SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Objective

To compare the incidence and extent of coronary artery calcification (CAC) as measured by electron beam computed tomography (EBCT) in patients with systemic lupus erythematosus (SLE) and controls, and to identify variables associated with CAC in patients with SLE.

Methods

Female patients with SLE and matched controls were recruited; EBCT of the coronary arteries was performed, and laboratory values (including the homocysteine concentration, the lipid level, the high-sensitivity C-reactive protein [hsCRP] concentration, the glomerular filtration rate [GFR], and the level of soluble CD154 [sCD154]) were determined. For patients, the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index and the SLE Disease Activity Index scores were recorded. Tests of association between the CAC score and the above-mentioned variables were performed.

Results

The incidence of CAC was higher in patients with SLE than in controls (P = 0.009), and patients had a higher mean raw CAC (rCAC) score (87.9 versus 9.6 in controls; P = 0.02). In particular, more CAC-positive patients than CAC-positive controls had rCAC scores above the 75th percentile (P = 0.003). Among both patients and controls, those with CAC were ∼10 years older than those without CAC. In addition to age, a significant determinant of positive CAC status in both groups was the number of cardiovascular risk factors. In patients with SLE, CAC was associated with a higher homocysteine concentration, a lower GFR, and longer disease duration. In controls, the total cholesterol level correlated positively with CAC. When multivariate logistic regression methods were applied to candidate explanatory variables, homocysteine concentration, age, and disease duration (but not the levels of sCD154 or hsCRP) contributed significantly to CAC status. The methylenetetrahydrofolate reductase C677T genotype was not a predictor of hyperhomocysteinemia or CAC status.

Conclusion

Among patients with SLE, the homocysteine concentration, the GFR, age, and disease duration were associated with CAC. CAC occurred more frequently and was more extensive in patients with SLE than in controls, suggesting that EBCT could be used to detect premature atherosclerosis in the former group. An elevated homocysteine concentration might identify patients with SLE who are likely to have premature atherosclerosis and who would benefit from evaluation of CAC by EBCT.

Patients with systemic lupus erythematosus (SLE) have a 4–10-fold greater risk of atherosclerotic cardiovascular disease (ASCVD) compared with the general population (1). Although traditional Framingham risk factors may be inadequate for assessing ASCVD risk in patients with SLE (2), the high incidence of cardiovascular morbidity and mortality in these patients underscores the need for objective and accurate means by which to identify those with subclinical disease and to apply targeted interventions.

Electron beam computed tomography (EBCT) is used to detect coronary artery calcification (CAC) within atheromatous plaques. CAC scoring by EBCT correlates with the total histopathologic and arteriographic burden and can be used to predict future cardiovascular events (3). CAC scoring by EBCT has been used in many large population-based studies to evaluate the prognostic utility of this approach in the primary prevention setting (4, 5).

The pathogenesis of ASCVD in SLE is likely multifactorial, involving traditional risk factors, early menopause, treatment side effects, and immunologic/inflammatory components. Although elevated homocysteine levels have been associated with stroke and thrombosis in patients with SLE, reports linking hyperhomocysteinemia with ASCVD are lacking (6). In the general population, however, hyperhomocysteinemia is independently associated with an increased risk of ASCVD (7), and the methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism is associated with hyperhomocysteinemia in healthy individuals with low-folate status (8).

The aim of this study was to determine the incidence and extent of CAC, an early manifestation of ASCVD, in patients with SLE compared with controls, and to identify variables that are associated with CAC in this population. We postulated that patients with SLE are more likely than controls to have CAC, and that EBCT could identify subclinical ASCVD in women with SLE.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

SLE and control samples.

Consecutive nonpregnant women older than age 18 years who fulfilled at least 4 of the American College of Rheumatology revised criteria for the classification of SLE (9) were invited to participate. Age-matched (±2 years) and race-matched nonpregnant female controls were recruited from the University of Pennsylvania clinics. The study was approved by the University of Pennsylvania Institutional Review Board, and informed consent was obtained from each participant.

Data collection.

A medical history was obtained for all study participants. All participants underwent a physical examination, electrocardiography, and EBCT, and had a fasting blood sample drawn. SLE disease activity and organ damage were assessed using the SLE Disease Activity Index (10) and the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (11).

Clinical assessments.

Laboratory analysis.

Fasting blood samples were tested to determine the lipid profile, the antiphospholipid antibody profile (dilute Russell's viper venom test, anticardiolipin antibodies by enzyme-linked immunosorbent assay [ELISA], antibodies to β2-glycoprotein I by ELISA), the levels of C3, C4, and anti–double-stranded DNA antibody (by ELISA), the Westergren erythrocyte sedimentation rate, the level of high-sensitivity C-reactive protein (hsCRP), the complete blood cell count, the level of creatinine, and the total plasma homocysteine concentration.

