http://france.elsevier.com/direct/BONSOI/

Joint Bone Spine 73 (2006) 379–387

Original article

* Corresponding author.E-mail address: maxim

1297-319X/$ - see front mdoi:10.1016/j.jbspin.2006.

Cardiovascular risk and rheumatoid arthritis:

clinical practice guidelines based on published evidence and expert opinion

Thao Pham a, Laure Gossec b, Arnaud Constantin c, Stephan Pavy d, Eric Bruckert e,Alain Cantagrel c, Bernard Combe f, René-Marc Flipo g, Philippe Goupille h, Xavier Le Loët i,Xavier Mariette j, Xavier Puéchal k, Thierry Schaeverbeke l, Jean Sibilia m, Jacques Tebib n,

Daniel Wendling o, Maxime Dougados b,*

a Service de rhumatologie, CHU de la Conception, Marseille, Franceb Service de rhumatologie B, hôpital Cochin, 27, rue du Faubourg-Saint-Jacques, 75014 Paris, France

c Service de rhumatologie, CHU Rangueil, Toulouse, Franced Service de rhumatologie A, CHU Cochin, 75014 Paris, France

e Service d’endocrinologie–métabolisme, CHU Pitié-Salpétrière, Paris, Francef Service d’immunorhumatologie, CHU Lapeyronie, Montpellier, France

g Service de rhumatologie, CHRU Roger-Salengro, Lille, Franceh Service de rhumatologie, CHU Trousseau, Tours, Francei Service de rhumatologie, CHU de Rouen, Rouen, France

j Service de rhumatologie, CHU Bicêtre, Le Kremlin-Bicêtre, Francek Service de rhumatologie, centre hospitalier, Le Mans, France

l Service de rhumatologie, CHU Pellegrin-Tripode, Bordeaux, Francem Service de rhumatologie, CHU Hautepierre, Strasbourg, Francen Service de rhumatologie, CHU Lyon-Sud, Pierre-Bénite, Franceo Service de rhumatologie, CHU Jean-Minjoz, Besançon, France

Received 5 January 2006; accepted 25 January 2006Available online 29 March 2006

This project was supported by a grant from Abbott France

Abstract

Objective: To develop clinical practice guidelines for the evaluation and management of cardiovascular risk in patients with rheumatoidarthritis (RA), using the evidence-based approach and expert opinion.

Methods: Recommendations were developed using the evidence-based approach and expert opinion: A scientific committee used a Delphiprocedure to select five questions, which formed the basis for developing the recommendations; Evidence providing answers to the five questionswas sought in the literature; Based on this evidence, recommendations were developed by a panel of experts.

Results: The recommendations were as follows: 1) In patients with RA, attention should be given to the risk of cardiovascular disease, whichis responsible for an excess burden of morbidity and mortality; 2) It must be recognized that RA may be an independent cardiovascular riskfactor. Persistent inflammation is an additional risk factor; 3) The cardiovascular risk should be evaluated, and modifiable risk factors should becorrected; 4) In patients with RA requiring glucocorticoid therapy, the need for cardiovascular risk minimization is among the reasons thatmandate the use of the minimal effective dose; 5) It should be recognized that methotrexate may protect against cardiovascular mortality inpatients with RA; 6) It should be recognized that TNFα antagonists remain contraindicated in patients with RA and severe heart failure. TNFαantagonists do not seem to worsen moderate heart failure and may protect against cardiovascular mortality; 7) AFSSAPS recommendations aboutLDL-cholesterol objectives should be followed, with active RA being counted as a cardiovascular risk factor; 8) In patients with RA, statintherapy should be considered only when cholesterol levels are elevated despite appropriate dietary treatment; 9) RA per se does not indicate

e.dougados@cch.ap-hop-paris.fr (M. Dougados).

atter © 2006 Elsevier SAS. All rights reserved.01.014

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387380

aspirin for primary prevention. When aspirin is used for secondary prevention, it should be recognized that concomitant treatment with non-steroidal antiinflammatory drugs (NSAIDs) may decrease the antiplatelet effects and increase the gastrointestinal side effects of aspirin therapy.© 2006 Elsevier SAS. All rights reserved.

Keywords: Rheumatoid arthritis; Cardiovascular risk; Everyday medical practice

1. Introduction

Cardiovascular disease is the main source of excess mortal-ity in patients with rheumatoid arthritis (RA) [1–3]. Cardiovas-cular morbidity is also increased in RA patients. Suggested ex-planations involve the inflammatory response that characterizesactive RA, the increased prevalence of specific cardiovascularrisk factors in the RA population, adverse effects of glucocor-ticoid therapy or other medications, and decreased attention tocardiovascular risk factors in patients whose joint disease is themedical priority.

The objective of this study was to evaluate the cardiovascu-lar risk associated with RA or its treatment and to help rheu-matologists prevent and manage cardiovascular disease in RApatients seen in everyday practice.

2. Methods

A multi-step approach was used to develop the recommen-dations. A scientific committee composed of hospital-basedrheumatologists (A.C., B.C., R.M.F., P.G., X.L.L., X.M.,X.P., J.S., T.S., J.T., D.W., and M.D.), a literature review taskforce (A.C., L.G., S.P., and T.P.), and a panel of experts parti-cipated in the development of the recommendations. The over-all method has been described elsewhere [4].

2.1. Selection of topics by the scientific committee members

The scientific committee selected three topics: Topic A:Cardiovascular risk and RA; Topic B: Use of methotrexate inRA [5]; Topic C: Non-pharmacological management of pa-tients with recent-onset RA [6]. Topic A is discussed in thisarticle.

