SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Objective

To determine whether anti–apolipoprotein A-1 (anti–Apo A-1) IgG are associated with major cardiovascular events in patients with rheumatoid arthritis (RA).

Methods

We determined anti–Apo A-1 IgG levels and the concentrations of cytokines, oxidized low-density lipoprotein (LDL), and matrix metalloproteinase 1 (MMP-1) MMP-2, MMP-3, and MMP-9 in sera from 133 patients with RA who did not have cardiovascular disease at baseline, all of whom were longitudinally followed up over a median period of 9 years. A major cardiovascular event was defined as a fatal or nonfatal stroke or acute coronary syndrome. The proinflammatory effects of anti–Apo A-1 IgG were assessed on human macrophages in vitro.

Results

During followup, the overall incidence of major cardiovascular events was 15% (20 of 133 patients). At baseline, anti–Apo A-1 IgG positivity was 17% and was associated with a higher incidence of major cardiovascular events (adjusted hazard ratio 4.2, 95% confidence interval 1.5–12.1). Patients who experienced a subsequent major cardiovascular event had higher circulating levels of anti–Apo A-1 IgG at baseline compared with those who did not have a major cardiovascular event. Receiver operating curve analysis showed that anti–Apo A-1 IgG was the strongest of all tested biomarkers for the prediction of a subsequent major cardiovascular event, with an area under the curve value of 0.73 (P = 0.0008). At the predefined and previously validated cutoff levels, the specificity and sensitivity of anti–Apo A-1 IgG to predict major cardiovascular events were 50% and 90%, respectively. Anti–Apo A-1 IgG positivity was associated with higher median circulating levels of interleukin-8 (IL-8), oxidized LDL, and MMP-9 and higher proMMP-9 activity as assessed by zymography. On human macrophages, anti–Apo A-1 IgG induced a significant dose-dependent increase in IL-8 and MMP-9 levels and proMMP-9 activity.

Conclusion

Anti–Apo A-1 IgG is an independent predictor of major cardiovascular events in RA, possibly by affecting vulnerability to atherosclerotic plaque.

In rheumatoid arthritis (RA), the burden of atherosclerosis and subsequent cardiovascular events is increased and represents the primary cause of mortality. RA is currently considered to be an independent risk factor for cardiovascular disease (1–5). A combination of traditional and nontraditional risk factors has been described to account for the accelerated atherosclerosis and the increased cardiovascular risk, but the underlying pathways are not well understood. It is not clear to what extent emergent nontraditional cardiovascular risk factors such as chemokines (6), cytokines (7), autoantibodies (8, 9), oxidized low-density lipoproteins (LDLs) (10), and matrix metalloproteinases (MMPs) (11) could contribute to atherosclerosis and atherothrombosis in RA.

High levels of anti–apolipoprotein A-1 (anti–Apo A-1) IgG have been identified in patients with autoimmune diseases associated with high cardiovascular risk, such as systemic lupus erythematosus (SLE) (12), as well as in patients with myocardial infarction (MI) (13, 14), in whom anti–Apo A-1 IgG have been shown to be an independent predictor of major cardiovascular events (15). Whether anti–Apo A-1 IgG exist in patients with RA and could predict the occurrence of major cardiovascular events remains unknown.

Therefore, we investigated 1) whether anti–Apo A-1 IgG are present in RA, 2) whether they could predict the occurrence of major cardiovascular events in RA, and 3) whether anti–Apo A-1 IgG are associated with higher levels of circulating mediators of inflammation and plaque instability, such as oxidized LDL, interleukin-6 (IL-6), IL-8, IL-1 receptor antagonist (IL-1Ra), tumor necrosis factor α, monocyte chemotactic protein 1, and MMP-1, MMP-2, MMP-3, and MMP-9, which are the MMPs most consistently associated with increased cardiovascular risk in humans (6, 7, 11).

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

The research Ethics Committee of the Geneva University Hospitals approved this protocol, which was performed in compliance with the principles of the Declaration of Helsinki.

Patient population and study design.

This was a nested cohort study of patients with RA for whom serum samples were available and who were included in the Geneva arthritis biobank patient cohort. The sample size of this study was computed based on preliminary results showing that the prevalence of Apo A-1 IgG in patients with RA was 11% (14), and by extrapolating our previous results from patients with MI (15) to patients with RA. Using a survival model (utilizing the log rank approach), a total sample size of 133 patients was estimated to detect a 3-fold increase in the risk of a major event (12% versus 36%) with a power of 90% and an alpha error of 5%. Inclusion criteria consisted of a diagnosis of RA, regardless of age and sex. Exclusion criteria were the presence of concomitant SLE based on the revised American College of Rheumatology (ACR) criteria (16), antiphospholipid syndrome (APS) according to the Sapporo criteria (17), or a major cardiovascular event (see definition below) before enrollment. Between January 1, 1990 and January 1, 2008, a total of 1,200 patients with a diagnosis of RA according to the revised ACR criteria (18) were followed up at the Geneva University Hospitals, for whom 144 samples were available in the serum biobank. Among those, 133 patients did not have any known cardiovascular event, SLE, or APS at the time of blood collection and were deemed eligible for inclusion in the present study.

The primary outcome was the occurrence of a major cardiovascular event during followup, defined prospectively by the presence of fatal or nonfatal acute coronary syndrome or stroke. The occurrence of major cardiovascular events was adjudicated by a study coordinator who was blinded to the results of biochemical analyses. Information was obtained by checking the medical files of the patients and by contacting the physician in charge of the patient. We also made contact with the patients individually and inquired about the occurrence of cardiovascular events. Only major cardiovascular events confirmed by the medical record and the treating physician were taken into account.

The secondary outcome involved analyzing the relationship between anti–Apo A-1 IgG and several markers of inflammation and plaque stability, such as oxidized LDL, IL-6, IL-8, IL-1Ra, MMP-1, MMP-2, MMP-3, and MMP-9.

Sample collection.

The median time between RA diagnosis and sample collection was 7 years (range 0–50 years; interquartile range [IQR] 1–14 years). Samples were immediately centrifuged, aliquoted, and frozen at −80°C until analyzed.

Biochemical analysis.

Determination of human antibodies to Apo A-1 by enzyme-linked immunosorbent assay (ELISA).

