Author + information
- Received October 22, 2011
- Revision received March 21, 2012
- Accepted March 29, 2012
- Published online August 7, 2012.
- Rachel H. Mackey, PhD, MPH⁎,⁎ (, )
- Philip Greenland, MD†,
- David C. Goff Jr, MD, PhD‡,
- Donald Lloyd-Jones, MD, ScM†,
- Christopher T. Sibley, MD§ and
- Samia Mora, MD, MHS∥
- ↵⁎Reprint requests and correspondence:
Dr. Rachel H. Mackey, University of Pittsburgh Graduate School of Public Health, 130 DeSoto Street, A531 Crabtree Hall, Pittsburgh, Pennsylvania 15261
Objectives The purpose of this study was to evaluate independent associations of high-density lipoprotein cholesterol (HDL-C) and particle (HDL-P) concentrations with carotid intima-media thickness (cIMT) and incident coronary heart disease (CHD).
Background HDL-C is inversely related to CHD, and also to triglycerides, low-density lipoprotein particles (LDL-P), and related metabolic risk. HDL-P associations with CHD may be partially independent of these factors.
Methods In a multiethnic study of 5,598 men and women ages 45 to 84 years old, without baseline CHD, excluding subjects on lipid-lowering medications, triglycerides >400 mg/dl, or missing values, we evaluated associations of HDL-C and nuclear magnetic resonance spectroscopy-measured HDL-P with cIMT and incident CHD (myocardial infarction, CHD death, and angina, n = 227 events; mean 6.0 years follow-up). All models were adjusted for age, sex, ethnicity, hypertension, and smoking.
Results HDL-C and HDL-P correlated with each other (ρ = 0.69) and LDL-P (ρ = −0.38, −0.25, respectively, p < 0.05 for all). For (1 SD) higher HDL-C (15 mg/dl) or HDL-P (6.64 μmol/l), cIMT differences were − 26.1 (95% confidence interval [CI]: −34.7 to −17.4) μm and −30.1 (95% CI: −38.8 to − 21.4) μm, and CHD hazard ratios were 0.74 (95% CI: 0.63 to 0.88) and 0.70 (95% CI: 0.59 to 0.82), respectively. Adjusted for each other and LDL-P, HDL-C was no longer associated with cIMT (2.3; 95% CI: − 9.5 to 14.2 μm) or CHD (0.97; 95% CI: 0.77 to 1.22), but HDL-P remained independently associated with cIMT (−22.2; 95% CI: − 33.8 to −10.6 μm) and CHD (0.75; 95% CI: 0.61 to 0.93). Interactions by sex, ethnicity, diabetes, and high-sensitivity C-reactive protein were not significant.
Conclusions Adjusting for each other and LDL-P substantially attenuated associations of HDL-C, but not HDL-P, with cIMT and CHD. Potential confounding by related lipids or lipoproteins should be carefully considered when evaluating HDL-related risk.
- cardiovascular disease
- high-density lipoprotein cholesterol
- high-density lipoprotein particles
There is great interest in raising levels of high-density lipoprotein cholesterol (HDL-C), given its well-established inverse association with atherosclerosis and coronary heart disease (CHD) (1). However, quantification of HDL-C, the cholesterol carried by HDL particles (HDL-P), may not fully capture HDL-related risk (1,2). For example, some forms of genetically low (3) or genetically high HDL-C (4) do not correspond to expected differences in CHD risk. Recent failures of drugs that raised HDL-C without reducing cardiovascular disease (CVD) events (5,6) or atherosclerosis (7) have also fueled interest in alternative indexes of HDL quantity (i.e., HDL-P or apolipoprotein A-I [apo A-I]) or possibly HDL “quality,” such as, particle size, subclass distribution (8), or various measures of HDL functionality (2).
