RT Journal Article
SR Electronic
T1 Quantification of Myocardial Blood Flow Using High-Resolution, High-Sensitivity and 214-ps SiPM<sup>82</sup>Rb-PET/CT: A comparison of three software packages.
JF Journal of Nuclear Medicine
JO J Nucl Med
FD Society of Nuclear Medicine
SP 166
OP 166
VO 60
IS supplement 1
A1 Mario Jreige
A1 Christel Kamani
A1 Gilles Allenbach
A1 Martin Pappon
A1 Patick Genoud
A1 Silvano Gnesin
A1 Marie Nicod-Lalonde
A1 Niklaus Schaefer
A1 Robert DeKemp
A1 John Prior
YR 2019
UL http://jnm.snmjournals.org/content/60/supplement_1/166.abstract
AB 166Aim: Myocardial blood flow (MBF) quantification is a promising noninvasive tool for evaluating CAD, assisting in diagnosis, treatment, and risk stratification. Several analysis software packages are validated for MBF quantification.We compared results of MBF obtained using three software packages used in clinical routine with this latest SiPM generation scanner. Methods: In total, 111 patients (70M/41F) underwent dynamic 82Rb-PET at rest and during adenosine stress on a latest generation SiPM PET/CT (Siemens Biograph Vision 600). Data were processed with: (1) Corridor4DM, (2) FlowQuant and (3) Syngo.PET MBF. MBF and myocardial flow reserve (MFR) polar maps were computed. Comparisons used Pearson’s correlation ρ (measuring precision), Bland-Altman limit-of-agreement and Lin’s concordance correlation ρc=ρ·Cb(Cbmeasuring systematic bias). Results: Lin’s concordance and Pearson’s correlation values were similar, suggesting no systematic bias between Corridor4DM and FlowQuant with an excellent precision ρ for rest and stress MBF (ρ=0.717, ρc=0.714, Cb=0.996 and ρ=0.908, ρc=0.845, Cb=0.930, respectively) and for MFR (ρ=0.803, ρc=0.726, Cb=0.904). A good precision was measured between Corridor4DM and Syngo.PET (MBF: ρ=0.579, ρc=0.419, Cb=0.724 for rest and ρ=0.923, ρc=0.892, Cb=0.967 for stress; MFR: ρ=0.531, ρc=0.465, Cb=0.874) and between FlowQuant and Syngo.PET (MBF: ρ=0.676, ρc=0.466, Cb=0.689 for rest and ρ=0.868, ρc=0.851, Cb=0.980 for stress; MFR: ρ=0.624, ρc=0.415, Cb=0.665). Although no mean bias was observed on Bland-Altman plots, Syngo.PET provided slightly lower MFR values than FlowQuant and Corridor4DMdue to higher rest flow values. Conclusions: Concordance between software packages was excellent for MBF and MFR, despite slightly lower values by Syngo.PET at higher MFR values. Therefore, comparison of outcomes from the three software packages assessing quantification of MBF can be applied in clinical practice.