In the late 1960s papers began to appear describing the development of a flatfoot deformity in association with a tear of the tibialis posterior tendon. Most of the early cases were traumatic, but in 1969 Kettlekamp and Alexander described several cases with no traumatic history. In 1989 Johnson and Strom classified the condition of "tibialis posterior tendon dysfunction" and proposed a system of treatment. Tibialis posterior tendon dysfunction came to be seen as " the commonest cause of the adult acquired flatfoot". Physical signs, as discussed later, were viewed as evidence of tibialis posterior insufficiency. Johnson and Strom discussed staging as though there were an established progression from tenosynovitis without foot deformity, through partial to complete rupture of the tendon with increasing but flexible deformity to fixed deformity without tendon function. Johnson and Strom’s treatment recommendations for most patients concentrated on replacing the function of the absent tendon.
However, further research over the last 10 years has shown that the situation is much more complex than previously thought
- Dyal et al showed that 70% of patients with unilateral tibialis posterior insufficiency had a contralateral flat foot, implying that the symptomatic foot was probably flat to start with.
- Division of the tibialis posterior tendon in a normal foot cadaver model does not in itself necessarily produce a flatfoot deformity – division of the plantar fascia, long and short plantar ligaments and deltospring ligament complex are required.
- The tibialis posterior is inactive in quiet standing although it is active during mid-stance phase of walking. In standing the main support of the arch is the plantar fascia.
- Singh et al showed that patients without tibialis posterior function due to neurological deficit or tibialis posterior tendon transfer do not necessarily develop a flat foot, even in the presence of peroneus brevis function.
- The cardinal physical sign of tibialis posterior insufficiency is generally considered to be inability to perform a single foot tiptoe test, but Briggs et al showed that over 50% of heel supination on tiptoe in the normal foot is due to the plantar fascia windlass effect. Results in flat feet are awaited.
- Some patients have an obviously functional tibialis posterior tendon yet have progressive foot deformity with the apex at their degenerate 1st TMT joints. This was noted in 1982 by Henceroth and Deyerle, but has been largely ignored since.
It is notable that a symposium in Clinical Orthopaedics and Related Research in 1999 was entitled "Adult Acquired Flatfoot". Probably adult flatfoot and tibialis posterior tendonopathy should be thought of as inter-related conditions rather than as a single disease entity. It seems likely that, in most patients, a pre-existing flatfoot may predispose to tendonopathy which permits the progression of deformity, and so on as a vicious cycle. This view of the syndrome would have the important corollary that attention to the tendon will be of little value unless the deformity is addressed to improve the biomechanics of the foot.
Epidemiology
The prevalence of adult acquired flatfoot or tibialis posterior tendonopathy has not been established. It is quite likely that there are many more or less asymptomatic persons in the community which makes it difficult to measure the numerator. In Blackburn we see about 20 new patients a year in a catchment area of 260,000.
Tibialis posterior tendonopathy is commonest in middle-aged females. Myerson et al (1989) also identified a group of younger male patients with seronegative arthropathies and synovitis. A few elderly patients present with stiff arthritic flat feet and a history which suggests they may have started with tibialis posterior insufficiency.
Some workers have examined the relationship between tibialis posterior tendonopathy and rheumatoid hindfoot deformity. Keenan et al (1991) found increased tibialis posterior EMG activity in rheumatoid flatfeet – although this would not exclude the possibility that the increased activity was ineffectual because of tendonopathy. Coakley et al (1994) found no clinical evidence of insufficiency in 18 rheumatoid valgus feet. Michelson et al(1995) found clinical evidence of tibialis posterior insufficiency in 11% of rheumatoid feet. Jernberg et al (1999)found tibialis posterior tendonopathy in 12/19 rheumatoid patients with flatfoot, but only 7/21 in non-flat feet. However, there were only 2 complete ruptures on MR scan and MR and clinical findings correlated poorly. Tibialis posterior tendonopathy probably contributes to the development of rheumatoid planovalgus deformity, but joint instability due to synovitis and abnormal biomechanics due to other deformities is probably more important. This may also be relevant to the "typical" adult acquired flatfoot.
