BlueCross and BlueShield of Montana Medical Policy/Codes
Total Ankle Replacement (TAR)
Chapter: Surgery: Procedures
Current Effective Date: September 24, 2013
Original Effective Date: December 18, 2009
Publish Date: September 24, 2013
Revised Dates: January 5, 2012; November 12, 2012; July 29, 2013

The ankle joint is a comparatively small joint in relation to the weight bearing and torque it must withstand.  These factors have made the design of total ankle joint replacements technically challenging.  The alternative to total ankle replacement (TAR) is arthrodesis, which may lead to alterations in gait and onset of arthrosis in joints adjacent to the fusion.  While both procedures are designed to reduce pain, total ankle replacement is also intended to improve function and reduce stress on adjacent joints.  TAR has been investigated since the 1970s, although during the 1980s the procedure was essentially abandoned due to a high long-term failure rate, both in terms of pain control and function.  Newer artificial ankles have since been developed, which can be broadly subdivided into two design types, fixed-bearing and mobile-bearing.

In the fixed-bearing (cemented) designs, the polyethylene component is locked into the baseplate, which tends to provide a greater stability, but also increases constraint and edge-loading stress at the bone implant interface, which potentially increases the risk of early loosening and failure.  The first fixed-bearing devices were implanted with cement fixation (cement fixation requires more removal of bone).  In 2002, the Food and Drug Administration (FDA) approved the Agility™ Ankle Revision Prosthesis (DuPuy Orthopaedics), which was intended for cemented use only in patients with a failed previous ankle surgery.  In 2005, the FDA reviewed a 510(k) marketing clearance application for the Topez™ Total Ankle Replacement (Topez Orthopedics, Inc., Boulder, Colorado) and determined that it was substantially equivalent to the existing DePuy Agility™ device.  The Topez Ankle is now called the Inbone Total Ankle (INBONE™ Technologies) and is also intended for cemented use only. 

The Agility LP (DuPuy Orthopaedics) and the Eclipse (Kinetikos Medical) received 510(k) marketing clearance in 2006.  The Salto® Talaris (Tornier) received 510(k) marketing clearance in 2006 and 2009.  These semi-constrained cemented prostheses are indicated in patients with end-stage ankle disorders (e.g., affected with severe rheumatoid, post-traumatic, or degenerative arthritis) as an alternative to ankle fusion.

Mobile-bearing systems (non-cemented) have a polyethelene component that is unattached and articulates independently with both the tibial and talar components.  These three-piece mobile-bearing prostheses were designed to reduce constraint and edge loading, but are less stable than fixed-bearing designs and have the potential for dislocation and increased wear.  Mobile-bearing designs have a porous coating to encourage osseo-integration.  These devices include the Scandinavian Total Ankle Replacement (STAR™, Small Bone Innovations) the TNK™ ankle (Kyocera Corporation) and the Buechel-Pappas™ system.

Three-component mobile-bearing systems are Class III devices that are considered under a different regulatory pathway (pre-market approval) than the fixed-component devices described above, which were cleared for marketing under the 510(k) regulatory pathway.  Pre-market approval (PMA) requires demonstration of clinical efficacy in FDA-regulated trials conducted under an investigational device exemption (IDE).  In May 2009, the FDA approved the STAR ankle as an alternative to fusion for replacing an ankle joint deformed by rheumatoid arthritis, primary arthritis or post-traumatic arthritis.  As a condition of the approval, the device manufacturer must continue to evaluate the safety and effectiveness of the device over the next eight years. 

Total ankle replacement has been performed in patients with severe rheumatoid arthritis, severe osteoarthritis, or post-traumatic osteoarthrosis.


Each benefit plan, summary plan description or contract defines which services are covered, which services are excluded, and which services are subject to dollar caps or other limitations, conditions or exclusions.  Members and their providers have the responsibility for consulting the member's benefit plan, summary plan description or contract to determine if there is any exclusion or other benefit limitations applicable to this service or supply.  If there is a discrepancy between a Medical Policy and a member's benefit plan, summary plan description or contract, the benefit plan, summary plan description or contract will govern.

Medically Necessary

Blue Cross and Blue Shield of Montana (BCBSMT) may consider total ankle replacement using a U.S. Food and Drug Administration (FDA) approved replacement device medically necessary in skeletally mature patients with moderate to severe ankle (tibiotalar) pain that limits daily activity and who have the following conditions:

  • Arthritis in adjacent joints (i.e., subtalar or midfoot); OR
  • Severe arthritis of the contralateral ankle; OR
  • Arthrodesis of the contralateral ankle; OR
  • Inflammatory (e.g., rheumatoid) arthritis.

Not Medically Necessary

BCBSMT considers ankle arthroplasty not medically necessary for patients with the following medical conditions:

  • Extensive avascular necrosis of the talar dome;
  • Compromised bone stock or soft tissue (including skin and muscle);
  • Severe malalignment (e.g., > 15 degrees) not correctable by surgery;
  • Active ankle joint infection;
  • Peripheral vascular disease;
  • Charcot neuroarthropathy;
  • Peripheral neuropathy;
  • Ligamentous instability;
  • Subluxation of the talus;
  • History of ankle joint infection;
  • Presence of severe deformities above or beneath the ankle.


BCBSMT considers total ankle replacement experimental, investigational and unproven for all other indications.


