This study was aimed to evaluate the usefulness of rifamycin SV as an agent for local treatment of rheumatoid synovitis. Rifamycin SV was compared with triamcinolone acetonide in a randomized controlled trial on 87 patients with rheumatoid arthritis and with persistent knee synovitis. The treatment with rifamycin consisted of several weekly intra-articular injections, whereas triamcinolone was given as a single intra-articular injection. At the end of the therapy, 27 (61.4%) of the 44 rifamycin patients and 39 (91%) of the 43 steroid patients responded well to the treatment, and this difference was significant (p < 0.01). Rifamycin SV was responsible for unpleasant local side effects in all cases. In both groups, after 1 year of follow-up the synovitis had relapsed in about 42% of cases. We conclude that rifamycin SV is less useful than triamcinolone acetonide in the local treatment of rheumatoid synovitis.