Hydroxyproline content of needle biopsies as an objective measure of liver fibrosis: Emphasis on sampling variability.

Needle liver biopsy is essential for the diagnosis of liver fibrosis and cirrhosis. However, sampling variability has been reported by semiquantitative methods. The aim of the present study was to investigate quantitatively the sampling variability of needle liver biopsy on the assessment of liver fibrosis/cirrhosis. Liver samples of 12 patients undergoing autopsy, liver transplantation, or liver tumor resection were investigated. Nine to 10 liver biopsies at least 5 mm apart were taken from each of 12 removed human livers using Trucut needles. Hydroxyproline level of each liver core was determined. Sections >1 cm(2) and stained by Sirius red were used for the reference staging of liver fibrosis in each patient according to the METAVIR scoring system. Hydroxyproline levels for the three groups, F0-1, F2-3, and F4, were significantly elevated as the fibrosis score increased. There was a wide range of hydroxyproline contents for the 9-10 biopsy samples from each liver, the coefficient of variation being between 13% and 36% (mean, 23%). Several differences were noted between the amount of collagen present according to hydroxyproline measurement and the stages of fibrosis according to histology. Although increases in collagen content (by hydroxyproline) approximately paralleled increases in fibrosis stage (by histology), there was considerable overlap and some discrepancies. Much work remains to be done to clarify the significance of the apparent huge variations in fibrosis within diseased livers. It is advisable for clinicians to interpret staging diagnoses by histological staging with some caution.