Biochemical markers of liver fibrosis and lymphocytic piecemeal necrosis in UDCA-treated patients with primary biliary cirrhosis
Version of Record online: 9 JUN 2004
Volume 24, Issue 3, pages 187–193, June 2004
How to Cite
Corpechot, C., Poujol-Robert, A., Wendum, D., Galotte, M., Chrétien, Y., Poupon, R. E. and Poupon, R. (2004), Biochemical markers of liver fibrosis and lymphocytic piecemeal necrosis in UDCA-treated patients with primary biliary cirrhosis. Liver International, 24: 187–193. doi: 10.1111/j.1478-3231.2004.0918.x
- Issue online: 9 JUN 2004
- Version of Record online: 9 JUN 2004
- Received 6 October 2003, accepted 2 February 2004
- biliary liver cirrhosis;
- ursodeoxycholic acid;
- biological markers;
Abstract: Background/Aim: We have previously shown that the histological stage and severity of lymphocytic piecemeal necrosis (LPN) are independent predictive factors of cirrhosis development in ursodeoxycholic acid (UDCA)-treated patients with primary biliary cirrhosis (PBC). Our aim during this study was to determine whether biochemical parameters classically used in PBC management and measured under UDCA could be considered as reliable surrogate markers for these histological prognostic indices in clinical practice.
Method: The study included 153 patients with PBC who had undergone a control liver biopsy after 2 years of UDCA therapy. The relationships between histological and biological features were assessed by variance analysis and logistic regression. The diagnostic value of independent markers was assessed in terms of their sensitivity, specificity, positive predictive value (PPV) and negative value (NPV) and receiver-operating characteristic curves.
Results: Two variables were independently associated with extensive fibrosis (i.e. advanced histological stages): serum levels of bilirubin and hyaluronic acid (HA). A fibrosis index ([bilirubin (μmol/l)/14]+[HA (μg/l)/143]) higher than 1.5 exhibited good PPV and specificity (>74%) but rather poor NPV and sensitivity (<64%) regarding a diagnosis of extensive fibrosis. The only independent marker of LPN was aspartate aminotransferase (AST) activity. AST activity of more than twice the upper limit of normal showed acceptable PPV (>70%) but very low sensitivity (<25%) for a diagnosis of LPN.
Conclusions: Serum bilirubin and HA levels measured under UDCA therapy are of acceptable diagnostic value for extensive fibrosis, but none of the biochemical tests commonly employed in the management of PBC can be considered as surrogate markers of LPN. Taken together with our previous results, these findings suggest that liver biopsy may be necessary to screen UDCA-treated patients who might require additional therapies.