Abstract
OBJECTIVE: There is limited evidence on the relationship between remnant cholesterol (RC) and low-density lipoprotein cholesterol (LDL-C) discordance and type 2 diabetes or prediabetes. This study aimed to investigate the association between RC and LDL-C discordance and type 2 diabetes or prediabetes in the general US adult population. METHODS: Data from 19,604 participants in the National Health and Nutrition Examination Surveys (NHANES) (2005-2018) were analyzed. The percentile difference between RC and LDL-C was used to define discordance. Logistic regression models were performed to assess the relationships between RC, LDL-C, RC and LDL-C percentile difference, LDL-C and RC discordant. Nonlinear relationships were explored using restricted cubic splines. Mediation analysis was conducted to evaluate the direct and indirect relationships between abdominal obesity and type 2 diabetes through RC. RESULTS: RC was non-linearly and J-shaped correlated with type 2 diabetes and prediabetes, whereas LDL-C was non-linearly and U-shaped correlated with type 2 diabetes and was S-shaped correlated with prediabetes. Compared to concordant participants, those with discordantly low RC population had lower type 2 diabetes odds, while those with the discordantly high RC population had higher type 2 diabetes and prediabetes odds. When a clinical LDL-C cut-off of 2.60 mmol/L was applied, participants in the high LDL-C and low RC group had the lowest odds of type 2 diabetes, while those in the low LDL-C and high RC group had the highest odds. Significant interactions between LDL-C and RC discordance and factors such as age, MetS, and lipid-lowering medications were observed in relation to type 2 diabetes odds. RC was found to mediate 15.86% of the association between abdominal obesity and type 2 diabetes, and 16.22% of the association between abdominal obesity and prediabetes in the non-diabetes population. CONCLUSIONS: Discordantly high RC was associated with higher odds of type 2 diabetes and prediabetes, while discordantly low RC was associated with lower odds of type 2 diabetes.