The puzzle of high-density lipoprotein in cardiovascular prevention

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

El-Harchaoui, A.

Publication date

2009

Document Version

Final published version

Link to publication

Citation for published version (APA):

El-Harchaoui, A. (2009). The puzzle of high-density lipoprotein in cardiovascular prevention.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

(2)

prevention

(3)

The puzzle of high-density lipoprotein in cardiovascular prevention Dissertation, University of Amsterdam, Amsterdam, the Netherlands ISBN 978-90-8559-604-2

Author A. El-Harchaoui

Layout Optima Grafische Communicatie, Rotterdam Cover Oguz Kurt, www.kurtontwerp.nl

Financial support for the printing of this thesis was provided by:

Stichting tot Steun Promovendi Vasculaire Geneeskunde, Universiteit van Amsterdam, Pfizer, Genzyme, AstraZeneca, Novartis, Merck Sharp & Dohme, Servier, Stichting AMSTOL, Schering-Plough, Menarini Farma and Astellas.

Financial support by the Netherlands Heart Foundation for the publication of this thesis is gratefully acknowledged.

(4)

prevenTion

ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam op gezag van de Rector Magnificus

prof. dr. D.C. van den Boom

ten overstaan van een door het college van promoties ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapel op donderdag 3 december 2009, te 10:00 uur

door

Abdelkarim El-Harchaoui geboren te Beni-Touizini, Marokko

(5)

promoTiecommissie

Promotor: Prof. dr. J.J.P. Kastelein Co-promotores: Dr. E.S.G. Stroes

Dr. J.A. Kuivenhoven Overige Leden: Prof. dr. R.J.G. Peters

Prof. dr. A. Sturk Prof. dr. J.G.P. Tijssen Dr. C. van ‘t Veer Prof. dr. J.W. Jukema Prof. F.L.J. Visseren Faculteit der Geneeskunde, Universiteit van Amsterdam

(6)

chapter 1

general introduction

(9)

(10)

inTroducTion

Despite impressive progress in the field of cardiovascular prevention during the last 3 decades, coronary artery disease (CAD) still remains the leading cause of death world-wide. The primary pathophysiological mechanism underlying CAD is atherosclerosis. Atherosclerosis is a degenerative process and as a consequence of aging everyone will eventually develop ath-erosclerosis. However, the rate and progression of this process depends on a combination of genetic as well as environmental factors. Important risk factors for developing atherosclerosis are gender, hypertension, smoking and diabetes (1). One of the main risk factors accelerating the atherosclerotic process is the presence of high cholesterol levels. In line, cardiovascular risk estimation engines, such as the Framingham Risk Score and the PROCAM risk score, all include measurements of cholesterol levels (2). A drawback of measuring cholesterol levels is that their predictive value is restricted to the population level, whereas their discriminative value for an individual is rather poor. The latter is emphasized by the fact that the majority of cardiovascular events occurs in people with normal cholesterol levels and/or low risk estimates (3). In view of these considerations, intensive efforts are being conducted to further improve the accuracy of risk estimation.

For this purpose, low-density lipoprotein (LDL) cholesterol is regarded as the most atherogenic fraction in cholesterol measurements and is traditionally used as the cornerstone in lipid lowering therapy. More recently, it has been suggested that various subclasses within the atherogenic LDL cholesterol fraction may be differently associated with CAD risk. Particularly, the small dense LDL particles are thought to be extremely atherogenic as compared to the larger LDL particles (4). In this context individuals with similar LDL cholesterol levels may have higher or lower numbers of atherogenic small LDL particles and, as a result, may differ in terms of absolute CAD risk.

This heterogeneity in terms of the cholesterol content per particle is not sufficiently reflected by measuring the cholesterol levels in plasma. This observation has led to the concept that risk estimates may be improved by directly measuring the different particles or indirectly by mea-suring the apolipoproteins of the particles, as a reflection of the atherogenecity of the particles. Refining current risk estimates with this information may more accurately identify individuals at high risk of cardiovascular events and may simultaneously reduce the number of ‘undertreated’ subjects. This might ultimately add to an optimal treatment benefit from lipid lowering therapy. After having identified who is at increased risk for developing a cardiovascular event, effective treatment of the lipid abnormality is the next step. Thus far, cardiovascular therapy has mainly focused on LDL cholesterol. Despite the reduction of LDL cholesterol to treatment goals, only a 20-30% overall reduction in the number of cardiovascular events has been achieved in

(11)

random-Chapt

er 1

10

ized clinical trials (5). These findings have led to the search for alternative treatment options in lipid therapy.

From epidemiological studies it is common knowledge that there is a strong and inverse rela-tionship between the levels of high density lipoprotein (HDL) and the risk of CAD(6). Studies in families with premature coronary disease have shown that a low HDL level was the most common abnormality (7) and epidemiological studies have reported a 20 to 30% increase in cardiovascular risk for each 0.26 mmol/l decrease in HDL cholesterol level (8). Even in subjects with spontaneously very low LDL cholesterol concentrations (< 1.55 mmol/l), HDL cholesterol is an independent predictor of coronary risk, with a 10% risk increase for every 0.26 mmol/l decrease in HDL cholesterol (9). These observations have contributed to the fact that treating low HDL may be a strategy in the fight against atherosclerosis even after LDL cholesterol lower-ing. In contrast to LDL, however, HDL is a complex molecule which is believed to have different anti-atherogenic functions. This finding has delayed the development of effective therapeutic interventions to increase low HDL cholesterol levels.

Traditionally, the anti-atherogenic effects of HDL were confined to its role in the reverse cholesterol transport (RCT) pathway. Reverse cholesterol transport is a term used to describe the efflux of excess cellular cholesterol from peripheral tissues and its return to the liver for excretion in the bile and ultimately in the feces (10). In this pathway, cholesterol is effluxed

Legend Scheme of the RCT pathway, adapted from Ashen NEJM 2005

(12)

from arterial macrophages to HDL through the action of membrane transporters such as ATP binding cassette transporter A1 (ABCA1) and ATP-binding cassette transporter G1 (ABCG1). Following its efflux to the HDL particle, cholesterol may then be esterified by the enzyme lecithin:cholesterol acyltranferase (LCAT), and is ultimately transported by HDL to the liver, either directly via the scavenger receptor BI (SR-BI) or following its transfer to apolipoprotein B-containing lipoproteins by the cholesteryl ester transfer protein (CETP). The final step is the excretion of cholesterol in the feces.

Promotion of RCT could therefore be an effective strategy for reducing risk associated with atherosclerotic vascular disease. Interfering in the different molecular mechanisms regulating RCT, results in changes in HDL cholesterol levels, and consequently in promotion of RCT. These developments could open new avenues in lipid prevention. In this thesis we mainly focus on the role of CETP and its relation to treatment of HDL levels.

ouTline of The Thesis

This thesis consists of two parts. Part one focuses on CAD risk assessment by refining current lipid tests. In this part we describe alternative lipoprotein parameters and their relation to CAD risk. In the second part we switch to the pathophysiology and treatment of low HDL cholesterol levels, with special emphasis on the role of CETP in HDL metabolism.

part i: refining current cholesterol lipid tests for assessment of future cad risk.

In chapter 2 and 3 we assess the independent relationship of LDL and HDL particle number as measured by nuclear magnetic resonance spectroscopy with risk of future CAD. In chapter 2 we focus on the pro-atherogenic LDL fractions. This chapter assesses whether measuring LDL particle number and LDL particle size have the capacity to improve the prediction of future CAD compared to measuring levels of traditional’ LDL cholesterol.