Anti–Apo A-1 IgG were measured as previously described (14). Briefly, Maxi-Sorb plates (Nunc) were coated with purified, human-derived delipidated Apo A-1 (20 μg/ml; 50 μl/well) for 1 hour at 37°C. After 3 washes with phosphate buffered saline (PBS)/2% bovine serum albumin (BSA; 100 μl/well), all wells were blocked for 1 hour with 2% BSA at 37°C. Samples were diluted 1:50 in PBS/2% BSA and incubated for 60 minutes. Additional patient samples at the same dilution were also added to an uncoated well to assess individual nonspecific binding. After 6 further washes, 50 μl/well of signal antibody (alkaline phosphatase–conjugated anti-human IgG; Sigma-Aldrich) diluted 1:1,000 in PBS/2% BSA solution was incubated for 1 hour at 37°C. After 6 more washes (150 μl/well) with PBS/2% BSA solution, the phosphatase substrate p-nitrophenyl phosphate disodium (50 μl/well; Sigma-Aldrich) dissolved in diethanolamine buffer (pH 9.8) was added. Each sample was tested in duplicate, and absorbance, determined as the optical density at 405 nm (OD405 nm), was determined after 20 minutes of incubation at 37°C (VersaMax, Molecular Devices). The corresponding nonspecific binding value was subtracted from the mean absorbance value for each sample.

The specificity of detection was assessed using previously described conventional saturation tests (13, 14) and was further confirmed by Western blot analysis (data not shown). The positivity cutoff was predefined and set at an OD value of 0.6 and 37% of the positive control value, as described earlier (15). OD values ranged from 0 to 1.33.

Oxidized LDL assessment.

Oxidized LDL levels were determined using a commercially available ELISA kit with 4E6 monoclonal antibodies (Mercodia). Samples were run in duplicate, with results given as the mean.

Serum C-reactive protein (CRP), creatinine, total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride levels.

CRP, creatinine, total cholesterol, HDL cholesterol, and triglyceride levels were determined using a DxI autoanalyzer (Beckman Coulter). LDL was conventionally calculated based on the Friedewald equation.

IL-6, IL-8, IL-1Ra, MMP-1, MMP-2, MMP-3, and MMP-9 quantification.

The levels of cytokines and MMPs were determined in patient sera and culture supernatants by using a commercially available multiplex beads immunoassay (Fluorokine MAP Multiplex Human Cytokine Panel and Human MMP Panel; R&D Systems) according to the supplier's instructions, using a Bio-Plex 200 array reader (Bio-Rad) with Luminex MAP Technology.

ProMMP-9 zymographic assay.

ProMMP-9 activity was detected by zymography on serum samples and supernatants from in vitro experiments (see below). Gels were copolymerized with gelatin (Sigma). Equal amounts of serum and supernatant (2 μl) and 1 ng of recombinant proMMP-9 standard (Calbiochem) were loaded on 9% sodium dodecyl sulfate–polyacrylamide gels in the absence of reducing agents. After electrophoresis overnight, gels were rinsed twice in 2.5% Triton X-100, incubated for 20 hours in 0.15M NaCl, 10 mM CaCl2, 0.2% Brij-35, and 50 mM Tris HCl buffer, pH 7.4, at 37°C, stained for 30 minutes in Coomassie blue R250, and destained in acetic acid–ethanol–water (1:3:6). The results of zymography were expressed as proMMP-9 proteolytic activity. The amount of gelatinase in the serum of patients with RA and in macrophage supernatant was estimated on the basis of the following formula: serum/supernatant proMMP-9 = (Iobs/Istd) × Wstd, where Iobs and Istd are the intensities of lytic areas produced in gels by samples and by standard proMMP-9, and Wstd is the weight (1 ng) of standard proMMP-9 loaded onto the gel. Zymography data were expressed as the logarithmic value of ng/ml of serum, as previously described (19). Gelatinolytic bands were measured using a gel analysis system (GeneGenius).

In vitro experiments.

Human monocytes were isolated from the peripheral blood mononuclear cells of healthy blood donors, as previously described (20). Monocytes were then further transformed into macrophages by 24-hour incubation with interferon-γ (500 units/ml). Macrophages (105/well) (Costar 3596, 96-well cell culture plate) were then exposed to increasing concentrations of polyclonal anti-human Apo A-1 IgG (Academy Biomedical) or the respective human IgG controls (Meridian Life Science) from 5 μg/ml to 40 μg/ml, or 1 μg/ml ultrapure lipopolysaccharide (LPS) (Enzo Life Sciences) for 24 hours, after which IL-8 levels were measured in culture supernatant. For MMP-9 measurement, macrophages were exposed to antibodies or LPS for 48 hours. For each donor, experiments were done in duplicate, and supernatants were pooled before MMP-9 or IL-8 assessment. Each experiment was repeated on 4 different healthy blood donors. LPS contamination in anti–Apo A-1 IgG and control IgG was ruled out using the Limulus test (21), which was performed according to the manufacturer's instructions. The results are expressed as the median value from the 4 different blood donors.

Statistical analysis.

Analyses were performed using Statistica software (StatSoft). Fisher's exact test, the chi-square test with Yates' correction, the Mann-Whitney U test, and the t-test were used, when appropriate, to compare groups of patients. Spearman's rank test was used to establish correlations between variables. Receiver operating curve (ROC) analysis was performed using Excel Analyse-it software (Microsoft). Time-to-event analyses were performed by Cox regression analysis, using anti–Apo A-1 IgG as a dichotomous variable. The adjusted Cox regression analysis was corrected for potential confounders, such as age, sex, diabetes mellitus, hypertension, dyslipidemia, smoking, obesity, and RA disease duration. Given the relatively limited size of the study sample, Cox regression analyses were also performed between anti–Apo A-1 IgG tertiles to determine whether the major cardiovascular event risk increased proportionally to anti–Apo A-1 IgG titers. Results are presented as crude and adjusted hazard ratios (HRs) with corresponding 95% confidence intervals (95% CIs). Survival analysis was performed by the Kaplan-Meier log rank method. Two-tailed P values less than 0.05 were considered significant; the value for significance was lower when Bonferroni correction was applied to take multiple testing into account.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Characteristics of the study subjects.

