Both 201Tl redistribution and persistent glucose uptake have been proposed as markers of viability after reperfusion. In the present study, they have been compared in the same open-chest canine preparation of occlusion and reperfusion. Ten fasting dogs were subjected to 2 hr of left anterior descending coronary artery occlusion and 4 hr of reperfusion. Myocardial blood flow was determined by a microsphere technique 100 min after occlusion and 3 hr after reperfusion. 201Tl was injected intravenously 20 min before reperfusion. Serial biopsy samples were obtained from ischemic and normal areas. 18F-2-deoxyglucose, a tracer of exogenous glucose uptake, was injected 3 hr after reperfusion. Thirty minutes before the animals were killed, simultaneous blood samples were taken from the femoral artery and the regional coronary veins draining the reperfused and the remote areas. Dogs were killed 4 hr after reperfusion was established. Area at risk was assessed by dye injection in vivo and area of necrosis by triphenyl tetrazolium chloride (TTC) staining, with confirmation by electron microscopy. Immediately after death, endocardial and epicardial samples were taken from regions characterized as risk regions, areas of necrosis, areas of patchy necrosis, and normal areas. These samples were counted in a scintillation well counter. Four hours after reperfusion, in ischemic myocardium (TTC positive) the relative 201Tl gradient between ischemic and normal regions was 26 +/- 13%, whereas in necrotic samples, this gradient was 71 +/- 26%.(ABSTRACT TRUNCATED AT 250 WORDS)