In chapter 3 we switch to the anti-atherogenic HDL fraction. This chapter addresses the inde-pendent relationship of HDL particle concentration and size to the risk of future CAD. In chapter

4, we evaluated whether the composite parameter representing both the pro-atherogenic and

anti-atherogenic lipoprotein fraction, i.e. the ratio of apolipoprotein B to A-I (apolipoprotein B/A-I ratio) adds to the accuracy of CAD risk assessments in the general population.

part ii: impact of low hdl cholesterol and ceTp on reverse cholesterol

transport and coronary artery disease.

Traditionally, HDL cholesterol levels were used as an ‘indirect’ measure for the reverse choles-terol transport pathway. In chapter 5 we evaluated the relation between HDL cholescholes-terol and fecal sterol excretion in patients with low HDL cholesterol levels due to a genetic defect. The

(13)

Chapt

er 1

12

protein CETP plays a key role in lipid metabolism and in reverse cholesterol transport by trans-ferring triglycerides and cholesterol between lipoproteins, leading to lower HDL cholesterol levels. chapter 6 evaluates whether the levels of CETP affects the postprandial changes in HDL cholesterol levels. In chapter 7 we present the results of a large case control study nested in a prospective population study. This study assesses the relation between a common CETP promoter polymorphism, which is associated with low CETP plasma levels and higher HDL cholesterol levels, and risk of CAD.

chapter 8 provides a review of CETP metabolism and CETP inhibitors. In chapter 9 we focus

on mechanisms by which inhibition of CETP might affect the process of atherosclerosis. Fur-thermore, we discuss the effects of CETP inhibition on fecal sterol excretion. In chapter 10 we describe the effects of CETP inhibition in subjects with isolated low HDL cholesterol levels. A summary of these studies is provided in chapter 11 as well as a perspective on the future of risk assessments and the role of CETP inhibition as a therapeutic target.

(14)

references

1. Yusuf S, Hawken S, Èunpuu S, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. The Lancet 364:937-952

2. Wilson PWF, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of Coronary Heart Disease Using Risk Factor Categories. Circulation 1998;97:1837-1847

3. Brindle P, Emberson J, Lampe F, et al. Predictive accuracy of the Framingham coronary risk score in British men: prospective cohort study. BMJ 2003;327:1267

4. Berneis KK, Krauss RM. Metabolic origins and clinical significance of LDL heterogeneity. J.Lipid Res. 2002;43:1363-1379

5. Shah PK. Low-Density Lipoprotein Lowering and Atherosclerosis Progression: Does More Mean Less? Circu-lation 2002;106:2039-2040

6. Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High density lipoprotein as a protective factor against coronary heart disease. The Framingham Study. Am J Med 1977;62:707-714

7. Genest JJ, Jr., Martin-Munley SS, McNamara JR, et al. Familial lipoprotein disorders in patients with prema-ture coronary artery disease. Circulation 1992;85:2025-2033

8. Gordon DJ, Probstfield JL, Garrison RJ, et al. High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies. Circulation 1989;79:8-15

9. deGoma EM, Leeper NJ, Heidenreich PA. Clinical Significance of High-Density Lipoprotein Cholesterol in Patients With Low Low-Density Lipoprotein Cholesterol. Journal of the American College of Cardiology 2008;51:49-55

10. Rader DJ, Alexander ET, Weibel GL, Billheimer J, Rothblat GH. Role of reverse cholesterol transport in animals and humans and relationship to atherosclerosis. J.Lipid Res. 2009;50:Suppl:S189-94

(15)

(16)

part I | chapter 2

value of ldl particle number and size as predictors of coronary

artery disease in apparently healthy men and women.

The epic-norfolk prospective population study

A Karim El Harchaoui, Wim A van der Steeg, Erik SG Stroes, Jan Albert Kuivenhoven, James D Otvos, Nicholas J Wareham, Barbara A Hutten, John JP Kastelein, Kay-Tee Khaw and S Matthijs Boekholdt

(17)

Chapt

er 2

16

absTracT

objectives: We assessed relations of low-density lipoprotein (LDL) particle number and LDL

particle size as measured by nuclear magnetic resonance (NMR) spectroscopy with LDL choles-terol and the risk of future coronary artery disease (CAD) in a cohort of healthy subjects.

background: Whereas LDL cholesterol is an established risk factor for CAD, its discriminative

power is limited. Measuring the number of LDL particles and size may have stronger associa-tions with CAD than LDL cholesterol.

methods: We conducted a case-control study nested in the prospective EPIC Norfolk study

which comprises 25663 apparently healthy subjects aged 45 to 79 years with moderately elevated LDL cholesterol (median 4.1 mmol/L). Cases (n= 1003) were individuals who devel-oped fatal or nonfatal CAD during 6 year follow-up. Controls (n=1885) were matched for age, sex and enrolment time. Odds ratios for future CAD were calculated by quartile of each LDL variable. We also evaluated whether LDL particle number could improve the Framingham Risk Score to predict CAD.

results: In univariate analyses LDL particle number (odds ratio in highest quartile: 2.00; 95% CI

1.58-2.59) and non-HDL cholesterol (odds ratio 2.14; 95% CI 1.69-2.69) were more closely associ-ated with CAD than LDL cholesterol (odds ratio in highest quartile: 1.73, 95% CI. 1.37 -2.18). The additional value of LDL particle number was lost after adjustment for high density lipoprotein cholesterol (HDL cholesterol) and triglyceride levels. Whereas LDL size was inversely related to CAD (odds ratio 0.60, 95% CI 0.47-0.76) in univariate analysis, this relation was abolished upon adjustment for LDL particle number. In a model adjusted for the Framingham Risk Score, LDL particle number retained its association with CAD (p for trend 0.02).

conclusion: In this large study of individuals with moderately elevated LDL cholesterol, LDL

particle number was related to CAD on top of Framingham Risk Score as well as after adjusting for LDL cholesterol. The additional value of LDL particle number was comparable to non-HDL cholesterol and it was abolished after adjusting for triglycerides and HDL cholesterol.

(18)

inTroducTion

The causal role of low-density lipoprotein (LDL) particles in the pathogenesis of coronary artery disease (CAD) is well established, as is the clinical benefit of lowering LDL in high risk patients. Hence, LDL cholesterol lowering is the principal target in cardiovascular preventive strategies (1). LDL cholesterol content is used as a parameter to estimate LDL-associated CAD risk. More recently, assessment of the number of LDL particles has been put forward as a more reliable method reflecting atherogenicity of the LDL-fraction (2). Since the cholesterol content per LDL particle exhibits large inter-individual variation due to differences in particle size as well as rela-tive content of cholesterol ester and triglycerides in the particle core, the information provided by LDL cholesterol and LDL particle number is not equivalent (3). Individuals with the same level of LDL cholesterol may have higher or lower numbers of LDL particles and, as a result, may differ in terms of absolute CAD risk. Prospective studies, in which LDL particle concentration was estimated by apolipoprotein B levels, have underscored stronger associations between LDL particle number and CAD risk compared to LDL cholesterol, particularly in subjects with normal LDL cholesterol concentration (4, 5). In addition, the size of LDL particles may also contribute to the atherogenicity of LDL cholesterol (6, 7). Thus, at a given level of LDL cholesterol, individuals with small LDL particles have greater atherosclerotic risk than those with large-size LDL (8, 9).

