Aims: The aim of the present study was to compare the functional outcome between silastic Swanson and Sutter implants in metacarpophalangeal joints in a prospective and randomised of study of patients with rheumatoid arthritis. Methods: The study was prospective and randomised. Thematerial comprised 53 patients (6 men, 47 women) and 58 hands; a total of 89 Swanson and 126 Sutter implants were installed. The mean follow-up time was 57 (40–80) and 55 (36–79) months, respectively. Results: Active extension was corrected similarly in both groups. In the Swanson group from median [inter-quartile rate (IQR)] 22° (8, 44) to 12° (0,20), and in the Sutter group from 19° (7, 37) to 16 (5, 25). The median active flexion decreased less in the Sutter group from 75° (69, 84) to 64° (53, 72) vs. Swanson from 83° (77, 90) to 58° (48, 64) during the follow-up, and with that respect the difference between the groups was statistically significant (p=0.01). Statistically significant difference (p=0.03) was detected in correction of ulnar deviation only in ring finger: in the Swanson group from median 25°(6°, 34°) ulnar deviation to 1°(2°radial, 9°) radial deviation and in the Sutter group from 14°(7°, 28°) to 5°(0°, 13∞) respectively. Conclusion: It appears that the Sutter implant yields at least as good functional results in the MCP replacement as the Swanson prosthesis.