Background The interchangeability of Disease Activity Score 28 joints (DAS28) calculated using the erythrocyte sedimentation rate (ESR) versus C-reactive protein (CRP) has been questioned (1).
Objectives To compare the 28-Joint Disease Activity Score (DAS28) calculated using C-reactive protein (CRP) with that using erythrocyte sedimentation rate (DAS28-ESR) in a multi-centre, usual care, early rheumatoid arthritis (ERA) cohort.
Methods The present study analyzed baseline and 12-month follow up data from the Canadian Early Arthritis Cohort. The study sample included patients with RA with symptom duration less than 12 months and complete data on DAS28-ESR and DAS28-CRP at both time points. The linear association between continuous DAS28-ESR and DAS28-CRP scores was estimated using Pearson's correlation coefficient and differences across the range of DAS28 scores were examined with Bland-Altman Plots. Receiver operating characteristic (ROC) analysis was used to identify optimal thresholds for disease activity for DAS28-CRP scores that would best correspond with established DAS28-ESR thresholds in the total sample as well as in stratified samples by age and sex. Agreement between newly calculated DAS28-CRP and established DAS28-ESR thresholds was examined (kappa statistic).
Results The study sample included 995 ERA patients. Sample baseline mean (SD) age was 53.7 (14.5) with 5.8 (2.9) months of symptom duration and 738 (74%) were female. Though continuous DAS28-CRP and DAS28-ESR scores were highly correlated (r=0.92, p<0.0001), Bland-Altman plots showed that DAS28-ESR scores were higher than DAS28-CRP scores across the spectrum of disease activity, with larger positive discrepancies at lower values. Positive discrepancies between DASCRP and DASESR were higher for women and patients over 60 years old. Optimal thresholds identified for the DAS28-CRP were 2.4 for remission, 2.9 for low disease activity, and 4.6 for high disease activity. Overall, there was moderate to good agreement between newly calculated DAS28-CRP and DAS28-ESR thresholds (k=0.6).
Conclusions Results from this large observational study of “real-world” ERA patients showed that DAS28-CRP and DAS28-ESR scores were strongly correlated but that DAS28-CRP scores were consistently lower than DAS28-ESR scores. Differences between DAS28-CRP and DAS28-ESR were greater at lower ranges of the DAS28 and were most pronounced in women and older patients. This suggests that substituting one continuous score for the other is likely reasonable for research purposes in data analysis and that interpreting the DAS28-CRP using established DAS28-ESR thresholds may be acceptable for higher levels of disease activity, but not for lower levels, particularly not for remission.
Siemons L, Vonkeman HE, ten Klooster PM, van Riel PL, van de Laar MA. Interchangeability of 28-joint disease activity scores using the erythrocyte sedimentation rate or the C-reactive protein as inflammatory marker. Clin Rheumatol. 2014 Jun;33(6):783–9.
Disclosure of Interest B. Kuriya: None declared, O. Schieir: None declared, D. Lin: None declared, J. Pope: None declared, G. Boire: None declared, B. Haraoui: None declared, C. Thorne: None declared, D. Tin: None declared, C. Hitchon: None declared, E. Keystone: None declared, V. Bykerk Grant/research support from: The CATCH study was designed and implemented by the investigators. The authors have received unrestricted grants from: Amgen Canada Inc and Pfizer Canada Inc - Founding sponsor since 2007, Hoffmann-La Roche Ltd, UCB Canada Inc, Bristol-Myers Squibb Canada, AbbVie Corporation and Janssen Inc since 2011, Medexus Inc since 2014 and Eli Lilly Canada Inc since 2015.