The following outcomes are relevant to the analysis of safety and efficacy of total ankle replacement (TAR), compared to ankle arthrodesis, the standard treatment alternative:

  • Resolution of pain;
  • Function of both the ankle and proximal joint in various activities, such as gait walking on flat or irregular surfaces, or walking up stairs, and return to recreational activities;
  • Long-term outcomes including time to revision, and the development of arthritis in the tarsus, knee or hip related to strain on adjacent joints.

For example, if an arthrodesis or ankle replacement is not properly aligned, significant gait abnormalities may result.  In addition, an arthrodesis puts additional strain on proximal joints, which may in turn accelerate the development of arthritis in the knee and hip.  The principal limitations of past total ankle replacement have been loosening of the prosthesis, requiring revision.  If the prosthesis requires removal, the success of a subsequent arthrodesis must be considered.  Different prostheses require different amounts of removal of bone stock, potentially compromising the success of a subsequent arthrodesis.

Numerous reviews have detailed the technical challenges of total ankle replacement, including the evolving prosthetic designs, optimizing biomechanics, and surgical complications.  A 2008 evidence-based review concluded that the current literature on TAR for any prosthesis consists of level IV evidence (case series).  No randomized trials comparing arthrodesis to arthroplasty were identified, and two small, non-randomized studies had conflicting results.  The review indicates that in general, older, thin, low-demand individuals are considered optimal candidates for TAR, but specific criteria have not been established.  Absolute contraindications were listed as extensive avascular necrosis of the talar dome, compromised bone stock or soft tissue, peripheral neuropathy, peripheral vascular disease, and Charcot neuroarthropathy.  Ligamentous instability, subluxation of the talus, and presence of severe deformities above or beneath the ankle were listed as relative contraindications to arthroplasty.  The authors noted that in 2007 the Canadian Orthopaedic Foot and Ankle Society launched a multi-center, randomized case-controlled trial comparing the clinical and radiographic outcomes of the two procedures in order to provide evidence for future clinical decision-making.

One publication on a clinical trial of the STAR prosthesis by Valderrabano and colleagues was identified.  In this report, 68 patients with total ankle replacements using the STAR prosthesis were followed up for an average of 3.7 years.  The authors reported that 35 patients were completely free of pain, 67 ankles were graded as good or excellent by overall clinical score, and scores on the American Orthopedic Foot and Ankle Society hindfoot improved on average from 24.7 points (range, 3-44 points) to 84.3 points at follow-up.  However, there were complications reported including ballooning bone lysis (three patients), periarticular hypertrophic bone formation (43 ankles) with associated decreased flexion, prosthesis problems requiring revision surgery (nine patients), and additional or secondary surgery (14 patients).  Despite complications, results appear encouraging. 

Haddad and colleagues conducted a systematic review and meta-analysis of 10 studies (including two abstracts) on total ankle replacement (852 patients) and 39 studies on ankle arthrodesis (1262 patients).  No studies that directly compared the two procedures were identified.  The patients treated with TAR were older (58 vs. 50 years) and the primary indication was rheumatoid arthritis (39%) whereas post-traumatic arthritis was the primary indication for arthrodesis (57%). The meta-analysis found similar overall scores for the Ankle-Hindfoot Scale (78 for TAR vs. 76 for arthrodesis) and revision rates (7% vs. 9%), although these results are limited due to the quality of the included studies, heterogeneity of results, variability in reporting of outcomes of interest, different populations for the two procedures, and different durations of follow-up. Loosening (28%), wear (15%), and dislocation/migration (11%) were the most frequently reported reasons for revision of ankle arthroplasty (n=46), while revision of ankle arthrodesis (n=66) was predominantly due to nonunion (65%) and infection (26%).  Conversion from arthroplasty to arthrodesis was reported in 5% of 572 patients.  Below-the-knee amputations were reported in one out of 126 (0.8%) patients who had an ankle replacement and 12 out of 242 (5%) patients who had undergone ankle fusion.

SooHoo and colleagues conducted a review of California’s hospital discharge database to compare short- and long-term outcomes of patients who had undergone TAR (n=480) or ankle arthrodesis (n=4705) during a 10-year period (1995-2004).  The type of prosthesis was not specified.  At 90 days, there were more device-related complications (hazard ratio of 2.68) and major revisions (hazard ratio of 3.65) in comparison with arthrodesis.  For example, there were six (1%) major revision procedures by 90 days in the TAR group, including three revision arthroplasties, two implant removals, and two ankle fusions.  In comparison, additional fusion was performed in 16 (0.35%) of the ankle arthrodesis patients.  At five years after surgery, major revision rates were 23% for TAR and 11% for arthrodesis (hazard ratio of 1.93), with reduced survival rates according to Kaplan-Meier analysis.  There was a 2% lower rate of subtalar fusion following TAR compared with ankle arthrodesis (0.7% vs. 2.8%, hazard ratio of 0.28).  Patients treated with ankle fusion were more likely to have lower median income and safety-net insurance, complicated diabetes and osteonecrosis, whereas patients with ankle replacement were more likely to have rheumatoid arthritis.

Prospective controlled trials directly comparing total ankle replacement with the established alternative of fusion are lacking.  Therefore, relevant publications reporting outcomes of ankle arthrodesis and ankle arthroplasty are reviewed below.

