ORIGINAL

Early Atherosclerosis in Rheumatoid Arthritis: A case Control Study

1
Sonaullah Shah, M.D., Nitin Gupta, M.D., Feroze Shaheen, M.D., Fayaz Ahmad Sofi, M.D.,
Umar Hafiz, M.D., Rafi Ahmed Jan, M.D., Shawkat Ali Mufti, M.D., Bashir Ahmed Shah, M.D.
Department of Internal Medicine and 1Department of Radio-diagnosis, Sher-i-Kashmir Institute of Medical Sciences, Srinagar, Kashmir.

ABSTRACT
BACKGROUND: Atherosclerosis remains the major cause of death and premature disability in developed countries.
OBJECTIVES: To assess the prevalence of early (accelerated) atherosclerosis in Rheumatoid Arthritis (RA) patients in the
absence of traditional risk factors and influence of various other parameters on it.
METHODS: The study was carried at a tertiary care university hospital in northern India (Kashmir) in year 2008-2009. Thirty
nine patients in the age group of 25- 55 years with RA fulfilling the American College of Rheumatology (ACR) Modified
criteria 1987 and a purposive sample of twenty healthy volunteers that served as controls were enrolled to judge any
difference in the studied parameters. Subjects with other risk factors for atherosclerosis were excluded from the study.
Disease Activity Score (DAS-28) was used to measure the disease activity. B-mode ultrasonography was used to measure
carotid artery intima media thickness (CIMT) in both the groups.
RESULTS: RA patients presented with elevated CIMT in the age group of 41-50 years (p = 0.665) whereas volunteers had
such tendency in the 5th decade of life (p = 0.550). Duration of disease greater than 5 years also positively influenced the
development of increased CIMT in the patient group (p 0.64).Patients in the RA group had a higher erythrocyte sedimenta-
tion rate (ESR) and mean CIMT as compared to the controls(p = 0.000). RA patients had lower hemoglobin concentration
when compared to age and sex matched controls. Intergroup comparison in patients with normal and increased CIMT
showed that increased BMI, elevated triglyceride (TG) concentration and raised ESR influenced the development of CIMT
which on binary logistic regression showed that TG (p = 0.043) and BMI (p = 0.053) had influence in progression of
CIMT.
CONCLUSION: Rheumatoid Arthritis patients have definite evidence of early (accelerated) atherosclerosis due to
inflammation even in this ethnic population. BMI and serum triglycerides even in normal range have noteworthy influence in
acceleration of atherosclerosis in them. B-mode ultrasonography is simple, non-invasive, and one of the sensitive methods
to detect earlier atherosclerotic changes in them. JMS 2012;15(2):106-10

Key words: Rheumatoid arthritis, atherosclerosis, carotid intima media thickness, B-mode ultrasonography

Atherosclerosis remains the major cause of death and atherosclerosis. Among these novel risk factors, rheumatoid
premature disability in developed societies.1 It is now arthritis (RA) has been implicated to cause accelerated
considered an immuno-inflammatory process. Apart from atherosclerosis and increased cardiovascular morbidity and
traditional risk factors like smoking, hypertension, diabetes mortality.2
mellitus, dyslipidemia, several novel inflammation related Atherosclerosis is emerging as an important complica-
risk factors are currently implicated in the pathophysiology of tion of rheumatoid arthritis with coronary artery disease as
Correspondence: major cause of mortality. Patients with RA have reduced life
Dr. Fayaz Ahmad Sofi expectancy and high cardiovascular morbidity and mortality
Department of Internal Medicine, as compared to that in the general population.3-7 Indian
Sher-i-Kashmir Institute of Medical Sciences, Srinagar, Kashmir. population is also at increased risk of developing
E-mail:[email protected] atherosclerotic coronary artery disease. Indian data on this

