///

Ankle Replacement-BOX total ankle replacement (MatOrtho)

Learn the Ankle Replacement-BOX total ankle replacement (MatOrtho) surgical technique with step by step instructions on OrthOracle. Our e-learning platform contains high resolution images and a certified CME of the Ankle Replacement-BOX total ankle replacement (MatOrtho) surgical procedure.

Ankle replacement has been available as an intervention for ankle arthritis since the 1970s. The initial implants were engineered on the assumption that the human ankle joint functioned as a true hinge . They were therefore designed only to allow uniplanar movement (plantar and dorsiflexion) and comprised just 2 components which were mechanically linked. The ankle joints they were implanted into however also functioned with a degree of rotation which had to occur at the weakest point in the “mechanism”. Given the robustness of the implanted ankle hinges this transpired to be the implant/joint interface which therefore led invariably to early implant failure.
The next generation of ankle replacements used a 3 component design, in which the Tibial and Talar components were linked by a UHMW polyethylene meniscus which allowed rotation to occur within the joint itself. These ‘mobile-bearing’ prostheses used the congruity of the ‘articulating’ surfaces, to reduce the constrained forces and overcome the high contact stresses resulting in a reduction in polyethylene wear and mechanical loosening of the fixed components. These initial replacements whose results still define what longevity an ankle replacement should attain are the STAR , Beuchal-Pappas and Salto implants. Advancements have been made in the instrumentation and reproducibility of implantation. In general their 10 year survivorships are lower than reported for hip and knee replacements UK National Joint Registry survival rates are now in the region of 80 percent.
Total ankle replacement is generally not be recommended for younger or higher demand patients due to concerns of longevity due to accelerated wear of the implant, the exception being patients with severe poly-articular inflammatory arthropathy.
The BOX (Bologna-Oxford) ankle manufactured by MatOrth is a three component prosthesis. The Box ankle replacement prosthesis has been designed to maximise congruency throughout the arc of motion, aiming to mimic normal ankle biomechanics. The bearing surface of the tibial component has a subtle curve in the coronal plane to accommodate for varus/ valgus force through the talo-crural joint. Biomechanical modelling has demonstrated both rolling and sliding motions take place at the talocrural joint. In theory, full congruence should reduce wear by avoiding edge-loading. Similar to the STAR prosthesis, two anchorage bars on the tibial platform of the BOX ankle replacement provide stable primary fixation to the tibial bone. A precisely cut talar component allows a good press fit of the talar component, which is further stabilised with two vertical pegs.
Primary stability of the Box ankle replacement components reduces micromotion assisting the circumstances necessary to provide reliable bone ingrowth.
Readers will also find the following techniques of interest:
Ankle replacement-Revision using Wright Invision Ankle replacement system
Ankle replacement-Wright Infinity ankle replacement
Ankle replacement-Wright Prophecy
Ankle replacement-Star ankle replacement (revision of mensical component)
Ankle Replacement -De Puy Mobility




