Mark E. Easley
DEFINITION
The procedure to fuse the tibiotalar joint for isolated endstage tibiotalar arthrosis
ANATOMY
Ankle
Tibial plafond with medial malleolus
Articulations with dorsal and medial talus
In sagittal plane, slight posterior slope
In coronal plane, articular surface is 88 to 92 degrees relative to lateral tibial shaft axis.
Fibula
Articulation with lateral talus
Responsible for one sixth of axial load distribution of the ankle
Talus
60% of surface area covered by articular cartilage
Dual radius of curvature
Distal tibiofibular syndesmosis
Anterior inferior tibiofibular ligament
Interosseous membrane
Posterior tibiofibular ligament
Ankle functions as part of the ankle–hindfoot complex much like a mitered hinge.
PATHOGENESIS
Posttraumatic arthrosis
Most common cause
Intra-articular fracture
Ankle fracture-dislocation with malunion
Chronic ankle instability
Primary osteoarthrosis
Relatively rare compared to hip and knee arthrosis
Inflammatory arthropathy
Most commonly rheumatoid arthritis
Other
Hemochromatosis
Pigmented villonodular synovitis
Charcot neuroarthropathy
Septic arthritis
NATURAL HISTORY
Posttraumatic arthrosis
Malunion, chronic instability, intra-articular cartilage damage, or malalignment may lead to progressive articular cartilage wear.
Chronic lateral ankle instability may eventually be associated with:
Relative anterior subluxation of the talus
Varus tilt of the talus within the ankle mortise
Hindfoot varus position
Primary osteoarthrosis of the ankle is rare and poorly understood.
Inflammatory arthropathy
Progressive and proliferative synovial erosive changes failing to respond to medical management
May be associated with chronic posterior tibial tendinopathy and progressive valgus hindfoot deformity, eventual valgus tilt to the talus within the ankle mortise, potential lateral malleolar stress fracture, and compensatory forefoot varus
PATIENT HISTORY AND PHYSICAL FINDINGS
History
Typically, history of trauma to the ankle
Intra-articular ankle fracture (bior tri-malleolar ankle fracture; tibial plafond [pilon] fracture)
Chronic ankle instability
Inflammatory arthropathy
Primary ankle arthritis
Symptoms and complaints
Pain in anterior ankle with weight bearing and particularly with forced dorsiflexion
Often relieved by rest, but patient may have pain even at rest after vigorous activity or prolonged standing
Ankle swelling
Ankle stiffness
Medications
If patient is taking anti-inflammatory agents, these will need to be stopped preoperatively to limit the risk of perioperative bleeding.
Rheumatoid medications; may need to be stopped perioperatively to optimize wound and bone healing
Physical examination
Alignment
Ipsilateral limb alignment (not simply ankle alignment). The surgeon should examine the lower extremity from the hip to the foot. Optimal limb alignment is essential for the ankle arthrodesis to function well. Any ability for the lower limb to compensate for malalignment through the ankle is forfeited with ankle arthrodesis.
Ankle–foot alignment
The ankle functions as part of an ankle–subtalar joint complex.
Ankle fusion must be positioned on a sufficiently supportive and plantigrade foot.
Hindfoot, midfoot, and even forefoot malalignment may need to be addressed simultaneous to or staged with ankle arthrodesis.
Range of motion (ROM)
Ankle ROM is not critical since the ankle will be stiff following arthrodesis.
Hindfoot ROM is essential for successful ankle arthrodesis. A stiff hindfoot and fused ankle allows very little accommodation and functions as a tibiotalocalcaneal or even pan-talar arthrodesis. Ankle arthritis associated with hindfoot stiffness, particularly if due to hindfoot arthritis, may be better treated with total ankle arthroplasty (TAA).
Soft tissues
An intact, relatively healthy soft tissue envelope surrounding the ankle is less likely to have soft tissue complications postoperatively, provided careful soft tissue handling is maintained.
Previous surgical scars must be considered. Either they can be incorporated into the surgical approach or the surgical approach may be modified to limit postoperative wound complications.
Vascular status: Intact pulses and satisfactory refill must be confirmed; if not, a Doppler ultrasound or noninvasive vascular studies must be performed before considering surgery.
Neurologic status: A peripheral neuropathy is a relative contraindication for TAA; in our opinion, well-controlled diabetes without neuropathy is not. However, if there is any question about risks, then arthrodesis should be considered in lieu of arthroplasty for end-stage ankle arthritis. Established neuropathy and either existing or high risk of Charcot neuroarthropathy is a contraindication for TAA. Ankle arthrodesis or even tibiotalocalcaneal arthrodesis is favored over TAA for end-stage ankle arthritis associated with a dense peripheral neuropathy and risk of or existing Charcot neuroarthropathy.
