Transient elastography for non-invasive evaluation of post-transplant liver graft fibrosis in children

As graft survival in pediatric LT is often affected by progressive fibrosis, numerous centers carry out protocol liver biopsies. Follow-up biopsy protocols differ from center to center, but all biopsies are progressively spaced out, as time from transplant increases. Therefore, there is a need for non-invasive techniques to evaluate graft fibrosis progression in those children who have no clinical or serological signs of liver damage. Indirect markers, such as the APRI, should be relied on with caution because their sensitivity in predicting fibrosis can be strongly influenced by the etiology of liver disease, severity of fibrosis, and patient age. A valid alternative could be TE, a non-invasive technique already validated in adults, which estimates the stiffness of the cylindrical volume of liver tissue, 100-fold the size of a standard needle biopsy sample. The aims of this study were to evaluate the reliability of TE in children after LT and to compare both the TE and the APRI index results with the histological scores of fibrosis on liver biopsies. A total of 36 pediatric LT recipients were studied. All patients underwent both TE and biopsy within a year (median interval -0.012 months) at an interval from LT of 0.36 to 19.47 years (median 3.02 years). Fibrosis was assessed on the biopsy specimens at histology and staged according to METAVIR. There was a statistically significant correlation between TE stiffness values and METAVIR scores (P = .005). The diagnostic accuracy of TE for the diagnosis of significant fibrosis (F ≥ 2) was measured as the area under the curve (AUROC = 0.865), and it demonstrated that the method had a good diagnostic performance. APRI was not so accurate in assessing graft fibrosis when compared to METAVIR (AUROC = 0.592). A liver stiffness cutoff value of 5.6 kPa at TE was identified as the best predictor for a significant graft fibrosis (METAVIR F ≥ 2) on liver biopsy, with a 75% sensitivity, a 95.8% specificity, a 90% positive predictive value, and an 88.5% negative predictive value. These data suggest that TE may represent a non-invasive, reliable tool for the assessment of graft fibrosis in the follow-up of LT children, alerting the clinicians to the indication for a liver biopsy, with the aim of reducing the number of protocol liver biopsies.