Background: Percutaneous liver biopsy is reference for rating the fibrosis stage both in hepatitis in chronic hepatitis C (CHC) and B (CHB). Therefore biopsy is an invasive procedure and has disadvantages such assembling errors, patient discomfort and risk of major complications. The aim of our study was to assess the diagnostic performance of non-invasive (NI) tests APRI, modified APRI (m-APRI), FIB-4, Fibro index (FI)in CHB and CHC compared to biopsy.
Methods: In this study 236 patients with CHB/CHC who experienced ultrasound guided liver biopsy between January 2007 and May 2014 at Katip Celebi University Ataturk Training and Research Hospital were enrolled. Histological grading of necroinflammation and fibrosis were performed according to Knodell and ISHAK scoring systems. APRI, m-APRI, FIB-4, FI scores were calculated according to formulas. Statistical analyses were performed by using SPSS packed programme.
Results: Patients were evaluated in two groups according to the viral etiology.183 (%77,54) patients were CHB and 53 (%22,46) of them were CHC. 136 (%57.9) of them were male and 99 (%42,1) were female. Number of patients for F0, F1, F2, F3, F4, F5, F6 fibrosis grade were 57 (24.2%), 59 (25%), 45 (19,1%), 35 (14,8%), 24 (10.2%), 10 (4,2%), 6 (2.5%) respectively.Results of NI tests were compared for each fibrosis stages of biopsy results in CHB/CHC patients. In patients with grade 0-1-2 fibrosis, m-APRI and FIB-4; in patients with grade 3-4 FIB-4 scores correlated with biopsy scores (Table 1). Optimal cutoffs were found with Youden Method, which maximizes the sum of sensitivity and specificity to assess difference in cutoffs depending to the viral etiology. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) were calculated for each fibrosis stages and viral etiology (Table 2).
Conclusion: For all NI tests, NPVs were statistically significant but PPVs were highly variable in CHB/CHC patients compared to biopsy. FI showed better performance with higher sensitivity and specificity rates compared to other NI tests.
A. Nemli, None
S. Ozdemir, None
S. Sari, None
S. Ural, None