A recent report described a pattern of "reverse redistribution" on poststreptokinase 201Tl studies which was believed to be due to rapid washout of 201Tl from the infarct area related to reperfusion of the infarct vessel. We have also observed the phenomenon of rapid washout of 201Tl from the area of infarction in the absence of thrombolytic therapy. This study was undertaken to test the hypothesis that rapid washout of 201Tl from an area of infarction is an artifact of background subtraction usually employed in analysis of washout. A total of 61 patients with previous myocardial infarction who underwent cardiac catheterization and exercise 201Tl imaging were examined. Thallium-201 images were analyzed using a validated quantitative method employing interpolative background correction. Abnormally increased 201Tl washout was noted in 11 infarct segments in 10 (18%) patients. Infarct segments with rapid washout had significantly less initial uptake, and more severe associated wall motion abnormalities than infarct segments with normal washout. When quantitative analysis was repeated without background subtraction, no segments with rapid washout were observed. A phantom model was constructed to further test our hypothesis. The frequency of observed rapid washout was directly related to the severity of the initial defect and was entirely dependent upon utilizing background correction during the quantitative analysis. Our study suggests that rapid washout of 201Tl in an area of previous infarction reflects an artifact of background subtraction involved with standard quantitative analysis.