Abstract
BACKGROUND: Non-alcoholic fatty liver disease (NAFLD) is a prevalent condition that often goes unrecognized in the population, and many risk factors for this disease are not well understood. Glyphosate (GLY) is one of the most commonly used herbicides worldwide, and exposure to this chemical in the environment is significant. However, studies exploring the association between GLY exposure and NAFLD remain limited. Therefore, the aim of this study was to assess the association between urinary glyphosate (uGLY) level and fatty liver index (FLI) using data from the National Health and Nutrition Examination Survey (NHANES), which includes uGLY measurements. METHODS: The log function of uGLY was converted and expressed as Log(e)(uGLY) with the constant "e" as the base and used for subsequent analysis. The association between Log(e)(uGLY) (the independent variable) level and FLI (the dependent variable) was assessed by multiple linear regression analysis. Smoothing curve fitting and a generalized additive model were used to assess if there was a nonlinear association between the independent and the dependent variables. A subgroup analysis was used to find susceptible individuals of the association between the independent variable and the dependent variable. RESULTS: A final total of 2238 participants were included in this study. Participants were categorized into two groups (< -1.011 and ≥ -1.011 ng/ml) based on the median value of Log(e)(uGLY). A total of 1125 participants had Log(e)(uGLY) levels ≥ -1.011 ng/ml and higher FLI. The result of multiple linear regression analysis showed a positive association between Log(e)(uGLY) and FLI (Beta coefficient = 2.16, 95% CI: 0.71, 3.61). Smoothing curve fitting and threshold effect analysis indicated a linear association between Log(e)(uGLY) and FLI [likelihood ratio(LLR) = 0.364]. Subgroup analyses showed that the positive association between Log(e)(uGLY) and FLI was more pronounced in participants who were female, aged between 40 and 60 years, had borderline diabetes history, and without hypertension history. In addition, participants of races/ethnicities other than (Mexican American, White and Black) were particularly sensitive to the positive association between Log(e)(uGLY) and FLI. CONCLUSIONS: A positive linear association was found between Log(e)(uGLY) level and FLI. Participants who were female, 40 to 60 years old, and of ethnic backgrounds other than Mexican American, White, and Black, deserve more attention.