Ankle Arthrodesis

Coester et al. reported 22-year follow-up with clinical and radiological evaluation on 23 patients who had ankle arthrodesis for the treatment of painful post-traumatic arthritis of the ankle.  A chart review of arthrodesis procedures at the author’s institution identified 64 patients who met the inclusion/exclusion criteria of isolated ankle arthrodesis, 48 (75%) of whom were located. Thirteen of these 48 patients had died, four (8%) had a below-the-knee amputation, two (4%) had an additional midfoot arthrodesis, and six (13%) declined to participate, resulting in 23 patients included in the follow-up evaluation (range 12 – 44 years).  The mean age of the study group at the time of the operation was 41 years (range, 12 – 70), and 64 years (range 38-89) at follow-up. Twenty-two (96%) of the patients demonstrated a slight-to-moderate limp on clinical evaluation, with no range of motion present in 39% and motion less than half the range of the contralateral side in 57%.  Eleven patients (48%) had tenderness and swelling in the hindfoot and nine (39%) in the midfoot.  Six patients (26%) used a cane and two (9%) used a walker or other assistance for support.  Self-reported questionnaire results indicated more foot pain (38 vs. 11 points), foot disability (47 vs. 15 points) and more severe activity limitation on the ipsilateral than contralateral side (27 vs. 10 points).  Twenty-three patients (96%) reported limitations in vigorous activities and 20 patients (83%) reported difficulty walking more than one mile.  For the uninvolved ankle, most of the patients (87%) had full and painless motion.  Pain did not differ significantly between the ipsilateral and contralateral knee.

Radiographic evaluation of other joints showed more degeneration in the ipsilateral than contralateral foot.  For example, 21 patients had moderate or severe osteoarthritis in the ipsilateral but not contralateral subtalar joint and 13 patients had moderate or severe osteoarthritis in the ipsilateral but not contralateral talonavicular joint.  No differences were found in the level of osteoarthritis in the ipsilateral and contralateral knees.  The effects of ankle fusion on other joints of the foot may be underestimated in this study due to the exclusion of patients who underwent additional procedures.

Buchner and Sabo evaluated long-term outcomes of 48 patients at an average nine years after ankle arthrodesis.  From a cohort of 60 patients who underwent fusion between 1979 and 1997, seven patients were excluded, three died and two were lost to follow-up, leaving 45 patients who had clinical and radiologic evaluation and three patients who responded to questionnaires only. The average age of the patients at the time of surgery was 51 years (range, 20 – 74).  Before surgery, 34 patients (71%) reported severe pain that was almost always present, 12 (25%) reported moderate daily pain and two (4%) had mild occasional pain.  At follow-up, 10 (21%) patients reported moderate-to-severe pain and 38 (79%) reported mild-to-no pain.  The visual analog scale (VAS) for pain improved from an average of 8.8 before surgery to 3.0 at follow-up. Nine patients (19%) had revision surgery due to infection (n=4), non-union (n=4), and malposition (n=1).  Clinical evaluation at nine years (range, 3 – 21) revealed that four patients (8%) had a marked gait abnormality and limp, 26 (54%) had some gait abnormality, and 18 patients (38%) had no abnormality while walking.  The average postoperative score on the American Orthopaedic Foot and Ankle Society (AOFAS) ankle and hindfoot scale was 74 out of 100.  Thirty-four patients (73%) scored as good-to-excellent and 13 (27%) scored as fair-to-poor. Arthrosis in the subtalar joint was severe in four, moderate in 17 and mild in 17.  The average tarsal mobility of the surgically treated foot was 54% of the contralateral side, and restriction of tarsal mobility was correlated with worse clinical outcome.

Another study reported average seven-year follow-up (range, 2 – 15) on 42 patients who underwent arthrodesis for primary or secondary osteoarthritis of the ankle.  Out of 48 patients treated between 1979 and 1995, three patients died and three did not return for clinical evaluation or radiography, resulting in an 88% follow-up.  The average age of the patients at the time of surgery was 58 years (range, 25 – 79).  The clinical score improved from 54 to 78 points (out of 100), with the pain subscore improving from 18 to 35 (out of 40).  No association was found between postoperative pain and range of motion.  Non-union was detected in three ankles (7%).  Degenerative arthritis developed and advanced in the subtalar joint in 33% of the patients.  The severity of arthritis in the subtalar and Chopart joints was exacerbated if patients had arthritis before surgery.  Based on these findings, the authors concluded that a treatment method that allows mobility of the ankle, such as total ankle arthroplasty, is indicated for patients in whom degenerative changes are detected in adjoining joints before surgery.

The literature indicates that treatment of a painful arthritic joint with arthrodesis can significantly reduce pain.  However, non-union and malposition may require additional surgery, and as many as 5% of patients have been reported to choose amputation due to continuing pain or loss of function.  With longer-term follow-up, increasing foot pain and degenerative changes in adjoining joints have been observed.  These longer term changes are associated with reduced range of motion in the fused ankle joint, and have been shown to be most severe in patients with pre-existing osteoarthritis of the subtalar joint.

