Purpose:To demonstrate the feasibility of diffusion tensor imaging (DTI) of kidneys with respiratory triggering, and determine the optimal imaging parameters for fraction anisotropy (FA) maps.
Materials and Methods:DTI of kidneys from 16 healthy volunteers was performed using a 1.5T scanner. Five different sequences with different parameters including respiration-triggered acquisition or multiple breath-holding, slice thicknesses of 3 or 5 mm, and different numbers of signal averaging and b values were compared. FA and apparent diffusion coefficients (ADCs) of the cortex and medulla were measured. Measurement error within the same and repeated examination was examined using within-individual standard deviation (Sw).
Results:FAs of the renal cortex were lower than the medulla (mean value of a sequence ranging 0.148 -0.224, 0.433-0.476) and the ADCs of the cortex were higher than the medulla (2.26 -2.69 ϫ 10 Ϫ3 mm 2 /s, 1.77-2.19 ϫ 10 Ϫ3 mm 2 /s) in all sequences (P Ͻ 0.001). The renal cortexmedulla difference was the largest, with respiratory triggering including a 3-mm slice thickness, three signal averages,and a b-value ϭ 0, 200, or 400 s/mm 2 (P Ͻ 0.001). Sw tended to be smaller in the sequence with a b-value of 400 s/mm 2 . Conclusion: DTI of kidneys with respiratory triggering is feasible with excellent cortex-medulla differentiation.