Motor function: Intact motor function of the ankle and foot is essential to successful ankle arthrodesis. Lack of active dorsiflexion, plantarflexion, inversion, or eversion is a relative contraindication to ankle arthrodesis. Tibialis anterior function is still required to dorsiflex the foot at the transverse tarsal (talonavicular and calcaneocuboid) joints. Gastrocnemius–soleus function is needed to plantarflex the hindfoot. Posterior tibial and peroneal tendon function is necessary to maintain a dynamic balance of the foot under the ankle arthrodesis. Without these functioning muscle groups, a tibiotalocalcaneal or pan-talar arthrodesis or possibly a bridle tendon transfer may be warranted.
IMAGING AND OTHER DIAGNOSTIC STUDIES
Weight-bearing AP, lateral, and mortise views of the ankle.
Weight-bearing AP, lateral, and oblique views of the foot, particularly with associated foot deformity
With associated or suspected lower leg deformity, we routinely obtain weight-bearing AP and lateral tibia–fibula views.
With deformity in the lower extremity, we routinely obtain weight-bearing mechanical axis (hip-to-ankle) views of both extremities.
We typically evaluate complex or ill-defined ankle–hindfoot patterns of arthritis with or without deformity using CT of the ankle and hindfoot.
If we suspect avascular necrosis of the talus or distal tibia, we obtain an MRI of the ankle.
DIFFERENTIAL DIAGNOSIS
See “Pathogenesis.”
NONOPERATIVE MANAGEMENT
Activity modification
Bracing
Ankle–foot orthosis (AFO)
Double upright brace attached to shoe
Stiffer-soled shoe with a rocker-bottom modification
Nonsteroidal anti-inflammatories or COX-2 inhibitors
Medications for systemic inflammatory arthropathy
Corticosteroid injection
Viscosupplementation
SURGICAL MANAGEMENT
The trend is to perform ankle arthrodesis through an anterior approach with preservation of the malleoli.
Recently there have been favorable outcomes in conversion of ankle fusion to TAA.
While ankle arthrodesis is typically successful in relieving symptoms related to end-stage ankle arthritis, over time the hindfoot may develop compensatory degenerative changes (ie, adjacent joint arthritis).
If one or both of the malleoli are sacrificed, then this potential conversion is compromised.
The anterior approach is also used for the majority of TAA cases.
Preoperative Planning
Vascular and neurologic examination
It is easy to focus on the patient's symptoms and radiographs demonstrating end-stage ankle arthritis.
Satisfactory circulation is essential to allow wound healing and fusion.
A neuropathy may warrant a more extensive ankle– hindfoot stabilization.
Deformity correction
A sound preoperative plan facilitates effective intraoperative deformity correction.
The surgeon should evaluate the contralateral extremity and ankle to have an understanding of what is physiologic for that patient.
Positioning
Supine
Plantar aspect of operated foot at end of operating table
Foot and ankle well balanced, with toes directed to the ceiling
A bolster under the ipsilateral hip prevents undesired external rotation of the hip.
We routinely use a thigh tourniquet and regional anesthesia.
A popliteal block provides adequate pain relief postoperatively, particularly if a regional catheter is used. Moreover, hip and knee flexion–extension is not forfeited, facilitating safe immediate postoperative mobilization.
However, to use a thigh tourniquet with a popliteal block typically requires a supplemental femoral nerve block (patients temporarily forfeit knee extension postoperatively) or general anesthesia.
Approach
Anterior approach to the ankle, using the interval between the tibialis anterior (TA) tendon and the extensor hallucis longus (EHL) tendon
TECHNIQUES
APPROACH
Make a longitudinal midline incision over the anterior ankle, starting about 10 cm proximal to the tibiotalar joint and 1 cm lateral to the tibial crest (TECH FIG 1A).
Continue the incision midline over the anterior ankle just distal to the talonavicular joint.
At no point should direct tension be placed on the skin margins; we perform deep, full-thickness retraction as soon as possible to limit the risk of skin complications.
Identify and protect the superficial peroneal nerve by retracting it laterally.
In our experience there is a consistent branch of the superficial peroneal nerve that crosses directly over or immediately proximal to the tibiotalar joint.
We then expose the extensor retinaculum, identify the course of the EHL tendon, and sharply but carefully divide the retinaculum directly over the EHL tendon (TECH FIG 1B,C).
We always attempt to maintain the TA tendon in its dedicated sheath.
Preserving the retinaculum over the TA tendon
Prevents bowstringing of the tendon and thereby reduces the stress on the anterior wound
Should there be a wound dehiscence, then the TA is not directly exposed.
Preserving the retinaculum over the TA tendon is not always possible; some patients do not have a dedicated sheath for the TA.
Use the interval between the TA and EHL tendon, with the TA and EHL tendons retracted medially and laterally, respectively.
