Reference values for the triglyceride to high-density lipoprotein ratio and its association with cardiometabolic diseases in a mixed adult population: The ELSA-Brasil study

Deborah de Farias Lelis,João Vitor S Calzavara,Raul D Santos,Andrei C Sposito,Rosane Härter Griep,Sandhi Maria Barreto,Maria Del Carmen B Molina,Maria Inês Schmidt,Bruce B Duncan,Isabella Bensenor,Paulo Andrade Lotufo,José Geraldo Mill,Marcelo Perim Baldo
DOI: https://doi.org/10.1016/j.jacl.2021.07.005
Abstract:Background: Among several lipid ratios available, the triglyceride/HDL-cholesterol (TG/HDL-C) may detect individuals at risk of cardiometabolic diseases. However, its reference values for different ethnicities are not well established. Objective: To define sex- and ethnicity-specific reference values for TG/HDL-C ratio in a large sample of healthy multiethnic adults and test its association with cardiometabolic conditions. Methods: An apparently healthy sample (n = 2,472), aged 35-74, free of major cardiovascular risk factors, was used to generate the reference values for the TG/HDL-C. Exclusion criteria were diabetes, elevated blood pressure, obesity, hypercholesterolemia, severe hypertriglyceridemia, and smoking history. Cut-offs based on the reference values were tested in the whole ELSA Brasil study (n = 13,245), stratified by sex and ethnicity, to identify cardiometabolic conditions. Results: TG/HDL-C ratio was higher in men than women, and did not change significantly with age, regardless of sex and ethnicity. Also, black individuals showed lower levels of TG/HDL-C as compared to other ethnic groups. ROC curve showed that the cut-off based on the 75th percentile displayed better sensitivities and specificities for men and women, regardless of ethnicity. Also, the sex- and ethnicity-specific cut-offs based on the 75th percentile were significantly associated with all tested cardiometabolic conditions (hypertension, diabetes, obesity, metabolic syndrome, and insulin resistance). Also, we observed that the use of a single sex-specific cut-off (men: 2.6; women: 1.7) could be used for the different ethnicities with good reliability. Conclusion: The defined TG/HDL-C cut-offs (men: 2.6; women: 1.7) are reliable and showed good clinical applicability to detect cardiometabolic conditions in a multiethnic population.
What problem does this paper attempt to address?