Assessment of left and right ventricular functional parameters using dynamic dual-tracer [13N]NH3 and [18F]FDG PET/MRI

Sazan Rasul, Dietrich Beitzke, Tim Wollenweber, Ivo Rausch, Martin Lyngby Lassen, Marie Elisabeth Stelzmüller, Markus Mitterhauser, Verena Pichler, Thomas Beyer, Christian Loewe, Marcus Hacker (Corresponding author)

    Publications: Contribution to journalArticlePeer Reviewed

    Abstract

    BACKGROUND: Cardiac positron emission tomography/magnetic resonance imaging (PET/MRI) can assess various cardiovascular diseases. In this study, we intra-individually compared right (RV) and left ventricular (LV) parameters obtained from dual-tracer PET/MRI scan. METHODS: In 22 patients with coronary heart disease (69 ± 9 years) dynamic [13N]NH3 (NH3) and [18F]FDG (FDG) PET scans were acquired. The first 2 minutes were used to calculate LV and RV first-pass ejection fraction (FPEF). Additionally, LV end-systolic (LVESV) and end-diastolic (LVEDV) volume and ejection fraction (LVEF) were calculated from the early (EP) and late-myocardial phases (LP). MRI served as a reference. RESULTS: RVFPEF and LVFPEF from FDG and NH3 as well as RVEF and LVEF from MRI were (28 ± 11%, 32 ± 15%), (32 ± 11%, 41 ± 14%) and (42 ± 16%, 45 ± 19%), respectively. LVESV, LVEDV and LVEF from EP FDG and NH3 in 8 and 16 gates were [71 (15 to 213 mL), 98 (16 to 241 mL), 32 ± 17%] and [50 (17 to 206 mL), 93 (13 to 219 mL), 36 ± 17%] as well as [60 (19 to 360 mL), 109 (56 to 384 mL), 41 ± 22%] and [54 (16 to 371 mL), 116 (57 to 431 mL), 46 ± 24%], respectively. Moreover, LVESV, LVEDV and LVEF acquired from LP FDG and NH3 were (85 ± 63 mL, 138 ± 63 mL, 47 ± 19%) and (79 ± 56 mL, 137 ± 63 mL, 47 ± 20%), respectively. The LVESV, LVEDV from MRI were 93 ± 66 mL and 153 ± 71 mL, respectively. Significant correlations were observed for RVFPEF and LVFPEF between FDG and MRI (R = .51, P = .01; R = .64, P = .001), respectively. LVESV, LVEDV, and LVEF revealed moderate to strong correlations to MRI when they acquired from EP FDG and NH3 in 16 gates (all R > .7, P = .000). Similarly, all LV parameters from LP FDG and NH3 correlated good to strongly positive with MRI (all R > .7, and P < .001), except EDV from NH3 weakly correlated to EDV of MRI (R = .54, P < .05). Generally, Bland-Altman plots showed good agreements between PET and MRI. CONCLUSIONS: Deriving LV and RV functional values from various phases of dynamic NH3 and FDG PET is feasible. These results could open a new perspective for further clinical applications of the PET examinations.
    Original languageEnglish
    Pages (from-to)1003-1017
    Number of pages15
    JournalJournal of Nuclear Cardiology
    Volume29
    Issue number3
    DOIs
    Publication statusPublished - Jun 2022

    Austrian Fields of Science 2012

    • 302054 Nuclear medicine

    Keywords

    • 13N-NH3
    • 18F-FDG
    • Cardiac PET/MRI
    • coronary artery disease
    • dynamic PET
    • VOLUMES
    • MRI
    • QUANTIFICATION
    • Cardiac PET
    • MYOCARDIAL BLOOD-FLOW
    • F-18-FDG PET
    • POSITRON-EMISSION-TOMOGRAPHY
    • CLINICAL-APPLICATIONS
    • EJECTION FRACTION
    • F-18-FDG
    • N-13-NH3
    • F-FDG
    • N-NH3

    Cite this