Identify the deep neurovascular bundle (anterior tibial–dorsalis pedis artery and deep peroneal nerve) and carefully retract it laterally throughout the remainder of the procedure (TECH FIG 1D).
Perform an anterior capsulotomy along with elevation of the tibial and dorsal talar periosteum to about 6 to 8 cm proximal to the tibial plafond and talonavicular joint, respectively (TECH FIG 1E).
Elevate this separated capsule and periosteum medially and laterally to expose the ankle, access the medial and lateral gutters, and visualize the medial and lateral malleoli (TECH FIG 1F,G).
Remove anterior tibial and talar osteophytes to facilitate exposure and avoid interference with the instrumentation (TECH FIG 1H,I).
TECH FIG 1 • A. Anterior approach to ankle (note sural nerve). B,C. The extensor retinaculum is divided. B. Initiating the longitudinal incision in the retinaculum immediately superficial to the extensor hallucis longus (EHL) tendon. C. EHL tendon exposed. D. The deep neurovascular bundle must be identified and protected. E. Tibiotalar joint exposed after arthrotomy. F,G. Exposure improved with capsular and periosteal elevation at the joint line. F. Laterally. G. Medially. H,I. Distal anterior tibial exostectomy. H. Rongeur. I. Chisel.
TIBIOTALAR JOINT PREPARATION
I routinely use joint distraction (TECH FIG 2A,B).
I prefer to maintain the subchondral bone architecture.
In preserving the essential anatomy of the talar dome and tibial plafond, I have the ability to adjust dorsiflexion–plantarflexion without compromising limb length or bony apposition at the arthrodesis site.
Flat cuts tend to forfeit limb length and the ability to adjust alignment without forfeiting optimal bony apposition.
Obviously, with deformity correction through the joint, some of the subchondral architecture may need to be sacrificed.
I remove the residual cartilage with a sharp elevator or chisel (TECH FIG 2A).
While preserving the subchondral architecture as best as possible I penetrate the subchondral bone with a drill bit, a narrow chisel, or both (TECH FIG 2C–E).
This increases surface area and promotes fusion.
While careful to preserve the malleoli, I still prepare the tibiotalar joint gutters to further increase the surface area for fusion (TECH FIG 2F,G).
Use of bone graft is at the surgeon's discretion.
I routinely use bone graft to fill any voids at the arthrodesis site.
Avoid excessive use of bone graft; the best chance for fusion is if the physiologic surfaces are appropriately prepared and well apposed.
TECH FIG 2 • A,B. Tibiotalar joint preparation. A. Using a lamina spreader for distraction and a sharp elevator to delaminate residual cartilage. B. Alternatively, an invasive joint distractor may be used, here with drilling of the subchondral bone to promote healing. Tibiotalar joint preparation. C. Lateral lamina spreader with medial talar dome subchondral bone drilling. D. Dual lamina spreaders to switch to using only the medial lamina spreader. E. Medial lamina spreader with sharp elevator to remove residual lateral talar dome cartilage. F,G. Preparation of the tibiotalar gutters. F.Medial gutter with sharp elevator. G. Lateral gutter using a rongeur.
TIBIOTALAR JOINT REDUCTION
For me, optimal tibiotalar joint alignment for arthrodesis is:
Neutral dorsiflexion–plantarflexion (TECH FIG 3A)
Many years ago, there was a tendency to fuse women's ankles in plantarflexion to facilitate wearing a heel. This is an idea that should be abandoned.
The tendency is to underestimate how much dorsiflexion is needed to get the ankle to neutral. Therefore, I typically dorsiflex the talus within the mortise just slightly more than what I think it may need. This usually results in neutral dorsiflexion– plantarflexion.
Slight hindfoot valgus
Balance the talus within the ankle mortise, but be sure that the hindfoot is in slight valgus.
If not, then contour the tibiotalar preparation to get the hindfoot in slight valgus.
A reasonable landmark is to have the lateral bony aspect of the calcaneus be in line with the fibula; if it is medial to the fibula, then a neutral to varus position is inappropriately set.
Rotation
Align the second metatarsal with the anterior tibial crest.
When the malleoli are preserved, rotation is often auto-adjusted.
External rotation is recommended by some authors, but I consider this only if the contralateral extremity dictates this position.
The goal is to avoid internal rotation.
Sagittal plane relationship of the talus to the tibia
Avoid anterior translation of the talus relative to the tibia. This places the ankle and foot at a biomechanical disadvantage.
With some deformity, it may be difficult to translate the talus posteriorly to a more physiologic position. In some cases, I have had to resect some of the posterior malleolus (through the joint from the anterior approach with joint distraction) to allow such posterior translation (TECH FIG 3B). Also, judiciously, the deltoid ligament may need to be partially released to allow posterior translation. Perform this cautiously, though, as some of the talar dome blood supply travels though the deltoid branch off the posterior tibial artery.
