Arthritis of the wrist is a painful disabling condition that has various causes and presentations. The traditional treatment has been a total wrist fusion at a price of the elimination of movement. However, forms of treatment which allow the preservation of movement are now preferred. Modern arthroplasties of the wrist are still not sufficiently robust to meet the demands of many patients, nor do they restore normal kinematics of the wrist. A preferable compromise may be selective excision and partial fusion of the wrist using knowledge of the aetiology and pattern of degenerative change to identify which joints can be sacrificed and which can be preserved.

This article provides an overview of the treatment options available for patients with arthritis of the wrist and an algorithm for selecting an appropriate surgical strategy.

Cite this article: Bone Joint J 2015;97-B:1303–8.

There is an high incidence of failures of total wrist arthroplasties. We review our experiences in revising total wrist implant arthroplasties to arthrodeses. The most common mode of failure of the arthroplasties in our series was metacarpal loosening with dorsal perforation of the stem. Loosening of the proximal stem, progressive mal-position of the wrist and other causes appeared. We used for the revision arthrodesis in all our cases tricortical iliac crest bone grafts and additional spongiosa transplants from this donor site region. In one case we used a vascularized iliac crest bone graft to bridge the bone defect because of a bad host quality of the recipient area. Fixation was achieved with plates and screws. Our average follow-up period was 32 month. 40 patients with 41 failed wrist implants (3 different types) were treated with this technique. 40 wrist undergoing arthrodesis attained a solid painless fusion after a single operation. In one case a non-union with a loosening of the screws due to using a non-rigid plate was seen. In this case a revision was necessary to achieve a bone healing. All patients were satisfied, pain free and achieved an increased pinch and grip strength after bony fusion (measured with Yamar-Vigorimeter). A persisting loss of carpal height was seen in all cases. Arthrodesis after failed total wrist arthroplasty is a satisfactory salvage procedure even in cases with a bad quality of the recipient area. We recommend a rigid fixation technique to prevent non-unions

There is an incidence of failures of total wrist arthroplasties. We review our experiences in revising total wrist implant arthroplasties to arthrodeses. The most common mode of failure of the arthroplasties in our series was metacarpal loosening with dorsal perforation of the stem. We used for the revision arthrodesis in all our cases tricortical iliac crest bone grafts and additional spongiosa transplants from this donor site region. In one case we used a vascularized iliac crest bone graft to bridge the bone defect because of a bad host quality of the recipient area. Fixation was achieved with plates and screws. Our average follow-up period was 28 month. 14 patients with 15 failed wrist implants were treated with this technique. 14 patients undergoing arthrodesis attained a solid painless fusion after a single operation. In one case a non-union with a loosening of the screws due to using a non-rigid plate was seen. In this case a revision was necessary to achieve a bone healing. All patients were pain free and achieved an increased grip strength after bony fusion. Arthrodesis after failed total wrist arthroplasty is a satisfactory salvage procedure even in cases with a bad quality of the recipient area. We recommend a rigid fixation technique to prevent non-unions

We have implanted 76 biaxial total wrist prostheses as a primary procedure in patients with rheumatoid arthritis of the wrist. A total of 66 was reviewed at a mean follow-up time of 52 months. Pain was relieved in 67% of the surviving wrist replacements. On the basis of the Hospital for Special Surgery scoring system, 49 wrists (74%) were graded as fair to excellent. More than half of the 27 patients who had an arthrodesis on the contralateral wrist would have preferred a second arthroplasty. Five replacements were revised or fused because of loosening and a further nine showed signs of radiological loosening, three of which were asymptomatic. The probability of survival of the biaxial total wrist replacement at eight years was 83% with revision surgery as the terminal event, 78% with radiological loosening as the endpoint and 82% with dorsal migration and displacement from the metacarpal as the terminal event. There was a linear relationship between subsidence of the component and distal loosening. There was no evidence that the length of the stem of the carpal component, within the third metacarpal, affected any of the terminal events. The position and alignment of the carpal component within the bone at the time of surgery significantly affect the outcome and can be used to predict failure