RT Journal Article
SR Electronic
T1 Respiration-averaged CT versus standard CT attenuation maps for correction of the 18F-NaF uptake in hybrid PET/CT
JF Journal of Nuclear Medicine
JO J Nucl Med
FD Society of Nuclear Medicine
SP 662
OP 662
VO 61
IS supplement 1
A1 Tzolos, Evangelos
A1 Lassen, Martin Lyngby
A1 Pan, Tinsu
A1 Kwiecinski, Jacek
A1 Sebastien, Cadet
A1 Dey, Damini
A1 Berman, Daniel
A1 Slomka, Piotr
YR 2020
UL http://jnm.snmjournals.org/content/61/supplement_1/662.abstract
AB 662Introduction: Absolute quantification is a necessity in the classification of coronary plaques, using 18F-NaF. Current protocols employ standard computed tomography attenuation correction (CTAC) maps to correct for photon attenuation in the body. However, fast transmission CT (cine-CT) acquisitions allow the creation of respiratory averaged CTAC maps (RACTAC), which takes respiratory translations of the thorax into account. In this study we aim evaluate the quantitative impact of using an RACTAC map against the standard CTAC maps. Methods: This study comprised of 10 patients examined for coronary artery disease in a hybrid PET/CT system, with 18F-NaF (200-250 MBq). All patients had a 4D cine-CT acquisition followed by a standard AC map. The 4D cine-CT data were acquired at the following settings: 120 kV, 10 mA, cine duration of 5 sec and 0.8 sec/rot and obtaining seven CT images over 5 sec. The standard AC map were acquired at the following settings: 100kV and 40mAs/rot. From the CINE CT acquisition, we averaged the cine CT images to create the RACTAC. Both the CTAC and the RACTAC maps were used for attenuation correction using a vendor provided reconstruction toolbox (4 cardiac gates, 256x256 image matrix, 2i21subsets, ToF and PSF correction, 5mm in-plane Gaussian filter). Following image-reconstructions with the two attenuation correction maps obtained for each patient, all 10 patients were evaluated in FusionQuant, a dedicated software (FusionQuant, Cedars-Sinai, Los Angeles) to evaluate coronary plaques using hybrid imaging modalities. All datasets were corrected for cardiac motion using a PET-PET image co-registration employing a diffeomorphic algorithm. Following motion correction, we co-registered the PET activity with the corresponding anatomical CT structures and all atherosclerotic calcified plaques using a dedicate software. We drew regions of interest around these areas and measured the maximum standardized uptake values (SUVmax). To characterize 18F-NaF coronary uptake in the context of background activity we divided the maximum standard uptake values (SUVmax) by the blood pool activity (mean standard uptake values, SUVmean) measured in the center of the left atrium by delimiting spherical volumes of interest (radius=10mm) and produced the maximum target to background ratios (TBRmax). We compared the results using box-plot and Bland-Altman plot analysis. All statistical analysis was performed in SPSS software (version 24, SPSS, Inc., Chicago, Illinois). Results: In the 10 patients, we recognized 20 coronary lesions for all 20 PET reconstructions (CTAC and RACTAC attenuation corrected datasets). There was good agreement between the RACTAC and the standard CTAC. The median [IQR] TBRMAX for the RACTAC reconstructions was 1.24[1.05-1.45], while the median TBRMAX for the standard images was 1.27[1.1-1.43]; p=0.82. There was no significant misalignment between measurements [Figure 1]. The coefficient of variation was 4.02%. There was minimal bias (1.1%) and narrow limits of agreements between the two different reconstructions (-10.6% to12.9%), as shown in the Bland -Altman plot (Figure 2). Conclusions: Applying respiration-averaged CT attenuation correction (RACTAC) maps does not improve the quantification of the coronary lesions as read in fusion PET/CT images. Scrapping cine-CT will lead to reduced total radiation dose and time of procedure without affecting lesion quantification analysis.