Some patients give a clear history of a forced valgus injury which precipitates their problem. 48% of Myerson’s series had such a history. Traditionally this has been viewed as a missed tibialis posterior tear, and such injuries have been described. However, it may be that these injuries are more complex and involve, in particular, the deltoid and spring ligament complex.
Holmes and Mann (1992) reviewed the general medical conditions in 67 patients with tibialis posterior tendonopathy and found that half had hypertension, diabetes mellitus or obesity – obesity had the greatest association followed by hypertension. While it is intuitive that obesity may promote deformity of an unstable foot, the other conditions may simply be markers for middle-aged relatively unfit patients.
Pathology
Mosier et al (1998) showed that the pathology of tibialis posterior tendonopathy is much the same as that of other tendons such as the Achilles tendon and rotator cuff. They compared 15 tendons with surgically proven tendonopathy with 15 cadaver controls. There was thickening of the surgical cases, but no loss of continuity. The alignment of collagen bundles was lost, there was mucin deposition, myxoid degeneration, hypercellularity and neovascularisation.
Frey et al (1990) identified a 14mm zone of relative ischaemia in the tendon between the medial malleolus and the navicular. This zone also has a sharp turn round the medial malleolus, so that Jahss (1982) suggested that this was an attritional rupture associated with stenosing tenovaginitis. However, it seems more likely that tibialis posterior tendonopathy is a principally degenerative condition.
Some recent studies have examined evidence of autoimmune processes and abnormalities of collagen metabolism – this approach may yield important information in future but as yet no conclusive data have emerged.
Staging and classification
The natural history is believed to be a progression from tendonopathy without excess deformity, through a mobile deformity to a foxed deformity. However, few patients have been followed to demonstrate progression of the condition. In the Blackburn series, Hattrup + Johnson stage 3 patients were older than those in stage 2.
Hattrup and Johnson (1989) proposed a staging system which is in general use in the orthopaedic foot and ankle community. They recognised three basic stages:
- stage 1 with tendonopathy but normal tendon length and no deformity
- stage 2 with tendon lengthening and flexible planovalgus deformity
- stage 3 with tendon lengthening and fixed planovalgus deformity
Myerson subsequently added a stage 4 where there is also valgus tilting of the ankle in the mortise.
Hattrup and Johnson’s classification was based on clinical examination, with treatment recommendations as follows:
Stage |
1 |
2 |
3 |
Tendon condition |
peritendonitis/ degeneration
|
elongation |
elongation |
Hindfoot |
mobile, normal |
mobile, valgus |
mobile, fixed |
Pain |
medial, focal |
medial, along tendon |
medial + sinus tarsi/lat ankle |
Single heel rise test |
mild weakness |
marked weakness |
marked weakness |
"Too many toes" sign |
normal |
positive |
positive |
Pathology |
synovitis/ degeneration
|
degeneration |
degeneration |
Treatment |
conservative/ debridement
|
FDL=>tib post transfer |
subtalar arthrodesis |
Myerson added a stage 4, where there is fixed foot deformity and tilting of the talus in the ankle mortise. Dereymaeker also proposed a stage 0, where there is biomechanical abnormality but no symptoms – this acknowledges the more importance of pre-existing biomechanics but implies that stage 0 feet are likely to progress.