106 Journal of Medical Sciences 2012;15(2):106-10

aspect of RA is sparse and there is no such study available in corresponded to the medial-adventitial border and the inner
this ethnic population till date. Keeping this in mind, we line corresponded to the luminal-intimal border. Distance
conducted a prospective study to assess the prevalence of between two parallel lines represented the CIMT as the
atherosclerosis in RA patients in the absence of traditional white area. Mean CIMT values (2SD) above the mean
risk factors and influence of various other factors on it. CIMT values of volunteers were taken as abnormal.9,10
Statistical analysis: Data was described as meanSD and
Methods percentages. Univariate analysis was done by Student's t-test
The study was carried at a tertiary care university and Mann-Whitney's U test. The factors with p value of
hospital in northern India (Kashmir) from 2008-2009. Thirty <0.05 on univariate analysis were subjected for multivariate
nine patients in the age group of 20- 54 years with RA analysis (Binary Logistic Regression) to evaluate best
fulfilling the American College of Rheumatology(ACR) predictors in the study group. Statistical package SPSS-16.0
was used for data analysis.
Modified criteria 1987 were taken up for the study.8 Besides;
twenty healthy volunteers were enrolled to judge any
difference in the studied parameters. Results
Smokers, males with age < 20 and 45 years, females Age of patients ranged from 20- 50 years (37.76.9) and
<20 and 55 years and patients having a BMI of > 30 were in controls 21-53 years (36.111.1). Females constituted
excluded from the study. Besides patients having a history of 71.8% of cases and 40% of controls. Gender ratio of the
coronary artery disease (CAD) or stroke, a history of CAD or respective two groups was 1:2.3 and 1.5:1 respectively.
stroke in first degree relatives in males > 55 years and females Duration of disease ranged from 1-20 years (mean 5.74.1
> 65 years, diabetes mellitus, hypertension (BP 140/90 years). 94.9% of patients were on some form of non-steroidal
mmHg), hyperlipidemia (fasting total cholesterol of >200 anti-Inflammatory drugs (NSAIDs), 48.7% on disease
mg/dL; fasting serum triglycerides > 150 mg/dL) were also modifying anti-rheumatic drugs (DMARDs), 59% on
excluded from the study.1 steroids and 7.7% on ayurvedic medicines. Rheumatoid
Formal written consent was taken from the studied factor was positive in 34(87.4%) and 35(89.7%) tested
subjects prior to recruitment in the study. Also the study was positive for C - reactive protein (CRP). There was evidence
approved by the hospital ethical committee. A detailed of X-ray erosions in X-rays of hands in 14(35.9%). Mean
history and physical examination was performed in all the CIMT was maximum in the age group of 41-50 years (p
subjects. Disease Activity Score-28 (DAS- 28) was used to 0.665), but in the controls it was maximum in the age group
measure the disease activity with a DAS-28 score of 3.2- of 51-60 years (p=0.55). Mean CIMT increased in patients
<5.1 classified as mild and a DAS-28 score of 5.1 classified after the age of 40 years and was higher in patients with
as severe disease.9 Patient was said to have diabetes mellitus if disease duration of greater than 5 years (p=0.64). Gender
it was diagnosed by the physician, patient was taking anti- difference was insignificant in cases (p=0.240) and controls
diabetic medications or had a fasting blood sugar 126 mg% (p=0.729).
or a 2 hour post 75 gram oral glucose load 200 mg% or a Haemoglobin(Hb), erythrocyte sedimentation rate
random blood glucose of 200 mg%. CAD was said to be (ESR), mean CIMT and female gender showed statistically
present if the patient had a history of angina or myocardial significant difference between study and control group
infarction or if the patient had been put on anti-ischemic (p<0.05) (Table 1).
medication by the physician. History of stroke or transient Additionally, CIMT among the subjects in the study
ischemic attack (TIA) in the patient or his or her first degree group was dichotomized into raised (2SD from mean of
relative was determined by historical evidence and previous controls) and normal (<2SD from mean of controls) groups.
records. These two groups were then subjected to intergroup
B mode ultrasonography or color Doppler was used to comparison. BMI, TG and ESR were found to be signifi-
measure CIMT in both the groups. A single expert radiolo- cantly different among the two groups.. The predicted three
gist conducted the test to exclude any individual bias in parameters were further evaluated by binary logistic regres-
interpretation. sion that showed TG (p=0.043) and BMI (p=0.053) having
The subjects were examined in supine position in the statistically significant influence in progression of CIMT
bed. Common carotid arteries on both the sides were (Table 2).
scanned along the three different planes in longitudinal axis DAS-28 scoring did not show any statistical influence
and in a cross sectional axis. CIMT was measured on a on CIMT in RA patients (p 0.707). Rheumatoid factor (p
longitudinal scan of the carotid artery at a point 10 mm =0.840), CRP status (p=0.540) and presence of radiological
proximal from beginning of the dilatation of the carotid bulb. abnormalities (p=0.503) as well did not influence mean
This scan pattern was characterized by two echogenic lines CIMT values. Patients taking NSAIDs, DMARDs and
separated by hypo-echoic and anechoic space. Outer line steroids showed no significant difference in mean CIMT in

