Advertisement

Relationship of lipid oxidation with subclinical atherosclerosis and 10-year coronary events in general population

  • Miquel Gómez
    Affiliations
    Cardiology Department, Parc de Salut Mar, Barcelona, Spain

    Medicine Department, Autonomous University of Barcelona, Barcelona, Spain

    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain
    Search for articles by this author
  • Joan Vila
    Affiliations
    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    Cardiovascular and Genetic Epidemiology Research Group (EGEC, REGICOR Study Group), IMIM, Spain

    CIBER of Epidemiology and Public Health (CIBERESP), Spain
    Search for articles by this author
  • Roberto Elosua
    Affiliations
    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    Cardiovascular and Genetic Epidemiology Research Group (EGEC, REGICOR Study Group), IMIM, Spain
    Search for articles by this author
  • Lluís Molina
    Affiliations
    Cardiology Department, Parc de Salut Mar, Barcelona, Spain

    Medicine Department, Autonomous University of Barcelona, Barcelona, Spain

    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    CIBER of Epidemiology and Public Health (CIBERESP), Spain
    Search for articles by this author
  • Jordi Bruguera
    Affiliations
    Cardiology Department, Parc de Salut Mar, Barcelona, Spain

    Medicine Department, Autonomous University of Barcelona, Barcelona, Spain

    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain
    Search for articles by this author
  • Joan Sala
    Affiliations
    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    Cardiology Department, Hospital Josep Trueta, Girona, Spain
    Search for articles by this author
  • Rafel Masià
    Affiliations
    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    Cardiology Department, Hospital Josep Trueta, Girona, Spain
    Search for articles by this author
  • Maria Isabel Covas
    Affiliations
    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    Cardiovascular Risk and Nutrition Research Group (CARIN, REGICOR Study Group), IMIM, Spain

    CIBER of Obesity Physiopathology and Nutrition (CIBEROBN), Spain
    Search for articles by this author
  • Author Footnotes
    1 Equally contributed as senior co-authors.
    Jaume Marrugat
    Footnotes
    1 Equally contributed as senior co-authors.
    Affiliations
    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    Cardiovascular and Genetic Epidemiology Research Group (EGEC, REGICOR Study Group), IMIM, Spain
    Search for articles by this author
  • Author Footnotes
    1 Equally contributed as senior co-authors.
    Montserrat Fitó
    Correspondence
    Corresponding author. Cardiovascular Risk and Nutrition Research Group (CARIN), IMIM (Hospital del Mar Medical Research Institute), Dr. Aiguader, 88, 08003 Barcelona, Spain. Tel.: +34 93 3160724; fax: +34 93 2257550.
    Footnotes
    1 Equally contributed as senior co-authors.
    Affiliations
    Inflammatory and Cardiovascular Disease Programme (RICAD), IMIM-Hospital del Mar Medical Research Institute, Barcelona, Spain

    Cardiovascular Risk and Nutrition Research Group (CARIN, REGICOR Study Group), IMIM, Spain

    CIBER of Obesity Physiopathology and Nutrition (CIBEROBN), Spain
    Search for articles by this author
  • Author Footnotes
    1 Equally contributed as senior co-authors.

      Highlights

      • Oxidized LDL was independently associated with 10-year coronary clinical events.
      • Oxidized LDL was not associated with 10-year coronary subclinical atherosclerosis.
      • Oxidized LDL improved the reclassification capacity of Framingham risk functions.

      Abstract

      Objectives

      To assess 1) the association of lipid oxidation biomarkers with 10-year coronary artery disease (CAD) events and subclinical atherosclerosis, and 2) the reclassification capacity of these biomarkers over Framingham-derived CAD risk functions, in a general population.

      Methods

      Within the framework of the REGICOR study, 4782 individuals aged between 25 and 74 years were recruited in a population-based cohort study. Follow-up of the 4042 who met the eligibility criteria was carried out. Plasma, circulating oxidized low-density lipoprotein (oxLDL) and oxLDL antibodies (OLAB) were measured in a random sample of 2793 participants.
      End-points included fatal and non-fatal acute myocardial infarction (AMI) and angina. Carotid intima-media thickness (IMT) in the highest quintile and ankle-brachial index <0.9 were considered indicators of subclinical atherosclerosis.

      Results

      Mean age was 50.0 (13.4) years, and 52.4% were women. There were 103 CAD events (34 myocardial infarction, 43 angina, 26 coronary deaths), and 306 subclinical atherosclerosis cases. Oxidized LDL was independently associated with higher incidence of CAD events (HR = 1.70; 95% Confidence Interval: 1.02–2.84), but not with subclinical atherosclerosis. The net classification index of the Framingham-derived CAD risk function was significantly improved when ox-LDL was included (NRI = 14.67% [4.90; 24.45], P = 0.003). No associations were found between OLAB and clinical or subclinical events. The reference values for oxLDL and OLAB are also provided (percentiles).

      Conclusions

      OxLDL was independently associated with 10-year CAD events but not subclinical atherosclerosis in a general population, and improved the reclassification capacity of Framingham-derived CAD risk functions.

