Abstract

AIMS: To validate a method of assessment of low hepatic iron concentration based on a biochemical colorimetric assay plus histological scoring. METHODS: The within-day and day to day precision of the iron colorimetric assay was determined on frozen rat liver. The coefficient of variation (CV) of iron measurement in two separate samples from the same liver was determined for 21 deparaffinised human biopsies. The intra- and interlaboratory variability of the colorimetric assay and histological scoring were assessed on 38 deparaffinised liver biopsies. RESULTS: For the within-day test, the CV was 11% (5.1 (0.6) mumol/g dry weight (dw), mean (SD) iron concentration). For the day to day test, the CV was 19.5% (8.2 (1.6) mumol/g dw). The CV was 14.7% for iron concentration determined in two separate samples from the same liver. By correlation and kappa concordance tests, the intra- and interlaboratory variability of the hepatic iron colorimetric assay and iron histological scoring was slight. Absence of stainable iron corresponded to a liver iron concentration < or = 20 mumol/g dw. CONCLUSIONS: A combination of two complementary methods, colorimetric measurement and histological scoring, is an accurate and reliable way of determining low iron concentrations in deparaffinised human liver biopsies. In secondary haemosiderosis, such methods would be essential for investigating the role of low iron overload in fibrogenesis and during the response to antiviral treatment in chronic viral hepatitis.