Background/Purpose: The threshold for reporting of C-reactive protein (CRP) differs from laboratory to laboratory. Moreover, CRP values are affected by the intra individual biological variability.[1] With respect to disease activity score in 28 joints (DAS28) and Rheumatoid Arthritis (RA), precise threshold for reporting CRP is important due to the direct effects of CRP on calculating DAS28, patient classification and subsequent treatment decisions[2]

Methods: This study consists of two sections: a theoretical consideration discussing the performance of CRP in calculating DAS28 with regard to the biological variation and reporting limit for CRP and a cross sectional study of  all RA patients from our department (n=876) applying our theoretical results. In the second section, we calculate DAS28 twice with actual CRP and CRP=9, the latter to elucidate the positive consequences of changing the lower reporting limit of CRP from <10mg/L to <3mg/L

Results: Theoretical considerations: lower reporting limit of <10mg/L leads to inaccurate patient classification in a great number of patients both because of biological variation, as well as the larger spectrum of numbers (0-9). In addition, reducing lower reporting limit for CRP to minimum<3mg/L results in optimizing patient classification (Figure 1). The logarithmic transformation of CRP in DAS28 formula has an important role Cross sectional study: 769 patients fulfilled the ACR criteria for RA were included, 107 excluded due to missing parameter used for DAS28 calculations. There was a statistically significant difference between patients´ DAS28 and new calculated DAS28 with CRP=9 (p<0.001). A total of 109 patients had a disease activity deviation (remission to low: 66, low to moderate: 39, moderate to high: 4)

Conclusion: Respecting DAS28 calculation, lower reporting limit for CRP<3 mg/L is acceptable and should be taken into consideration. A lower reporting limit for CRP<10 mg/l is too high. It is particularly relevant if treatment decisions are solely made on the basis of DAS28. Furthermore, we conclude that DAS28 in studies where the reporting limit of CRP is unknown are incomparable References:1.Macy EM, et al. Variability in the measurement of C-reactive protein in healthy subjects: implications for reference intervals and epidemiological applications. Clin Chem. 1997;43:52-8 2.Asmussen R, Antonsen S, Hansen IMJ. The influence of variation in C-reactive protein values on the DAS28 score. Ann Rheum Dis 2013;72:844

Figure 1: Dependency of Disease Activity Score in 28 joints to CRP (theoretical consideration). Tender joint: 3, Swollen joint: 1, patient global assessment: 34 and CRP ranges from 0 to 40.  CRP: C-reactive protein.  CV-Within: Intra-individual biologic variability