Validation of transient elastography and comparison with spleen length measurement for staging of fibrosis and clinical prognosis in primary sclerosing cholangitis

Abstract

Background: Patients with primary sclerosing cholangitis (PSC) develop progressive liver fibrosis and end-stage liver disease. Non-invasive and widely available parameters are urgently needed to assess disease stage and the risk of clinical progression. Transient elastography (TE) has been reported to predict fibrosis stage and disease progression. However, these results have not been confirmed in an independent cohort and comparison of TE measurement to other non-invasive means is missing. Methods: In a retrospective study we collected data from consecutive PSC patients receiving TE measurements from 2006 to 2014 (n = 139). Data from 62 patients who also underwent a liver biopsy were used to assess the performance of TE and spleen length (SL) measurement for the staging of liver fibrosis. Follow-up data from this cohort (n = 130, Hamburg) and another independent cohort (n = 80, Paris) was used to compare TE and SL as predictors of clinical outcome applying Harrel's C calculations. Results: TE measurement had a very good performance for the diagnosis and exclusion of higher fibrosis stages (≥F3: AUROC 0.95) and an excellent performance for the diagnosis and exclusion of cirrhosis (F4 vs. < F4: AUROC 0.98). Single-point TE measurement had very similar predictive power for patient outcome as previously published. In a combined cohort of PSC patients (n = 210), SL measurements had a similar performance as TE for the prediction of patient outcome (5 x cross-validated Harrel's C 0.76 and 0.72 for SL and TE, respectively). Conclusions: Baseline TE measurement has an excellent performance to diagnose higher fibrosis stages in PSC. Baseline measurements of SL and TE have similar usefulness as predictive markers for disease progression in patients with PSC.