Predictors for incidence and remission of NAFLD in the general population during a seven-year prospective follow-up Shira Zelber-Sagi 1,2,⇑,  , Roni Lotan 1,3,  , Amir Shlomai 1 , Muriel Webb 1 , Gil Harrari 2 , Assaf Buch 1,3 , Dorit Nitzan Kaluski 3 , Zamir Halpern 1,3 , Ran Oren 1,3 1 Department of Gastroenterology and Liver diseases, Tel Aviv Sourasky Medical Center, Israel; 2 School of Public Health, Faculty of Social Welfare and Health Sciences, University of Haifa, Israel; 3 Sackler Faculty of Medicine, Tel-Aviv University, Israel Background & Aims: Data on the incidence and remission rates of non-alcoholic fatty liver disease (NAFLD) as well as predictive factors are scant. This study aims at evaluating NAFLD’s epidemi- ology in prospective follow-up of individuals sampled from the general population. Methods: Evaluation of metabolic parameters and ultrasono- graphic evidence of NAFLD was performed in 213 subjects, with no known liver disease or history of alcohol abuse. The evaluation was performed at baseline and after a 7-year period by identical protocols. Results: Of the 147 patients who did not have NAFLD at baseline, 28 (19%) were found to have NAFLD at a 7-year follow-up. Base- line BMI, HOMA score, blood cholesterol, triglycerides, leptin lev- els, and weight gain (5.8 ± 6.1 vs. 1.4 ± 5.5 kg, p <0.001) were significantly higher and adiponectin was lower among those who developed NAFLD at 7-year follow-up, compared with those who remained NAFLD-free. However, only weight gain and base- line HOMA were independent predictors for the development of NAFLD. Of the 66 patients who were found to have NAFLD at baseline, as many as 24 patients (36.4%) had no evidence of NAFLD at 7 years. Weight loss of 2.7 ± 5.0 kg was significantly associated with NAFLD remission. Moreover, there was a 75% remission rate among NAFLD patients who lost 5% or more from their baseline weight. Conclusions: Among the general population, weight gain, and baseline insulin resistance are predictors for NAFLD incidence. One third of NAFLD patients may have remission of disease within a 7-year follow-up, mostly depending on modest weight reduction. Ó 2012 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Introduction Non-alcoholic fatty liver disease (NAFLD) is one of the most common liver diseases in the Western world nowadays affecting 20–30% of adults in developed countries and up to 80% of obese people [1]. A recent study indicates a higher prevalence of NAFLD, than previously estimated, of 46% and 74% among the general pop- ulation and among patients with diabetes, respectively [2]. Obesity and insulin resistance are two major risk factors for NAFLD [3], and patients with NAFLD are prone to develop type 2 diabetes and car- diovascular diseases [4–6]. NAFLD is a heterogeneous disease that may range from a relatively mild and subtle disease (hepatic steatosis) to a much more active and progressive disease, in which significant liver inflammation and damage are involved, desig- nated non-alcoholic steatohepatitis (NASH). Patients with NASH are at a substantial risk of developing a severe liver disease result- ing in cirrhosis and even hepatocellular carcinoma (HCC) [7,8]. Therefore, primary prevention and treatment of NAFLD are of par- amount importance with a potential to save lives of millions of patients worldwide. Moreover, a detailed knowledge about the incidence and the natural course of NAFLD is essential for a proper design of screening and follow-up programs. Although several studies have attempted to evaluate the inci- dence and the natural course of NAFLD, it is still difficult to draw firm conclusions regarding people from the general population who do not drink alcohol and for whom known liver disease was ruled-out. A Japanese study performed on a convenience sample of employees attending a medical health check-up found the met- abolic syndrome to be a strong predictor for NAFLD [9], during a relatively short follow-up of only 1.1 years. An analysis performed on patients from the Dionysos study [10], revealing ethanol intake as the only risk factor for NAFLD incidence and remission during a long follow-up of 8.5 years, included only patients with elevated liver enzymes and many patients with a history or concurrent ethanol consumption, therefore raising concerns that this study population does not accurately reflect a general naïve population with primary NAFLD. The same concerns are true for a recent study assessing the natural course of NAFLD over a 3-year period among a hospital-based cohort of patients [11]. Therefore, we aimed at evaluating a representative sample of the general population without history or concurrent significant ethanol consumption and with no apparent liver disease for the Journal of Hepatology 2012 vol. 56 j 1145–1151 Keywords: Weight change; Insulin resistance; Predictors; Hepatorenal index; Ultrasonography. Received 19 August 2011; received in revised form 6 December 2011; accepted 9 December 2011; available online 13 January 2012 ⇑ Corresponding author. Address: Department of Gastroenterology, Tel Aviv Sourasky Medical Center, 64239 Tel-Aviv, Israel. Tel.: +972 3 6973984; fax: +972 3 6974622. E-mail address: zelbersagi@bezeqint.net (S. Zelber-Sagi).   The authors contributed equally to this work. Abbreviations: NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic st- eatohepatitis; HCC, hepatocellular carcinoma; HOMA, homeostasis model assess- ment; HRI, hepato-renal index; ALT, alanine aminotransferase; BMI, body mass index; PA, physical activity. Research Article