Objectives Disease Activity Score in 28 joints calculated with C-reactive protein (DAS28-CRP) is used instead of erythrocyte sedimentation rate (DAS28-ESR) to assess rheumatoid arthritis disease activity; however, values for remission and low disease activity (LDA) for DAS28-CRP have not been validated. American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) guidelines suggest remission should be calculated by Simplified Disease Activity Index (SDAI) rather than DAS28-ESR. We examined values of remission and LDA of DAS28-CRP that correspond to the respective cut-off points for DAS28-ESR and SDAI from five clinical trials.

Methods DAS28-CRP cut-offs that best correspond to DAS28-ESR remission <2.6 and LDA ≤3.2 were obtained by cumulative distribution plots, receiver operating curves and maximum concordance and averaged for each approach, treatment group and study. Level of agreement between DAS28-CRP and DAS28-ESR remission and LDA cut-offs was compared against each other and versus SDAI remission ≤3.3 and LDA ≤11.

Results Percentage of patients who achieved remission and LDA by DAS28-ESR cut-offs was greater for DAS28-CRP versus DAS28-ESR regardless of patient population or treatment group. Discordance between CRP and ESR cut-offs ranged from 4%–26% and 8%–23% for remission and LDA, respectively, and 19%–40% and 6%–11% for DAS28-CRP versus SDAI, respectively. Estimated (range) remission and LDA thresholds were 2.4 (2.2–2.6) and 2.9 (2.6–3.3), 1.9 (1.6–2.2) and 3.1 (3.1–3.3) and 2.2 (1.1–2.9) and 3.6 (3.4–4.0) for DAS28-CRP versus DAS28-ESR, DAS28-CRP versus SDAI and DAS28-ESR versus SDAI, respectively.

Conclusions DAS28-CRP underestimates disease activity when using cut-off points validated for DAS28-ESR; therefore, DAS28-ESR cut-off values should not be applied to DAS28-CRP. Although DAS28-CRP and DAS28-ESR cut-offs for LDA ≤3.2 correspond to SDAI LDA, neither corresponds well to SDAI remission.