Original Article on Translational Imaging in Cancer Patient Care
Quantitative assessment of liver fibrosis and its stage in a rabbit model by using intravoxel incoherent motion diffusion-weighted imaging at a 3T magnetic resonance system
Abstract
Background: To explore whether and how intravoxel incoherent motion (IVIM) diffusion-weighted imaging (DWI) at a 3T magnetic resonance scanner could monitor and stage liver fibrosis in a rabbit model.
Methods: Fifty-six New Zealand white rabbits were given carbon tetrachloride to model liver fibrosis, and eight treated with normal saline served as control subjects. IVIM-DWI with eight b values of 0, 10, 20, 50, 100, 200, 800, 1,000 s/mm2 was performed on the 0, 6th, 8th, 10th and 12th weekends after modeling this disease. The stages of liver fibrosis (stages F0 to F4) were identified based on METAVIR classification system. The IVIM derived parameters (D*, pseudo-diffusion; D, pure molecular diffusion; and f, perfusion fraction) were quantitatively measured, and statistical analysis were performed for detecting and staging liver fibrosis.
Results: Significant difference was found in D between normal and fibrotic liver (P<0.05). D could distinguish stages between F0 and F3 or F4 (P<0.05); D*, f could not discriminate between normal and fibrotic liver, nor could discriminate any stages of liver fibrosis (all P>0.05). A negative correlation was found between fibrosis stage and D (r=−0.605, P<0.05). According to receiver operating characteristic curve, D could differentiate between stage F0 and F3–4 with an area under the corresponding curve of 0.937.
Conclusions: The IVIM derived parameter D can quantitatively monitor the progression of liver fibrosis.
Methods: Fifty-six New Zealand white rabbits were given carbon tetrachloride to model liver fibrosis, and eight treated with normal saline served as control subjects. IVIM-DWI with eight b values of 0, 10, 20, 50, 100, 200, 800, 1,000 s/mm2 was performed on the 0, 6th, 8th, 10th and 12th weekends after modeling this disease. The stages of liver fibrosis (stages F0 to F4) were identified based on METAVIR classification system. The IVIM derived parameters (D*, pseudo-diffusion; D, pure molecular diffusion; and f, perfusion fraction) were quantitatively measured, and statistical analysis were performed for detecting and staging liver fibrosis.
Results: Significant difference was found in D between normal and fibrotic liver (P<0.05). D could distinguish stages between F0 and F3 or F4 (P<0.05); D*, f could not discriminate between normal and fibrotic liver, nor could discriminate any stages of liver fibrosis (all P>0.05). A negative correlation was found between fibrosis stage and D (r=−0.605, P<0.05). According to receiver operating characteristic curve, D could differentiate between stage F0 and F3–4 with an area under the corresponding curve of 0.937.
Conclusions: The IVIM derived parameter D can quantitatively monitor the progression of liver fibrosis.