Lipoprotein particle analysis by nuclear magnetic resonance (NMR) spectroscopy is a rela-tively new method by which both LDL particle number and LDL particle size can be efficiently measured (10). We evaluated the associations between LDL particle number and LDL size, in comparison with LDL cholesterol and non-high density lipoprotein cholesterol (non-HDL cholesterol) as traditional markers, and risk of future CAD in apparently healthy men and women enrolled in a large prospective cohort with moderately elevated LDL cholesterol. Since LDL particle number and LDL size are closely related to traditional lipid factors such as HDL cholesterol and triglycerides, we performed multivariable analyses to assess independency of the correlations. We also assessed clinical usefulness of these novel parameters by determining their effect on the discriminative accuracy of the Framingham Risk Sore.

meThods

We performed a nested case-control study among participants of the EPIC (European Prospec-tive Investigation into Cancer and Nutrition)-Norfolk study, a prospecProspec-tive population study of 25663 men and women aged between 45 and 79 years, resident in Norfolk, UK, who completed a baseline questionnaire survey and attended a clinic visit (11). Participants were recruited from age-sex registers of general practices in Norfolk as part of the ten-country collaborative EPIC study designed to investigate dietary and other determinants of cancer. Additional data were obtained in EPIC-Norfolk to enable the assessment of determinants of other diseases.

(19)

Chapt

er 2

18

The design and methods of the study have been described in detail(11). In short, eligible participants were recruited by mail. At the baseline survey between 1993 and 1997, partici-pants completed a detailed health and lifestyle questionnaire. Non-fasting blood was taken by vein puncture into plain and citrate bottles. Blood samples were processed for assay at the Department of Clinical Biochemistry, University of Cambridge, or stored at –80˚C. All individu-als were flagged for death certification at the UK Office of National Statistics, with vital status ascertained for the entire cohort. In addition, participants admitted to hospital were identified using their unique National Health Service number by data linkage with ENCORE (East Norfolk Health Authority database). CAD was defined as codes 410-414 according to the International Classification of Diseases 9th revision. Participants were identified as having CAD during follow-up if they had a hospital admission and/or died with CAD as underlying cause. We report results with follow-up up to January 2003, an average of about 6 years. The study was approved by the Norwich District Health Authority Ethics Committee and all participants gave signed informed consent.

participants

We excluded all individuals who reported a history of heart attack or stroke at the baseline clinic visit. None of the cases or controls were on statin treatment. Cases were individuals who developed a fatal or non-fatal CAD during follow-up. For each case, two controls matched for gender, age (within 5 years), and time of enrolment (within 3 months), were identified who had remained free of CAD during follow-up.

nmr spectroscopy

Lipoprotein subclass particle concentrations and average size of LDL particles were mea-sured by proton NMR spectroscopy (LipoScience, Inc., North Carolina) as previously described (10). Particle concentrations of lipoprotein subclasses of different size were obtained directly from the measured amplitudes of their spectroscopically distinct lipid methyl group NMR sig-nals. The following LDL subclasses were defined: lDL (23-27 nm), large LDL (21.2-23 nm), small LDL (18-21.2 nm). LDL subclass particle concentrations are given in units of nmol/L. Summation of the LDL subclass levels provides total LDL (including IDL) particle concentrations. Weighted-average LDL particle sizes (in nm diameter units) are computed as the sum of the diameter of each subclass multiplied by its relative mass percentage as estimated from the amplitude of its methyl NMR signal. LDL subclass distributions determined by NMR and gradient gel electrophoresis are highly correlated (12). LDL subclass diameters, which are consistent with electron microscopy data (13), are uniformly ~5 nm smaller than those estimated by gradient gel electrophoresis.

(20)

biochemical analyses

Serum levels of total cholesterol, HDL cholesterol, and triglycerides were measured with the RA 1000 (Bayer Diagnostics, Basingstoke, UK), and LDL cholesterol levels were calculated with the Friedewald formula (14). NMR analysis was performed on stored serum samples which were analyzed in random order to avoid systemic bias. Researchers and laboratory personnel were blinded to identifiable information, and could identify samples by number only.

statistical analysis

Baseline characteristics were compared between cases and controls taking into account the matching. A mixed effect model was used for continuous variables and conditional logistic regression was used for categorical variables. Because triglyceride levels had a skewed distri-bution, values were log-transformed before being used as continuous variables in statistical analyses.

Spearman correlation coefficients and corresponding p-values were calculated to assess associations between the various biomarkers and established continuous CAD risk factors. To asses the strength of association between a risk factor and the occurrence of CAD, we calcu-lated odds ratios and corresponding 95% confidence intervals (95% CI) by conditional logistic regression analysis, taking into account matching for sex, age and enrolment time. Odds ratios were calculated per quartile of each risk factor, with the first quartile as the referent group. P-values represent significance for linearity across the odds ratios for the four quartiles of each risk factor. To compare the individual strengths of disease association of LDL particle number, non-HDL cholesterol, LDL cholesterol, HDL cholesterol and triglycerides, we calculated odds ratios for future CAD per quartile of each variable in separate models adjusted for smoking (yes/no) and systolic blood pressure. Since our objective was to determine relations of lipids/ lipoproteins with CAD, we did not adjust additionally for body mass index (BMI) and diabetes, two lipid-altering risk factors. Multivariable models were also examined to determine how relations of each variable were affected by adjustment for the other lipid/lipoprotein variables.

We further assessed the relation of LDL particle number and non-HDL cholesterol with future CAD using a model which included the Framingham Risk Score. We calculated this score using a previously reported algorithm(15) based on age, sex, levels of total and HDL choles-terol, systolic and diastolic blood pressure, diabetes and smoking and categorized subjects into three groups: low (< 10%), intermediate (10-20%) or high (> 20%) risk. Odds ratios for future CAD were calculated per quartile of the risk factor, adjusting for the Framingham Risk Score category.

Statistical analyses were performed using SPSS software (version 12.0.1, Chicago, Illinois). A p-value < 0.05 was considered to indicate statistical significance.

(21)

Chapt

er 2

20

resulTs

baseline characteristics

We identified 1003 participants who were apparently healthy at baseline and developed CAD during follow-up. A total of 882 cases were matched to two controls each, whereas the remain-ing 121 cases could be matched to one control only, givremain-ing a total number of 1885 controls. Baseline characteristics of cases and controls are listed in Table 1. As expected, cases were more likely to be smokers and diabetics, and to have a higher blood pressure and BMI than controls. Levels of total cholesterol, LDL cholesterol, non-HDL cholesterol and triglycerides were signifi-cantly higher in cases than in controls, whereas HDL cholesterol levels were signifisignifi-cantly lower

(p < 0.0001 for each). Baseline LDL particle number was higher in cases compared to controls (p < 0.0001). Levels of the large LDL subclass were not different, but cases had more IDL and small LDL particles. Thus, the increased LDL cholesterol in cases is attributable mainly to increased cholesterol in small LDL. The average LDL particle size was smaller in cases than controls.