I routinely obtain intraoperative fluoroscopic views in the AP and lateral planes to confirm appropriate alignment and bony apposition.
TECH FIG 3 • A. Tibiotalar joint reduction, with neutral dorsiflexion–plantarflexion, slight hindfoot valgus, and second metatarsal rotated to anterior tibial crest. B. If the talus fails to translate posteriorly in the ankle mortise, then the posterior malleolus may need to be weakened to allow the talus to reduce under the tibial axis.
INTERNAL FIXATION WITH ANTERIOR PLATING–SCREW FIXATION
Internal fixation is contraindicated or less than optimal in the face of:
Infection
Osteopenic bone
Traditionally, I performed screw fixation and added an anterior plate for further stability; more recently, I have switched to a technique where anterior plating is the primary technique, and I supplement with screws (other than those in the plate) only if I feel further stability is needed.
Provisional fixation once optimal reduction is achieved
Traditional screw fixation and supplemental anterior plate
55-year-old high-demand patient with anterior translation of the talus within the ankle mortise (TECH FIG 4A–C)
Patient is positioned supine on the operating table with a bump under the ipsilateral hip to resist external rotation of the extremity.
I typically use a medial screw first (TECH FIG 4D).
Next, I place the posterior-to-anterior screw, the “home-run” screw.
With the newer anterior plating techniques that provide satisfactory stability, this screw has been largely abandoned; it is awkward to place and equally difficult to remove (TECH FIG 4E).
TECH FIG 4 • A–C. Fifty-five-year-old man with chronic instability and posttraumatic arthritis. A. AP view with comparison to contralateral ankle. B. Mortise view. C. Lateral view. There is considerable anterior translation of the talus from the ankle mortise. Medial screw placed first from the medial tibia to the talar dome, placed through a medial stab incision. E. Traditional posterior-to-anterior screw, placed via a posterolateral stab incision (care must be maintained to avoid injury to the sural nerve). F. Anterolateral screw placed through the anterior approach. Provisional fixation was placed adjacent to this screw. G–I. Anterior plating. G. Proximal screw fixation. H. Talar screw fixation. I. Final view of plate before closure. J–N. Postoperative weight-bearing radiographs of example patient with traditional screw fixation and supplemental anterior plate. J. AP radiograph. K. Mortise view. L. Lateral view (talus is reduced under tibial axis). M. Dorsiflexion view. N. Plantarflexion view. The patient lacks some hindfoot compensation for dorsiflexion and plantarflexion.
I add an anterolateral screw, one that is relatively vertical (TECH FIG 4F).
Finally, I augment the fixation with an anterior plate. In this case a small fragment, non-locking plate was used (TECH FIG 4G–I).
In my experience, adding a supplemental anterior plate to an ankle arthrodesis construct adds considerable stability.
Follow-up radiographs (TECH FIG 4J–N)
Patient returned to full activities, even playing doubles tennis.
He lacks some plantarflexion; time will tell what effect this will have on the hindfoot articulations that are attempting to compensate.
The talus is again in a physiologic relationship with the tibia, improving his biomechanics despite ankle arthrodesis.
Plate fixation as the primary fixation
33-year-old man with posttraumatic ankle arthritis and syndesmotic disruption (TECH FIG 5A–C)
Same joint preparation as described above
Provisional fixation with desired joint reduction
Plate locked to the dorsolateral talar neck with locking screws
Plate is precontoured based on average anterior ankle morphology.
Compression device is secured and compression is applied, thereby approximating the arthrodesis surfaces (TECH FIG 5D–F).
TECH FIG 5 • A–C. Preoperative radiographs of patient undergoing double anterior plating arthrodesis technique. A,B.AP and mortise views with end-stage ankle arthritis and chronic syndesmosis disruption. C. Lateral view. D. Lateral anterior plate applied and secured to talus and proximal compression device in place. E,F. Intraoperative fluoroscopic views of ankle of a different patient undergoing dual anterior plating, with provisional fixation and lateral plate in place. E. Lateral view. F. AP view. G. Example patient with both plates in place. H.Intraoperative fluoroscopic view of different patient with both plates in place.
While the locking plate creates axial compression, a mild but desirable valgus moment may be introduced since the lateral plate is being used for compression.
To obtain optimal compression, provisional fixation is removed before compression is applied but after the screws are locked into the talar neck and the compression device is secured proximally.
After performing compression and securing the lateral plate in the tibia, the medial plate is applied (TECH FIG 5G,H).
Since compression has already been performed, this medial plate, which is also precontoured, serves to statically lock the arthrodesis.
Each plate has a screw hole to allow non-locking screw fixation from the plate to the posterior talar body (TECH FIG 6A,B).
