Purpose. To evaluate the capabilities of the noncontrast cardiac magnetic resonance tomography (MRI) method (T1-mapping) in comparison with delayed contrast MRI in the detection of acute myocardial infarction with ST-segment elevation and determination of viable myocardium.

Materials and Methods. 40 patients with revascularized acute myocardial infarction with ST-segment elevation underwent cardiac time-delayed contrast-enhanced MRI no later than 7 days from the onset of the disease on a Siemens Magnetom Aera, 1.5 Tesla MR tomograph. Cardiac MRI protocol included cine mode, T2-weighted imaging, delayed contrast, T1, T2, and T2* mapping. Data were analyzed by semi-quantitative method on syngo.via workstation (Siemens). Location of lesion was assessed according to 17-segment scheme of left ventricular, 4 groups – 0-25%, 26-50%, 51-75% and 76-100% (segments with lesion over 50% of left ventricular wall thickness were considered as nonviable myocardium areas). The severity of myocardial lesion was calculated by calculating the fibrosis/necrosis index.

Results and discussion. We analyzed 612 segments on MR imaging scans with delayed contrast and on T1 maps. Myocardial lesions were detected in 191 segments (31.2% of the total number of segments) on T1 maps, while gadolinium accumulation was seen on MR imaging with delayed contrast in 170 segments (27.6%). In 43 segments (7.1%), the detection of myocardial lesions on T1 charts was false. And in 22 segments (3.5%) gadolinium accumulation was not reflected on T1 maps. Thus, T1 mapping detected acute myocardial infarction with a sensitivity of 86.5% (95% CI: 80.4-91.2) and specificity of 90.3% (95% CI: 87.1-92.9) compared with delayed contrast-enhanced MRI. The diagnostic accuracy of native T1 maps was 89.4%.

The median relaxation time (T1) for myocardial areas corresponding to gadolinium accumulation zones was 1211.5 ms [1186.3;1249.6] and for unaffected myocardium was 1073.9 ms [1049.5,3;1094.6]. Detection of acute myocardial infarction on native T1 maps is possible at T1 equal to or greater than 1126.55 ms with sensitivity and specificity of 81.8% and 78.5%, respectively (AUC=0.826(95% CI:0.785-0.867), p < 0.0001).  T1 values for subendocardial and transmural lesions were not significantly different.

Fibrosis index values on T1 maps correlated significantly with fibrosis index on delayed con

trast (R=0.67, p<0.001). The strongest correlation between the fibrosis indices was seen for myocardial infarction of anterior localization.

Conclusions. 1. Native T1 imaging detects acute myocardial infarction with high sensitivity and specificity and can be used in the diagnosis of acute myocardial infarction. 2. The possibility of assessing myocardial viability with native T1 imaging requires further investigation.