Abstract
BACKGROUND AND AIM: One of the unmet needs in subjects with metabolic risks is the prediction of metabolic liver disease by noninvasive tests. The construction of performant tests is dependent on the appropriateness of the histological reference definition. The aim of this study was to analyze the limitations of similar European (Fatty Liver Inhibition of Progression) and USA (Clinical-Research-Network) standard definitions and their impact on the construction of tests. METHODS: We hypothesized that a simpler histological definition of non-alcoholo steato-hepatitis (NASH), which does not require the presence of steatosis and the presence of both lobular inflammation and ballooning, should improve the concordance rates with previously validated blood tests. We reviewed the landmark studies in metabolic liver disease, sources of the standard definitions, and we compared the adequacy of these standards to other possible definitions in 1081 subjects with biopsies, by concordance and accuracy rates. RESULTS: The limitations of standard definitions included the presence of appropriate controls in only 6.6% of landmark studies, an arbitrary definition of steatosis and NASH covering only four (15%) out of 27 possible combinations of features, compared with 18 (67%) for a simplified NASH definition, which did not require steatosis. A total of 39/1081 (3.6%) cases were not identified by standard definition, but were identified by the simplified definition as significant active disease, including 15 cases with significant fibrosis. The simplified definition increased the κ concordance (P<0.0001) between test prediction and histological reference. CONCLUSION: A simplified definition of NASH could help in the construction of biomarkers with higher performances.