Concordance of non-invasive serology-based scoring indices and transient elastography for liver fibrosis and cirrhosis in chronic hepatitis C.

Aim To assess concordance of eight frequently used serology-based scoring indices for liver fibrosis and cirrhosis with transient elastography (TE) in chronic hepatitis C (CHC) patients in order to determine serum indices with the highest concordance and clinical usability in clinical practice. Methods In this prospective study, 63 CHC patients were included and TE results were compared with eight non-invasive indices. The diagnostic performance of these tests was assessed using receiver operating characteristic curves with kappa index calculated for the concordance analysis. Results Median age of 63 patients was 54 years (interquartile range: 42 to 63); 27 (42.9%) were females. According to areas under the Receiver Operating Characteristics (AUROC), the best performing serum markers for significant liver fibrosis (METAVIR ≥F2), advanced liver fibrosis (≥F3) and cirrhosis (F4) determined by TE measurements (≥7.1kPa, ≥9.5kPa and ≥12kPa, respectively) were Fibrotest (AUROC=0.727 for ≥F2) and FIB-4 score (AUROC=0.779 for ≥F3 and AUROC=0.889 for F4). Fibrotest cut-off at >0.50 was concordant with TE for presence of significant fibrosis in 30 (out of 45; 66.7%), FIB-4 cut-off at <1.45 was concordant for absence of significant fibrosis in 13 (out of 18; 72.2%) and Goeteborg University Cirrhosis Index (GUCI) cut-off at >1 was concordant for presence of cirrhosis in 16 (out of 22; 72.7%) patients, but not for exclusion of cirrhosis. Conclusion Serology-based scoring indices had moderate overall concordance with TE. We propose that FIB-4 score, Fibrotest and GUCI be used in routine practice to exclude and diagnose significant fibrosis and diagnose cirrhosis, respectively.