Measurement of soluble CD154 (sCD154).

Plasma was tested in duplicate for sCD154 levels, by ELISA (Chemicon, Temecula, CA).

Measurement of homocysteine.

Whole blood (5 ml) was drawn into EDTA, placed on ice, and centrifuged at 2,500 revolutions per minute for 5 minutes at room temperature. Plasma homocysteine concentrations were determined by fluorescence polarization immunoassay (AxSYM Homocysteine; Abbott Laboratories, Abbott Park, IL).

Genetics analysis.

DNA was isolated using Generation Capture Column Kits (Gentra Systems, Minneapolis, MN). MTHFR C677T genotypes were analyzed using a heteroduplex generator method, as previously described (12).

Glomerular filtration rate (GFR).

To assess possible renal insufficiency, the GFR for each participant was calculated using the Modification of Diet in Renal Disease equation (13).

Risk factors.

The following cardiovascular risk factors were assessed: diabetes, hypertension (systolic blood pressure [BP] >140 mm Hg and/or diastolic BP >90 mm Hg on 2 occasions), postmenopausal status (i.e., >1 year since last menstrual period), history of ASCVD (i.e., previous myocardial infarction [MI], coronary artery bypass surgery, or angiographically proven stenosis), a significant family history of heart disease (i.e., MI, sudden cardiac death, or a revascularization procedure in a first-degree male relative younger than age 55 years and/or a first-degree female relative younger than age 65 years), and current-smoker status. Each category was assigned a value of 1 point, and the summed value (range 0–6) was determined for each participant (14). Hypercholesterolemia was analyzed as a separate risk factor and was not included in the summed value.

EBCT.

Patients with SLE and controls had electrodes placed for gated data acquisition and were positioned supine in a GE Imatron C150 electron beam CT scanner (General Electric Medical Systems, South San Francisco, CA) for 2 image acquisitions. Electrocardiographic triggering of scans ensured that images were obtained during diastole. Serial, contiguous, 3-mm–thick transverse images were obtained (during breath-holding), commencing at the root of the aorta cephalad to the coronary sinuses and proceeding caudally through the entire coronary tree (Aquarius software; TeraRecon, San Mateo, CA). Data were quantitatively scored in a blinded manner by a registered radiology technologist. The total CAC burden was based on the area of calcification, the average density, and the number of plaques. Scores were calculated using the method described by Agatston et al (15). Total CAC was quantified by summing across the 4 major epicardial arteries, resulting in a raw CAC [rCAC] score. These scores were compared with rCAC scores from a previously established population of (age- and sex- matched) individuals whose data were preprogrammed as the control standard, yielding the following EBCT percentile categories (EBCT-PCs): <25th, 25th–50th, 50th–75th, 75th–90th, or >90th (16). EBCT-PCs may be preferable to rCAC scores, because they are easier to interpret.

Statistical analysis.

Descriptive analyses.

For continuous variables (e.g., age), the means and standard deviations were computed, stratified by group (SLE versus control), race, EBCT-PC (e.g., >75th percentile versus other EBCT-PCs), and rCAC score (e.g., >0 versus 0). For discrete variables, (e.g., smoking), the stratified frequencies and percentages were computed.

Inferential analysis.

Spearman's correlation coefficients for the rCAC score and all continuous variables of interest were computed, overall and stratified by group. Additionally, correlation coefficients were computed for EBCT-PCs above or below the candidate values of 75% and 90%. This exploratory component was used to assess agreement of analyses involving rCAC scores compared with those involving the more traditional EBCT-PCs. Associations between EBCT-PC classification and categorical explanatory variables were assessed by chi-square analyses. Variables showing significant association with rCAC score (>0) status were included as candidate explanatory variables in a stepwise multiple logistic regression with CAC status as the outcome. Clinically plausible models were explored to identify variables predictive of CAC status. All statistical analyses were produced using SAS software (SAS Institute, Cary, NC), with a Type I error rate of 0.05 for 2-sided tests of significance.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Sample demographics.

The study group comprised 152 women with SLE (mean ± SD age 43.3 ± 11.3 years, mean ± SD disease duration 11.1 ± 8.5 years). Eighty-one (53%) of the patients were African American, 58 (38%) were white, 7 (5%) were Asian, and 5 (3%) were Hispanic; the racial distribution among matched controls was similar (Table 1).

