Objective: Non-alcoholic fatty liver disease (NAFLD), a condition associated with multiple metabolic abnormalities, is frequently observed in normal weight individuals (lean NAFLD). The metabolic consequences of lean NAFLD, however, are not well characterized. Thus, this study aimed to evaluate the risk of incident diabetes in lean NAFLD. Methods: This is a cohort study of 51,463 adults without diabetes, history of liver disease or cancer at baseline who participated in a regular health

more »

... ular health screening exam. Fatty liver was diagnosed by ultrasonography. The study outcome was the development of diabetes during follow-up. Results: During 236,446.6 person-years of follow-up (median follow-up of 4.0 years), 5,370 participants developed diabetes. In fully-adjusted models, the hazard ratios (HRs) for incident diabetes comparing lean participants with NAFLD, overweight/obese participants without NAFLD, and overweight/obese participants with NAFLD to lean participants without NAFLD, were 1.18 (95% CI 1.03–1.35), 1.06 (0.98–1.14), and 1.45 (1.34–1.57), respectively. The fully-adjusted HR for incident diabetes for lean NAFLD participants with low NAFLD fibrosis score (NFS) (〈-1.455) and with intermediate to high NFS (≥-1.455) compared to lean participants without NAFLD were 1.32 (1.14–1.53) and 2.73 (2.10–3.55), respectively. Conclusions: In this large cohort study, the presence and severity of NAFLD in normal-weight adults was associated with an increased incidence of diabetes independently of established risk factors. Indeed, isolated lean NAFLD was a stronger risk factor for incident diabetes than the presence of overweight/obesity without NAFLD. Subjects with lean NAFLD require careful monitoring for the development of metabolic abnormalities.