Two-Sample Mendelian Randomization Study of Lipid levels and Ischemic Heart Disease
- Author:
Su Hyun LEE
1
;
Ji-Young LEE
;
Guen hui KIM
;
Keum Ji JUNG
;
Sunmi LEE
;
Hyeon Chang KIM
;
Sun Ha JEE
Author Information
- Publication Type:Original Article
- From:Korean Circulation Journal 2020;50(10):940-948
- CountryRepublic of Korea
-
Abstract:
Background and Objectives:Associations between blood lipids and risk of ischemic heart disease (IHD) have been reported in observational studies. However, due to confounding and reverse causation, observational studies are influenced by bias, thus their results show inconsistency in the effects of lipid levels on IHD. In this study, we evaluate whether lipid levels have an effect on the risk of IHD in a Korean population.
Methods:A 2-sample Mendelian randomization (MR) study, using the genetic variants associated with lipid levels as the instrumental variables was performed. Genetic variants significantly associated with lipid concentrations were obtained from the Korean Genome and Epidemiology Study (n=35,000), and the same variants on IHD were obtained from the Korean Cancer Prevention Study-II (n=13,855). Inverse variance weighting (IVW), weighted median, and MR-Egger approaches were used to assess the causal association between lipid levels and IHD. Radial MR methods were applied to remove outliers subject to pleiotropic bias.
Results:Causal association between low-density lipoprotein-cholesterol (LDL-C) and IHD was observed in the IVW method (odds ratio, 1.013; 95% confidence interval, 1.007–1.109).However, high-density lipoprotein-cholesterol (HDL-C) and triglyceride (TG) did not show causal association with IHD. In the Radial MR analysis of the relationship between HDL-C, TG and IHD, outliers were detected. Interestingly, after removing the outliers, a causal association between TG and IHD was found.
Conclusions:High levels LDL-C and TG were causally associated with increased IHD risk in a Korean population, these results are potentially useful as evidence of a significant causal relationship.
