• Joanna Górka-Dynysiewicz, Monika Pazgan-Simon, and Jolanta Zuwała-Jagiełło.
• Department of Pharmaceutical Biochemistry, Wroclaw Medical University, Wrocław, Poland. joanna.gorka-dynysiewicz@umed.wroc.pl
• Pol. Arch. Med. Wewn. 2020 Feb 27; 130 (2): 112-120.

IntroductionNoninvasive methods are increasingly used in the clinical assessment of patients with chronic hepatitis C (CHC).ObjectivesWe aimed to develop a predictive model for the evaluation of significant fibrosis in patients with CHC, based on serum biomarkers. We compared the accuracy of our model in detecting significant fibrosis with currently known markers / models of fibrosis (such as the aspartate aminotransferase to platelet ratio index [APRI], the Fibrosis‑4 [FIB-4] score, and the Forns index).Patients And MethodsA total of 242 patients with CHC not receiving antiviral treatment were divided into 2 groups: training group (n = 150) and validation group (n = 92). Significant fibrosis was defined as F2 or higher on the Meta‑analysis of Histological Data in Viral Hepatitis (METAVIR) scale.ResultsMultivariable analysis revealed that age (P <0.001), pentraxin 3 (PTX3) levels (P = 0.009), γ‑glutamyl transpeptidase (GGT) to platelet count (PLT) ratio (P = 0.08), and hyaluronic acid levels (HA) (P = 0.07) were independent predictors of significant fibrosis. Based on that, we developed a model for predicting significant fibrosis: Pentra score = 0.176 × PTX3 (ng/ml) + 0.522 × HA (ng/ml) + 0.29 × GGT (IU/l) to PLT (×109/l) ratio + 0.14 × age (years) - 3.9346. Then, we compared our model with the biomarkers and models currently used to predict liver fibrosis. The Pentra score yielded the largest area under the receiver operating characteristic curve for predicting significant fibrosis in the training and validation groups (0.894 and 0.867, respectively). It also had the highest diagnostic accuracy in both groups (90.6% and 87.0%, respectively).ConclusionsOur model for detecting significant fibrosis in patients with CHC using pentraxin 3 and other serum biomarkers compares well with the existing and previously published indices. However, further validation in larger cohorts is needed.

35 keywords...

hide…