Results of Total Ankle Arthroplasty

Historically, the only option for a patient with painful ankle arthritis was arthrodesis. Although ankle motion was lost, arthrodesis successfully relieved pain; however, the biomechanical alterations from ankle arthrodesis often resulted in the development or exacerbation of arthritis in other joints, as well as complications such as non-union, malalignment, infection, and decreased gait speed and mobility.^1–4 The success of total hip and knee replacement gave hope that equally good results could be obtained with total ankle arthroplasty (TAA). Beginning in the 1970s, multiple designs of ankle prostheses were created and many appeared to provide good shortterm results.^5–8 However, as longer-term follow-up data were gathered, it became apparent that these designs had a number of flaws. Subsequent reports of severe osteolysis, component loosening, impingement, infection, and soft-tissue breakdown led most surgeons to abandon TAA.^9–14 Patient satisfaction with these first-generation implants ranged from 19 to 81%, with results deteriorating with longer follow-up. Stauffer and Segal¹² reported a 41% complication rate in 102 ankle arthroplasties, while Bolton- Maggs et al.⁹ reported that only 30% of their 41 patients were satisfied with their results. Takakura et al.¹³ reported only 27% satisfactory results in 30 cemented TAAs, and 67% satisfactory results in 30 uncemented TAAs. In 1988, McGuire et al.¹¹ compared 23 TAAs with 19 arthrodeses and found 95% good or excellent results at three-year follow-up in those with arthrodesis compared with 69% good or excellent results in those with TAA. In a 1996 report¹⁵ of 160 arthroplasties with the Mayo total ankle prosthesis, only 19% had good results; 36% had failed and had to be removed. The authors concluded: “On the basis of these findings, we do not recommend ankle arthroplasty with a constrained Mayo implant for rheumatoid arthritis or osteoarthritis of the ankle.

”These disappointing results with TAA led many to return to arthrodesis as the treatment of choice for painful ankle arthritis, but they also spurred renewed efforts to design a prosthesis that would be stable and long-lasting and better replicate the complex anatomy of the ankle joint. Secondgeneration ankle prostheses were of two basic designs: three-component mobile-bearing and two-component fixed-bearing with varying degrees of constraint (constrained, semi-constrained, or unconstrained). Mobilebearing prostheses have a moving polyethylene bearing separating the convex talar component from the flat tibial component, forming two separate articular surfaces. Fixed-bearing prostheses have only one articulation between the tibial and talar components. Both designs have specific advantages and disadvantages. Fixed-bearing designs are less likely to break or dislocate, but mobile-bearing designs provide greater congruence and, theoretically, less wear. The Salto Talaris prosthesis has combined these two design features by placing the mobile-bearing concept into the instrumentation.

Differences in patient selection, prosthesis used, length of follow-up, endpoint for survival determinations (re-operation or failure leading to arthrodesis), and outcome measurement methods make it difficult to determine an overall success rate for the second- and third-generation TAA implants, but reported results in large numbers of patients are dramatically superior to those reported with earlier designs. Some of the largest series include 200 STAR arthroplasties with 93% survival at five years,¹⁶ 306 Agility arthroplasties with 89% survival at five years,¹⁷ 93 Buechel- Pappas arthroplasties with 84% survival at eight years, ¹⁸ and 93 Salto arthroplasties with 98% survival at five years.¹⁹ Other studies have reported patient satisfaction rates ranging from 79 to 97%.^20–28

To compare the intermediate and long-term outcomes of second-generation TAA and ankle arthrodesis, Haddad et al. ²⁹ performed a comprehensive search of all relevant articles published in English between January 1990 and March 2005, as well as all relevant abstracts from the 2003 and 2004 annual proceedings of major orthopaedic meetings. They identified 49 primary studies, 10 of which evaluated TAA in 852 patients and 39 of which evaluated ankle arthrodesis in 1,262 patients. Excellent or good results were reported in 69% of TAA patients and in 68% of arthrodesis patients; 7% of TAA patients required revision surgery compared with 9% of arthrodesis patients; and below-knee amputation was required in 1% of TAA patients and 5% of arthrodesis patients. Survival rates for TAA were 78% at five years and 77% at 10 years. The authors concluded that, on the basis of their findings, the intermediate outcome of TAA appeared to be similar to that of ankle arthrodesis; however, they noted that data were sparse and more comparative studies are needed. Another recent comparison of TAA and ankle arthrodesis that included 480 TAAs and 4,705 ankle fusions ³⁰ found that the rates of major revision were 9% at one year and 23% at five years for TAA compared with 5 and 11%, respectively, after ankle arthrodesis. The authors noted that TAA has potential advantages in terms of a decreased risk of a subtalar fusion being required and that appropriate indications for each procedure need to be clarified.

A meta-analysis of studies of third-generation TAAs ³¹ found that of 1,830 citations identified only 18 (1,086 patients) met the inclusion criteria: a minimum of 20 patients, a minimum 24-month follow-up, and clinically relevant end-points, such as ankle scoring system, range of motion, complications, and survival rates. Of the 1,086 patients in these studies, 12.5% required secondary surgery and 6.3% required secondary arthrodesis; the weighted five-year prosthesis survival rate averaged 90.6%.