Issue: May 2005
May 01, 2005
14 min read

Experts discuss what to expect after ankle surgery

Patients can expect to return to many of their activities after ankle fusion and total ankle replacement.

Issue: May 2005
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Part I: [How top physicians make the call]

Ankle arthritis is a disease of relatively young and active patients, yet treatments are often time-limited. The emergence of improved joint replacement designs for the ankle has stemmed a renewed enthusiasm to find an alternative to fusion for disabling ankle pain.

In part two of this round table, some experts on ankle arthritis discuss their approach to this challenging clinical problem.

Charles Saltzman, MD


Charles Saltzman, MD [photo]Charles Saltzman, MD
Orthopaedic Surgery
Biomedical Engineering
University of Iowa
Iowa City

Beat Hintermann, MD [photo]

Beat Hintermann, MD
Associate Professor
Orthopaedic University Clinic
University Hospital
Basel, Switzerland

Sandro Giannini, MD [photo]

Sandro Giannini, MD
Full Professor
Clinical Orthopaedics
Department, Sixth Division
University of Bologna,
Rizzoli Orthopaedic Insitute
Bologna, Italy

Mark Myerson, MD [photo]

Mark Myerson, MD
President, American Orthopaedic Foot &
Ankle Society Director, Institute for Foot and Ankle Reconstruction at Mercy Hospital

Michel Bonnin, MD [photo]

Michel Bonnin, MD
Orthopaedic Surgeon
Clinique Sainte Anne
Lyon, France

Michael Coughlin, MD [photo]

Michael Coughlin, MD
Orthopedic Surgeon
Private Practice
Boise, Idaho
Clinical Professor of Orthopedic Surgery
Oregon Health Sciences University,

Saltzman: What should patients expect to be able to do after an ankle fusion or replacement? Are there any restrictions on patient to pursue activity after these surgeries? Are their activity restrictions different?

Hintermann: Successful ankle fusion may allow the patient most activities. However, some activities such as running and jumping, where active plantar flexion and/or dorsiflexion are needed, may be limited despite of shoe modifications. Besides relieving pain, successful total ankle replacement, in contrast, may restore some motion, allowing patients to return to certain physical activities that they were not able to perform prior to implantation. Consequently, the physical demands of patients may increase, and successful primary replacements may fail because the replaced ankle is subjected to too much stress. This is probably one of the biggest concerns I have with total ankle replacement in young patients with high demands.

Sagittal plane motion of a total ankle replacement requires excellent fluoroscopic imaging. In this example, the Salto Implant shows about 22° of dorsiflexion and 25° of plantar flexion.


When comparing fusion with replacement of the ankle, the activity restrictions may be basically similar. There is, however, a totally conversely self-limiting mechanism. The patient with a fused ankle will limit his or her activities to what he or she is able to perform with the help of shoe modifications. For instance, he or she will avoid uphill walking because difficulties with the fixed foot to adapt the ground contact. The patient with a successfully replaced ankle, in contrast, will not feel the restriction of the performed surgery, and thus he or she may tend to subject the replaced ankle to too much stress.

Giannini: Patients should expect to be able to return to a nearly normal life. They can walk, ascend and descend stairs, and perform pretty well all the daily living motor tasks. Light sports like swimming, golf and cycling can also be performed. Certainly they should avoid all more demanding sport activities. When the remaining motion at the other foot joints allows 20º to 25° flexion and no signs of arthritis are evident, ankle arthrodesis still allows patients to perform sports even with high impact, which certainly is not the case when total ankle replacement is performed. Perhaps, when flexion between the forefoot and the shank is achieved and fusion was indicated, more demanding activity can also be performed.

Coughlin: After an ankle fusion I believe that a patient can return to heavy-duty labor, can expect to play sports such as basketball, can jog within the constraints of their arthrodesis and basically proceed with activities of their choice. I do educate them to the fact that increased activity may lead to degenerative arthritis of adjacent joints. Following a total ankle arthroplasty my recommendations are for a patient to be much more careful. Inadvertent fracture of a medial or lateral malleolus, or excess activities may lead to subsidence or failure of the ankle arthroplasty. I think that the patient who has a total ankle arthroplasty must be more judicious in their choice of activities. Thus the activity restrictions following surgery are very different following a total ankle arthroplasty in my hands vs. an ankle arthrodesis.

