Technetium-99m 2,3-dimercaptosuccinic acid (Tc-99m DMSA) has been used successfully for imaging acute myocardial infarction in a canine model. The application in humans, however, has not been previously reported. In order to determine the feasibility of using this agent in clinical studies and to compare the agent to technetium-99m pyrophosphate (Tc-99m PPi), ten patients with proven myocardial infarction were studied. While imaging of transmural infarctions in humans was achieved using Tc-99m DMSA, scores for the Tc-99m DMSA images (1.8 +/- 0.96) were not as high as for Tc-99m PPi (2.5 +/- 0.45) (P less than 0.05). Discordance among four independent interpreters was greater for images obtained with Tc-99m DMSA. The superiority of Tc-99m PPi was evident whether images were obtained early (within 24 hours) or late (within five days). Although DMSA images were not obscured by rib uptake, they were less sensitive (63%) than Tc-99m PPi (97%). A potential advantage of Tc-99m DMSA in imaging acute myocardial infarction is that radiotracer concentration in the infarct occurs primarily in the early postinfarction period. The longer postinfarction that Tc-99m DMSA imaging was attempted, the lower the concentration of radiotracer. Thus, Tc-99m DMSA would not be expected to have the same persistence pattern as Tc-99m PPi into the remote postinfarction period. The persistent positivity of Tc-99m PPi has made it difficult to diagnose reinfarction.