Quantifying Triglyceride-Rich Lipoprotein Atherogenicity, Associations With Inflammation, and Implications for Risk Assessment Using Non-HDL Cholesterol

Background Triglyceride-rich lipoproteins and remnants (TRL/remnants) have a causal but not yet quantified relationship with coronary heart disease (CHD - myocardial infarction plus revascularisation). Objectives To estimate TRL/remnant per-particle atherogenicity, investigate causal relationships with inflammation, and determine whether difference in the atherogenicity of TRL/remnants and LDL impact the causal association of non-HDL-C with CHD. Methods Single nucleotide polymorphisms (SNPs) (n=1357) identified by genome-wide association in the UK Biobank were ranked into 10 clusters according to the effect on TRL/remnant-C versus LDL-C. Mendelian randomisation analysis was used to estimate for each SNP cluster CHD odds ratios per 10 mg/dL apoB and per 0.33 mmol/L non-HDL-cholesterol, and to evaluate association of TRL/remnants with biomarkers of systemic inflammation. Results SNPs in cluster 1 predominantly affected LDL-C while SNPs in cluster 10 predominantly affected TRL/remnant-C. CHD risk per genetically predicted increase in apoB and in non-HDL-C rose across clusters. Odds ratio per 10 mg/dL higher apoB was 1.1595%CI:1.11-1.19 in cluster 1 versus 1.7095%CI:1.52-1.90 in cluster 10. Comparing odds ratios between these TRL/remnant predominant- and LDL predominant-clusters, we estimated that TRL/remnants were at least 3.995%CI:2.8-5.4 times more atherogenic than LDL on a per-particle basis. For non-HDL-C, CHD odds ratios per 0.33 mmol/L rose from 1.151.11−1.19 for cluster 1 to 1.401.30−1.50 for cluster 10. TRL/remnants exhibited causal relationships with inflammation, but this did not explain their greater atherogenicity. Conclusion TRL/remnants are about 4 times more atherogenic than LDL. Variation in the causal association of non-HDL-C with CHD indicates that adjustment for percentage TRL/remnant-C may be needed for accurate risk prediction.