The purpose of this study was to assess the influence of success rate and interquartile range on the accuracy of transient elastography for the diagnostic of fibrosis in hepatitis C virus infection. Two-hundred fifty-four consecutive patients had liver stiffness measurements and liver biopsy of at least 15 mm. Discordances of at least two stages between transient elastography and histological assessment were observed in 28 cases (11%). Factors of discordance were assessed by comparing the 28 misclassified cases with the 226 others. In multivariate analysis, fibrosis stage (F0-F2 versus F3-F4) and the ratio interquartile range/median value of liver stiffness measurement (IQR/M) were associated with discordances (P <or= 0.05). The most significantly discriminant cutoff value was 0.21. For IQR/M < 0.21 versus IQR/M >or= 0.21, discordances of at least two stages of fibrosis were respectively observed in 10 of 135 cases (7.4%) versus 18 of 119 cases (15.1%) (P <or= 0.05). In patients with IQR/M >or= 0.21 versus IQR/M < 0.21, for the diagnosis of liver fibrosis F >or= 2, F >or= 3, F = 4, areas under the receiver operating characteristic curve (AUROCs) were 0.80 (95% confidence interval [CI], 0.73-0.89) versus 0.81 (95% CI, 0.70-0.90), (P = NS); 0.80 (95% CI, 0.72-0.88) versus 0.89 (95% CI, 0.83-0.95) (P = 0.04); and 0.86 (95% CI, 0.77-0.94) versus 0.95 (95% CI, 0.92-0.99) (P = NS). No association was found between success rate and discordance.
Conclusion: IQR/M is a factor of overestimation of liver fibrosis, and the most discriminant cutoff value is 0.21. Success rate is not a factor of accuracy for the diagnosis of hepatic fibrosis.