Predictive cut-off values for the triglyceride-to-high-density lipoprotein cholesterol ratio to predict metabolic syndrome in the middle-aged Korean population

Abstract

Background and aimsThe triglyceride-to-high density lipoprotein cholesterol (TG/HDL) ratio is associated with insulin resistance related diseases, including metabolic syndrome (MetS). However, specific TG/HDL values that can predict MetS have not been well identified. In this study, we analyzed both cross-sectional and longitudinal data from two national Korean datasets to obtain TG/HDL cut-off values that can identify MetS and predict its occurrence. Methods and resultsTo distinguish the presence and occurrence of MetS, the cut-off values were determined using the maximum F-score calculated through a logistic regression analysis. To predict new-onset MetS within 10 years, Cox proportional hazard models were used to consider the time of occurrence. The TG/HDL cut-off values of 3.97, 3.24, and 3.24 were optimal for identifying current MetS and predicting new-onset MetS within 10 years and five years, respectively, in Korean men. In Korean women, the optimal values for each task were 3.18, 2.38, and 2.26, respectively. ConclusionsWe suggest the TG/HDL ratio as a potential candidate predictor for MetS. Therefore, we anticipate that future studies will apply individual lipid levels as well as their combinatory values to establish models that predict the prevalence and occurrence of MetS, diabetes, and cardiovascular disease.