This study compared the results of planar and single-photon emission computed tomographic (SPECT) imaging with tetrofosmin with those of201Tl and coronary angiography. In three normal volunteers the images were normal by both201Tl and tetrofosmin (planar and SPECT). In 23 patients with coronary artery disease, the images were abnormal in 20 patients by SPECT tetrofosmin, in 19 by planar tetrofosmin, in 20 by SPECT thallium, and in 18 by planar thallium (difference not significant). Both planar and SPECT images were divided into five segments per patient. There were 58 perfusion defects by SPECT tetrofosmin, 50 by planar tetrofosmin (difference not significant), 47 by SPECT thallium, and 42 by planar thallium (difference not significant). Perfusion defects were reversible in 47 segments (36%) by SPECT tetrofosmin, 35 (27%) by planar tetrofosmin, 31 (24%) by SPECT thallium (p<0.05 vs SPECT tetrofosmin), and 31 (24%) by planar thallium (difference not significant). Among the 23 patients with coronary artery disease, 19 underwent coronary angiography. In these patients there were 32 diseased coronary arteries. Perfusion defects were present in 21 territories (66%) by SPECT tetrofosmin, 19 (59%) by planar tetrofosmin, 20 (63%) by SPECT thallium, and 18 (56%) by planar thallium. There was agreement between thallium and tetrofosmin in 108 of 130 segments (κ statistics=0.69±0.06). The images, especially with SPECT, are better with tetrofosmin than with201Tl. Thus myocardial imaging with tetrofosmin provides results that are at least as good as those of201Tl. Slightly more abnormal segments and more reversible defects are detected by tetrofosmin than by thallium imaging, especially with SPECT imaging. © 1994 American Society of Nuclear Cardiology.