Unfortunately, there are a number of problems with the Hattrup and Johnson staging system:
- the organising concept in the text of the paper seems to be the state of the tendon, but the basis of the actual staging system is the physical signs
- most of the physical signs are those of planovalgus deformity rather than tendon disease, yet conclusions are drawn about the state of the tendon
- the paper and the staging system imply that the foot begins in a neutral posture, yet many patients have a pre-existing flatfoot
- no account is taken of different degrees of flexibility in the hindfoot and forefoot – indeed Johnson denied the existence of this (the two-piece concept of the foot – see below under subtalar fusion)
- no account is taken of the severity of deformity
- Conti et al described a separate MR classification of tendon abnormalities which cannot be readily integrated into the Hattrup and Johnson classification
- Hattrup and Johnson’s paper contains no actual data and no subsequent publication has examined the reliability and reproducibility of the classification, or how it affects clinical decision making
For all these reasons, it is probably time to re-evaluate the classification and staging of adult acquired flatfoot, acknowledging both the seminal work of Hattrup and Johnson and subsequent work. A revised classification should probably incorporate:
- a clinical classification of flatfoot deformity based on that of Hattrup and Johnson, but taking account of severity of deformity and stiffness at different levels
- radiological staging of deformity
- a classification of tibialis posterior tendonopathy, probably based on the MR classification of Conti et al and recognising that some patients with adult flatfoot deformity have no tibialis posterior disease
- rigorous evaluation for reliability, validity and effect on clinical decision making
A development of the Johnson and Strom classification was described by Parsons (the Truro classification). It principally divides stage 2 into three stages, depending on the severity and reducibility of the classification, and also recognises that stage 1 patients may have a pre-existing flatfoot deformity:
Stage 1: no deformity, or flatfoot deformity which has not progressed and is similar to the opposite foot.
Stage 2: Flatfoot deformity which has progressed but is fully passively correctible and in which forefoot varus is less than 15deg.
Stage 3: Flatfoot deformity which has progressed but is fully passively correctible and in which forefoot varus is greater than 15deg.
Stage 4: Flatfoot deformity which has progressed; the hindfoot is fully passively correctible but the forefoot is not.
Stage 5: Flatfoot deformity which is not correctible
Stage 6: Flatfoot deformity in association with tilting of the talus in the ankle mortise on standing radiography.
The Truro classification is usable by different professional groups and is fairly reproducible.
Myerson has proposed that there should be radiological criteria of the severity of deformity and is evaluating such criteria. It seems likely that there will be a revised classification within the next five years.
Clinical features
The typical patient is a woman between 45 and 65 with a history which often extends over some years. Like many patients with an insidious, relatively uncommon problem they may be used to their complaints not being taken seriously or put down purely to obesity. The commonest complaints are:
- posteromedial ankle pain and swelling, usually clearly related to the tibialis posterior tendon
- anteromedial ankle pain, related to the deltoid ligament
- sinus tarsi pain due to lateral subtalar impingement
- medial arch aching, sometimes going into the sole
- progressive planovalgus deformity – this is the primary complaint in relatively few patients
- a few have neurological complaints in the forefoot – there is a recognised association with tarsal tunnel syndrome
It is important to enquire about:
- inflammatory joint disease, including extra-articular features, inflammatory bowel disease and psoriasis
- trauma to the foot
- whether the foot has always been flat and whether the shape has actually changed
- what treatment has been tried before, particularly orthotics
All patients are routinely asked about:
- diabetes
- neurological disease
- circulatory problems
- family history of foot problems, arthritis or neurological disease
Sometimes the patient was not conscious of having "flat feet" although the opposite foot is obviously flat. Some patients are not particularly conscious of even quite gross planovalgus deformity and complain mainly of pain.
Examination often shows some degree of obesity. The opposite foot is usually at least somewhat flat but usually asymptomatic. Some patients have a markedly antalgic or stiff gait.
Patients with flat feet usually have excess wear of the medial heel and medial wall of the shoe at the level of the talonavicular joint. The affected foot is usually flat but about 10-20% of patients have neutral feet and complaints localised to the tendon. Examination from behind shows hindfoot valgus and forefoot abduction demonstrated by the "too many toes sign". Normally, no more than 3 toes can be seen lateral to the ankle when viewed from directly behind the heel, but in an abducted forefoot 5 or "more" toes may be visible. This sign simply demonstrates forefoot abduction, not the presence of tibialis posterior tendonopathy.
