Uncemented STAR total ankle prostheses.

Abstract

BACKGROUND: The feasibility of replacing the ankle joint has been a matter of speculation for a long time. In recent years, the designs of ankle prostheses have been improved, and three designs, all used without bone cement, currently dominate the market. However, documentation of the clinical results of the use of these prostheses is sparse. We reviewed the intermediate-term results of fifty-one consecutive Scandinavian Total Ankle Replacements (STAR). METHODS: Between 1993 and 1999, fifty-one consecutive ankles were replaced with an uncemented, hydroxyapatite-coated STAR total ankle prosthesis. Clinical examination for the present study was performed by one surgeon who had not taken part in the operations. Standardized radiographs were used. Complications and failures were recorded, and patient satisfaction and functional outcome scores were determined for all patients with an unrevised implant. RESULTS: Twelve ankles had to be revised. Seven were revised because of loosening of at least one of the components; two, because of fracture of the meniscus; and three, for other reasons. A component was exchanged in seven of the twelve revisions, whereas the ankle was successfully fused in the other five. An additional eight ankles had radiographic signs of loosening. The thirty-nine unrevised ankles (thirty-seven patients) were examined after thirty-six to ninety-seven months (median, fifty-two months). The patient was satisfied with the result after thirty-one of the ankle replacements, somewhat satisfied after two, and not satisfied after six. The median Kofoed score increased from 39 points before the surgery to 70 points at the time of the follow-up examination. A median follow-up score of 74 points was recorded when the system described by Mazur et al. and the AOFAS (American Orthopaedic Foot and Ankle Society) system were used. The median range of motion was approximately the same preoperatively and postoperatively. The estimated five-year survival rate, with revision for any reason as the end point, was 0.70. When radiographic loosening of either component was used as the end point, the estimated five-year radiographic survival rate was significantly better for the last thirty-one ankles treated in the series (p = 0.032). CONCLUSIONS: Total ankle replacement may be a realistic alternative to arthrodesis, provided that the components are correctly positioned and are of the correct size. However, the risks of loosening and failure are still higher than are such risks after total hip or total knee replacement.