Abstract 684: Principal Component Analysis Identifies Lipoprotein Subfractions Associated with Coronary Artery Calcification in a Young High Cardiometabolic Risk Population: The Baptist Employee Healthy Heart Study

Abstract

Introduction: There is limited evidence relating lipoprotein subfractions or their composite measures to subclinical markers of atherosclerosis. In this study, we tested whether combinations of lipoprotein subfractions independently predict the presence of Coronary Artery Calcification (CAC) among presumably healthy people with high cardiovascular disease (CVD) risk. Methods: Cardio IQ™ Ion Mobility lipoprotein fractionation was measured at baseline in 172 high CVD risk participants of the Baptist Employee Healthy Heart Study (BEHHS). Principal component analysis was used to check for eventual components of CVD risk. Using bootstrap techniques, we created 1000 data sets of size 5000 and analyzed results of logistic regression. Results: Three principal components (PC) accounted for 88% variability. PC1 represented an increase in all lipoprotein measures; PC2 a decrease in HDL, large and medium LDL with an increase in VLDL and small and very small LDL; PC3 an increase in HDL, and a decrease in LDL subfractions and large IDL particles. The adjusted odds ratios for the presence of CAC were: 1.01(0.98 - 1.03) for PC1, 1.22(1.17-1.27) for PC2 and 0.90 (0.85 - 0.97) for PC3. PC2 and PC3 were associated with p-values smaller than 0.05 in 100% and 88% of all models. Conclusion: PCA of lipoprotein subfractions identified 3 independent components of CVD risk. PC2, proxy for the “atherogenic lipoprotein phenotype”, was consistently associated with the presence of underlying coronary artery calcification. There is potential to explore how lipoprotein subfractions as determined by a safe, inexpensive blood test can be used to stratify CVD risk in younger but high risk individuals.