Table 1. Characteristics of patients with systemic lupus erythematosus and age- and race-matched controls*
CharacteristicPatients (n = 152)Controls (n = 142)P
  • *

    Except where indicated otherwise, values are the mean ± SD. Percentages are based on patients and controls for whom data on each variable were available. Electron beam computed tomography percentile category (EBCT-PC) denotes a computer-generated percentile rank of raw calcium scores of individuals compared with those of a previously established population of age- and sex-matched (not race-matched) controls. CAC = coronary artery calcification; NA = not applicable; ND = not done; SDI = Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index; SLEDAI = Systemic Lupus Erythematosus Disease Activity Index; HRT = hormone replacement therapy; GFR = glomerular filtration rate; CRP = C-reactive protein.

  • Calculated for race distribution within each population, not for individual races.

  • History of at least 1 event.

  • §

    Calculated in a subset of 125 patients and 103 controls.

Age, years43.3 ± 11.343.6 ± 10.60.81
Race distribution, no. (%)  0.81
 African American81 (53.3)75 (52.8) 
 White58 (38.2)57 (40.1) 
 Asian7 (4.6)6 (4.2) 
 Hispanic5 (3.3)2 (1.4) 
 Other1 (0.7)2 (1.4) 
Raw CAC (Agatston) score87.9 ± 385.99.6 ± 49.00.02
 Positive CAC, no. (%)45 (29.6)23 (16.2)0.009
 EBCT-PC >75%, no. (%)38 (25.0)16 (11.3)0.003
 EBCT-PC >90%, no. (%)24 (15.8)10 (7.0)0.021
Disease duration, years11.1 ± 8.5NAND
SDI, mean (median)2 (2)NAND
SLEDAI, mean (median)5 (4)NAND
Vascular events, no. (%) of total events40 (26.3)7 (4.9)<0.0001
 Myocardial infarction4 (8.3)2 (25.0)0.45
 Deep vein thrombosis24 (50.0)0 (0.0)<0.0001
 Pulmonary embolism2 (4.2)2 (25.0)0.95
 Stroke18 (37.5)4 (50.0)0.0030
HRT user, no. (%)8 (5.3)3 (2.2)0.16
Oral contraceptive user, no. (%)8 (5.3)16 (11.3)0.07
Body mass index, kg/m228.5 ± 7.029.6 ± 7.20.23
Hypertension, no. (%)79 (52.0)33 (23.2)<0.0001
Systolic blood pressure, mm Hg123.1 ± 15.4122.5 ± 17.10.76
Diastolic blood pressure, mm Hg76.9 ± 9.975.7 ± 10.60.31
Homocysteine, μmoles/liter11.9 ± 4.59.8 ± 2.5<0.0001
Total cholesterol, mg/dl189.7 ± 47.9198.2 ± 41.40.11
Very low-density lipoproteins, mg/dl24.9 ± 31.316.0 ± 11.90.0017
Low-density lipoproteins, mg/dl107.5 ± 35.4121.8 ± 35.90.0007
High-density lipoproteins, mg/dl56.1 ± 17.660.4 ± 14.40.03
Triglycerides, mg/dl124.4 ± 185.786.5 ± 56.30.02
GFR, ml/minute88.7 ± 32.694.2 ± 21.10.09
Erythrocyte sedimentation rate, mm/hour32.2 ± 27.8NAND
High-sensitivity CRP, mg/liter8.5 ± 22.45.6 ± 12.00.17
Double-stranded DNA, IU/ml257.5 ± 740.4NDND
CD154, ng/ml§7.2 ± 4.49.7 ± 4.4<0.0001

CAC.

Direct associations between the rCAC score and EBCT-PCs were evident for both the SLE and control groups; however, the magnitude of the association was greater for patients with SLE. The percentages of rCAC scores >0 in patients with SLE and controls were 29.6% and 16.2%, respectively (P = 0.009). Twenty-four patients (15.8%) had rCAC scores above the 90th EBCT-PC, compared with 10 controls (7.0%) (P = 0.02), and 38 patients (25.0%) had scores above the 75th EBCT-PC, compared with 16 controls (11.3%) (P = 0.003) (Table 1). The rCAC scores for patients and controls in the age categories 26–35 years, 36–45 years, 46–55 years, and 56–65 years (comprising 87.8% of the study subjects) are shown in Figure 1. Participants with a rCAC score >0 were older than those with a rCAC score of 0, among both patients (51.0 versus 40.1 years; P < 0.0001) and controls (51.6 versus 42.0 years; P = 0.002) (Tables 2 and 3, respectively). In each of the 4 age categories, patients had a greater CAC burden than did controls.

thumbnail image

Figure 1. Coronary artery calcification in patients with systemic lupus erythematosus (SLE) and controls, according to age group (years). Values are the mean and SD. The numbers above each bar represent the number of individuals in each category. rCAC = raw coronary artery calcification.

