New thresholds in semi-quantitative [18F]FDG PET/CT are needed to assess large vessel vasculitis with long-axial field-of-view scanners
Luisa Knappe, Carola Bregenzer, Nasir Gözlügöl, Clemens Mingels, Ian Alberts, Axel Rominger, Federico Caobelli
Abstract
Abstract Aim [ 18 F]FDG PET/CT proved accurate in the diagnostic work-up of large vessel vasculitis (LVV). While a visual interpretation is currently considered adequate, several attempts have been made to integrate it with a semiquantitative evaluation. In this regard, there is the need to validate current or new thresholds for the semiquantitative parameters on long-axial field of view (LAFOV) scanners. Methods We retrospectively evaluated 100 patients (50 with LVV and 50 controls) who underwent [ 18 F]FDG LAFOV PET/CT. Semiquantitative parameters (SUVmax and SUVmean) were calculated for large vessels in 3 districts (supra-aortic [SA], thoracic aorta [TA], and infra-aortic [IA]). Values were also normalized to liver activity (SUV max /L-SUV max , and SUV max /L-SUV mean ). Results Of the 50 patients diagnosed with LVV, SA vessels were affected in 38 (76%), TA in 42 (84%) and IA vessels in 26 (52%). To-liver normalized values had higher diagnostic accuracy than non-normalized values (AUC always ≥ 0.90 vs. 0.74–0.89). For the SA vessels, best thresholds were 0.66 for SUV max /L-SUV max and 0.88 for SUV max /L-SUV mean ; for the TA, 1.0 for SUV max /L-SUV max and 1.30 for SUV max /L-SUV mean ; finally, for IA vessels, the best threshold was 0.83 for SUV max /L-SUV max and 1.11 for SUV max /L-SUV mean . Conclusion LAFOV [ 18 F]FDG-PET/CT is accurate in the diagnostic workup of LVV, but different threshold in semi-quantitative parameters than reported in literature for standard scanners should be considered.