      Keywords

      1. Introduction

      Several markers of inflammation and oxidation are associated with the atherosclerotic process. Because of the relationship between their systemic levels and the occurrence and extent of atherosclerosis, these biomarkers have been evaluated as predictors of cardiovascular events, promoters of disease progression, and determinants of the effectiveness of cardiovascular therapies [
      • Ross R.
      Atherosclerosis: an inflammatory disease.
      ,
      • Holvoet P.
      • Mertens A.
      • Vehamme P.
      • et al.
      Circulating oxidized LDL is a useful marker for identifying patients with coronary artery disease.
      ,
      • Ehara S.
      • Ueda M.
      • Naruko T.
      • et al.
      Elevated levels of oxidized low density lipoprotein show a positive relationship with the severity of acute coronary syndromes.
      ]. The oxidative conversion of low-density lipoprotein (LDL) to oxidized low-density lipoprotein (oxLDL) is considered to be a key factor in the physiopathologic process that initiates and accelerates the development of the early atherosclerotic lesion [
      • Holvoet P.
      • Mertens A.
      • Vehamme P.
      • et al.
      Circulating oxidized LDL is a useful marker for identifying patients with coronary artery disease.
      ]. Several short- to medium-term follow-up studies have reported that circulating oxLDL could be a predictor of acute coronary syndromes in the general population [
      • Meisinger C.
      • Baumert J.
      • Khuseyinova N.
      • Loewel H.
      • Koenig W.
      Plasma oxidized low-density lipoprotein, a strong predictor for acute coronary heart disease events in apparently healthy, middle-aged men from the general population.
      ] and in coronary artery disease (CAD) patients [
      • Gomez M.
      • Valle V.
      • Arós F.
      • et al.
      Oxidized LDL, lipoprotein (a) and other emerging risk factors in acute myocardial infarction (FORTIAM Study).
      ,
      • Fraley A.E.
      • Schwartz G.G.
      • Olsson A.G.
      • et al.
      MIRACL Study Investigators
      Relationship of oxidized phospholipids and biomarkers of oxidized low-density lipoprotein with cardiovascular risk factors, inflammatory biomarkers, and effect of statin therapy in patients with acute coronary syndromes: results from the MIRACL (Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering) trial.
      ]. A few studies have looked at the relationship between oxLDL and subclinical atherosclerotic markers [
      • Choi S.H.
      • Chae A.
      • Miller E.
      • et al.
      Relationship between biomarkers of oxidized low-density lipoprotein, statin therapy, quantitative coronary angiography, and atheroma: volume observations from the REVERSAL (Reversal of Atherosclerosis with Aggressive Lipid Lowering) study.
      ,
      • Nishi K.
      • Itabe H.
      • Uno M.
      • et al.
      Oxidized LDL in carotid plaques and plasma associates with plaque instability.
      ,
      • Wallenfeldt K.
      • Fagerberg B.
      • Wikstrand J.
      • Hulthe J.
      Oxidized low-density lipoprotein in plasma is a prognostic marker of subclinical atherosclerosis development in clinically healthy men.
      ]. In addition, levels of antibodies against oxLDL (OLAB), the immune response to the oxLDL particle by lymphocytes, could be predictive for the progression of carotid or femoral atherosclerosis in apparently healthy subjects [
      • Salonen J.T.
      • Yla-Herttuala S.
      • Yamamoto R.
      • et al.
      Autoantibody against oxidized LDL and progression of carotid atherosclerosis.
      ,
      • Fukumoto M.
      • Shoji T.
      • Emoto M.
      • Kawagishi T.
      • Okuno Y.
      • Nishizawa Y.
      Antibodies against oxidized LDL and carotid artery intima-media thickness in a healthy population.
      ].
      Carotid intima-media thickness (IMT) and ankle-brachial index (ABI) have been used as simple, inexpensive, and safe non-invasive indicators for the atherosclerotic process, even at the pre-symptomatic phase when intervention can improve prognosis and prevent or delay severe complications [
      • Barth J.
      An update on carotid ultrasound measurement of intima-media thickness.
      ].
      The objectives of this study were to assess 1) the association of lipid oxidation biomarkers with 10-year CAD events, and with subclinical signs of atherosclerosis and 2) the reclassification capacity of lipid oxidation biomarkers over the Framingham-derived CAD risk functions in general population. The general population reference values of circulating oxLDL and OLAB are also presented.

      2. Methods

      2.1 Design and participants

      A population-based cohort study was designed and recruited 4782 individuals, aged between 25 and 74 years, in the province of Girona, Spain (1748 in the period 1994-96 and 3034 in the period 1999–2002), within the framework of the REGICOR study [
      • Grau M.
      • Subirana I.
      • Elosua R.
      • et al.
      on behalf of the REGICOR Investigators
      Trends in cardiovascular risk factor prevalence (1995-2000-2005) in northeastern Spain.
      ]. Participants who declined follow-up, or who already had symptomatic heart disease at study inclusion, or who moved out of the area under study or presented a terminal disease were excluded from the present analysis. Follow-up of the 4042 who met the eligibility criteria was carried out (See flowchart, Fig. 1). From this group 107 patients (3.8%) were lost during the long-term follow up, especially due to change in their permanent address.
      Figure thumbnail gr1
      Fig. 1Flowchart of participants included in the study. AMI: acute myocardial infarction; CV: cardiovascular.
      Participants were initially chosen by 2-phase random sampling. In the first phase, populations were selected; in the second, the same number of men and women for each population, stratified by 10-year age groups, was selected from the most recent census (1991 and 1999, respectively). The participation rate was above 71% [
      • Grau M.
      • Subirana I.
      • Elosua R.
      • et al.
      on behalf of the REGICOR Investigators
      Trends in cardiovascular risk factor prevalence (1995-2000-2005) in northeastern Spain.
      ]. All participants were duly informed and signed their consent. The local ethics committee approved the study.

