Cutoff values for alcoholic liver fibrosis using magnetic resonance elastography technique.

BACKGROUND Due to the lack of cutoff values validated for specific liver diseases, the purpose of this study was to set up specific magnetic resonance elastography (MRE) cutoff values for asymptomatic liver fibrosis in alcoholic patients. METHODS Ninety patients underwent 3 clinical exams. The liver stiffness was measured locally with the Fibroscan, and globally through cartographies of shear modulus generated with MRE. The Fibroscan method was chosen as the gold standard to classify the fibrosis. The liver score was also obtained with the Fibrometer A, and the diagnostic performance of the methods was analyzed with receiver-operating characteristic (ROC) curves and cutoff values were calculated. RESULTS Spearman correlation and area under the ROC curve revealed that MRE is a better diagnostic method than the Fibrometer A, to identify various levels of fibrosis. The results showed that the Fibrometer A was adapted for severe fibrosis. The MRE cutoff values are F1:2.20 kPa, F2:2.57 kPa, F3:3.31 kPa, and F4:4 kPa and were not influenced by the glutamic oxaloacetic transaminase level. By using the ultrasound cutoff values attributed for alcoholism, 66% of patients had a similar liver fibrosis diagnosis as the MRE cutoffs. However, both imaging techniques did not provide the same distribution for minor fibrosis. CONCLUSIONS None of the imaging techniques (Fibroscan, MRE) could replace the gold standard of the biopsy. However, due to the risk and the unnecessary procedure for the present recruited alcoholic patients, the Fibroscan method was chosen as the reference. Since MRE is currently being used as a clinical exam, the present MRE cutoffs could aid clinicians with their diagnosis of liver fibrosis for alcoholism disease.