INDICATIONS
–Isolated Ankle arthritis in a well aligned ankle: Generally a replacement is an operation for the lower demand and fifty plus age group with limited angular deformity and good soft tissue cover around the ankle and with no history of deep infection or neuropathy. With an ankle replacement the failure rate of most implants (which have been in use for long enough) is 2%/annum which equates to a 10 year survivorship of 80%.
A fusion in general is for higher demand/ younger patients or those wishing a greater degree of predictability than afforded by Ankle replacement. With a fusion the “risk” in the majority of patients can be regarded as “front-loaded”. As long as a non-union does not occur (5-10% chance, technique dependent) then in the majority no subsequent / later intervention is likely though the subtalar and midfoot joints are highly likely to become degenerate. Function will reduce with this if this occurs but the lead time is likely to be 10-20 years. .
-Ankle arthritis with deformity : This will self evidently be a more challenging primary operation. Increased failure rates are seen if normal alignment of the limb has not been restored post-operatively. Cases of deformity with bone loss effecting the weight bearing surfaces then cutting the deformity out of the bone may lead to large bony resections .This can produce issues with adequate bony support for the prosthesis and potentially relative laxity of the soft tissues stabilisers . In those ankles with deformity and no bone loss, the soft tissue restraints are likely to be significantly attenuated. The context most often is a varus arthritic ankle where a robust lateral ligament reconstruction may be required at the time of primary operation (such as an Evans peroneal re-routing stabilisation).
Relative indications for ankle replacement (when compared to Ankle Fusion) are also for cases of inflammatory arthropathies involving the ankle, where subtalar and mid foot articulations have (or are more likely to develop) inter-current disease and on one side for a patient with bilateral ankle arthritis.
Increased failure rates are seen if normal alignment of the limb is not restored post-operatively. Consideration should be given to arthrodesis if there is significant deformity at the ankle. Associated procedures may be required to correct hindfoot or proximal limb deformity.
SYMPTOMS & EXAMINATION:
Most patients with severe ankle arthritis localise the pain well to the level of the joint. Very much as with arthritis elsewhere symptoms tend to progress from early activity /start up pain which eases off through to progressively more disabling and continual weight bearing pain and on occasion patients can experience pain at night or at rest. A much less common symptom which can co-exist with pain is that of ankle instability. If gait is becoming altered due to the arthritis pain proximal to the ankle may occur secondary to alteration of the weight-bearing axis of the limb. Some patients complain of swelling which restricts the use of some footwear.
The vast majority of patients will either have a history of a significant injury (such as an ankle fracture), chronic deformity (for example Cavo-varus) or a past history of chronic lateral ligament instability. More rarely the cause is a more generalised tendency to osteoarthritis or an inflammatory arthropathy.
On examination swelling and tenderness well localised to the ankle is common. Range of movement is often reduced and may be uncomfortable. More important than ankle movement is what the subtalar and midfoot mobility is like. If both are very mobile then it is likely that post-fusion good compensatory movement in these joints will allow normal gait and in fitter ,younger patients even the ability to return to running. Conversely if movement here is restricted these joints should be carefully inspected with CT or MRI to identify any arthritic change. If still equivocal then an injection into the ankle joint with inta-articular contrast (see below) is indicated.
Any deformity should be noted. Varus is most common and valgus and equinus less common. The key issues with any deformity are A:Whether it is passively correctable (or not) and B.:Being sure of its anatomical location(s). The former is easily clinically determined .The latter can be more difficult to assess, in particular in the presence of severe deformity and CT and/or long leg alignment films are useful to locate the level of deformity.
Another feature to examine carefully in any varus ankle is the position of the 1st Ray , in particular whether it is plantar or dorsi-flexed and fixed. If this becomes apparent with the ankle corrected to neutral then the first ray will need correcting as part of the procedure.
In assessing equinus it should be appreciated at what level(s) the deformity rests. Beware of associated fixed midfoot equinus which will leave the mid/forefoot in a plantar flexed position once the ankle is fused in neutral if it is ignored. If dealing with isolated ankle equinus, a Silverskiold test will identify whether this is related to gastrocnemius or generalised calf tightness, be prepared to add a triple cut (or open )Achilles release or an isolated gastrocnemius release dependent on the severity of the deformity.
The rest of the lower limbs alignment should not be forgotten. In general correction of deformity should start proximally and proceed distally.
A vascular examination must be made and if abnormal dealt with appropriately.
INVESTIGATION:
Weight-Bearing Plain X-Ray: This is the initial imaging for most patients with ankle arthritis of any degree. Though the ankle is relatively well visualised, the subtalar and midfoot joints aren’t included, in the presence of associated deformity weight-bearing foot X-Rays should be acquired.
CT scan. This is better in defining how much relevant arthritic change exists and where it is than MRI. It is also easier to differentiate the level of deformity from CT than MRI. There are cases where there are significant subchondral cysts which may require bone grafting, or if very large present a contra-indication to surgery. the location and extent is again best defined with CT.
MRI scan: An MRI is more sensitive for early degenerative change but will be degraded by any internal fixation and is not 100% sensitive for early arthritis. It can be more difficult to be objective about the severity of more advanced arthritic change as bone oedema ( a reversible phenomenum) complicates the MRI images. A CT lacks this sensitivity which is a positive and not a negative. Some surgeons prefer to use MRI rather than CT pre-fusion as imaging.
X-Ray guided injection: This should be into whichever joint (ankle or subtalar ) appears more likely the location of symptoms. Contrast is needed as in a proportion of patients the two joints will inter-connect and improvement of symptoms after injection into one cannot under these circumstances be regarded as discriminatory.
ALTERNATE OPERATIVE MANAGEMENT:
Ankle fusion: Can be performed open or arthroscopically, pain levels are lower with the arthroscopic technique and hospital stay in half of patients is just one night. Ultimately no longer term difference with a successful Arthroscopic versus Open ankle fusion , just more patients get there arthroscopically and the journey is easier. In the presence of severe and fixed deformity however it should be the procedure of choice.
Arthroscopic Ankle debridement: This has a role for the treatment of those with lesser degrees of arthritic change & “intermediate” symptoms. There are no clear criteria for this but patients with severe levels of pain on minor activity and possibly at rest or night are unlikely to be appropriate candidates for joint sparing surgery.
Distal Tibial osteotomy: In the much smaller subgroup of patients with isolated arthritis in association with significant distal tibial deformity, where there is a large area of well preserved talo-crural joint, a closing or opening wedge tibial osteotomy can be considered.
NON-OPERATIVE MANAGEMENT:
Activity modification and analgesia.
Local anaesthetic & steroid injection.
Orthotics & Shoewear modifications.
CONTRAINDICATIONS:
Active infection, active smoking, poor vascular inflow: require correction before replacement is considered.
Osteonecrosis, large subchondral cysts, very high body mass index.

