Utility of the Framingham risk score to predict the presence of coronary atherosclerosis in patients with rheumatoid arthritis
© Chung et al.; licensee BioMed Central Ltd. 2006
Received: 24 July 2006
Accepted: 14 December 2006
Published: 14 December 2006
The prevalence of ischemic heart disease and atherosclerosis is increased in patients with rheumatoid arthritis (RA). In the general population, but not in patients with systemic lupus erythematosus, the Framingham risk score identifies patients at increased cardiovascular risk and helps determine the need for preventive interventions. We examined the hypothesis that the Framingham score is increased and associated with coronary-artery atherosclerosis in patients with RA. The Framingham score and the 10-year cardiovascular risk were compared among 155 patients with RA (89 with early disease, 66 with long-standing disease) and 85 control subjects. The presence of coronary-artery calcification was determined by electron-beam computed tomography. The Framingham score was compared in patients with RA and control subjects, and the association between the risk score and coronary-artery calcification was examined in patients. Patients with long-standing RA had a higher Framingham score (14 [11 to 18]) (median [interquartile range]) compared to patients with early RA (11 [8 to 14]) or control subjects (12 [7 to 14], P < 0.001). This remained significant after adjustment for age and gender (P = 0.015). Seventy-six patients with RA had coronary calcification; their Framingham risk score was higher (14 [12 to 17]) than that of 79 patients without calcification (10 [5 to 14]) (P < 0.001). Furthermore, a higher Framingham score was associated with a higher calcium score (odds ratio [OR] = 1.20, 95% confidence interval [CI] 1.12 to 1.29, P < 0.001), and the association remained significant after adjustment for age and gender (OR = 1.15, 95% CI 1.02 to 1.29, P = 0.03). In conclusion, a higher Framingham risk score is independently associated with the presence of coronary calcification in patients with RA.
Patients with rheumatoid arthritis (RA) have increased mortality, largely attributable to cardiovascular disease , and there is increasing evidence from controlled clinical studies that patients with RA have more extensive extra-coronary atherosclerosis [2–5] and coronary calcification  than age- and gender-matched control subjects. The contribution of traditional and novel cardiovascular risk factors has been examined, but the mechanism for this increased cardiovascular risk in RA remains unclear.
Hypertension, smoking, and increased concentrations of C-reactive protein are more frequent in patients with RA than in control subjects [7, 8]. Furthermore, patients with RA and coronary-artery atherosclerosis are older and have a higher cumulative exposure to cigarettes and a higher erythrocyte sedimentation rate than patients without atherosclerosis, suggesting that traditional risk factors and inflammation may both play a role in the process .
The Framingham risk score is an extensively studied index to predict cardiovascular risk in the general population . It includes age, gender, smoking, blood pressure, and cholesterol concentrations and estimates the risk of coronary events by stratifying individuals into three risk categories: low (<10% risk of an event in 10 years), intermediate (10% to 20%), and high (>20%) . Although the Framingham risk score is widely used in the general population to determine prognosis and the need for intervention , the value of this risk score is less clear in younger patients, women, and patients with inflammatory diseases [11, 12]. For example, in patients with systemic lupus erythematosus (SLE), another rheumatic disease associated with increased coronary calcification, there were no differences in the Framingham risk score in patients and controls, and the majority of the SLE patients with coronary calcification had a low 10-year risk according to the Framingham calculations .
In contrast to the general population, there is no information about the relationship between the Framingham cardiovascular risk score and coronary calcification in patients with RA. Therefore, we tested the hypothesis that the Framingham cardiovascular risk score would be associated with coronary calcification in patients with RA.
Materials and methods
Eighty-nine patients with early RA who had a median (interquartile range) disease duration of 2 (1 to 3) years and mean age of 51 ± 11 (mean ± standard deviation) years, 66 patients with long-standing RA (median disease duration 19 [14 to 24] years), aged 57 ± 10 years, who met the classification criteria for RA , and 85 control subjects aged 52 ± 10 years were studied. Control subjects did not meet the classification criteria for RA or any other autoimmune disease and were frequency-matched for age, gender, and race with patients with RA (early and long-standing disease combined). These patients and control subjects are part of an ongoing cohort study to identify cardiovascular risk factors in RA . The study was approved by the Institutional Review Board of Vanderbilt University Hospital, and all subjects gave written informed consent.