Fixed-Bearing Total Ankle Replacement

Spirt and colleagues reported outcomes from 306 consecutive TARs (303 patients) fitted with the Agility ankle system had the surgical procedure performed between 1995 and 2001.  The majority of the patients had post-traumatic osteoarthrosis (65%) or primary osteoarthrosis (25%), and had an average age of 54 years (range of 19-85).  Loosening of the talar component was observed in 22 joints (7%).  At an average 33 months follow-up, 40% of cases had required reoperation and 33 TARs (11%) were considered to have failed.  The five-year implant survival rate was 80%.  Age at the time of the primary total ankle arthroplasty was the only covariate related to the rate of reoperation and failure, with each one-year increase in age associated with a 3.5% decrease in the hazard of failure.  Another case series of 100 consecutive total ankle replacements with the Agility Ankle, implanted between 1984 and 1993, reported follow-up of 2-12 years.  Patients were evaluated with an interview focusing on pain and activities of daily living, and clinical and radiologic examination.  Of the 85 ankles in 83 patients that were available for follow-up, 98% were associated with some level of pain relief.  A total of 74% of patients reported an increase in their functional level.  Based on radiologic exam, 36% of these prostheses were associated with a delayed union or nonunion.  Migration of talar or tibial components of the prosthesis were also noted; migration of the tibial component was associated with nonunion.  Nonunion was associated with ballooning lysis at the interface between the bone and tibial component, although lysis was also seen in cases when a solid union was present.  The authors conclude that these intermediate results are encouraging, although the radiographic findings created concerns about long-term outcomes.  Another case series of 86 cases has been published and reported similar results.  While 79 of the 86 cases (92%) reported a favorable outcome, there were similar radiographic findings.  A total of 22% of prosthetic components had migrated, and 8 of the 12 tibial components that had migrated involved a delayed union or nonunion.  Kopp and colleagues reported minimum two-year follow-up (range of 26-64 months) on 43 consecutive ankle replacements with the Agility prosthesis; two patients were lost to follow-up and one patient required revision due to aseptic loosening.  Pain was reported to have improved in all patients, rated postoperatively as “none” in 16 patients, “mild, occasional” in 21 patients, and “moderate, daily” in three patients.  Twelve perioperative and 12 postoperative complications occurred (60%), requiring additional operative procedures.  The authors note that the high rate of complications and need for reoperation is consistent with other reports on the Agility prosthesis, but most of the complications can be adequately treated.  Radiolucency or lysis was noted at follow-up in 34 of 40 ankles, and migration or subsidence of components was noted in 18.  The authors concluded that, “the overall intermediate-term clinical results of total ankle replacement using the Agility prosthesis are promising, but the longevity of the prosthesis is questionable because of the frequency of periprosthetic lucency, lysis, and component subsidence.”

Mobile-Bearing Total Ankle Replacement

Wood et al. reported mid-term outcomes from 200 patients who had been randomized to receive one of two mobile-bearing ankle replacement systems (STAR or Buechal-Pappas) between 2000 and 2003.  The mean follow-up (date last seen for surviving ankles or for failure) was 49 months, with a range of 1-85 months.  At the time of follow-up, 163 implants had survived, 21 patients had died, and 16 (8%) implants had failed (12 Buechal-Pappas and 4 STAR).  These were treated with fusion (n=14) or revision (n=2).  There was a trend toward higher failure with the Buechal-Pappas ankle compared to the STAR (p=0.09), with a hazard ratio of 2.7.  The presence of a varus or valgus deformity before surgery was associated with failure for either prosthesis, with a hazard ratio of 1.64 for every five-degree increment in deformity.  Edge-loading was observed in 12 Buechal-Pappas and six STAR prostheses, 39% of which were subsequently revised.  A patient who had a varus or valgus deformity of 15 degrees or more had a 6.5 greater likelihood of developing edge-loading than if the ankle was well aligned before surgery.  Pain and function, as measured by the American Orthopaedic Foot and Ankle Society (AOFAS) hindfoot score, improved to a similar extent in the two groups.  The study found that few patients (less than 20%) had marked increases (10 degrees or more) in range of ankle movement with either prosthesis.  Results were not compared with arthrodesis.

The pivotal trial for the STAR prosthesis, reported to the FDA in 2007, consisted of a two-year non-inferiority design with 158 patients from 10 sites treated with arthroplasty and 66 patients from five additional sites treated with arthrodesis.  Patients were included if they had primary ankle arthritis, posttraumatic arthritis, or rheumatoid arthritis, moderate to severe pain (Buechal-Pappas pain score of 20 or less), loss of mobility and function (total Buechal-Pappas score of less than 50 out of 100), and a minimum of six months of conservative treatment including a three-month trial of orthosis and/or analgesic medication.  Exclusion criteria included hindfoot or forefoot malalignment, avascular necrosis, severe osteopenia or inadequate bone stock, insufficient ligament support, neuropathy, or neuromuscular impairment.  There were no differences between groups in the operative time, estimated blood loss, or length of stay.  In the STAR arm, 142 patients (90%) completed the 24-month follow-up; three patients died and two were transferred to a bilateral treatment study.  Only 78% completed 24-month follow-up in the arthrodesis arm due to non-compliance by patients and investigators.  The average total Buechal-Pappas score increased from 41 to 82 in the STAR group and from 43 to 70 in the arthrodesis arm, achieving non-inferiority for this outcome.  Statistical superiority was driven primarily by the improvement in range of motion (ROM), with slight improvements in deformity (increased by 1.9 vs. 0.4 for arthrodesis) and function (increased by 13.4 vs. 9.7 for arthrodesis).  Safety success was achieved in fewer STAR patients (71%) than arthrodesis patients (83%).  Major adverse events were reported in 9% of STAR patients and 1.5% of controls.  Implant-related adverse events included bone fracture (18%), bony changes (8%) nerve injury (20%), soft tissue edema (16%) decreased ROM (6%) and wound problems (20%).  Painful adverse events were similar in the two groups (44% for STAR and 49% for arthrodesis).  Surgical instrumentation and technique were modified during the study to address the wound problems and sensory loss from damage to a branch of the peroneal nerve.  Approximately 450 additional patients have been enrolled in a FDA-approved multi-center continued access registry from the same sites that had participated in the pivotal study.  The STAR prosthesis received final FDA approval in 2009.

