A 48-week phase II open, uncontrolled study of auranofin (AF) in patients juvenile rheumatoid arthritis (JRA) was conducted to assess efficacy, tolerance and serum gold levels, and to consider the feasibility of further controlled studies (phase III) in such patients. The study group included 25 patients (20 F, 5 M) with active pauciarticular (n = 4) or polyarticular (n = 21) JRA. Median age was 100 months (range 62-176); median disease duration was 55 months (range 13-155). AF was given at 0.1 mg/kg/day divided into 2 doses, and increases to maximum of 6 mg/day were permitted if clinical improvement was insufficient. Nonsteroidal anti-inflammatory drugs and physiotherapy also were allowed. Significant improvement (p less than or equal to 0.05) was observed in the various joint counts and articular indices, as well as in the physician's global assessment of health. Some subjective functional measurements also were improved, but the differences were not statistically significant. There were no clinically important trends in the laboratory measurements. No patient was withdrawn because of adverse reactions; 1 was withdrawn because of disease exacerbation and 3 for lack of response. Mild adverse reactions were seen in 9 patients: 6 had abdominal pain, 6 diarrhea/loose stools, 1 nausea, 3 rash, 2 pruritus. Two patients had dosage reduced because of loose stools. Serum gold levels varied greatly; increased dosage usually resulted in increased serum levels. Occurrence of adverse reactions or response to therapy was not related to increases in dose or to serum gold levels.(ABSTRACT TRUNCATED AT 250 WORDS)