Table 1. Baseline characteristics of coronary artery disease cases and matched controls

controls cases

(n = 1885) (n = 1003) P

Male sex % (n) 63.2 (1192) 63.9 (641) Matched

Age, yr 65 ± 8 65 ± 8 Matched

Diabetes % (n) 1.6 (30) 6.1 (61) < 0.0001

Smoking % (n) 8.3 (155) 15.7 (157) < 0.0001

BMI, kg/m2 _{26.2 ± 3.4} _{27.3 ± 3.9} _{< 0.0001}

Systolic blood pressure, mmHg 139 ± 18 144 ± 19 < 0.0001

Diastolic blood pressure, mmHg 84 ± 11 86 ± 12 < 0.0001

chemical lipid measures

Total cholesterol (mmol/L) 6.2 [5.5-6.9] 6.4 [5.6-7.2] <0.0001

HDL cholesterol (mmol/L) 1.3 [1.1-1.6] 1.2 [1.0-1.8] <0.0001

LDL cholesterol (mmol/L) 4.0 [3.4-4.7] 4.2 [3.6-4.9] <0.0001

Triglycerides (mmol/L) 1.6 [1.1-2.2] 1.8[1.3-2.6] <0.0001

non-HDL cholesterol (mmol/L) 4.8 [4.1-5.6] 5.2 [4.4-5.9] <0.0001

nmr ldl particle measures

LDL particle number (nmol/L) 1525 [1278-1812] 1640 [1383-1955] <0.0001

IDL (nmol/L) 36 [14-66] 43 [19-78] 0.003

Large LDL (nmol/L) 572 [448-704] 568 [427- 708] 0.6

Small LDL (nmol/L) 885 [637-1190] 999 [747-1330] <0.0001

LDL size (nm) 21.1 [20.7-21.5] 21.0 [20.1-21.4] 0.002

Data are presented as mean ±SD, percentage (n), or median [interquartile range]. Means, percentages, and medians may be based on fewer observations than the indicated number of subjects. LDL indicates low density lipoprotein; IDL indicates intermediate density lipoprotein.Triglyceride levels were log-transformed before analysis

(22)

associations of ldl measures with other cad risk factors

As shown in Table 2, LDL size was inversely correlated with LDL particle number (r= -0.58), but not with LDL cholesterol (r= 0.01). We identified a strong inverse relation of LDL size with triglyc-eride levels (r=-0.53) and BMI (r=-0.19) and a positive correlation with HDL cholesterol (r = 0.54). LDL cholesterol and LDL particle number were strongly interrelated (r= 0.63), but both

param-eters had markedly different associations with HDL cholesterol and triglycerides. Whereas LDL cholesterol was only weakly associated with HDL cholesterol (r= -0.03) and triglyceride levels (r= 0.18), LDL particle number was more strongly correlated with HDL cholesterol (r= -0.29) and triglycerides (r= 0.55). Non-HDL cholesterol was strongly correlated with LDL particle number (r= 0.79) and triglycerides (r= 0.47), and inversely correlated with HDL cholesterol (- 0.16) and LDL size (r= -0.18).

ldl particle number, traditional lipid risk factors and risk for future cad

Shown in the upper panel of Table 3 are the univariate odds ratios for future CAD associated with increasing quartiles of non-HDL cholesterol, LDL cholesterol, LDL particle number, HDL cholesterol, and triglycerides. The five lipid/lipoprotein measures exhibited comparable strengths of association with CAD, with odds ratios differing approximately two-fold comparing the highest and lowest quartiles. The magnitude of the predictive value of LDL particle number and non-HDL cholesterol were greater than that of LDL cholesterol. Comparing the highest to lowest quartile, the odds ratio for LDL particle number was 2.00 (95% CI 1.58-2.59) and 2.14 (95% CI 1.69-2.69) for non-HDL cholesterol versus 1.73 (95% CI 1.37-2.18) for LDL cholesterol.

The lower panels of Table 3 show the results of multivariate analyses in which each lipid/ lipoprotein variable was adjusted for the other 2 parameters to assess the independence of their mutual relations with CAD. The associations of both LDL cholesterol and LDL particle number with CAD were attenuated after adjustment for HDL cholesterol and triglycerides, whereas attenuation was more pronounced for LDL particle number than for LDL cholesterol (4th_{quartile odds ratios reduced from 2.00 to 1.37 for LDL particle number vs 1.73 to 1.55 for} Table 2. Spearman correlation coefficients between measured variables*

Non-HDL cholesterol LDL cholesterol LDL particle number LDL size LDL cholesterol 0.94 LDL particle number 0.76 0.63 LDL size -0.18 0.01† _-0.58 Total cholesterol 0.94 0.93 0.65 HDL cholesterol -0.16 -0.03§ _-0.29 _0.54 Triglycerides 0.47 0.18 0.55 -0.53 BMI 0.11 0.02‡ _0.14 _-0.19

*p <0.0001 for comparison unless otherwise indicated. †_{p= 0.6,}§_p=0.09,‡_p=0.2

(23)

Chapt

er 2

22

LDL cholesterol). Similarly, relations of HDL cholesterol and triglycerides with CAD were attenu-ated more by adjustment for LDL particle number than LDL cholesterol.

concordance/discordance between ldl cholesterol and ldl particle number

In a conditional logistic regression model that included both parameters and corrected for smoking and systolic blood pressure, LDL cholesterol was no longer statistically significantly associated with future CAD (Table 4). LDL particle number, however, remained a significant risk

Table 3. Odds ratios for future coronary artery disease by quartile of lipid/lipoprotein variable in univariable and multivariable models*

Quartiles 1 2 3 4 univariable models P† LDL cholesterol 1.00 1.37 1.38 1.73 <0.0001 (1.09-1.73) (1.09-1.74) (1.37-2.18) LDL particle number 1.00 1.23 1.48 2.00 <0.0001 (0.97-1.56) (1.17-1.87) (1.58-2.59) HDL cholesterol 1.00 0.76 0.60 0.50 <0.0001 (0.60-0.95) (0.48-0.75) (0.39-0.64) Triglycerides 1.00 1.27 1.50 2.01 <0.0001 (1.00-1.62) (1.20-1.88) (1.61-2.51) Non-HDL cholesterol 1.00 1.31 1.57 2.14 <0,0001 (1.04-1.66) (1.25-1.97) (1.69-2.69) multivariable models LDL cholesterol 1.00 1.26 1.27 1.55 0.001 (0.99-1.60) (1.00-1.61) (1.22-1.96) HDL cholesterol 1.00 0.83 0.71 0.66 0.001 (0.66-1.04) (0.56-0.89) (0.50-0.87) Triglycerides 1.00 1.12 1.22 1.52 0.001 (0.88-1.43) (0.96-1.54) (1.19-1.95) LDL particle number 1.00 1.13 1.21 1.37 0.02 (0.89-1.44) (0.94-1.54) (1.04-1.83) HDL cholesterol 1.00 0.86 0.74 0.70 0.005 (0.68-1.09) (0.59-0.94) (0.53-0.92) Triglycerides 1.00 1.11 1.19 1.36 0.03 (0.87-1.42) (0.93-1.51) (1.04-1.79) Non-HDL cholesterol 1.00 1.31 1.57 2.14 <0.0001 (1.04-1.66) (1.25-1.97) (1.69-2.69) HDL cholesterol 1.00 0.83 0.71 0.66 0.001 (0.66-1.05) (0.56-0.90) (0.50-0.87) Triglycerides 1.00 1.08 1.13 1.30 0.06 (0.85-1.39) (0.88-1.43) (0.99-1.70)

*Odds ratios (95% CIs) were calculated by conditional logistic regression, taking into account matching for sex, age and enrollment time and adjusted additionally for smoking and systolic blood pressure. Univariable models examined each lipid/lipoprotein variable in a separate model. Multivariable models examined each variable in a model adjusted for the other 2 lipid/lipoprotein variables as continuous variables.

(24)

factor, such that subjects in the highest quartile had an odds ratio of 1.78 (95% CI, 1.34 to 2.37; p<0.0001 for linearity) (Table 4).