2.2. Selection of questions

Using a Delphi prioritization procedure, the following fivequestions were selected: 1) Are patients with RA at increasedrisk for arterial and venous thromboembolism? Which eventsoccur at an increased rate? 2) What factors determine the riskof arterial and venous thromboembolism that is specifically re-lated to RA? 3) Should patients with RA undergo specific in-vestigations to evaluate arterial and venous thromboembolismand/or receive special follow-up (what tests and how often)? 4)Do second-line drugs influence the occurrence of cardiovascu-lar disease in patients with RA? 5) What is the best strategy forpreventing arterial and venous thromboembolism in patientswith RA?

2.3. Literature review

The strategy used to identify and select articles and to clas-sify levels of evidence has been described elsewhere [4].Searches were run using the following keywords: Arthritis,Rheumatoid, Thrombosis, Coronary Thrombosis, Thromboem-bolism, Venous Thrombosis, Atherosclerosis, Cardiovasculardiseases, Death, Mortality, Epidemiology, EpidemiologicMethods, Complications, Prevalence, Incidence, Epidemiolo-gic Factors, Inflammation, Inflammation mediators, C ReactiveProtein, Blood pressure, Hypertension, Tobacco, Smoking, Li-pids, Lipoproteins, Body mass index, Obesity, Homocysteine,Hyperhomocysteinemia, Insulin, Insulin resistance, Hyperinsu-linism, Safety Management, Health Surveys, AntirheumaticAgents, Antirheumatic Agents–Pharmacological Action, Anti-rheumatic Agents–Adverse effects, Glucocorticoids, Lefluno-mide, Methotrexate, Salazopyrine, Gold, Adalimumab, Etaner-cept, Infliximab, TNF-blocking drugs, TNF, Prevention andcontrol, Primary Prevention, and Safety Management.

All publications in English or French retrieved using thesekeywords were examined, including systematic reviews, me-taanalyses, randomized controlled trials (RCTs), non-randomized controlled trials, and observational studies. Ab-stracts of communications delivered at the French Society forRheumatology (SFR) and European League Against Rheuma-tism (EULAR) meetings between 2003 and 2005 and at theAmerican College of Rheumatology (ACR) meetings in 2003and 2004 were examined also. The results of the literature re-view were presented in June 2005. Based on this presentation,the moderators (X.M., X.P., and J.S.) determined the modal-ities of the expert panel conference.

2.4. Expert opinion

2.4.1. Selection of the expertsThe experts invited to participate in the conferences were

French rheumatologists with extensive experience in the man-agement of RA. Among them, 39% had participated in the de-velopment of the previous recommendations [4]. Of the 94 ex-perts who attended the meetings in June 2005, 49% workedfull-time in hospitals, 35% worked part-time in hospitals andpart-time in office practice, and 16% worked full-time in officepractice.

2.4.2. Presentation of the evidence from the literatureThe evidence from the literature review was presented by

the person in charge of the literature review for Topic A (TP)during three 90-min workshops, each attended by 20–25 ex-perts and moderated by a scientific committee member(X.M., X.P., or J.S.). The evidence was reported for each ques-

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387 381

tion, and the level of evidence in each article was specified. Atthe end of each workshop, the experts suggested answers to thequestions, as a preliminary step to developing the recommen-dations.

2.5. Development of the recommendations

The suggestions made by the experts at the end of the threeworkshops were reviewed by the three moderators and by theperson in charge of the literature review. When the suggestionspertaining to a given recommendation were identical at thethree workshops, they were coalesced into a single statement;when the suggestions differed one statement was developed foreach suggestion, in order to preserve the diversity of the sug-gested answers.

The final wording of each recommendation was chosen dur-ing a plenary session after an initial vote among all the expertsand members of the scientific committee and literature reviewtask force.

Fig. 1. The procedure used to de

2.6. Evaluation of the recommendations

The strength of each recommendation was graded from A toD [7]. The strength of a recommendation depends not only onthe level of evidence used to develop the recommendation butalso on the treatment effect, adverse effects, applicability of therecommendation to the target population, ease of administra-tion, and economic considerations.

The extent to which the experts agreed with the final word-ing was evaluated using a five-category Likert scale (rejectscompletely, rejects with reservation, neither accepts nor rejects,agrees with reservation, agrees completely), after a second voteheld during the plenary session (Fig. 1).

3. Results–discussion

Fig. 1 summarizes the procedure used to develop the recom-mendations. Nine recommendations about the evaluation andmanagement of cardiovascular risk factors and disease in RA

velop the recommendations.

Table 1The nine recommendations about rheumatoid arthritis and cardiovascular risk, with the degree of agreement among experts at the 2005 Rheumatology ExpertMeetings

Level of evidence (7) Strength of therecommendation (7)

Agreement amongexperts (% yes) a

1 In patients with RA, attention should be given to the riskof cardiovascular disease, which is responsible for an excess burden ofmorbidity and mortality.

2b/3 B 88.7

2 It must be recognized that RA may be an independent cardiovascular riskfactor. Persistent inflammation is an additional risk factor.

2b/3 B 79.7

3 The cardiovascular risk should be evaluated, and modifiable risk factorsshould be corrected.

4 D 96.0

4 In patients with RA requiring glucocorticoid therapy, the need forcardiovascular risk minimization is among the reasons that mandate theuse of the minimal effective dose.

4 D 95.9

5 It should be recognized that methotrexate may protect againstcardiovascular mortality in patients with RA.

2b/3 B 74.6

6 It should be recognized that TNFα antagonists remain contraindicated inpatients with RA and severe heart failure. TNFα antagonists do not seemto worsen moderate heart failure and may protect against cardiovascularmortality.