      2.2 Lipid oxidation biomarkers and routine laboratory determinations

      Biological samples from the population controls were obtained in the baseline recruitment visit after 10–14 h fasting, coded, shipped to a central laboratory, and frozen at −80 °C until the assay. To guarantee the technical quality of the assays, no previous freeze–thaw cycles were permitted.
      Two markers of systemic lipid oxidation, oxLDL and OLAB, were determined in a random sample of 2793 participants out of the 4042 that were potentially eligible. Plasma oxLDL was determined by an ELISA sandwich procedure using the murine monoclonal antibody mAB-4E6 as capture antibody and a peroxidase-conjugated antibody against oxidized ApoB bound to the solid phase (oxLDL, Mercodia AB, Uppsala, Sweden) [
      • Holvoet P.
      • Mertens A.
      • Vehamme P.
      • et al.
      Circulating oxidized LDL is a useful marker for identifying patients with coronary artery disease.
      ]. Levels of OLAB were measured in plasma by ELISA, using copper-oxidized LDL coated onto microtiter strips as antigen and specific peroxidase-conjugated anti-human IgG antibodies to detect the presence of bound antibodies (OLAB, Biomedica, Vienna, Austria). The interserial coefficients of variation were 11.45% and 9.98% for OLAB and oxLDL, respectively. Analytical inter-assay imprecision, expressed as coefficient of variation, was assessed from 20 day-to-day measurements of control samples.
      Serum glucose, total cholesterol, high-density lipoprotein cholesterol and triglycerides were determined by enzymatic methods (Roche Diagnostics, Basel, Switzerland) in a Cobas Mira Plus autoanalyzer (Roche Diagnostics, Basel, Switzerland). Whenever triglycerides were <300 mg/dL, LDL cholesterol was calculated by the Friedewald equation. High-sensitivity C-reactive protein (hs-CRP) was determined by immunoturbidimetry (ABX Diagnostic, Montpellier, France). Quality control was performed with the External Quality Assessment-WHO Lipid Program (WHO, Prague, Czech Republic) and Monitrol-Quality Control Program (Baxter Diagnostics, Dudingen, Switzerland).

      2.3 Follow-up for CAD events of interest

      All participants were contacted by telephone every two years until 2010 to detect incident CAD events. Follow-up of those participants included in the first survey (1994-96) was censored at 10 years. Non-fatal events during follow-up were ascertained by a standardized telephone questionnaire, review of medical records, linkage with a regional myocardial infarction population registry, and cross-check of all these data. Fatal events were identified from the regional and national mortality registers covering the whole population included in this study. Deaths and suspected cardiovascular events were classified by an Event Committee using standardized criteria. Admission due to fatal or non-fatal acute myocardial infarction, angina pectoris, or death from coronary artery disease were the events of interest.

      2.4 Follow-up for subclinical atherosclerosis

      All the REGICOR participants were invited to a follow-up examination undertaken between 2007 and 2010. ABI and IMT were measured at the end of the follow up period in 1427 participants of those with lipid oxidation biomarkers. ABI was measured with a continuous 8 MHz probe Doppler device (SONICAID 421, Oxford Instruments). Systolic blood pressure was measured in both brachial arteries and posterior tibial and dorsalis pedis arteries of both lower limbs after 5 min rest. Right and left ABIs were then calculated, taking the highest distal vessel pressure of each lower limb and the mean brachial pressure. ABI <0.9 in either leg was considered pathological, ABI of 0.9–1.39 was considered normal, and ABI >1.39 was excluded from evaluation because the possible influence of arterial wall stiffness made it impossible to discard arterial obstruction [
      • Barth J.
      An update on carotid ultrasound measurement of intima-media thickness.
      ].
      An Acuson Aspen® ultrasound instrument equipped with an L7 5–12 MHz near-field small-parts vascular transducer was used to non-invasively measure carotid IMT. The IMT was defined as the distance between the leading edges of the lumen-intima interface and the media-adventitia interface of the far wall. The provided results are the maximum IMT (in millimeters) between both sides at common carotid artery. We considered the top-quintile of IMT as a significant plaque.

      2.5 Other measurements

      Smoking habit and cholesterol treatment were determined by a standardized questionnaire. Hypertension was defined as use of antihypertensive agents or systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg, calculated as the average of 2 readings taken at least 5 min apart. Diabetes was defined as use of insulin or hypoglycaemic agents or fasting blood glucose >125 mg/dl. Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared.

      2.6 Statistical analyses

      The OxLDL values were log-transformed owing to their non-normal distribution. Chi-square tests for categorical variables and Student t-test or Mann Whitney U-tests were used. Areas under the receiver operating characteristics (ROC) curves were calculated using Somers Dxy rank correlation for a censored response variable. Net reclassification index (NRI) was computed to assess whether the inclusion of oxLDL improved individual risk stratification calculated with REGICOR-Framingham adapted cardiovascular risk charts [
      • Marrugat J.
      • Subirana I.
      • Comin E.
      • et al.
      Validity of an adaptation of the Framingham cardiovascular risk function: the VERIFICA study.
      ]. To deal with missing values in covariates, a multiple imputation analysis was done. The statistical analyses were performed by R version 2.15.1 (R: a language and environment for statistical computing R Foundation for Statistical Computing, Vienna, Austria).

