Triglyceride-Rich Lipoprotein Cholesterol, Small Dense LDL Cholesterol, and Incident Cardiovascular Disease

Abstract

Background: Elevated triglyceride-rich lipoprotein (TRL) and small-dense low-density lipoprotein (sdLDL) particles are hallmarks of atherogenic dyslipidemia, and their cholesterol content is hypothesized to drive atherosclerotic risk. Prospective epidemiological data pertaining to cholesterol content of TRLs and sdLDL in primary prevention populations are mostly limited to coronary heart disease.

Objectives: The purpose of this study was to prospectively evaluate whether triglyceride-rich lipoprotein cholesterol (TRL-C) and small-dense low-density lipoprotein cholesterol (sdLDL-C) concentrations associate with composite and individual incident cardiovascular disease (CVD) outcomes including myocardial infarction (MI), ischemic stroke (IS), and peripheral artery disease (PAD).

Methods: In a prospective case-cohort study within the Women's Health Study, TRL-C and sdLDL-C (mg/dl) were directly measured in baseline blood specimens of case subjects (n = 480) and the reference subcohort (n = 496). Risk associations were evaluated for total CVD (MI, IS, PAD, and CVD death), coronary and cerebrovascular disease (MI, IS, CVD death), and individual outcomes (MI, IS, and PAD). Models were adjusted for traditional risk factors, low-density lipoprotein cholesterol, and high-sensitivity C-reactive protein.

Results: The risk of both composite outcomes significantly increased across quartiles of TRL-C and sdLDL-C. TRL-C was significantly associated with MI and PAD (MI hazard ratio [HR]_Q4: 3.05 [95% confidence interval (CI): 1.46 to 6.39]; p_trend = 0.002; PAD HR_Q4: 2.58 [95% CI: 1.18 to 5.63]; p_trend = 0.019), whereas sdLDL-C was significantly associated with MI alone (HR_Q4: 3.71 [95% CI: 1.59 to 8.63]; p_trend < 0.001). Both markers weakly associated with IS. Association patterns were similar for continuous exposures and, for TRL-C, among subjects with low atherogenic particle concentrations (apolipoprotein B <100 mg/dl).

Conclusions: TRL-C strongly associates with future MI and PAD events, whereas sdLDL-C strongly associates with MI alone. These findings signal that the cholesterol content of TRLs and sdLDL influence atherogenesis independently of low-density lipoprotein cholesterol, and high sensitivity C-reactive protein, with potentially different potency across vascular beds. (Women's Health Study; NCT00000479).

Keywords: atherogenesis; biomarkers; dyslipidemia; epidemiology; primary prevention; vascular disease.

FIGURE 1. Baseline Concentrations of TRL-C and sdLDL-C for Subjects With Incident CVD and the Subcohort

Baseline concentrations of TRL-C (left) and sdLDL-C (right) for subjects in the reference subcohort and individual cardiovascular outcomes. Concentrations of TRL-C and sdLDL-C were significantly higher in each case group compared with the reference subcohort (p < 0.001 for all except PAD and sdLDL-C [p = 0.002]). Values shown are geometric means and 95% confidence interval. IS = ischemic stroke; MI = myocardial infarction; PAD = peripheral artery disease; sdLDL-C = small-dense low-density lipoprotein cholesterol; TRL-C = triglyceride-rich lipoprotein cholesterol.

FIGURE 2. Extreme Quartile Adjusted Hazard Ratios for Associations With Incident CVD Outcomes

Estimates above the dashed horizontal had a statistically significant trend across quartiles (p_trend <0.05), whereas those below did not (p_trend >0.05). All analyses adjusted for model 4 covariates. CCVD = coronary and cerebrovascular disease; CVD = cardiovascular disease; other abbreviations as in Figure 1.

FIGURE 3. Spline Analyses for Associations With Incident CVD Outcomes

For each plot, the first knot identified is used as a reference point (hazard ratio: 1.00). *Nonlinear relationships. TRL-C had a nonlinear relationship with incident total CVD (p_nonlinearity <0.001; knots at 19.0, 48.0, and 100 mg/dl) as did sdLDL-C (p_nonlinearity <0.001; knots at 19.0, 40.0, and 84.0 mg/dl). Associations with CCVD were found to be nonlinear for TRL-C and linear for sdLDL-C (TRL-C p_nonlinearity = 0.001; sdLDL-C p_nonlinearity = 0.19; p_linearity <0.001). Both associations with MI were nonlinear (p_nonlinearity <0.001 for both). TRL-C and PAD association was more consistent with linearity (p_nonlinearity = 0.17; p_linearity = 0.002), whereas sdLDL-C and PAD were not significantly associated. Neither marker significantly associated incident IS. All analyses adjusted for model 4 covariates. Abbreviations as in Figures 1 and 2.

FIGURE 4. Effects of TRL-C and sdLDL-C Stratified by Apolipoprotein B

Adjusted cardiovascular risk associations stratified by Apolipoprotein B (<100 mg/dl or ≥100 mg/dl) are shown for TRL-C (A to E) and sdLDL-C (F to J). Panels for total CVD (A, F), CCVD (B, G), MI (C, H), PAD (D, I), and IS (E, J). All analyses adjusted for model 4 covariates. Abbreviations as in Figures 1 and 2.

CENTRAL ILLUSTRATION. Cholesterol Content of Triglyceride-Rich Lipoproteins and Small-Dense Low-Density Lipoprotein

Cholesterol-esterase transfer protein exchange of triglycerides and cholesterol esters occurs between triglyceride-rich lipoproteins (TRL) and low-density lipoprotein particles in hypertriglyceridemia, while hepatic lipase activity promotes conversion of large low-density lipoprotein particles to small dense low-density lipoprotein (sdLDL) particles. The result is an increase in cholesterol-enriched TRL particles and in sdLDL particles of varying triglyceride and cholesterol content. The cholesterol content of TRL and sdLDL is associated with an increased risk of cardiovascular disease outcomes of varying magnitude across vascular territories. Apo = apolipoprotein; CE = cholesteryl ester; CETP = cholesterylester transfer protein; FFA = free fatty acids; HL = hepatic lipase; IDL = intermediate-density lipoprotein; IS = ischemic stroke; LDL = low-density lipoprotein; LpL = lipoprotein lipase; MI = myocardial infarction; PAD = peripheral artery disease; TG = triglycerides; VLDL = very-low-density lipoprotein.