E

Tt

a

KUERDC

iwcSmdtdscaimfctp[

mowtcotcficttf

ca

1d

Joint Bone Spine 78 (2011) 539–541

Available online at

www.sciencedirect.com

ditorial

he new classification criteria for rheumatoid arthritis and their impact onherapeutic decisions

r t i c l e i n f o

eywords:ndifferentiated arthritisarly arthritisheumatoid arthritisiagnosislassification criteria

Classification criteria for rheumatoid arthritis (RA) were issuedn 1987 [1]. Because they were developed in a cohort of patients

ith established disease, their usefulness for the diagnosis andlassification of recent-onset RA came under increasing criticism.tudies of early undifferentiated arthritis showed poor perfor-ance of the 1987 criteria in separating patients who subsequently

eveloped definite RA from those who had nonserious and poten-ially self-limiting arthritis or other forms of inflammatory jointisease [2,3]. In addition, the 1987 criteria set contains no exclu-ion criteria and fails to take into account a number of testsommonly used by rheumatologists for diagnostic purposes suchs acute-phase reactants and anti-citrullinated peptide antibod-es (ACPAs), although this last test is considered the most specific

arkers of RA [4]. The poor performance of the 1987 criteriaor diagnosing and classifying recent-onset RA is a major short-oming, as many studies and recent recommendations emphasizehe need for early specific treatment in patients at risk forersistent and/or erosive RA, two features used to define RA5–8].

Consequently, a revision of the 1987 criteria was in order. Twoajor international societies, the American College of Rheumatol-

gy (ACR) and the European League Against Rheumatism (EULAR),orked together to develop new classification criteria of use for

he early diagnosis of RA [9]. This recent and extremely positiveollaboration between the ACR and EULAR started with the devel-pment of recommendations for reporting disease activity andreatment responses in clinical trials of patients with RA [10]. Itulminated with the publication of the 2010 ACR/EULAR classi-cation criteria for RA [11], which was followed immediately byriteria for RA remission in clinical trials [12]. This close coopera-ion between European and American rheumatologists is expectedo considerably benefit the field of rheumatology in the near

uture.

When developing the new 2010 ACR/EULAR criteria, two majoroncerns were the early diagnosis of RA and the use of testsvailable to rheumatologists in everyday practice. The criteria

297-319X/$ – see front matter © 2011 Published by Elsevier Masson SAS on behalf of thoi:10.1016/j.jbspin.2011.10.002

development process involved three steps [11]. The first phase,conducted under the aegis of the EULAR, consisted in a pooled anal-ysis of observational cohorts of patients with early arthritis, mostof whom were from Europe or Canada [13]. The objective was toidentify the main clinical and laboratory markers that promptedmethotrexate prescription by rheumatologists, the reference stan-dard first-line treatment of RA [5,7]. Plain radiographs were notincluded in the criteria set, as the investigators considered thatradiographic changes were often absent in patients with early RA.Instead, the presence of typical radiographic erosions was con-sidered to indicate definitive RA. Thus, the new criteria apply topatients who do not have bone erosions at baseline. The secondphase of the ACR/EULAR collaboration [14] was the developmentof a consensus among experts who reviewed the results of the firstphase and the literature on autoantibodies [4]. The 12 European and12 North American experts identified and weighted factors associ-ated with developing RA, based on 50 real cases of patients withearly undifferentiated arthritis. The probability of developing RAranged across the case scenarios from low to high. The data werefed to decision-support software (1000 minds), which generatedmultiple pairwise simulations to determine, based on the experts’responses, the relative importance and weight of each domain ofinterest. After a very large number of simulations, the software sug-gested a weight for each item. Finally, in the third phase of thecollaboration, the results of the two earlier phases were used todevelop a scoring system capable of identifying patients at high riskfor subsequent persistent and/or erosive arthritis among patientswith early undifferentiated arthritis. The optimal score thresholdindicating definitive RA was determined based on both the casescenario classification by the experts and application of the scoringsystem to the data of the three cohorts having the highest rates ofexplanatory items (ESPOIR, REACH, and a Norvegian cohort). Thebest cutoff was 66/100 points. The criteria were then simplifiedto improve ease of use, and the 0–100 scale was converted to a0–10 scale. Finally, the items were validated in new patient casescenarios and in three new cohorts of patients with early arthri-tis. A total score of 6 or greater was found to indicate definiteRA.