Patients and control subjects were assessed according to a standardized clinical interview, physical examination, laboratory tests, and (in patients) chart review. Family history of coronary disease was defined as a first-degree relative who had had a myocardial infarction or stroke before the age of 55 in males or 65 in females. Height and weight were measured, and body mass index was calculated by dividing the weight in kilograms by the square of the height in meters. Blood pressure was recorded as the mean of two measurements obtained 5 minutes apart after subjects had rested in a supine position for 10 minutes.
After an overnight fast, blood was collected for measurement of glucose, total cholesterol, high-density lipoprotein (HDL), triglycerides, Lp(a) lipoprotein, and homocysteine concentrations and the calculation of low-density lipoprotein (LDL).
RA measurements of disease activity and other outcomes
In patients with RA, disease activity was measured with the use of the disease activity score using 28 joint counts (DAS28) . All patients and control subjects completed a Modified Health Assessment Questionnaire that provided scores for physical function, pain (0 to 10 cm, visual analog scale), and fatigue (visual analog scale) .
The composite simplified coronary prediction model built on the blood pressure and cholesterol categories proposed by the Joint National Committee on Blood Pressure and the National Cholesterol Education Program was used. This model includes age, total and HDL cholesterol, blood pressure, and smoking and was designed in the setting of a community-based cohort (Framingham) of more than 5,000 people followed for 12 years .
All subjects underwent imaging with an Imatron C-150 scanner (GE Imatron, now part of GE Healthcare, Little Chalfont, Buckinghamshire, UK) as described previously . Briefly, imaging was performed with a 100-millisecond scanning time and a single-slice thickness of 3 mm. Forty slices were obtained during a single breath-holding period starting at the aortic arch and proceeding to the level of the diaphragm. Tomographic imaging was electrocardiographically triggered at 60% of the interval between R waves. All areas of calcification within the borders of a coronary artery with a minimal attenuation of 130 Hounsfield units (HU) were computed. A calcified coronary plaque was considered present if at least three contiguous pixels were detected (voxel size, 1.03 mm3). All the scans were read by a single expert investigator (PR), unaware of the subjects' clinical status.
Calculation of calcium scores
The degree of coronary-artery calcification was calculated as described by Agatston et al. . The area of each calcified plaque is multiplied by the peak radiological attenuation inside this area expressed as a coefficient (1 = for an attenuation of 130 to 199 HU; 2 = 200 to 299 HU; 3 = 300 to 400 HU; and 4 = more than 400 HU). The sum of the scores for all coronary arterial lesions provides an overall score for each individual.
Clinical characteristics and the cardiovascular risk scores were compared in patients with early RA, patients with long-standing RA, and control subjects using Kruskal-Wallis or Wilcoxon rank sum tests for continuous variables and Fisher's exact test for categorical variables. The correlation between Framingham score and continuous clinical characteristics was performed using the Spearman correlation coefficient (ρ). A multiple linear regression model with adjustment for age and gender was used to assess the difference in the Framingham score among control subjects and patients with early and long-standing RA.
Coronary calcium scores were classified as absent (0), minimal to mild (1 to 100), moderate (101 to 400), and severe (>400), and the association between these categories and the risk score was examined by proportional logistic regression in patients with RA and control subjects. In the multivariable analyses, non-linear effect of age was evaluated using a restricted cubic spline . All analyses used a 5% two-sided significance level and were performed with STATA 9.1 and R 2.1.0 (STATACorp, College Station, TX, USA) .
Clinical characteristics of patients with rheumatoid arthritis and control subjects.
