Thallium-201 (201T1) uptake and redistribution kinetics were examined in an open-chest canine preparation of occlusion and reperfusion. Seven dogs (group I) underwent 3 hr of sustained occlusion and received 1.5 mCi of 201TI after 40 min of occlusion of the left anterior descending coronary artery (LAD). Group II (n = 18) underwent 60 min of LAD occlusion followed by sudden and total release of the ligature. Group lla (n = 8) received intravenous 201T1 during occlusion of the LAD, whereas group JIb (n 10) received intravenous 201T1 at the time of peak reflow. Group III dogs (n = 26) also underwent 60 min of LAD occlusion that was followed by gradual reflow through a residual critical stenosis. Animals in this group also received 20MT1 either before (Illa; n = 16) or after reflow was established (IlIb; n = 10). In group I, the relative 201T1 gradient (nonischemic minus ischemic activity) decreased from 88 + 8% (mean + SEM) to 59 + 6% during 3 hr of coronary occlusion (p = .034). After rapid and total reperfusion (group lla), this gradient decreased from 71 + 6% during occlusion to 26 + 5% after reflow (p < .001). After slow reperfusion through a residual stenosis (group Illa), the gradient decreased from 81 ± 5% to 31 ± 5% (p < .001) (p = .56 compared with group Ha). In rapidly reperfused dogs receiving intravenous thallium during peak reflow (Ilb), initial 201T1 activity in the ischemic zone was 155 ± 20% of initial normal activity and fell to 93 ± 13% of normal after 2 hr of reperfusion. Similarly, in dogs reperfused slowly through a critical stenosis (IlIb), which received 201T1 during reflow, 201T1 activity soon after reflow was 94 ± 4% of initial normal and decreased to 80 + 6% at 2 hr of reperfusion (p = .10). Histochemical evidence of necrosis was present in the biopsy region in 80% of the 20 dogs subjected to triphenyl tetrazolium chloride (TTC) staining. Microsphere-determined transmural blood flow was similar in all groups during LAD occlusion and final flows after 2 hr were comparable in all subgroups undergoing reflow. Ischemic zone flow (% normal) was significantly higher at the time of 201TI administration in groups Ilb (192 ± 25%) and IfIb (1 10 + 5%), which received 201T1 during reflow, than in groups lla (31 ± 9%) and Illa (22 + 5%), which received 201T1 during occlusion. These differences in flow at the time of administration of 20MT1 explain the different thallium uptake patterns observed. These data suggest that after 1 hr of LAD occlusion there is no difference between rapid reperfusion through a totally patent vessel and slow reperfusion through a critical stenosis with regard to ultimate degree of flow restoration or magnitude of 201T1 redistribution in instances in which 201T1 is given before reflow. With both methods of reperfusion a residual 201T1 gradient is seen. Administration of 201T1 during reflow, however, probably overestimates the degree of myocardial salvage as reflected by final 201T1 uptake values. In dogs rapidly reperfused, a relative "hot spot"' of 201T1 activity was observed in the ischemic zone when 201T1 was administered at peak reflow, despite histochemical evidence of necrosis. These results have clinical implications with respect to the timing of 201T1 administration and interpretation of serial 201T1 scintigrams in patients with acute myocardial infarction undergoing thrombolysis.