Longitudinal changes in liver fibrosis in children with sickle cell disease undergoing chronic transfusion therapy

Background and study aims: The progression of liver injury from transfusional iron overload in sickle cell disease (SCD) is poorly understood. We sought to identify predictors liver fibrosis development over time. Patients and methods: We performed a retrospective cohort study of chronically transfused SCD patients who had ≥ 2 serial liver biopsies. Core biopsies were scored for fibrosis in a blinded fashion. Primary analyses evaluated longitudinal changes in liver fibrosis and changes in surrogate markers. Secondary analyses determined the relationship between liver iron concentration (LIC) and serum biomarkers. Results: 26 people had ≥ 2 serial biopsies for evaluation (n = 70 biopsies total). Fibrosis was Ishak grade 0 or 1 in all biopsies. Evaluation of the first 2 biopsies showed fibrosis regression (n = 6), development (n = 2), persistence (n = 1), and absence (n = 17). There was no consistent association of fibrosis with LIC over time, or between changes in fibrosis status and surrogate markers. For predicting fibrosis on a cross-sectional basis, ALT and ferritin performed moderately (AUCs 0.80 and 0.63, respectively) but LIC performed poorly (AUC 0.30). The highest positive likelihood ratios for fibrosis were for ferritin cutoff of 5000 ng/mL (LR + 5.7) and ALT cutoff of 65 U/L (LR + 5.2). Conclusions: Liver fibrosis progression is minimal in chronically transfused SCD. LIC does not correlate well with fibrosis development. We propose routine liver biopsies are not necessary components in the standard monitoring of chronically transfused SCD patients.