[Total elbow replacement in patients with rheumatoid arthritis]

Total elbow arthroplasty is associated with a higher occurrence of complications than is usual for large-joint replacements. Two kinds of prostheses, unconstrained and semi-constrained, are currently used and each has its supporters or opponents. In this study the results of the two techniques used in our patients are evaluated and compared.

Two groups of elbows in patients with rheumatoid arthritis were evaluated. One comprised 58 elbows treated by Souter-Strathclyde total elbow arthroplasty (S-S group). The mean age of the patients at the time of surgery was 53 years (range, 22 to 71) and the mean follow-up was 9.5 years (range, 0.7 to 16.7). The other group included 63 elbows treated by Coonrad-Morrey elbow arthroplasty (C-M group). The mean age of the patients at the time of surgery was 54 years (range, 26 to 75) and the mean follow-up was 4.21 years (range, 0.28 to 7.87).

The Kaplan-Meier analysis was used to estimate implant survival in each group. Clinical assessment included range of motion and pain experience. The Mayo Elbow Performance Score (MEPS) was used as a clinical rating scale for the whole group. Radiographs were taken in two basic projections. The elbows with an implant removed or re-implanted were excluded from the evaluation. The patients were studied prospectively. The results were statistically analysed, with the level of significance set at 0.05.

All patients experienced pain relief after surgery. In the S-S group, 35 elbows were free from pain (77.7 %), in the C-M group this was 53 elbows (88.3 %). The range of motion improved after arthroplasty in both groups. Flexion more than 110° was achieved in twice as many elbows in the C-M group than in the S-S group. Flexion contracture in the S-S group did not improve significantly. MEPS values after surgery improved in both groups, with significantly better results in the C-M group. In the S-S group, four elbow arthroplasties (6.9 %) showed instability, which was treated by replacement with a semi-constrained implant in one case and managed by articulated external fixation of the elbow for 6 weeks in three cases. Radiolucent lines were detected in five replacements (11.1 %) along the whole ulnar component width, in 12 (26.6 %) in the olecranon region and in 14 (31.1 %) in the distal humeral component. In the C-M group no radiolucency was recorded around the component. In the S-S group, revision surgery was carried out in 13 arthroplasties (22.4 %); of these, 10 (17.2%) were due to aseptic loosening, one (1.7 %) due to instability and one (1.7%) because of deep infection. In the C-M group, three elbows required revision (4.8 %), one for periprosthetic fracture (1.6 %) and two for deep infection (3.2 %). The results of survival analysis did not differ between the two groups.

The weak point of Souter-Strathclyde total elbow arthroplasty is the ulnar component whose impairment and subsequent wear are involved in all cases of aseptic loosening. An insufficient length of the humeral component is another risk factor. Even natural movements of the elbow joint produce concentration of stresses on a small surface; this gradually weakens component fixation in bone and results in aseptic loosening. A higher risk of failure in Coonrad-Morrey elbow arthroplasty is associated with polyethylene lining of the hinge mechanism.

The Coonrad-Morrey total elbow replacement is at present considered the method of choice. It is easier to perform and provides better functional outcomes than the Souter-Strathclyde elbow implant.