The theatre team should confirm that the implants and instruments are sterile prior to anaesthetic commencing.
AP and lateral weight bearing radiographs should be displayed, with alignment views if necessary.
The procedure is done under a general or spinal anaesthetic in combination with a popliteal regional nerve block.
The patient is positioned supine on the operation table.
The side and procedure should be confirmed as part of the WHO check.
Pre-operative antibiotics are administered intravenously.
A sandbag is placed under the ipsilateral buttock so that the 2nd ray is aligned vertically.
The leg is exsanguinated and a thigh tourniquet inflated just prior to skin preparation.
The skin is prepared up to the tourniquet and the leg should be draped above the knee.

Incision site marked between Extensor Hallucis Longus and Tibialis Anterior tendons.
An anterior approach to the ankle provides good access to the distal tibia and superior talus. Here the incision is planned just medial to the Extensor Hallucis Longus tendon. An incision lateral to Extensor digitorum longus is an alternative approach. Patients should be informed that the superficial peroneal nerve is at risk of injury from these approaches which may leave an area of sensory loss over the dorsum of the foot.

1. Full thickness skin flaps developed.
2. Superior extensor retinaculum.
The planned skin incision is performed taking care to protect branches of the superficial peroneal nerve, where possible. Full thickness flaps should be elevated, avoiding undermining, and the skin edges gently retracted to preserve vascularity.

1.Crenulated Superior Extensor Retinaculum incision.
2.Tibialis Anterior.
3.Extensor Hallucis Longus (EHL).


Leash blood vessels anterior to ankle joint.
The plane between Tibialis Anterior and EHL is developed, taking care to identify and protect the deep peroneal nerve and anterior tibial artery, which lie deep and lateral to the EHL tendon.
A consistent leash of blood vessels is encountered anterior to the ankle joint capsule which should be quarterised or tied off as necessary.

1.Tibial osteophyte.
2.Talar osteophyte.
The dissection proceeds subperiosteally to expose the anterior tibial and talar surfaces, access to the ankle joint is often restricted by osteophytes of the tibia and talus, as is the case in this procedure.

A sharp osteotome is used to remove the anterior tibial osteophyts and expose the anterior margin of the tibial plafond, care should be taken to protect the medial malleolus.

The talar osteophytes are similarly removed to expose the superior surface of the talar neck

Clearance should be sufficient to allow seating of the talar cutting block with the ankle dorsiflexed to 90 degrees.

The Tibial Alignment Guide is attached, inserting the talar cutting block into the joint aligned in the inter-malleolar plane.

The Tibial Alignment Guide is now attached and the proximal strap secured at the level of the tibial tuberosity.

The Proximal Screw is loosened to allow positioning of the Tibial alignment guide.

The position of the alignment Tibial Alignment Guide will determine the slope of the talar cut in the sagittal plane. To avoid plantar or dorsiflexion of the tibial cut, ensure that the rod is parallel to the posterior border of the tibia as demonstrated by the lower skin marking.

Tighten The Proximal Screw to secure the position of The Tibial Alignment Guide.