Download figure to PowerPoint

Table 2. Characteristics of SLE patients with CAC and those without CAC*
VariableCAC positive (n = 45)CAC negative (n = 107)P, unadjustedP, age-adjusted
  • *

    Except where indicated otherwise, values are the mean ± SD. Percentages are based on patients and controls for whom data on each variable were available. CVD = cardiovascular disease; COX-2 = cyclooxygenase 2 (see Table 1 for other definitions).

  • P values calculated for race distribution within each population, not for individual races.

  • Calculated in a subset of 125 patients (37 CAC positive, 88 CAC negative).

  • §

    P values calculated for distribution of risk factors between groups.

  • Defined as the presence of anticardiolipin antibodies or anti-β2 glycoprotein I antibodies in titers >4 SD above normal values.

  • #

    Average daily dose multiplied by the number of years receiving the drug.

Age, years51.0 ± 9.640.1 ± 10.4<0.0001NA
Race distribution, no. (%)  0.640.12
 African American26 (57.8)55 (51.4)  
 White13 (28.9)45 (42.1)  
 Asian3 (6.7)4 (3.7)  
 Hispanic2 (4.4)3 (2.8)  
 Other1 (2.2)0 (0.0)  
Disease duration, years15.7 ± 9.69.1 ± 7.2<0.00010.0050
SDI, mean (median)2 (2)2 (1)0.510.75
SLEDAI, mean (median)4 (4)5 (2)0.47>0.99
Vascular events, no. (%)    
 Myocardial infarction3 (6.7)1 (0.9)0.04520.37
 Deep vein thrombosis8 (18.2)16 (15.2)0.660.90
 Pulmonary embolism0 (0.0)2 (1.9)0.350.91
 Stroke5 (11.4)13 (12.5)0.850.52
Body mass index, kg/m228.9 ± 7.628.2 ± 6.70.700.99
Hypertension, no. (%)30 (66.7)49 (46.2)0.02140.10
Systolic blood pressure, mm Hg129.8 ± 16.3120.3 ± 14.30.00110.20
Diastolic blood pressure, mm Hg76.5 ± 9.277.0 ± 10.20.740.76
Homocysteine, μmoles/liter14.0 ± 4.811.0 ± 4.10.0010.002
Total cholesterol, mg/dl199.5 ± 47.6185.5 ± 47.60.110.57
Low-density lipoproteins, mg/dl111.1 ± 32.3105.9 ± 36.40.400.75
Very low-density lipoproteins, mg/dl32.2 ± 47.021.8 ± 20.90.130.16
High-density lipoproteins, mg/dl56.5 ± 18.456.0 ± 17.30.870.20
Glomerular filtration rate, ml/minute75.4 ± 35.194.3 ± 29.90.00150.043
High-sensitivity CRP, mg/liter6.5 ± 11.19.4 ± 25.80.490.57
Double-stranded DNA, IU/ml294.5 ± 942.4242.7 ± 648.90.730.49
Soluble CD154, ng/ml8.2 ± 4.96.8 ± 4.10.120.22
Postmenopausal, no. (%)29 (64.4)33 (30.8)0.00020.53
HRT user, no. (%)3 (6.8)5 (4.7)0.600.69
Oral contraceptive user, no. (%)0 (0.0)8 (8.0)0.060.88
No. of CVD risk factors, no. (%)§  0.0030.07
 02 (4.8)30 (30.9)  
 116 (38.1)23 (23.7)  
 27 (16.7)31 (32.0)  
 312 (28.6)10 (10.3)  
 44 (9.5)3 (3.1)  
 51 (2.3)0 (0.0)  
Current-smoker status, no. (%)12 (27.3)19 (17.9)0.200.072
Pack-years of smoking10.3 ± 16.24.1 ± 9.80.01560.62
Antiphospholipid antibody positive, no. (%)16 (35.6)45 (43.7)0.360.68
History of renal disease, no. (%)17 (37.8)37 (34.6)0.710.09
History of proteinuria, no. (%)35 (79.5)72 (67.3)0.140.81
History of nephritis, no. (%)3 (6.7)21 (19.6)0.0570.20
Cumulative dose of prednisone, mg#67.0 ± 97.052.6 ± 90.20.420.32
Medications, no. (%)    
 Hydroxychloroquine, ever use36 (80.0)88 (88.2)0.740.71
 Azathioprine, ever use9 (20.0)26 (24.3)0.570.85
 Mycophenolate mofetil, ever use15 (27.3)15 (14.0)0.610.46
 Methotrexate, ever use5 (11.1)12 (11.2)0.970.98
 Cyclophosphamide, ever use3 (6.7)20 (18.7)0.070.42
 Beta-blocker, current use17 (37.8)12 (11.2)0.00030.002
 Statin, current use8 (17.8)11 (10.3)0.210.27
 COX-2 inhibitors, current use5 (11.6)17 (16.3)0.460.06
Table 3. Characteristics of controls with CAC and those without CAC*
VariableCAC positive (n = 23)CAC negative (n = 117)P, unadjustedP, age-adjusted
  • *

