Abstract Archives of the RSNA, 2009
SST03-05
Clinical Evaluation of Dual-energy CT for Detection of Chronic Myocardial Infarction
Scientific Papers
Presented on December 4, 2009
Presented as part of SST03: Cardiac (CT Angiography: Dual Energy)
Research and Education Foundation Support
Monvadi B. Srichai MD, Presenter: Nothing to Disclose
Hersh Chandarana MD, Abstract Co-Author: Nothing to Disclose
Irene Isabel P Lim BS, Abstract Co-Author: Nothing to Disclose
Robert Donnino MD, Abstract Co-Author: Nothing to Disclose
Christianne Leidecker PhD, Abstract Co-Author: Employee, Siemens AG
James S. Babb PhD, Abstract Co-Author: Nothing to Disclose
Jill E. Jacobs MD, Abstract Co-Author: Speaker, General Electric Company
Research grant, Seimens AG
00030490-DMT et al, Abstract Co-Author: Nothing to Disclose
Dual energy (DE) computed tomography (CT) aids tissue characterization and may increase conspicuity of myocardial infarct compared to standard cardiac CT angiography (CCTA). This study evaluated the diagnostic accuracy of different DECT reconstruction and post-processing algorithms for detection of chronic myocardial infarction.
Seventeen consecutive patients (88% male, mean age 67y) with chronic myocardial infarction had DECT acquisition as part of their clinical CCTA examination (Definition, Siemens Healthcare). First pass (n=4), early 1 min delay (n=5) or late 6-10 min delay (n=10) scans were performed with DE (140kV and 100kV) acquisition. First pass reconstructions were done at 10% phase intervals of R-R wave (0-90%) and analyzed with left ventricular 17-segment model for function; abnormal wall motion segments were classified as infarct segments. DE scans were reconstructed at the least motion diastolic phase into: 140kV, 100kV, and merged datasets, and were evaluated for possible (grade 2) or definite (grade 3) perfusion abnormalities (hypoperfusion on first pass and early delay scans and hyperenhancement on late delay scans) with the 17-segment model. Using DE software, iodine only images were also evaluated.
Grade 3 perfusion abnormalities (hypoperfused or hyperenhanced) had higher specificity but lower sensitivity for infarct detection compared to combined grade 2 and 3 abnormalities. Grade 3 hypoperfusion defects had similar diagnostic accuracy (82% vs 84%, p>0.05) with significantly higher sensitivity (53% vs 25%, p<0.05) and significantly lower specificity (98% vs 92%, p<0.001) compared to grade 3 hyperenhanced regions. Segments with grade 3 hypoperfused defects on 1 min delay scan with 100kV reconstruction had highest diagnostic accuracy of 93%, with 79% sensitivity and 97% specificity. Diagnostic accuracy with DE iodine images was similar to other reconstructions for hyperenhancement, but inferior to 100kV reconstruction for hypoperfused defects.
Early hypoperfusion abnormalities at 100kV CCTA had the best diagnostic accuracy for detection of chronic myocardial infarct regions compared to late hyperenhanced perfusion abnormalities and DE iodine images.
CCTA allows accurate detection of segmental chronic myocardial infarction.
Srichai, M,
Chandarana, H,
Lim, I,
Donnino, R,
Leidecker, C,
Babb, J,
Jacobs, J,
et al, 0,
Clinical Evaluation of Dual-energy CT for Detection of Chronic Myocardial Infarction. Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL.
http://archive.rsna.org/2009/8006477.html