Abstracts

RETURN TO ABSTRACT LISTING


E2406. Validation of Clinically Established Cut-Off Values for MRE Liver Fibrosis Assessment in a Non-Transplant Center
Authors
  1. Hailey Choi; University of California San Francisco
  2. Kevin Zand; University of California San Francisco
  3. Michael Ohliger; University of California San Francisco
Objective:
Although MRE is integrated as an integral part of liver imaging in many institutions, cut-off values for estimating fibrosis varies in different studies, and there is no widely accepted, consensus agreement on cut-off values to stratify liver stiffness values to fibrosis risk. At our institution, we use clinically established threshold values to estimate liver fibrosis. The objective of our study was to validate our institution’s clinically established MRE liver stiffness cut-off values in our non-transplant setting clinical practice.

Materials and Methods:
In this IRB approved, HIPAA compliant study, radiology database was queried for MRE exams between 2016-2020. The radiology database was also queried for liver biopsies performed by interventional radiology within the same time interval. For patients who underwent MRE and liver biopsy within 12 months, the MRE liver stiffness measurement was compared with histologic diagnosis of fibrosis from liver biopsy. Our institution uses clinically established cut-off values for fibrosis assessment. Liver stiffness < 3 kPa is designated as low risk (F0-F1), = 3 and < 4.5 kPa as intermediate risk (F1-F2), and = 4.5 kPa as high risk for fibrosis (F3 and F4). Descriptive statistics were performed.

Results:
Total 478 MRE exams were performed. 88 MRE exams were performed in 84 patients who had liver biopsies. 57 patients had MRE within 12 months of liver biopsy. Of the 57, 8 patients had nondiagnostic elastography and were excluded for analysis. Final study population included 49 patients. Most common indication for biopsy was transaminitis (33%) and non-alcoholic fatty liver disease (33%), followed by hepatitis B viral infection (15%). In 41/49 (84%), MRE risk stratification was concordant with the pathologic diagnosis. In 2/49 (4%), MRE mildly understaged fibrosis (low vs. intermediate, or intermediate vs. high risk); MRE mildly upstaged fibrosis in 1/49 (2%). In 5/49 (10%), MRE significantly understaged hepatic fibrosis, with a discrepancy of at least 2 risk categories.

Conclusion:
Our clinically established liver stiffness values perform well, with 84% concordance with liver biopsy, in patients with multiple underlying etiologies. Only 5 cases demonstrated significant discrepancy. These clinically established cut-off values for MRE are easy to remember and can be a promising tool for risk stratification in patients with chronic liver disease.