It is important also to look at the tarsometatarsal joints, especially the first. Some patients with an acquired flatfoot have primarily 1st TMTJ instability or arthritis and may have an obvious breach at this level, or large osteophytes.
Palpation of the tendon may demonstrate tenderness, thickening, synovitis or a thinned or apparently absent tendon. The deltoid and spring ligaments should be carefully palpated. Tenderness may be found in the sinus tarsi, especially with hindfoot abduction.
In a flexible flatfoot the heel normally corrects into varus and the arch reconstitutes on tiptoe standing. Two abnormal signs have been described in tibialis posterior insufficiency:
- failure of the heel to move into varus on double tiptoe standing; if the opposite heel goes into varus "windswept heels" may be noted
- inability, or extreme difficulty, in doing a single foot tiptoe test; in a few patients the heel comes off the ground but a midfoot breach develops and the forefoot does not rise.
Obviously, if the hindfoot is stiff these tests will be meaningless. Tansey also demonstrated that heel varus on tiptoe is mainly controlled by the plantar fascia windlass mechanism. No study has been reported correlating these findings with radiological, gait analysis or surgical findings. Hence the accuracy of these tests is unknown and they may be much less useful than commonly thought.
It is important to establish the range of ankle, subtalar and midtarsal movement. In assessing the range of ankle dorsiflexion, it is essential to test with the heel in neutral, otherwise a large range of apparent ankle dorsiflexion may be produced by subtalar eversion. Most patients have a very tight Achilles tendon which may not be detected unless ankle dorsiflexion is carefully assessed.
Even if the subtalar joint can be completely reduced, there may be fixed forefoot varus/supination which can be appreciated from behind with the patient prone, or from the front. This is of considerable importance on planning treatment (see below). Some patients have a gross peritalar subluxation and the reducibility of this can be assessed.
The strength of the tibialis posterior is best tested in plantarflexion-inversion to exclude the help of tibialis anterior. Resisted eversion tests muscle strength and irritability.
At the end of the clinical examination, the examiner should be able to answer the following questions, which will guide treatment:
- what is the main problem – tendon pain, ankle pain, deformity, lateral impingement, arthritic pain
- is there an apparent underlying cause such as inflammatory arthritis?
- is the foot deformed?
- is the deformity mainly peritalar or cuneometatarsal?
- is the hindfoot deformity correctable?
- is there fixed hindfoot/forefoot malalignment?
- is the Achilles tendon tight?
- are there any associated problems – hallux valgus, secondary arthritis, tarsal tunnel syndrome?
Imaging
The diagnosis of adult acquired flatfoot is made clinically. If there is no controversy about diagnosis or management we do not routinely obtain radiographs. However, in many patients there is a need to assess the joints and alignment for treatment planning, and in some there may be the possibility of inflammatory arthropathy or a tarsal coalition.
The standard plain films are:
- standing hindfoot alignment view – we use a technique slightly modified from that of Saltzman and el-Khoury (1995) to improve the visualisation of the ankle – many people refer to this as the Cobey view. It shows the hindfoot alignment and any tilting in the ankle mortise
- standing lateral foot view – shows peritalar and talometatarsal alignment and allows assessment of arthritis
- standing dorsoplantar foot view – as above
On these views we assess the following parameters:
- tibiotalar and tibiocalcaneal alignment
- peritalar malalignment using:
- lateral talometatarsal (Meary’s) angle
- dorsal talometatarsal angle
- talonavicular coverage angle
- cuneometatarsal alignment in lateral and dorsoplantar projections
- other deformities such as hallux valgus
- presence of arthritis, accessory navicular or tarsal coalition
- bone stock
- any previous surgery
If we are planning surgery, or if there is any other need to assess the tendons and ligaments, we obtain an MR scan. We have not found ultrasound reliable in assessing the tibialis posterior tendon, although other authors have described it and it has the advantage of often being quicker to obtain.
CT shows bony changes well, but does not show soft tissues clearly.