The association of HDL-C with CHD risk is complicated by the inverse association of HDL-C with triglycerides, insulin resistance, obesity, high-sensitivity C-reactive protein (hsCRP), and atherogenic lipoprotein particles, (i.e., apolipoprotein B [apo B] and LDL particle [LDL-P] concentration) (1). Recent reports showed that adjusting for apo B and apo A-I abolished the inverse association of HDL-C with CHD risk (9), but HDL-P remained inversely associated with CHD, adjusted for apo B, triglycerides, and HDL particle size (10). Therefore, we hypothesized that HDL-C associations with carotid intima-media thickness (cIMT) and incident CHD events partly reflect correlated lipid, apolipoprotein, or lipoprotein concentrations, particularly LDL-P, but that HDL-P associations are less affected by these metabolic risk factors, including HDL-C. Because HDL functionality has been reported to be altered in diabetes (11,12), with inflammation (13), by ethnicity (14), or sex (15), we also evaluated potential interactions by baseline diabetes, hsCRP, sex, and ethnicity.
Participants and risk factor measurement
Participants eligible for the present study were 6,814 men and women enrolled at baseline (2000 to 2002) in the National Heart, Lung, and Blood Institute (NHLBI) sponsored multicenter community-based cohort, Multi-Ethnic Study of Atherosclerosis (MESA), the design and objectives of which have been previously described (16). Briefly, MESA participants were community-dwelling men and women 45 to 84 years old, of African American, Hispanic, white, and Chinese American ethnicity. Baseline exclusion criteria included self-reported CVD (heart attack, angina, coronary, or any other arterial revascularization procedure; pacemaker or defibrillator implantation; valve replacement; heart failure or cerebrovascular disease), pregnancy, cancer, cognitive impairment, or weight >136 kg. The present study excluded participants with baseline lipid-lowering medication use (n = 1,100), triglycerides >400 mg/dl (n = 57), or missing values for lipid-lowering medication use, HDL-C, HDL-P, or smoking (n = 59). Of the remaining 5,598 participants, cIMT was missing for 57 (1.0%), and incident CHD data was missing for 1, leaving 5,541 participants for the cIMT analyses and 5,597 participants for the CHD analyses. Participants provided informed written consent at their field centers. The study was approved by the institutional review boards of the participating institutions and the University of Pittsburgh.
Height, weight, blood pressure, and medications were collected at the baseline MESA examination. Smoking was defined as never, former (smoked ≥100 cigarettes in lifetime), or current (smoked cigarettes in last 30 days). Hypertension was defined as systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg, or self-reported hypertension and antihypertensive medication use. Hormone therapy use was defined as current user (yes/no). Type 2 diabetes was defined as fasting glucose >125 mg/dl or use of antidiabetic medication. Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as: insulin (μIU/l) × (glucose [mg/dl] × 0.055)/22.5 for those not on antidiabetic medication (17).
Lipid, lipoprotein, and other laboratory assays
Blood was drawn after a 12-h fast, and samples were stored at −70°C. Lipids, insulin, and glucose were measured at a central laboratory (Collaborative Studies Clinical Laboratory at Fairview University Medical Center, Minneapolis, Minnesota). Lipids were assayed on thawed ethylenediaminetetraacetic acid plasma within 2 weeks of the sample collection, using Centers for Disease Control Prevention/NHLBI standards. HDL-C was measured using the cholesterol oxidase method (Roche Diagnostics, Indianapolis, Indiana) after precipitation of non–HDL-C with magnesium/dextran (coefficient of variation 2.9%). LDL-C was calculated using the Friedewald equation (18). Plasma lipoprotein particle concentrations were measured at LipoScience, Inc. (Raleigh, North Carolina) by nuclear magnetic resonance (NMR) spectroscopy using the LipoProfile-3 algorithm. HDL-P and LDL-P (coefficient of variation <4%) are the sums of the particle concentrations of their respective subclasses, which are quantified based on particle size using the amplitudes of their lipid methyl group NMR signals, and mean particle sizes are the weighted average of related subclasses (19).
Carotid atherosclerosis was measured using high-resolution B-mode ultrasound as previously described for the Cardiovascular Health Study (20). cIMT was calculated from maximal thickness measured at 8 sites (right and left, near and far walls of the common and internal carotid arteries) as previously described (21). Incident CHD events (myocardial infarction, CHD death, resuscitated cardiac arrest, or definite or probable angina (followed by revascularization) were ascertained and adjudicated for MESA as previously described (22).