The largest consecutive series of 200 implants (184 patients) with the STAR prosthesis was reported by Wood and colleagues.  Eighty-one patients had osteoarthritis (25 fracture-related) and 119 had inflammatory joint disease (112 with rheumatoid arthritis).  The cumulative five-year survival rate was 93%, and the 10-year survival rate was 80%.  Twenty-four ankles (12%) failed at a mean of 48 months (range of 1-108); five had delayed wound healing, nine sustained an intraoperative malleolar fracture, and 10 sustained a postoperative fracture within 12 months of the index procedure.  The authors suggested that survivorship figures are similar to those of early reports of total knee replacement when techniques and designs were being developed.  Valderrabano et al. reported outcomes of 68 patients with total ankle replacements using the STAR prosthesis and followed up for an average of 3.7 years.  The authors reported that 35 patients (51%) were completely free of pain, 67 ankles were graded as good or excellent by overall clinical score, and scores on the AOFAS hindfoot score improved on average from 24.7 points (range, 3-44 points) to 84.3 points at follow-up.  Complications included ballooning bone lysis (three patients), periarticular hypertrophic bone formation (43 ankles) with associated decreased flexion, prosthesis problems requiring revision surgery (nine patients), and additional or secondary surgery (14 patients).  A Swedish group reported median 52 month follow-up on 51 consecutive ankle replacements with the STAR prosthesis implanted between 1993 and 1999.  Twelve ankles (24%) had to be revised and another 16% showed radiographic signs of loosening. The estimated five-year survival rate was 70%, with significant improvement in survival after the first 20 cases were performed.

Fevang et al. reported analysis of 257 primary ankle replacements from the Norwegian arthroplasty registry between 1994 and 2006, with 82% of all TARs being registered.  The cement-less STAR prosthesis was used in 82% of the cases; 14% were with the fixed-bearing Norwegian TPR prosthesis.  Median follow-up was 9.2 years for the TPR and 3.3 years for the STAR.  Revision was performed in 27(11%) of the cases, with a mean time until revision of 2.3 years.  Six revisions (19% of 32) were from the TPR prosthesis and 21 (10% of 216) with the STAR.  Survival of the single-coated STAR prosthesis (used until 1999) was significantly lower than survival of the double-coated STAR prosthesis (used from 2000 to 2006).  The five-year survival for all prostheses was 89%, with a 10-year survival of 76%, rates that that the authors considered low compared to survival of hip and knee prostheses in Norway.

Buechel, Buechel and Pappas reported on 2- to 20-year follow-up on two versions (shallow-sulcus and deep-sulcus) of the device that bears their names.  The shallow-sulcus study (n=40) was conducted for an FDA clinical trial, the deep-sulcus study patients (n=75) were treated under an exemption from premarket notification.  The authors reported 20-year survivorship of 74% for the shallow-sulcus design and 12-year survival of 92% for the deep-sulcus design.  However, Kaplan-Meier survival curves show that 74% survival was reached after 13 years, while 92% survival with the deep-sulcus design was reached after six years.  Thus, these results may be limited by loss to follow-up.  Doets et al. reported a prospective observational study of 93 total ankle arthroplasties with two mobile bearing designs (80% Buechal-Pappas devices, performed between 1988 and 1999) in patients with inflammatory arthritis.  Seventeen patients died due to unrelated causes during follow-up (21 ankles, 5 to 171 months after surgery), and 15 ankles (16%) were revised; no other patients were lost to follow-up.  Early postoperative complications included malleolar (n=20) or tibial (n=11) fracture or wound healing disturbance or infection (n=11).  In most cases, the fractures were not considered to have affected long-term outcomes.  At one year after surgery, significant improvements were observed in low contact stress (36 pre-surgery versus 82 post-surgery), the AOFAS score (27 vs. 78) and the Kofoed ankle score (27 vs. 74).  Analysis of radiographs revealed complete osseo integration of the talar component in 50 ankles; all but nine ankles had partial or complete radiolucent lines.  With final follow-up ranging from 0.4 to 16 years (average of 7.6), the eight-year overall survival rate was 84% (eight-year average survival, range of 5-16 years).  San Giovanni et al. reported results from 31 consecutive ankle replacements with the Buechal-Pappas prosthesis in 23 patients with rheumatoid arthritis.  Fracture of the medial malleolus was the most common complication, occurring in 32% of the cases.  Overall survivorship at eight years was estimated to be 93%. Seventy-five percent of the patients rated their post-implant pain as “none,” 21% as “mild,” and 4% as “moderate.”  Radiographic analysis revealed stable, well-positioned implants with biologic ingrowth in 82% of the ankles.  The remaining 18% were interpreted as being at risk for impending failure.  The Buechal-Pappas TAR system is no longer available for use in the United States.

Kofoed and colleagues reported 1-15 years of follow-up on a prospective series of 100 consecutive cases treated with ankle arthroplasty between 1981 and 1996.  From 1981 to 1985, a two-component device was used, and from 1986 to 1996, a three-component prosthesis was used. Patients were divided into two groups for analysis, those who were younger than 50 years of age at implantation (30 ankles, median age 46, range, 22-49 years) and those 50 years of age or older (70 ankles, median age 63, range 51-83 years).  In the patients younger than 50 years of age, four cases (13%) were revised and three (10%) were converted to arthrodesis at a median of five years (range, 5-9).  For the older group, four cases (6%) were revised and four (6%) were converted to arthrodesis after a median of 5.5 years (range, 2-8).  Estimated survival at 15 years was 75% in patients younger than 50 (confidence interval of 63-87%) and 81% for patients 50 or older (confidence interval of 74-87%).  This difference was not significant.