Table 5. Odds ratios for future coronary artery disease by quartile of LDL size, with and without adjustment for LDL particle number*

ldl size Quartile 1 2 3 4 †_P Range (nm) <20.6 20.7-21.0 21.1-21.4 >21.4 Unadjusted for LDL 1.00 0.77 0.76 0.60 <0.0001 Particle number (0.62-0.97) (0.60-0.95) (0.47-0.76) Adjusted for LDL 1.00 0.92 0.99 0.86 0.5 Particle number 0.72-1.16) 0.77-1.28) 0.65-1.15)

*Odds ratios (95% CIs) were calculated by conditional logistic regression adjusted for smoking and systolic blood pressure, with and without additional adjustment for LDL particle number.

†_{P for linear trend}

Table 6. Odds ratios for future coronary artery disease after taking into account the Framingham Risk Score* Quartiles 1 2 3 4 †_P LDL cholesterol 1.00 1.17 1.09 1.24 0.15 (0.93-1.48) (0.86-1.39) (0.97-1.58) LDL particle number 1.00 1.10 1.22 1.34 0.02 (0.86-1.39) (0.96-1.54) (1.03-1.73) Non-HDL cholesterol 1.00 0.98 1.03 1.38 0.04 (0.73-1.32) (0.76-1.38) (1.01-1.90)

*Odds ratios were calculated by conditional logistic regression, with adjustment for risk categories based on the Framingham Risk Score.

†_{P for linear trend}

Table 4. Odds ratios for future coronary artery disease by quartile of LDL cholesterol and LDL particle number, both entered into one model*

ldl size Quartile 1 2 3 4 †_P

LDL cholesterol 1.00 1.23 1.13 1.22 0.3

(0.97-1.57) (0.88-1.45) (0.92-1.61)

LDL particle number 1.00 1.18 1.39 1.78 <0.0001

(0.93-1.51) (1.08-1.79) (1.34-2.37)

*Odds ratios (95% CIs) were calculated by conditional logistic regression adjusted for smoking and systolic blood pressure.

(25)

Chapt

er 2

24

ldl size and risk for future cad

Table 5 shows the odds ratios for future CAD associated with increasing quartiles of LDL size adjusted for smoking and systolic blood pressure, with and without additional adjustment for the potentially confounding inverse correlation of LDL size with LDL particle number. Without adjustment for LDL particle number, there was a significant relation of smaller LDL size with CAD, with an odds ratio for individuals in the highest quartile compared with those in the low-est quartile of 0.60 (95% CI 0.47-0.76). However, upon adjustment for LDL particle number, the relation of LDL size with CAD was greatly attenuated and was no longer significant.

ldl particle number and the framingham risk score

The results in Table 6 indicate that LDL particle number and non-HDL cholesterol retain a similar association with future CAD after accounting for the Framingham Risk Score (odds ratio of 1.34 and 1.38 respectively in the highest versus lowest quartile).

discussion

Measurements of LDL particle number and LDL size have the potential to improve coronary disease risk assessment as well as decisions about LDL-treatment intensity, since they account for aspects of lipid-atherogenicity that are incompletely reflected by values of LDL cholesterol. In this large prospective case-control study, we show that LDL particle number and non-HDL cholesterol were more closely associated with the occurrence of future CAD than levels of LDL cholesterol. Upon adjusting for HDL cholesterol and triglyceride levels, the predictive capacity of LDL particle number was comparable to that of LDL cholesterol. Whereas LDL size was related to CAD risk, this relationship was abolished after adjusting for LDL particle number. Both LDL particle number and non-HDL cholesterol had incremental value on top of the Framingham Risk Score in multivariate analyses. Overall, these findings do not support routine use of LDL particle number for CAD risk assessment in primary prevention setting.

ldl particle number and risk for future cad

The cholesterol content of LDL particles, which can create discordance between levels of LDL cholesterol and LDL particle number, is mainly influenced by cholesterol ester transfer protein (CETP) activity, which is enhanced under circumstances of hypertriglyceridemia (8). This has two important consequences. First, transfer of cholesteryl ester from HDL particles to apoli-poprotein B-containing liapoli-poproteins causes low HDL cholesterol levels. Second, cholesterol depletion and triglyceride enrichment of LDL particles facilitates the formation of small dense LDL particles (3, 8). As a result, LDL cholesterol levels generally underestimate the number of LDL particles in individuals with elevated triglycerides, as clearly illustrated in the Framingham Study (2, 3). Discordance between LDL cholesterol and LDL particle number is also a feature of

(26)

diabetic patients (2, 16) and the number of metabolic syndrome components (2, 17). Our data in EPIC-Norfolk showing that triglyceride and HDL cholesterol levels are much more strongly correlated with LDL particle number than LDL cholesterol are in agreement with these findings. Accordingly, it was not unexpected that the relation of LDL particle number with CAD risk was weakened more than that of LDL cholesterol by multivariate adjustment for triglycerides and HDL cholesterol. Conversely, adjusting for LDL particle number weakened relations of CAD with triglycerides and HDL cholesterol, suggesting that some of the risk associated with these non-LDL risk factors may actually stem from elevations of LDL particle number not reflected by levels of LDL cholesterol.

We found a moderate degree of discordance between LDL cholesterol and LDL particle number in our study population. While the source of this excess risk may not be due entirely to the elevated LDL particle number, since those with discordantly high LDL particle number often have increased triglycerides and decreased HDL cholesterol, it is biologically plausible that LDL particle number makes a contribution. Current understanding of the pathophysiology of atherosclerotic vascular disease is that LDL particles are active participants from the time they enter the artery wall, are retained in the intima by binding to extracellular matrix, become chemically modified by oxidation, and are subsequently ingested by macrophages to create foam cells and increased plaque burden (2, 4).

Despite the finding that LDL particle number predicted CAD independently of the Framingham Risk Score, our results showing a loss of discriminative power of LDL particle number over LDL cholesterol when HDL cholesterol and triglyceride levels are accounted for, do not argue for the routine implementation of LDL particle number assessment for CAD risk assessment. LDL particle number may, however, play a useful role in patient management by helping judge the adequacy of LDL lowering therapy, particularly among those with elevated triglycerides and reduced HDL cholesterol. Such a role has been proposed for apolipoprotein B (4,5), and both non-HDL cholesterol and apolipoprotein B have been put forward as secondary treatment targets after LDL cholesterol goals have been achieved (1,4). In fact, data supporting LDL par-ticle number as an alternative treatment target has gained support from clinical intervention studies showing that on-treatment levels of apolipoprotein B or NMR-measured LDL particle number were more reliable indicators of residual CAD risk than on-treatment LDL cholesterol (2, 4, 5, 18, 19).

ldl size and risk for future cad

Small LDL size is another factor associated with high triglycerides, low HDL cholesterol, obesity, insulin resistance, diabetes, and metabolic syndrome (6-9). Our finding of a relation between smaller LDL size and greater CAD risk is consistent with previous studies reporting that small LDL particles have higher atherogenic potential than large LDL particles (8, 9). However, upon adjustment for LDL particle number, the relationship of LDL size with CAD was abolished. This result is consistent with findings in the Multi-Ethnic Study of Atherosclerosis indicating

(27)

Chapt

er 2

26

that NMR-measured numbers of small and large LDL particles are related similarly to carotid atherosclerosis (20).

