Methodological aspects of the interpretation of non-invasive biomarkers of liver fibrosis: a 2008 update.

This review summarizes the methodological aspects of the interpretation of non-invasive biomarkers in liver fibrosis. A scoring system has been updated to better compare the quality of fibrosis biomarkers. Several methodological issues are related to the classical methodology using biopsy, as this is considered the gold standard. However, from evidence-based data, it appears that the methodology needs to change to prevent flawed conclusions among key opinion leaders as well as in obsolete guidelines. As waiting for the perfect biomarker for the diagnosis of advanced fibrosis to come along is probably a waste of time, in the meantime, methods can be improved. The main proposals for improving the methodology are, to take into account the spectrum bias, to assess accuracy between adjacent stages, to compare biomarkers in the same patient, to assess the cause of failure among discordant cases and to use specific statistical methods adapted for imperfect gold standards.

Relationship between the Fibrotest and portal hypertension in patients with liver disease.

BACKGROUND: The best technique to estimate portal hypertension (PHT) is to measure the hepatic venous pressure gradient (HVPG), which is an invasive method. AIM: To assess the relationship between the Fibrotest (Biopredictive, Paris, France) and the presence and degree of PHT in patients with liver disease, and to determine if the Fibrotest can diagnose severe PHT, defined by HVPG >or= 12 mmHg, in cirrhotic patients. METHODS: Patients who underwent a transjugular liver biopsy were prospectively included. HVPG was measured, and classification of histological lesions assessed. The same day, blood samples for Fibrotest were performed. RESULTS: A total of 130 patients were included (no or minimal fibrosis: 12%, moderate fibrosis 17%, cirrhosis 71%). There was a significant correlation between Fibrotest and HVPG (Pearson correlation coefficient = 0.58, P < 0.0001), also weaker in cirrhotic patients (Pearson correlation coefficient = 0.24, P = 0.02). In cirrhotic patients, Fibrotest was significantly higher when there was a severe PHT (0.87 +/- 0.15 vs. 0.73 +/- 0.14, respectively, P = 0.02). The areas under the receiver operating characteristic curves for the diagnosis of severe PHT was 0.79 +/- 0.07, not different from that of platelets and Child-Pugh score. CONCLUSION: In patients with liver disease or cirrhosis, Fibrotest is correlated with the presence and degree of PHT. Other studies are needed to confirm these results, especially in non-decompensated cirrhotic patients.