Nonalcoholic fatty liver disease (NAFLD) and associated mortality in individuals with type 2 diabetes, pre-diabetes, metabolically unhealthy, and metabolically healthy individuals in the United States

Background: The prevalence of nonalcoholic fatty liver disease (NAFLD) is high among subjects with type 2 diabetes (T2D). However, the prevalence and outcomes of NAFLD among individuals with pre-diabetes (Pre-D) and metabolically healthy and metabolically unhealthy individuals without T2D are not known. Our aim was to assess prevalence and mortality of NAFLD among these four groups.Methods: The Third National Health and Nutrition Examination Survey (NHANES) III (1988-1994) with mortality data (follow up to 2019) via linkage to the National Death Index was utilized. NAFLD was defined by ultrasound and absence of other liver diseases and excess alcohol use. Pre-D was defined as fasting plasma glucose values of 100-125 mg/dL and/or HbA1c level between 5.7 %-6.4 % in the absence of established diagnosis of T2D. Metabolically healthy (MH) was defined if all of the following criteria were absent: waist circumference of >102 cm (men) or > 88 cm (women) or BMI of >30; blood pressure (BP) > 130/85 mmHg or using BP-lowering medication; triglyceride level > 150 mg/dL or using lipid-lowering medication; lipoprotein cholesterol level of <40 mg/dL (men) or < 50 mg/dL (women); homeostasis model assessment of insulin resistance (HOMA-IR) score > 2.5; C-reactive protein (CRP) level of >2 mg/L; Pre-D and T2D. Metabolically unhealthy (MU) individuals were defined as the presence of any component of metabolic syndrome but not having Pre-D and T2D. Competing risk analyses of cause-specific mortality were performed.Findings: 11,231 adults (20-74y) were included: mean age 43.4 years; 43.9 % male; 75.4 % white, 10.8 % Black, and 5.4 % Mexican American, 18.9 % NAFLD, 7.8 % T2D; 24.7 % Pre-D; 44.3 % MU; and 23.3 % in MH individuals. In multivariable adjusted logistic model, as compared to MH individuals, the highest risk of having NAFLD were in T2D individuals (Odd Ratio [OR] = 10.88 [95 % confidence interval: 7.33-16.16]), followed by Pre-D (OR = 4.19 [3.02-5.81]), and MU (OR = 3.36 [2.39-4.71]). During a median follow up of 26.7 years (21.2-28.7 years), 3982 died. NAFLD subjects had significantly higher age-adjusted mortality than non-NAFLD (32.7 % vs. 28.7 %, p < .001). Among subjects with NAFLD, the highest age-standardized cumulative mortality was observed among those with T2D (41.3 %), followed by with Pre-D (35.1 %), MU subjects (30.0 %), and MH subjects (21.9 %) (pairwise p-values<.04 vs. MH). Multivariable adjusted cox models showed that NAFLD with T2D had a higher risk of all-causes and cardiac-specific deaths (Hazard Ratio [HR] = 4.71 [2.23-9.96] and HR = 20.01 [3.00-133.61]), followed by NAFLD with Pre-D (HR = 2.91 [1.41-6.02] and HR = 10.35 [1.57-68.08]) and metabolically unhealthy NAFLD (HR = 2.59 [1.26-5.33] and HR = 6.74 [0.99-46.03]) compared to metabolically healthy NAFLD. In addition to older age, independent predictors of mortality among NAFLD with T2D included high CRP, CVD, CKD, high FIB-4, and active smoking. Similarly, among NAFLD with Pre-D, high CRP, CKD, CVD, hypertension, and active smoking were associated with mortality. Finally, CVD and active smoking were predictors of mortality among metabolically unhealthy NAFLD, and active smoking was the only mortality risk among metabolically healthy NAFLD subjects. Interpretation: Metabolic abnormality impacts both prevalence and outcomes of subjects with NAFLD.