Purpose. To evaluate a standardized definition of delayed hyperenhancement in the analysis of contrast-enhanced cardiac magnetic resonance (ceCMR) imaging. Patients and Methods. CeCMR was performed in 15 patients with chronic ischemic heart disease. Delayed hyperenhancement was analyzed both by visual analysis by an experienced team of observers, and after thresholding the window setting of the images at 2, 3, 4, 5, and 6 SD above the mean signal intensity of remote, normal myocardium in the same slice. In each patient, total infarct size (TIS) and segmental infarct extent (SIE) were calculated. Results. TIS and SIE were 22.9 ± 12.2 mL and 32 ± 28% after visual analysis. Thresholding the window setting at 2, 3, 4, and 6 SD above signal intensity of remote caused a 40%, 31%, and 17% increase (p < 0.007) and a 7% decrease (p = NS) in TIS, and a 75%, 41%, and 16% increase and 22% decrease in SIE (p < 0.001), respectively. There was no difference between visual analysis and analysis after thresholding at 5 SD. Conclusion. Analyzing ceCMR with a standardized definition of hyperenhancement related to the signal of remote, nonenhanced myocardium may result in considerable overestimation of infarct size at the usual cut-off of 2 SD.