Background. Both thallium 201 and technetium 99m sestamibi have been used to quantitate infarct size at rest. Exercise Tl-201 scintigraphy has been shown to have powerful prognostic information after myocardial infarction. A single study using these agents that could provide data on infarct size and prognosis would be of value. The purpose of this study was to compare estimates of infarct size by use of Tl-201 and Tc-99m sestamibi and to correlate these measurements with left ventricular ejection fraction in patients after acute myocardial infarction. Methods and Results. The study group consisted of 20 patients who underwent low-level Tl-201 stress studies with reinjection and Tc-99m sestamibi resting studies within 4 days. Acute reperfusion was attempted in 18 of 20 patients. For Tc-99m sestamibi tomographic imaging, infarct size was quantitated with 60% of maximal counts per slice for five short-axis slices as described in multiple previous studies. The postreinjection delayed Tl-201 images acquired 4 hours after stress were quantitated according to the same threshold method. Tl-201 patient images were also quantitated with a commercially available polar map program and compared with sex-matched control subjects. Ejection fraction was determined for each patient by radionuclide ventriculography 6 weeks later. Ejection fraction was well preserved for the group: mean 0.53 +/- 0.10. Infarct size with Tc-99m sestamibi was 12% +/- 13% of the left ventricle, which was significantly smaller than either method with Tl-201: threshold method, 29% +/- 18% of left ventricle; polar map method, 25% +/- 17% of left ventricle (both Tl-201 estimates, p < 0.0001 vs Tc-99m sestamibi; Tl-201, 70% threshold vs Tl-201 polar map, p = 0.04). There was a significant correlation between infarct size with Tc-99m sestamibi and that with Tl-201 (r = 0.72 to 0.73; p < 0.001). Infarct size with Tc-99m sestamibi, however, provided the closest correlation with ejection fraction (r = 0.81; p < 0.001), with the two Tl-201 quantitative methods providing very similar correlations (r = 0.69; p < 0.001). Conclusions. Infarct size with reinjection Tl-201 imaging correlates significantly with resting infarct size with Tc-99m sestamibi, although it provides significantly larger estimates. Although both approaches can be combined with a same day exercise protocol, the closer correlation of infarct size with ejection fraction at 6 weeks suggests that resting infarct size with Tc-99m sestamibi may be slightly more accurate.