The tongue of The Talar Cutting Block is placed, as far as it will go, into the joint space, centred between the malleoli, secure The Frontal Screw.

adjust The Tibial Alignment Guide to position it parallel to the anterior border of the tibia in the frontal plane. It is very important at this stage to ‘dial’ in adequate external rotation, to ensure that the ankle is implanted in the correct axial alignment. There is a tendency towards internal rotation. A loose guidewire passed into the medial gutter, skirting the medial border of the talus, gives some indication of the ankle axis, and fixing the tibial alignment guide slightly externally rotated relative to this pin is important. Alignment can be further assessed by ensuring that the tibial alignment jig is roughly aligned towards the 3rd ray(preoperative assessment of the intermalleolar axis on a CT scan can be very informative when assessing axial alignment intra-operatively).
Once alignment is optimal in all 3 planes, secure with three long pins.

Confirm that The Talar Block is well seated in the intermalleolar plane. Dorsiflex the foot to 90 degrees. if the ankle does not dorsiflex to 90 degrees, consider an achilles or gastrocnemius lengthening.

The talar cut is made through The Talar Cutting Block with the Tewkes saw using controlled probing and gliding strokes. The foot must be positioned at 90 degrees (plantigrade) for the talar cut, if the cut is made in plantar flexion, the talar component will be inserted in a malrotated, dorsiflexed position.

Loosen The Frontal Screw to free Talar Cutting Block.

Remove Talar Cutting Block.

The medial and lateral edges of the talar cut will need to be completed, using a fine power saw.

Select appropriate sized Tibial Cutting Block (small, medium and large).

Prepare The Tibial Cutting Block. Medial and lateral alignment can be adjusted using the horizontal scale to ensure optimal placement and to avoid excessive bone removal from the medial or lateral malleoli.
This image demonstrates an offset placement.

Insert the Tibial Cutting Block onto the Tibial Alignment Guide, whilst maintaining the selected medial/lateral displacement on the block. Confirm central placement relative to the malleoli. This can be assessed with image intensifier.

Tighten The Frontal Screw to secure the Tibial Alignment Guide.
Insert the appropriate Tibial Tensioner through the slot in the Tibial Cutting Block. This will determine the thickness of the distal tibial resection. A 5mm Tensioner will is used for minimum bone resection. In most circumstances I use a 6mm.

The tibial tension ratchet is used to balance the joint and achieve optimal resection of the tibia. The tab of The Ratchet Button is flipped upwards to unlock it. The Knob Tightener is inserted into the Ratchet Knob and turned in anti-clockwise direction using gentle finger tension. The amount of tension applied will represent the amount of tension in the replaced joint. Increasing tension reduces the amount of bone resected from the distal tibia. If the selected position of the tibial resection cut appears too distal go back to the Ratchet starting position and select a larger (6 0r 7 mm) Tensioner.

The central lug holes are completed using the 4.5mm drill which should be inserted through the Tibial Cutting Block up to the drill stop.
Select the appropriate depth of drill (S,M,L) for the cutting block.

The 3.2 mm Tibial Corner Drill is laser marked, select the appropriate depth (S,M,L) to pass the drill.

The Tibial Corner Drill holes are completed. The rounded corner the drill hole provides reduces the risk of stress fracture of the malleoli. a pin can be placed in the corner drill holes to protect the malleoli during saw resection.

The Tewkes saw is used to complete the medial and lateral vertical cuts. care should be taken to protect the soft tissues and with the medial cut, be aware that the neurovascular bundle, Tibialis posterior tendon and the flexor hallucis longus tendon are vulnerable to injury if the saw is passed too deeply.

The transverse tibial cut is made using the Tewkes saw, taking care medially and laterally to avoid notching the malleoli. A pin left in the medial tibial corner drill holewill help protect against notching of the medial malleolus.

Loosen the Frontal Screw.

Remove the tibial jig, it is sometimes helpful to gently lever this off with an osteotome.

The tibial cut is completed using The Tibial Corner Gouge, which is inserted into the two corner drill holes. The gouge tends to jam tight and usually needs to be extracted using the slap-hammer to avoid levering against the malleoli.

An osteotome is used to free up the transverse cut.

