CLINICAL STUDIES Serum ferritin is a discriminant marker for both ¢brosis and in£ammation in histologically proven non-alcoholic fatty liver disease patients Pinelopi Manousou 1 , George Kalambokis 1 , Federica Grillo 2 , Jennifer Watkins 2 , Elias Xirouchakis 1 , Maria Pleguezuelo 1 , Gioacchino Leandro 1 , Vasiliki Arvaniti 1 , Giacomo Germani 1 , David Patch 1 , Vincenza Calvaruso 1 , Dimitri P. Mikhailidis 3 , Amar P. Dhillon 2 and Andrew K. Burroughs 1 1 The Royal Free Sheila Sherlock Liver Centre, Division of Surgery and Interventional Sciences, University College London, London, UK 2 Department of Histopathology, Royal Free Hospital, London, UK 3 Department of Clinical Biochemistry, Royal Free Hospital, London, UK Keywords ferritin – fibrosis – inflammation – NAFLD – NASH Abbreviations ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotranfrerase; BMI, body mass index; CRP, C-reactive protein; DM, diabetes mellitus; Fe, serum iron; g-GT, gamma-glutamyl transferase; NS, non-significant; TBL, total bilirubin; TIBC, total iron-binding capacity. Correspondence Andrew K. Burroughs, The Royal Free Sheila Sherlock Liver Centre, Division of Surgery and Interventional Sciences, University College London, Pond Street, Hampstead, London NW3 2QG, UK Tel: 144 20 74726229 Fax: 144 20 74726226 e-mail: andrew.burroughs@royalfree.nhs.uk Received 11 April 2010 Accepted 31 January 2011 DOI:10.1111/j.1478-3231.2011.02488.x Abstract Introduction: Differentiation between steatosis and non-alcoholic steatohepa- titis (NASH) in non-alcoholic fatty liver disease (NAFLD) is important as NASH progress to cirrhosis. No specific laboratory/imaging technique exists either to diagnose NASH or to select patients for liver biopsy. Patients and methods: We evaluated serum ferritin and the features of metabolic syndrome with respect to histological inflammation and/or fibrosis in NAFLD patients. The Kleiner scoring system was used to classify NAFLD in consecutive liver biopsies. One hundred and eleven patients: median age 52.6, 64 males, obesity 62, diabetes mellitus (DM) 58, arterial hypertension 26 and hyperlipidaemia 40%. Results: Histologically, 40.7 had fatty liver, 30.6% had borderline NASH, 28.7% had NASH and 11% had cirrhosis. Multivariate regression showed that diabetes, serum ferritin concentrations, body mass index (BMI) and AST were independently associated with NASH: together, the areas under the receiver operating characteristic (AUROC) was 0.91 (95% confidence interval 0.86–0.96); fibrosis was associated with ferritin concentrations and BMI: AUROC 0.87, portal inflammation with ferritin and DM: AUROC 0.82, while lobular inflammation was associated with BMI, DM and ferritin: AUROC 0.85. Conclusion: Serum ferritin concentrations and BMI are strongly asso- ciated with fibrosis, portal and lobular inflammation in NAFLD patients. Both ferritin and BMI are potential discriminant markers to select patients for liver biopsy and are associated with inflammation and fibrosis. Non-alcoholic fatty liver disease (NAFLD), encompass- ing simple steatosis, non-alcoholic steatohepatitis (NASH) to cirrhosis, is emerging as one of the most common liver disorders in developed countries (1). In the UK, 34 and 32% of patients with unexplained abnormal liver function tests are subsequently diagnosed as having NASH or fatty liver, respectively, by liver biopsy (2). Although fatty liver is now established to have a benign clinical course (3), NASH can be a cause of progressive liver fibrosis leading to cirrhosis, liver failure and hepatocellular carcinoma (4, 5). Severe fibrosis is noted in 7–49% of NASH patients and cirrhosis develops in 2–28% (6–10). Those at a high risk of progression include patients with histological advanced fibrosis (5, 11). The diagnosis of NAFLD is clinicopathological and, therefore, although the clinical component of the diagnosis is one of exclusion of significant alcohol ingestion and other causes of chronic liver disease, the histopathological examination is vital in diagnosing and staging suspected NAFLD (12). The important distinction is between fatty liver alone and NASH (i.e. inflammation with or without fibrosis). Given the prevalence of fatty liver, a liver biopsy cannot be considered for all patients. Indeed, in a recent study of an Italian group, fatty liver regressed in 50% of all cases examined and had a benign course (13). Therefore, there have been efforts to determine the predictive markers of fibrosis and/or steatohepatitis with the intention of avoiding biopsies, but to date, none have good clinical applicability (14). Liver International (2011) 730 c  2011 John Wiley & Sons A/S Liver International ISSN 1478-3223