      3. Results

      The median follow-up for coronary events was 9.06 years. There were no differences in the sociodemographic and main clinical characteristics between participants with and without lipid oxidation biomarkers determined.
      The baseline characteristics of the population included in the study are shown in Table 1 by development of CAD event (Panel A) and by presence of subclinical signs of atherosclerosis (Panel B). Of the 2793 subjects (mean age 50.0 (13.4), 52.4% women), 103 had a CAD clinical event (34 non-fatal AMI, 43 angina and 26 CAD deaths). Of the 306 participants with a subclinical atherosclerosis, 268 were at the top quintile of IMT values and 57 had an ABI <0.9 (19 had both). In our study, oxLDL was higher in those with clinical and subclinical events. OLAB concentration was not associated with subclinical atherosclerosis or coronary clinical events.
      Table 1Baseline characteristics of a general population cohort by coronary artery disease event in Panel A and by subclinical atherosclerosis in Panel B.
      Panel AClinical eventP-value
      NoYes
      N = 2690N = 103
      Age mean (years)
      Standard deviation.
      49.5 (13.5)60.0 (9.94)<0.001
      Sex: female1451 (53.9%)31 (30.1%)<0.001
      Diabetes301 (11.4%)26 (26.0%)<0.001
      Hypertension1045 (39.8%)75 (75.0%)<0.001
      Smoker687 (25.9%)27 (26.5%)0.992
      Cholesterol treatment137 (5.09%)12 (11.7%)0.007
      Total cholesterol (mg/dl)
      Standard deviation.
      221 (42.1)241 (45.9)<0.001
      LDL cholesterol (mg/dl)
      Standard deviation.
      149 (37.4)167 (41.6)<0.001
      HDL cholesterol (mg/dl)
      Standard deviation.
      52.0 (13.2)44.6 (12.6)<0.001
      C-reactive protein (mg/dL)0.11 [0.04; 0.34]0.29 [0.12; 0.75]<0.001
      Triglycerides (mg/dl)88.0 [67.0; 121]125 [91.0; 176]<0.001
      Body mass index (Kg/m2)
      Standard deviation.
      27.3 (4.51)29.1 (4.41)0.003
      Ankle-brachial index < 0.957 (3.00%)6 (12.2%)0.004
      Maximum IMT (mm)0.66 [0.58; 0.77]0.82 [0.71; 0.90]<0.001
      Top-quintile IMT268 (19.0%)17 (53.1%)<0.001
      Oxidized LDL (U/L)53.2 [38.5; 71.8]65.1 [47.1; 84.2]<0.001
      Top-quintile oxidized LDL509 (18.9%)31 (30.1%)0.007
      Antibodies against oxLDL (U/L)343 [193; 627]291 [153; 569]0.118
      Panel BSubclinical eventP-value
      NoYes
      N = 1121N = 306
      Age mean (years)
      Standard deviation.
      45.4 (12.4)58.1 (10.0)<0.001
      Sex: female642 (57.2%)132 (43.1%)<0.001
      Diabetes87 (7.87%)50 (16.7%)<0.001
      Hypertension341 (30.9%)195 (65.4%)<0.001
      Smoker289 (26.0%)63 (20.9%)0.082
      Cholesterol treatment40 (3.57%)20 (9.48%)<0.001
      Total cholesterol (mg/dl)
      Standard deviation.
      216 (41.4)232 (42.9)<0.001
      LDL cholesterol (mg/dl)
      Standard deviation.
      143 (36.6)157 (37.6)<0.001
      HDL cholesterol (mg/dl)
      Standard deviation.
      53.0 (13.1)51.2 (13.2)0.043
      C-reactive protein (mg/dL)0.08 [0.03; 0.29]0.18 [0.08; 0.40]<0.001
      Triglycerides (mg/dl)85.0 [65.0; 117]102 [76.0; 138]<0.001
      Body mass index (Kg/m2)
      Standard deviation.
      26.9 (4.48)28.3 (4.22)<0.001
      Oxidized LDL51.8 [37.1; 69.3]57.8 [42.6; 76.7]<0.001
      Top-quintile oxidized LDL187 (16.7%)75 (24.5%)0.002
      Antibodies against oxLDL (U/L)342 [192; 649]334 [183; 657]0.683
      OLAB, oxidized low-density lipoprotein antibodies; IMT, intima-media thickness; LDL, low-density lipoprotein; HDL, high-density lipoprotein.
      a Standard deviation.
      Bivariate analysis of the top oxLDL quintile versus all other quintiles (1–4 quintiles) with other participant characteristics is summarized in Table 2. In the follow-up, subjects in the top oxLDL quintile had a higher incidence of AMI or angina (P = 0.007), carotid atheroma plaques (top IMT quintile) (P = 0.002), and abnormal ABI (P = 0.029). Of all the risk factors, only diabetes and smoking were not associated with the higher quintile of oxLDL. In the multivariate analyses, oxLDL was independently associated with CAD events, but not with subclinical events (Table 3). In the multiple imputation analyses the magnitude of the association between oxLDL and clinical events was similar and consistent with the main analyses, although the association in model 2 and 3 was not statistically significant. From the overall group, 248 participants (8.88%) commenced lipid-lowering therapy during the follow-up. The inclusion of this variable in both previous models did not modify the independent association of oxLDL in the analysis for clinical events (P = 0.235) and for subclinical events (P = 0.222). The discriminant capacity of the predictive models (REGICOR-Framingham adapted cardiovascular risk function) with classical risk factors did not change when oxLDL was added to the model (c-statistics = 0.79 for both models, without and with oxLDL). However, reclassification improved when oxLDL was added to the classical predictive model (NRI = 14.67% [4.90; 24.45], P = 0.003).
      Table 2Participant characteristics by top quintile versus rest of oxidized low-density lipoprotein values.
      Oxidized LDLP-value
      1-4 QuintilesTop quintile
      N = 2253N = 540
      AMI or angina in follow-up3.20%5.74%0.007
      Ankle-brachial index < 0.92.76%5.10%0.029
      Top-quintile IMT18.2%26.6%0.002
      Maximum IMT (mm)0.80 [0.70; 0.95]0.85 [0.73; 1.01]0.001
      Antibodies against oxLDL (U/L)346 [197; 638]326 [171; 570]0.082
      Age mean (years)
      Standard deviation.
      49.1 (13.6)53.4 (12.3)<0.001
      Sex: female54.9%45.4%<0.001
      Diabetes11.7%13.2%0.367
      Hypertension39.1%49.7%<0.001
      Smoker26.7%22.6%0.058
      Cholesterol treatment4.62%8.33%0.001
      Total cholesterol (mg/dl)
      Standard deviation.
      216 (40.5)246 (41.3)<0.001
      LDL cholesterol (mg/dl)
      Standard deviation.
      144 (35.9)173 (36.3)<0.001
      HDL cholesterol (mg/dl)
      Standard deviation.
      52.5 (13.3)48.6 (12.3)<0.001
      C-reactive protein (mg/dL)0.11 [0.04; 0.33]0.16 [0.06; 0.40]0.002
      Triglycerides (mg/dl)85.0 [65.0; 117]110 [80.0; 150]<0.001
      Body mass index (Kg/m2)
      Standard deviation.
      27.2 (4.53)28.1 (4.38)<0.001
      AMI, acute myocardial infarction; OLAB, oxidized low-density lipoprotein antibodies; IMT, intima-media thickness; LDL, low-density lipoprotein; HDL, high-density lipoprotein.
      a Standard deviation.
      Table 3Hazard ratio of coronary artery disease event [panel A] and odds ratio of subclinical atherosclerotic disease [panel B] for one unit of the log (oxidized LDL).
      Multivariate analyses 95% CIMultiple imputation of covariates 95% CI
      HRLowerUpperHRLowerUpper
      Panel A
      Coronary artery disease event
      Model 11.931.223.031.751.072.84
      Model 21.721.032.871.530.922.53
      Model 31.701.022.841.520.922.52
      Panel B
      Subclinical atherosclerosis
      Model 11.240.901.701.100.811.51
      Model 20.970.681.391.000.711.40
      Model 30.960.671.370.990.701.39
      LDL, low-density lipoprotein; CI, confidence interval; HR, hazard ratio.
      Model 1: adjusted for age, sex and cohort (1995–96 or 1999–2002).
      Model 2: adjusted as model 1 and LDL cholesterol, smoking, hypertension and diabetes mellitus.
      Model 3: adjusted as model 2 and lipid lowering treatment and 2000 cohort.
      The Spearman correlation coefficient between IMT and oxLDL was 0.133 (P < 0.001), and between ABI and oxLDL was 0.001 (P = 0.967). There was no significant correlation between IMT and ABI (r = 0.005, P = 0.836). The relationship between level of oxLDL and the age of patient at the time of an event was not significant too (P = 0.774).
      Fig. 2 shows the oxLDL and OLAB distribution for men and women by age group in our general population.
      Figure thumbnail gr2
      Fig. 2Percentile population distribution of oxidized low-density lipoprotein (LDL) -for the Mercodia method- and oxidized low-density lipoprotein antibodies (OLAB) by age and sex in a general population cohort. (OxLDL (mean): 60.0 U/L (SD 27.0) U/L and 54.4 U/L (SD 26.0) for men and women respectively. OLAB (median): 300 U/L (5th–95th percentile: 74.4–1681) and 363 U/L (5th–95th percentile: 80.4–1810), for men and women respectively)