Journal of Medical Sciences 2012;15(2):106-10 107

TABLE 1. Baseline parameters in the study and control groups TABLE 2. Correlation of CIMT across respective parameters among the study
group
Study (n=39) Control (n=20) p value
Raised CIMT Normal cIMT p value
Age(yrs) 37.76.9 36.111.1
(n=25) (n=14)
(21,50) (20, 53) 0.481
Female 28 (71.8) 0.019 Female 18 (72.0) 10 (71.4) 0.970
NSAIDS 37 (94.9) 0 (0.0) 0.000 Duration(yrs) 5.44.0 (0,15) 5.15.7(0,20) 0.814
DMARDS 19 (48.7) 0 (0.0) 0.000 NSAIDS 24 (96.0) 13 (92.9) 0.674
Steroids 23 (59.0) 0 (0.0) 0.000 DMARDS 12 (48.0) 7 (50.0) 0.906
Ayurvedic 3 (7.7) 0 (0.0) 0.000 Steroids 15 (60.0) 8 (57.1) 0.864
BMI(kg/m2) 24.62.5 23.72.8 Ayurvedic 3 (12.0) 0 (0.0) 0.183
(17.8, 28.6) (19.0, 27.9) 0.324 BMI(kg/m2) 25.12.2 23.32.8
Hb(g/dl) 11.11.8 12.51.8 (18.7, 28.6) (17.8, 26.7) 0.038
(7.3, 15) (9.1,15.5) 0.009 DAS- 28 6.61.2 6.51.2
TLC(cumm) 7.72.6 6.71.8 (3.7, 8.4) (4.1, 8.4) 0.707
(4.1,15.0) (4.4,11.0) 0.112 Hb(g/dl) 10.81.6 11.71.9
ESR(mm/h) 41.618.2 23.75.7 (7.3, 14.8) (9.2, 15.0) 0.121
(13, 92) (13, 40) 0.000 TLC/l 7.52.4 8.22.9
CRP(mg/l) 35(89.7) 0(0.0) 0.000 (4.1, 15.0) (4.2, 14.5) 0.405
RF(IU/ml) 34(87.2) 0(0.0) 0.000 ESR(mm/h) 46.218.8 33.214.3
BSF(mg/dl) 89.512.3 91.717.2 (13, 92) (16, 60) 0.030
(61, 116) (68, 146) 0.591 CRP(mg/l) 23 (92.0) 12 (85.7) 0.540
Urea(mg/dl) 35.09.0 34.69.5 RF(IU/ml) 22 (88.0) 12 (85.7) 0.840
(20, 58) (23, 51) 0.888 BSF(mg/dl) 88.710.2 91.116.3
Creatinine(mg/dl) 0.80.2 (0.6, 0.9) (61, 109) (61, 116) 0.603
(0.4, 1.3) 0.80.1 0.549 Urea(mg/dl) 34.48.1 36.411.3
TG(mg/dl) 126.117.1 126.413.3 (20, 58) (22, 55) 0.575
(90, 150) (95, 144) 0.959 Creatinine(mg/dl) 0.80.2 0.70.2
CHOL(mg/dl) 125.214.9 127.3 14.4 (0.5, 1.3) (0.4, 1.0) 0.153
(98, 149) (95, 147) 0.632 TG(mg/dl) 130.311.2 117.723.5
Mean cIMT 0.6 0.1 0.5 0.1 (110, 150) (90, 150) 0.044
(0.4, 0.9) (0.3, 0.6) 0.000 CHOL(mg/dl) 125.014.4 125.716.4
Erosions 14 (35.9) 0 (0.0) 0.000 (98, 149) (101,148) 0.897
Erosions 8 (32.0) 6 (42.9) 0.503
Note: Quantitative parameters are expressed as mean SD and the
subsequent values separated by a comma within parenthesis represent Note: Quantitative parameters are expressed as mean SD and the
greatest lower bound and least upper bound of the respective variable. subsequent values separated by a comma within parenthesis represent
Otherwise the numerals associated in parenthesis represent respective greatest lower bound and least upper bound of the respective variable.
frequency and its percentage. Otherwise the numerals associated in parenthesis represent respective
frequency and its percentage.
comparative analysis (Table 2). Results of multivariage
analysis are shown in table 3. being increasingly employed as marker of early atherosclero-
sis and vascular risk.2 The gold standard for measurement of
atherosclerosis is Magnetic Resonance Imaging, which is
TABLE 3. Results of multivariate analysis
neither freely available nor cost effective.12
Score Sig.
Mean age of our patients and females which constituted
BMI 3.7583 0.053 70% was similar to study done by Mahajan et al.13 Pattern of
ESR 2.5755 0.109 age and gender was similar to that seen in a study done in this
TG 4.106 0.043
population previously and worldwide.14,15-17 Mean duration of
illness in our patients was 5.7 years while it was 8.035.48
Discussion years in earlier studies.10,13-14 Sero-positive patients constituted
Forty percent of RA patients have extra articular 87.2% being similar to an earlier study, though in some
manifestations of which 15% are severe.11 Increasing studies it was around 35%. Moreover percentage of patients
attention is now paid to accelerated atherosclerosis in RA. on various medications was similar to other studies as well.16-17
Both traditional and inflammation related risk factors Patients above 40 years and duration of illness>5 years
contribute to atherosclerosis in them. We sought evidence of had higher CIMT which is in accordance to the studies
atherosclerosis by B-mode carotid ultrasonography in conducted earlier.18-19 Gender difference did not show any
patients of RA by measuring CIMT, which is simple, cost- significant difference in CIMT between study and control
effective and one of the sensitive methods although with group. There was no such study available for comparison. On
some limitations in focal lesions. This non-invasive tool is subgroup analysis, BMI showed significant influence on

