Category:Ankle Arthritis; Ankle; OtherIntroduction/Purpose:With the introduction of improved implants and long-term outcome data, total ankle arthroplasty (TAA) is becoming an increasingly common surgical treatment for end-stage ankle osteoarthritis. However, the treatment of a failed primary TAA remains a significant challenge. Ankle arthrodesis as a salvage procedure results in high rates of non-union and collapse. Revision arthroplasty is an alternative to arthrodesis, but there is little published data on the outcomes of revision total ankle arthroplasty (RTAA). This study presents 2-year outcomes after RTAA using a modular prosthesis and metal/cement augmentation to reconstitute talar height, as well as restore sagittal and coronal alignment.Methods:A retrospective review was performed on 23 patients who underwent RTAA after failed primary TAA. Demographic data, talar height, coronal and sagittal alignment, and range of motion pre-revision and at most recent follow up were recorded. Failure was defined as need for revision surgery during the follow up period. Radiographic measurements were performed on weight bearing lateral and AP radiographs. For Agility implants, measurement on weight-bearing computed tomography (WBCT) was required. The measurement methodology was performed on both radiographs and WBCT for validation and measurements were consistent across all implants.Results:Patient follow-up ranged from 2 to 3.9 years, with a mean of 2.56 years. 17 of 22 RTAA did not require further revision. Of the five failures, one was due to deep infection, four to subsidence of the talar component. For the 17 successful revisions, average pre-operative coronal malalignment was 3.8◦ (range 7.4◦ varus to 15◦ valgus), and average post-operative malalignment improved to 2.6◦ (range 0◦ varus to 7.9◦ valgus), but this difference was not statistically significant (p=.09). Average pre-operative sagittal malalignment was 8.7◦ (range 20.7◦ plantarflexion to 20.1◦ dorsiflexion), and average post-operative malalignment improved to 3.6◦ (range 8.3◦ plantarflexion to 9.3◦ dorsiflexion), which was statistically significant (p=.01). Talar height improved by 3.9mm (p< 0.001), and range of motion from 16.9◦ to 25.0◦ (p,0.001).Conclusion:At a minimum of two years of follow up, revision arthroplasty shows improved alignment, talar height, and range of motion. While the failure rate remains significantly higher than primary ankle arthroplasty, it is comparable or superior to that of conversion to arthrodesis. The complexity of RTTA varies greatly due to surgical risk, soft tissue quality, and residual bone stock amongst other factors, which limits the generalizability of this patient cohort. RTAA is a viable option for the salvage of failed primary TAA, with functional and radiographic improvements shown at mid-term follow-up.