Background/aim: Accurate histological assessment of liver fibrosis is essential in the management of chronic hepatitis B (CHB). Although semi-quantitative scoring systems describe well the pathological patterns of hepatic structure, they produce fibrosis evaluation that is not very precise. Image analysis or morphometry has the theoretical advantage of providing truly quantitative data.
Patients and methods: The present study aimed at validating a new image analysis system, Bioquant Nova Prime, in estimating collagen content in liver biopsy samples from patients with CHB. The biopsies were stained with picrosirius red and the areas of collagen were measured. The results were correlated with laboratory parameters and Ishak modified histological scores. Discriminative reliability of morphometry was determined using receiver operating characteristics (ROC) analysis.
Results: There was excellent interobserver agreement (r=0.84-0.94, P<0.01) in the morphometric analysis. Significant correlations between the quantitative morphometric data and the semi-quantitative score (Spearman's r=0.68-0.78, P<0.001) were also demonstrated. Excellent discriminative power of morphometry in differentiating mild from advanced fibrosis and cirrhosis from absence of cirrhosis was shown by the ROC analysis.
Conclusions: Our results validated the use of Bioquant Nova Prime in estimating collagen content in liver biopsies. We showed that morphometry is a sensitive method of liver fibrosis quantification in CHB and complements semi-quantitative histological scoring system. This tool, with its reliable intraassay variability, could be of special value in assessing histological response to treatment after anti-viral or anti-fibrotic therapy.