Introduction Half of all the liver biopsies performed are to assess the severity of pathology including grading of fat, iron accumulation as well as fibrosis. Liver biopsies are invasive tests associated with sampling errors; the coefficient of variation for fibrosis measurement is 45% even with 25mm long specimens. We aimed to develop and validate non-contrast, non-breath-holding, quantitative MRI methodology to estimate the amout of fibrosis, fat and iron accumulation within the whole liver.
Methods MRI relaxation time data (T1, T2and T2*) were acquired (over 15–20 minutes) using a novel Echo Planar Imaging technique with a respiratory-triggered (r.t.) acquisition method. 1H MR spectra were acquired (r.t.) using a multiple echo PRESS acquisition which allowed for individual T2correction to the spectrum for accurate quantification of the fat fraction in a 30x30x30mm3voxel.
Results 115 patients (67 Training; 48 Validation cohort) with suspected chronic liver disease aged 19 to 72 years [alcoholic (13%), non-alcoholic (56%) fatty liver disease, chronic viral hepatitis (21%) and haemochromatosis (3%)] who had a liver biopsy ≥25 mm were included in the study. The diagnostic accuracy of the T1 parameter in the detection of different histological stages of fibrosis, using receiver operator curves and areas under the curve (AUC), in the training and validation cohort are summarised in Table 1. There were also significant correlations between MR measures of fat fraction and staging of steatosis with a Spearman’s correlation coefficient of 0.760 (p < 0.001) and T2* with hepatic iron staging with Spearman’s correlation coefficient of –0.588 (p < 0.001). The T1 relaxation time of the liver correlated with the percentage of fibrosis measured as a continous variable on morphometry within the entire study population (Pearson correlation coefficient of 0.712, p < 0.001).
Conclusion Across a range of chronic liver diseases, MR measures of fat fraction, hepatic iron content and fibrosis of the whole liver correlate well with related histological measures.
Disclosure of Interest None Declared
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.