Purpose: To determine the diagnostic performance of liver apparent diffusion coefficient (ADC) measured with conventional diffusion-weighted imaging (CDI) and diffusion tensor imaging (DTI) for the diagnosis of liver fibrosis and inflammation.
Materials and methods: Breathhold single-shot echo-planar imaging CDI and DTI with b-values of 0 and 500 second/mm(2) was performed in 31 patients with chronic liver disease and 13 normal volunteers. Liver biopsy was performed in all patients with liver disease with a median delay of two days from MRI. Fibrosis and inflammation were scored on a 5-point scale (0-4). Liver ADCs obtained with CDI and DTI were compared between patients stratified by fibrosis stage and inflammation grade. Receiver operating characteristic (ROC) curve analyses were conducted to evaluate the utility of the ADC measures for prediction of fibrosis and inflammation.
Results: Patients with liver fibrosis and inflammation had significantly lower liver ADC than subjects without fibrosis or inflammation with CDI and DTI. For prediction of fibrosis stage > or = 1 and stage > or = 2, area under the ROC curve (AUC) of 0.848 and 0.783, sensitivity of 88.5% to 73.7%, and specificity of 73.3% to 72.7% were obtained, for ADC < or =1.40 x 10(-3) mm(2)/second and < or =1.30 x 10(-3) mm(2)/second (using CDI), respectively. For prediction of inflammation grade > or = 1, AUC of 0.825, sensitivity of 75.0%, and specificity of 78.6% were obtained using ADC < or = 1.30 x 10(-3) mm(2)/second (using CDI). CDI performed better than DTI for diagnosis of fibrosis and inflammation.
Conclusion: Liver ADC can be used to predict liver fibrosis and inflammation with acceptable sensitivity and specificity.
(c) 2008 Wiley-Liss, Inc.