2b/3 B 75.4

7 AFSSAPS recommendations about LDL-cholesterol objectives should befollowed, with active RA being counted as a cardiovascular risk factor.

4 D 74.2

8 In patients with RA, statin therapy should be considered only whencholesterol levels are elevated despite appropriate dietary treatment.

2b/3 B 69.1

9 RA per se does not indicate aspirin for primary prevention. When aspirinis used for secondary prevention, it should be recognized that concomitanttreatment with nonsteroidal antiinflammatory drugs (NSAIDs) maydecrease the antiplatelet effects and increase the gastrointestinal sideeffects of aspirin therapy.

2b/3 C 87.5

* see Table 2.a percentage of “agrees completely” and “agrees with reservation” votes.

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387382

patients were developed. Table 1 lists these recommendationswith the level of evidence, strength of the recommendation,and extent of agreement among experts.

Recommendation 1: In patients with RA, attention shouldbe given to the risk of cardiovascular disease, which is respon-sible for an excess burden of morbidity and mortality.

A comprehensive review of studies evaluating survival inRA patients confirmed that RA is associated with excess mor-tality, the standardized mortality ratio being 1.70 [2]. Excessmortality was ascribable to cardiovascular causes in 34% of448 RA patients in the cohort studied by Symmons et al. [1]and 40% of 898 RA patients in the cohort studied by Wolfe etal. [8]. In a recent cohort of 11,633 RA patients investigated byWatson et al., the relative risk (RR) of death from cardiovas-cular disease was 1.5 (95% confidence interval [95%CI], 1.4–1.6) [9].

Cardiovascular morbidity is increased in patients with RA.[3,10–13]. For instance, after adjusting for age, sex, smokinghistory, body mass index (BMI), diabetes, and hypertension,the hazards ratio (HR) for myocardial infarction requiring hos-pital admission was 1.09 (95%CI, 0.71–1.68) and the HR forsilent myocardial infarction was 2.13 (95%CI, 1.13–4.03) inthe Rochester cohort (603 patients with RA) [13]. Even afteradjusting for cardiovascular risk factors, the risk of myocardialinfarction increased with disease duration: in the Nurses’Health Study (527 patients with RA, follow-up 2.4 million pa-tient-years), the RR was 3.10 (95%CI, 1.64–5.87) in patientswith disease durations longer than 10 years and 1.16 (95%CI,

0.52–2.59) in those with disease durations of less than 10 years[12].

Recommendation 2: It must be recognized that RA may bean independent cardiovascular risk factor. Persistent inflamma-tion is an additional risk factor.

This recommendation stems from the increase in cardiovas-cular risk associated with RA. Two studies show that RA is arisk factor for cardiovascular disease [10,12]. The first is acase-control study comparing the rates of myocardial infarctionand stroke in 236 RA patients [10] with a follow-up of 252patient-years and in 4635 individuals participating in an epide-miological study of atherosclerosis. The RR of cardiovascularevents was 3.17 (95%CI, 1.33–6.36) in the RA group. Thesecond study compared patients with and without RA in theNurses’ Health Study. The adjusted RR of myocardial infarc-tion was 2.0 (95%CI, 1.23–3.29) in the RA group and the RRfor fatal myocardial infarction was 1.82 (95%CI, 0.75–4.41).

Inflammation is a cardiovascular risk factor in its own right.In the Reykjavik study (2459 cases with myocardial infarctionand 3969 controls), the odds ratio for myocardial infarction inthe general population was 1.45 (95%CI, 1.25–1.68) in indivi-duals with C-reactive protein elevation and 1.30 (95%CI, 1.13–1.51) in those with erythrocyte sedimentation rate elevation[14]. In the Women’s Health Study cohort of 27,939 initiallyhealthy women followed up for 8 years, C-reactive protein le-vel was a strong independent predictor of myocardial infarc-tion: the RR increased from one quintile to the next, from 1.4to 1.6, 2.0, and 2.3, respectively (P < 0.001) [15], after adjust-

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387 383

ment for age, smoking history, presence of diabetes, and bloodpressure values.

Recommendation 3: The cardiovascular risk should be eval-uated, and modifiable risk factors should be corrected.

Increased rates of several traditional cardiovascular risk fac-tors have been reported in populations with RA. Solomon et al.[16] compared patients with and without RA in the Nurses’Health Study and found no significant differences for BMI,current smoking, blood pressure, presence of diabetes, physicalactivity, or total cholesterol levels. The RA group was charac-terized by a higher proportion of ex-smokers, higher C-reactiveprotein levels, and lower HDL-cholesterol levels, compared tothe control group [16]. A comparison of patients with andwithout RA in the North Glasgow MONICA Study also foundno significant differences regarding BMI, physical activity, orcurrent smoking [17]. Diastolic blood pressure was signifi-cantly higher in the RA group (83 vs. 80 mmHg, P = 0.02),total cholesterol was significantly lower (5.3 vs. 6.2 mmol/l,P < 0.0001), and markers for inflammation were significantlyhigher.