Hypotheses were also tested using secondary endpoints of: 1) “hard” CHD, which excluded angina; 2) “all CVD,” which was “all CHD” plus stroke, stroke death, other atherosclerotic death, or other CVD death; 3) “hard” CVD, which included “hard CHD” plus stroke and stroke death; and 4) cIMT considered separately for the common versus the internal (more susceptible to plaque) carotid artery.
Analyses were performed using SAS version 9.2 (SAS Institute, Cary, North Carolina). Two-tailed p values <0.05 were considered significant. Spearman-rank correlations were adjusted for age, sex, and ethnicity. HDL-C and HDL-P were analyzed as continuous variables (results reported per 1-SD increment) or categorized as tertiles or quartiles. Triglycerides were log-transformed. Associations with cIMT were modeled using analysis of covariance and linear regression, and with incident CHD events using Cox proportional hazards regression. All models were adjusted for a base set of covariates: age, sex, ethnicity, hypertension, and smoking. A p value for linear trend was calculated using contrasts. Hypothesized differences by sex, ethnicity, diabetes, and hsCRP were tested by including interaction terms with HDL-C and HDL-P for each of the models, with the main effect included in the model. Sensitivity analyses included excluding hormone therapy users (n = 837), stratifying by sex, and also testing hypotheses using the previously described secondary endpoints (i.e., hard CHD, all CVD, hard CVD, internal carotid artery cIMT, and common carotid artery cIMT).
To illustrate multivariable regression results, a stratified analysis was used to calculate adjusted mean cIMT for HDL-C tertiles within HDL-P tertiles, and then further stratified by above and/or below median LDL-P. Tertiles (HDL-C and HDL-P) were used rather than quartiles to allow for a sufficient number of individuals in discordant cells (i.e., high HDL-C/low HDL-P or low HDL-C/high HDL-P). Finally, we sought to replicate a report of an increased risk of CHD for very high HDL-C (≥80 mg/dl) relative to low HDL-C (<40 mg/dl) in adjusted models (9), and also evaluated CHD risk for very high versus low HDL-P, using the corresponding 95th and 25th percentiles of HDL-P (≥45.7 and <29 μmol/l).
Study participants were multiethnic men and women 45 to 84 years old from MESA, without baseline clinical CVD or lipid-lowering medication use (Table 1). HDL-C and HDL-P concentrations were positively correlated (Fig. 1). HDL-C and HDL-P were inversely correlated with LDL-C, weakly (ρ = −0.08, −0.13, respectively), and with LDL-P, more strongly (ρ = −0.38, −0.25), and with other metabolic risk factors (e.g., small LDL-P, triglycerides, body mass index, waist circumference, and HOMA-IR), but for all, correlations were stronger for HDL-C than for HDL-P (Table 2).
LDL-C and LDL-P
Positive associations of LDL-C and LDL-P with cIMT showing that CVD events in MESA have been published (21,23), and when LDL-P and LDL-C differed, associations were stronger for LDL-P (23). In our study, adjusted for base covariates, the cIMT difference per 1-SD increment was 28.8 (95% confidence interval [CI]: 13.4 to 44.3) μm for LDL-C and 36.5 (95% CI: 20.8 to 52.1) μm for LDL-P. The hazard ratios (HRs) for CHD per 1-SD increment were 1.24 (95% CI: 1.09 to 1.42) for LDL-C and 1.29 (95% CI: 1.13 to 1.47) for LDL-P, adjusted for base covariates. LDL-C and LDL-P each remained associated (p < 0.05) with cIMT and incident CHD in the models reported in the tables, which were also adjusted for HDL-P, HDL-C, or both.
Associations with carotid atherosclerosis
Adjusted for base covariates (age, sex, ethnicity, hypertension, and smoking), mean cIMT was lower with higher quartiles of HDL-C (Fig. 2A) or HDL-P (Fig. 2B). The inverse linear association between HDL-C and cIMT was attenuated by adjusting for LDL-P or for HDL-P, and was abolished when adjusted for both. In contrast, HDL-P remained inversely associated with cIMT after adjusting for LDL-P, HDL-C, or both (p for trend <0.05, for all).
