Twelve anesthetized mongrel dogs underwent left thoracotomy with placement of a removable ligature around the left circumflex coronary artery. Following a 3 to 6 hour delay, ECG-gated spin-echo MRI was performed. The ligature was then removed reperfusing the heart, and after a 10-15 min period, MRI repeated. Finally, post-sacrifice images were obtained, and the hearts chemically stained for infarct evaluation. The MR images were subjectively and quantitatively evaluated for visibility of the endocardial border and of the injured myocardium, and for changes after reperfusion. The injured tissue was variably visible in vivo, the major limitation a result of motion blurring and artifact. The abnormal tissue was easily visible on MRI in 11 animals, and not clearly visible in one. The endocardial border was easily seen in 10 animals. The variation of calculated relaxation times was high for both normal and ischemic/infarcted myocardium in the beating hearts (normal: T1 = 566 ± 288, T2 = 38 ± 6; injured myocardium: T1 = 637 ± 250, T2 = 41 ± 12) in contrast, relatively stationary skeletal muscle measured in the same images had narrower ranges (T1 = 532 ± 199, T2 = 28 ± 2). Changes with reperfusion were seen, but not reliably. The infarcted or ischemic zones were easily visible on post-sacrifice images in all animals imaged. Post-sacrifice relaxation times were T1 = 564 ± 69 msec, T2 = 39 ± 3 msec for normal heart muscle, and 725 ± 114, T2 = 47 ± 5 for ischemic/infarcted tissue. We conclude that acute myocardial infarction can usually be detected by MRI, given a prior knowledge of its location. However, the technique is at present likely to be of only limited value clinically in the prospective diagnosis of acute myocardial infarction, though this may improve as technology advances. Finally, signal changes following reperfusion may be visible in some cases, but not reliably so.
ASJC Scopus subject areas
- Biomedical Engineering
- Radiology Nuclear Medicine and imaging