The existence of these risk factors led the experts to recom-mend an evaluation of cardiovascular risk factors in patientswith RA. Evaluation of the overall cardiovascular risk is de-scribed in the latest recommendations issued by the FrenchAgency for Healthcare Product Safety (AFSSAPS) [18] andinvolves an inventory of specific risk factors (Table 2). Simplyadding the number of risk factors in the patient is adequate ineveryday clinical practice. Alternatively, the overall cardiovas-cular risk can be estimated using risk equations (Framingham,SCORE) [19,20]. The Framingham equation estimates the 5-and 10-year risks of cardiovascular disease based on data froman epidemiological study conducted in the USA. The SCOREequation was developed from a large number of prospectivestudies from Europe and predicts the risk of cardiovasculardeath in the individual patient. It should be borne in mind thatthe risk of cardiovascular disease is higher in the USA than inFrance. Nevertheless, using risk equations may help the patientmonitor his or her progress, which may promote compliancewith risk factor interventions [18]. Inflammation is not in-

Table 2Risk factors that should be sought when evaluating the overall cardiovascular risk, arecommendations [18])

Risk factors to consider when evaluatingthe overall cardiovascular risk, according toAFSSAPS recommendations

Age (male ≥ 50 years, female ≥ 6Family history of premature coroMyocardial infarction or sudden dMyocardial infarction or sudden drelativeCurrent smoking or smoking cessTreated or untreated sustained hyTreated or untreated type 2 diabeHDL-cholesterol < 0.40 g/l (1 mmProtecting factor: HDL-cholestero

LDL-cholesterol target according tothe number of risk factors, according to Afssapsrecommendations

Number of risk factors L0 <1 <2 <> 2 <Secondary prevention orequivalent risk

<

cluded in the Framingham equation; therefore, if RA is an in-dependent cardiovascular risk factor, the equation may under-estimate the cardiovascular risk in RA patients.

Recommendation 4: In patients with RA requiring glucocor-ticoid therapy, the need for cardiovascular risk minimization isamong the reasons that mandate the use of the minimal effec-tive dose.

Two case-control studies suggest that glucocorticoid therapymay increase the risk of cardiovascular disease in patients withRA. In the MEMO (MEdicine MOnitoring unit) cohort, gluco-corticoid therapy (N = 68,781) was associated with a dose-de-pendent increase in cardiovascular events, compared to the82,202 controls [21]. In patients taking more than 7.5 mg/dayprednisone-equivalent, the RR of myocardial infarction was3.26 (95%CI, 2.60–4.09), compared to 0.98 (95%CI, 0.87–1.09) in patients taking lower doses. Similarly, the risk of heartfailure was greater in the higher-dose group (RR, 3.72; 95%CI,2.71–5.12) than in the lower-dose group (RR, 1.5; (95%CI,1.29–1.75). Among the 1165 MEMO cohort patients who hadinflammatory joint disease, those taking more than 7.5 mg/dayhad a higher risk of cardiovascular events (RR, 3.3; 95%CI,1.56–6.96) than those on lower doses (RR, 1.5; 95%CI, 0.98–2.30). Similarly, in the General Practice Research Database, theodds ratio for cardiovascular events was 1.25 (95%CI, 1.21–1.29) in the group taking glucocorticoids [22], with no evi-dence of dose-dependency. Of the 50,656 patients, 1515 hadRA; in this subgroup, the odds ratios (95%CIs) for myocardialinfarction, ischemic stroke, and heart failure were 1.2 (1.11–1.29), 0.91 (0.84–0.99), and 2.66 (2.46–2.87), respectively.

On the other hand, glucocorticoids might conceivably de-crease the cardiovascular risk by combating the inflammatoryresponse. No evidence supporting this hypothesis has beenpublished. In one study, however, the risk-increasing effect ofglucocorticoid therapy was confined to those patients who hada negative history for cardiovascular disease: in a cohort of 603RA patients matched to 603 controls and having a mean fol-low-up of 15.0 ± 9.9 years, the HR in patients who were takingglucocorticoid therapy and had no history of coronary heartdisease was 1.78 (95%CI, 1.19–2.67), when patients not taking

nd LDL-cholesterol targets according to the number of risk factors (AFSSAPS

0 years)nary heart diseaseeath before 55 years of age in the father or another male first-degree relativeeath before 65 years of age in the mother or another female first-degree

ation within the last 3 yearspertensiontes mellitusol/l) in males and femalesl ≥ 0.6 g/l (1.6 mmol/l): subtract one risk from the risk scoreDL-cholesterol target2.20 g/l (5.7 mmol/l)1.90 g/l (4.9 mmol/l)1.60 g/l (4.1 mmol/l)1.30 g/l (3.4 mmol/l)1 g/l (2.6 mmol/l)

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387384

glucocorticoids served as the reference (HR = 1) [11]. Amongpatients with a history of coronary heart disease, those with RAand glucocorticoid therapy had a HR of 2.47 (95%CI, 1.37–4.26), compared to 3.07 (95%CI, 1.91–4.95) in patients withRA but no glucocorticoid therapy.

Available epidemiological data led the experts to concludethat the long-term risk/benefit ratio of glucocorticoids was un-favorable in terms of the cardiovascular risk and that, conse-quently, the minimal effective dosage should be used in RApatients who require glucocorticoid therapy.

Recommendation 5: It should be recognized that methotrex-ate may protect against cardiovascular mortality in patientswith RA.

Three recent studies suggest that methotrexate may protectagainst cardiovascular mortality in patients with RA. In a pro-spective observational study of 1240 RA patients with a meanfollow-up of 6 years, 191 patients died, including 84 from car-diovascular causes [23]. Of the 1240 patients, 588 were takingmethotrexate (mean dose, 13 mg/week); their disease activitymarkers indicated greater severity compared to the patientsusing other second-line drugs. After adjustment for cardiovas-cular risk factors, disease activity, and demographic character-istics, the HR for mortality was 0.2 (95%CI, 0.1–0.7) in themethotrexate group and 1.0 (95%CI, 0.6–1.6) in the group tak-ing other second-line drugs. Although observational studiesmay be less reliable than RCTs, these findings suggest a pro-tective effect of methotrexate. They contradict data obtained byLandewe et al. [24] in a retrospective cohort of 633 patientswith RA. During follow-up, 73 patients died, and the RR was3.40 (P = 0.0054) in the group taking methotrexate. Theauthors suggested a deleterious effect of methotrexate mediatedby an increase in serum homocysteine levels. In most studies,however, despite evidence of homocysteine elevation, cardio-vascular morbidity was identical in the groups with and with-out folate supplementation [23,25,26].

