Modeling the excretion of FDG in human kidneys using dynamic PET

In order to understand the excretion function of kidneys, dynamic scan was performed using the positron emission tomography (PET), the process of FDG's (2-[(18)F]fluoro-2-deoxy-D-glucose) excretion was detected, and the kidney model was established. The model in this study consisted of two parts: the fore part of the model described the transportation of FDG from plasma and the accumulation of FDG in kidney, and the latter part of the model described the transportation of FDG from kidney to ureter and then to bladder. Since there was a time delay between the fore part and the later part, which occurred when FDG was filtered into urine and accumulated in pelvis temporarily, a new parameter, delay constant t(0), was introduced in the model. Twelve healthy adult volunteers took part in the dynamic FDG-PET experiment. Ten subjects received dynamic scan on kidneys, and the data extracted from the PET scans were used for parameter estimation and model analysis. The other two subjects received dynamic scan on bladder in order to confirm the time delay constant. The output of the model fit well with the original curve, and the model built in this study could not only describe the excretion process of FDG, but also be used to quantitatively estimate urinary excretion of FDG and plasma clearance. Moreover, the model kept good accordance with physiological characteristics.