Additional models are shown in Table 3, which reports adjusted mean cIMT differences for a 1-SD increment in HDL-C or HDL-P. Separately, higher HDL-C and HDL-P were each associated with lower mean cIMT, and were modestly attenuated by adjusting for LDL-C or HDL-P size. However, adjusting for LDL-P substantially attenuated cIMT estimates, more for HDL-C (12.2; 95% CI: − 21.4 to −3.1 μm) than for HDL-P (−20.7; 95% CI: −29.6 to −11.8 μm). Further adjustment for LDL-C and triglycerides had little effect.
In joint models (HDL-C and HDL-P adjusted for each other and base covariates [Table 3, lower panel]), HDL-C and HDL-P associations were only mildly attenuated by adjusting for LDL-C. However, adjusted for HDL-P, HDL-C was no longer significantly associated with cIMT (−11.1; 95% CI: −22.7 to 0.42 μm), and became positive, but not statistically significant, in models that also adjusted for mean HDL size or LDL-P, with or without LDL-C and triglycerides. Conversely, HDL-P remained inversely associated with cIMT (−22.2; 95% CI: −33.8 to −10.6 μm) when adjusted for HDL-C, and also for LDL-P or HDL size, or LDL-P, LDL-C, and triglycerides.
To visualize the multivariable model results, Figure 3 shows adjusted mean cIMT for cross-classified tertiles of HDL-C and HDL-P (Fig. 3A), and further stratified by median LDL-P (Figs. 3B and C). Power is limited for these analyses because of the small numbers in the discordant groups (e.g., high HDL-C/low HDL-P). Within HDL-P tertiles, HDL-C was generally not inversely associated with cIMT (HDL-C trend, p = NS for all) (Fig. 3A). However, within each HDL-C tertile, each HDL-P was inversely associated with cIMT in both low and high HDL-C tertiles (p trend < 0.05 for both). When further stratified by median LDL-P (Figs. 3B and 3C), trends were generally not statistically significant, but HDL-C was positively associated with cIMT within 5 of the 6 HDL-P tertiles, whereas HDL-P remained inversely associated with cIMT within 5 of the 6 HDL-C tertiles.
Sensitivity analyses for cIMT associations
Interaction terms for sex, ethnicity, diabetes status, or hsCRP were not significant. In sex-stratified models, adjusted for LDL-P and each other, HDL-C was not inversely associated with cIMT for women (−1.2; 95% CI: −15.0 to 12.7 μm) or men (8.3; 95% CI: −13.4 to 30.1 μm), whereas HDL-P was significantly inversely associated with cIMT for women (−17.8; 95% CI: −31.4 to −4.2 μm) and men (−27.1; 95% CI −47.7 to −6.4 μm). Evaluated separately, the internal and common carotid arteries showed similar results to our combined cIMT measure, although associations were slightly stronger for the internal carotid artery, which is more prone to plaque.
Associations with incident coronary heart disease events
Among the 5,597 participants with incident CHD data, 227 CHD events occurred during a 6.0 ± 1.4 years of follow-up. The proportionality assumption appeared valid (i.e., interactions of time with HDL-C or HDL-P were not significant). CHD risk was reported for quartiles and for a 1-SD increment in HDL-C and HDL-P, for separate and joint models (Table 4). Separately, higher HDL-C and HDL-P were similarly associated with lower CHD risk, adjusted for base covariates. Adjusting for LDL-C or HDL particle size had little effect. Adjustment for LDL-P attenuated the association for HDL-C to a greater extent than for HDL-P. In joint models (adjusted for HDL-P), HDL-C HRs were not statistically significant and became weakly positive when adjusted for LDL-P, LDL-C, and (log) triglycerides. In contrast, the inverse association of HDL-P with CHD remained statistically significant when adjusted for HDL-C, LDL-P, LDL-C, and triglycerides.
