Patients with rheumatoid arthritis attending rheumatology out-patients who had a symptomatic knee effusion were randomly allocated to receive one of three treatments: group I, a steroid injection without washout; group II, a joint washout with normal saline; group III, a joint washout with normal saline and steroid injection. Sixty knees in all were studied. Laboratory parameters for disease activity (erythrocyte sedimentation rate, C-reactive protein) were monitored in all patients prior to the study and at 3 months. Clinical assessment of disease activity (pain, morning stiffness involving the knee, circumference of the knee, walking distance and range of movement) were recorded prior to the study, at 1 month and at 3 months. All three treatments, resulted in a reduction of pain and increased movement. However, patients who had a joint washout alone showed significantly less improvement as compared with the other two groups. Symptomatic improvement was marginally greater in patients following joint washout and injection than in those who had had joint injection alone. The results of the study indicate that the simple procedure of joint aspiration and steroid injection, which can be carried out in the out-patient clinic, provides satisfactory relief of symptoms in rheumatoid patients with knee effusions. Joint washout alone was less beneficial.