The demographic characteristics of the patients with RA are summarized in Table 1. The incidence of anti–Apo A-1 IgG positivity was 17% in patients with RA versus 1–3% in healthy blood donors, consistent with the results from our previous work (13, 14). Male sex was significantly overrepresented in patients positive for anti–Apo A-1 IgG when compared with patients who had negative test results for these autoantibodies (Table 1). Anti–Apo A-1 IgG–positive and anti–Apo A-1 IgG–negative patients with RA did not differ in terms of most of the important potential cardiovascular confounders such as age, traditional cardiovascular risk factors, RA treatment, and disease duration at the time of blood collection (Table 1).

Table 1. Baseline demographic and clinical characteristics of the patients with RA*
 RA patients (n = 133)Anti–Apo A-1 IgG negative (n = 110)Anti–Apo A-1 IgG positive (n = 23)P
  • *

    Except where indicated otherwise, values are the number (%). Normally distributed continuous values are expressed as the mean ± SD, with P values calculated using the 2-tailed t-test. Continuous variables with skewed distribution are expressed as the median (interquartile range [IQR]). The 2-sided Fisher's exact test was used to compare dichotomous variables. RA = rheumatoid arthritis; anti–Apo A-1 = anti–apolipoprotein A-1; CVD = cardiovascular disease; RF = rheumatoid factor; DMARDs = disease-modifying antirheumatic drugs; anti-TNFα = anti–tumor necrosis factor α; OD450 nm = optical density at 450 nm.

Age, mean ± SD years65 ± 16.365.2 ± 16.765.4 ± 14.80.95
Female sex95 (71)83 (75)12 (52)0.04
Cardiovascular risk factors    
 Smoker16 (12)16 (15)0 (0)0.31
 Diabetes mellitus16 (12)11 (13)3 (23)0.71
 Hypertension43 (32)34 (31)9 (39)0.46
 Obesity6 (5)5 (5)1 (4)1
 Dyslipidemia10 (8)7 (6)3 (13)0.37
 Positive family history1 (0.07)1 (1)0 (0)1
CVD incidence20 (15)10 (9)10 (43)0.001
RF positive96 (72)83 (75)13 (57)0.08
Disease duration, mean ± SD years19.7 ± 13.219.1 ± 13.419.8 ± 12.70.80
Time from sampling to CV event, mean ± SD years9 ± 5.49.4 ± 5.49.5 ± 5.80.95
Time from RA diagnosis to sampling, mean ± SD years9.8 ± 1110 ± 11.68.8 ± 90.64
Treatment    
 DMARDs95 (71)79 (72)16 (70)0.8
 Anti-TNFα27 (20)25 (23)2 (9)0.1
 Anti-CD2015 (11)13 (12)2 (9)1
Anti–Apo A-1 IgG positivity Anti–Apo A-1 IgG titer22 (17)
  Median OD450 nm0.23
  IQR (range)0.06–0.48 (0–1.33)

During a median followup of 9 years (IQR 5–15 years), the rate of major cardiovascular events was 15% (20 of 133 patients). During followup, 6 patients died of cardiovascular complications (4 fatal MIs and 2 fatal ischemic strokes), 10 patients experienced a nonfatal acute coronary syndrome (5 patients with ST elevation MI and 5 with non–ST elevation MI), and 4 patients had a nonfatal ischemic stroke.

Baseline characteristics associated with major cardiovascular events during followup.

As shown in Table 2, patients who had a major cardiovascular event during followup were older and were more likely to be diabetic, hypertensive, or dyslipidemic, and had higher circulating levels of IL-1Ra and oxidized LDL and lower HDL concentrations. However, a multivariable analysis in which age, sex, diabetes mellitus, hypertension, dyslipidemia, obesity, smoking, disease duration, oxidized LDL, IL-1Ra, HDL, and anti–Apo A-1 IgG were set as confounders, only hypertension (χ2 = 11.3, P = 0.0007), dyslipidemia (χ2 = 5.1, P = 0.02), and anti–Apo A-1 IgG (χ2 = 10, P = 0.001) remained significant predictors of major cardiovascular events. The confounder disease duration was close to significant in terms of predicting major cardiovascular events (χ2 = 3.38, P = 0.06). For these reasons, only anti–Apo A-1 IgG, traditional cardiovascular disease risk factors (age, sex, diabetes mellitus, hypertension, dyslipidemia, smoking, obesity), and disease duration were considered for further risk analysis.

Table 2. Baseline demographic and clinical characteristics of the patients with RA according to the occurrence or lack of occurrence of major adverse cardiac events during followup*
ParameterMACE positive (n = 20)MACE negative (n = 113)P
  • *

    Normally distributed continuous values are expressed as the mean ± SD, and P values were calculated using the 2-tailed t-test. Continuous variables with skewed distribution are expressed as the median (range [interquartile range (IQR)]), and P values were calculated using the Mann-Whitney U test. The two-sided Fisher's exact test was used to compare dichotomous variables. MACE = major cardiovascular event; anti–Apo A-1 = anti–apolipoprotein A-1; OD450 nm = optical density at 450 nm; TNF = tumor necrosis factor; IL-6 = interleukin-6; MCP-1 = monocyte chemotactic protein 1; IL-1Ra = interleukin-1 receptor antagonist; CRP = C-reactive protein; HDL = high-density lipoprotein; LDL = low-density lipoprotein; MMP-1 = matrix metalloproteinase 1; DMARDs = disease-modifying antirheumatic drugs.

  • Mean ± SD ng/ml.

