Optimal correlation between different instruments for Fibrotest-Actitest protein measurement in patients with chronic hepatitis C

Abstract

Combination of alpha 2-macroglobulin, haptoglobin, apolipoprotein-A1, gamma-glutamyl transpeptidase, total bilirubin and alanine aminotransferase measurements allows to determine the Fibrotest-Actitest score, an alternative to liver biopsy in hepatitis C virus infection. The aims of this study were to evaluate the analytical variability of the Fibrotest-Actitest proteins alpha 2-macroglobulin, haptoglobin and apolipoprotein-A1, and to assess their impact on the Fibrotest-Actitest scores. We compared 129 sera from hepatitis C virus infected patients for alpha 2-macroglobulin, haptoglobin and apolipoprotein-A1 levels obtained with the Immage (Beckman-Coulter) and the BNProspec (Dade-Berhing) automates. We evaluated Fibrotest-Actitest results obtained with the two nephelemeters. Optimal correlation was found for alpha 2-macroglobulin (Y=1.05X + 0.01, correlation coefficient: 0.98) and haptoglobin (Y=1.05X - 0.07, correlation coefficient: 0.98). Apolipoprotein-A1 levels, as determined by Immage, were slightly lower than those obtained by BNProspec (Y=0.86X - 0.02, CC=0.95). When Fibrotest-Actitest scores obtained with the two protein measurements were compared adjusting for apolipoprotein-A1 from Immage, the concordance rate was 0.903+/-0.096, with only 2/107 patients showing minimal discordance>0.10 for Fibrotest, and 1.00+/-0.06 for Actitest, with no discordance>0.10. Measurement of apolipoprotein-A1, included in the Fibrotest-Actitest score, depends on the equipment used. Such discordance is of little clinical consequence for liver fibrosis evaluation in hepatitis C virus patients.