Patients with early rheumatoid arthritis (n = 89)
Patients with long-standing rheumatoid arthritis (n = 66)
Control subjects (n = 85)
Age in years
Female gender, n (percentage)
Caucasian, n (percentage)
Components of the Framingham risk scoreb
Total cholesterol (mg/dl)
Smoking, n (percentage)
High-density lipoprotein (mg/dl)
Systolic blood pressure (mm Hg)
Framingham risk score (units)
Ten-year risk of a cardiovascular event
Other cardiovascular risk factors
Family history of coronary heart disease, n (percentage)
Body mass index (kg/m2)
Low-density lipoprotein (mg/dl)
Lp(a) lipoprotein (mg/dl)
Spearman correlations between the Framingham risk score and clinical variables in patients with rheumatoid arthritis
Systolic blood pressure (mm Hg)
Diastolic blood pressure (mm Hg)
Body mass index (kg/m2)
Disease duration (years)
Pack-years of smoking
Coronary calcium score (Agatston units)
Total cholesterol (mg/dl)
HDL cholesterol (mg/dl)
LDL cholesterol (mg/dl)
Lp(a) lipoprotein (mg/dl)
Cumulative use of steroids (grams)
Cumulative use of hydroxychloroquine (grams)
Modified Health Assessment Questionnaire (0–3)
Disease activity score (DAS28)
The control group was frequency-matched for age and gender with the entire group of RA patients and thus predictably did not exactly match with either the early or late RA groups. Minor differences between the groups were dealt with by the use of multivariable adjusted analyses. To further confirm our results, we performed sensitivity analyses between all patients with long-standing RA (mean age 56.7 ± 10.5 years, 25.8% male) and 66 matched control subjects (mean age 55.0 ± 7.6 years, 25.8% male). The association between RA and Framingham score was of borderline significance (β = 1.62, 95% CI -0.01 to 3.26, P = 0.05) and remained significant after adjusting for age, gender, and homocysteine (β = 1.40, 95% CI 0.17 to 2.64, P = 0.03). Also, the association between the Framingham risk and coronary calcium scores was significant (unadjusted OR = 1.16, 95% CI 1.07 to 1.25, P < 0.001) and remained so after adjustment for age, gender, and homocysteine (OR = 1.21, 95% CI 1.08 to 1.35, P = 0.001).
The two primary findings in this report are that patients with long-standing RA had higher Framingham risk scores compared with patients with early disease and control subjects and that patients with coronary calcification also have higher Framingham scores than those without coronary calcification.
The Framingham risk score includes age, gender, total and HDL cholesterol, blood pressure, diabetes, and smoking to derive an estimated risk of developing coronary heart disease within 10 years . However, despite widespread use, the guidelines underestimate the presence of coronary calcification in certain populations, particularly women . Because RA affects predominantly women, the utility of the Framingham risk score in this patient population is of interest, particularly considering that in patients with SLE, the Framingham risk score does not differ among patients and control subjects, and is inadequate for cardiovascular risk stratification [12, 19].
Smoking and hypertension, two of the components of the Framingham risk score, differed among patients with early RA, patients with long-standing RA, and control subjects. There were more smokers among patients with early and late RA than among control subjects, and patients with late disease had higher systolic blood pressure. These differences explain in part why patients with long-standing disease had higher risk scores after adjustment for age and gender. This finding appears to be important as both risk factors can be modified, thus emphasizing the need for behavioral and therapeutic interventions and better control of common comorbidities in patients with RA.
The association between the Framingham risk score and additional cardiovascular risk factors and RA-specific characteristics such as cumulative exposure to corticosteroids and disease duration is of interest, as is the association between the Framingham risk score and the severity of coronary calcification in patients with RA. However, although the Framingham score and the 10-year risk estimates were higher in patients with coronary calcification, the majority of patients with coronary calcification were classified as being at 'low' 10-year risk and only 29 of 76 patients (38%) were at moderate or high risk according to the Framingham risk score. To improve prediction of future cardiovascular events in the general population, novel risk factors such as coronary calcium score have been studied.