Follow-up of case example (TECH FIG 6C–G)
A supplemental screw may be added from the medial tibia to the talar body, but often this is unnecessary (TECH FIG 6H,I).
Closure
I use a drain for 24 hours.
Standard wound closure
I routinely close the capsule, extensor retinaculum, subcutaneous layer, and skin (to a tensionless closure).
The deep neurovascular bundle, extensor tendons, and superficial peroneal nerve need to be protected during closure.
Sterile dressings on wound
Padding
Posterior–sugar-tong splint
TECH FIG 6 • A,B. Intraoperative fluoroscopic views of two screws placed through the plate into the posterior talus for additional stability. A. Lateral view. B. Broden view to confirm that screws do not violate the subtalar joint. C–G. Postoperative radiographs of example patient for dual anterior plating. C. AP view. D. Mortise view. E. Lateral view. F. Dorsiflexion view. G. Plantarflexion view. H, I. Intraoperative fluoroscopic views of different patient with supplemental screw to anterior plating. H. AP view. I. Lateral view (note broken guide pin; it is important to follow the exact trajectory of the guide pin with cannulated screw systems).
EXTERNAL FIXATION
Infection is not a contraindication for external fixation.
There will be no implant directly at the tibiotalar joint.
In some cases, I have performed a staged arthrodesis, with initial débridement and antibiotic bead placement. The external fixator may be placed at that initial procedure or at the definitive procedure when the antibiotic beads are removed and the joint is reduced and compressed with the external fixator.
TECH FIG 7 • A–C. Preoperative radiographs of example patient for ankle arthrodesis with external fixation; patient has failed ankle arthrodesis with internal fixation. A. AP view. B. Mortise view. C. Lateral view. D,E. Poor skin condition. D. Medial ankle with prior anteromedial approach. E.Dorsolateral aspect with residual scarring. F. Patient positioned supine on the operating table. The external rotation malunion of the distal tibia creates excessive external rotation of the foot relative to the tibial axis.
Forty-five-year-old patient with posttraumatic arthritis and deformity of the ankle, failing to respond to a prior attempt at ankle arthrodesis.
Radiographs demonstrate nonunion and residual deformity (TECH FIG 7A–C).
Clinically, there are poor soft tissues anteriorly and a prior medial incision that will need to be incorporated into the surgical approach (TECH FIG 7D,E).
A standard anterior approach is too risky and, in my opinion, would leave an insufficient skin bridge to the prior incision.
Supine on the operating table, again with a bolster under the ipsilateral hip to direct the ankle anteriorly
This patient also had a distal tibial external rotation malunion and an ankle nonunion with residual ankle external rotation (TECH FIG 7F).
Hardware removal
I used the prior incision and added another “mini-arthrotomy” incision laterally, thereby avoiding the unhealthy skin directly anteriorly over the ankle (TECH FIG 8A).
I prepared the joint through the medial incision and used the lateral incision to provide joint distraction (TECH FIG 8B). I also switched the lamina spreader to the medial wound so that I could prepare the remainder of the joint via the lateral incision.
From the preoperative radiographs it is obvious that there is distal tibial deformity and nonanatomic malleolar anatomy (TECH FIG 8C,D).
For this reason, the talus is not locked within the ankle mortise and rotation will need to be carefully controlled. However, this is more important with internal fixation; with external fixation such malrotation could still be corrected postoperatively with external fixator frame adjustment.
TECH FIG 8 • A. Modified “mini-open” arthrotomy approach to ankle arthrodesis. Previous medial incision used and a separate mini-lateral incision. Medial incision is being used for joint preparation while joint is being distracted by lamina spreader placed via lateral incision. Skin bridge between two wounds is adequate and previously compromised skin is not violated. B. Medial joint preparation. C,D. With distortion of the malleolar anatomy, the talus is not “locked” within the ankle mortise. C. Ankle tends to externally rotate. D. Ankle can be manually reduced to a physiologic position with the second metatarsal aligned to the anterior tibial shaft axis.
Joint reduction
Neutraldorsiflexion–plantarflexion
Slight hindfoot valgus
Correct malrotation
Align second metatarsal with the anterior tibial crest.
Provisionally pin the joint
I usually place two Steinmann pins axially. While this violates the subtalar joint, I do not believe that this has significant consequences in these patients with deformity, severe ankle arthritis, and compensatory hindfoot alignment.
I routinely close the wounds at this point because once the external fixator is in place, suturing is particularly tedious.
However, if you prefer to delay the wound closure until the external fixator is in place, one or two struts can easily be reflected to allow adequate access to the wound or wounds.
Proximal ring block (TECH FIG 9A)
I place the proximal ring block (I usually use two rings to create the “block”) orthogonally to the tibia.
Initially, I stabilize the rings with two thin wires but do not tension them at this point.
I supplement the proximal ring block fixation with three half-pins (TECH FIG 9B).
