Fifty consecutive patients with RA and clinical wrist synovitis were randomly allocated to either receive an injection of glucocorticoid into the radiocarpal joint or have the same amount of drug divided into the radiocarpal and midcarpal joints. Inferior radioulnar joint synovitis was treated with extra steroid injection only in the latter group. Patient's and doctor's assessments as well as ultrasonography improved significantly at month 3 in both groups. A statistical difference between the groups was found in the midcarpal joint measurement favouring the extra midcarpal injection. Ten wrists (20%) were normal when assessed with ultrasound at month 3 while 34 wrists (68%) were normal on clinical assessment.