Male sex, no. (%)9 (45)29 (26)0.1
Age, mean ± SD years73 ± −1263 ± −160.02
Hypertension, no. (%)14 (70)29 (26)<0.001
Diabetes mellitus, no. (%)6 (30)10 (8)0.01
Dyslipidemia, no. (%)7 (35)3 (2)<0.001
Obesity, no. (%)1 (5)5 (4)0.48
Smoker, no. (%)2 (10)14 (12)1
Rheumatoid factor positive, no. (%)13 (65)83 (69)0.42
Disease duration, mean ± SD years23 ± −1619 ± −130.26
Biochemical parameters, median (range) [IQR]   
 Anti–Apo A-1 IgG, OD450 nm0.58 (0–1.26) [0.23–0.93]0.21 (0–1.3) [0.05–0.4]0.001
 TNFα, pg/ml2.1 (0–5) [1.2–2.6]1.6 (0–249) [1.2–2.2]0.42
 IL-6, pg/ml3.7 (0–56) [1.3–14]3.4 (0–140) [1.2–7.7]0.67
 IL-8, pg/ml11 (2–197) [7–24]7.5 (0.7–290) [2.8–15.5]0.05
 MCP-1, pg/ml97 (33–2,011) [53–200]99.3 (51–163) [44–200]0.6
 IL-1Ra, pg/ml581 (120–3,044) [415–1,197]388 (0–6,481) [190–849]0.04
 CRP, mg/liter13.5 (5.5–93) [7–29]15.3 (1–287) [4.2–37]0.6
 HDL, mmoles/liter0.8 (0.6–1.7) [0.7–1.3]1.05 (0.3–2.2) [0.9–1.4]0.02
 LDL, mmoles/liter3.2 (1.8–4.4) [3.8–5.6]2.8 (1.4–4.7) [2.4–3.6]0.52
 Triglycerides, mmoles/liter1.2 (0.6–7.4) [0.8–1.6]1.08 (0.4–4.1) [0.8–1.5]0.56
 Cholesterol, mmoles/liter4.9 (2.9–9.8) [3.8–5.6]4.6 (2.1–6.7) [4–5.3]0.54
 Oxidized LDL, units/ml65 (13–91) [46–70]46 (0–84) [36–63]0.006
 MMP-1, ng/ml6.6 (0.8–26) [1.8–6.5]4.3 (0–45) [2.2–8.6]0.41
 MMP-2, ng/ml144.4 (0.1–249) [129–161]140.9 (0.7–202) [127–154]0.36
 MMP-3, ng/ml65.9 (27–106) [52–90]57.4 (0.2–118) [31–80]0.16
 MMP-9, ng/ml593 (56–3,737) [265–1,662]891 (3–3,019) [326–1,547]0.76
 Log proMMP-9 activity9.96 ± 0.879.86 ± −0.620.52
Treatment   
 DMARDs, no. (%)15 (75)80 (71)0.79
 Anti-TNF, no. (%)2 (10)25 (22)0.36
 Anti-CD20, no. (%)0 (0)15 (13)0.3

Association of anti–Apo A-1 IgG positivity with future major cardiovascular events.

RA patients positive for anti–Apo A-1 IgG had a significantly higher rate of major cardiovascular events (43% versus 9% of anti–Apo A-1 IgG–negative patients; P = 0.001) (Table 1). Translated into a crude HR, the presence of anti–Apo A-1 IgG positivity increased the risk of major cardiovascular events 5-fold (HR 4.7, 95% CI 1.9–11.2). This association remained significant and of the same order of magnitude after adjustment for age, sex, hypertension, dyslipidemia, smoking, diabetes mellitus, obesity, and RA disease duration (HR 4.2, 95% CI 1.5–12.1). The presence of a significant dose response (P for trend = 0.02) suggests that the risk of a major cardiovascular event increases proportionally to increasing anti–Apo A-1 IgG concentrations. Indeed, in the second tertile (for anti–Apo A-1 IgG, OD450 nm = 0.12–0.37), the major cardiovascular event risk did not significantly increase (HR 1.3, 95% CI 0.3–6, P = 0.67), but in the third tertile (for anti–Apo A-1 IgG, OD = 0.38–1.33), the risk of major cardiovascular events increased significantly (2-fold) (HR 2.2, 95% CI 1.2–14.4). Major cardiovascular event–free survival was 86% in RA patients negative for anti–Apo A-1 IgG and 40% in RA patients positive for anti–Apo A-1 IgG (P = 0.0001 by log rank test) (Figure 1). Furthermore, patients with RA who had an incident major cardiovascular event during followup had higher median values for anti–Apo A-1 IgG at baseline when compared with patients without major cardiovascular events (Table 2), and this difference remained significant after Bonferroni correction (data not shown).

thumbnail image

Figure 1. Kaplan-Meier plot showing the time to cardiovascular (CV) events according to the anti–apolipoprotein A-1 (anti–Apo A-1) IgG status of the patients. Survival analysis was performed without adjustment for traditional cardiovascular risk factors.

Download figure to PowerPoint

ROC analysis.

ROC analysis showed that the presence of anti–Apo A-1 IgG is a good predictor of major cardiovascular events in patients with RA, with an area under the curve value of 0.73 (95% CI 0.59–0.86, P = 0.0008). At the predefined positivity cutoff, set at an OD value of 0.6, corresponding to the 97.5th percentile of the distribution defined earlier (14) and recently validated prospectively in patients with MI (15), the sensitivity and specificity of anti–Apo A-1 IgG to predict major cardiovascular events were 90% (95% CI 83–94) and 50% (95% CI 28–72), respectively. Negative and positive predictive values were 90% (95% CI 84–95) and 48% (95% CI 26–70), respectively.

Anti–Apo A-1 IgG and inflammation parameters.

As shown in Table 3, anti–Apo A-1 IgG–positive patients with RA had higher median serum levels of IL-8, MMP-9, oxidized LDL, and proMMP-9 activity, but none of these differences was significant after Bonferroni correction (data not shown).

Table 3. Baseline biologic markers according to anti–Apo A-1 IgG status*
Biologic biomarkersAnti–Apo A-1 IgG–negative patients (n = 110)Anti–Apo A-1 IgG–positive patients (n = 23)P
  • *

    Except where indicated otherwise, values are the median (interquartile range). Anti–Apo A-1 = anti–apolipoprotein A-1; CRP = C-reactive protein; IL-6 = interleukin-6; TNFα = tumor necrosis factor α; IL-1Ra = IL-1 receptor antagonist; MCP-1 = monocyte chemotactic protein 1; MMP-9 = matrix metalloproteinase 9; LDL = low-density lipoprotein; HDL = high-density lipoprotein.

  • By Mann-Whitney U test.

