Rheumatoid arthritis near remission: clinical rather than laboratory inflammation is associated with radiographic progression

Background Disease activity in rheumatoid arthritis (RA) can be measured clinically (eg, swollen joint count (SJC)) or systemically (eg, C reactive protein (CRP)). In general, both contribute to the progression of joint damage, but the relevance of residual inflammation in patients near remission is unclear. Objective To determine the independent contribution of SJC and CRP to progression of joint damage in patients near remission. Methods Data from methotrexate monotherapy arms of the ASPIRE, ERA, Leflunomide, PREMIER and TEMPO trials (n=1184) were pooled and the average SJC and CRP values from visits at 6, 9 and 12 months were determined. The two variables were then dichotomised into active and inactive, where inactive was defined as a mean. Radiographic outcomes were assessed according to these definitions. Results The greatest progression was seen in patients in whom both SJC and CRP were active and the smallest in those in whom both were inactive. If SJC was inactive, radiographic progression was not different between those with inactive or active CRP (0.7 +/- 4.3/year and 0.8 +/- 5.4/year, respectively, p=0.19). However, if CRP was inactive (<1 mg/dl), SJC status still determined radiographic progression (0.7 +/- 4.3/year and 1.8 +/- 5.6/year, for inactive and active SJC, respectively, p=0.004). The importance of SJC in patients with inactive CRP was also shown in a linear model (p=0.019), while CRP was not significantly different in patients with inactive SJC (p=0.40). Conclusion In patients with RA who are near remission, the amount of joint swelling appears to be more predictive of radiographic progression than the amount of CRP.