Table 1 shows the 2010 ACR/EULAR criteria for RA classification.The score is intended to be used only in patients who have clinicalsynovitis (swelling) in at least one joint and no alternative diagnosis

better explaining the synovitis. These eligibility criteria are, in fact,exclusion criteria, since they eliminate patients with other causes ofsynovitis such as spondylarthropathy, lupus, and other connectivetissue diseases.

e Société Française de Rhumatologie.

540 Editorial / Joint Bone Spine 78 (2011) 539–541

Table 12010 American College of Rheumatology (ACR)/ European League Against Rheumatism (EULAR) for classifying rheumatoid arthritis.

A. Joint involvement1 large joint 02–10 large joints 11–3 small joints (with or without involvement of large joints) 24–10 small joints (with or without involvement of large joints) 3> 10 joints (at least 1 small joint) 5

B. Serology (at least 1 test result is needed for classification)Negative RF and negative ACPA 0Low-positive RF or low-positive ACPA 2High-positive RF or high-positive ACPA 3

C. Acute-phase reactants (at least 1 test result is needed for classification)Normal CRP and normal ESR 0Abnormal CRP or abnormal ESR 1

D. Duration of symptoms< 6 weeks 0≥ 6 weeks 1

Patients eligible for classification: at least one joint with definite clinical synovitis (swelling); no alternative diagnosis better explaining the synovitis.C ssifiesp

oofp

TD

Rie

lassification criteria: sum the scores for categories A through D. A score ≥ 6/10 claeptide antibodies.

The classification criteria are grouped into four domains, each

f which is defined in detail (Table 2): number and site (larger small joint) of involved joints, autoantibodies (rheumatoidactors and ACPA, scored as absent, low-titer, or high-titer), acute-hase response (erythrocyte sedimentation rate [ESR] or C-reactive

able 2efinition of the domains used in the 2010 ACR/EULAR criteria set.

Objective Method

Eligible patients Defines the population to which thnew criteria set can be applied

Joints “involved” Definition of the characteristics of“involved joint” domain

Small joints of the hands/feet Belongs to the “involved joint” domMedium-sized to large joints Belongs to the “involved joint” domNumber and type of involved joints Belongs to the “involved joint” dom

Duration of symptoms

Erosion typical for RA Part of the mandatory initial evaluabefore application of the criteria

Serological abnormalities Autoantibodies: ACPA and RF

Acute-phase reactants ESR and CRP

A: rheumatoid arthritis; MCP: metacarpophalangeal; DIP: distal interphalangeal; IP: intmal interphalangeal; MTP: metacarpotarsal; ACPA: anticitrullinated protein antibodies;rythrocyte sedimentation rate; CRP: C-reactive protein.

a After exclusion of other causes of an acute-phase response.

the patient as having definite RA. CRP: C-reactive protein; ACPA: anticitrullinated

protein [CRP]), and symptom duration (more or less than 6 weeks).

A score of 6/10 or greater indicates definite RA. Three other pointsdeserve attention. First, patients with typical erosions on plainradiographs are classified as having definitive RA. However, thetypical erosion remains to be defined, and the ACR and EULAR

Results

e Based on evaluation by an expert, the patient has clinicalsynovitis in at least one joint (except the first MCP or DIPjoint but including the thumb IP joint) not better explainedby an alternative diagnosis

the Any joint with swelling or tenderness to pressure(detected clinically or by US/MRI) in an eligible patient

ain Joints of the wrists, MCP, PIP, MTP 2-5, and thumb IP jointsain Shoulders, hips, knees, ankles, and elbowsain The patient is classified based on the number and highest

site category of involved jointsThe patient self-assesses the maximum time since the firstpersistent synovitis (swelling, pain, and stiffness) in a jointthat is still involved at the time of the evaluation

tion Radiographic erosions assessed by an expert as typical forRA (in an involved joint)

