AIM To identify a panel of biomarkers that can distinguish between non-alcoholic fatty liver disease(NAFLD) and non-alcoholic steatohepatitis(NASH), and explore molecular mechanism involved in the process of developing NASH from NAFLD.METHODS Biomarkers may differ during stages of NAFLD. Urine and blood were obtained from non-diabetic subjects with NAFLD and steatosis, with normal liver function(n = 33), from patients with NASH, with abnormal liver function(n = 45), and from healthy age and sex-matched controls(n = 30). Samples were subjected to metabolomic analysis to identify potential non-invasive biomarkers. Differences in urinary metabolic profiles were analyzed using liquid chromatography tandem mass spectrometry with principal component analysis and partial least squares-discriminate analysis.RESULTS Compared with NAFLD patients, patients with NASH had abnormal liver function and high serum lipid concentrations. Urinary metabonomics found differences in 31 metabolites between these two groups, including differences in nucleic acids and amino acids. Pathway analysis based on overlapping metabolites showed that pathways of energy and amino acid metabolism, as well as the pentose phosphate pathway, were closely associated with pathological processes in NAFLD and NASH.CONCLUSION These findings suggested that a panel of biomarkers could distinguish between NAFLD and NASH, and could help to determine the molecular mechanism involved in the process of developing NASH from NAFLD. Urinary biomarkers may be diagnostic in these patients and could be used to assess responses to therapeutic interventions. AIM To identify a panel of biomarkers that can distinguish between non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH), and explore molecular mechanism involved in the process of developing NASH from NAFLD. METHODS Biomarkers may differ in stages of NAFLD. Urine and blood were obtained from non-diabetic subjects with NAFLD and steatosis, with normal liver function (n = 33), from patients with NASH, with abnormal liver function (n = 45), and from healthy age and sex-matched controls (n = 30). Samples were subjected to metabolomic analysis to identify potential non-invasive biomarkers. Differences in urinary metabolic profiles were analyzed using liquid chromatography tandem mass spectrometry with principal component analysis and partial least squares-discriminate analysis. patients, patients with NASH had abnormal liver function and high serum lipid concentrations. Urinary metabonomics found differences in 31 metabolites between th Pathway analysis based on overlapping metabolites showed that pathways of energy and amino acid metabolism, as well as the pentose phosphate pathway, were closely associated with pathological processes in NAFLD and NASH .CONCLUSION These findings suggested that a panel of biomarkers could distinguish between NAFLD and NASH, and could help to determine the molecular mechanism involved in the process of developing NASH from NAFLD. Urinary biomarkers may be diagnostic in these patients and could be be assess to to respond to therapeutic interventions .