PURPOSE: To assess the efficacy and side effects of immunosuppressive therapy in patients with Behçet's disease uveitis. METHODS: A nonrandomized retrospective case-series study analyzed data from 22 patients with Behçet's disease uveitis, from a single Uveitis Service, São Paulo, Brazil (period 1978-2007), under systemic chlorambucil and/or cyclosporine A, for at least 6 months with a minimum one-year follow-up. Drug efficacy was measured by reduction in relapse rate and reduction of prednisone dose. RESULTS: Patients (10M/12F) mean age was 29 (range 10-43) years-old at the onset of uveitis. The median duration of followup was 11 (range 1-29) years-old. Chlorambucil (2-6 mg/day) was used in 13 patients and cyclosporine A (3-5 mg/kg/day) in 9 patients at initiation. Drugs were switched because of no effectiveness or side-effects. Chlorambucil was effective in 78.5% (11/14) and induced disease remission in 43% (6/14) of patients, whereas cyclosporine A was effective in 57% (8/14) of patients. Chlorambucil and cyclosporine A were discontinued due to side effects in 21% (leucopenia) and in 57% of patients (nephrotoxicity, 36% and gastrointestinal complications, 21%), respectively. No case of late malignancy was observed. 36% (16/44) of eyes had final visual acuity < or =0.1, among which 69% (11/16) had already this visual acuity at the first visit. CONCLUSION: This study reiterates previous data that chlorambucil can induce long-term remission of Behçet's disease uveitis, whereas cyclosporine is effective but side effects limit its use. Chlorambucil therapy may still be a reasonable option in patients with intractable, sight-threatening Behçet's disease uveitis.