      4. Discussion

      4.1 OxLDL and clinical events

      In this prospective 10-year population-based cohort study, circulating oxLDL was independently associated with the onset of CAD events but not with subclinical atherosclerosis indicators. Moreover oxLDL improves the predictive capacity of classical risk function measured by the reclassification.
      Our findings concur with previous reports [
      • Ehara S.
      • Ueda M.
      • Naruko T.
      • et al.
      Elevated levels of oxidized low density lipoprotein show a positive relationship with the severity of acute coronary syndromes.
      ,
      • Meisinger C.
      • Baumert J.
      • Khuseyinova N.
      • Loewel H.
      • Koenig W.
      Plasma oxidized low-density lipoprotein, a strong predictor for acute coronary heart disease events in apparently healthy, middle-aged men from the general population.
      ,
      • Gomez M.
      • Valle V.
      • Arós F.
      • et al.
      Oxidized LDL, lipoprotein (a) and other emerging risk factors in acute myocardial infarction (FORTIAM Study).
      ] showing the association between circulating oxLDL and cardiovascular diseases. In a series of 135 CAD patients, oxLDL levels were positively correlated with the severity of acute coronary syndromes [
      • Ehara S.
      • Ueda M.
      • Naruko T.
      • et al.
      Elevated levels of oxidized low density lipoprotein show a positive relationship with the severity of acute coronary syndromes.
      ]. Nishi et al. reported that oxLDL levels in plasma and carotid plaques from patients undergoing carotid endarterectomy were related to plaque instability [
      • Nishi K.
      • Itabe H.
      • Uno M.
      • et al.
      Oxidized LDL in carotid plaques and plasma associates with plaque instability.
      ]. Other authors have also reported high oxLDL levels in CAD patients versus control subjects [
      • Holvoet P.
      • Mertens A.
      • Vehamme P.
      • et al.
      Circulating oxidized LDL is a useful marker for identifying patients with coronary artery disease.
      ,
      • Weinbrenner T.
      • Cladellas M.
      • Covas M.I.
      • et al.
      High oxidative stress in patients with stable coronary heart disease.
      ]. In a recent 10-year follow-up study including 1260 elderly inhabitants, circulating oxLDL stand out as a risk factor for all-cause mortality independent of major confounding attributes [
      • Linna M.
      • Ahotupa M.
      • Lopponen M.K.
      • Irjala K.
      • Vasankari T.
      Circulating oxidised LDL lipids, when proportioned to HDL-c, emerged as a risk factor of all-cause mortality in a population-based survival study.
      ]. Another population-based study recently showed that increased oxLDL concentrations were associated with an increased risk of incident CAD, in the follow-up of 2060 individuals [
      • Koenig W.
      • Karakas M.
      • Zierer A.
      • et al.
      Oxidized LDL and the risk of coronary heart disease: results from the MONICA/KORA Augsburg study.
      ]. Nevertheless, this significance is attenuated after additional adjustment for lipid and inflammatory profile and endothelial dysfunction markers. It must be highlighted that the available assays to measure oxLDL, and lipid oxidation markers in general, vary considerably and this makes it difficult to compare data [
      • Salonen J.T.
      • Yla-Herttuala S.
      • Yamamoto R.
      • et al.
      Autoantibody against oxidized LDL and progression of carotid atherosclerosis.
      ,
      • Fraley A.E.
      • Tsimikas S.
      Clinical applications of circulating oxidized low-density lipoprotein biomarkers in cardiovascular disease.
      ].
      Possible associated mechanisms linked to LDL oxidation that could explain plaque instability and the association of oxLDL with incident coronary events includes collagen synthesis in smooth muscle cells, induction of apoptosis, inhibition of nitric oxide function (vasospasm), and increases in a wide variety of proinflammatory cytokines in macrophages [
      • Steinberg D.
      The LDL modification hypothesis of atherogenesis: an update.
      ].