A meta-analysis including asymptomatic individuals indicated that those with coronary-artery calcification above the median have an 8.7-fold increased risk of future coronary events . In addition, there are data indicating that progression in calcium scores is associated with higher risk of myocardial infarction , and coronary-artery calcification adds information to the prediction of overall mortality . Thus, efforts have been made to improve the prediction of risk by combining the Framingham risk score with coronary calcium score, and this combination appears of value to estimate risk in the general population [23, 24]. In RA patients with coronary calcification, the use of such a modified Framingham score would increase the estimation of 10-year risk from a mean of 7% (4% to 12%) to 12% (5% to 22%) and 51 of 76 patients with coronary calcification would be classified as being at moderate to high risk and therefore would be candidates for aggressive therapeutic interventions.
Patients with RA and coronary-artery atherosclerosis have higher Framingham cardiovascular risk scores even after adjustment for age and gender. However, current risk stratification classified most patients with RA as having a low 10-year risk of hard events. The use of coronary-artery calcium scores may add information to the assessment of cardiovascular risk in patients with RA and may lead to better guidelines for therapeutic interventions in this patient population. Further prospective studies are needed to examine the utility of the Framingham and the modified Framingham scores to predict hard events in patients with RA.
= confidence interval
= disease activity score using 28 joint counts
= high-density lipoprotein
= Hounsfield units
= low-density lipoprotein
= odds ratio
= rheumatoid arthritis
= systemic lupus erythematosus.
We thank Carol Brannon and Elizabeth Simpson for helping with the recruitment of patients.
This study was supported by grants (HL04012, HL67964, and GM5M01-RR00095) from the National Institutes of Health and by the Arthritis Foundation. IA is partially supported by a grant from the American College of Rheumatology.
- Pincus T, Callahan LF: Taking mortality in rheumatoid arthritis seriously–predictive markers, socioeconomic status and comorbidity. J Rheumatol. 1986, 13: 841-845.PubMedGoogle Scholar
- Park YB, Ahn CW, Choi HK, Lee SH, In BH, Lee HC, Nam CM, Lee SK: Atherosclerosis in rheumatoid arthritis: morphologic evidence obtained by carotid ultrasound. Arthritis Rheum. 2002, 46: 1714-1719. 10.1002/art.10359.View ArticlePubMedGoogle Scholar
- Kumeda Y, Inaba M, Goto H, Nagata M, Henmi Y, Furumitsu Y, Ishimura E, Inui K, Yutani Y, Miki T, et al: Increased thickness of the arterial intima-media detected by ultrasonography in patients with rheumatoid arthritis. Arthritis Rheum. 2002, 46: 1489-1497. 10.1002/art.10269.View ArticlePubMedGoogle Scholar
- Gonzalez-Juanatey C, Llorca J, Testa A, Revuelta J, Garcia-Porrua C, Gonzales-Gay MA: Increased prevalence of severe subclinical atherosclerotic findings in long-term treated rheumatoid arthritis patients without clinically evident atherosclerotic disease. Medicine. 2003, 82: 407-413. 10.1097/01.md.0000101572.76273.60.View ArticlePubMedGoogle Scholar
- Roman MJ, Moeller E, Davis A, Paget SA, Crow MK, Lockshin MD, Sammaritano L, Devereux RB, Schwartz JE, Levine DM, Salmon JE: Preclinical carotid atherosclerosis in patients with rheumatoid arthritis. Ann Intern Med. 2006, 144: 249-256.View ArticlePubMedGoogle Scholar
- Chung CP, Oeser A, Raggi P, Gebretsadik T, Shintani AK, Sokka T, Pincus T, Avalos I, Stein CM: Increased coronary-artery atherosclerosis in rheumatoid arthritis: relationship to disease duration and cardiovascular risk factors. Arthritis Rheum. 2005, 52: 3045-3053. 10.1002/art.21288.View ArticlePubMedGoogle Scholar
- Solomon DH, Curhan GC, Rimm EB, Cannuscio CC, Karlson EW: Cardiovascular risk factors in women with and without rheumatoid arthritis. Arthritis Rheum. 2004, 50: 3444-3449. 10.1002/art.20636.View ArticlePubMedGoogle Scholar
