356 ENHANCED EXPRESSION OF LIVER X RECEPTOR AND ITS LIPOGENIC AND INFLAMMATORY TARGETS IN NONALCOHOLIC FATTY LIVER DISEASE AND CHRONIC HEPATITIS C

Abstract

Our aim was to compare two expert pathologists with digital image analysis (DIA) using both H&E and ORO stained sections to assess the degree of fat in mice with NAFLD. Methods: We studied histological steatohepatitis exhibiting a spectrum of microvesicular and macrovesicular fat in male C57BL/6J mice with high fat and/or high fructose diets. Tissue sections were stained with both H&E and ORO and examined by two expert pathologists. DIA was used to calculate the percentage of steatosis on H&E and ORO stained sections and size distribution of fat droplets, in the latter. Triglyceride concentrations in the tissue were measured biochemically as a true reflection of tissue lipid content. Results: ORO staining identified fat in the control and fructose groups where no fat was identified by expert pathologists on H&E staining. Most of this fat was microvesicular. The two expert pathologists showed strong interobserver agreement (Pearson correlation R=0.991, p ≤ 0.001). However, compared to ORO DIA, they overestimated the amount of fat for the high fat (71.6% vs. 46.2%) and high fat/fructose (71.9% vs. 47.0%) groups where there were large amounts of macrovesicular fat (p = 0.011 and p=0.003). ORO DIA accurately reflects the liver triglyceride concentrations and is therefore an accurate reflection of liver steatosis (R = 0.706, p = 0.001). Conclusions: ORO staining can identify fat in cases where it cannot be seen on H&E staining. Although NAS is clinically valid, in research situations ORO DIA is warranted as the most reliable way to accurately assess liver steatosis and also give additional information on fat droplet size.