Two studies presented at the last EULAR meeting support aprotective effect of methotrexate against cardiovascular death,in particular via improved disease control. In one of these stu-dies, the HR for cardiovascular death was compared acrosssecond-line treatments taken at the time of death in 19,580 pa-tients with RA (63,811 patient-years) [27]. After adjustment,methotrexate was associated with a HR of 0.82. In the otherstudy, 554 patients with RA were followed up for 11 years,and the incidence of cardiovascular disease was 24.8/1000 pa-tient-years [28]). The HR for cardiovascular disease after ad-justment for traditional risk factors was 0.3 in the group takingmethotrexate (P = 0.03).

Recommendation 6: It should be recognized that TNFα an-tagonists remain contraindicated in patients with RA and se-vere heart failure. TNFα antagonists do not seem to worsenmoderate heart failure and may protect against cardiovascularmortality.

TNFα antagonists are considered to be contraindicated inpatients with severe heart failure, based on findings from threetherapeutic trials of TNFα antagonists for the treatment of con-gestive heart failure. All three studies were discontinued pre-

maturely because of adverse events or lack of efficacy [29–31].It should be pointed out that most of the adverse events oc-curred with dosages greater than those used to treat RA. In2003, the Food and Drug Administration reviewed RCTs ofthe three TNFα antagonists (> 7300 patients with RA). Only18 new cases of congestive heart failure were reported, includ-ing seven in patients taking TNFα antagonists (< 0.2%) and 11in patients taking the placebo (0.7–2.1%) [32]. In addition,Wolfe and Michaud [33] compared heart failure rates in a co-hort of 13,171 RA patients including 45% treated with TNFαantagonists (etanercept or infliximab) and in 2568 patients withosteoarthritis, over a 2-year period. After adjustment for demo-graphic characteristics, the risk of heart failure was 3.9% (95%CI, 3.4–4.3) in the RA cohort and 2.3% (95%CI, 1.6–3.3] inthe control group, confirming the increased risk in RA. AmongRA patients, 2.8% of those taking TNFα antagonists experi-enced heart failure, compared to 3.9% of the other RA patients(P = 0.03).

Preliminary data suggest that TNFα antagonists may havebeneficial effects on cardiovascular risk factors. In a study byPopa et al. [34] comparing 33 RA patients given adalimumabfor 15 days and 13 RA patients given a placebo, HDL-choles-terol levels at the end of the treatment period were significantlyhigher in the adalimumab group. At the 2005 EULAR meeting,Ferrante et al. [35] presented a study of intima-media thicknessin two groups of 40 RA patients matched on age, age at diseaseonset, disease activity score (DAS), and serum lipid profile.One group was given conventional treatment (methotrexateand glucocorticoid) and the other was also given TNFα antago-nist therapy. After 2 years, a significant decrease in intima-media thickness was found in the TNFα antagonist group butnot in the conventional treatment group.

Finally, TNFα antagonists may protect against cardiovascu-lar events. A Swedish registry of RA patients was used to eval-uate the risk of experiencing a first coronary event in 412 pa-tients taking TNFα antagonist therapy (follow-up 667 patient-years) and in 580 patients not treated with TNFα antagonists(follow-up 2303 patient-years) [36]. The standardized inci-dence ratio (SIR) in the TNFα antagonist group was 0.6(95%CI, 0.3–1.1) versus 1.0 in the reference group.

Recommendation 7: AFSSAPS recommendations aboutLDL-cholesterol objectives should be followed, with activeRA being counted as a cardiovascular risk factor.

According to AFFSAPS recommendations, the serum levelof LDL-cholesterol (the atherogenic fraction of cholesterol)governs treatment decisions in patients with dyslipidemia: re-ducing LDL-cholesterol levels is clearly the goal. All patientsat high risk for cardiovascular events and all patients withLDL-cholesterol levels > 1.60 g/l (4.1 mmol/l) should receivedietary treatment. The target LDL-cholesterol level depends onthe number of risk factors, as shown in Table 2. These targetlevels were defined as an aid to treatment decisions, via a con-sensus-building procedure (based on expert opinion and cur-rent European and international recommendations) [18,37].They were not determined experimentally through interven-tional studies or cost-benefit studies.

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387 385

Available epidemiological data led the experts to considerthat RA is a cardiovascular risk factor in its own right thatshould be counted alongside traditional cardiovascular risk fac-tors.

Recommendation 8: In patients with RA, statin therapyshould be considered only when cholesterol levels are elevateddespite appropriate dietary treatment.

