AbstractTo assess the validity of thallium‐201 myocardial imaging in the diagnosis of non‐transmural ischemia and infarction, the proximal left anterior descending coronary artery was partially occluded for 60 minutes with a balloon‐tip catheter in intact anesthetized dogs in a basal state or during atrial pacing. In vivo sclntigrams of myocardium were compared with those obtained in the isolated heart and in the incised ventricle spread flat. None of the animals with partial occlusion wtth or without pacing demonstrated abnormal scintiscans in vivo. Removal of background by isolating the heart increased positive images to 30%; positive images were associated wtth an isotope count ratio between ischemlc and normal muscle of less than 0.67. Removal of superimposed nonischemic muscle in the heart enface increased image detection after pacing to 11 of 15. Since animals with subendocardlal scar failed to demonstrate a “cold area” in vivo, unfavorable geometry as well as extent and degree of ischemia appear to be important Thus, thallium radioactivity in superimposed and adjacent myocardium, as well as background, may limit the detection of nontransmural ischemia