed to obtain thrombolysis in myocardial infarction grade-3 flow. Myocardial contrast echocardiography was performed 15 and 60 minutes after recanalization to define perfusion defect size. (99m)Tc-autoradiography and infarct size (IS) measurement were performed postmortem. An additional 5 dogs with coronary artery ligation followed by reperfusion served as control animals. These dogs also underwent myocardial contrast echocardiography and in vivo labeling with (99m)Tc-DMP-44. (99m)Tc uptake was significantly higher in the reperfused bed in dogs with thrombus compared with control dogs (2.7 +/- 0.9 vs 1.4 +/- 0.3 counts/pixel(-1)/min(-1), P =.01) indicating the presence of microthromboemboli. Perfusion defect size early (15 minutes) after recanalization was smaller than the hot spot on autoradiography and overestimated IS in dogs with thrombus. Perfusion defect size decreased with time and was closer to IS 60 minutes after recanalization. The dogs with thrombi demonstrated larger IS/risk area ratios compared with the 5 control dogs (46 +/- 6% vs 27 +/- 12%, P =.04). We conclude that part of the no-reflow phenomenon seen after angioplasty in acute coronary thrombosis is a result of microthromboemboli and is mostly reversible. No reflow late after reperfusion is a result of tissue necrosis. The thrombus burden also affects ultimate IS.