108 Journal of Medical Sciences 2012;15(2):106-10

CIMT which was in contrast to that observed by Del Rencon varying methods.22, 26 One study even large enough took more
et al.7 Various studies have shown significant increase in men, age>60years, having severe hypertention, diabetes
CIMT in patients with high DAS.1,16,20-21 But in our study mellitus (10%) and hypercholesterolemia (13%).31
there was no such influence observed, which is also in In conclusion, rheumatoid arthritis patients have
accordance to the observation made by Carotti et al,18 definite evidence of accelerated atherosclerosis in absence of
probably because of score recorded in the first week. Effect of traditional risk factors owing to inflammation. This is
erosions on CIMT was also insignificant in contrast to study comparable to what is seen in other populations' globally.
done by Kumeda et al.22 This could be because of smaller size BMI and serum triglycerides even in normal range showed
of our study or use of different method of documentation. significant influence in acceleration of atherosclerosis in
Marker of inflammation, ESR was found to be associ- these patients. Various other factors like age>40 years, male
ated with increase in CIMT although this association could gender, longer duration of disease, BMI, elevated DAS-28
not be observed in subgroup analysis and association is score, ESR, CRP positivity and erosions showed positive
conflicting in various other studies as well.20,23-26 CRP correlation though not statistically significant. B-mode
positivity in our study had no impact on CIMT in contrast to ultrasonography is simple, non-invasive, and one of the
other studies,16,25,27-28 reason could be that high sensitivity CRP sensitive methods to detect earlier atherosclerotic changes in
was not measured in our sample which could have shown the them.
true reflection. Long half life of ESR as compared to CRP
could also explain our observation. Serum triglycerides even Conflict of interest:
in normal ranges showed positive correlation with CIMT in
our patients as was also observed by Cuomo, et al.29 This The authors declare that they have no conflicting
observation needs further study. In our study, it could interest, support or funding from any agency.
probably be because of steroids or DMARDs use or female
preponderance in whom there is lesser burden of References
atherosclerotic cardiovascular disease. This observation 1. Libby P. The pathogenesis, prevention, and treatment of
might point to the fact that even normal levels are not atherosclerosis. In: Harrison's principle of internal
protective. medicine. New York: McGraw-Hill. 17th ed. 2008; 2:
Patients on NSAIDs had higher CIMT although not 2252-8.
statistically significant (p 0.674). This has been seen in earlier 2. Doornum S, McColl G, Wicks IP. Accelerated atheroscle-
studies as well implicating that this could lead to increased rosis-an extra-articular feature of rheumatoid arthritis.
risk of cardiovascular disease, infarction and heart failure.30-31 Arthritis and Rheum 2002;46:862-73.
Kumeda, et al observed no difference on CIMT in patients on 3. Pasceri V, Yeh ETH. A tale of two diseases: Atherosclerosis
various medications.22 There was no effect of DMARDs in and rheumatoid arthritis. Circulation 1999;100:2124-6.
our study as compared to previous study,16 possibly we took 4. Cathcart ES, Spondick DH. Rheumatoid heart disease. A
all drugs as one category. Patients on ayurvedic medications study of the incidence and nature of cardiac lesion in
had higher CIMT; however, no study was available for rheumatoid arthritis. N Engl J Med 1962;266:959-64.
comparison. 5. Allebeck P. Increased mortality in rheumatoid arthritis.
Twenty four(60%) of our patients had significantly Scand J Rheumatol 1982;11:81-6
abnormal CIMT as compared to controls. Only one (5%) 6. Pincus T, Callahan IF, Sale WG. Severe functional
control had abnormal CIMT. The frequency of abnormal declines, work disability and increased mortality in 75
CIMT in our patient population was much higher than that rheumatoid arthritis patients studied over 9 years. Arthritis
observed by Grover, et al (33.4%),10 and Mahajan, et al (21%).13 and Rheum 1984;27:864-72.
This higher incidence could be explained by geographical 7. Del Rincon I, Williams's K, Stern MP, Freeman GL,
variation and our patients being mostly referred with severe O'Leary DH, Escalante A. High risk of cardiovascular
disease than those in the community. This could also be events in a rheumatoid arthritis cohort not explained by
because of higher age at the time of enrollment and greater traditional cardiac risk factors. Arthritis and Rheum 2001;
duration of disease; this observation needs further study in 44:2737-45.
our population. Tyrrel, et al in his meta-analysis observed 8. Prevoo ML, Van Hoff MA, Kuper HM. Modified disease
significantly increased CIMT in autoimmune disorders activity scores that include twenty eight joint count,
including rheumatoid arthritis patients.32 Similar observation development and validation in a prospective longitudinal
of increased prevalence and higher severity of CIMT study of patients of rheumatoid arthritis. Arthritis and
especially in bulb-ICA (Internal Carotid Artery) was made by Rheum 1995;38:44-8.
another study as well.33 9. Mannami T, Konishi M, Baba S, Nishi N, Terao A.
The studies that failed to show an increase in carotid Prevalence of asymptomatic carotid atherosclerotic lesion
atherosclerosis in RA in past were relatively small and used detected by high resolution ultrasonography and its