Inflammation marker   
 CRP, mg/liter15.4 (4.6–37.2)14.1 (6–31)0.96
 IL-6, pg/ml3.4 (1.2–7.5)3.5 (1–11)0.66
 IL-8, pg/ml7.2 (2.9–13.8)14 (6.3–29.1)0.01
 TNFα, pg/ml1.7 (1.2–2.3)1.9 (1–2.6)0.49
 IL-1Ra, pg/ml398 (198–864)550 (281–1,075)0.28
 MCP-1, pg/ml102 (44–201)89 (46–195)0.89
 MMP-9, ng/ml787 (283–1,490)1,064 (563–2,103)0.03
 Log proMMP-9 activity, mean ± SD ng/ml9.88 ± 0.3310.34 ± 1.460.008
 MMP-1, ng/ml4 (2–8)4.4 (2–7)0.95
 MMP-2, ng/ml141 (127–154)142 (122–157.6)0.75
 MMP-3, ng/ml61.6 (33–80)63 (33–91)0.48
 Oxidized LDL, units/ml46.5 (36–65)59 (46–72)0.02
Lipid profile   
 Total cholesterol, mmoles/liter4.75 (2.1–9.8)4.55 (3–6.1)0.4
 HDL, mmoles/liter1.05 (0.3–2.2)0.9 (0.5–1.8)0.07
 LDL, mmoles/liter2.9 (1.4–4.7)3 (1.4–4.2)0.88
 Triglycerides, mmoles/liter1.03 (0.5–3.3)1.25 (0.8–1.6)0.61

Anti–Apo A-1 IgG–stimulated IL-8 and MMP-9 production by cultured human macrophages.

To further explore the possibility of a causal relationship between high levels of circulating anti–Apo A-1 IgG and inflammation mediators such as IL-8 and MMP-9, human monocyte–derived macrophages were cultured with increasing concentrations of commercial anti–Apo A-1 IgG or respective IgG controls for 24 hours and 48 hours, respectively.

As shown in Figure 2, anti–Apo A-1 IgG induced dose-dependent production of IL-8 (P for trend = 0.0005 by Kruskal-Wallis nonparametric test) and MMP-9 (P for trend = 0.03 by Kruskal-Wallis nonparametric test) by human macrophages. The results of trend tests for control IgG were not significant (P = 0.62 and P = 0.75, respectively). For IL-8, the maximal effect was observed at 40 μg/ml, whereas the optimal stimulating concentration for MMP-9 was 10 μg/ml. Control IgG did not significantly stimulate macrophages, and the differences between anti–Apo A-1 IgG and control IgG for IL-8 and MMP-9 production were statistically significant at 20 μg/ml and 10 μg/ml, respectively (P = 0.03 for both) (Figure 2).

thumbnail image

Figure 2. Anti–apolipoprotein A-1 (anti–Apo A-1) IgG–induced dose-dependent production of interleukin-8 (IL-8) and matrix metalloproteinase 9 (MMP-9) by human macrophages. Bars show the median (range). ∗ and ∗∗∗ = P = 0.03 versus control, by Mann-Whitney U test. LPS = lipopolysaccharide.

Download figure to PowerPoint

Furthermore, the dose-dependent increase in MMP-9 production was accompanied by a concomitant increase in proMMP-9 activity as assessed by zymography (Figure 3A), which followed the same dose-dependent pattern (Figure 3B). There was a significant correlation between MMP-9 levels and proMMP-9 activity in the supernatant of anti–Apo A-1 IgG–stimulated macrophages (r = 0.59, P < 0.01) (data not shown). The absence of LPS contamination of anti–Apo A-1 IgG and the control IgG preparation was confirmed by Limulus assay (0.1 EU/ml for both antibodies) (data not shown).

thumbnail image

Figure 3. Anti–apolipoprotein A-1 (anti–Apo A-1) IgG–induced dose-dependent increase in pro–matrix metalloproteinase 9 (proMMP-9) activity in human macrophage supernatants, as assessed by zymography. A, White band (arrow) on the gel represents the proMMP-9 gelatinolytic activity of human macrophage supernatant in the presence of standard (ST), control medium (NT), and increasing concentrations of control lgG and anti–Apo A-1 IgG (5–40 μg/ml). Dark bands represent the amount of protein stained by Coomassie blue. B, Mean ± SD results of zymography. ∗∗∗ = P = 0.03 versus control, by t-test. NT = not treated.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

The main findings of this study are that anti–Apo A-1 IgG exist in a significant subset of patients with RA (17%), and that these autoantibodies are predictive of major cardiovascular events in patients with RA, independent of age, sex, diabetes mellitus, smoking, obesity, disease duration, and IL-1Ra and oxidized LDL levels. Furthermore, our results suggest that among all emergent markers of cardiovascular risk tested in the present study, anti–Apo A-1 IgG positivity was the strongest predictor of major cardiovascular events in patients with RA.

Our results show that anti–Apo A-1 IgG positivity tended to be associated with enhanced systemic inflammatory responses, as reflected by higher levels of IL-8, oxidized LDL, MMP-9, and MMP-9 activity (P not significant after Bonferroni adjustment). However, because Bonferroni correction is known to be prone to a lack of power (Type II error) by missing dose-response relationships between variables (22), we tested the ability of anti–Apo A-1 IgG to induce the production of MMP-9 and IL-8 by human monocyte–derived macrophages, which are key players in inflammation and atherosclerosis. Our results show that physiologically relevant anti–Apo A-1 IgG concentrations induced dose-dependent IL-8 and MMP-9 production by human macrophages with the same order of magnitude as LPS (1 μg/ml), whereas control IgG were devoid of any stimulatory effect.

MMP-9 levels have been reported to transiently increase during acute coronary syndrome (23), stroke (24), and heart failure (25), and increased levels are associated with more unstable atherosclerotic plaques (26). Other MMPs such as MMP-1, MMP-2, and MMP-3 have also been associated with increased cardiovascular risk, but to the best of our knowledge, the most consistent data have been observed with MMP-9. These clinical results are corroborated by results of in vitro and in vivo studies, showing that MMP-9 increases atherosclerotic plaque fragility by degrading type IV collagen within the fibrous cap, by increasing transendothelial migration of immune cells within atherosclerotic plaques, and by stimulating the proliferation of vascular smooth muscle cells (11, 27). In patients with RA, circulating MMP-9 levels have been reported to be higher than those in control individuals (28) and are associated with higher disease activity and progression of joint damage (29). The significance of elevated MMP-9 levels in RA with respect to cardiovascular events remains elusive. However, our data do not suggest that MMP-9 is an independent predictor of major cardiovascular events in RA.