Myerson: Following ankle replacement, I allow patients to walk, hike, bike, swim, play doubles tennis and cross country ski. My biggest concern following ankle replacement is repetitive impact activities that are likely to lead to implant subsidence.

To some extent their ability to perform these activities is limited not by the prosthesis but by their age and activities prior to implantation. Although impact activities are tolerated and permitted following ankle arthrodesis, with the limitation of ankle dorsiflexion, less physical or sporting activity is likely.

Saltzman: How natural is ankle motion after replacement? What factors play into restoring motion?

Hintermann: Independent of the complexity of the deformity instability or malalignment, the main goal of total ankle arthroplasty remains the restoration of a normal, well-balanced, stable, and aligned ankle and hindfoot, where the ankle prosthesis is implanted to replace destroyed surfaces. It does not make sense to replace an ankle that never worked normally (for example, in clubfoot deformity), or where muscular function is lacking (for example, post-polio foot deformity or a paralytic foot).

Recently, Valderrabano and co-workers investigated, in vitro, the range of motion, movement transfer, and talar movement in the normal ankle, the fused ankle, and the replaced ankle (with three different ankle prostheses: STAR, Hintegra and Agility). Motion at the ankle joint complex was restricted less by each of the three ankle prostheses than by ankle fusion. The prostheses also changed the movement transfer within the ankle joint complex less than ankle fusion, especially during dorsiflexion/plantar flexion of the foot. The two-component ankle (Agility) restricted talar motion within the ankle mortise, whereas the three-component ankles (Hintegra and STAR) seemed to allow talar motion comparable to that in the normal ankle. Using an anatomically shaped talar component (Hintegra) evidenced motion closer to that of the normal ankle than did the use of a non-anatomically shaped talar component (STAR). They concluded that a successful design for total ankle arthroplasty should be shaped as anatomically as possible and should provide a physiological range of motion at the ankle joint, full transmission of movement transfer between foot and lower leg, and unconstrained talar motion within the ankle mortise.

What I have learned over the years is that restoration of the original center of rotation of the talus is essential to get motion close to the normal ankle. Using an anatomically shaped talar component (that is a conical shape with medially a smaller radius than laterally) is also essential to get unconstrained motion. The use of a cylindrically shaped talar component, as is the case for most of the recent designs, potentially overstresses the medial ankle ligaments particularly on posterior aspect that, in turn, provokes painful limitation of dorsiflexion. Ossifications may typically result from this chronic overload of ligaments.

Giannini: Many factors play into restoring ankle motion after surgical treatments, the first being a marked arc of motion before surgery. In many cases limited motion and rigidity affect significantly the status of the soft tissues, which can likely cause limited motion also after surgery. It is important, therefore, to use a prosthesis that can be set to be compatible with the soft tissues retained.

Bonnin: ROM of the ankle is generally improved after TAR but never returns to normal. In our series, the overall ROM as measured from dynamic radiographs improved from 15.2° ± 10° preoperatively to 28.3° ± 7° at follow up (P<.05). Clinical measurements are 22.6° ± 12° and 32.9° ± 13° respectively (P<.05), vs. an average of 46° ± 19° on the controlateral side.

In my experience, the quality of the functional result depends mainly on the capacity of recovering good dorsiflexion: The quality of life is better with a fusion in neutral position than with a stiff TAR in equinus position.

The factors influencing ROM are as follows: (1) the preop ROM; (3) the surgical technique; (2) the way you manage the postop period; and (3) the etiology.

  • In a case of very stiff ankle preoperatively (especially if stiff in equinus) even if a very aggressive release is done during surgery, the ankle remains stiff postoperatively. In these cases, the challenge is to obtain 50 to 10° of dorsiflexion. At the end of the procedure with the testing components, if the dorsiflexion of the ankle doesn’t reach 10° easily (observed on the lateral side of the foot) we must consider few tricks to improve the mobility: Achilles lengthening, tenolysis of posterior tibial tendon, checking with an image intensifier that the AP positioning of the talar component is correct and lastly doing a new distal tibial cut a few millimeters.

  • In the postop period it is critical to maintain immobilization in maximal dorsiflexion otherwise the ankle develops equinus contracture because of the tonicity of the gastrocnemius muscles.

