Abstract
3143
Introduction: In this study we investigate the performance of total-body PET/CT with ultra-low-activity 18F-FDG from both dynamic and static imaging aspects, compared with those of total-body PET/CT with low-activity 18F-FDG.
Methods: Thirty patients who underwent 75-min dynamic total-body PET/CT on a uEXPLORER scanner with ultra-low-activity (0.37 MBq/kg) and low-activity (1.85 MBq/kg) 18F-FDG after 24 hours interval, were prospectively studied. Among the enrolled patients, there were 24 patients confirmed with malignance by surgery or puncture. Kinetic metrics (k1–k3, and Ki), kinetic constants, corresponding coefficients of variance (CoVs), and image contrast were generated and compared between ultra-low-activity and low-activity groups. Ultra-low-activity PET raw data were acquired within 60-75 minutes and reconstructed using data from the first 65 min, 68 min, 70 min and 75 min (as G5, G8, G10, G15). Low-activity PET raw data were reconstructed using data from the first 61 min, 62 min and 63 min (as G1’, G2’, G3’). In static imaging, G5 and G1’, G8 and G2’, G10 and G2’, G15 and G3’ were selected for inter-group comparisons. Static image qualities were assessed and compared between activity groups, by subjectively assessing the 5-point Likert scales (5-PS) and objectively calculating the mean and maximum of standardized uptake value (SUVmean and SUVmax) and the signal-to-noise ratio (SNR) of liver, blood pool and lesions. Cohen’s kappa test was performed to investigate the intra-reader and inter-reader agreement.
Results: All constant rates, corresponding CoVs, and image contrast in different organs were comparable with none significant differences between ultra-low-activity and low-activity groups. None significant difference of kinetic metrics (k1–k3 and Ki) of normal organs and lesions were identified between both groups. In ultra-low-activity static PET imaging, image quality scores in G5 (2.97 ± 0.18), G8 (3.01 ± 0.13), G10 (3.21 ± 0.42) and G15 (3.97 ± 0.18) were higher than those in the low-activity group G1’ (2.92 ± 0.27), G2’ (3.05 ± 0.22), G3’ (3.88 ± 0.32), with no difference (all p > 0.05). Qualitative analysis of image quality showed good intra- and inter-reader agreements (all kappa > 0.7). None significant difference of SUVmean andSUVmax of liver, blood pool and lesions were identified among all inter-group comparisons (all p > 0.05). Only three lesions were missed on G1’, and there were no missed lesions in other group.
Conclusions: Significant reduction in administered 18F-FDG tracer activities with Ultra-low-activity is feasible in total-body PET/CT.