      4.2 OxLDL and subclinical events

      We observed an association between oxLDL levels with subclinical events (maximal IMT and ABI <0.9) in the bivariate analysis. However, the odds ratio (OR) for oxLDL was no longer significant after adjustment for potential confounders.
      Carotid IMT measurement is useful in screening for atherosclerosis and as a surrogate measurement for response to disease interventions [
      • Grobbee D.E.
      • Bots M.L.
      Carotid artery intima-media thickness as an indicator of generalized atherosclerosis.
      ]. In a series of 146 men with previous coronary artery bypass graft surgery, the IMT was related to risk for coronary events beyond that predicted by coronary artery measures of atherosclerosis in the coronary angiography and lipid measurements [
      • Hodis H.
      • Mack W.
      • LaBree L.
      • et al.
      The role of carotid arterial intima-media thickness in predicting clinical coronary events.
      ]. In a meta-analysis of 8 studies with 37,197 subjects, carotid IMT was a strong predictor of future vascular events, although the relative risk per IMT difference was slightly higher for stroke than for myocardial infarction [
      • Lorenz M.
      • Markus H.
      • Bots M.
      • Rosvall M.
      • Sitzer M.
      Prediction of clinical cardiovascular events with carotid intima-media thickness a systematic review and meta-analysis.
      ].
      In very specific population subgroups, oxLDL has been related to subclinical events, as in Swedish men with a mean age of 58 years and higher levels than our population for total cholesterol (233 mg/dl versus 221 mg/dl) and oxLDL (85 U/l versus 58 U/l) [
      • Wallenfeldt K.
      • Fagerberg B.
      • Wikstrand J.
      • Hulthe J.
      Oxidized low-density lipoprotein in plasma is a prognostic marker of subclinical atherosclerosis development in clinically healthy men.
      ]. In another subgroup, asymptomatic members of Familial Combined Hyperlipidemia families, oxLDL was independently associated with carotid IMT and could be used as a marker of early atherosclerosis [
      • Liu M.
      • Ylitalo K.
      • Salonen R.
      • Salonen J.
      • Taskinen M.R.
      Circulating oxidized low-density lipoprotein and its association with carotid intima-media thickness in asymptomatic members of Familial Combined Hyperlipidemia families.
      ].
      In our study, oxLDL was not related to ABI <0.9, which has a sensitivity and specificity of 95% for detecting angiographically positive peripheral artery disease [
      • Doobay A.V.
      • Anand S.S.
      Sensitivity and specificity of the ankle-brachial index to predict future cardiovascular outcomes: a systematic review.
      ]. Other authors report that oxLDL was independently associated with femoral plaques, but not with carotid artery wall damage [
      • Langlois M.
      • Rietzschel E.
      • De Buyzere M.
      • et al.
      on behalf of the Asklepios Investigators
      Femoral plaques confound the association of circulating oxidized low-density lipoprotein with carotid atherosclerosis in a general population aged 35 to 55 Years. The Asklepios study.
      ]. In their study, oxLDL was also related to the presence of femoral plaques in 54% of cases with carotid plaques and elevated oxLDL concentrations. Circulating oxLDL has been proposed as a marker of unstable echolucent plaque phenotype in the femoral artery in men as well as a marker of progression in small-vessel disease stroke [
      • Cuadrado-Godia E.
      • Ois A.
      • Garcia-Ramallo E.
      • et al.
      Biomarkers to predict clinical progression in small vessel disease strokes: prognostic role of albuminuria and oxidized LDL cholesterol.
      ].

      4.3 Antibodies against oxidized LDL

      In our work and others [
      • Tsouli S.G.
      • Kiortsis D.N.
      • Xydis V.
      • Argyropoulu M.I.
      • Elisaf M.
      • Tselepis A.D.
      Antibodies against various forms of mildly oxidized low-density lipoprotein are not associated with carotid intima-media thickness in patients with primary hyperlipidemia.
      ], OLAB concentration was not related to clinical events or subclinical signs of atherosclerosis. In contrast, low IgM antibodies against phosphorylcholine could predict cardiovascular disease in a 5- to 7-year follow-up [
      • de Faire U.
      • Su J.
      • Hua X.
      • et al.
      Low levels of IgM antibodies to phosphorylcholine predict cardiovascular disease in 60-year old men: effects on uptake of oxidized LDL in macrophages as a potential mechanism.
      ]. In addition, Fukumoto et al. reported an inverse relationship between OLAB and carotid IMT in healthy subjects [
      • Fukumoto M.
      • Shoji T.
      • Emoto M.
      • Kawagishi T.
      • Okuno Y.
      • Nishizawa Y.
      Antibodies against oxidized LDL and carotid artery intima-media thickness in a healthy population.
      ]. OLAB indirectly measures the immunologic individual response to lipid oxidation, thus a high inter-individual variability exists. Together with the fact that the available OLAB laboratory methodologies vary considerably, this limits comparability of the scarce number of publications [
      • Salonen J.T.
      • Yla-Herttuala S.
      • Yamamoto R.
      • et al.
      Autoantibody against oxidized LDL and progression of carotid atherosclerosis.
      ]. This variability probably renders OLAB assessment less useful than the more specific oxLDL analysis.