non-hdl cholesterol and risk for future cad

The present findings confirm that non-HDL cholesterol is a better predictor of CAD than LDL cholesterol (21,22). Previous studies have also found that the number of most atherogenic lipoprotein particles, as measured by apolipoprotein B, were more strongly related to CAD risk than was non-HDL cholesterol (21, 22). In the present study we show that the association of LDL particle number with CAD is almost equal to that of non-HDL cholesterol. The fact that apolipoprotein B captures all atherogenic apolipoprotein B particles (including VLDL and LDL), whereas LDL particle number only measures LDL particles, may have contributed to this dis-tinction. There has been an intense debate concerning clinical relevance of measuring particle numbers and/or cholesterol content of the particles. In the present study, we observed that both LDL particle number as well as non-HDL cholesterol conferred predictive value on top of the Framingham Risk Score. Since the association between LDL particle number and CAD was equal to that of non-HDL cholesterol, the present findings do not advocate routine use of LDL particle number in CAD risk assessment. The potential value of LDL particle number measurement for monitoring patients on lipid lowering medication needs to be addressed in separate intervention trials.

study limitations

Our study has several limitations. The study population was relatively elderly, which may limit the generalizability of our results. Case-control differences in CAD among older populations are more weakly related to lipoprotein levels than in younger populations because many of the control subjects have extensive subclinical disease, yet have not experienced a coronary event (23). LDL cholesterol levels in the study population were also considerably higher than in the general U.S. population, with a mean LDL cholesterol value in EPIC-Norfolk corresponding to the 80th_{percentile of Framingham subjects of similar age and gender (24). CAD events in our}

study were ascertained through death certification and hospital admission data, which may lead both to under ascertainment and to misclassification of cases. Previous validation studies in this cohort, however, indicate high specificity of such case ascertainment (25).

conclusion

In conclusion, in this large cohort of apparently healthy men and women, LDL particle number and non-HDL cholesterol were more closely associated than LDL cholesterol with the occur-rence of future CAD. After adjustment for HDL cholesterol and triglycerides, LDL particle num-ber was no longer more predictive than LDL cholesterol. These findings do not support routine use of LDL particle number in CAD risk assessment strategies in primary prevention strategies. However recognition that patients with low HDL cholesterol and/or high triglycerides often

(28)

have elevated numbers of LDL particles without having elevated LDL cholesterol may enable their LDL-related CAD risk to be managed more effectively.

acknowledgemenTs

Liposcience Inc (Jim Otvos) kindly performed all NMR spectroscopy measurements in the cohort.

(29)

Chapt

er 2

28

references

1. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) Final Report. Circulation 2002; 106:3143.

2. Cromwell WC, Otvos JD. Low-density lipoprotein particle number and risk for cardiovascular disease. Curr Atheroscler Rep 2004; 6:381-7.

3. Otvos JD, Jeyarajah EJ, Cromwell WC. Measurement issues related to lipoprotein heterogeneity. Am J Cardiol 2002; 90:22i-9i.

4. Barter PJ, Ballantyne CM, Carmena R, et al. Apo B versus cholesterol in estimating cardiovascular risk and in guiding therapy: report of the thirty-person/ten-country panel. J Intern Med 2006; 259:247-58.

5. Sniderman AD, Furberg CD, Keech A, et al. Apolipoproteins versus lipids as indices of coronary risk and as targets for statin treatment. Lancet 2003; 361:777-80.

6. Gardner CD, Fortmann SP, Krauss RM. Association of small low-density lipoprotein particles with the inci-dence of coronary artery disease in men and women. JAMA 1996; 276:875-81.

7. Lamarche B, Tchernof A, Moorjani S, et al. Small, Dense Low-Density Lipoprotein Particles as a Predictor of the Risk of Ischemic Heart Disease in Men: Prospective Results From the Quebec Cardiovascular Study. Circulation 1997; 95:69-75.

8. Berneis KK, Krauss RM. Metabolic origins and clinical significance of LDL heterogeneity. J Lipid Res 2002; 43:1363-79.

9. Sacks FM, Campos H. Clinical review 163: Cardiovascular endocrinology: Low-density lipoprotein size and cardiovascular disease: a reappraisal. J Clin Endocrinol Metab 2003; 88:4525-32.

10. Otvos JD. Measurement of lipoprotein subclass profiles by nuclear magnetic resonance spectroscopy. Clin Lab 2002; 48:171-80.

11. Day N, Oakes S, Luben R, et al. EPIC-Norfolk: study design and characteristics of the cohort. European Pro-spective Investigation of Cancer. Br J Cancer 1999; 80 Suppl 1:95-103.

12. Blake GJ, Otvos JD, Rifai N, Ridker PM. Low-Density Lipoprotein Particle Concentration and Size as Deter-mined by Nuclear Magnetic Resonance Spectroscopy as Predictors of Cardiovascular Disease in Women. Circulation 2002; 106:1930-7.

13. Rumsey SC, Galeano NF, Arad Y, Deckelbaum RJ. Cryopreservation with sucrose maintains normal physical and biological properties of human plasma low density lipoproteins. J Lipid Res 1992; 33:1551-61. 14. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein

choles-terol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972; 18:499-502.

15. Wilson PWF, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of Coronary Heart Disease Using Risk Factor Categories. Circulation 1998; 97:1837-47.

16. Garvey WT, Kwon S, Zheng D, et al. Effects of insulin resistance and type 2 diabetes on lipoprotein subclass particle size and concentration determined by nuclear magnetic resonance. Diabetes 2003; 52:453-62. 17. Kathiresan S, Otvos JD, Sullivan LM, et al. Increased small low-density lipoprotein particle number: a

promi-nent feature of the metabolic syndrome in the Framingham Heart Study. Circulation 2006; 113:20-9. 18. Otvos JD, Collins D, Freedman DS, et al. Low-density lipoprotein and high-density lipoprotein particle

subclasses predict coronary events and are favorably changed by gemfibrozil therapy in the Veterans Affairs High-Density Lipoprotein Intervention Trial. Circulation 2006; 113:1556-63.

19. Rosenson RS, Otvos JD, Freedman DS. Relations of lipoprotein subclass levels and low-density lipoprotein size to progression of coronary artery disease in the Pravastatin Limitation of Atherosclerosis in the Coronary Arteries (PLAC-I) trial. Am J Cardiol 2002; 90:89-94.

20. Mora S, Szklo M, Otvos JD, et al. LDL particle subclasses, LDL particle size, and carotid atherosclerosis in the Multi-Ethnic Study of Atherosclerosis (MESA). Atherosclerosis 2006.

21. Pischon T, Girman CJ, Sacks FM, Rifai N, Stampfer MJ, Rimm EB. Non-High-Density Lipoprotein Cholesterol and Apolipoprotein B in the Prediction of Coronary Heart Disease in Men. Circulation 2005; 112:3375-83.

(30)

22. Cui Y, Blumenthal RS, Flaws JA, et al. Non-High-Density Lipoprotein Cholesterol Level as a Predictor of Cardiovascular Disease Mortality. Arch Intern Med 2001; 161:1413-9.

23. Kuller L, Arnold A, Tracy R, et al. Nuclear magnetic resonance spectroscopy of lipoproteins and risk of coro-nary heart disease in the cardiovascular health study. Arterioscler Thromb Vasc Biol 2002; 22:1175-80. 24. Freedman DS, Otvos JD, Jeyarajah EJ, et al. Sex and age differences in lipoprotein subclasses measured by

nuclear magnetic resonance spectroscopy: the Framingham Study. Clin Chem 2004; 50:1189-200. 25. Boekholdt SM, Peters RJ, Day NE, et al. Macrophage migration inhibitory factor and the risk of myocardial

infarction or death due to coronary artery disease in adults without prior myocardial infarction or stroke: the EPIC-Norfolk Prospective Population study. Am J Med 2004; 117:390-7.

