Search

Objective: To evaluate the feasibility of image subtraction in late iodine enhancement CT (LIE-CT) for assessment of myocardial infarction (MI)., Methods: A comprehensive cardiac CT protocol and late gadolinium enhancement MRI (LGE-MRI) was used to assess coronary artery disease in 27 patients. LIE-CT was performed after stress CT perfusion (CTP) and CT angiography. Subtraction LIE-CT was created by subtracting the mask volume of the left ventricle (LV) cavity from the original LIE-CT using CTP dataset. The %MI volume was quantified as the ratio of LIE to entire LV volume, and transmural extent (TME) of LIE was classified as 0%, 1-24%, 25-49%, 50-74% or 75-100%. These results were compared with LGE-MRI using the Spearman rank test, Bland-Altman method and chi-square test., Results: One hundred twenty-five (29%) of 432 segments were positive on LGE-MRI. Correlation coefficients for original and subtraction LIE-CT to LGE-MRI were 0.79 and 0.85 for %MI volume. Concordances of the 5-point grading scale between original and subtraction LIE-CT with LGE-MRI were 75% and 84% for TME; concordance was significantly improved using the subtraction technique (p <0.05)., Conclusion: Subtraction LIE-CT allowed more accurate assessment of MI extent than the original LIE-CT., Key Points: • Subtraction LIE-CT allows for accurate assessment of the extent of myocardial infarction. • Subtraction LIE-CT shows a close correlation with LGE-MRI in %MI volume. • Subtraction LIE-CT has significantly higher concordance with TME assessment than original LIE-CT.

We evaluated the image quality and diagnostic performance of late iodine enhancement computed tomography (LIE-CT) with knowledge-based iterative model reconstruction (IMR) for the detection of myocardial infarction (MI) in comparison with late gadolinium enhancement magnetic resonance imaging (LGE-MRI). The study investigated 35 patients who underwent a comprehensive cardiac CT protocol and LGE-MRI for the assessment of coronary artery disease. The CT protocol consisted of stress dynamic myocardial CT perfusion, coronary CT angiography (CTA) and LIE-CT using 256-slice CT. LIE-CT scans were acquired 5 min after CTA without additional contrast medium and reconstructed with filtered back projection (FBP), a hybrid iterative reconstruction (HIR), and IMR. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were assessed. Sensitivity and specificity of LIE-CT for detecting MI were assessed according to the 16-segment model. Image quality scores, and diagnostic performance were compared among LIE-CT with FBP, HIR and IMR. Among the 35 patients, 139 of 560 segments showed MI in LGE-MRI. On LIE-CT with FBP, HIR, and IMR, the median SNRs were 2.1, 2.9, and 6.1; and the median CNRs were 1.7, 2.2, and 4.7, respectively. Sensitivity and specificity were 56 and 93% for FBP, 62 and 91% for HIR, and 80 and 91% for IMR. LIE-CT with IMR showed the highest image quality and sensitivity (p < 0.05). The use of IMR enables significant improvement of image quality and diagnostic performance of LIE-CT for detecting MI in comparison with FBP and HIR.

Objectives: To evaluate the feasibility of three-dimensional (3D) maximum principal strain (MP-strain) derived from cardiac computed tomography (CT) for detecting myocardial infarction (MI)., Methods: Forty-three patients who underwent cardiac CT and magnetic resonance imaging (MRI) were retrospectively selected. Using the voxel tracking of motion coherence algorithm, the peak CT MP-strain was measured using the 16-segment model. With the trans-mural extent of late gadolinium enhancement (LGE) and the distance from MI, all segments were classified into four groups (infarcted, border, adjacent, and remote segments); infarcted and border segments were defined as MI with LGE positive. Diagnostic performance of MP-strain for detecting MI was compared with per cent systolic wall thickening (%SWT) assessed by MRI using receiver-operating characteristic curve analysis at a segment level., Results: Of 672 segments excluding16 segments influenced by artefacts, 193 were diagnosed as MI. Sensitivity and specificity of peak MP-strain to identify MI were 81 % [95 % confidence interval (95 % CI): 74-88 %] and 86 % (81-92 %) compared with %SWT: 76 % (60-95 %) and 68 % (48-84 %), respectively. The area under the curve of peak MP-strain was superior to %SWT [0.90 (0.87-0.93) vs. 0.80 (0.76-0.83), p < 0.05]., Conclusions: CT MP-strain has a potential to provide incremental value to coronary CT angiography for detecting MI., Key Points: • CT MP-strain allows for three-dimensional assessment of regional cardiac function. • CT-MP strain has high diagnostic accuracy for detecting myocardial infarction. • CT-MP strain may assist in tissue characterisation of myocardium assessed by LGE-MRI. • CT-MP strain provides incremental values to coronary CTA for detecting myocardial infarction.

Background: Adenosine triphosphate (ATP) induces relative hypoperfusion in significantly stenotic coronary arteries, but its effect on myocardial strain has not been used clinically for the detection of ischemia. The purpose of this study was to quantify ATP-stress-induced contractile impairment as altered myocardial strain in coronary artery disease (CAD) using tagged magnetic resonance (MR) and to evaluate its diagnostic capability in comparison with ATP-stress myocardial perfusion-MR., Methods and Results: Tagged MR and perfusion-MR under ATP-stress and rest conditions and late gadolinium enhancement (LGE) MR imaging were performed in 22 patients with suspected CAD. The peak absolute value of the circumferential strain (C-strain) was measured in 12 segments. Myocardial segments were categorized as non-ischemic (n=201), ischemic (n=42), or infarcted (n=21) according to the perfusion-MR and LGE MR imaging results. The absolute C-strain was significantly greater under ATP-stress (19±13%) (mean±SD) than under at-rest (15±11%, P<0.001) conditions in non-ischemic segments. Conversely, the absolute C-strain was significantly lower under ATP-stress (10±13%) than under at-rest (16±6%, P<0.01) conditions in ischemic segments. Optimal cutoff values for stress C-strain (-17.5%) could successfully differentiate segments with ischemia or infarction from non-ischemic ones with a sensitivity of 86%, a specificity of 65%, and an area under the curve of 0.79., Conclusions: C-strain analysis using tagged MR could detect ATP-stress-induced contractile impairment in ischemic myocardium.

Objective: To determine LV function at different distances from myocardial infarction (MI) by using 3-T tagged MRI and late gadolinium enhancement (LGE)., Methods: Cardiac MR images were acquired from 21 patients with previous MI. The harmonic phase (HARP) method was used to calculate radial and circumferential strain (RS, CS). The two strains were synchronised by subtracting the CS from the RS at the same time, and this was defined as the efficient strain (ES). Peak strain (P-RS, P-CS, P-ES) and time to peak strain (T-RS, T-CS, T-ES) were used as estimates of contractile function. Based on the presence of LGE, myocardium was classified into infarct, border zone, adjacent and remote areas., Results: P-RS and P-ES were significantly greater for remote than for adjacent and infarct areas. P-CS values were significantly greater for remote and border zone than for infarct areas. T-RS and T-ES were significantly shorter for remote and border zone than for infarct areas. T-CS was significantly shorter for border zone than for infarct areas., Conclusion: Contractile dysfunction demonstrated by peak strain was correlated with location at different distances from the infarct. In the border zone, contractile deformation was characterised as earlier T-RS, T-CS and T-ES and greater P-CS than in the infarct area.

A case of severe aortic stenosis with normal coronary arteriograms was associated with myocardial infarction involving the circumferential subendocardial wall of the left ventricle. The infarct was caused solely by the severe aortic stenosis and resulted from the extreme disparity between myocardial oxygen supply and demand.