Once the half-pins are secured, I tension the thin wires (TECH FIG 9C).
Foot plate
I suspend the foot plate (“horseshoe”) from a transverse forefoot wire. This way I can control the foot's position within the foot plate (TECH FIG 10A).
Once I am satisfied with the foot's position relative to the foot plate, I secure the hindfoot with two crossed thin wires, making sure the plantar surface of the foot is distal to the foot ring (TECH FIG 10B).
I typically place a midfoot wire as well.
Before tensioning the thin wires, I close the horseshoe-shaped foot plate anteriorly.
This can be done by adding a half-ring to the anterior foot plate, or I can have a double-decker foot plate and close the more proximal of the two foot plates (TECH FIG 10C).
Having two foot plate components affords less interference between the struts (that will connect the proximal ring block to the foot plate) and the thin wires to be passed through the foot from the foot plate.
I then tension the thin wires in the foot (TECH FIG 10D).
I also place one or two talar wires to provide greater support and to protect the subtalar joint (TECH FIG 10E,F).
These two wires either need to be built up from a single foot plate or connected to the proximal component of a two-ring foot plate set-up.
TECH FIG 9 • A. Building the proximal ring block, first with thin wires. Wounds were closed before applying external fixator. B. Half-pins added to stabilize the proximal ring block. C. Thin wires are tensioned within the proximal ring block.
TECH FIG 10 • A. Forefoot wire placed to suspend the foot plate. B. Foot balanced within the foot plate. Foot plate suspended from forefoot wire and calcaneal wires being passed to stabilize the hindfoot. C. Tensioning the thin wires in the foot. The ring has been closed on the foot frame so that tension in all wires can be effectively maintained. D. In this case, two rings were used for the foot plate portion of the frame. Closing the top ring allows the foot frame to be closed even without placing a half-ring on the anterior portion of the “horseshoe.” E,F.Two talar wires are passed. Without talar wires compression would be placed not only on the tibiotalar joint but also on the subtalar joint. G. Calcaneal half-pin for added foot frame stability.
This is also essential to protect the subtalar joint from compression. If fixation from the foot plate to the foot is limited to the forefoot, midfoot, and calcaneus and no fixation is added to the talus, then axial compression will not be isolated to the tibiotalar joint but will also include the subtalar joint (with potential detrimental effects to the subtalar joint cartilage and motion). A perhaps more sophisticated (but not more complicated) construction of the foot plate is to distract between the two components of the foot plate, so that the subtalar joint is distracted while the tibiotalar joint is compressed. Although unproven, this may have a protective effect on the subtalar joint.
I routinely add a calcaneal half-pin for added foot plate stability (TECH FIG 10G).
Connect the proximal ring block and foot plate by struts and apply tibiotalar compression (TECH FIG 11A).
I make subtle adjustments at this point, which sometimes warrants removing one or both of the provisional fixation pins (TECH FIG 11B).
If the alignment is optimal, then I can leave one provisional pin in place (provided it is truly axial) to act as a rail as I compress the tibial talar joint with the external fixator.
TECH FIG 11 • A. Adding struts to be used for compression between the proximal ring block and the foot frame. B. Proper position of the foot and leg within the external fixator. Ankle with neutral dorsiflexion–plantarflexion and plantar foot is distal to most distal ring–plate. The provisional fixation was removed for compression. C,D. Physiologic hindfoot valgus, with varus avoided. C. AP view. D. Posteroanterior view.
If no translation, angulation, or rotation is required, which is often the case if the initial reduction was appropriate, then simply tightening the struts uniformly leads to satisfactory axial compression (TECH FIG 11C,D).
If adjustments need to be made, the computer program may be used to run an effective correction at this time. However, on the operating table, the struts may simply be loosened, a gross manual adjustment can be made (with the provisional fixation removed), and the struts again secured. Then, uniform tightening of all struts can be performed.
Final fluoroscopic views in the AP and lateral planes are sometimes difficult to interpret with an external fixator in place, but with subtle rotation of the limb, appropriate alignment and bony apposition can be confirmed.
Final check to be sure that all bolts and connections are stable
Sterile dressings on the wound
Sterile dressings on the wires and half-pins
Pin irritation typically occurs because of skin motion or tension about the half-pins or thin wires.
I routinely place thick dressings around the thin wires and half-pins, creating moderate pressure from the dressing on the skin immediately adjacent to the halfpin or wire and thereby stabilizing the skin.
Prefabricated bolsters are also available to stabilize the skin around the pins.
Final follow-up for external fixation case example (TECH FIG 12)
Alignment restored
Fusion apparent despite distorted distal tibial alignment
TECH FIG 12 • Follow-up radiographs suggesting successful revision ankle arthrodesis using external fixation. A. AP view. B. Mortise view. C. Lateral view.
