Computer morphometry for quantitative measurement of liver fibrosis: comparison with Knodell's score, colorimetry and conventional description reports.

Liver fibrosis is currently described quite subjectively or, at best, semiquantitatively by scoring systems. In order to measure the severity of liver fibrosis quantitatively and to compare this with established methods, such as Knodell's scoring system, the colorimetric method and conventional description reports, we undertook the present study. A personal computer with an image grabber card and a microscope equipped with a computer-controlled slide-driver was used for computer morphometry. The principle behind morphometry is based on the different colours of hepatocytes and fibres following staining with Masson's trichrome stain. There were 31 patients (25 male, six female) recruited into the present study with a mean +/- SD age of 41.6 +/- 15 years (range 24-66 years). Of these patients, 16 had chronic hepatitis B, 12 had chronic hepatitis C and three were alcoholics. Colorimetric methods and Knodell's fibrosis score were performed according to established protocols. Conventional description reports were obtained from reviews of patient charts. The results from computer morphometry were highly correlated with results from the colorimetric method, with a correlation coefficiency gamma = 0.85 (P<0.0001). The results from computer morphometry also correlated with both Knodell's scoring system (gamma = 0.69; P<0.001) and conventional description reports (gamma = 0.46; P<0.01). Results from Knodell's scoring system were significantly correlated with computer morphometry, as follows: score 0, 2.7 +/- 1.4; score 1, 5.7 +/- 1.2; score 2, 7.7 +/- 2.3; score 3, 10.7 +/- 3.2; score 4, 21.8 +/- 14.1. The trend was statistically significant by the Wilcoxon rank sum test. In conclusion, our computerized morphometry system is a reliable tool for the evaluation of the severity of liver fibrosis and can be used as a tool for the objective quantification of liver fibrosis.