OBJECTIVES To evaluate the benefits of the addition of leflunomide (LEF) in children with polyarticular course juvenile idiopathic arthritis (JIA), non-responsive to standard dose parenteral methotrexate (MTX). METHODS In an observational study, 32 children with polyarticular course JIA failing standard dose MTX (up to 15 mg/m2/week sc for at least 3 and up to 6 months) received additional LEF. Permitted concomitant drugs included pulse steroids for flares and/or low bridging dose of prednisolone, intra-articular steroids and non-steroidal anti-inflammatory drugs. No other DMARDs had been used before enrolment. Patients underwent 8-12 weekly assessment. At each visit, core set of outcome variables and laboratory parameters, viz. haemogram and liver enzymes were recorded. The primary efficacy outcome was the ACR Pedi 30 criteria. At the last follow up, Wallace's criteria were used to determine children achieving remission. RESULTS 25 of 32 children who followed up for at least 3 months were analysed. Mean follow up duration following addition of LEF was 1.61 years (range: 0.29 to 3.0 years). At 3 months, 68% of the patients met the ACR Pedi 30 response. 17 of the 20 children (85%) showed an ACR Pedi 30 response at 6 months and 16 out of 18 (88.8%) at 1 year. Of the 18 children followed up till the end of the study, 12 (66.6%) met the ACR Pedi 30 criteria and 9 (50%) were in clinical remission on medications (off steroids). Adverse effects were observed in 2 children (gastritis in one and elevated liver enzymes in the other). CONCLUSIONS Our findings support further study of the role of this combination in the management of polyarticular course JIA refractory to standard dose MTX, especially in resource challenged settings where biologicals are unaffordable. The open observational nature of the study is its limitation.