The established standard for the painful arthritic ankle is fusion, which usually results in a pain-free but rigid ankle in the short term.  Complications associated with ankle fusion are non-union, an increase in arthrosis and pain in adjoining joints, and not-uncommonly, amputation.  For specific conditions, including presence of bilateral, subtalar or midfoot arthritis, fusion is not indicated.  Total ankle replacement is in the development stage with designs that are continuing to evolve.  Although long-term evidence is lacking, short-term results suggest similar improvements in pain and function in comparison with arthrodesis, and mid-term results indicate 75-80% survival at 10-15 years. 

2011 Update

A search of peer-reviewed literature was conducted through November 2011.  Following is a summary of the key literature to date

Krause et al. reported a comparison of complications between total ankle replacement (TAR) and arthrodesis in 2011.  From February 2002 through August 2007, data collected from 516 patients following TAR or ankle arthrodesis were entered into the database.  Indications for ankle arthrodesis were severe deformity and instability, poor ankle motion, no or mild adjacent joint arthritis, and younger age.  Indications for TAR were older age, severe adjacent joint arthritis, a diagnosis of rheumatoid arthritis, and no or only mild deformity or instability.  Patient preference was also a factor.  A total of 114 TARs and 47 ankle fusions met the inclusion criteria for the study, with a complete data set and minimum two-year follow-up.  Sixty-one of the TARs were performed with the fixed-bearing Agility prosthesis, while the remaining 53 were performed with one of three types of mobile-bearing prostheses (HINTEGRA, STAR, and Mobility).  The mean age was 64 years for the patients who underwent TAR and 59 years for the patients who underwent arthrodesis.  The validated self-administered Ankle Osteoarthritis Scale (AOS) was used to evaluate all patients at 6, 12, and 24 months and 3.5, 5, and 10 years postoperatively. Radiographic evaluations were performed at a mean of 39 months following TAR and 37 months following arthrodesis.  Both groups had significant improvement in the validated AOS (30.9 points for TAR and 30.6 for arthrodesis, p<0.001 for pre-/post-comparison).  There were significantly more complications following TAR than ankle arthrodesis (54% vs. 21%, respectively, p=0.003).  Aseptic loosening occurred in 17 (15%) of the 114 TARs, and 11 of these had revision surgery.  A technical error occurred in 17 (15%) of the TARs, which included lateral gutter impingement, excessive polyethylene wear or breakage, and malalignment.  There were eight (7%) intraoperative fractures, which were treated during the index operation, and seven cases (6%) of deep infection.  The highest rate of complications were reported for the Agility prosthesis (61%), followed by the Mobility (47%), the STAR (44%), and the HINTEGRA (18%).  Complications in patients treated with arthrodesis included adjacent joint arthritis (6%), non-union (4%), and technical error (2%).  Other complications (9%) included medial-gutter-related discomfort and nonspecific ongoing pain.  For both groups, there was a significant impact of major complications on the AOS outcome score.  The comparison of adverse rates between groups is limited by differences in the patient populations selected for each procedure.

Gougoulias and colleagues published a 2010 systematic review on outcomes from ankle replacement.  Thirteen level IV (case series) studies were published between 2003 and 2008 that included at least 20 subjects and had at least two years of follow-up.  The studies included a total of 1,105 total ankle replacements (TARs) (including 234 Agility, 344 STAR, 153 Buechel-Pappas, 152 HINTEGRA, 70 TNK, and 54 Mobility).  The failure rate, with revision, arthrodesis, or amputation as an endpoint, was 9.8%, with a weighted follow-up of 5.2 years.  The available evidence was insufficient to determine superiority of any implant design over another.  Studies of fixed-bearing and mobile-bearing devices are described below.

In 2009, Jensen and Linde reported follow-up of up to 23 years for 26 patients (33 ankles) with rheumatoid arthritis who had received a TAR between 1980 and 1993.  The median age of the patients was 60.5 years (range: 31 to 75 years) at the time of surgery.  At the latest follow-up, prostheses in four patients had been removed (15%, 4-13 years after implantation); two patients with three prostheses were alive at 23 years after surgery.  Two patients had received amputation due to unrelated causes, and the remaining 18 patients had died with the prosthesis in place (median 9.5 years after TAR, range: 0.5 to 23.3 years).  Survival based on radiographic loosening was 64% at 10 years, while the prosthesis survival rate was 85% at 10 years.

The STAR prosthesis received final FDA approval in 2009.  Results from this trial, and from 435 patients enrolled in the FDA-regulated, multicenter, continued-access registry, were published in 2009. 

Prospective medium- to long-term follow-up on 84 ankles implanted with the STAR prosthesis in the U.S. was reported by Mann et al. in 2011.  Included were patients with primary (25%), post-traumatic (56%), or inflammatory (18%) arthritis who had failed conservative treatment. Contraindications for TAR were a history of deep infection, severe uncorrectable deformity, paralysis, neuropathy, severe peripheral vascular disease, poor skin coverage, and morbid obesity.  The average age at implantation was 61 years (range 33-86 years), and the average body mass index (BMI) was 28.8 kg/m2.  Forty percent of patients met the definition for obesity (BMI > 30 kg/m2).  At an average follow-up of 9.1 years (range, 2.6 to 11 years), the AOFAS ankle-hindfoot score improved an average 39 points, from a mean of 43 to 82 points at latest follow-up. The gain was primarily due to an improvement in pain, from a mean of seven at baseline to 33 at follow-up (40-point scale).  Ninety-one percent of the prostheses remained implanted at the time of latest follow-up (or death).  The probability of implant survival was calculated to be 96% at five years and 90% at 10 years.  In total, 14% of patients had component angulation or subsidence, and 17% of patients underwent additional surgical procedures for complications, including osteolysis around the tibial or talar components.  At final follow-up, 88% of joints adjacent to the STAR prosthesis had no progression of arthritis.