(31)

(32)

part I | chapter 3

high density lipoprotein size and particle concentration and

coronary risk

A Karim El Harchaoui, Benoit J Arsenault, Remco Franssen, Jean-Pierre Després, G Kees Hovingh, Erik SG Stroes, James D Otvos, Nicholas J Wareham, John JP Kastelein, Kay-Tee Khaw, S Matthijs Boekholdt.

(33)

Chapt

er 3

32

absTracT

background: High-density lipoprotein (HDL) cholesterol levels are inversely related to risk of

coronary artery disease (CAD). Since HDL particles are heterogeneous in size and composition, they may be differentially associated with other cardiovascular risk factors and cardiovascular risk per se.

objective: To study the independent relationships of HDL size and HDL particle concentration

to risk of future CAD.

design: Nested case-control study within the EPIC-Norfolk cohort. Baseline survey between

1993 and 1997, follow-up until November 2003.

setting: Norfolk, United Kingdom.

participants: Cases were 822 apparently healthy men and women who developed CAD during

follow-up. Controls were 1401 participants who remained free of CAD, and were matched to cases by sex, age and enrollment time.

measurements: First CAD events leading to either hospitalization or death.

results: Nuclear magnetic resonance spectroscopy-measured HDL particle concentration

(33.9±5 vs 32.9±6 µmol/l; p<0.001) and HDL size (8.9±0.5 vs 8.8±0.5 nm; p < 0.001) as well as gradient gel electrophoresis-measured HDL size (88.6±4.2 vs 88.1±4.3 nm; p=0.005) were lower in cases compared to controls. HDL size and HDL particle concentration were only weakly correlated (for nuclear magnetic resonance spectroscopy-measured r= 0.08, for gradient gel electrophoresis-measured r=0.10). HDL size was strongly associated with risk factors character-istic of the metabolic syndrome, including waist-to-hip ratio, triglycerides, and apolipoprotein B, whereas HDL particle concentration was not. HDL size and HDL particle concentration were independently associated with CAD risk. The association between HDL size and CAD risk was abolished upon adjustment for apolipoprotein B and triglycerides (adjusted odds ratio,1.00 [95% CI 0.71 to 1.39] for top versus bottom quartile), whereas HDL particle concentration remained independently associated with CAD risk (adjusted odds ratio, 0.50 [95% CI 0.37 to 0.66]).

limitations: Measurements were performed in non-fasting blood samples and residual

con-founding cannot be excluded.

conclusions: HDL size and HDL particle concentration were independently associated with

other cardiovascular risk factors and with the risk of developing CAD. The relationship between HDL size and CAD risk was completely explained by markers associated with the metabolic syndrome, indicating that part of the relationship between HDL cholesterol and CAD risk is merely a reflection of this constellation of metabolic risk.

(34)

inTroducTion

The strong inverse relationship between high-density lipoprotein (HDL) cholesterol levels and risk of coronary artery disease (CAD) is well established (1). However, at any given level of HDL cholesterol, HDL particle concentration and HDL size distribution (see glossary table for definitions) may differ substantially between individuals. The atheroprotective role of HDL cholesterol is believed to be mediated mainly by its role in reverse cholesterol transport, the biological pathway that facilitates removal of cholesterol from macrophages in the arterial wall back to the liver (2). Substantial evidence suggests that HDL particles are heterogeneous in their efficacy to facilitate ATP-binding cassette A1 (ABCA1)-mediated cholesterol efflux from macrophages and scavenger receptor class BI (SR-BI)-mediated hepatic uptake of cholesterol from HDL particles. In particular, this heterogeneity has been associated with HDL particle size (3-5). In addition to its role in reverse cholesterol transport, accumulating evidence suggests that HDL particles have anti-coagulant, anti-oxidative and anti-inflammatory properties which contribute to the anti-atherogenic capacity of HDL. In fact, HDL lipoproteins were recently shown to carry a wide array of proteins that mediate these properties (6). The binding affinity of these proteins to the surface of HDL lipoproteins may depend on their size (7;8). Under certain conditions, HDL lipoproteins may lose their anti-atherogenic capacity and become dysfunc-tional (9;10). As a consequence, various HDL lipoprotein subpopulations may differ substan-tially in their capacity to play an atheroprotective role. Despite this functional heterogeneity, HDL has thus far been regarded as a single entity in epidemiological studies and the amount of cholesterol transported by HDL particles is traditionally being used for this purpose. However, we have recently shown that higher HDL cholesterol levels may not necessarily be associated with lower cardiovascular risk (11) and the HDL cholesterol raising drug torcetrapib has recently been shown to have detrimental effects (12). These findings emphasize that a broader perspec-tive on HDL metabolism is warranted.

We hypothesized that HDL particle concentration and HDL size distribution are differen-tially associated with cardiovascular risk factors and with cardiovascular risk. We tested these hypotheses in a case-control study nested in the EPIC-Norfolk cohort.

meThods

We performed a nested case-control study among participants of the European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk study, a prospective population study of 25,663 men and women aged between 45 and 79 years, resident in Norfolk, United Kingdom, who completed a baseline questionnaire survey and attended a clinic visit (Figure). Participants were recruited from age-sex registers of general practices in Norfolk as part of the ten-country collaborative EPIC study designed to investigate dietary and other determinants of cancer.

(35)

Chapt

er 3

34

Additional data were obtained in EPIC-Norfolk to enable the assessment of determinants of other diseases.

The design and methods of the study have been described in detail (13). In short, eligible participants were recruited by mail. At the baseline survey between 1993 and 1997, partici-pants completed a detailed health and lifestyle questionnaire. Non-fasting blood samples were obtained by venipuncture into plain and citrate bottles. Blood samples were processed for assay at the Department of Clinical Biochemistry, University of Cambridge, or stored at –80˚C. All individuals have been flagged for death certification at the United Kingdom Office of National Statistics, with vital status ascertained for the entire cohort. In addition, participants admitted to hospital were identified using their unique National Health Service number by data linkage with the East Norfolk Health Authority (ENCORE) database, which identifies all hospital contacts throughout England and Wales for Norfolk residents. CAD was defined as codes 410-414 according to the International Classification of Diseases 9th revision. Participants were identified as having CAD during follow-up if they had a hospital admission and/or died with CAD as underlying cause. Previous validation studies in our cohort indicate high specificity of such case ascertainment (14). We report results with follow-up up to November 2003, an aver-age of about 6 years. The study was approved by the Norwich District Health Authority Ethics Committee and all participants gave signed informed consent.

Glossary

Adenosine triphospate-binding cassette A1 (ABCA1)

A cell membrane transporter that facilitates the delivery of cholesterol from cells to lipid-poor apolipoprotein A-I in the extracellular space High density lipoproteins (HDL) The smallest (7.0 nm and 13 nm) and most dense (between 1.036 g/

mL and 1.25 g/mL) of the plasma lipoproteins. They are heterogenous, comprising several subpopulations that vary in shape, size, composition and surface charge

HDL cholesterol The cholesterol content carried in all HDL particles

HDL particle concentration Number of HDL particles per plasma volume, expressed in µmol/L.

HDL size The mass-weighted average diameter of the HDL particles in a particular

plasma sample

Large HDLs HDL particles with diameter between 7.3-8.2 nm

Medium HDLs HDL particles with diameter between 8.2-8.8 nm

Myeloperoxidase Leukocyte-derived enzyme with a role in the immune system.