    Except where indicated otherwise, values are the mean ± SD. Percentages are based on patients and controls for whom data on each variable were available. CVD = cardiovascular disease; COX-2 = cyclooxygenase 2 (see Table 1 for other definitions).

  • P values calculated for race distribution within each population, not for individual races.

  • P values calculated for distribution of risk factors between groups.

  • §

    Defined as the presence of anticardiolipin antibodies or anti-β2 glycoprotein I antibodies in titers >4 SD above normal values.

  • Controls had no overt inflammatory disease and were not receiving prednisone, hydroxychloroquine, azathioprine, mycophenolate mofetil, methotrexate, or cyclophosphamide.

Age, years51.6 ± 10.142.0 ± 10.00.002NA
Race distribution, no. (%)  >0.99>0.99
African American14 (60.9)60 (51.3)  
White9 (39.1)47 (40.2)  
Asian06 (5.1)  
Hispanic02 (1.7)  
Other02 (1.7)  
Vascular events, no. (%)    
 Myocardial infarction2 (8.7)0 (0.0)0.00130.88
 Deep vein thrombosis0 (0.0)0 (0.0)NANA
 Pulmonary embolism0 (0.0)2 (1.7)0.540.90
 Stroke1 (4.5)3 (2.6)0.620.92
Body mass index, kg/m231.9 ± 6.428.8 ± 6.70.0450.0339
Hypertension, no. (%)11 (47.8)21 (17.9)0.00180.0517
Systolic blood pressure, mm Hg127.5 ± 19.4121.6 ± 16.70.140.77
Diastolic blood pressure, mm Hg79.4 ± 13.374.9 ± 10.00.070.30
Homocysteine, μmoles/liter11.2 ± 3.09.7 ± 2.40.01130.22
Total cholesterol, mg/dl224.3 ± 43.0193.4 ± 39.50.00190.0138
Very low-density lipoproteins, mg/dl22.0 ± 16.315.0 ± 10.60.020.03
Low-density lipoproteins, mg/dl139.9 ± 32.7118.6 ± 35.80.01220.0334
High-density lipoproteins, mg/dl62.5 ± 18.459.8 ± 13.60.400.61
Glomerular filtration rate, ml/minute90.1 ± 22.094.7 ± 20.90.340.25
High-sensitivity CRP, mg/liter4.8 ± 5.95.6 ± 13.00.770.53
Soluble CD154, ng/ml10.9 ± 3.59.5 ± 4.60.230.36
Postmenopausal, no. (%)8 (36.4)23 (19.8)0.090.39
HRT user, no. (%)1 (4.8)2 (1.7)0.390.85
Oral contraceptive user, no. (%)0 (0.0)16 (13.9)0.070.89
No. of CVD risk factors, no. (%)  0.020.46
 04 (19.1)65 (56.5)  
 17 (33.3)26 (22.6)  
 25 (23.8)16 (13.9)  
 35 (23.8)8 (7.0)  
 40 (0.0)0 (0.0)  
 50 (0.0)0 (0.0)  
Current-smoker status, no. (%)6 (28.6)17 (14.5)0.120.56
Pack-years of smoking9.9 ± 14.84.1 ± 10.80.060.35
Antiphospholipid antibody positive, no. (%)§5 (22.7)26 (22.4)0.970.91
History of renal disease, no. (%)1 (4.3)0 (0.0)0.870.90
History of proteinuria, no. (%)0 (0.0)1 (0.9)0.910.91
Medications used at time of study, no. (%)    
 Beta-blockers3 (13.6)1 (0.9)0.01380.0149
 Statins4 (18.2)6 (5.1)0.04160.32
 COX-2 inhibitors0 (0.0)2 (1.9)0.910.91

Predictors of CAC among patients with SLE and controls.