Mid-term follow-up from large case series (from 96 patients in each study) have been reported for the three-component mobile-bearing Salto and Mobility prostheses.  Two smaller case series have reported a high rate of osteolysis with the Ankle Evolutive System (AES) total ankle in mid-term follow-up.  Outside of an investigational device exemption (FDA-regulated) trial for the Mobility total ankle system, these devices are not available for use in the U.S.

Practice Guidelines and Position Statements

In 2009, the AOFAS issued the following position statement on total ankle arthroplasty:  “Over the past decade, total ankle replacement surgery has evolved as an acceptable and viable alternative to ankle arthrodesis in select patients with ankle arthritis.  These include adult patients with primary, post-traumatic, and rheumatoid arthritis who have moderate or severe pain, loss of mobility, and loss of function of the involved ankle.  Before considering total ankle replacement, patients should have completed several months of conservative treatment, should have satisfactory vascular perfusion in the involved extremity, and must have adequate soft-tissue coverage about the ankle that affords a safe surgical approach to total ankle replacement.  In such patients, high-level evidence indicates that total ankle replacement safely relieves pain and may provide superior functional results when compared to ankle fusion.  Additional concomitant or sequential surgical procedures may be required in some patients to optimize outcome.”

A search of peer-reviewed literature through November 2011 identified no new clinical trial publications or any additional information that would change the coverage position of this medical policy.


Disclaimer for coding information on Medical Policies           

Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy.  They may not be all-inclusive.           

The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers.  Only the written coverage position in a medical policy should be used for such determinations.           

Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps. 

