Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.


The Diagnostic Performance of the ACR/EULAR 2010 Criteria If One of Its Parameters Is Not Available. Results from the Rotterdam Early Arthritis CoHort (REACH).

Alves2,  Celina, Luime1,  Jolanda, van Zeben4,  Jendé, Huisman4,  Margriet, Weel3,  Angelique, Barendregt3,  Pieternella, Hazes1,  Johanna

Erasmus MC
Erasmus MC, Rotterdam, The Netherlands
Maasstad Ziekenhuis
Sint Franciscus Gasthuis

Background:

Recently, criteria were developed for the diagnosis of RA to improve early diagnosis of RA, devised as a predictive algorithm for persistent and/or erosive arthritis. They contain 4 domains: joints, serology, acute phase reactants (APR) and duration. However, in daily clinical practice components may not be available due to various reasons. Therefore we set out to evaluate the diagnostic value of the ACR/EULAR criteria if one of its parameters is not available.

Methods:

Eligible patients for REACH presented with synovitis or >= 2 tender joints and symptom duration less than 12 months. The outcome used was persistent disease, defined as synovitis or the need for DMARDs at 1 year. Statistical analysis was done using ROC-curves for discriminative power and sensitivity and specificity when using the proposed cut off of 6 or higher. Data were analysed for the full model and 9 models with a missing or modified parameter: (i) use of swollen joint count (SJC) instead of SJC and tender joint count (TJC) for joints, (ii) missing RF, (iii) anti-CCP, (iv) ESR or (v) CRP and (vi, vii, viii, ix) serology low or high values were replaced by a binary result attaching 2 or 3 points if the biomarker was present.

Results:

Patients (n=736) had mean age 51 yr (sd 14), median ESR 15 (0–121), RF+ 21%, Anti-CCP+ 20%, median SJC 1 (0–38), median TJC 6 (0–42), erosions 5% at baseline. Complete information was not obtained for 36 patients. Figure 1 shows the ROC-curves for the full criteria set and all modified criteria sets. The model with only SJC provided the best AUC was 0.82 (95%CI 0.79–0.85) while the lack of anti-CCP provided the poorest model (AUC 0.71 (95%CI 0.68–0.75)). Table 1 shows sensitivity and specificity for the full set and all modified sets. All models performed well in identifying the non-diseased (specificity), but moderate performance was shown for identifying those who were ill (sensitivity).

Conclusion:

The ACR/EULAR 2010 criteria are robust against missing values shown by a consistent diagnostic performance. However, the model with all 4 variables available but only swollen joints taken into account performed better than the original criteria including tender joints.

Table 1. AUC (95% CI), sensitivity and specificity. Sensitivity and specificity were calculated for the proposed cut-off of 6 or higher. (n=700)

 AUC, 95% CISensitivitySpecificity
(0) ACR/EULAR 2010 criteria0.73 (0.70–0.77)0.530.80
(i) ACR/EULAR 2010 criteria with only SJC0.82 (0.79–0.85)0.370.97
(ii) ACR/EULAR 2010 criteria; RF missing0.73 (0.70–0.77)0.490.81
(iii) ACR/EULAR 2010 criteria: Anti-CCP missing0.71 (0.68–0.75)0.490.80
(iv) ACR/EULAR 2010 criteria: ESR missing0.73 (0.70–0.77)0.520.80
(v) ACR/EULAR 2010 criteria: CRP missing0.73 (0.69–0.77)0.510.80
(vi) ACR/EULAR 2010 criteria: no value for RF, RF pos (2 points)0.73 (0.70–0.77)0.520.81
(vii) ACR/EULAR 2010 criteria: no value for anti-CCP, anti-CCP pos (2 points)0.73 (0.69–0.76)0.520.80
(viii) ACR/EULAR 2010 criteria: no value for RF, RF pos (3 points)0.73 (0.69–0.77)0.530.79
(ix) ACR/EULAR 2010 criteria: no value for anti-CCP, anti-CCP pos (3 points)0.73 (0.70–0.77)0.530.79

To cite this abstract, please use the following information:
Alves, Celina, Luime, Jolanda, van Zeben, Jendé, Huisman, Margriet, Weel, Angelique, Barendregt, Pieternella, et al; The Diagnostic Performance of the ACR/EULAR 2010 Criteria If One of Its Parameters Is Not Available. Results from the Rotterdam Early Arthritis CoHort (REACH). [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1776
DOI: 10.1002/art.29541

Abstract Supplement

Meeting Menu

2010 ACR/ARHP