      4.4 Role of hypocholesterolemic therapy

      In our follow-up recruiting of a general population, in which only 5.33% received lipid-lowering drugs, higher levels of oxLDL in the treated group were observed but had no effect on outcome after adjusting by other co-variables (Table 3). Statins interfere with oxidation in several ways that may contribute to reducing the atherogenic process. In addition to direct antioxidant effects, statins reduce circulating oxLDL and inhibit their uptake by macrophages. They also reduce circulating markers of oxidation such as F2-isoprostane and nitrotyrosine and inhibit oxidant enzymes activity. Otherwise, in acute coronary syndromes patients, MIRACL researchers reported that atorvastatin treatment could increase oxidized phospholipids on apolipoprotein B [
      • Fraley A.E.
      • Schwartz G.G.
      • Olsson A.G.
      • et al.
      MIRACL Study Investigators
      Relationship of oxidized phospholipids and biomarkers of oxidized low-density lipoprotein with cardiovascular risk factors, inflammatory biomarkers, and effect of statin therapy in patients with acute coronary syndromes: results from the MIRACL (Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering) trial.
      ]. Similarly, statin therapy could increase levels of lipoprotein (a) and of oxidized phospholipids and malondialdehyde on apolipoprotein B particles [
      • Choi S.H.
      • Chae A.
      • Miller E.
      • et al.
      Relationship between biomarkers of oxidized low-density lipoprotein, statin therapy, quantitative coronary angiography, and atheroma: volume observations from the REVERSAL (Reversal of Atherosclerosis with Aggressive Lipid Lowering) study.
      ]. These studies hypothesize that mobilization of oxidized phospholipids may occur from sites of inflammation of atherosclerotic lesions, even via transfer from other lipoproteins to oxidized phospholipids acceptors, such as lipoprotein(a), which strongly binds oxidized phospholipids [
      • de Faire U.
      • Su J.
      • Hua X.
      • et al.
      Low levels of IgM antibodies to phosphorylcholine predict cardiovascular disease in 60-year old men: effects on uptake of oxidized LDL in macrophages as a potential mechanism.
      ].

      4.5 Study strengths and limitations

      The cohort analyzed in the present study was recruited in two periods (1994-6 and 1999–2002). To ensure consistency we compared their baseline characteristics and no clinically relevant differences were observed. In addition, we stratified the analyses by survey and adjusted our models for period of recruitment; the main results and conclusion remained similar.
      Careful sample preservation, centralization of analyses, and rigorous quality control assurance were performed to avoid variability between methodologies and guarantee the comparability of results between surveys conducted in different time periods. Longitudinal stability of the oxLDL determination has been reported when samples were well preserved [
      • Holvoet P.
      • Macy E.
      • Landeloos M.
      • et al.
      Analytical performance and diagnostic accuracy of immunometric assays for the measurement of circulating oxidized LDL.
      ]. In addition, the randomized subsample with no oxLDL determination in the study did not differ from the overall group. Nevertheless, the intra-individual variability of oxLDL during the follow-up was not assessed and this is a limitation inherent to the design of our study. Further studies are warranted to analyze the association between changes in oxLDL and CAD incidence.
      A limitation of this study, and most other current population series, is that subclinical coronary atherosclerosis was not directly assessed by coronary angiography. Carotid IMT and ABI are indirect methods to measure subclinical atherosclerosis. However, their predictive capacity is well established for both in clinical practice. To our knowledge, this is one of the largest cohort, with the longest follow-up, reported with carotid IMT, ABI, OLAB and oxLDL measurements. Even so, our findings must be confirmed in other cohorts, and cost-benefit studies are required before oxLDL can be embraced as a clinically useful screening tool for the general population. Data from this cohort also provide reference values for oxLDL and OLAB in our general population.

      5. Conclusions

      In a population-based cohort with low incidence of CAD, increased plasma levels of oxLDL are an independent risk factor of future clinical events and improve the reclassification capacity of Framingham-derived CAD risk functions. This relationship is independent of other traditional cardiovascular risk factors. In contrast, this biomarker was not related to subclinical atherosclerosis as assessed by carotid IMT or ABI examinations. Also we presented the table of normal values of oxLDL and OLAB in the general population.

      Acknowledgments

      This study was funded by a grant from the Instituto de Salud Carlos III FEDER-Fondo de Investigación Sanitaria (FIS) PI010105 , PI081327 , and PI1101801 ; FEDER-Ministerio de Economía y Competitividad , Instituto de Salud Carlos III (Red HERACLES RD06/0009); MARATO TV3 (081630), Miguel Servet SNS contract (CP06/00100); Comissionat per la Universitat i Recerca , Departament d’Innovació , Universitats i Empresa of the Generalitat de Catalunya , Agència de gestió ajuts universitaris de recerca (AGAUR 2009 SGR 1195); Sociedad Española De Cardiología (SEC 2010); The CIBEROBN and CIBERESP are initiatives of the Instituto de Salud Carlos III, Madrid, Spain. We appreciate the English revision of the manuscript by Elaine Lilly, PhD.