The resected tibia, often cannot be removed en-bloc and need to be fragmented using a small osteotome. Posterior fragments are best accessed by distracting the joint with a laminer spreader, the soft metaphyseal distal tibia should be protected from the spreader with a broad osteotome, as shown in the image. The posterior fragments are often adherent to the posterior ankle capsule and it takes considerable care, attention and patience to ensure that the joint is cleared.

Confirm sizing (S,M,L) using the Tibial Length Gauge. If there is a need to increase the tibial implant size, select the next biggest Tibial Cutting Block, The central 4.5mm and corner holes must be re-drilled and the tibia re-cut.

The Tibial Keyhole Cutter is passed into the central drill holes with the fin positioned perpendicular to the resected plafond. Take care to pass the cutter directly down the drilled hole to avoid widening the drilled tunnel.

The Talar Champfer Guide is applied to the flat cut surface of the talus slide it forward until the anterior champfer abuts the front of the talus. insert the appropriate thicknes Flat (blue) Spacer and assess tension in the joint moving from plantigrdade through plantar flexion. The tension should be equal, throughout the range of movement, thus confirming adequate resection.

Anterior talar osteophytes often cause The Talar Champfer Guide to be placed too anteriorly. It is usually necessary to resect the anterior talar articular margin using a rongeur so that the guide can be seated more posteriorly to gain the optimum antero-posterior position.

Optimum position of the Talar Champfer Guide should be confirmed on Image Intensifier.

The Talar Champfer Guide is held seated with the Talar Lever and secured in place using two short pins . The guide should be placed centrally with no overhang medially or laterally, and parallel to the longitudinal axis of the talus

The talar pegs are created using Talar Peg Drill passed through the Talar Drill Tube. Use the Talar Lever distract the joint and hold down The posterior part of The Talar Champfer guide whilst drilling.

The posterior talar champfer cut is made using probing, sweeping cuts with the Tewkes saw.

The Talar Champfer guide is removed and the champfer cut is to be completed using a fine saw. Plantar flexing the ankle allows excellent visualisation of the posterior talus.

the resected bone is removed

There are often areas of sclerotic bone on the dorsal talus. A short pin can be used to ‘pepper-pot’ the subchondral plate, aiming to maximise bony ingrowth into the talar component.

Talar trial is inserted carefully lining up the two pegs with their corresponding peg holes.

The talar trial is impacted using the Talar Impactor. Inspect the trial circumferentially to ensure that it is flush to the cut talar surface.

the Tibial trial is inserted using the two pronged tibial impactor and lining up the implant with the two tibial lugs. The spacer protects the talar component, and ensures that the tibial trial is pushed up against the cut surface of the distal tibia as it is inserted. It is helpful to use a spacer that is 1 mm thicker than planned to ensure a snug fit.

Insert the appropriate sized and thickness Meniscal Trial, using the Meniscal Trial Inserter.

With the trial in situ, test dorsiflexion and plantar flexion, plus varus and valgus stability. If too tight, a smaller spacer might be selected. The Meniscal trial should traverse antero-posteriorly by approximately 5mm during ankle movement. The meniscus should remain in full contact with both of the metal trials throughout flexion/extension and internal/external rotation. Fixed equinus may require an achilles lengthening or gastrocnemius recession.
Check the position of the trials with image intensifier. Antero-posterior adjustment of the tibial component may be needed to optimise tibial coverage.

The spacer is removed with The Meniscal Trial Extractor.

the tibial trial is removed with The Tibial Trial Remover, which is clipped over the posterior edge of the trial and withdrawn with the slap-hammer.

the talar trial is removed with the Talar Extractor, which is clipped around the corresponding recesses in the anterior champfer of the trial.

Clean the resected bone surfaces with pressurised lavage to remove debris and dry thoroughly.

The appropriate sized implants are prepared. The Talar Component is lined up, with the pegs lined up to their corresponding peg-holes. using a non-touch technique.

The Talar Component is impacted in-line with the pegs using The Talar Impactor. Inspect the implant circumferentially to ensure that it is flush to the resected surface.

The tibial component is placed on to The Tibial Inserter, using a non-touch technique. Beware, the Tibial Implant is not secured on the Inserter, and is at risk of slipping off if the inserter is tilted from the vertical position.

Insert the Tibial Implant with the Tibial Inserter, use the green Profile Spacer to protect the bearing surfaces of the implants and to keep the Tibial Component pressed firmly to the resected tibial surface. It is helpful to use a spacer one size thicker than the planned meniscus. Insert the Tibial Component to the optimum depth as established by the Tibial Trial.

