Feasibility of 18F-FDG dose reductions in whole body hybrid TOF PET/MR

Objectives: The study aims to investigate the feasibility of reduced tracer dose in ¹⁸F-FDG whole body integrated PET/MR imaging and its effects on the semi-quantification analysis and lesion detection.

Methods: 20 patients (41.85±22.42 y; BMI 20.40±4.20 kg/m²) with malignant tumors were performed 6min/bed whole body PET/MR (GE SIGNA) imaging with respiratory gating at 44.60±2.28min after injected standard dose (3.70MBq/kg). The imaging of reduced doses was simulated by reconstruction with 100%, 75%, 50% and 25% of the original data, and the abilities of lesion detection were compared among simulated images of different doses. The signal-to-noise ratio (SNR), SUVmax and SUVmean of liver background in each simulated dose were analyzed by paired t test while the SUVmax, SUVmean, metabolic tumor volume (MTV), total lesion glycolysis (TLG) of the lesions and the image quality score by visualization were analyzed with Wilcoxon test.

Results: The image SNR decreased significantly (t=5.040, 9.638, 11.166, P < 0.01) and the SUVmax of liver background increased (t=3.466, 4.974, 6.964, P < 0.05) with reduced dose. No significant difference was found both in the SUVmean(t=0.567, 0.442, 0.269, P>0.05) of liver background and lesions at each dose, as well as in the SUVmax(Z=-0.673, -0.797, P>0.05), SUVmean(Z=-1.924, -1.127, P>0.05), MTV(Z=-0.43, -1.080, P>0.05) and TLG(Z=-0.615, -1.958,P>0.05) of lesions at 100%, 75%, and 50% simulated dose Yet at 25% simulated dose, the SUVmax was significant higher than that of 100% with 6% deviation (Z=-2.141,P < 0.05), and the MTV and TLG of lesions were much lower with 24%, 18% deviation respectively (Z=-2.107, -2.550, P < 0.05). There was no significant difference for the image quality score between 100% dose and 75% simulated dose (Z=-0.703, P > 0.05), but the image quality scores for both doses were significant higher than that of 50% and 25%( Z=-2.384, -3.746,P < 0.05). The 283 PET-positive lesions have been detected in images from both 75% and 50% simulated doses, while in 25% simulated dose image, 12 lesions were missed, and 8 false-positive lesions were detected.

Conclusions: A 1.85Mbq/kg dose can be feasible in ¹⁸F-FDG PET/MR 6min/bed whole body examination. The detection rate and semi-quantitative analysis results will not be affected. It’s not suggested to do imaging with too low dose in case of potential deviation of semi-quantification analysis and lesion detection, and prolonged scan duration should be applied when necessary.Male, 32 years old, right lung cancer with extensive lymph node and bone metastasis. The signal-to-noise ratio (SNR) of simulated dose images reconstructed with 100% (a, e,i), 75% (b, f, j), 50% (c, g, k), 25% (d, h, l) decreased in turn. MIP images showed that the lesions in 100% (a) dose simulated images could be clearly displayed in 75% (b) and 50% (c) simulated dose images. (red arrow) False positive lesion appeared in 25% simulated dose images. (blue arrow) The right acetabulum lesions could be clearly displayed on100%, 75% and 50% dose simulated images, while 25% of dose simulated images were missed the lesion.

• Download figure
• Open in new tab
• Download powerpoint