  • In our series the dorsiflexion measured on dynamic XR is better in the RA group (14.6° ± 17°) than in the OA group (9.7° ± 5.6°) (P<.05), although Achilles’ tendon lengthening was less frequent (17% in RA vs. 45% in OA) and preoperative equinus contracture more frequent in the RA group (mean preoperative dorsiflexion was -3.5° ± 10° in the RA group vs. 0° ± 8.2° in the OA group).

Saltzman: Does premature adjacent joint arthritis develop after ankle replacements?

The Hintegra Total Ankle Prosthesis is a nonconstrained, three-component system. This new version uses pegs, thus screw fixation of the talar component is optional. It is manufactured by Newdeal SA in Lyon, France, with permission.


Hintermann: Based on recent reports, there is evidence that ankle replacement protects, to a certain extent, the adjacent joints from further degenerative disease. Personally I was in the beginning too aggressive with contemporary subtalar and/or talonavicular fusion while performing total ankle arthroplasty due to the fear that the observed degenerative changes in these joints may be a source of pain. I then changed my attitude and, today, I limit such additional fusions to severe osteoarthrtitic conditions, severe instability and/or deformities. Most surprisingly, there were less than 1% of my cases where I had to consider a fusion of the subtalar or talonavicular joint after replacement. I thus support the belief that ankle replacement effectively preserves the adjacent joints from degenerative disease. This has some practical relevance.

For example, if severe osteoarthritis developed in a young patient after a trauma, total ankle arthroplasty may allow him or her to regain some motion and to properly bear the affected ankle, which, in turn, may protect the foot from further degenerative joint disease. When the replaced ankle wears out and revision of the prosthesis is no longer possible, ankle arthrodesis remains an option for salvage. The patient, nevertheless, may have benefited greatly from this intermediate solution, which allowed continued participation in some activities (including professional activities) during a significant period of his or her life that would not have been possible with a primary ankle arthrodesis. The intermediate total ankle arthroplasty may additionally have preserved his or her foot from the development of secondary osteoarthritis. Further clinical research is needed, however, to understand the pathologic changes taking place in the neighboring joints of a replaced and fused ankle, respectively, and how they can be influenced.

Bonnin: Arthritis of adjacent joints can be observed after TAR. However, this is mainly in the case of an undiagnosed, preexisting arthritis. This is true particularly in post-traumatic ankle — fracture of the talus — and in rheumatoid cases. In these patients I recommend a careful evaluation of the subtalar and midtarsal joints before considering a TAR. A preop CT scan, MRI or bone scan is often useful. In case of pantalar arthritis a triple arthrodesis can be done before the TAR.

Myerson: I have not identified this problem and believe that the contrary probably occurs. The range of motion after joint replacement is not considerable, and averages about 25º, but with this limited range of motion, this seems sufficient to offload the adjacent joints (subtalar and talonavicular), limiting the development of arthritis.

Saltzman: After ankle fusion is the development of adjacent joint arthritis inevitable? If it occurs, how do you treat it?

Giannini: After ankle fusion, the development of arthritis at the adjacent joints is very likely, though it progresses very slowly. This is true particularly when arthrodesis is at ankle neutral (90°) and rear-foot in slight valgus position. In these cases, the use of insoles, drugs and physical activities can be indicated. In more severe cases, subtalar and mid-tarsal arthrodesis can be considered or even conversion of the arthrodesis into ankle replacement in special cases.

Bonnin: Development of subtalar, mid tarsal or Lisfranc degeneration after ankle fusion is a challenging problem. The risk of this phenomenon can be limited with optimal positioning of the arthrodesis (no equinus) but it seems to be inevitable on long-term evaluations. The risk is particularly high in rheumatoid arthritis and it is the reason why I try to never fuse a rheumatoid ankle.

Treatment is difficult. The widely used attitude in these cases is to fuse the involved joints. The risk of this attitude is to transfer the problem on other joints and to increase the difficulty: When ankle, subtalar and midtarsal joints are fused what is the solution if the Lisfranc joint is involved? In the case of malpositionned arthrodesis, considering a reorientation osteotomy can be a viable option. In two complex cases I have done “desarthrodesis-prosthesis” of the ankle joint with quite satisfactory results.