Analysis of SLE patient data (Table 2) revealed that rCAC scores >0 were significantly associated with the number of ASCVD risk factors, disease duration, the homocysteine concentration, and (inversely) the GFR. Among controls, rCAC scores >0 were significantly associated with the levels of total cholesterol and low-density lipoprotein, and with the body mass index (Table 3).

Regression analyses.

Due to the expected correlation among candidate explanatory variables, ordinal logistic regression with rCAC score as the outcome variable was performed on various pairings with the explanatory variables ASCVD risk, smoking status, the GFR, history of hypertension, disease duration, age, and homocysteine. For each model, a test of second-order interaction was performed; if the results were not significant, the analysis was repeated without the interaction term.

In single risk variable logistic regression models, significant determinants of a positive rCAC score (other than age) for both patients with SLE and controls were the number of ASCVD risk factors (for SLE, P = 0.003 [age-adjusted P = 0.07]; for controls, P = 0.02 [age-adjusted P = 0.46]), postmenopausal status (for SLE, P = 0.0002 [age adjusted P = 0.53]; for controls, P = 0.09 [age-adjusted P = 0.39]), and pack-years of smoking (for SLE, P = 0.02 [age-adjusted P = 0.62]; for controls, P = 0.06 [age-adjusted P = 0.35]). Among patients, the GFR (P = 0.0015 [age-adjusted P = 0.043]), the homocysteine concentration (P = 0.001 [age-adjusted P = 0.002]), and SLE disease duration (P < 0.0001 [age-adjusted P = 0.005]) were all significantly associated with a positive rCAC score.

Application of stepwise multivariate logistic regression to the multiple candidate explanatory variables identified age (P < 0.0001, r = 0.43), homocysteine concentration (P = 0.002, r = 0.24), and disease duration (P = 0.01, r = 0.19) as significant predictors of a rCAC score >0. For this model, R2 = 0.28. In multivariate analysis among controls, age (P = 0.0002, r = 0.38) and cholesterol level (P = 0.006, r = 0.25) were significant predictors of a rCAC score >0.

Soluble CD154.

Paradoxically, sCD154 concentrations in the 125 patients with SLE for whom such data were available were lower than those in their matched controls (P < 0.0001). Associations between sCD154 concentrations and rCAC scores for patients and controls were not statistically significant.

Homocysteine and GFR.

To reduce the effects of skew in their respective distributions, homocysteine concentrations >13 μmoles/liter and GFRs in the lowest quartile (i.e., <71 ml/minute) were classified as abnormal. A nominal logistic regression of CAC status (positive versus negative) using these 2 categorical variables produced a significant odds ratio of 3.6 (P = 0.0002) for homocysteine concentrations >13 μmoles/liter, and a nonsignificant odds ratio of 1.8 (P = 0.18) for a GFR <71 ml/minute. Results of a test for second-order interaction between these 2 factors were not significant (P = 0.79), indicating that the relationship between hyperhomocysteinemia and CAC positivity is not dependent on a reduced GFR.

Medications.

Patients with SLE who were receiving beta-blockers were more likely to have positive rCAC scores than were those who were not receiving these drugs (P = 0.0079). In contrast, prednisone, hydroxychloroquine, azathioprine, mycophenolate mofetil, methotrexate, cyclophosphamide, statins, cyclooxygenase 2 inhibitors, angiotensin-converting enzyme inhibitors, aspirin, hormone replacement therapy, and oral contraceptives were not associated with positivity for CAC or an rCAC score >0 in patients with SLE (Table 2).

The MTHFR C677T polymorphism, homocysteine, and CAC.

Mean rCAC scores did not differ significantly between MTHFR C677T genotypes (TT/CT versus CC) in either patients with SLE (P = 0.49) or controls (P = 0.33). Furthermore, no significant association between the MTHFR C677T genotype and the homocysteine concentration was observed. When analysis was restricted to the younger half of participants in each group, or when African Americans and whites were considered separately, the lack of association between genotype and phenotype (i.e., CAC or homocysteine) persisted.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

We have shown that female patients with SLE have a statistically significantly higher incidence of CAC, as quantified by EBCT, compared with age- and race-matched female controls. Although many traditional risk factors were significantly associated with CAC, in a multivariate analysis only age, hyperhomocysteinemia, and disease duration were associated with an elevated rCAC score in patients.