Journal of Medical Sciences 2012;15(2):106-10 109

 relation to cardiovascular risk factors in the general 22. Kumeda Y, Inaba M, Goto H, Nagata M, Henmi Y,
population in Japanese city- A Suita study. Stroke 1997;28: Furumitsu Y, et al. Increased thickness of the arterial intima-
518-27. media detected by ultrasonography in patients with
10. Grover S, Sinha RP, Singh U, Tewari S, Aggarwal A, Misra rheumatoid arthritis. Arthritis Rheum 2002;46:1489-97.
R. Subclinical atherosclerosis in rheumatoid arthritis in 23. Jadhav UM, Kadam NN. Impact of glycosylated hemo-
India. J Rheumatol 2006 ;33:244-7. globin on atherosclerosis - context with carotid intima-
11. Del Rincon I, Williams K, Stern MP, Freeman GL, O'Leary media thickening (IMT) APICON 2002. Free Papers
DH, Escalante A. Association between carotid atheroscle- Session Platform Presentation.
rosis and markers of inflammation in rheumatoid arthritis 24. Keser G, Aksu K, Tamsel S, Ozmen M, Kitapcioglu G,
patients and healthy subjects. Arthritis and Rheum 2003; Kabaroglu C, et al. Increased thickness of the carotid artery
48:1833-40. intima-media assessed by ultrasonography in behcet's disease.
12. Saam T, Underhill HR, Chu B, Takaya N, Cai J, Polissar Clinical and Experimental Rheumatology 2005;23:S71-6.
NL, et al. Prevalence of American Heart Association Type 25. Cao JJ, Arnold AM, Manolio TA, Polak JF, Psaty BM,
VI Carotid Atherosclerotic Lesions Identified by Magnetic Hirsch CH, et al. Association of carotid artery intima-
Resonance Imaging for Different Levels of Stenosis as media thickness, plaques, and C-reactive protein with
Measure by Duplex Ultrasound. J Am Coll Cardiol 2008; future cardiovascular disease and all-cause mortality: the
51:1014-2. Cardiovascular Health Study. Circulation 2007;116:32-8.
13. Mahajan V, Handa R, Kumar U, et al. Assessment of 26. Jonsson CW, Backman C, Johnson O, Karp K, Lundstrm
atherosclerosis by carotid intima-media thickness in E, Sundqvist KG, et al. Increased prevalence of atheroscle-
patients with rheumatoid arthritis. JAPI 2008;56:587-90. rosis in patients with medium term rheumatoid arthritis. J
14. Buhroo AM, Baba AN. Adverse effects of low dose Rheumatol 2001;28:2597-602.
methotrexate in patients with rheumatoid arthritis. IJPMR 27. Niki P, Savi M, Zari N and Duri D. Common carotid
2006;17:21-5. artery intima-media thickness, carotid atherosclerosis and
15. Coaccioli S, Capit G, Valentini M, Pinoca F, Landucci P, subtypes of ischemic cerebral disease. Med Pregl 2003;56:
Falati G, et al. Intima-media thickness of common carotid 85-91.