Recent studies suggest that statins may exert antiinflamma-tory effects. In the PROVE IT study (Pravastatin or Atorvas-tatin Evaluation and Infection Therapy Trial), 3745 patientswith acute coronary syndrome were included to compare theefficacy of atorvastatin 80 mg/d and pravastatin 40 mg/d[38]. In both treatment groups, ultrasensitive C-reactive proteinlevels declined from day 30 through the end of the study(36 months). Another study in RA patients (Trial of Atorvas-tatin in RA, TARA) compared atorvastatin 40 mg/d to a place-bo [39] in 116 patients with a mean DAS28 of 5.8 and a meanC-reactive protein level of 12.4 mg/l. After 6 months of treat-ment, the mean DAS28 change was –0.50 (95%CI, –0.75; –0.25) in the atorvastatin group and 0.03 (95%CI, –0.23; 0.28)in the placebo group (P = 0.004). In addition, ultrasensitive C-reactive protein levels showed larger decreases in the atorvas-tatin group than in the placebo group. However, the decrease inultrasensitive C-reactive protein levels (which do not reflectinflammation as seen in rheumatology) and the slight DAS28improvement documented in TARA are not sufficient to sup-port statin therapy in RA patients outside conventional indica-tions for lipid-lowering therapy with target LDL-cholesterolvalues adapted to the number of risk factors and with activeRA counted as an independent risk factor.

Recommendation 9: RA per se does not indicate aspirin forprimary prevention. When aspirin is used for secondary pre-vention, it should be recognized that concomitant treatmentwith nonsteroidal antiinflammatory drugs (NSAIDs) may de-crease the antiplatelet effects and increase the gastrointestinalside effects of aspirin therapy.

Although epidemiological studies have documented an in-crease in the risk of cardiovascular disease in patients withRA, the magnitude of this increase is unclear. National andinternational cardiology societies recommend low-dose aspirinfor primary prevention when the individual risk of coronaryheart disease is greater than 0.7–1.5% per year. Two epidemio-logical studies report data that can be used to estimate the riskof coronary heart disease in populations; the results range from0.12 to 0.59%/year. Individual risk calculation relies on riskequations. The Framingham equation, from the US, predictsmorbidity from coronary heart disease, whereas the SCOREequation predicts 10-year mortality from coronary heart disease[19,20]. The experts concluded that RA per se is not sufficientreason to use aspirin for primary prevention. Aspirin for pri-mary prevention may be appropriate in RA patients at in-creased risk for cardiovascular disease.

Many studies have established that low-dose aspirin is ef-fective for the secondary prevention of cardiovascular events[40]. Nevertheless the experts pointed out that concomitant

treatment with some NSAIDs, including COX-2 inhibitors,may decrease the antiplatelet effects of aspirin [41,42].

These recommendations intended for everyday practice reston a recent literature review (updated through June 2005) andon the opinion of rheumatologists with extensive experience inthe management of RA. The recommendations seek to increaseawareness among rheumatologists that the risk of cardiovascu-lar disease requires special attention in RA patients; they alsosuggest a standardized management strategy that is suitable foreveryday practice and that encompasses both screening for andtreatment of cardiovascular risk factors, most notably regardingmedications used to treat RA (glucocorticoids, methotrexate,and TNFα antagonists).

A literature review of evidence on the cardiovascular riskassociated with inflammatory joint disease was published afterthe recommendations were developed [43]. One of its conclu-sions is that RA should be included among cardiovascular riskfactors, which is in agreement with our recommendation 2. Theauthors of the review developed a simplified risk-evaluationalgorithm and suggested three levels of risk management (stan-dard, substantial, and intensive). Clearly, there is a need forprospective studies aimed at improving the evaluation of car-diovascular risk in RA patients and the effects of treatments, inorder to provide a basis for developing risk-reduction strate-gies.

Acknowledgments

We are grateful to Catherine Mazzacco, Gérard Goldfarb,and Fabrice Michiels at Abbott France for their support; Janinede Palmas and Fabien Moll-François at Margaux Orange fortheir help; and the 75 experts who used their knowledge andtheir experience of rheumatoid arthritis to develop the recom-mendations:

Laure Artru (Le Mans); Alain Beaulieu-Camus (Sainte-Gen-eviève-des-Bois); Patrick Boumier (Amiens); Pierre Bourgeois(Paris); Philippe Breuillard (Gonesse); Pascal Brochot (Reims);Christian Cadene (Carcassonne); Thierry Cardon (La Made-leine); Jeanne Charrin (Lyon); Daniel Chenebit (Montluçon);Luzia Chevalier (Ferney-Voltaire); Gaelle Clavel-Refregiers(Amiens); Nadine Cornet-Droguet (Grenoble); Thierry Cosser-at (Amiens); Xavier Déprez (Valenciennes); Emmanuelle Der-nis (Le Mans); Brigitte Di Maio (Lyon); Christine Dosne(L’Isle-sur-la-Sorgue); Marie-Solange Doualla (Paris); Géral-dine Durand (Poitiers); Jérôme Fannius (Versailles); AmarFenni (Argenteuil); Véronique Ferrazzi (Montpellier); PatrickGaborit (Orange); Laurence Gagneux (Reims); Philippe Gau-din (Grenoble); Jean Godde (Marseille); Danièle Goldberg(Paris); Marie-Christine Goutet-Postel (Beauvais); Jean-LucGrauer (Aix-en-Provence); Laurent Grimault (Mâcon); Gene-viève Guaydier-Souquières (Caen); Isabelle Hau (Billère); JoëlHautin (Tours); Sylvain Lanot (Alençon); Frédéric Lavie (LeKremlin-Bicêtre); José Le Noach (Cholet); France Lecoqd’André (Paris); Alain Lejeune (Roubaix); Olivia Lemaire(Toulouse); Laurence Lequen (Pau); Thierry Lequerré (Rouen);Anne Lohse (Belfort); Isabelle Ludot (Toulouse); Cédric Lukas

