MRI of total sodium (Na) content may allow assessment of myocardial viability, but information on Na content in normal myocardium, necrotic/scar tissue, and stunned or hibernating myocardium is lacking. Thus, the aims of the study were to: 1) quantify the temporal changes in myocardial Na content post‐myocardial infarction (MI) in a rat model (Protocol 1); 2) compare Na in normally perfused, hibernating, and stunned canine myocardium (Protocol 2); and 3) determine whether, in buffer‐perfused rat hearts, infarct scar can be differentiated from intact myocardium by ²³Na‐MRI (Protocol 3). In Protocol 1, rats were subjected to LAD ligation. Infarct/scar tissue was excised at control and 1, 3, 7, 28, 56, and 128 days post‐MI (N = 6–8 each), Na content was determined by ²³Na‐NMR spectroscopy (MRS) and ion chromatography. Na content was persistently increased at all time points post‐MI averaging 306*–160*% of control values (*P < 0.0083 vs. control). In Protocol 2, ²³Na‐MRS of control (baseline), stunned and hibernating samples revealed no difference in Na. In Protocol 3, ²³Na‐MRI revealed a mean increase in signal intensity, to 142 ± 6% of control values, in scar tissue. A threshold of 2 standard deviations of the image intensity allowed determination of infarct size, correlating with histologically determined infarct size (r = 0.91, P < 0.0001). Magn Reson Med 45:756–764, 2001. © 2001 Wiley‐Liss, Inc.