as cardiovascular risk factor in rheumatoid arthritis and 28. Abu-Shakra M, Polychuck I, Szendro G, Bolotin A,
metabolic disorders. Clin Ter 2007;158:505-8. Jonathan BS, Flusser D, et al. Duplex study of the carotid
16. Dessein PH, Joffe BI and Singh S. Biomarkers of endothe- and femoral arteries of patients with rheumatoid arthritis:
lial dysfunction, cardiovascular risk factors and atheroscle- a controlled study. Semin Arthritis Rheum 2005;35:18-23.
rosis in rheumatoid arthritis. Arthritis Res Ther 2005;7: 29. Cuomo G, Di Micco P, Niglio A, Montagna GL,Valentini
R634-43. G. Atherosclerosis and rheumatoid arthritis: relationships
17. Johnsen SP, Larsson H, Tarone RE, McLaughlin JK, between intima-media thickness of the common carotid
Nrgrd B, Friis S, et al. Carefully controlled study show arteries and disease activity and disability. Reumatismo
that all class of non-aspirin NSAIDs cause increased risk of 2004; 56:242-6.
myocardial infarction. Arch Intern Med 2005;165:978-8. 30. Wllberg-Jonsson S, Ohman M and Rantap-Dahlqvist
18. Carotti M, Salaffi F, Mangiacotti M, Cerioni A, S. Which factors are related to the presence of atheroscle-
Giuseppetti GM, Grassi W. Atherosclerosis in rheumatoid rosis in rheumatoid arthritis? Scand J Rheumatol 2004;
arthritis: the role of high-resolution B-mode ultrasound 33:373-9.
in the measurement of the arterial intima-media thick- 31. Bennett JS, Daugherty A, Herrington D, Greenland P,
ness. Reumatismo 2007;59:38-49. Roberts H , Taubert K. The Use of Nonsteroidal Anti-
19. Arnett FC, Edworthy SM, McShane DJ, et al. The Inflammatory Drugs (NSAIDs). Circulation 2005; 111:
American Rheumatism Association revised criteria for 1713-6.
classification of rheumatoid arthritis. Arthritis Rheum 1988; 32. Tyrrell PN, Beyene J, Feldman BM, McCrindle BW,
31:315-24. Silverman ED, Bradley TJ. Rheumatic disease and carotid
20. Sattar N, McCarey DW, Capell H, McInnes LB. Explain- intima-media thickness: a systematic review and meta-
ing how "High-Grade" systemic inflammation accelerates analysis. Arterioscler Thromb Vasc Biol 2010;30:892-3.
vascular risk in rheumatoid arthritis. Circulation 2003; 33. Kobayashi H, Giles JT, Polak JF, Blumenthal RS, Leffell
108:2957-63. MS, Szklo M, et al. Increased prevalence of carotid artery
21. Hannawi S, Haluska B, Marwick TH and Thomas R. atherosclerosis in rheumatoid arthritis is artery-specific. J
Atherosclerotic disease is increased in recent-onset Rheumatol 2010 ;37:890-3.
rheumatoid arthritis: a critical role for inflammation.
Arthritis Res Ther 2007; 9(6):R116.

110 Journal of Medical Sciences 2012;15(2):106-10