Serum connective tissue markers as predictors of advanced fibrosis in patients with chronic hepatitis B and D

2011 
Background/aims: Liver biopsy to assess fibrosis is invasive and prone to sampling error. While algorithms of serum markers to predict fib- rosis stage have been described for chronic hepatitis C, these cannot be applied equally well to hepatitis B. Methods: We therefore determined 9 serum fibrosis markers, liver biochemical tests and ultrasound parameters in 109 consecutive adult patients with chronic hepatitis B and D. All patients had compensated liver disease. Using the METAVIR score, advanced disease was defined as fibrosis stage ≥F2, and active inflam- mation as grade ≥A2. A gold standard was created considering splenomegaly and/or platelets <150,000 as indicators of advanced fibrosis ir- respective of histology. Area under receiver operating characteristics curves was used for assessment of single markers and odds ratio for their combinations. Results: Patients with advanced disease were older, had lower albumin, higher gamma glutamyl transferase and lower plate- let. Levels of 6 of the 9 fibrosis markers, tissue inhibitor of metalloproteinases-1, procollagen type III aminoterminal propeptide, matrix metal- loproteinase-2, laminin, hyaluronan and collagen IV correlated with advanced fibrosis. Markers useful for fibrosis prediction also predicted marked inflammation. Using the gold standard, age, prothrombin time, gamma glutamyl transferase and albumin were independent predic- tors of fibrosis with odds ratio's of 3.11, 4.18, 3.35 and 5.25, respectively. Their combined use predicted fibrosis with an odds ratio of 228.8. Tis- sue inhibitor of metalloproteinases-1 and hyaluronan were powerful predictors of fibrosis (Odds ratio's of 8.65 and 8.38). Their combined use revealed an odds ratio of 28.6, when compared with the gold standard. Conclusion: In conclusion, advanced liver fibrosis in chronic hepati- tis B and D may be predicted with use of these two fibrosis markers.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    58
    References
    15
    Citations
    NaN
    KQI
    []