Coughlin: It has been shown in several studies that following an ankle arthrodesis that adjacent joints will develop arthritis. Some arthritis obviously precedes the ankle arthrodesis. In the case where that has not occurred, there is no question that an ankle arthrodesis puts more stress on adjacent joints. While I do believe that adjacent joint arthritis is inevitable, with an appropriate position of the ankle fusion I believe that this arthritis progresses much more slowly. When it occurs, all the conservative methods such as anti-inflammatory medications and bracing are choices. In more severe cases a hindfoot arthrodesis may be necessary.

Saltzman: How long do total ankles last? When they fail why do they fail?

Hintermann: The question of how long the current ankles last cannot be answered with certainty. Based on most recent reports, a 10-year survivorship of more than 80% is realistic. Although the outcome of ankle replacement may further improve in the future because of improved implants, more reliable instrumentation and increased experience, it may never reach a success over time similar to that after hip or knee replacement. There are several reasons for this as follows:

(1) The ankle is a part of a functional unit, and any pathology involves other joints as well, particularly the subtalar joint.

(2) Malalignment and instability of the talus within the mortise are extremely difficult to be sufficiently addressed, as, again, the neighboring joints might be involved in the pathologic process.

(3) The bone stock may be critical with respect to load resistance and vascularity.

(4) Generally younger patients present with this pathology, thus not only their life expectation but also their activity level are higher.

Probably the biggest source of failures is the use of non-anatomic implants that do not use the whole resection surface for bony support, which may lead to subsidence of component. Non-anatomical surfaces, particularly on the talar side, may provoke ligament stress and thereby intrinsic forces that increase the shear forces at the bone-implant interface, particularly if malalignment has not been properly corrected, and/or if implants, particularly on talar side, have not been properly implanted. The created intrinsic forces may to lead component loosening and polyethylene wear. A final concern in total ankle arthroplasty is the vascularity of the talus below the inserted component over time, particularly in post-traumatic cases were some osteonecrosis may have occurred before component insertion.

Painful post-traumatic osteoarthritis 19 years after an ankle fracture with progressive anterior extrusion of the talus (female, 45 years old) (A-B). While performing ankle replacement, attention was paid to restore the center of rotation of talus, thereby balancing the ligaments in the sagittal plane (C-D, intraoperative view). Minimal bone resection was necessary (E) to insert the implants and tension the ligaments (F).


Bonnin: The survival rate of TAR with long follow-up is variable in the different series of the literature. H. Kitaoka with the Mayo Clinic cemented cylindrical TAR reported 64% at nine years. Kofoed with a cemented STAR reported 72% to 75% at nine years and Wood with a non-cemented STAR reported 88% at 3.8 years. For Kofoed the use of noncemented components improved the longevity and Wood emphasized the improvement due to dual coatings of Ti and calcium phosphate.

In terms of survival rate, the results of the Salto prosthesis seems slightly better than those reported for other three components with a survival rate at last review of 95% to 98%. However, the first implantation was done in January 1997 and we need longer-term evaluation.

Failures in TAR leading to revision can be due to loosening, instability, poly wear or residual pain.

Coughlin: A total ankle may fail for a number of reasons. If it is malaligned at surgery this is probably the most common cause of failure. When it is appropriately aligned it may fail due to subsidence or loosening of the components either at the tibial component interface or the talar component interface. The polyethylene may have early wear. A major concern in my mind is the possibility of polyethylene wear as we have seen in more proximal total joints. A total ankle’s longevity is variable. In a more sedentary older patient, I believe that the longevity is decidedly better than in a younger, heavier, more active patient. That is why a frank discussion about the pros and cons of ankle arthroplasty is most important prior to surgery.

Myerson: I can only base this on my own limited experience of eight years of implantation, and that which has been reported in the literature. This is a common question posed by patients but is difficult to answer. Furthermore, I explain to patients that this is the wrong question to ask, since the longevity of the implant does not correlate with the perceived success, nor their satisfaction with the joint replacement.

There seem to be two phases by which one can determine the success of the surgery: the initial year when the position and alignment of the implant, and syndesmosis arthrodesis are important. Early failure occurs commonly when associated with deformity of the foot, which has not been adequately corrected. The later success is determined by the absence of osteophyte formation and subsidence.

