Abstract

Objectives: To evaluate the ability of tumour necrosis factor (TNF) antagonist therapy to produce remission and prevent progression to rheumatoid arthritis (RA) in patients with poor prognosis undifferentiated inflammatory arthritis (UA).

Methods: Patients with UA of <12 months’ duration and having relapsed after a single parenteral corticosteroid injection were recruited into a double-blind, placebo-controlled trial of infliximab or placebo monotherapy administered at weeks 0, 2, 6 and 14. Methotrexate was added at week 14 if no clinical response (raised C-reactive protein (CRP) and clinical synovitis) was achieved. Standard outcomes were collected at baseline, infusion visits and weeks 26 and 52. The primary outcome was clinical remission at week 26.

Results: 17 patients were randomised (10 infliximab, 7 placebo) all with poor prognostic features. At week 14, the infliximab group had greater improvements in CRP and Health Assessment Questionnaire (HAQ) but by week 26 there was just a trend favouring infliximab for early morning stiffness, tender joint score, swollen joint score and HAQ; there was no significant difference in 28 joint count Disease Activity Score between the two groups. Furthermore, only three patients were in clinical remission (two infliximab, one placebo). By week 52, 100% patients in the infliximab group and 71% (5/7) patients in the placebo group had developed RA.

Conclusions: In poor prognosis UA, a short course of TNF antagonist therapy provided modest short-term relief but did not prevent the development of RA. Patients with UA with a poor prognosis relapsing after corticosteroid have a high risk of evolving to RA and are suitable candidates for interventional treatment.