AUTHOR=Kwon Yiyoung , Kim Eun Sil , Choe Yon Ho , Kim Mi Jin TITLE=Stratification by Non-invasive Biomarkers of Non-alcoholic Fatty Liver Disease in Children JOURNAL=Frontiers in Pediatrics VOLUME=10 YEAR=2022 URL=https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2022.846273 DOI=10.3389/fped.2022.846273 ISSN=2296-2360 ABSTRACT=Background

The spectrum of non-alcoholic fatty liver disease (NAFLD) ranges from isolated hepatic steatosis to non-alcoholic steatohepatitis to fibrosis. We aimed to introduce useful biomarkers released during liver inflammation and fibrogenesis that are easy to use in outpatient clinic and adjust to children to evaluate each NAFLD stage without biopsy.

Methods

This prospective study included 60 patients aged under 19 years whose alanine aminotransferase (ALT) levels were elevated from March 2021. All patients were proven to have NAFLD by ultrasonography and laboratory work-up to exclude other causes of hepatitis. Fibroscan and additional laboratory tests for biomarkers [procollagen type1 amino-terminal propeptide (P1NP), osteocalcin, interleukin-6 (IL-6), and Mac-2 binding protein glycosylated isomer (M2BPGi)] were performed. Fibroscan-AST (FAST) score was used for the comparison of steatohepatitis and liver stiffness measurement (kPa) was used for the comparison of advanced fibrosis.

Results

The biomarker that showed a significant difference between the FAST-positive and negative groups was the P1NP/osteocalcin ratio with a p-value of 0.008. The area under receiver operating characteristic (AUROC) of P1NP/osteocalcin ratio*ALT values (values obtained through multivariate analysis) was 0.939 with the cut-off value of 305.38. The biomarkers that showed a significant difference between the LSM-positive and negative groups were IL-6 and M2BPGi with a p-values of 0.005 and <0.001. AUROC of IL-6 *AST values (values obtained through multivariate analysis) was 0.821 with the cut-off value of 228.15. M2BPGi showed a significant linear relationship with LSM in Pearson correlation analysis (Pearson correlation coefficient = 0.382; p = 0.003). The diagnostic capability of M2BPGi to evaluate advanced fibrosis showed an acceptable result (AUROC = 0.742; p = 0.022).

Conclusions

Non-invasive biomarkers can be used to predict each stage of NAFLD in children. The measurements of P1NP, IL-6 or M2BPGi along with the basic chemistry tests would help determine the stage of NAFLD they correspond to at the time of initial diagnosis and predict responsiveness after the treatment.