Abstract 17033: Inflammation and Lipids Components: Insights From the VLDL Study

Abstract

Background: Dyslipidemia and inflammation independently contribute to atherosclerosis. The associations between different lipid parameters and inflammatory markers is not fully understood. Hypothesis: LDL-C, triglycerides, and HDL-C do not predict inflammation equally. Methods: We analyzed data from 784,233 patients from the second harvest of the Very Large Database of Lipid study with lipids and hsCRP measured. The prevalence of having hsCRP≥2 mg/L was compared in 20 quantiles of non-HDL-C, LDL-C, HDL-C and triglycerides. Using linear regression, we estimated the correlations between hsCRP and lipids and to what degree individual lipid components explain the variation in hsCRP values. We then examined these association by sex and age (≥65 vs <65 years) categories. Results: The median hsCRP of the population was 2.3 mg/L (IQR, 1.0-5.7). The proportion of patients with hsCRP≥2 mg/L progressively increased with higher non-HDL-C, LDL-C, triglycerides quintiles, but decreased with higher HDL-C quintiles. All lipid measures directly correlated with hsCRP value except HDL-C which was inversely correlated (P<0.001 for all). LDL-C and non-HDL-C values explained very little of the variance seen in hsCRP (univariate R 2 = 0.5% and 0.1%, respectively). Triglyceride levels were the strongest predictor of hsCRP (standardized β, 0.21; P<0.001) and explained 4.5% of variability in hsCRP values. HDL-C was the second best (albeit an inverse) predictor of hsCRP (standardized β, -0.19; P<0.001) and explained 3.8% of its variance. Regardless LDL-C quantile, the prevalence of having hsCRP ≥2 mg/L was lower in those also have low triglyceride or high HDL-C levels (Figure). These associations were consistent by sex and age categories. Conclusion: Of major lipid and lipoprotein cholesterol fractions, triglycerides and HDL-C correlated most strongly and non-HDL-C and LDL-C contributed the least to hsCRP.