Abstract 13843: High-density Lipoprotein Cholesterol Subfractions and Risk for Cardiovascular Events in the Framingham Offspring and Jackson Heart Studies: A Meta-analysis
Introduction: High-density lipoprotein cholesterol (HDL) is inversely proportional to risk for coronary heart disease (CHD) related events. The relationship between large (HDL2) and small (HDL3) subfractions and risk for CHD events is more controversial, with results being decidedly mixed.
Hypothesis: We quantified the relationship between HDL and its subfractions with risk for CHD related events by performing a meta-analysis of persons enrolled in the Framingham Offspring (FOS) and Jackson Heart (JHS) Studies. Based on our prior work, we hypothesized that HDL3 correlates with reduced CHD risk.
Methods: This analysis incorporated the results for 818 (exam 6) and 4114 (from start of study) men and women without prevalent CHD from the FOS and JHS, respectively, followed up over 8 years. Serum lipoproteins were subfractionated by vertical auto profile (VAP) testing. The relationships between HDL and its subfractions to incident CHD events (myocardial infarction, CV mortality, and revascularization procedures) was assessed using multivariate Cox proportional hazards regression, adjusting for age, sex, BMI, smoking status, SBP, DBP, lipid treatment, and diabetes mellitus. Hazard ratios per standard deviation increase of each parameter and their 95% confidence intervals were calculated and hazard ratios then combined across FHS and JHS using a fixed-effects meta-analysis.
Results: For the combined group, HDL (HR 0.81; 95% CI 0.67-0.98; p<0.033) and HDL3 (HR 0.77; 95% CI 0.64-0.93, p<0.007) correlate with reduced risk for CV events. In contrast, HDL2 did not correlate significantly (HR 0.92; 95% CI 0.76-1.10, p=0.353). Similarly, when evaluating risk for CHD by the ratio of total atherogenic lipoprotein cholesterol to HDL and its subfractions, non-HDL/HDL and non-HDL/HDL3 are predictive of increased risk (HR 1.32; 95% CI 1.13-1.53, p<0.001; and HR 1.36; 95% CI 1.18-1.58, p<0.001), while non-HDL/HDL2 is not (HR 1.12; 95% CI 0.95-1.32, p=0.188).
Conclusions: In these cohorts, total HDL and HDL3 correlate highly with reduced risk for CV events, while HDL2 does not. These data may help explain why effort aimed at reducing risk by predominately raising HDL2 with niacin and cholesterol ester transfer protein inhibitors have been negative to date.
- © 2013 by American Heart Association, Inc.