Background/Purpose: Intra articular (IA) corticosteroid injections aim at amelioration of symptoms of local inflammation in rheumatoid arthritis (RA). Some patients also report a systemic effect. In treat-to-target strategies IA injections may help to achieve low disease activity. Our aim was to investigate the immediate and late effects on pain, disease activity and functional ability over time of IA injections and to compare the subsequent disease course in injected versus non-IA injected RA patients.

Methods: In the BeSt study IA injections were allowed in all four treatment strategy arms at the discretion of the treating rheumatologist. We evaluated the effect of IA injections administered in large joints in the first year of treatment. We compared pain scores (Visual Analogue Scale (VAS)), Disease Activity Score (DAS) and local joint swelling and tenderness before and after injection. The subsequent disease course in IA injected versus non-IA injected patients was compared using a linear mixed model (LMM) to model the DAS and the Health Assessment Questionnaire (HAQ) score over time. Considered confounders were age, gender, DAS at baseline, treatment strategy, body mass index (BMI), anticitrullinated peptide antibodies (ACPA), rheumatoid factor (RF) and DAS at time of outcome HAQ or HAQ at time of outcome DAS.

Results: 61 patients received >=1 IA injections in the large joints, mostly in shoulders (34%) and knees (31%). Pre-injection, local joint swelling was scored in 48% of injected joints, and local tenderness in 64%. Three months after the injection, 56% of the swollen joints were no longer swollen and 52% of the tender joints were no longer tender. Mean (SD) DAS before was 3.7 (1.3) and after was 3.3 (1.4) respectively (p<0.01), with a mean (SD) VAS for pain of 52 (26) before and 39 (26) after IA injection (p<0.01). Significantly more IA injections were given in the first year of treatment with initial methotrexate monotherapy than in the initial combination therapy arms (including either prednisone or infliximab), p<0.01. LMM showed a significantly higher DAS and HAQ in patients who received IA injections in year 0-1 compared to those who did not (mean DAS (SD) 3.07 (0.10) versus 2.75 (0.04), p<0.01) mean HAQ (SD) 0.90 (0.04) versus 0.82 (0.02), p=0.03), and no significant difference in subsequent years (table 1). Patients who were injected in year 0-1 had a similar number of treatment adjustments in subsequent years compared to patients who were not. Radiological damage after 8 years was present in 18% of the X-rays of injected joints, and in14% of the X-rays of non-injected joints.

Conclusion: Our data suggest that IA injections have moderate short term and limited long term efficacy in patients with early RA.

Table 1: Mean DAS and mean HAQ in injected patients versus non injected patients