- Dessein PH, Joffe BI, Stanwix AE: Inflammation, insulin resistance, and aberrant lipid metabolism as cardiovascular risk factors in rheumatoid arthritis. J Rheumatol. 2003, 30: 1403-1405.PubMedGoogle Scholar
- Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB: Prediction of coronary heart disease using risk factor categories. Circulation. 1998, 97: 1837-1847.View ArticlePubMedGoogle Scholar
- Ford ES, Giles WH, Mokdad AH: The distribution of 10-year risk for coronary heart disease among US adults: findings from the National Health and Nutrition Examination Survey III. J Am Coll Cardiol. 2004, 43: 1791-1796. 10.1016/j.jacc.2003.11.061.View ArticlePubMedGoogle Scholar
- Mahoney LT, Burns TL, Stanford W, Thompson BH, Witt JD, Rost CA, Lauer RM: Usefulness of the Framingham risk score and body mass index to predict early coronary artery calcium in young adults (Muscatine Study). Am J Cardiol. 2001, 88: 509-515. 10.1016/S0002-9149(01)01728-3.View ArticlePubMedGoogle Scholar
- Chung CP, Oeser A, Avalos I, Raggi P, Stein CM: Cardiovascular risk scores underestimate the presence of subclinical coronary-artery atherosclerosis in women with systemic lupus erythematosus. Lupus. 2006, 15: 562-569. 10.1177/0961203306071870.View ArticlePubMedGoogle Scholar
- Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, Healey LA, Kaplan SR, Liang MH, Luthra HS, et al: The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988, 31: 315-324.View ArticlePubMedGoogle Scholar
- Prevoo ML, van't Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, van Riel PL: Modified disease activity scores that include twenty-eight-joint counts. Development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum. 1995, 38: 44-48.View ArticlePubMedGoogle Scholar
- Pincus T, Sokka T, Kautiainen H: Further development of a physical function scale on a MDHAQ [corrected] for standard care of patients with rheumatic diseases. J Rheumatol. 2005, 32: 1432-1439.PubMedGoogle Scholar
- Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Vaimonte M, Detrano R: Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990, 15: 827-832.View ArticlePubMedGoogle Scholar
- Harrell F: Regression Modeling Strategies. 2001, New York: SpringerView ArticleGoogle Scholar
- The R Project for Statistical Computing. [http://www.r-project.org/]
- Esdaile JM, Abrahamowicz M, Grodzicky T, Li Y, Panaritis C, du Berger R, Cote R, Grover SA, Fortin PR, Clarke AE, Senecal JL: Traditional Framingham risk factors fail to fully account for accelerated atherosclerosis in systemic lupus erythematosus. Arthritis Rheum. 2001, 44: 2331-2337. 10.1002/1529-0131(200110)44:10<2331::AID-ART395>3.0.CO;2-I.View ArticlePubMedGoogle Scholar
- O'Malley PG, Taylor AJ, Jackson JL, Doherty TM, Detrano RC: Prognostic value of coronary electron-beam computed tomography for coronary heart disease events in asymptomatic populations. Am J Cardiol. 2000, 85: 945-948. 10.1016/S0002-9149(99)00906-6.View ArticlePubMedGoogle Scholar
- Raggi P, Cooil B, Shaw LJ, Aboulhson J, Takasu J, Budoff M, Callister TQ: Progression of coronary calcium on serial electron beam tomographic scanning is greater in patients with future myocardial infarction. Am J Cardiol. 2003, 92: 827-829. 10.1016/S0002-9149(03)00892-0.View ArticlePubMedGoogle Scholar
- Shaw LJ, Raggi P, Schisterman E, Berman DS, Callister TQ: Prognostic value of cardiac risk factors and coronary artery calcium screening for all-cause mortality. Radiology. 2003, 228: 826-833.View ArticlePubMedGoogle Scholar
- Becker CR, Majeed A, Crispin A, Knez A, Schoepf UJ, Boekstegers P, Steinbeck G, Reiser MF: CT measurement of coronary calcium mass: impact on global cardiac risk assessment. Eur Radiol. 2005, 15: 96-101. 10.1007/s00330-004-2528-5.View ArticlePubMedGoogle Scholar
- Schisterman EF, Whitcomb BW: Coronary age as a risk factor in the modified Framingham risk score. BMC Med Imaging. 2004, 4: 1-10.1186/1471-2342-4-1.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.