Paraoxonase-1 Enzyme located on the surface of HDL is believed to protect against the

oxidation of low-density lipoprotein (LDL) and therefore to affect the risk of coronary artery disease

Small HDL HDL particles with diameter between 8.8-13 nm

Scavenger receptor class B1 (SR-BI) HDL receptor, mainly present in the liver, that promotes the selectively uptake of HDL cholesterol

(36)

participants

We have previously described other analyses within this prospective nested case-control study (14;15). Briefly, we excluded all individuals who reported a history of heart attack/stroke or use of lipid lowering drugs at the baseline clinic visit. Cases were individuals who developed fatal or non-fatal CAD during follow-up until November 2003. Controls were study participants who remained free of any cardiovascular disease during follow-up. We matched two controls

Cases (n=1138) Controls (n=2237) EPIC-Norfolk Prospective Population Study (n=25.663)

Nested case-control study (n=3375) Controls (n=1807) Excluded (total n =270)* •no LDL cholesterol (n=75) •no HDL cholesterol (n=77) •no Triglycerides (n=15) •no blood pressure (n=2) •no ApoA1 level (n=161) •no ApoB level (n=109) •no CRP level (n=30) •no NMR data (n=10) •no Waist hip ratio (n=2)

Excluded (total n =430)*

•no LDL cholesterol (n=112) •no HDL cholesterol (n=111) •no Triglycerides (n=31) •no blood pressure (n=5) •no ApoA1 level (n=289) •no ApoB level (n=171) •no CRP level (n=27) •no NMR data (n=7) •no Waist hip ratio (n=2)

Cases (n=868)

Excluded for lack of matching

control participants (n=46) Excluded for lack of matching case patients (n=406)

Analyzed Analyzed (n=822) (n=1401)

figure. Study flow diagram

ApoAI = apolipoprotein A-I; ApoB = apolipopotein B; CRP = C-reactive protein; EPIC = European Prospective Investigation into Cancer and Nutrition; HDL = high-density lipoprotein; LDL = low-density lipoprotein; NMR = nuclear magnetic resonance. *_{Individual case patients and control participants may} have several missing variables.

(37)

Chapt

er 3

36

to each case by age (within 5 years), sex and time of enrolment (within 3 months). The current analysis was performed on all participants who had a complete dataset available for baseline characteristics, apolipoproteins A-I and B, gradient gel electrophoresis-measured HDL size and lipoprotein nuclear magnetic resonance (NMR) spectroscopy.

We selected 1138 people who were apparently healthy at baseline but did develop CAD during follow-up. We aimed to select 2 controls for every case who were healthy at baseline and remained free of cardiovascular disease during follow-up. From the original dataset of 1138 cases and 2237 controls, 270 cases and 430 control patients were excluded because at least one value was missing for any of the parameters mentioned above. A total of 46 cases were excluded because they had no matching control patients, as were 406 control patients because they had no matching cases. The analyses are therefore based on a dataset containing 822 cases and 1401 control patients (243 cases with 1 matching control patient, 579 cases were matched with 2 control patients).

measurements

Data on smoking and alcohol consumption were obtained by health questionnaires at the baseline clinic visit. Physical activity was obtained with a Physical Activity questionnaire (EPAQ2) (13). In the original cohort smoking was classified into current cigarette smokers, former smok-ers and never smoksmok-ers. In the present study smoking was recoded into yes/no. Physical activity was classified in four categories: inactive, moderately active, moderately inactive and active. Use of large amounts of alcohol was defined as more than 21 units of alcohol/week. Diabetes mellitus and use of hormone replacement therapy was self-reported. Participants were asked about medical history with the question “Has a doctor ever told you that you have any of the following?”, followed by a number of choices including diabetes. Blood pressure was recorded by taking two measurements of diastolic and systolic blood pressure using the Accutor Sphyg-momanometer (Datascope, UK), after 3 min of resting.

Serum levels of total cholesterol, HDL cholesterol and triglycerides were measured on fresh samples with the RA 1000 (Bayer Diagnostics, Basingstoke, United Kingdom). Low-density lipo-protein (LDL) cholesterol levels were calculated with the Friedewald formula to closely approach current clinical procedures. Plasma concentrations of C-reactive protein were measured with a sandwich-type enzyme linked immuno sorbent assay as previously described (16). Serum levels of apolipoprotein A-I and apolipoprotein B were measured by rate immunonephelometry (Behring Nephelometer BNII, Marburg, Germany) with calibration traceable to the International Federation of Clinical Chemistry primary standards (17). Serum concentration of myeloperoxi-dase was measured by use of a commercially available enzyme-linked immuno sorbent assay (CardioMPO Test, Prognostix, Cleveland, Ohio, USA) (15). Paraoxonase-1 activity was measured as previously described (18). HDL size was measured by 4-30% nondenaturing polyacrylamide gradient gel electrophoresis (GGE) as previously described (19). Lipoprotein subclass particle concentrations and average size of particles were also measured by proton NMR spectroscopy

(38)

(LipoScience, Inc., NC, USA) as previously described (20). In brief, particle concentrations of lipoprotein subclasses of different size were obtained directly from the measured amplitudes of their spectroscopically distinct lipid methyl group NMR signals. HDL particle concentrations are expressed in µmoles of particles per liter (µmol/L). Summation of the HDL subclass levels provides total HDL particle concentration. The following 5 HDL subclasses were defined: H5 (10-13 nm, mean 11.5 nm), H4 (8.8-10 nm, mean 9.4 nm), H3 (8.2-8.8 nm, mean 8.5 nm), H2 (7.8-8.2 nm, mean 8.0 nm), H1 (7.3-7.7 nm, mean 7.5 nm). These diameter range estimates were based on size measurements of the isolated HDL subclass reference standards by GGE. The HDL subclasses H5, H4, H3, H2 and H1 are closely related to the GGE subclass designations 2b, 2a, 3a, 3b and 3c, respectively (21). To simplify data analysis and interpretation, we grouped these HDL subclasses as follows: Small HDL (H1 + H2), Medium HDL (H3), Large HDL (H4 + H5).(22) NMR-measured HDL size was calculated as the mass-weighted average diameter of the HDL particles in a particular plasma sample. The average HDL particle size is computed as the sum of the diameter of each subclass multiplied by its relative mass percentage as estimated from the amplitude of its measured NMR signal. It has been shown that the reproducibility of NMR lipoprotein profile assessments is very good, storing at -70°C and thawing has no significant influence on the quality of the HDL associated measurements (20). Samples were analyzed in random order to avoid systemic bias. Researchers and laboratory personnel were blinded to identifiable information, and could identify samples by number only.

statistical analysis

The hypotheses of the present study were generated prior to data analysis. Baseline character-istics were compared between cases and controls taking into account the matching. A mixed effect model was used for continuous variables and conditional logistic regression was used for categorical variables. Pearson’s correlation coefficients and corresponding p-values were cal-culated to assess associations between HDL-related parameters, metabolic and inflammatory covariates. Mean values of risk factors were calculated per quartile of HDL particle concentration and HDL size. Differences between categories were calculated by analysis of variance (ANOVA). In order to assess the strength of the associations between HDL particle concentration or HDL size and the risk of future CAD, we calculated odds ratios and corresponding 95% confidence intervals (95% CI) using conditional logistic regression, taking into account matching for sex, age and enrollment time and additionally adjusting for smoking. Regression analyses were also performed with additional adjustment for the inflammatory covariates myeloperoxidase, para-oxonase and C-reactive protein levels and for the metabolic covariates apolipoprotein B, and log-transformed triglycerides. The first quartile was used as reference group. P-values represent significance for linearity across the quartiles. Statistical analyses were performed using SPSS software (version 12.0.1, Chicago, Illinois). A P-value < 0.05 was considered to indicate statistical significance.

The puzzle of high-density lipoprotein in cardiovascular prevention - Thesis

UvA-DARE (Digital Academic Repository)