Introduction: Intestinal failure associated liver disease (IFALD) is a significant complication in pediatric intestinal failure. Liver fibrosis with or without cholestasis is a frequent occurrence in IFALD. However, the diagnosis of fibrosis in IFALD is challenging and currently limited to liver biopsy. Our objective was to evaluate the use of transient elastography (TE) in pediatric IF patients to determine its efficacy for monitoring liver fibrosis as an alternative to liver biopsy.

Methods: A retrospective cohort study of IF patients between January 1, 2015 to December 31, 2017. The study cohort included a sample of patients who had a routine liver biopsy during an operative procedure and completed transient elastography at the same time. Liver biopsies were evaluated for stage of fibrosis using the modified Scheuer score. Statistical analysis included univariate analysis and Chi-square methods. High (grade 3-4) versus low (grade 0-2) fibrosis and TE scores were compared to determine sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Receiver operating characteristic (ROC) curves were also used to determine the ability of TE to discriminate various stages of fibrosis. Data is presented as medians with interquartile ranges and frequencies.

Results: 30 patients (21 male [70%]) with a median age at biopsy of 320 days (154-1776) were evaluated. Majority of patients had an etiology of abdominal wall defects (12[40%]) and necrotizing enterocolitis (9[30%]). Median length of PN therapy was 190 days (120-789). All TE measurements were completed within a median of 8.5 days (5.0-13.3) of the liver biopsy. Individuals with histologic fibrosis scores of 0-1 (n=18) had a median TE score of 4.9 (3.9-7.1), fibrosis scores of 2 (n=5) had a median TE of 6.7 (4.4-11.4) and those with a fibrosis score of 3 (n=7) had a median TE of 12.1 (6.3-14.6). No patients had grade 4 fibrosis. Dichotomous evaluation of the fibrosis scores (low vs high grade) compared to the chronic cholestatic scale for Fibroscan demonstrated 95.7% sensitivity, 57.1% specificity, 88.0% PPV and 80.0% NPV for low grade fibrosis. The c-statistic for the ROC curve was 0.764.

Conclusion: Fibrosis in patients with IF is often difficult to assess without liver biopsy. TE has only been evaluated in small cohorts but shows potential as a non-invasive method to monitor low vs high grade fibrosis in children with IF. Further studies in a larger cohort are required.