POSTOPERATIVE CARE
With advances in anesthesia, ankle arthrodesis may be performed on an outpatient basis.
However, we typically keep these patients at least overnight for pain control, nasal oxygen (which may have some positive effect on anterior wound healing), and prophylactic intravenous antibiotics.
Follow-up in 10 to 14 days
Internal fixation
Suture removal
Short-leg, touch-down weight-bearing cast
External fixation
Suture removal
Radiographs to assess bony apposition at the arthrodesis site and alignment. If a subtle adjustment needs to be made, it is done at this time, typically with the computer program.
We routinely add more compression to the arthrodesis site at this and subsequent visits. Simple axial compression does not require use of the computer program; instead, uniform tightening of all struts creates axial compression at the arthrodesis site. This is a major advantage of external fixation over internal fixation. With internal fixation, bony apposition at the arthrodesis site cannot be altered after the index procedure.
The patient is instructed how to perform pin care. We do not usually have the patient perform pin care in the first 10 to 14 days in order to protect the wound. My routine pin care includes once-a-day pin cleaning with a sponge moistened with a 50–50 mixture of sterile saline and hydrogen peroxide. I instruct the patients to “shoeshine” the pins with the sponge so that the debris is removed at the pin–skin interface. If a pin is irritated, then we recommend placing an antibiotic ointment at that pin's interface with the skin and to continue to stabilize that particular pin with dressings that stabilize the skin adjacent to the pin. Oral antibiotics may be required in some situations.
We have the orthotist create a tread for the foot plate. Once the wounds have healed adequately and edema is controlled, the tread can be added and weight bearing through the external fixator is possible, another potential advantage of external over internal fixation.
Follow-up at about 6 weeks
Internal fixation
Ankle radiographs
If healing is progressing well, the patient is progressed to a cam boot.
If more healing is necessary, a short-leg cast is continued.
Weight bearing may be progressively increased if healing is progressing, but we typically restrict the patient from full weight bearing until 10 weeks (longer if healing is delayed).
External fixation
Radiographs
We routinely add more axial compression.
Pin care is reinforced.
Weight bearing is encouraged with the tread on the foot plate.
Follow-up at 10 to 12 weeks and beyond
Internal fixation
Radiographs
If healing is suggested, then the patient can progress to full weight bearing, first in the cam boot and then transitioning to a regular shoe by 12 to 14 weeks. If healing is delayed, then this protocol is delayed.
External fixation
Radiographs
More axial compression is added.
If healing is suggested radiographically, then the surgeon should plan for external fixator removal between 12 and 16 weeks.
If healing is delayed, more axial compression is added and follow-up is set for 3 to 4 more weeks. External fixator removal is delayed until healing is suggested.
Frame removal may be performed in the office, but removal of half-pins may be particularly uncomfortable for the patient (especially if hydroxyapatite-coated pins are used).
A short operating room procedure should be considered for frame removal with the patient under anesthesia.
We routine add a short-leg walking cast for an additional 2 to 4 weeks, then transition to a cam boot and regular shoe.
OUTCOMES
The literature suggests favorable outcomes of ankle arthrodesis, with good relief of ankle pain and high rates of patient satisfaction (mostly level IV retrospective studies without standardized foot and ankle outcome measures).
At intermediate follow-up, good to excellent results have been reported in 66% to 90% of patients (mostly level IV retrospective studies without standardized foot and ankle outcome measures).
In long-term follow-up, a considerable number of patients with ankle arthrodesis develop adjacent joint (subtalar and, to a lesser degree, transverse tarsal joint) arthrosis.
Although most patients with arthrodesis report satisfactory pain relief, functional outcome, particularly gait analysis, is not physiologic.
COMPLICATIONS
Both internal and external fixation
Infection
Wound dehiscence or delayed wound healing
Nonunion
Malunion
Late development of subtalar (and, to a lesser degree, transverse tarsal joint) arthritis (adjacent joint arthritis)
Internal fixation
Prominent hardware
Residual gapping at tibiotalar arthrodesis site that cannot be compressed postoperatively
External fixation
Pin tract infection
REFERENCES
1. Agel J, Coetzee JC, Sangeorzan BJ, et al. Functional limitations of patients with end-stage ankle arthrosis. Foot Ankle Int 2005;26:537–539.
2. Anderson T, Montgomery F, Besjakov J, et al. Arthrodesis of the ankle for non-inflammatory conditions—healing and reliability of outcome measurements. Foot Ankle Int 2002;23:390–393.
3. Coester LM, Saltzman CL, Leupold J, et al. Long-term results following ankle arthrodesis for post-traumatic arthritis. J Bone Joint Surg Am 2001;83A:219–228.
4. Colman AB, Pomeroy GC. Transfibular ankle arthrodesis with rigid internal fixation: an assessment of outcome. Foot Ankle Int 2007;28: 303–307.
