Quantification and parametric imaging of renal cortical blood flow in vivo based on Patlak graphical analysis

Abstract

Patlak graphical analysis was applied to quantify renal cortical blood flow with N-13 ammonia and dynamic positron emission tomography. Measurements were made in a swine model of kidney transplantation with a wide range of normal and abnormal renal blood flows (N = 57 studies) and in 20 healthy human volunteers (N = 45 studies). Estimates of renal cortical blood flow by the Patlak method were compared to those from a two-compartment model for N-13 ammonia. In addition, estimates of renal cortical blood flow by the N-13 ammonia PET approach were compared in 10 normal human volunteers to estimates by the metabolically inert, freely diffusible O-15 water and a one-compartment model. Patlak graphical analysis estimates of renal cortical blood flow correlated linearly with the standard two-compartment model in pigs (y = -0.05 + 1.01x, r = 0.99) and in humans (y = 0.57 + 0.88x, r = 0.93). Estimates of renal cortical blood flow by O-15 water in human volunteers were also linearly correlated with those by N-13 ammonia and the Patlak graphical analysis (y = 0.71 + 0.84x, r = 0.86). Renal cortical blood flow estimates were highly reproducible both with N-13 ammonia and O-15 water measurements in humans. It is concluded that the Patlak graphical analysis with N-13 ammonia dynamic positron emission tomograpic imaging renders accurate and reproducible estimates of renal cortical blood flow. Moreover, the graphical analysis approach is 1,000 times faster than the standard model fitting approach and suitable for generating parametric images of renal blood flow in the clinical setting.