Hyperhomocysteinemia has been observed in other SLE cohorts and has been identified as a risk factor for atherothrombotic events in SLE (17, 18). This study is the first to identify hyperhomocysteinemia as a statistically significant risk marker for CAC, as assessed by EBCT, although a nonsignificant trend in another SLE cohort was previously reported (19). Additionally, a nonsignificant trend was observed using carotid ultrasonography (20).

These results are consistent with a growing body of literature in which the relationship between hyperhomocysteinemia and atherothrombotic disease has been reported in the general population. An interaction between hyperhomocysteinemia and inflammation may predispose members of vulnerable populations to premature ASCVD. Because folic acid can lower homocysteine concentrations, SLE patients with hyperhomocysteinemia may be able to lower their risk of ASCVD and consequent excess morbidity and mortality by using vitamin supplements that contain folic acid.

In healthy persons and in patients with ASCVD but without overt inflammatory disease, hyperhomocysteinemia is underpinned, at least partly, by the MTHFR C677T genotype (21). Surprisingly, we observed no association between the MTHFR genotype and the homocysteine concentration in either patients with SLE or controls. Published reports that have linked the MTHFR 677TT genotype and hyperhomocysteinemia have often been biased toward older male subjects. The lack of association between the MTHFR genotype and hyperhomocysteinemia in the controls reported here may reflect their relatively young age, female sex, African American race, or other demographic feature. The lack of association between the MTHFR genotype and hyperhomocysteinemia or CAC in patients with SLE suggests that other mechanisms, not directly involving differential MTHFR activity, generate an SLE-related phenotype characterized by high homocysteine levels and premature ASCVD. Nonetheless, in our study, hyperhomocysteinemia was clearly associated with higher rCAC scores in patients with SLE and may be one of the few laboratory markers that correlate with ASCVD in such patients.

In contrast to homocysteine, the levels of hsCRP and sCD154, each of which has received attention as a predictive marker in patients who have an enhanced risk of cardiovascular events (22, 23), were not associated with CAC in our SLE patients. This is not entirely surprising, because most patients with SLE have some systemic inflammation; i.e., a serologic marker of inflammation that varies over time may not identify unique subsets of patients with SLE who are at risk for ASCVD.

More than half of our SLE population was African American. Before undertaking this study, it was unclear how race might affect the analysis of ASCVD detected by EBCT in a racially diverse population, because the software of the EBCT scanners compares each rCAC score with an age- and sex-matched, but not-race matched, control group of subjects who are predominantly white (16). The Coronary Artery Risk Development in Young Adults study (4) previously demonstrated that race does not significantly affect the prevalence of CAC in either men or women, even after adjustment for traditional risk factors. However, in another study (24), CAC was almost twice as prevalent in whites as in African Americans. In the current study, the incidence and extent of CAC were similar for African American and white participants in both the SLE and control groups, validating the use of EBCT as an effective screening tool in patients of both races.

