Determination of glomerular filtration rate in anesthetized pigs by use of three-phase whole-kidney computed tomography and Patlak plot analysis

Abstract

Objective - To develop a whole-kidney computed tomography (CT) technique that would allow 3-point Patlak plot determination of glomular filtration rate (GFR) and assess the correlation of GFR determined via CT (CT-GFR) with GFR determined via renal plasma clearance of inulin (Inu-GFR) in pigs. Animals - 6 healthy anesthetized pigs. Procedures - Each pig underwent 3-phase whole-kidney helical CT (arterial, early, and late parenchymal phases) before and after contrast medium administration. After contrast medium administration, corrected Hounsfield unit values were determined for each kidney and the aorta. A 3-point Patlak plot for each kidney was generated, and plasma clearance per unit volume was multiplied by renal volume to obtain whole-animal CT-GFR. Correlations of mean Inu-GFR for the left and right kidneys (and combined [total] values) with the corresponding CT-GFRs were assessed via linear regression and Bland-Altman analyses. Results - Left kidney, right kidney, and total CTGFRs were good predictors of the respective lnu-GFR values (r2 = 92.3%, r2 = 85.5%, and r2 = 93.7%, respectively). For the left kidney, no significant bias between lnu-GFR and CT-GFR was detected. Right kidney and total CT-GFRs underestimated the corresponding lnu-GFRs (mean underestimation, -8.4 mL·min-1 and -12.6 mL·min-1, respectively). Conclusions and Clinical Relevance - Three-phase whole-kidney CT with Patlak plot analysis of GFR may underestimate right kidney and total Inu-GFRs in pigs. The Patlak plot generated may be sensitive to nonlinearity caused by temporal variation in GFR. Nonetheless, the 3-phase CT approach offers some practical advantages for simultaneous evaluation of renal morphology and measurement of GFR.