MR shows:
- tibialis posterior tendon and pathology
- other long flexor tendon pathology – FHL tendonopathy is sometimes seen in association with tibialis posterior tendonopathy
- the deltoid and, to some extent, the spring ligament
- intra-articular lesions in the ankle and other joints
- bone oedema, sometimes seen in the lateral calcaneum in lateral impingment, or in the navicular
Conti et al (1992) classified the MR appearances:
- type 1: longitudinal splits without tendonopathy
- type 2: swelling and degeneration
- type 3: replacement of tendon substance with scar
Conti et al found that the MR appearances were a better guide to outcome than surgical findings. In particular tendon transfers were significantly more successful in type 1 tendons, but tendons graded type 1 by the surgeon were graded type 2 by MR in 10/17 patients - intra-operative assessment may lead to inappropriate choice of treatment.
Non-surgical treatment
Non-surgical treatment includes analgesics, shoewear advice and sometimes adaptations, weight reduction, Achilles tendon stretching and local treatment to irritated joints and tendons. A few patients appear to be able to voluntarily substitute the use of FHL for the tibialis posterior in walking, with reduction in medial foot pain. Achilles tendon stretching must be done with the subtalar joint in neutral otherwise the stretching manoevre will simply force the subtalar joint into eversion. We teach the patient’s partner or other available person to do this.
There have been two studies of non-surgical treatment. Both Wapner and Chao (1999) and Jari et al (2002) found that 70% of patients were satisfactorily treated with orthoses and shoe modifications. Only about 10% were operated on, although a number of patients with significant continuing symptoms decided against surgery
Non-surgical treatment should be tried in every patient unless there are cogent reasons not to do so. Even imminent skin breakdown over the prominent talar head has been successfully treated with a Scotchcast diabetic boot followed by the use of a total contact inshoe orthosis.
Because the initial treatment is conservative in almost every case, patients whose referral letters imply that they have adult acquired flatfoot are primarily seen, in Blackburn, by the podiatrist rather than the surgical clinic.
Surgical options
Tendon debridement
For patients with tendonopathy but no deformity a simple debridement may be considered.
Two small series have been reported. 90% had good pain relief and 90% had normal single foot tiptoe tests.
This procedure may be satisfactory for a small number of patients. However, many patients have a pre-existing flatfoot and there is a strong case for improving the biomechanical environment of the debrided tendon with a posterior medial displacement calcaneal osteotomy. In addition, Conti et al (1992) showed that surgical inspection tends to underestimate tendonopathy and therefore early good results may not last. 10% of patients in the reported series developed progressive deformity and underwent further surgery.
Tendon reconstruction with transfers
The early series recommended reconstruction by side-to-side anastomosis of the tibialis posterior to FHL or FDL (Jahss 1975). Other authors recommended excision of the abnormal tendon and bridging of this gap, or of a rupture, by transferring the FDL into the navicular (Johnson 1989). Initial enthusiasm for this method reduced as it was realised that the arch was not generally recreated, the medial pain tended to recur and other lesions such as deltoid and spring ligament tears were not dealt with.
Another tendon transfer method was described by Cobb and popularised in the southern UK and some other circles by Helal. The tibialis anterior tendon was split, passed through the medial cuneiform or the tibialis posterior insertion, then through the tibialis posterior sheath and attached to the tibialis posterior stump proximally. This gave a large, strong reconstruction which seemed more able to restore the arch.
Gazdag and Cracchiolo (1997) reported 22 patients who had reconstruction of the tibialis posterior, mostly with FDL transfer, and most also had spring ligament reconstructions. 70% had good pain relief but all had residual deformity. There have been only anecdotal reports of the Cobb procedure.
Several other authors have described in vitro models of flatfoot deformity and reconstructive procedures ( Deland et al 1992, Thordarson et al 1995, Kitaoka et al 1997, 1998) but, while interesting, these have not translated into clinical results. It began to be realised that symptomatic improvement required measures to improve the biomechanics of the foot.
