Enhanced Liver Fibrosis (ELF) test can predict clinical outcomes in patients with chronic liver disease

Parkes, J., Roderick, P., Harris, S., Day, C, Mutimer, D, Collier, J., Lombard, M., Alexander, G., Ramage, J., Dusheiko, G., Wheatley, M., Gough, C, Burt, A and Rosenberg, W. (2010) Enhanced Liver Fibrosis (ELF) test can predict clinical outcomes in patients with chronic liver disease. Gut, 59 (9), 1245-1251. (doi:10.1136/gut.2009.203166).

Abstract

Assessment of liver fibrosis is important in determining prognosis and evaluating interventions.

Due to limitations of accuracy and patient hazard of liver biopsy, non-invasive methods have been sought to provide information on liver fibrosis, including the European liver fibrosis (ELF) test, shown to have good diagnostic accuracy for the detection of moderate and severe fibrosis.

Access to independent cohorts of patients has provided an opportunity
to explore if this test could be simplified. This paper reports the simplification of the ELF test and its ability to identity severity of liver fibrosis in external validation studies in patients with chronic hepatitis C (CHC).

Paired biopsy and serum samples from 347 na?¨ve patients with
CHC in three independent cohorts were analysed. Diagnostic
performance characteristics were derived (AUROC, sensitivity and specificity, predictive values), and clinical utility modelling performed to determine the proportion of biopsies that could have been avoided if ELF test was used in this patient group. It was possible to simplify the original ELF test without loss of performance and the new
algorithm is reported.

The simplified ELF test was able to predict severe fibrosis [pooled AUROC of 0.85 (95% CI 0.81–0.89)] and using clinical utility modelling to predict severe fibrosis (Ishak stages 4–6; METAVIR stages 3 and 4) 81% of biopsies could have been avoided (65% correctly).

Issues of spectrum effect in diagnostic test evaluations are discussed. In chronic hepatitis C a simplified ELF test can detect severe liver fibrosis with good accuracy.

This record has no associated files available for download.

Contributors

Author: J. Parkes

Author: P. Roderick

Download statistics