Category:Ankle; Ankle Arthritis; HindfootIntroduction/Purpose:Despite substantial advancements in Total Ankle Arthroplasty (TAA), catastrophic failure due to implant subsidence remains a common and serious problem. While there are many studies of short and some of intermediate term follow- up of TAA, the number of long-term studies, especially with prospectively collected data, is much fewer, and the data on the incidence and impact of talar subsidence on those results is even more limited. Talar subsidence can be especially challenging to manage, given the limited talar bone stock and revision options as compared to the larger tibial bone stock. The purpose of this study was to evaluate the long-term incidence and characteristics of TAA implant failure.Methods:A prospectively collected database of TAAs performed in a tertiary referral center began enrolling in 1999. Medical records/radiographs were reviewed to evaluate outcomes of TAA and final radiographs were reviewed to identify implant failure from 1999-2016. Minimum radiographic follow-up was two years. Pre-operative and post-operative radiographs were measured for ankle, foot, and implant coronal and sagittal alignment Failure and reoperation rates of two different implants, the mobile bearing Scandinavian Total Ankle Replacement (STAR) and the fixed bearing Salto Talaris (Salto) were analyzed. Paired Student t- tests were performed between groups. TAA revision surgeries were excluded if the primary TAA was performed at another institution. Reoperations were recorded according to the Reoperations Coding System (CROCS) classification of the Canadian Orthopaedic Foot and Ankle Society (COFAS). There were 149 TAAs reviewed in 146 patients (136 STARs and 13 Saltos), with an average follow-up of 7.03 years (range 2-20 years).Results:Implant survival was 85.8% at mean 7 years. 25 TAAs demonstrated radiographic catastrophic failures, 21 underwent re- operation, and 4 declined revision surgery (Table 1). One failure was due to infection (CROCS 10), others were aseptic (CROCS 9). 13/25 failures were related to talar subsidence, all STARs (9.6% of all STARs in the series). Nine STARs (6.4%) had less severe talar subsidence but were asymptomatic, The STAR talar component, in total, subsided in 17.6% (24/136) of STARs, requiring revision in 13 (9.6%) of cases at mean 3.8 years post-op. There was no difference in pre-operative/post-operative radiographic alignment between the TAAs that did and did not fail. Twenty-six ankles (17.4%) underwent additional surgery with retention of metal components, bringing the total reoperation rate to 47/149 (31.5%).Conclusion:At long-term follow-up, TAA demonstrates reasonable implant survival rates, especially given the learning curve that includes STARs implanted prior to the Food and Drug approval study. We describe a unique mode of failure in STARs that highlights one of the risk factors for failure in total ankle arthroplasty. This represents the highest reported rate of talar subsidence in the literature with almost 10% of STARs demonstrating talar-sided failure at mean 7-year follow-up. Surgeons who utilize the STAR total ankle should be vigilant for talar subsidence given the high rates reported in this series.