Medial displacement calcaneal osteotomy
The simplest addition to the tendon transfer procedures was a transverse osteotomy of the posterior calcaneum with medial translation of about 1cm, and internal fixation with one or more screws. Originally described by Koutsogiannis (1971), this technique has been promoted particularly by Myerson. The osteotomy is made 1cm posterior and parallel to the peroneal tendons. A chevron osteotomy with the apex anteriorly has been proposed to increase stability and allow earlier weightbearing. The biomechanical rationale is that realigning the vector of the Achilles tendon will transform this tendon from an evertor of the subtalar joint to a neutral effect.
The largest series is that of Myerson (2004). In 120 patients, pain and function were markedly improved in 90%, and arch height was also restored with few recurrences. Poor results were mainly in those with severe deformity pre-operatively.
Two smaller series using FDL (Guyton 2001, Wacker et al (2002)) and one using FHL (Sammarco 2001) reprted good clinical results but less improvement in the arch: Sammarcon had no patients with improvement in clinical or radiological measures of arch, Guyton had 50% and Wacker 75% (with 10% recurrence at 3-5y).
This procedure appears able to give a more durable result than tendon transfer alone. Outcome measures are relatively consistent between series, with reasonably good improvement in pain and function but variable effects on arch height. It will be important to see whether these results are maintained in a procedure which does not fully correct foot alignment.
Lateral column lengthening
Because of dissatisfaction with the ability of the medial displacement calcaneal osteotomy to restore arch anatomy, a number of authors have studied the effects of lengthening the relatively short lateral column of a planovalgus foot with forefoot abduction. This concept was originally popularised by Evans for treating over-corrected clubfeet and most of the older papers describe the treatment of paediatric flatfoot deformities.
Two techniques have been described:
- A transverse osteotomy of the calcaneum 10mm proximal to the calcaneocuboid joint is distracted to correct forefoot abduction and a tricortical iliac crest bone graft inserted. Various fixation techniques have been used. This preserves calcaneocuboid movement but one study found a marked increase in the pressures in the calcaneocuboid joint, raising concern that patients might eventually need calcaneocuboid fusion for OA. A subsequent study using a better flatfoot model suggested that calcaneocuboid pressure is increased in the flatfoot anyway and that calcaneal osteotomy does not chage this. The reported rate of calcaneocuboid fusion for OA after osteotomy is probably about 5%, although not all papers report this clearly. Pressures in the subtalar joint may also be increased.
- Distraction fusion of the calcaneocuboid joint also lengthens the lateral column and avoids a second procedure for OA. However, the non-union rate is about 15%.
There has been only one (retrospective) comparative series between osteotomy and fusion. All patients also had FDL transfer to the navicular. Mean AOFAS scores were 81 for fusion and 88 for osteotomy. Overall the results of the fusions were no better than the osteotomies, they took longer to recover and had more complications. However, the numbers were small and more osteotomies than fusions were lost to follow-up. The authors appear to adopted the fusion procedure later than the osteotomy, though without abandoning the osteotomy. Frustratingly, they do not comment on this change in practice, whether the treatment groups were comparable or whether there were particular indications for the fusions in later patients, or implications for the results.
Simialr results have been reported from single-procedure studies (Hintermann 1999, Moisier-LaClair 2001). Moisier-LaClair’s series had anterior and posterior (as described in the previous section) osteotomies, intending to share the displacements between the two osteotomies and reduce skin tension; wound problems may have been less frequent but 14% failed because of persistent pain.
Theoretically, lateral column lengthening should give superior correction to tendon transfer plus posterior calcaneal osteotomy alone. Series of lateral column lengthening report improvement in the arch more consistently, but the clinical scores appear similar and all studies report a lot of wound problems and lateral column pain, which may offset the possibility of improved correction. There have been no published series comparing the calcaneal osteotomy plus tendon transfer to lateral column lengthening. It may be that the lateral procedure is more effective in patients with more severe deformities (Truro stage 3), but so far there is no evidence on which to judge this hypothesis.