[14

[15

[16

[17

[18

[19

[20

[21

[22

[23

[24

[25

[26

[27

[28

[29

[30

[31

[32

[33

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387386

(Montpellier); Frédéric Marin (Albi); Charles Masson (An-gers); Frédéric Maury (Béthune); Othmane Mejjad (Rouen);Isabelle Mennecier (Clermont); Jacques Morel (Montpellier);Yves Mornand (Lyon); Hélène Moyen-Mignonat (Billère);Henri Nataf (Mantes-la-Jolie); Béatrice Pallot-Prades (Saint-Etienne); Daniel Pariente (Aix-en-Provence); Marc Prudat(Colmar); Sébastien Quennesson (Chateauroux); SylvieQuéré-Schmidt (Bois-Guillaume); Denis Rau (Martigues);Saad Rouaghe (Bondy); Sid-Ahmed Rouidi (Dreux); Jean Rou-vière (Arles); Antoinette Sacchi (Mantes-la-Jolie); MartinSoubrier (Clermont-Ferrand); Huyen Tran-Lambert (Cagnes-sur-mer); Marie-Jeanne Tricard (Limoges); Eve Van derSchueren (Nice); Stéphane Vancostenoble (Lagny-sur-Marne);Valérie Vedere (Tours); Olivier Vittecoq (Rouen); Philippe Viu(Lezignan-Corbières); Laurence Voisin-Becquet (Rouen);Claude Werlé (Haguenau); Djamila Zerkak (Pari).

References

[1] Symmons DPM, Jones MA, Scott DL, Prior P. Long term mortality out-comes in patients with rheumatoid arthritis: early presenters continue todo well. J Rheumatol 1998;25:1072–7.

[2] Ward MM. Recent improvements in survival in patients with rheumatoidarthritis: better outcomes or different study designs? Arthritis Rheum2001;44:1467–9.

[3] Wolfe F, Freundlich B, Straus WL. Increase in cardiovascular and cere-brovascular disease prevalence in rheumatoid arthritis. J Rheumatol2003;30:36–40.

[4] Pham T, Gossec L, Fautrel B, Combe B, Flipo RM, Goupille P, et al.Physical examination and laboratory tests in the management of patientswith rheumatoid arthritis: development of recommendations for clinicalpractice based on published evidence and expert opinion. Joint BoneSpine 2005;72:222–8.

[5] Pavy S, Constantin A, Pham T, Gossec L, Cantagrel A, Combe B, et al.Methotrexate therapy for rheumatoid arthritis: clinical practice guidelinesbased on publised evidence and expert opinion. Joint Bone Spine 2006;73 (in press).

[6] Gossec L, Pavy S, Pham T, Constantin A, Combe B, Flipo RM, et al.Nonpharmacological treatments in early rheumatoid arthritis: clinicalpractice guidelines based on publised evidence and expert opinion. JointBone Spine 2006;73 (in press).

[7] Shekelle PG, Woolf SH, Eccles M, Grimshaw J. Clinical guidelines: de-veloping guidelines. BMJ 1999;318:593–6.

[8] Wolfe F, Mitchell DM, Sibley JT, Fries JF, Bloch DA, Williams CA.The mortality of rheumatoid arthritis. Arthritis Rheum 1994;37:481–94.

[9] Watson DJ, Rhodes T, Guess HA. All-cause mortality and vascularevents among patients with rheumatoid arthritis, osteoarthritis, or no ar-thritis in the UK General Practice Research Database. J Rheumatol 2003;30:1196–202.

[10] del Rincon ID, Williams K, Stern MP, Freeman GL, Escalante A. Highincidence of cardiovascular events in a rheumatoid arthritis cohort notexplained by traditional cardiac risk factors. Arthritis Rheum 2001;44:2737–45.

[11] Maradit-Kremers H, Nicola PJ, Crowson CS, Ballman KV, Gabriel SE.Cardiovascular death in rheumatoid arthritis: a population-based study.Arthritis Rheum 2005;52:722–32.

[12] Solomon DH, Karlson EW, Rimm EB, Cannuscio CC, Mandl LA, Man-son JE, et al. Cardiovascular morbidity and mortality in women diag-nosed with rheumatoid arthritis. Circulation 2003;107:1303–7.

[13] Maradit-Kremers H, Crowson CS, Nicola PJ, Ballman KV, Roger VL,Jacobsen SJ, et al. Increased unrecognized coronary heart disease andsudden deaths in rheumatoid arthritis: a population-based cohort study.Arthritis Rheum 2005;52:402–11.

] Danesh J, Wheeler JG, Hirschfield GM, Eda S, Eiriksdottir G, RumleyA, et al. C-reactive protein and other circulating markers of inflammationin the prediction of coronary heart disease. N Engl J Med 2004;350:1387–97.

] Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the pre-diction of first cardiovascular events. N Engl J Med 2002;347:1557–65.

] Solomon DH, Curhan GC, Rimm EB, Cannuscio CC, Karlson EW. Car-diovascular risk factors in women with and without rheumatoid arthritis.Arthritis Rheum 2004;50:3444–9.

] McEntegart A, Capell HA, Creran D, Rumley A, Woodward M,Lowe GD. Cardiovascular risk factors, including thrombotic variables,in a population with rheumatoid arthritis. Rheumatology (Oxford) 2001;40:640–4.

] Afssaps. Prise en charge du patient dyslipidémique ; mars 2005. http://agmed.sante.gouv.fr.

] Executive Summary of The Third Report of The National CholesterolEducation Program (NCEP) Expert Panel on Detection, Evaluation, AndTreatment of High Blood Cholesterol In Adults (Adult Treatment PanelIII). JAMA 2001;285:2486–97.

] Conroy RM, Pyorala K, Fitzgerald AP, Sans S, Menotti A, De Backer G,et al. Estimation of ten-year risk of fatal cardiovascular disease in Eur-ope: the SCORE project. Eur Heart J 2003;24:987–1003.

