Keywords:
Cardiac, Molecular imaging, PET-MR, Technology assessment, Diagnostic procedure, Ischaemia / Infarction
Authors:
F. Nensa1, T. D. Poeppel1, J. Schelhorn1, A. A. Mahabadi1, C. Buchbender2, R. A. Erbel1, A. Bockisch1, M. Forsting1, T. Schlosser1; 1Essen/DE, 2Düsseldorf/DE
DOI:
10.1594/ecr2013/C-0895
Methods and Materials
Eight patients with confirmed acute ST segment elevation myocardial infarction underwent CMR-PET with 18F-fluorodeoxyglucose within 2-7 days after interventional revascularization.
For comparison of infarct identification and characterization the left ventricle was mapped to the 17-segment model recommended by the American Heart Association. Tracer accumulation in each myocardial segment was compared to regional wall motion abnormalities and to signal intensity in late gadolinium-enhanced (LGE) images.
The size of the infarct zone was measured with freehandedly drawn regions on the LGE MR images and PET images and was expressed as the percentage of the entire left ventricular myocardium. Also the contrast ratio (CR) between infarcted and non-infarcted myocardium and between non-infarcted myocardium an the left ventricular cavity was assessed in LGE images as CRinf-noninf = |(SIinf - SInoninf)/(SIinf + SInoninf)| and CRnoninf-LVC = |(SInoninf - SILVC)/(SInoninf + SILVC)|,
respectively.
Categorical inter-method agreement between PET and cine imaging or LGE imaging,
respectively,
was calculated using Cohen’s kappa.
Significance testing was performed using paired and unpaired t-Tests as appropriate.
All statistical calculations were performed using the R-software environment for statistical computing.