5. Easley ME, Montijo HE, Wilson JB, et al. Revision tibiotalar arthrodesis. J Bone Joint Surg Am 2008;90A:1212–1223.
6. Eylon S, Porat S, Bor N, et al. Outcome of Ilizarov ankle arthrodesis. Foot Ankle Int 2007;28:873–879.
7. Fuchs S, Sandmann C, Skwara A, et al. Quality of life 20 years after arthrodesis of the ankle: a study of adjacent joints. J Bone Joint Surg Br 2003;85:994–998.
8. Glazebrook M, Daniels T, Younger A, et al. Comparison of healthrelated quality of life between patients with end-stage ankle and hip arthrosis. J Bone Joint Surg Am 2008;90:499–505.
9. Greisberg J, Assal M, Flueckiger G, et al. Takedown of ankle fusion and conversion to total ankle replacement. Clin Orthop Relat Res 2004;424:80–88.
10. Haddad SL, Coetzee JC, Estok R, 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;89A: 1899–1905.
11. Hintermann B, Barg A, Knupp M, et al. Conversion of painful ankle arthrodesis to total ankle arthroplasty. J Bone Joint Surg Am 2009;91A:850–858.
12. Holt ES, Hansen ST, Mayo KA, et al. Ankle arthrodesis using internal screw fixation. Clin Orthop Relat Res 1991;268:21–28.
13. Johnson EE, Weltmer J, Lian GJ, et al. Ilizarov ankle arthrodesis. Clin Orthop Relat Res 1992;280:160–169.
14. King HA, Watkins TB Jr, Samuelson KM. Analysis of foot position in ankle arthrodesis and its influence on gait. Foot Ankle 1980;1:44–49.
15. Kovoor CC, Padmanabhan V, Bhaskar D, et al. Ankle fusion for bone loss around the ankle joint using the Ilizarov technique. J Bone Joint Surg Br 2009;91B:361–366.
16. Mann RA, Rongstad KM. Arthrodesis of the ankle: a critical analysis. Foot Ankle Int 1998;19:3–9.
17. Monroe MT, Beals TC, Manoli A II. Clinical outcome of arthrodesis of the ankle using rigid internal fixation with cancellous screws. Foot Ankle Int 1999;20:227–231.
18. Muir DC, Amendola A, Saltzman CL. Long-term outcome of ankle arthrodesis. Foot Ankle Clin 2002;7:703–708.
19. Myerson MS, Quill G. Ankle arthrodesis: a comparison of an arthroscopic and an open method of treatment. Clin Orthop Relat Res 1991;268:84–95.
20. Nielsen KK, Linde F, Jensen NC. The outcome of arthroscopic and open surgery ankle arthrodesis: a comparative retrospective study on 107 patients. Foot Ankle Surg 2008;14:153–157.
21. Ogut T, Glisson RR, Chuckpaiwong B, et al. External ring fixation versus screw fixation for ankle arthrodesis: a biomechanical comparison. Foot Ankle Int 2009;30:353–360.
22. Plaass C, Knupp M, Barg A, et al. Anterior double plating for rigid fixation of isolated tibiotalar arthrodesis. Foot Ankle Int 2009;30: 631–639.
23. Salem KH, Kinzl L, Schmelz A. Ankle arthrodesis using Ilizarov ring fixators: a review of 22 cases. Foot Ankle Int 2006;27:764–770.
24. Saltzman CL, Mann RA, Ahrens JE, et al. Prospective controlled trial of STAR total ankle replacement versus ankle fusion: initial results. Foot Ankle Int 2009;30:579–596.
25. Sealey RJ, Myerson MS, Molloy A, et al. Sagittal plane motion of the hindfoot following ankle arthrodesis: a prospective analysis. Foot Ankle Int 2009;30:187–196.
26. SooHoo NF, Zingmond DS, Ko CY. Comparison of reoperation rates following ankle arthrodesis and total ankle arthroplasty. J Bone Joint Surg Am 2007;89:2143–2149.
27. Takakura Y, Tanaka Y, Sugimoto K, et al. Long-term results of arthrodesis for osteoarthritis of the ankle. Clin Orthop Relat Res 1999;361:178–185.
28. Thomas R, Daniels TR, Parker K. Gait analysis and functional outcomes following ankle arthrodesis for isolated ankle arthritis. J Bone Joint Surg Am 2006;88:526–535.
29. Thomas RH, Daniels TR. Ankle arthritis. J Bone Joint Surg Am 2003;85A:923–936.
30. Trouillier H, Hansel L, Schaff P, et al. Long-term results after ankle arthrodesis: clinical, radiological, gait analytical aspects. Foot Ankle Int 2002;23:1081–1090.
31. White AA III. A precision posterior ankle fusion. Clin Orthop Relat Res 1974;98:239–250.