Correlation of nonalcoholic fatty liver disease at different ages of onset with new-onset diabetes mellitus

Abstract

Objective: To investigate the effect of nonalcoholic fatty liver (NAFLD) at different ages of onset with new-onset diabetes mellitus. Methods: The cohort study was conducted in Kailuan Group Company. Active and retired employees were used as study subjects. After excluding NAFLD diagnosed at baseline, previous history of diabetes mellitus, and long-term history of heavy drinking, 43 317 cases were finally included in the cohort. The study subjects were divided into five groups according to age (<30 years old as group 1, 30-39 years old as group 2, 40-49 years as group 3, 50-59 years as group 4, and ≥60 years as group 5). The prevalence and incidence density of new-onset diabetes mellitus were compared between each NAFLD and non-fatty liver population group. The effect of NAFLD at different ages of onset with new-onset diabetes mellitus was analyzed by multivariate Cox's regression model. Statistical analysis was performed using one-way ANOVA, χ2 test or multivariate Cox's regression model. Results: The prevalence and incidence density of diabetes mellitus was significantly higher in NAFLD than non-fatty liver population. The prevalence of diabetes mellitus in different age groups were 6.45%, 6.88%, 9.94%, 10.83%, and 11.43%, respectively. The incidence density of each age group was 9.21/1 000 person-years, 11.10/1 000 person-years, 16.17/1 000 person-years, 18.72/1 000 person-years, and 22.13/1 000 person-years, and the differences were statistically significant (P<0.001). Multivariate Cox's regression model result showed that after adjusting for confounding factors such as gender, systolic blood pressure, and fasting blood glucose, the HRs (95%CI) for diabetes mellitus in each age group were 3.992 (1.897, 8.400), 2.321 (1.589, 3.392), 2.041 (1.667, 2.500), 2.007 (1.708, 2.360), and 1.908 (1.570, 2.319), and the differences were statistically significant (P＜0.001). Conclusion: Newly developed NAFLD is an independent risk factor for new-onset diabetes mellitus. Early exposure to NAFLD increases the risk of developing diabetes mellitus compared with the same age group. Younger age of onset of NAFLD should be given attention and active treatment.