CLINICAL STUDIES Serum adipokine levels predictive of liver injury in non-alcoholic fatty liver disease Maud Lemoine 1,2,3 , Vlad Ratziu 1,4 , Minji Kim 1,2 , Mustapha Maachi 1,2,5 , Dominique Wendum 1,6 , Franc ¸ois Paye 1,7 , Jean Philippe Bastard 1,2,5 , Raoul Poupon 1,3 , Chantal Housset 1,2,5 , Jacqueline Capeau 1,2,6 and Lawrence Serfaty 1,2,3 1 Universite Pierre et Marie Curie, Facult ´ e de M ´ edecine, Paris, France 2 Inserm, UMRS 938, Paris, France 3 AP-HP, Ho ˆ pital Saint-Antoine, Service d’H ´ epatologie, Paris, France 4 AP-HP, Ho ˆ pital Piti ´ e-Salp ´ etrie ` re, Service d’H ´ epatologie, Paris, France 5 AP-HP, Ho ˆ pital Tenon, Service de Biochimie et d’Hormonologie, Paris, France 6 AP-HP, Ho ˆ pital Saint-Antoine, Service d’Anatomo-pathologie, Paris, France 7 AP-HP, Ho ˆ pital Saint-Antoine, Service de Chirurgie Digestive, Paris, France Keywords adiponectin – adipokines – fatty liver – insulin- resistance – leptin – NAFLD – NASH – steatotis Correspondence Dr Lawrence Serfaty, Service d’H ´ epatologie, Ho ˆ pital Saint-Antoine, 184 rue du Faubourg Saint-Antoine, 75571 Paris Cedex 12, France Tel: 133 1 49 28 29 23 Fax: 133 1 49 28 21 07 e-mail: lawrence.serfaty@sat.aphp.fr Received 14 December 2008 Accepted 2 March 2009 DOI:10.1111/j.1478-3231.2009.02022.x Abstract Aims: The aim of this study was to determine whether serum levels of adipokines, including the ratio of serum adiponectin to leptin (A/L) levels could predict the severity of liver injury in patients with non-alcoholic fatty liver disease (NAFLD). Patients and methods: Fifty-seven patients with biopsy-proven non-alcoholic steatohepatitis (NASH) (mean age 51  12, sex ratio 1), 17 with simple steatosis (mean age 47  12, sex ratio 1.4) and 10 controls without steatosis (mean age 51  11, sex ratio 4) were investigated. In all subjects, serum concentrations of triglycerides, ultrasensitive C reactive protein, leptin, adiponectin, soluble tumour necrosis factor receptor 1, interleukin (IL)-6 and Homeostasis Model Assessment Method (HOMA) were measured. Hepatic expression of adiponectin and its two receptors was assessed by quantitative reverse transcriptase polymerase chain reaction. Results: Body mass index (BMI) and HOMA were correlated positively with leptin levels (r = 0.44 and 0.28 respectively) and negatively with the A/L ratio (r = 0.51 and 0.41 respectively). Independent parameters associated with NASH vs steatosis were HOMA 4 3 [odds ratio (OR) = 6.9] and A/L ratio o 1.4 10 3 (OR = 5.2). The combination of HOMA with A/L ratio showed an area under the receiver operating characteristic curve of 0.82 for distinguishing between NASH and steatosis. Extensive portal fibrosis was present in 17 (23%) patients with NAFLD. Three independent parameters were associated with fibrosis: age (OR = 1.1), BMI (OR = 1.3) and high IL-6 levels (OR = 1.6). The hepatic expression of adiponectin receptor 2 was significantly higher in patients with NASH compared with controls and was related with necroin- flammatory injury. Conclusions: This study shows that in patients with NAFLD, the combination of HOMA with A/L ratio may be a useful non- invasive approach to appreciate the severity of liver damage. Non-alcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver disease in Western coun- tries, with an estimated incidence of 10–24% in the general population (1). NAFLD, frequently associated with insulin resistance (IR), represents a feature of the metabolic syndrome (2). Most often, fatty liver has a benign course, but in approximately 20–30% of the cases, histological signs of fibrosis and necroinflammatory injury that define non-alcoholic steatohepatitis (NASH) are present (3). Patients with NASH are at higher risk of developing fibrosis, cirrhosis and hepatocellular carcinoma (3–5). The pathogenesis of NASH remains unclear. It has been proposed that IR represents a first hit, leading to steatosis, and that pro-inflammatory cytokines could cause a second hit, leading to NASH (6). Both adipo- kines, i.e., fat-secreted proteins such as leptin or adipo- nectin and fat- or liver-derived cytokines such as tumour necrosis factor a (TNFa) and interleukin 6 (IL-6) could act as pathogenic factors in NASH (7). These factors are able to modify insulin sensitivity (8). In addition, TNFa , IL-6 and leptin have been shown to exert pro-inflamma- tory effects and adiponectin has been shown to exert anti-inflammatory effects at the liver level (7). It has also been suggested that leptin could be involved in fibrogen- esis (9) while adiponectin was reported to have antifi- brogenic and antisteatogenic effects in the liver (10, 11). Liver International (2009) c  2009 John Wiley & Sons A/S 1431 Liver International ISSN 1478-3223