Luigia D'Errico, Abstract Co-Author: Nothing to Disclose

Francesco Salituri, Abstract Co-Author: Nothing to Disclose

Marco Ciardetti, Abstract Co-Author: Nothing to Disclose

Riccardo Favilla, Abstract Co-Author: Nothing to Disclose

Mathis Schlueter, Abstract Co-Author: Nothing to Disclose

Lorenzo Faggioni MD, Presenter: Nothing to Disclose

Michele Coceani, Abstract Co-Author: Nothing to Disclose

Massimiliano Bianchi, Abstract Co-Author: Nothing to Disclose

Alessandro Mazzarisi, Abstract Co-Author: Nothing to Disclose

Giuseppe Coppini, Abstract Co-Author: Nothing to Disclose

Carlo Bartolozzi MD, Abstract Co-Author: Nothing to Disclose

Paolo Marraccini, Abstract Co-Author: Nothing to Disclose

Epicardial fat (EF) is an independent predictor of cardiac prognosis. Computed Tomography (CT) allows quantification of EF, but information about accuracy and reproducibility is lacking. The aim of the study was to evaluate EF volume (EFV) and density from CT series for coronary calcium scoring (CS, without contrast medium) and coronary angiography (CTCA), and to assess inter- and intra-observer variability.

30 patients (60±10 years, 73% male, body mass index 27.6±3.9 kg/m2) underwent CS and CTCA acquisition. Reformatted slices were reconstructed along the vertical long axis of the heart with a 2.5mm slice thickness. Pericardial boundaries were traced in three axial slices and in two MPR planes (corresponding to 4- and 2-chamber views). A dedicated application allowed an automatic selection of EF and a calculation of total, right, and left ventricular EFV (ml) and density (HU). Three physicians analyzed CS and CTCA twice, for a total of 120 evaluations for each operator. Intra-observer variability was evaluated in two physicians (one trained [O1] and one untrained [O2] in the use of the program). Inter-observer variability among all 3 operators was also determined.

Total EFV was 111±54.8 ml and 73.3±37.1 ml (p<0.001) in CS and CTCA series, respectively. The intra-observer coefficient of repeatability (CR) in CS was 12.6 ml in O1 and 28.3 ml in O2, whereas in CTCA was 9.76 ml in O1 and 17.5 ml in O2. The inter-observer CR was 29.2 ml (95% CI 27.3-30.8) and 24.3 ml (95% CI 22.6-25.6) in CS and CTCA, respectively. Right ventricular EFV in CS and CTCA was 69.7±27.4 ml and 43.7±22.2 ml, respectively; the corresponding values for left ventricular EFV were 41.7±18.6 ml and 28.2±14.7 ml. The fat density resulted slightly (but statistically significantly), lower in the CS series (-87.9±5.2 vs -85.6±4.7 HU (p< 0.04). Moreover, the range of fat density was larger in CS [(-99.56,-74.1) vs (-94,-75.82 HU)].

CT may allow the computation of EFV in clinical setting. CTCA and CS had a clinical acceptable intra- and inter-observer variability of computer-assisted measurements. EFV was lower in CTCA than in CS series, so this observation must be taken into account in clinical and follow-up studies.

Cardiac CT may allow a reliable computation of epicardial fat volume, an important predictor of prognosis in cardiac patients.

D'Errico, L, Salituri, F, Ciardetti, M, Favilla, R, Schlueter, M, Faggioni, L, Coceani, M, Bianchi, M, Mazzarisi, A, Coppini, G, Bartolozzi, C, Marraccini, P, CT-derived Epicardial Fat Volume: Influence of Scanning Protocol and Reproducibility of Measurements.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034482.html