Retrospective assessment of at-risk myocardium in reperfused acute myocardial infarction patients using contrast‐enhanced balanced steady‐state free‐precession cardiovascular magnetic resonance at 3T with SPECT validation

Figures

• 

  Fig. 1. Breakdown of Diagnostic and Non-diagnostic Cases. Diagnostic refers to cases where all images were scored as ‘good’ or ‘adequate’. Non-diagnostic refers to cases where at least one imaging slice was scored as non-diagnostic and had to be excluded from per-patient analysis. Criteria for ‘good’, ‘adequate’ and ‘non-diagnostic’ scores are defined in the text

• 

  Fig. 2. Representative Images from non-ST elevation myocardial infarction (NSTEMI) Patients with left anterior descending (LAD), left circumflex (LCx) and right coronary artery (RCA) myocardial infarctions. X-ray Digital subtraction angiography (DSA) images acquired before and after PCI, along with corresponding short-axis myocardial T2-STIR and contrast enhanced (CE) steady state free precession (SSFP) showing at-risk regions and infarct territories (late gadolinium enhancement (LGE)) acquired at 3T are shown. On X-ray DSA images, the red arrows refer to occlusive or substantially narrowed vessels, and the red box with dotted line demonstrates the re-establishment of blood flow following recanalization of the corresponding vessel. On CMR images, epicardial and endocardial borders are delineated by green and red borders, respectively. The left columns shows the raw images and the right columns shows the same images processed using semi-automatic threshold-based signal detection. The edematous myocardium delineated by the T2-STIR and CE-SSFP images are shown as a region enclosed by the pale blue areas and irreversibly damaged myocardium is identified on LGE images as yellow areas

• 

  Fig. 3. Representative Images from STEMI Patients with LAD, LCx and RCA Myocardial Infarctions. X-ray Digital subtraction angiography (X-ray DSA) images acquired before and after PCI, along with corresponding short-axis myocardial T2-STIR and CE-SSFP showing at-risk regions and infarct territories (LGE) acquired at 3T are shown. On X-ray DSA images, the red arrows refer to occlusive or substantially narrowed vessels, and the red box with dotted line demonstrates the re-establishment of blood flow following recanalization of the corresponding vessel. On CMR images, epicardial and endocardial borders are delineated by green and red borders, respectively. The left columns shows the raw images and the right columns shows the same images processed using semi-automatic threshold-based signal detection. The edematous myocardium delineated by the T2-STIR and CE-SSFP images are shown as a region enclosed by the pale blue areas and irreversibly damaged myocardium is identified on LGE images as yellow areas

• 

  Fig. 4. Image Quality: T2-STIR versus CE-SSFP. Diagnostic scores of T2-STIR and CE-SSFP are shown (score are as follows: ‘non-diagnostic’ is scored 1; ‘adequate’ scored 2; and ‘good’ scored 3). In total, 365 matched imaging slices from T2-STIR and CE-SSFP had the highest image quality scores and were used for head-to-head comparison

• 

  Fig. 5. Comparison of T2-STIR and CE-SSFP AAR in Per-Slice Analysis. Linear regression between area at risk (AAR) determined using diagnostic quality T2-STIR and CE-SSFP images (blue line showing line of best fit; red line represents line of identity) for both STEMI and NSTEMI patients are shown in (A) and the corresponding Bland-Altman analysis is shown in (B). Similar regression and Bland-Altman plots are shown for STEMI ((C) and (D)) and NSTEMI (E) and (F)). In all cases, the AAR is expressed as area of hyperintensity in the imaging slice

• 

  Fig. 6. SPECT Validation in Acute MI Patient with LAD, LCx and RCA Myocardial Infarctions. The AAR determined using SPECT, CE-SSFP, T2-STIR, as well as infarct size determined using LGE images are shown from representative LAD, LCx and RCA myocardial infarctions. Yellow arrows demarcate the at-risk myocardium and the white arrows point to the infarct zone

• 

  Fig. 7. AAR from SPECT vs. AAR from T2-STIR and CE-SSFP. a There was good agreement between the CE-SSFP (mean 30 ± 7 %LV) and SPECT (mean 28 ± 7 %LV) with an R2 of 0.86; (p < 0.001) with a low bias on Bland Altman analysis (mean ± 2SD: bias − 1.3 ± 7.8 %LV, p = 0.39) as seen on b. c, d T2-STIR (mean 31 ± 11 %) vs. SPECT on per patient-based analysis showed good correlation (R2 of 0.81; p < 0.001) and a mean difference of 0.5 ± 11.1 %LV; p = 0.81). e, f A strong agreement between CE-SSFP and T2-STIR was observed (R2 = 0.84; p < 0.001) with a low bias (− 2.1 ± 10.3 %LV; p = 0.31)