Sensitivity analysis for CHD associations
In multivariable models, interaction terms for sex, ethnicity, diabetes, or hsCRP were not statistically significant. Results were similar if adjusted for baseline diabetes status, or if current hormone users were excluded, or if stratified by sex. With few cases among women (n = 66), CHD risk estimates for women were not statistically significant, but the base covariate-adjusted HR (95% CI) for (1 SD) higher HDL-C was 0.81 (95% CI: 0.63 to 1.05) and became 1.00 (95% CI: 0.71 to 1.43) when adjusted for HDL-P and LDL-P, whereas for HDL-P, the base covariate-adjusted HR of 0.77 (95% CI: 0.59 to 1.00) became 0.81 (95% CI: 0.57 to 1.14) when adjusted for HDL-C and LDL-P. Results were also similar for the secondary outcomes of all CVD or hard CHD or hard CVD (Online Table 1).
Finally, we evaluated CHD risk associations at very high levels of HDL-C and HDL-P. Adjusted for base covariates, LDL-P, HDL-P, and log-triglycerides, the HR for very high HDL-C (≥80 mg/dl) compared with low HDL-C (<40 mg/dl) became positive and statistically significant (HR: 2.59; 95% CI: 1.11 to 6.02). Conversely, adjusted for base covariates, LDL-P, HDL-C, and log triglycerides, the HR for analogous very high versus low HDL-P (≥45.7 vs. <29 μmol/l) was 0.50 (95% CI: 0.19 to 1.35).
Among multiethnic men and women without clinical CVD or lipid-lowering medication use at baseline, HDL-C associations with cIMT and incident CHD were substantially attenuated by adjusting for atherogenic lipoproteins, particularly LDL-P. In contrast, HDL-P associations with cIMT and incident CHD were relatively unaffected by adjusting for atherogenic lipoproteins, HDL-C, and mean HDL particle size. Results were similar for secondary outcomes of all CVD, and hard CHD or CVD events.
Few studies have evaluated HDL-P associations with CHD risk, and we know of none that evaluated it jointly with HDL-C and LDL-P. Low HDL-P levels predicted CHD death over 18 years of follow-up among men with metabolic syndrome in the MRFIT (Multiple Risk Factor Intervention Trial) cohort (24). In EPIC (European Prospective Investigation into Cancer and Nutrition)-Norfolk, lower HDL-P levels predicted incident events independent of age, sex, apo B, triglycerides, mean HDL particle size, smoking, myeloperoxidase, paraoxonase-1, and hsCRP (10). Lower HDL-P also predicted CVD events among HIV patients (25). In the VA-HIT (Veterans Affairs High-Density Lipoprotein Cholesterol Intervention Trial), lower levels of baseline and on-trial HDL-P predicted CHD events among men with low HDL-C randomized to gemfibrozil versus placebo (26). In the large Women's Health Study, the inverse association of HDL-P with incident CVD over an 11-year follow-up was not significant (27). However, HDL-P was inversely associated with incident CHD among postmenopausal women in the Women's Health Initiative Hormone Trial, adjusted for treatment arm (28), and the inverse association of HDL-P with cIMT was statistically significant for women in the present study. Future studies may help to reconcile these results.
In this study, HDL-C was not inversely associated with cIMT or CHD after adjusting for LDL-P and HDL-P, similar to reported attenuation of HDL-C associations when adjusted for apo B, as an index of atherogenic lipoproteins, and apo A-I (9). In our study, very high HDL-C (≥80 mg/dl) became positively associated with CHD risk (p < 0.05) when adjusted for LDL-P, HDL-P, and triglycerides, as reported in the IDEAL (Incremental Decrease in Clinical Endpoints Through Aggressive Lipid Lowering) study (adjusted for apo B and apo A-I) (9). In contrast, very high HDL-P (≥45.7 μmol/l) remained inversely associated with CHD, in models adjusting for LDL-P, HDL-C, and triglycerides. However, the results of our study suggest that, adjusted for LDL-P and HDL-P, HDL-C loses its inverse association with atherosclerotic CHD across the range of HDL-C, not just at very high HDL-C.
