Dynamic rubidium‐82 PET/CT as a novel tool for quantifying hemodynamic differences in renal blood flow using a one‐tissue compartment model

Purpose Assessing renal perfusion in‐vivo is challenging and quantitative information regarding renal hemodynamics is hardly incorporated in medical decision‐making while abnormal renal hemodynamics might play a crucial role in the onset and progression of renal disease. Combining physiological stimuli with rubidium‐82 positron emission tomography/computed tomography (82Rb PET/CT) offers opportunities to test the kidney perfusion under various conditions. The aim of this study is: (1) to investigate the application of a one‐tissue compartment model for measuring renal hemodynamics with dynamic 82Rb PET/CT imaging, and (2) to evaluate whether dynamic PET/CT is sensitive to detect differences in renal hemodynamics in stress conditions compared to resting state. Methods A one‐tissue compartment model for the kidney was applied to cardiac 82Rb PET/CT scans that were obtained for ischemia detection as part of clinical care. Retrospective data, collected from 17 patients undergoing dynamic myocardial 82Rb PET/CT imaging in rest, were used to evaluate various CT‐based volumes of interest (VOIs) of the kidney. Subsequently, retrospective data, collected from 10 patients (five impaired kidney functions and five controls) undergoing dynamic myocardial 82Rb PET/CT imaging, were used to evaluate image‐derived input functions (IDIFs), PET‐based VOIs of the kidney, extraction fractions, and whether dynamic 82Rb PET/CT can measure renal hemodynamics differences using the renal blood flow (RBF) values in rest and after exposure to adenosine pharmacological stress. Results The delivery rate (K1) values showed no significant (p = 0.14) difference between the mean standard deviation (SD) K1 values using one CT‐based VOI and the use of two, three, and four CT‐based VOIs, respectively 2.01(0.32), 1.90(0.40), 1.93(0.39), and 1.94(0.40) mL/min/mL. The ratio between RBF in rest and RBF in pharmacological stress for the controls were overall significantly lower compared to the impaired kidney function group for both PET‐based delineation methods (region growing and iso‐contouring), with the smallest median interquartile range (IQR) of 0.40(0.28–0.66) and 0.96(0.62–1.15), respectively (p