Obesity Facts (Jan 2024)

Triglycerides Mediate the Influence of Body Mass Index on Non-Alcoholic Fatty Liver Disease in a Non-Obese Chinese Population with Normal Low-Density Lipoprotein Cholesterol Levels

  • Xixi Han,
  • Jingwen Kong,
  • Hemin Zhang,
  • Yuan Zhao,
  • Yafeng Zheng,
  • Chao Wei

DOI
https://doi.org/10.1159/000536447

Abstract

Read online

Introduction: Over 25% of the world’s population has non-obese or lean non-alcoholic fatty liver disease (NAFLD), and the prevalence is higher than average in Asia. The present study focused on the relationship between body mass index (BMI) and non-obese NAFLD in non-overweight people in China, particularly the influence of triglycerides (TG) in the pathogenesis of non-obese NAFLD. The findings suggest new treatments for NAFLD patients with normal BMI, as well as provide an early warning system for the understanding and prevention of NAFLD in non-obese patients. Methods: This cross-sectional study enrolled 159,959 Chinese subjects with BMI <24 kg/m2 and normal levels of low-density lipoprotein cholesterol (LDL-c). The average age was 40.21 ± 13.88 years, and males accounted for 45.7%. A total of 15,907 (9.94%) patients with NAFLD were diagnosed by ultrasonography. Biochemical indicators were measured using an automated analyzer (Abbott AxSYM). The BMI (kg/m2) was calculated from the weight (kg)/height in square meters (m2). The BMI quartile was used as the column-stratified variable to determine the baseline distribution, and logistic regression analysis was used to assess the relationship between NAFLD and its risk factors, with multiple logistic regression used to assess the relationships between BMI or TG and NAFLD and multivariate linear regression used to analyze the association between BMI and TG, while mediation analysis was used to assess the mediation effect of TG. Results: After adjustment of all covariates, the odds ratios were 1.788 (95% CI: 1.749–1.829; p < 0.00001) and 1.491 (95% CI: 1.451–1.532; p < 0.00001) for the association between BMI and TG with NAFLD incidence. The multivariate linear regression coefficient of BMI and TG was β = 0.027 (95% CI: 0.023–0.030; p < 0.00001). Mediation analysis showed that BMI contributed to 10.81% of lean NAFLD with a mediation effect of 2.98%. Conclusion: In a Chinese population with BMI <24 kg/m2 and normal LDL-c levels, BMI and TG were found to be independent predictors of NAFLD. The direct effect of BMI on non-obese NAFLD was 10.41%. The TG level was found to partially mediate the association.

Keywords