] Wei L, MacDonald TM, Walker BR. Taking glucocorticoids by prescrip-tion is associated with subsequent cardiovascular disease. Ann InternMed 2004;141:764–70.

] Souverein PC, Berard A, Van Staa TP, Cooper C, Egberts AC, Leuf-kens HG, et al. Use of oral glucocorticoids and risk of cardiovascularand cerebrovascular disease in a population based case-control study.Heart 2004;90:859–65.

] Choi HK, Hernan MA, Seeger JD, Robins JM, Wolfe F. Methotrexateand mortality in patients with rheumatoid arthritis: a prospective study.Lancet 2002;359:1173–7.

] Landewe RB, van den Borne BE, Breedveld FC, Dijkmans BA. Metho-trexate effects in patients with rheumatoid arthritis with cardiovascularcomorbidity. Lancet 2000;355:1616–7.

] Hernanz A, Plaza A, Martin-Mola E, De Miguel E. Increased plasma le-vels of homocysteine and other thiol compounds in rheumatoid arthritiswomen. Clin Biochem 1999;32:65–70.

] Jensen OK, Rasmussen C, Mollerup F, Christensen PB, Hansen H, Eke-lund S, et al. Hyperhomocysteinemia in rheumatoid arthritis: influence ofmethotrexate treatment and folic acid supplementation. J Rheumatol2002;29:1615–8.

] Michaud K, Wolfe F. Reduced mortality among RA patients treated withanti-TNF therapy and methotrexate. Ann Rheum Dis 2005;64(supp III):87.

] Welsing P, Kievit W, Fransen J, Van Riel P. The relation of disease ac-tivity and methotrexate with the risk of cardiovascular disease in patientswith rheumatoid arthritis. Ann Rheum Dis 2005;64(Suppl III):90 [ab-stract].

] Coletta AP, Clark AL, Banarjee P, Cleland JG. Clinical trials update: re-newal (renaissance and recover) and attach. Eur J Heart Fail 2002;4:559–61.

] Anker SD, Coats AJ. How to recover from renaissance? The significanceof the results of recover, renaissance, renewal and attach. Int J Cardiol2002;86:123–30.

] Chung ES, Packer M, Lo KH, Fasanmade AA, Willerson JT, Anker SD.Randomized, double-blind, placebo-controlled, pilot trial of infliximab, achimeric monoclonal antibody to tumor necrosis factor-alpha, in patientswith moderate-to-severe heart failure: results of the anti-TNF therapyagainst congestive heart failure (attach) trial. Circulation 2003;107:3133–40.

] FDA meeting March 2003: Update on the safety on new drugs for rheu-matoid arthritis. Part II: CHF, Infection and other safety issues. http://www.fda.gov/ohrms/dockets/ac/03/transcripts/3930T1.htm.

] Wolfe F, Michaud K. Heart failure in rheumatoid arthritis: rates, predic-tors, and the effect of anti-tumor necrosis factor therapy. Am J Med2004;116:305–11.

[34

[35

[36

[37

[38

[39

[40

[41

[42

[43

T. Pham et al. / Joint Bone Spine 73 (2006) 379–387 387

] Popa C, Netea MG, Radstake T, Van der Meer JW, Stalenhoef AF, vanRiel PL, et al. Influence of anti-tumour necrosis factor therapy on cardi-ovascular risk factors in patients with active rheumatoid arthritis. AnnRheum Dis 2005;64:303–5.

] Ferrante A, Parrinello G, Avellone G, Ciccia F, Bologna P, Giardina E,et al. Atherosclerosis in females patients with rheumatoid arthritis: a twoyears follow-up. Ann Rheum Dis 2005;64(Suppl III):91 [abstract].

] Jacobsson L, Turesson C, Gulfe A, Crncik M, Petersson I, Saxne T, et al.Low incidence of first cardiovascular event in rheumatoid arthritis pa-tients treated with TNF-blockers. Arthritis Rheum 2003;48(Suppl 9):S241 [abstract].

] Summary of the second report of the National Cholesterol Education Pro-gram (NCEP) Expert Panel on Detection, Evaluation, and Treatment ofHigh Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA1993;269:3015–23.

] Ridker PM, Cannon CP, Morrow D, Rifai N, Rose LM, McCabe CH,et al. C-reactive protein levels and outcomes after statin therapy. N EnglJ Med 2005;352:20–8.

] McCarey DW, McInnes IB, Madhok R, Hampson R, Scherbakov O,Ford I, et al. Trial of Atorvastatin in Rheumatoid Arthritis (TARA): dou-ble-blind, randomised placebo-controlled trial. Lancet 2004;363:2015–21.

] Collaborative meta-analysis of randomised trials of antiplatelet therapyfor prevention of death, myocardial infarction, and stroke in high riskpatients. BMJ 2002;324:71–86.

] Catella-Lawson F, Reilly MP, Kapoor SC, Cucchiara AJ, DeMarco S,Tournier B, et al. Cyclooxygenase inhibitors and the antiplatelet effectsof aspirin. N Engl J Med 2001;345:1809–17.

] Kurth T, Glynn RJ, Walker AM, Chan KA, Buring JE, Hennekens CH,et al. Inhibition of clinical benefits of aspirin on first myocardial infarc-tion by nonsteroidal antiinflammatory drugs. Circulation 2003;108:1191–5.

] Hall FC, Dalbeth N. Disease modification and cardiovascular risk reduc-tion: two sides of the same coin? Rheumatology (Oxford) 2005;44:1473–82.