In RA joints, IL-8–secreting cells have been detected within rheumatoid synovium at the pannus–cartilage junction (30). IL-8 is a major neutrophil chemoattractant, and neutrophil activation is currently considered to play a critical role in the development of joint inflammation (31), suggesting that IL-8 contributes to the mechanisms leading to tissue damage. However, the pathogenic role of IL-8 in RA-related cardiovascular complications is unknown. IL-8 has been shown to be predictive of fatal and nonfatal coronary artery disease in healthy subjects (32), to be a powerful and independent predictor for cardiovascular events in patients with coronary artery disease (33), and to stimulate MMP-9 production (34). These data are part of the compelling body of evidence suggesting an important role of neutrophil activation in atherosclerosis-related cardiovascular complications in humans (35, 36). Indeed, neutrophils have been shown to localize within atherosclerotic plaques (37), to enhance local vascular inflammation and myeloperoxidase release (38). Neutrophils increase the production of radical oxygen species, thus promoting oxidized LDL formation (36) and impeding the atheroprotective functions of HDL by mediating specific Apo A-1 chlorination (39).

Oxidized LDL, produced upon lipid peroxidation, has been reported to play a major role in all stages of atherogenesis, including induction of endothelial dysfunction, transendothelial migration of immune cells, and foam cell formation (40). Plasma levels of oxidized LDL have been described as promising diagnostic and prognostic markers of cardiovascular complications (40) as well as markers of the angiographic severity of coronary heart disease in patients with acute coronary syndromes (41, 42). In RA, oxidized LDL has been associated with disease that is more active (10), but the association of oxidized LDL with cardiovascular events remains elusive. The results of our study do not suggest that oxidized LDL is an independent predictor of cardiovascular complications in RA. This result contrasts with previous findings in other autoimmune diseases such as SLE and APS, in which high levels of oxidized LDL were associated with arterial disease (43) and thrombotic events (44), respectively. In the context of acute coronary syndrome, anti–Apo A-1 IgG positivity was associated with 5-fold higher levels of oxidized LDL compared with anti–Apo A-1 IgG negativity (14). The link between anti–Apo A-1 IgG and oxidized LDL is unclear but could be mediated by IL-8–related lipid peroxidation, possibly through leukocyte myeloperoxidase activation (37, 39).

Current thinking supports the notion that RA and atherosclerosis share several common inflammatory proteins intervening with endothelium and hemostatic factors, which in turn leads to plaque formation and rupture (9). Whether some autoantibodies could actively take part in this process remains elusive, and our results showing that anti–Apo A-1 IgG per se are able to induce IL-8 and MMP-9 production by human macrophages add weight to this hypothesis and further underline the role of humoral autoimmunity in RA-related cardiovascular complications.

Our results also raise several questions such as how anti–Apo A-1 antibodies are involved in the increased incidence of cardiovascular events. Anti–Apo A-1 IgG induces production of proinflammation mediators by macrophages, which are involved in atherogenesis and plaque rupture. To the best of our knowledge, Apo A-1 is not expressed by human macrophages. However, anti–Apo A-1 antibodies induce specific concentration-dependent stimulatory responses in cultured macrophages, suggesting that these antibodies may exert their stimulatory effects by interacting with Fcγ receptors or innate immune receptors, as recently described for anticardiolipin antibodies (45). Alternatively, these antibodies may recognize a common conformational epitope expressed by macrophages. Another nonexclusive hypothesis is that anti–Apo A-1 IgG could interfere with autonomic cardiac regulation, which is increasingly recognized as a vector of sudden cardiac death in RA (46). In acute MI, anti–Apo A-1 IgG positivity is associated with a higher basal heart rate, and plasma from patients positive for anti–Apo A-1 antibodies induced a strong positive chronotropic effect in vitro (15). Furthermore, spiking anti–Apo A-1 IgG into control plasma induced a dose-dependent chronotropic response, suggesting that anti–Apo A-1 IgG mediate the chronotropic effects of plasma samples from patients with MI, and may affect autonomic cardiac function in MI (15). However, it remains to be demonstrated whether this applies to cardiovascular events in patients with RA.

Despite this study being appropriately powered, the number of events was still relatively low, which prevented us from drawing definite conclusions. Although risk analyses were adjusted for the traditional cardiovascular risk factors used to compute the Framingham risk score, we could not formally compare the predictive accuracy of the Framingham risk score and that of anti–Apo A-1 IgG, because the exact values for systolic blood pressure were not available for the total study sample. Because the Framingham risk score does not appear to perform particularly well in patients with RA (47), knowing whether anti–Apo A-1 IgG could outperform the Framingham risk score for predicting major cardiovascular events in RA clearly warrants further study. Additionally, the antibodies used for in vitro experiments in this study were of commercial origin and were not extracted from the serum of patients; extraction of serum was not possible given the insufficient amount of material available. However, because we used LPS-free anti-human Apo A-1 IgG, we believe that our results established proof of principle that anti–Apo A-1 IgG per se are sufficient to trigger significant proinflammatory responses. Another limitation of this study is that disease activity measurements were not available. Because disease activity has been shown to predict the cardiovascular outcome in patients with RA (48), analyzing anti–Apo A-1 IgG with respect to the disease activity score is definitely of interest and warrants further study.

Nevertheless, to our knowledge, no simple diagnostic test is currently available to assess cardiovascular risk in patients with RA, and our results suggest that anti–Apo A-1 IgG testing could represent a useful tool for stratifying patients according to the risk of cardiovascular events. Thus, the presence of anti–Apo A-1 IgG may permit the identification of a subset of patients who could benefit from early and specific cardiovascular risk prevention, including the management of traditional risk factors and better control of RA disease activity.

Finally, because the presence of rheumatoid factor (RF) in serum samples is known to induce false-positive results in immunoassays, it can be contended that the high anti–Apo A-1 IgG positivity rate observed in RA could be attributable to this analytical pitfall. The facts that RF-positive patients with RA had significantly lower median values for anti–Apo A-1 IgG (OD = 0.22) compared with RF-negative patients with RA (OD = 0.33; P = 0.02 by Mann-Whitney U test), and that no correlation was observed between the presence of RF (in terms of latex titer or IgM RF concentrations) and anti–Apo A-1 IgG titers provide evidence against that hypothesis. Therefore, although it is not formally excluded, RF interference is very unlikely to have blunted the results and the conclusions of this study.

