Although infarct size correlates generally with prognosis after acute myocardial infarction, an absolute measure of infarct size may have differing prognostic significance depending on absolute left ventricular mass. To test the hypothesis that single photon emission computed tomography can accurately measure myocardial infarct size as a percent of total left ventricular mass ("infarction fraction"), thallium-201 and technetium-99m pyrophosphate tomograms were acquired in 21 dogs 24 to 48 hours after fixed occlusion of the left anterior descending or circumflex coronary artery. Pathologic infarct weight was measured as the myocardial mass that showed no staining with triphenyltetrazolium chloride. Scintigraphic infarct mass by technetium-99m pyrophosphate was calculated from the total number of left ventricular volume elements (voxels) demonstrating technetium-99m pyrophosphate uptake X voxel dimension [( 0.476 cm]3) X specific gravity of myocardium (1.05 g/cm3). Scintigraphic left ventricular mass was calculated in a similar fashion using an overlay of the thallium-201 and technetium-99m pyrophosphate scans. The "infarction fraction" was calculated as: infarction fraction = infarct mass/left ventricular mass. There was good correlation between single photon emission computed tomography and pathologic measurements of infarct mass (technetium-99m pyrophosphate mass = 1.01 X pathologic infarct mass + 0.96; r = 0.98), left ventricular mass (single photon emission computed tomographic left ventricular mass = 0.60 X pathologic left ventricular mass + 37.4; r = 0.86) and "infarction fraction" (single photon emission computed tomographic infarction fraction = 1.09 X pathologic infarction fraction - 1.7; r = 0.94).(ABSTRACT TRUNCATED AT 250 WORDS)