Serum levels of YKL-40 and hyaluronic acid as noninvasive markers of liver fibrosis in haemodialysis patients with chronic hepatitis C virus infection L. L. Schiavon, 1 J. L. Narciso-Schiavon, 1 R. J. Carvalho Filho, 1 J. P. Sampaio, 1 J. O. Medina- Pestana, 2 V. P. Lanzoni, 3 A. E. B. Silva 1 and M. L. G. Ferraz 1 1 Hepatitis Section, Division of Gastroenterology, 2 Division of Nephrology, and 3 Department of Pathology, Federal University of Sao Paulo, Sao Paulo, Brazil Received January 2008; accepted for publication March 2008 SUMMARY. Hepatitis C virus (HCV) infection is highly pre- valent among end-stage renal disease (ESRD) patients undergoing haemodialysis and it is an important cause of morbidity and mortality in this population. The aim of this study was to evaluate the diagnostic value of YKL-40 and hyaluronic acid (HA) as noninvasive markers of liver fibrosis in 185 ESRD HCV-infected patients. Significant liver fibrosis was defined as METAVIR F2, F3 or F4 stages. Significant fibrosis was observed in 45 patients (24%). By univariate analysis, higher levels of YKL-40, HA, aspartate aminotransferase (AST), alanine aminotransferase (ALT) and gamma-glutamyltransferase (GGT) as well as reduced platelet count were associated with fibrosis. However, by multivariate analysis, only AST (P = 0.001), platelet count (P = 0.004) and HA (P = 0.042) were independently associated with significant fibrosis. For the prediction of significant fibrosis, the areas under receiver operating characterictic curve (AUROC) of the regression model (0.798) was significantly higher than the AUROC of YKL-40 (0.607) and HA (0.650). No difference was noted between the AUROC of the regression model and AST to platelet ratio index (APRI) (0.787). Values <8.38 of the regression model showed a negative predictive value of 94% and scores ‡9.6 exhibited a positive predictive value of 65%. If biopsy indication was restricted to scores in the intermediate range of the regression model, it could have been correctly avoided in 61% of the cases. In conclusion, APRI and a model based on AST, platelet count and HA showed better accuracy than YKL-40 and HA (when used solely) for the prediction of significant fibrosis in ESRD HCV-infected patients. Keywords: end-stage renal disease, haemodialysis, hepatitis C, hyaluronic acid, YKL-40. INTRODUCTION End-stage renal disease (ESRD) patients undergoing hae- modialysis represent a high-risk group for hepatitis C virus (HCV) infection, with reported prevalence ranging from 10% to as high as 59% [1,2]. In addition, HCV infection among dialysis patients has been associated with higher morbidity and mortality [3], as well as impaired survival and higher rates of graft loss after kidney transplantation [4–7]. More- over, given the contraindication of ribavirin and the higher frequency of adverse events and discontinuation rates, the efficacy of interferon-based therapy in haemodialysis indi- viduals with chronic HCV infection is limited [8,9]. For that reason, knowledge of the stage of liver fibrosis in those subjects is essential for therapeutic decisions and proper selection for kidney transplantation. Although liver biopsy is considered the gold standard in assessing liver fibrosis, it is an invasive technique with associated morbidity. Patient discomfort is relatively com- mon and pain in the right upper quadrant or right shoulder after liver biopsy is observed in about one-fourth of the cases [10]. Additionally, coagulation disorders are common among ESRD individuals and poses additional risk for pa- tients undergoing invasive procedures, including liver biopsy [11–13]. Furthermore, the heterogeneity of liver fibrosis in chronic hepatitis C [14,15], the potential inter- and intra- observer variability in the assessment of fibrosis [16], and the need for adequately sized specimens [17] are major dis- advantages of liver biopsy. Hence, the development of accurate noninvasive tests to predict liver fibrosis would be particularly useful in this population. Abbreviations: ESRD, end-stage renal disease; HA, hyaluronic acid; GGT, gamma-glutamyltransferase; AUROC, areas under receiver operating characteristic curve; APRI, AST to platelet ratio index; ROC, receiver operating characteristic curve; ALP, alkaline phosphatase. Correspondence: Leonardo L. Schiavon, MD, Rua Pedro de Toledo, 650 – 2 andar, Vila Clementino, Sa ˜o Paulo – SP, Brazil, Zip code: 04023-900. E-mail: leo-jf@uol.com.br Journal of Viral Hepatitis, 2008, 15, 666–674 doi:10.1111/j.1365-2893.2008.00992.x Ó 2008 The Authors Journal compilation Ó 2008 Blackwell Publishing Ltd