Saltzman: Is it easy to convert a failed TAR to a fusion or revision replacement?

Giannini: It is definitely not difficult to convert failed replacement to a fusion, particularly in a few of the current designs, and revision replacement is very rare at the ankle. In the former case, the leg shortening needs to be discussed with the patient. In the latter case, bone allografts are essential and longer immobilization is necessary (four to six weeks). The bone stock maintenance, with technical difficulties and use of bone graft, revision replacement can also be considered. Revision replacement with proximal fixation stems should be considered.

Bonnin: The conversion of a TAR into fusion can be difficult if the design required major bone resections, if it was cemented or in case of migration of the components with osteolysis. In these difficult cases revision is a challenging technical problem requiring a pantalar arthrodesis. The use of a calcaneo-tibial intramedullary nail or external fixator can be a salvage solution, used in association with massive bone grafting.

Fusion is easier if it was a “third generation” TAR revised before any migration or osteolysis. In these cases, the gap created by the prosthesis is simply filled with iliac crest bone graft and fixation with screws respecting the subtalar joint is generally possible. Fusion is then easy to obtain.

Revision TAR with TAR is an unusual procedure. In my experience it is limited to simple cases such as changing an oversized component or exchanging a mobile component. In case of loosening, instability or osteolysis I prefer revising with fusion.

Coughlin: It is not as difficult as one thinks. Both the Agility and STAR ankle can often be successfully salvaged with a fusion. Likewise many times a failed ankle arthroplasty can be salvaged with a thicker polyethylene component or a revision of either the talar or the tibial components or both.

Myerson: Conversion of a failed replacement to an arthrodesis is a difficult surgery to perform but can be performed without inclusion of the subtalar joint. With the availability of intramedullary fixation for arthrodesis of the tibiotalar and subtalar joints, too many are performed with unnecessary inclusion of the subtalar joint. Arthrodesis must be performed with a structural bone graft, bone stimulation and rigid fixation. Revision ankle replacement is a frustrating and difficult endeavor. The same principles of joint replacement apply to the revision as to the primary procedure, but due to bone loss, instability and deformity, are more daunting procedures.

Dr. Bonnin has a financial interest in a product mentioned in this round table.

For more information:

  • Anderson T, Montgomery F, Carlsson A. Uncemented S.T.A.R. total ankle prosthesis. Three to eight-year follow-up of fifty-one consecutive ankles. J Bone Joint Surg Am. 2003;85:1321-1329.
  • Bonnin M, Judet T, Colombier JA, et al. Mid-term results of the Salto Total Ankle Prosthesis. Clin Orthop. 2000.
  • Hintermann B. (2005) Total ankle replacement. Springer Vienna.
  • Hintermann B, Valderrabano V, Dereymaeker G, Dick W: The HINTEGRA Ankle. Rational and short-term results of 122 consecutive ankles. Clin Orthop. 2003.
  • Kofoed H. Cylindrical cemented ankle arthroplasty: a prospective series with long-term follow-up. Foot Ankle Int. 1995;16:474-479.
  • Kofoed H, Sorensen TS. Ankle arthroplasty for rheumatoid arthritis and osteoarthritis: prospective long-term study of cemented replacements. J Bone Joint Surg Br. 1998;80:328-332.
  • Valderrabano V, Hintermann B, Nigg BM, et al. Kinematic changes after fusion and total replacement of the ankle: part 1: range of motion. Foot Ankle Int. 2003;24:881-887.
  • Valderrabano V, Hintermann B, Nigg BM, Stefanyshyn D, Stergiou P. Kinematic changes after fusion and total replacement of the ankle: part 2: movement transfer. Foot Ankle Int. 2003;24:888-896.
  • Valderrabano V, Hintermann B, Nigg BM, Stefanyshyn D, Stergiou P. Kinematic changes after fusion and total replacement of the ankle: part 3: talar movement. Foot Ankle Int. 2003;24:897-900.
  • Valderrabano V, Hintermann B, Dick W. Scandinavian total ankle replacement: a 3.7 year average follow-up of 65 patients. Clin Orthop. In press.
  • Wood PL, Deakin S. Total ankle replacement. The results in 200 ankles. J Bone Joint Surg Br. 2003;85:334-341.