In conclusion, this study is the first to demonstrate that anti–Apo A-1 IgG predict major cardiovascular events in patients with RA, and that anti–Apo A-1 IgG positivity is associated with higher levels of MMP-9, IL-8, and oxidized LDL, which are 3 markers and mediators of atherosclerotic plaque destabilization. In addition, the results of in vitro experiments showing that anti–Apo A-1 IgG per se function in a manner similar to that of a proinflammatory molecule indicate that anti–Apo A-1 IgG are not only a marker of RA-associated cardiovascular complications but also could contribute to such complications by promoting inflammation and atherosclerotic plaque destabilization.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Vuilleumier had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Vuilleumier, Bas, Lovis, Hochstrasser, Roux-Lombard, Gabay.

Acquisition of data. Vuilleumier, Pagano, Montecucco, Guerne, Finckh, Mach.

Analysis and interpretation of data. Vuilleumier, Pagano, Montecucco, Finckh, Mach, Hochstrasser.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

We are indebted to C. Viglino for skillful technical assistance.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES
  • 1
    Kaplan MJ. Management of cardiovascular disease risk in chronic inflammatory disorders. Nat Rev Rheumatol 2009; 5: 20817.
  • 2
    Del Rincon I, O'Leary DH, Freeman GL, Escalante A. Acceleration of atherosclerosis during the course of rheumatoid arthritis. Atherosclerosis 2007; 195: 35460.
  • 3
    Solomon DH, Karlson EW, Rimm EB, Cannuscio CC, Mandl LA, Manson JE, et al. Cardiovascular morbidity and mortality in women diagnosed with rheumatoid arthritis. Circulation 2003; 107: 13037.
  • 4
    Wolfe F, Freundlich B, Straus WL. Increase in cardiovascular and cerebrovascular disease prevalence in rheumatoid arthritis. J Rheumatol 2003; 30: 3640.
  • 5
    Sodergren A, Stegmayr B, Lundberg V, Ohman ML, Wallberg-Jonsson S. Increased incidence of and impaired prognosis after acute myocardial infarction among patients with seropositive rheumatoid arthritis. Ann Rheum Dis 2007; 66: 2636.
  • 6
    Aukrust P, Yndestad A, Smith C, Ueland T, Gullestad L, Damas JK. Chemokines in cardiovascular risk prediction. Thromb Haemost 2007; 97: 74854.
  • 7
    Hansson GK. Inflammation, atherosclerosis, and coronary heart disease. N Engl J Med 2005; 352: 168595.
  • 8
    Edwards CJ, Syddall H, Goswami R, Goswami P, Dennison EM, Arden NK, et al. The autoantibody rheumatoid factor may be an independent risk factor for ischaemic heart disease in men. Heart 2007; 93: 12637.
  • 9
    Montecucco F, Mach F. Common inflammatory mediators orchestrate pathophysiological processes in rheumatoid arthritis and atherosclerosis. Rheumatology (Oxford) 2009; 48: 1122.
  • 10
    Kim SH, Lee CK, Lee EY, Park SY, Cho YS, Yoo B, et al. Serum oxidized low-density lipoproteins in rheumatoid arthritis. Rheumatol Int 2004; 24: 2303.
  • 11
    Newby AC. Metalloproteinases and vulnerable atherosclerotic plaques. Trends Cardiovasc Med 2007; 17: 2538.
  • 12
    O'Neill SG, Giles I, Lambrianides A, Manson J, D'Cruz D, Schrieber L, et al. Antibodies to apolipoprotein A-1, high-density lipoprotein, and C-reactive protein are associated with disease activity in patients with systemic lupus erythematosus. Arthritis Rheum 2010; 62: 84554.
  • 13
    Vuilleumier N, Reber G, James R, Burger D, de Moerloose P, Dayer JM. Presence of autoantibodies to apolipoprotein A-1 in patients with acute coronary syndrome further links autoimmunity to cardiovascular disease. J Autoimmun 2004; 23: 35360.
  • 14
    Vuilleumier N, Charbonney E, Fontao L, Alvarez M, Turck N, Sanchez JC, et al. Anti-apolipoprotein A-1 IgG are associated with high oxidised low-density lipoprotein levels in acute coronary syndrome. Clin Sci 2008; 115: 2533.
  • 15
    Vuilleumier N, Rossier MF, Pagano S, Python M, Charbonney E, Nkoulou R, et al. Anti-apolipoprotein A-1 autoantibody of IgG subtype as an independent cardiovascular prognostic marker affecting basal heart rate in myocardial infarction. Eur Heart J 2010; 31: 81523.
  • 16
    Hochberg MC, for the Diagnostic and Therapeutic Criteria Committee of the American College of Rheumatology. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus [letter]. Arthritis Rheum 1997; 40: 1725.
  • 17
    Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette JC, et al. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum 1999; 42: 130911.
  • 18
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 19
    Altieri P, Brunelli C, Garibaldi S, Nicolino A, Ubaldi S, Spallarossa P, et al. Metalloproteinases 2 and 9 are increased in plasma of patients with heart failure. Eur J Clin Invest 2003; 33: 64856.
  • 20
    Hyka N, Dayer JM, Modoux C, Kohno T, Edwards CK, Roux-Lombard P, et al. Apolipoprotein A-I inhibits the production of interleukin-1β and tumor necrosis factor-α by blocking contact-mediated activation of monocytes by T lymphocytes. Blood 2001; 97: 23819.
  • 21
    Bryans TD, Braithwaite C, Broad J, Cooper JF, Darnell KR, Hitchins VM, et al. Bacterial endotoxin testing: a report on the methods, background, data, and regulatory history of extraction recovery efficiency. Biomed Instrum Technol 2004; 38: 738.
  • 22
    Quan H, Luo X, Capizzi T. Multiplicity adjustment for multiple endpoints in clinical trials with multiple doses of an active treatment. Stat Med 2005; 24: 215170.
  • 23