These results suggest that because HDL-C, the cholesterol content of HDL, varies inversely with triglycerides, LDL-P, or apo B, and other metabolic risk factors, the risk attributed to HDL-C might come from several sources other than low levels of particles (i.e., HDL-P). In contrast, HDL-P remained inversely associated with atherosclerotic risk relatively independently of both atherogenic lipoprotein levels, and its own cholesterol content (i.e., HDL-C). More HDL-P might equal higher reverse cholesterol transport capacity. Cholesterol efflux, an index of the capacity of HDL for reverse cholesterol transport, was inversely correlated with cIMT and angiographic coronary disease independent of HDL-C, and cholesterol efflux was also associated with higher levels of apo A-I (a rough measure of HDL-P) independent of HDL-C (29). Furthermore, in a study of diabetic patients, cholesterol efflux was positively associated with total HDL-P but not with HDL-C or apo A-I (30). Antiatherogenic benefits of higher HDL-P (antioxidation, anti-inflammation, and so on) may also be related to the protein or other cargo of HDL (e.g., apo A-I, paraoxonase-1, myeloperoxidase) rather than to its cholesterol cargo (1,2). In EPIC-Norfolk, the HDL-associated antioxidant paroxonase-1 was more strongly correlated with HDL-P than with HDL-C or apo A-I (10). However, given the complexity of HDL, many potential mechanisms require further investigation (2).
Lifestyle and pharmacological interventions to increase low HDL-C have been reviewed elsewhere (31), but few studies have evaluated intervention effects on HDL-P compared with HDL-C. A few studies have reported that HDL-C and HDL-P were both higher with hormone therapy (28) and alcohol intake (32). Active smokers had lower levels of both HDL-C and HDL-P (33), and both increased with smoking cessation in a recent randomized clinical trial (34). The limited existing data suggested that physical activity (35) and diet and/or exercise interventions might increase HDL-C but not HDL-P (36), which would occur if large cholesterol-rich particles increased at the expense of smaller particles.
Niacin also raised HDL-C with little effect on HDL-P (37). Torcetrapib reportedly raised HDL-P by only 1% despite a 53% increase in HDL-C (38). In contrast, gemfibrozil increased HDL-P more than HDL-C in VA-HIT, and as noted, on-treatment HDL-P predicted lower CHD events (26). Statins also increased HDL-P more than HDL-C, as well as decreasing LDL-P less than LDL-C (39). Among diabetic individuals, effects of vitamin E intake on HDL function (i.e., cholesterol efflux) differed by haptoglobin genotype, increasing it for Hp2-2 and decreasing it for Hp 2-1 (12). Whether effects on HDL-P levels would parallel these effects is unknown. Additional research is needed to quantify differential effects of pharmacological interventions as well as types of diet, omega-3 fatty acids, haptoglobin genotype (12), and other influences on HDL in relation to outcomes.
These results showed that associations of HDL-C with CHD risk might be partially due to metabolic correlations with atherogenic lipoprotein concentrations. In contrast, associations of HDL-P with CHD risk were substantially independent of atherogenic lipoprotein concentrations and of HDL-C. HDL-C is one measured parameter of HDL. It might be important to assess parameters other than HDL-C in clinical trials of interventions to raise HDL. HDL-P might be an alternative to HDL-C as a marker of HDL-related cardiovascular risk, if these findings are confirmed in other studies and found to be cost-effective.
These results, although robust, were observational, did not prove causality, and might be subject to bias and confounding, measurement error, or unmeasured confounders. Lipid and lipoprotein levels are dynamically metabolically interrelated (8,11) and statistical modeling in a large observational study is only one tool to investigate potential effects of these interrelationships, which warrant additional metabolic studies. However, despite statistical adjustment for several correlated lipoproteins, multicollinearity was not a problem in this study, as assessed by variance inflation factors. Apo A-I and apo B are not currently available in MESA, so their influence could not be evaluated in the present study. Finally, there were very few events at the high end of the HDL parameter ranges and few among women, limiting our ability to evaluate independent effects on CHD risk in those groups.