Recommendation regarding ACPAsNegative result: ≤ upper limit of normal (ULN, for thelaboratory and test); low-positive: > ULN but ≤ 3 × ULN;high-positive: > 3 × ULNIn the future, ACPA titers will probably be reported instandard IUs; when this occurs the definition will beupdatedRecommendation regarding FR IgM FR (in IU):Negative: ≤ upper limit of normal (ULN, for the laboratoryand test); low-positive: > ULN but ≤ 3 × ULN; high-positive:> 3 × ULNWhen the RF result is reported as positive or negative,without the titer, the patients with a positive result areclassified in the low-positive categoryCRP Normal/abnormala based on the normal range for thelaboratoryESR Normal/abnormala based on the normal range for thelaboratoryIf the results of at least one of these two tests areabnormal, the patient is classified as having abnormalacute-phase reactants

erphalangeal; US: ultrasonography; MRI: magnetic resonance imaging; PIP: prox-RF: rheumatoid factors; ULN: upper limit of normal; IU: international units; ESR:

Spine

ascpm

fatwaditpetimjataoi

ddbeucswopdictnasaot

D

n

[

[

[

[

[

[

[

Editorial / Joint Bone

re currently working on this point. Second, patients with long-tanding disease in whom available past data indicate that theriteria are met can be classified as having definite RA. Finally, someatients may fail to meet the new criteria during the first visit theneet them later on.The 2010 ACR/EULAR criteria [11] exhibit several marked dif-

erences from the 1987 criteria [1]. Two criteria in the 1987 setre missing from the 2010 set, namely, morning stiffness dura-ion, which shows little change over time; and rheumatoid nodules,hich are rarely found early in the disease. Radiographic findings

re used in a different way in the new criteria set. A symmetricalistribution of the joint involvement is absent from the new set, as

ts impact seems very limited [14,15]. Finally, important new cri-eria have been introduced. Among them, the exclusion criteria arearticularly important, and the absence of such criteria was consid-red one of the main drawbacks of the 1987 criteria set. In addition,he new set uses parameters that are evaluated by rheumatologistsn their everyday practice, such as the ACPAs, new joint imaging

ethods (ultrasonography and magnetic resonance imaging) foroint count determination (Table 1), and acute-phase reactants (ESRnd CRP). Although the new criteria were developed in adults (olderhan 18 years), they can be used in polyarticular juvenile idiopathicrthritis with onset in adolescence, which is identical to adult-nset RA. However, they cannot be used in other forms of juvenilediopathic arthritis.

It is still too early to predict the impact of these new criteria,esigned as classification criteria, on the diagnosis and treatmentecisions in everyday practice. Validation studies will have toe performed first. Recent studies indicate better sensitivity inarly RA compared to the 1987 criteria [15,16]. Specificity is stillnknown but is expected to be good, given the use of exclusionriteria. Although the 2010 criteria were developed to serve as clas-ification criteria and should be used primarily for clinical research,e believe they will also prove useful in everyday rheumatol-

gy practice. Indeed, they are very easy to use and include onlyarameters that are routinely evaluated by rheumatologists for theiagnosis and first-line treatment decisions in patients with early

nflammatory joint disease. Thus, although they are not diagnosticriteria strictly speaking, they may be of assistance to rheuma-ologists, particularly in difficult cases. Patients who meet theew criteria will be considered eligible for a disease-modifyingntirheumatic drug effective on potentially persistent and/or ero-ive arthritis, such as methotrexate. Finally, however, the clinicalcumen and confidence of the rheumatologist will predominatever the criteria when making decisions about the diagnosis andreatment.

isclosure of interest

The author declares that he has no conflicts of interest concering this article.

