Abstract
BACKGROUND: It is well established that the consumption of trans-fatty acids (TFAs) can increase the incidence of total mortality, cardiovascular disease, cancer, and diabetes. However, there are still no demographic studies on the effects of circulating TFA isoforms on the albumin-creatinine ratio (ACR), an early marker of chronic kidney disease. Our goal was to explore the possible relationships between TFAs and ACR. METHODS: In this study, complete TFAs and urinary ACR data were collected from the National Health and Nutrition Examination Survey (NHANES) (2009-2010 and 1999-2000 cycles). The independent linear relationships between different circulating TFA isoforms and the ACR were examined by performing multivariable linear regression models. Machine learning was used to analyze the contribution of the different TFA isoforms to the ACR. To assess the nonlinearity of the relationship, smooth curve fitting and an analysis of threshold effect were performed, and a stratified analysis was conducted to identify possible susceptible populations. RESULTS: Our analysis included a total of 3785 individuals. Elaidic acid, linolelaidic acid, and sum TFAs were shown to be positively associated with the ACR after full adjustment by weighted multivariable regression analysis. In the subgroup analysis, the positive associations were maintained in participants with hypertension and without diabetes. In the XGBoost model of the ACR, Sum TFAs were found to be the most crucial factor. In addition, smooth curve fitting showed that there was a nonlinear relationship between the different TFAs and the ACR, and there was a saturation point. CONCLUSIONS: Our study demonstrated that TFA isoforms were positively and independently correlated with urinary albumin excretion, especially in participants with hypertension and without diabetes. This suggested that reducing trans fatty acid intake may reduce the risk of renal events.