Spin-echo cardiac magnetic resonance imaging studies were performed in 20 patients with a first 7- to 14-day-old (mean 10) myocardial infarction. The magnetic resonance imaging findings were compared with coronary angiography (14 patients), myocardial enzyme release (18 patients), radionuclide angiography (19 patients), and thallium-201 perfusion scintigraphy (19 patients). Regional T2 relaxation times determined from the signal intensities at echo times 30 msec and 90 msec were significantly prolonged in the infarcted areas. Based on abnormal T2 times for every patient, a regional and a total myocardial damage score was determined. The infarct-related artery was correctly identified in 93% of patients by magnetic resonance imaging, in 79% of patients by thallium-201 scintigraphy, and in 62% of patients by radionuclide angiography. The total damage score correlated well with enzymatic infarct size (r = 0.75, p less than 0.001). The correlation between left ventricular end-systolic volume index determined by magnetic resonance imaging and by radionuclide angiography was r = 0.89 (p less than 0.002). The left ventricular end-systolic volume index correlated significantly with enzymatic infarct size (r = 0.72, p less than 0.001), total damage score (r = 0.68, p less than 0.002), and radionuclide left ventricular ejection fraction (r = -0.68, p less than 0.002). Correlations between the magnetic resonance damage score and the thallium-201 perfusion score were r = 0.60 (p less than 0.01) for the exercise images, and r = 0.72 (p less than 0.001) for the redistribution images. This study shows that spin-echo magnetic resonance imaging is quite comparable with the established noninvasive imaging modalities currently used in patients with acute myocardial infarction.