78 (2011) 539–541 541

References

[1] Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Associa-tion 1987 revised criteria for the classification of rheumatoid arthritis. ArthritisRheum 1988;31:315–24.

[2] van der Helm-van Mil AH, Detert J, le Cessie S, et al. Validation of a predic-tion rule for disease outcome in patients with recent-onset undifferentiatedarthritis: moving toward individualized treatment decision-making. ArthritisRheum 2008;58:2241–7.

[3] Saraux A, Berthelot JM, Chalès G, et al. Ability of the American College ofRheumatology 1987 criteria to predict rheumatoid arthritis in patients withearly arthritis and classification of these patients two years later. ArthritisRheum 2001;44:2485–91.

[4] Aggarwal R, Liao K, Nair R, et al. Anti-citrullinated peptide antibody assaysand their role in the diagnosis of rheumatoid arthritis. Arthritis Rheum2009;61:1472–83.

[5] Combe B, Landewe R, Lukas C, et al. EULAR recommendations for the manage-ment of early arthritis. Ann Rheum Dis 2007;66:34–45.

[6] Combe B. Early rheumatoid arthritis: strategies for prevention and manage-ment. Best Pract Res Clin Rheumatol 2007;21:27–42.

[7] Smolen JS, Landewé R, Breedveld FC, et al. EULAR recommendations for themanagement of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs. Ann Rheum Dis 2010;69:964–75.

[8] Lukas C, Combe B, Ravaud P, et al. Very early DMARD initiation in inflammatoryarthritis is effective in inhibition of radiographic progression. Arthritis Rheum2011;63:1804–11.

[9] Cohen S, Emery P. The American College of Rheumatology/European LeagueAgainst Rheumatism criteria for the classification of rheumatoid arthritis: agame changer. Arthritis Rheum 2010;62:2592–4.

10] Aletaha D, Landewe R, Karonitsch T, et al. Reporting disease activity inclinical trials of patients with rheumatoid arthritis: EULAR/ACR collabora-tive recommendations. Ann Rheum Dis 2008;67:1360–4 [Arthritis Rheum2008;59:1371–7].

11] Aletaha D, Neogi T, Silman AJ, et al. 2010 rheumatoid arthritis classificationcriteria: an American College of Against Rheumatism collaborative Rheuma-tology/European League initiative. Ann Rheum Dis 2010;69:1580–8 [ArthritisRheum 2010;62:2582–9].

12] Felson DT, Smolen JS, Wells G, et al. Preliminary definition of remission inrheumatoid arthritis for clinical trials. Ann Rheum Dis 2011;70:404–13 [Arthri-tis Rheum 2011;63:573–86; doi:10.1002/art.30129].

13] Funovits J, Aletaha D, Bykerk V, et al. The 2010 American College ofRheumatology/European League Against Rheumatism classification criteriafor rheumatoid arthritis: methodological report phase I. Ann Rheum Dis2010;69:1589–95.

14] Neogi T, Aletaha D, Silman AJ, et al. The 2010 American College ofRheumatology/European League Against Rheumatism classification criteriafor rheumatoid arthritis: Phase 2 methodological report. Arthritis Rheum2010;62:2582–91.

15] Fautrel B, Rincheval N, Combe B, et al. Level of Agreement of 1987 ACR and 2010ACR/EULAR Classification Criteria: An Analysis Based on the ESPOIR CohortData. Arthritis Rheum 2010;62:S731.

16] Van der Linden MPM, Knevel R, Huizinga TW, et al. Classification of rheumatoidarthritis: comparison of the 1987 American College of Rheumatology crite-ria and the 2010 American College of Rheumatology/European League AgainstRheumatism criteria. Arthritis Rheum 2011;63:37–42.

Bernard CombeDépartement de rhumatologie, université MontpellierI, CHU Lapeyonie, 34295 Montpellier cedex 5, France

E-mail address: b-combe@chu-montpellier.fr

29 September 2011Available online 9 November 2011