Introduction Outcome of liver disease in children is mainly determined by severity and progression of liver fibrosis. Liver biopsy is the accepted standard for evaluating fibrosis but is limited by the need for sedation in children, sampling error and risks including bleeding.

Aim The aim of this study was to compare tools for non-invasive assessment of liver fibrosis in a paediatric cohort.

Method Children were recruited at the time of liver biopsy and underwent transient elastography (TE) and blood collection. Liver biopsies were scored by a hepato-histopathologist from F0 (no fibrosis) to F4 (cirrhosis). Serum samples underwent analysis for the Enhanced Liver Fibrosis (ELF) test; comprising hyaluronic acid, P3NP and TIMP1 (iQur, UK). CK18-M30 levels (caspase cleavage fragments) were measured using ELISA. Biomarkers were compared to biopsy score.

Results During the study period 101 children (62 boys) were enrolled. Median age: 13.6 (range 6–18) years. Diagnosis was autoimmune liver disease (AILD) in 27; non-alcoholic fatty liver disease (NAFLD) in 37; 13 children were post-transplant; 8 children had Hepatitis B/C; 5 had Wilson disease and the remainder miscellanous. TE was a good discriminator of fibrosis ≥ F2 (p<0.001), ≥ F3 (p<0.001) and F4 (p=0.003) with areas under the ROC curve of 0.78, 0.8 and 0.96 respectively. Data derived cut-offs for ≥ F1 were 6.1 kPa, ≥ F2; 6.8 kPa, ≥ F3; 8.3 kPa and F4; 14.1 kPa.

Blood biomarkers were not as accurate in distinguishing severity of disease. ELF performed better with increasing stages of fibrosis. Area under the curve for cirrhosis was 0.77. CK18-M30 levels could distinguish significant fibrosis (≥ F2) (p=0.009) with an area under the ROC curve of 0.77, severe fibrosis (≥ F3) with an AUROC of 0.69 and cirrhosis (F4) with an AUROC of 0.69. Within the different diagnostic groups, again TE performed better than serum biomarkers with AUROC of 0.75 for ≥ F2 and 0.81 for ≥ F3 in NAFLD; 0.85 for ≥ F2, 0.94 for ≥ F3 and 1.0 for F4 in AILD; and in children post-transplant; 0.9 for ≥ F2 and 0.83 for ≥ F3.

Conclusion To our knowledge, this is the largest paediatric study reported to date comparing TE and serum biomarkers for the non-invasive evaluation of fibrosis. TEwas a reliable tool in distinguishing different stages of liver fibrosis in paediatric patients. Blood biomarkers may be of use in combination with TE especially in the stratification of more severe disease. Routine use of these techniques may serve as a useful adjunct to liver biopsy for diagnostic purposes and provide a reliable method of non-invasively monitoring liver disease progression in children.