Orientate The Meniscal Implant, it has two raised Marker Ball Pads anteriorly, and on posteriorly.

The Meniscal Implant is inserted, applying (often considerable) pressure with one thumb on each Marker Ball Pad.

Assess movement and stability of the implant. The Meniscal Component should remain in contact with the two metal bearing surfaces throughout the range of movement.

It is advisable to release the tourniquet and control local bleeding before closure. There is usually a cuff of anterior ankle capsule that can be sutured over the implants, care should to be taken to avoid incorporating the adjacent neurovascular structures laterally.

The superior extensor retinaculum is repaired suturing adjacent crenelations. Repairing the extensor retinaculum is important to prevent bow-stringing of the extensor tendons.
The subcutaneous tissues should be repaired over the retinaculum, taking care to protect the superficial peroneal nerve.

Ensure that the ankle is plantigrade. if not consider an achilles or gastrocnemius release.

How much weight bearing allowed early depends upon the quality of the press fit of components and a Surgeons preference. My own is to limit all weight-bearing for 2 weeks with a light weight cast. Weight bearing in a below cast as pain allows for 2-4 weeks, this allows the soft tissue inflammation to settle, and does not have the disastrous effect upon range of movement (or indeed I think any effect) that one would expect if treating a knee replacement in a similar fashion. from 4(6)-12 weeks mobilisation fully weight-bearing in a pneumatic boot working on active and passive range of movement. The incidence of early post-operative peri-articular fracture is significantly reduced.
Range of movement can be worked upon once anterior wound has healed.
Post operative follow up xrays weeks 6,12 and annually.
Follow up is advised longer term follow up has shown that with all total ankle replacements, there is an incidence of cyst formation and aseptic loosening, these maybe managed early with bone grafting to potentially increase longevity and prevent presentation with catastrophic failure, as revision surgery, whether to a fusion or a replacement is more successful when there is a reasonably preserved bone-stock.

Giannini et al. CORR. 2010; 468(10): 2746-53
Report 51 BOX ankle replacements in 7 centres. Outcomes were assessed using AOFAS scores and radiographs with minimum 2 year follow up. The mean preoperative AOFAS score of 38.5 increased to a mean of over 75. with a survival rate of 97 percent. They concluded that the BOX provided good short term results with high patient satisfaction and low revision rates, but conceded that longer follow-up studies are needed.
Zaidi et al. Bone Joint J. 2013;95-B:1500–7
Meta-analysis, of over 50 papers providing clinical data on TARs concluded that TAR has a positive impact on patients’ lives with benefits continuing over 10 years. The review included data functional data on a number of prostheses including the BOX Ankle, whose results were comparable with other modern designs.
Affatato et al. J Biomech. 2007; 40: 1871-76
Biomechanical study: BOX Ankle replacements tested in a wear simulator, put through load-motion patterns typical of a replaced ankle. Demonstrated BOX Ankle uniform motion through range, with full conformity maintained between components and produced very low level of wear.

Bianchi et al. JBJS-Br. 2012; 94(6): 793-8
62 BOX TARs overall survival was 91.9% at 4 years. The mean AOFAS score had improved by over 40 points at final follow-up. Single surgeon (not involved in design).
Cenni et al. Foot Ankle Int. 2012; 33(4): 290-300
3D video-fluoroscopic analysis of 12 patients, satisfactory recovery of the natural overall motion at the three anatomical planes at 6 months and maintained at last follow up 3 yrs postoperatively, Restoration of the natural parameters of joint rotation axes.
Sopher, Med Eng Phys 2017, Apr 42.
Total ankle replacement design and positioning affect implant-bone micromotion and bone strains
This study assessed micromotion at the implant bone interface, excessive micromotion is related to early implant failure. They found of the three most commonly used implants at that time, the BOX total ankle replacement demonstrated the lowest level of micromotion. Fixation close to the implant, more than one fixation ‘peg’ were thought to be factors that contributed to stability in the BOX ankle.
UK NJR Data 2017
734 primary total ankle replacements were implanted in the UK in 2017. 14.4% were BOX ankles.
The NJR estimated that the overall cumulative percentage probability of revision surgery was 8.7% at 7 years


Reference

  • orthoracle.com
Dark mode powered by Night Eye