ICD-9 Codes
81.56, 357.1, 359.6, 443.9, 446.0, 710.0-710.2, 714.0, 715.17, 715.97, 718.57, 730.27, 731.3, 824.9
ICD-10 Codes
G63- G72.49, i73.9, M05.59, M19.071 - M19.079- M19.90, M24.671 - M24.676   - M35.03, M30.0 - M31.7, M32.0 - M32.9, M34.0 - M34.9, M35.00 - M35.09, M05.40 - M05.719, M86.9 - M86.9, M89.70 - M89.79 
Procedural Codes: 27702, 27703, 27704
  1. Pyevich, M.T., Saltzman, C.L., et al.  Total ankle arthroplasty: a unique design.  Two to twelve-year follow-up.  J Bone Joint Surg Am (1998) 80(10):1410-20.
  2. Kofoed, H., and A. Lundberg-Jensen.  Ankle arthroplasty in patients younger and older than 50 years: a prospective series with long-term follow-up.  Foot Ankle Int (1999) 20(8):501-6.
  3. Takakura, Y., Tanaka, Y., et al.  Long-term results of arthrodesis for osteoarthritis of the ankle. Clin Orthop Relat Res (1999) (361):178-85.
  4. Conti, S.F., Bisignani, G., et al.  Update on total ankle replacement.  Semin Arthroplasty Reconstruct Foot Ankle (1999) 10:62-71.
  5. Coester, L.M., Saltzman, C.L, et al.  Long-term results following ankle arthrodesis for post-traumatic arthritis.  J Bone Joint Surg Am (2001) 83-A (2):219-28.
  6. Buchner, M., and D. Sabo.  Ankle fusion attributable to posttraumatic arthrosis: a long-term follow-up of 48 patients.  Clin Orthop Relat Res (2003) (406):155-64.
  7. Buechel, F.F. Sr., Buechel, F.F. Jr., et al.  Ten-year evaluation of cementless Buechel-Pappas meniscal bearing total ankle replacement.  Foot Ankle Int (2003) 24(6):462-72.
  8. American Orthopaedic Foot and Ankle Society (AOFAS).  Position statement on total ankle arthroplasty. (2003 June 6) (available at: ). (accessed April 2009).
  9. Wood, P, L, R., and S. Deakin. Total ankle replacement.  The results in 200 ankles.  J Bone Joint Surg Br (2003) 85(3):334-41.
  10. Anderson, T., Montgomery, F., et al.   Uncemented STAR total ankle prosthesis.  Three to eight-year follow-up of fifty-one consecutive ankles.  J Bone Joint Surg Am (2003) 85-A (7):1321-9.
  11. Valderrabano, V., Hintermann, B.  Scandinavian total ankle replacement: a 3.7-year average follow up of 65 patients.  Clin Orthop Relat Res (2004) (424):47-56.
  12. Buechel, F.F. Sr., Buechel, F.F. Jr.  Twenty-year evaluation of cementless mobile-bearing total ankle replacements.  Clin Orthop Relat Res (2004) (424):19-26.
  13. Spirt, A.A., Assal, M., et al.  Complications and failure after total ankle arthroplasty.  J Bone Joint Surg Am (2004) 86-A (6):1172-8.
  14. Kopp, F.J., Patel, M.M., et al.  Total ankle arthroplasty with the Agility prosthesis: clinical and radiographic evaluation.  Foot Ankle Int (2006) 27(2):97-103.
  15. Doets, H.C., Brand, R., et al.  Total ankle arthroplasty in inflammatory joint disease with use of two mobile-bearing designs.  J Bone Joint Surg Am (2006) 88(6):1272-84.
  16. San Giovanni, T.P., Keblish, D.J., et al.  Eight-year results of a minimally constrained total ankle arthroplasty.  Foot Ankle Int (2006) 27(6):418-26.
  17. Haddad, S.L., Coetzee, J.C., et al.  Intermediate and long-term outcomes of total ankle arthroplasty and ankle arthrodesis.  A systematic review of the literature.  J Bone Joint Surg Am (2007) 89(9):1899-905.
  18. U.S. Food and Drug Administration.  STAR Ankle Premarket Approval Panel Meeting Presentation 4-24-07.  (2007).  Available online at:  accessed November 2011.
  19. SooHoo, N.F., Zingmond, D.S., et al.  Comparison of reoperation rates following ankle arthrodesis and total ankle arthroplasty.  J Bone Joint Surg Am (2007) 89(10):2143-9.
  20. Fevang, B.T., Lie, S.A., et al.  257 ankle arthroplasties performed in Norway between 1994 and 2005.  Acta Orthop (2007) 78(5):575-83.
  21. STAR Ankle Premarket Approval Panel Meeting Presentation 4-24-07.  Available at: (accessed 2008 December).
  22. Guyer, A.J., and G. Richardson.  Current concepts review: total ankle arthroplasty.  Foot Ankle Int (2008) 29(2):256-64.
  23. Wood, P.L., Prem, H., et al.  Total ankle replacement: medium-term results in 200 Scandinavian total ankle replacements. J Bone Joint Surg Br (2008) 90(5):605-9.
  24. Glazebrook, M.A., Arsenault, K., et al.  Evidence-based classification of complications in total ankle arthroplasty.  Foot Ankle Int (2009) 30(10):945-9.
  25. Jensen, N.C., and F. Linde.  Long-term follow-up on 33 TPR ankle joint replacements in 26 patients with rheumatoid arthritis.  Foot Ankle Surg (2009) 15(3):123-6.
  26. Saltzman, C.L., Mann, R.A., et al.  Prospective controlled trial of STAR total ankle replacement versus ankle fusion: initial results.  Foot Ankle Int (2009) 30(7):579-96.
  27. Wood, P.L., Sutton, C., et al.  A randomised, controlled trial of two mobile-bearing total ankle replacements.  J Bone Joint Surg Br (2009) 91(1):69-74.
  28. Total Ankle Replacement.  Chicago, Illinois: Blue Cross Blue Shield Association Medical Policy Reference Manual (2009 September) Surgery 7.01.77.
  29. Gougoulias, N., Khanna, A., et al.  How successful are current ankle replacements?  A systematic review of the literature.  Clin Orthop Relat Res (2010) 468(1):199-208.
  30. Coetzee, J.C., and J.K. Deorio.  Total ankle replacement systems available in the United States.  Instr Course Lect (2010) 59:367-74.
  31. Wood, P.L., Karski, M.T., et al.  Total ankle replacement: the results of 100 mobility total ankle replacements.  J Bone Joint Surg Br (2010) 92(7):958-62.
  32. Rodriguez, D., Bevernage, B.D., et al.  Medium term follow-up of the AES ankle prosthesis: High rate of asymptomatic osteolysis.  Foot Ankle Surg (2010) 16(2):54-60.
  33. Bonnin, M., Gaudot, F., et al.  The Salto Total Ankle Arthroplasty: Survivorship and Analysis of Failures at 7 to 11 years. Clin Orthop Relat Res (2010) 469(1):225-36.
  34. Mann, J.A., Mann, R.A., et al.  STAR Ankle: long-term results.  Foot Ankle Int (2011) 32(5):473-84.
  35. American Orthopaedic Foot and Ankle Society (AOFAS).  Position statement: Total ankle replacement surgery. (2009).  Available online at:  accessed 2011 November.
  36. Krause, F.G., Windolf, M., et al.  Impact of complications in total ankle replacement and ankle arthrodesis analyzed with a validated outcome measurement.  J Bone Joint Surg Am (2011) 93(9):830-9.
  37. Kokkonen, A., Ikavalko, M., et al.  High rate of osteolytic lesions in medium-term followup after the AES total ankle replacement.  Foot Ankle Int (2011) 32(2):168-75.
  38. Total Ankle Replacement.  Chicago, Illinois: Blue Cross Blue Shield Association Medical Policy Reference Manual (2011 September) Surgery 7.01.77.
January 2012  Policy updated with literature review through July 2011; references added and reordered; Added MN criteria to MN, changed from NMN to investigational statement.
November 2012 Policy updated with literature review through July 2012; references added and reordered; policy statements unchanged.
July 2013 Policy formatting and language revised.  Policy statement unchanged.  Title changed from "Ankle Replacement" to "Total Ankle Replacement (TAR)".  Added CPT code 27704.
®Registered marks of the Blue Cross and Blue Shield Association, an association of independent Blue Cross and Blue Shield Plans. ®LIVE SMART. LIVE HEALTHY. is a registered mark of BCBSMT, an independent licensee of the Blue Cross and Blue Shield Association, serving the residents and businesses of Montana.
CPT codes, descriptions and material only are copyrighted by the American Medical Association. All Rights Reserved. No fee schedules, basic units, relative values or related listings are included in CPT. The AMA assumes no liability for the data contained herein. Applicable FARS/DFARS Restrictions Apply to Government Use. CPT only © American Medical Association.
Total Ankle Replacement (TAR)