Influence of perfusion on hepatic MR diffusion measurement

Abstract

Percutaneous liver biopsy is the current gold standard for the diagnosis and assessment of hepatic fibrosis: it is an uncomfortable procedure with associated risks of infection and internal bleeding. Diffusion-weighted MRI may be important as a non-invasive measure of the grade of hepatic fibrosis, owing to the morphological and physiological changes in hepatic tissues as fibrosis develops. The factors that influence hepatic diffusion measurements are complex and are thought to include hepatic perfusion. This study assesses whether hepatic diffusion measurements are influenced by hepatic perfusion changes as this could be an important variable when investigating the relationship between diffusion measurements and grade of hepatic fibrosis. Hepatic apparent diffusion coefficients (ADCs) were measured in 10 healthy volunteers after an 8-h fast and 1-h after a standard meal (which increases portal vein flow and hepatic sinusoidal perfusion) using a breath-hold technique at b-values of 750, 500 and 200 s/mm(2). ADC measurements using b = 750 and 500 s/mm(2) in the posterior right lobe were unaffected by the perfusion changes (P = 0.15 and 0.14, respectively) whereas the results from the anterior right lobe significantly increased post-prandially (P = 0.03 and 0.008, respectively): inter-subject variability was low ( approximately 5%) for these b-values. Portal perfusion had a significant effect in both the posterior and anterior segments of the right lobe when using a b-value of 200 s/mm(2) (P = 0.01 and 0.00001 respectively): inter-subject variability was found to be high ( approximately 25%). The most reproducible measurements of the ADC can be made in the posterior right lobe and the effect of perfusion is minimised at the higher b-values (b = 500 and 750 s/mm(2)).