Among multiethnic men and women, associations of HDL-C with cIMT and incident CHD were substantially attenuated by adjusting for atherogenic lipoproteins and HDL-P, whereas HDL-P remained significantly inversely associated with cIMT and incident CHD, independent of atherogenic lipoproteins (LDL-P, triglycerides, and LDL-C) and HDL-C. These results might have implications both for risk assessment and for evaluation of therapeutic interventions, particularly pharmacological interventions that might differentially affect several lipid and lipoprotein parameters concurrently. Quantitative and metabolic interrelationships between lipids and the lipoprotein particles that carry them should be considered when evaluating associations between single parameters (e.g., HDL-C and atherosclerotic cardiovascular risk).
The authors thank the MESA investigators, staff, and participants for their valuable contributions; the MESA investigators/institutions are listed at http://www.mesa-nhlbi.org. The information contained herein was derived in part from data provided by the Bureau of Vital Statistics, NY City Department of Health and Mental Hygiene. Liposcience, Inc. measured the nuclear magnetic resonance lipoproteins at no additional cost to MESA.
For a supplementary table, please see the online version of this article.
Dr. Mackey was supported by a research grant from LipoScience, Inc. to the University of Pittsburgh. The grant was unrestricted because LipoScience exercised no control over the design, management, analysis, or interpretation of the data, or in the preparation, review, or approval of the paper. This research was supported by contracts N01-HC-95159 through N01-HC-95169 from the National Heart, Lung, and Blood Institute, an unrestricted grant from LipoScience, Inc., and by grant K08 HL094375 to Dr. Mora. Dr. Goff is a research consultant for a clinical trial of a glucose lowering medication marketed by Merck; and a Data and Safety Monitoring Board member for a clinical trial of a glucose lowering medication marketed by Takeda. Dr. Mora received a research grant from AstraZeneca; and honorarium/consultant fees from Pfizer, Abbott, Quest Diagnostics, and AstraZeneca. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- apo A-I
- apolipoprotein A-I
- apo B
- apolipoprotein B
- coronary heart disease
- confidence interval
- carotid intima-media thickness
- cardiovascular disease
- high-density lipoprotein cholesterol
- high-density lipoprotein particles
- homeostasis model of insulin resistance
- hazard ratio
- high-sensitivity C-reactive protein
- low-density lipoprotein cholesterol
- low-density lipoprotein particles
- National Heart, Lung, and Blood Institute
- nuclear magnetic resonance
- Received October 22, 2011.
- Revision received March 21, 2012.
- Accepted March 29, 2012.
- American College of Cardiology Foundation
- Vergeer M.,
- Holleboom A.G.,
- Kastelein J.J.,
- Kuivenhoven J.A.
- deGoma E.M.,
- deGoma R.L.,
- Rader D.J.
- Rosenson R.S.,
- Brewer H.B. Jr..,
- Chapman M.J.,
- et al.
- van der Steeg W.A.,
- Holme I.,
- Boekholdt S.M.,
- et al.
- Bild D.E.,
- Bluemke D.A.,
- Burke G.L.,
- et al.
- Friedewald W.T.,
- Levy R.I.,
- Fredrickson D.S.
- O'Leary D.H.,
- Polak J.F.,
- Kronmal R.A.,
- Manolio T.A.,
- Burke G.L.,
- Wolfson S.K. Jr..
- Bluemke D.A.,
- Kronmal R.A.,
- Lima J.A.,
- et al.
- Otvos J.D.,
- Collins D.,
- Freedman D.S.,
- et al.
- Mora S.,
- Otvos J.D.,
- Rifai N.,
- Rosenson R.S.,
- Buring J.E.,
- Ridker P.M.
- Hsia J.,
- Otvos J.D.,
- Rossouw J.E.,
- et al.
- Rashedi N.,
- Brennan D.,
- Kastelein J.J.,
- Nissen S.E.,
- Nicholls S.
- Rosenson R.S.,
- Otvos J.D.,
- Hsia J.