In summary, we have shown that EBCT can be used to detect CAC, an objective indicator of the early stages of ASCVD, in a racially diverse population of patients with SLE. The only SLE-related variable significantly associated with CAC was disease duration. Although the incidence and extent of CAC are known to increase with age, this alone did not explain the CAC variances in our model. However, a simple laboratory measure, an elevated homocysteine level, was also associated with CAC in both patients with SLE and controls. Therefore, our findings suggest that hyperhomocysteinemia is a potentially useful marker for CAC and subclinical ASCVD. If our results are confirmed by other investigators, patients with SLE who have raised homocysteine levels might benefit from EBCT screening. Clinical management of those with CAC could then incorporate strategies, perhaps including aggressive folic acid supplementation, designed to limit cardiovascular comorbidity.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Manzi S, Meilahn EN, Rairie JE, Conte CG, Medsger TA Jr, Jansen-McWilliams L, et al. Age-specific incidence rates of myocardial infarction and angina in women with systemic lupus erythematosus: comparison with the Framingham Study. Am J Epidemiol 1997; 145: 40815.
  • 2
    Esdaile JM, Abrahamowicz M, Grodzicky T, Li Y, Panaritis C, du Berger R, et al. Traditional Framingham risk factors fail to fully account for accelerated atherosclerosis in systemic lupus erythematosus. Arthritis Rheum 2001; 44: 23317.
  • 3
    Sangiorgi G, Rumberger JA, Severson A, Edwards WD, Gregoire J, Fitzpatrick LA, et al. Arterial calcification and not lumen stenosis is highly correlated with atherosclerotic plaque burden in humans: a histologic study of 723 coronary artery segments using nondecalcifying methodology. J Am Coll Cardiol 1998; 31: 12633.
  • 4
    Bild D, Folson A, Lowe L, Sidney S, Kiefe C, Westfall A, et al. Prevalence and correlates of coronary calcification in black and white young adults: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Arterioscler Thromb Vasc Biol 2001; 21: 8527.
  • 5
    Greenland P, LaBree L, Azen SP, Doherty TM, Detrano RC. Coronary artery calcium score combined with Framingham score for risk prediction in asymptomatic individuals. JAMA 2004; 291: 2105.
  • 6
    Refai TM, Al-Salem IH, Nkansa-Dwamena D, Al-Salem MH, for the European Concerted Action Project. Hyperhomocysteinaemia and risk of thrombosis in systemic lupus erythematosus patients. Clin Rheumatol 2002; 21: 45761.
  • 7
    Graham IM, Daly LE, Refsum HM, Robinson K, Brattstrom LE, Ueland PM, et al. Plasma homocysteine as a risk factor for vascular disease. JAMA 1997; 277: 177581.
  • 8
    Molloy AM, Daly S, Mills JL, Kirke PN, Whitehead AS, Ramsbottom D, et al. Thermolabile variant of 5,10-methylenetetrahydrofolate reductase associated with low red-cell folates: implications for folate intake recommendations. Lancet 1997; 349: 15913.
  • 9
    Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus [letter]. Arthritis Rheum 1997; 40: 1725.
  • 10
    Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH, and the Committee on Prognosis Studies in SLE. Derivation of the SLEDAI: a disease activity index for lupus patients. Arthritis Rheum 1992; 35: 63041.
  • 11
    Gladman D, Ginzler E, Goldsmith C, Fortin P, Liang M, Urowitz M, et al. The development and initial validation of the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index for systemic lupus erythematosus. Arthritis Rheum 1996; 39: 3639.
  • 12
    Barbaux S, Kluijtmans LA, Whitehead AS. Accurate and rapid “multiplex heteroduplexing” method for genotyping key enzymes involved in folate/homocysteine metabolism. Clin Chem 2000; 46: 90712.
  • 13
    Levey AS, Coresh J, Balk E, Kausz AT, Levin A, Steffes MW, et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003; 139: 13747.
  • 14
    Grundy SM, Pasternak R, Greenland P, Smith S, Fuster V. Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: a statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation 1999; 100: 148192.
  • 15
    Agatston AS, Janowitz WR, Hildner FJ, Zustmer NR, Viamonte M, Detrans R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990; 15: 82732.
  • 16
    Hoff JA, Chomka EV, Krainik AJ, Daviglus M, Rich S, Kondos GT. Age and gender distributions of coronary artery calcium detected by electron beam tomography in 35,246 adults. Am J Cardiol 2001; 87: 13359.
  • 17
    Petri M, Roubenoff R, Dallal GE, Nadeau MR, Selhub J, Rosenberg IH. Plasma homocysteine as a risk factor for atherothrombotic events in systemic lupus erythematosus. Lancet 1996; 348: 11204.
  • 18
    Bruce IN, Urowitz MB, Gladman DD, Ibanez D, Steiner G. Risk factors for coronary heart disease in women with systemic lupus erythematosus: the Toronto Risk Factor Study. Arthritis Rheum 2003; 48: 315967.
  • 19
    Asanuma Y, Oeser A, Shintani AK, Turner E, Olsen N, Fazio S, et al. Premature coronary-artery atherosclerosis in systemic lupus erythematosus. N Engl J Med 2003; 349: 240715.
  • 20
    Roman MJ, Shanker BA, Davis A, Lockshin MD, Sammaritano L, Simantov R, et al. Prevalence and correlates of accelerated atherosclerosis in systemic lupus erythematosus. N Engl J Med 2003; 349: 2399406.
  • 21
    Meleady R, Ueland PM, Blom H, Whitehead AS, Refsum H, Daly LE, et al. Thermolabile methylenetetrahydrofolate reductase, homocysteine, and cardiovascular disease risk: the European Concerted Action Project. Am J Clin Nutr 2003; 77: 6370.
  • 22
    Danesh J, Wheeler JG, Hirschfield GM, Eda S, Eiriksdottir G, Rumley A, et al. C-reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med 2004; 350: 138797.
  • 23
    Pai JK, Pischon T, Ma J, Manson JE, Hankinson SE, Joshipura K, et al. Inflammatory markers and the risk of coronary heart disease in men and women. N Engl J Med 2004; 351: 2599610.
  • 24
    Lee TC, O'Malley PG, Feuerstein I, Taylor AJ. The prevalence and severity of coronary artery calcification on coronary artery computed tomography in black and white subjects. J Am Coll Cardiol 2003; 41: 3944.