    Furenes EB, Arnesen H, Solheim S, Grogaard HK, Hoffmann P, Seljeflot I. The profile of circulating metalloproteinases after PCI in patients with acute myocardial infarction or stable angina. Thromb Res 2009; 124: 5604.
  • 24
    Jefferis BJ, Whincup P, Welsh P, Wannamethee G, Rumley A, Lennon L, et al. Prospective study of matrix metalloproteinase-9 and risk of myocardial infarction and stroke in older men and women. Atherosclerosis 2010; 208: 5576.
  • 25
    Kelly D, Cockerill G, Ng LL, Thompson M, Khan S, Samani NJ, et al. Plasma matrix metalloproteinase-9 and left ventricular remodelling after acute myocardial infarction in man: a prospective cohort study. Eur Heart J 2007; 28: 7118.
  • 26
    Loftus IM, Naylor AR, Goodall S, Crowther M, Jones L, Bell PR, et al. Increased matrix metalloproteinase-9 activity in unstable carotid plaques: a potential role in acute plaque disruption. Stroke 2000; 31: 407.
  • 27
    Ram M, Sherer Y, Shoenfeld Y. Matrix metalloproteinase-9 and autoimmune diseases. J Clin Immunol 2006; 26: 299307.
  • 28
    Chang YH, Lin IL, Tsay GJ, Yang SC, Yang TP, Ho KT, et al. Elevated circulatory MMP-2 and MMP-9 levels and activities in patients with rheumatoid arthritis and systemic lupus erythematosus. Clin Biochem 2008; 41: 9559.
  • 29
    Tchetverikov I, Lard LR, DeGroot J, Verzijl N, TeKoppele JM, Breedveld FC, et al. Matrix metalloproteinases-3,-8,-9 as markers of disease and joint damage progression in early rheumatoid arthritis. Ann Rheum Dis 2003; 62: 10949.
  • 30
    Deleuran BW. Cytokines in rheumatoid arthritis: localization in arthritic joint tissue and regulation in vitro. Scand J Rheumatol Suppl 1996; 104: 913.
  • 31
    Ottonello L, Cutolo M, Frumento G, Arduino N, Bertolotto M, Mancini M, et al. Synovial fluid from patients with rheumatoid arthritis inhibits neutrophil apoptosis: role of adenosine and proinflammatory cytokines. Rheumatology (Oxford) 2002; 41: 124960.
  • 32
    Boekholdt SM, Peters RJ, Hack CE, Day NE, Luben R, Bingham SA, et al. IL-8 plasma concentrations and the risk of future coronary artery disease in apparently healthy men and women: the EPIC-Norfolk prospective population study. Arterioscler Thromb Vasc Biol 2004; 24: 15038.
  • 33
    Inoue T, Komoda H, Nonaka M, Kameda M, Uchida T, Node K. Interleukin-8 as an independent predictor of long-term clinical outcome in patients with coronary artery disease. Int J Cardiol 2008; 124: 31925.
  • 34
    Dorweiler B, Torzewski M, Dahm M, Kirkpatrick CJ, Lackner K, Vahl C. Subendothelial infiltration of neutrophil granulocytes and liberation of matrix-destabilizing enzymes in an experimental model of human neo-intima. Thromb Haemost 2008; 99: 37381.
  • 35
    Madjid M, Awan I, Willerson JT, Casscells SW. Leukocyte count and coronary heart disease: implications for risk assessment. J Am Coll Cardiol 2004; 44: 194556.
  • 36
    Quinn K, Henriques M, Parker T, Slutsky AS, Zhang H. Human neutrophil peptides: a novel potential mediator of inflammatory cardiovascular diseases. Am J Physiol Heart Circ Physiol 2008; 295: 181724.
  • 37
    Van Leeuwen M, Gijbels MJ, Duijvestijn A, Smook M, van de Gaar MJ, Heeringa P, et al. Accumulation of myeloperoxidase-positive neutrophils in atherosclerotic lesions in LDLR-/- mice. Arterioscler Thromb Vasc Biol 2008; 28: 849.
  • 38
    Mazor R, Shurtzswirski R, Farah R, Kristal B, Shapiro G, Dorlechter F, et al. Primed polymorphonuclear leukocytes constitute a possible link between inflammation and oxidative stress in hyperlipidemic patients. Atherosclerosis 2008; 197: 93743.
  • 39
    Nicholls SJ, Hazen SL. Myeloperoxidase, modified lipoproteins and atherogenesis. J Lipid Res 2009; 50 Suppl: S34651.
  • 40
    Holvoet P, Mertens A, Verhamme P, Bogaerts K, Beyens G, Verhaeghe R, et al. Circulating oxidized LDL is a useful marker for identifying patients with coronary heart disease. Arterioscler Thromb Vasc Biol 2001; 21: 8448.
  • 41
    Yamashita H, Ehara S, Yoshiyama M, Naruko T, Haze K, Shirai N, et al. Elevated plasma levels of oxidized low-density lipoprotein relate to the presence of angiographically detected complex and thrombotic coronary artery lesion morphology in patients with unstable angina. Circ J 2007; 71: 6817.
  • 42
    Anselmi M, Garbin U, Agostoni P, Fusaro M, Pasini AF, Nava C, et al. Plasma levels of oxidized-low-density lipoproteins are higher in patients with unstable angina and correlated with angiographic coronary complex plaques. Atherosclerosis 2006; 185: 11420.
  • 43
    Svenungsson E, Jensen-Urstad K, Heimburger M, Silveira A, Hamsten A, de Faire U, et al. Risk factors for cardiovascular disease in systemic lupus erythematosus. Circulation 2001; 104: 188793.
  • 44
    Matsuura E, Kobayashi K, Koike T, Shoenfeld Y, Khamashta MA, Hughes GR. Oxidised low-density lipoprotein as a risk factor of thrombosis in antiphospholipid syndrome. Lupus 2003; 12: 5504.
  • 45
    Satta N, Dunoyer-Geindre S, Reber G, Fish RJ, Boehlen F, Kruithof EK, et al. The role of TLR2 in the inflammatory activation of mouse fibroblasts by human antiphospholipid antibodies. Blood 2007; 109: 150714.
  • 46
    Stojanovich L. Autonomic dysfunction in autoimmune rheumatic disease. Autoimmun Rev 2009; 8: 56972.
  • 47
    Chung CP, Oeser A, Solus JF, Avalos I, Gebretsadik T, Shintani A, et al. Utility of Framingham risk score to predict the presence of coronary atherosclerosis in patients with rheumatoid arthritis. Arthritis Res Ther 2006; 8: R186.
  • 48
    Turesson C, Jacobsson LT, Matteson EL. Cardiovascular co-morbidity in rheumatic disease. Vasc Health Risk Manag 2008; 4: 60514.