      References

        • Ross R.
        Atherosclerosis: an inflammatory disease.
        N Engl J Med. 1999; 340: 115-126
        • Holvoet P.
        • Mertens A.
        • Vehamme P.
        • et al.
        Circulating oxidized LDL is a useful marker for identifying patients with coronary artery disease.
        Arterioscler Thromb Vasc Biol. 2001; 21: 844-848
        • Ehara S.
        • Ueda M.
        • Naruko T.
        • et al.
        Elevated levels of oxidized low density lipoprotein show a positive relationship with the severity of acute coronary syndromes.
        Circulation. 2001; 103: 1955-1960
        • Meisinger C.
        • Baumert J.
        • Khuseyinova N.
        • Loewel H.
        • Koenig W.
        Plasma oxidized low-density lipoprotein, a strong predictor for acute coronary heart disease events in apparently healthy, middle-aged men from the general population.
        Circulation. 2005; 112: 651-657
        • Gomez M.
        • Valle V.
        • Arós F.
        • et al.
        Oxidized LDL, lipoprotein (a) and other emerging risk factors in acute myocardial infarction (FORTIAM Study).
        Rev Esp Cardiol. 2009; 62: 373-382
        • Fraley A.E.
        • Schwartz G.G.
        • Olsson A.G.
        • et al.
        • MIRACL Study Investigators
        Relationship of oxidized phospholipids and biomarkers of oxidized low-density lipoprotein with cardiovascular risk factors, inflammatory biomarkers, and effect of statin therapy in patients with acute coronary syndromes: results from the MIRACL (Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering) trial.
        J Am Coll Cardiol. 2009; 53: 2186-2196
        • Choi S.H.
        • Chae A.
        • Miller E.
        • et al.
        Relationship between biomarkers of oxidized low-density lipoprotein, statin therapy, quantitative coronary angiography, and atheroma: volume observations from the REVERSAL (Reversal of Atherosclerosis with Aggressive Lipid Lowering) study.
        J Am Coll Cardiol. 2008; 52: 24-32
        • Nishi K.
        • Itabe H.
        • Uno M.
        • et al.
        Oxidized LDL in carotid plaques and plasma associates with plaque instability.
        Arterioscler Thromb Vasc Biol. 2002; 22: 1649-1654
        • Wallenfeldt K.
        • Fagerberg B.
        • Wikstrand J.
        • Hulthe J.
        Oxidized low-density lipoprotein in plasma is a prognostic marker of subclinical atherosclerosis development in clinically healthy men.
        J Intern Med. 2004; 256: 413-420
        • Salonen J.T.
        • Yla-Herttuala S.
        • Yamamoto R.
        • et al.
        Autoantibody against oxidized LDL and progression of carotid atherosclerosis.
        Lancet. 1992; 339: 883-887
        • Fukumoto M.
        • Shoji T.
        • Emoto M.
        • Kawagishi T.
        • Okuno Y.
        • Nishizawa Y.
        Antibodies against oxidized LDL and carotid artery intima-media thickness in a healthy population.
        Arterioscler Thromb Vasc Biol. 2000; 20: 703-707
        • Barth J.
        An update on carotid ultrasound measurement of intima-media thickness.
        Am J Cardiol. 2002; 89: 32B-39B
        • Grau M.
        • Subirana I.
        • Elosua R.
        • et al.
        • on behalf of the REGICOR Investigators
        Trends in cardiovascular risk factor prevalence (1995-2000-2005) in northeastern Spain.
        Eur J Cardiovasc Prev Rehabil. 2007; 14: 653-659
        • Marrugat J.
        • Subirana I.
        • Comin E.
        • et al.
        Validity of an adaptation of the Framingham cardiovascular risk function: the VERIFICA study.
        J Epidemiol Community Health. 2007; 61: 40-47
        • Weinbrenner T.
        • Cladellas M.
        • Covas M.I.
        • et al.
        High oxidative stress in patients with stable coronary heart disease.
        Atherosclerosis. 2003; 168: 99-106
        • Linna M.
        • Ahotupa M.
        • Lopponen M.K.
        • Irjala K.
        • Vasankari T.
        Circulating oxidised LDL lipids, when proportioned to HDL-c, emerged as a risk factor of all-cause mortality in a population-based survival study.
        Age Ageing. 2013; 42: 110-113
        • Koenig W.
        • Karakas M.
        • Zierer A.
        • et al.
        Oxidized LDL and the risk of coronary heart disease: results from the MONICA/KORA Augsburg study.
        Clin Chem. 2011; 57: 1196-1200
        • Fraley A.E.
        • Tsimikas S.
        Clinical applications of circulating oxidized low-density lipoprotein biomarkers in cardiovascular disease.
        Curr Opin Lipidol. 2006; 17: 502-509
        • Steinberg D.
        The LDL modification hypothesis of atherogenesis: an update.
        J Lipid Res. 2009; 50: S376-S381
        • Grobbee D.E.
        • Bots M.L.
        Carotid artery intima-media thickness as an indicator of generalized atherosclerosis.
        J Intern Med. 1994; 236: 567-573
        • Hodis H.
        • Mack W.
        • LaBree L.
        • et al.
        The role of carotid arterial intima-media thickness in predicting clinical coronary events.
        Ann Intern Med. 1998; 128: 262-269
        • Lorenz M.
        • Markus H.
        • Bots M.
        • Rosvall M.
        • Sitzer M.
        Prediction of clinical cardiovascular events with carotid intima-media thickness a systematic review and meta-analysis.
        Circulation. 2007; 115: 459-467
        • Liu M.
        • Ylitalo K.
        • Salonen R.
        • Salonen J.
        • Taskinen M.R.
        Circulating oxidized low-density lipoprotein and its association with carotid intima-media thickness in asymptomatic members of Familial Combined Hyperlipidemia families.
        Arterioscler Thromb Vasc Biol. 2004; 24: 1492-1497
        • Doobay A.V.
        • Anand S.S.
        Sensitivity and specificity of the ankle-brachial index to predict future cardiovascular outcomes: a systematic review.
        Arterioscler Thromb Vasc Biol. 2005; 25: 1463-1469
        • Langlois M.
        • Rietzschel E.
        • De Buyzere M.
        • et al.
        • on behalf of the Asklepios Investigators
        Femoral plaques confound the association of circulating oxidized low-density lipoprotein with carotid atherosclerosis in a general population aged 35 to 55 Years. The Asklepios study.
        Arterioscler Thromb Vasc Biol. 2008; 28: 1563-1568
        • Cuadrado-Godia E.
        • Ois A.
        • Garcia-Ramallo E.
        • et al.
        Biomarkers to predict clinical progression in small vessel disease strokes: prognostic role of albuminuria and oxidized LDL cholesterol.
        Atherosclerosis. 2011; 219: 368-372
        • Tsouli S.G.
        • Kiortsis D.N.
        • Xydis V.
        • Argyropoulu M.I.
        • Elisaf M.
        • Tselepis A.D.
        Antibodies against various forms of mildly oxidized low-density lipoprotein are not associated with carotid intima-media thickness in patients with primary hyperlipidemia.
        Angiology. 2006; 57: 615-622
        • de Faire U.
        • Su J.
        • Hua X.
        • et al.
        Low levels of IgM antibodies to phosphorylcholine predict cardiovascular disease in 60-year old men: effects on uptake of oxidized LDL in macrophages as a potential mechanism.
        J Autoimmun. 2010; 34: 73-79
        • Holvoet P.
        • Macy E.
        • Landeloos M.
        • et al.
        Analytical performance and diagnostic accuracy of immunometric assays for the measurement of circulating oxidized LDL.
        Clin Chem. 2006; 52: 760-764