Nonalcoholic fatty liver disease without overlapping metabolic-associated fatty liver disease and the risk of incident type 2 diabetes.

Abstract

Re-classifying NAFLD as metabolic-associated fatty liver (MAFLD) has been proposed. While some people fulfil criteria for NAFLD, they do not have MAFLD; and whether NAFLD-only subjects have increased the risk of type 2 diabetes remains unknown. We compared risk of incident T2D in individuals with: (a) NAFLD-only; and (b) MAFLD, to individuals without fatty liver, considering effect modification by sex. 246 424 Koreans without diabetes or a secondary cause of ultrasound-diagnosed hepatic steatosis were studied. Subjects were stratified into: (a) NAFLD-only status and (b) NAFLD that overlapped with MAFLD (MAFLD). Cox proportional hazards models with incident T2D as the outcome were used to estimate hazard ratios (HRs) for: (a) and (b). Models were adjusted for time-dependent covariates, and effect modification by sex was analysed in subgroups. A total of 5439 participants had NAFLD-only status and 56 839 met MAFLD criteria. During a median follow-up of 5.5 years, 8402 incident cases of T2D occurred. Multivariable-adjusted HRs (95% CI) for incident T2D comparing NAFLD-only and MAFLD to the reference (neither condition) were 2.39 (1.63-3.51) and 5.75 (5.17-6.36) (women), and 1.53 (1.25-1.88) and 2.60 (2.44-2.76) (men), respectively. The increased risk of T2D in the NAFLD-only group was higher in women than in men (p for interaction by sex <0.001) and consistently observed across all subgroups. Risk of T2D was increased in lean participants regardless of metabolic dysregulation (including prediabetes). NAFLD-only participants without metabolic dysregulation and the criteria for MAFLD are at increased risk of developing T2D. This association was consistently stronger in women than in men.