Abstract 678: Delayed Enhancement Dual-Energy CT versus Cardiac MRI Viability Imaging for Quantifying Myocardial Infarct Size
PURPOSE Previous studies have shown that delayed enhancement CT can detect myocardial infarction, although CT systematically underestimated true infarct size as compared to cardiac MRI (cMRI) viability imaging. We aimed to evaluate, whether delayed enhancement dual-energy CT (DECT) enables more accurate quantification of myocardial infarction.
METHOD AND MATERIALS 26 patients with suspected ischemic heart disease underwent delayed enhancement DECT (Somatom Definition, Siemens) with 1.5mm collimation, 140kV/80kV, and 120 –360mAs. Scans were acquired six minutes after injection of 120mL of 370mgI/mL iopromide (Ultravist, Bayer). Data were reconstructed and analyzed using the Perfusion Blood Volume application (Siemens). All patients also underwent 1.5T (Avanto, Siemens) cMRI 15 minutes after the injection of gadopentetate dimeglumine (Magnevist, Bayer) using phase sensitive inversion recovery sequences. cMRI studies were analyzed using the Argus software (Siemens). Two independent, blinded observers performed segmental evaluation of all studies for myocardial iodine/gadolinium uptake compatible with infarct and performed planimetric measurements of infarct size. The percentage of infarct area per section (PIS) along with the ratio of infarct to the total area of myocardium (infarction fraction) were calculated.
RESULTS Delayed enhancement cMRI showed gadolinium uptake indicating infarction in 72/442 myocardial segments. Analysis using Receiver Operating Characteristics showed a mean Area Under the Curve of the two observers for detection of myocardial infarct at delayed enhancement DECT of 0.92 on a per-segment level and of 0.91 on a per-patient level. Interobserver agreement was excellent (κ=0.85). Linear regression analysis revealed a close, linear correlation between cMRI and DECT for determining the percentage of infarct area per section (y=1.2x+0.6, R=0.9), as well as the infarction fraction (y=1.1x+2.3, R=0.88).
CONCLUSION Delayed enhancement DECT has high accuracy for the detection of myocardial infarction and correlates well with delayed enhancement cMRI for quantifying infarct size. DECT appears less prone to systematic underestimation of infarct size, which has been reported as a limitation of single-energy CT.