Quantitative Ultrasound Techniques for Diagnosis of Nonalcoholic Steatohepatitis

BACKGROUND - Nonalcoholic fatty liver disease (NAFLD) is the most prevalent liver disease in Western countries. The more advanced form, nonalcoholic steatohepatitis (NASH) may evolve to fibrosis, cirrhosis, liver failure and liver cancer. Liver biopsy is the current reference standard for diagnosis of NASH. However, its invasiveness prevent its use for large-scale screening and diagnosis. Members of our team have developed innovative quantitative ultrasound (QUS) techniques (shear wave viscoelastography and sub-resolution cellular imaging) that have a high diagnostic potential. Advantages over magnetic resonance (MR) include the ability to characterize viscosity, cellular organizations and cost-effectiveness. We hypothesize that a combination of QUS techniques offering complementary assessment of tissue characteristics will differentiate NASH from NAFLD and quantify evolving grades of liver fat, inflammation and fibrosis.

OBJECTIVES - Primary objective: 1) Determine the accuracy of QUS parameters for diagnosis of NASH. Secondary objectives: 2) Determine the diagnostic accuracy of QUS for grading histology-determined liver fat, inflammation and fibrosis. 3) Compare the diagnostic accuracy of QUS and MR techniques for liver fat quantification.

Hypotheses: We hypothesize that QUS methods, including viscoelasticity, homodyned-K parameters and attenuation parameters can characterize properties of tissues encountered in NASH (including liver fat, inflammation and fibrosis).

METHODOLOGY - Design: This will be a cross-sectional imaging trial to evaluate the diagnostic accuracy of QUS techniques in 10 non-obese volunteers and 92 patients, using histopathology as the reference standard for patients. Paired index tests QUS-based (mechanical and cellular parameters) and MR-based (mechanical parameters and relaxation times reflecting tissue content) techniques will be performed as research procedures in close temporal proximity to the reference test (liver biopsy).

Inclusion criteria: Consecutive adult patients with known or suspected NAFLD or NASH undergoing a liver biopsy for clinical indications.

Exclusion criteria: Contraindication to MR, refusal to participate, any other cause of chronic liver disease or prior liver transplantation.

Data analysis: Random forest classifier to develop a predictive model. Cross-validated receiver operating characteristic curve analysis, sensitivity and specificity to assess diagnostic accuracy.

RATIONALE AND IMPACT - NASH has become the second leading cause of liver transplantation in North America and is predicted to become the leading indication in the near future. While liver biopsy is the established reference standard for NASH diagnosis and MR constitutes a noninvasive alternative, they are both impractical for large-scale application. Thus, there is an urgent need to investigate non-invasive and cost-effective techniques for diagnosis of NASH. Our proposed QUS techniques are experimental and currently not available on clinical US systems. A major impact of this work, for patients and medical institutions, will be to lower the need for liver biopsy, the risk of complications and reduce the cost for NASH diagnosis while providing quantitative therapeutic targets.