Subtalar fusion
Johnson and Strom (1989) recommended subtalar fusion for correction of the acquired flatfoot when the deformity of the foot was fixed (stage 3). This was based on the concept that the midfoot and forefoot would follow the calcaneum – the "two-piece concept".
Three series of around 20 patients, one from Kitaoka (Johnson’s successor at the Mayo Clinic), report good radiographic correction, with clinical scores similar to those of lateral column lengthening (mean AOFAS scores about 80) and no non-unions. Most patients also had FDL transfers, so the fusion was really an alternative to posterior calcaneal osteotomy. One series used fusion particularly for lateral impingement pain. Interestingly, Kitaoka commented that subtalar fusion is not appropriate in patients with midfoot supination as the forefoot deformity is not corrected – a denial of Johnson’s “two-piece concept”.
Subtalar fusion seems to be a useful procedure, although for indications other than described by Johnson. In particular it may be useful for patients whose main complaint is lateral impingement pain.
Talonavicular fusion
As the commonest deformity of adult acquired flatfoot is maximal at the talonavicular joint (Kitaoka et al 1998) it seems logical to consider stabilising this joint to restore the anatomy. Talonavicular fusion produces marked limitation of hindfoot motion in a cadaver model: 70% of pronation/supination and 25% of dorsiflexion/plantarflexion (Gellman et al 1987). Fogel et al (1982) performed gait analysis on patients 9 years after talonavicular fusion and found up to 30% loss of inversion/eversion, particularly on slopes, and reduction in late-stance plantarflexion.
One clinical series reported 26 patients who had talonavicular fusions for "the disorder known as insufficiency of the posterior tibial tendon". There was one non-union and one major wound problem. Over half had some residual pain, usually in the lateral midfoot. OA had developed or progressed in five surrounding joints. It is not clear whether the patients were selected for severe deformities or the procedure was applied “across the board”; there was no record of pre-op or post-op deformity.
In Blackburn we use talonavicular fusion particularly for some patients with severe deformities and fixed hindfoot-forefoot malalignment. Good correction of deformity is usually achieved, but we have also seen patients who develop lateral column pain.
Double fusion
This procedure fuses the talonavicular and calcaneocuboid joints. It aims to increase the stability of the midfoot correction and to avoid lateral column pain after isolated talonavicular fusion. It is particularly indicated where both midtarsal joints are arthritic, and in very obese patients.
One series (Mann and Beaman 1999) reported 16 patients who had double arthrodesis for tibialis posterior insufficiency. Mean AOFAS score was 79. The non-union rate was 25%, none of which had compression fixation. Although radiological correction was good, most feet appeared flat "but symmetrical".
Triple fusion
For patients with severe fixed deformities or degenerate changes in all hindfoot joints a triple fusion is generally indicated. This allows correction of all elements of the deformity and stabilisation of the correction, but greatly alters the biomechanics of the foot.
Three series (Graves, Mann and Graves (1993), Fortin and Walling (1999) and Coetzee and Hansen (2001)) report good improvement in pain and AOFAS scores about 75. Patients tended to have problems with slopes, uneven ground and stairs. Radiological correction was good. The non-union rate was about 5%, and there were some overcorrections, wound problems, fibular tip pain and nerve injuries.
Most patients were corrected by minimal resection of the joint surfaces and rotation/translation of the joints to the corrected position. Two modifications of this technique are worth having in the armamentarium:
- Horton and Olney (1995) used a bone block to distract the calcaneocuboid fusion site and thus lengthen the lateral column.
- Henderson advocated radical joint resections in very severe deformity.
Extended tarsal fusion
A few patients have ankle OA or develop it after initial treatment. The surgical options here are:
- extended hindfoot fusion with at least a tibiotalocalcaneal fusion and usually fusion of the talonavicular, calcaneocuboid and possibly lesser tarsal joints also. The use of a retrograde nail helps with stabilisation of such procedures
- osteotomies and/or arthrodesis to correct the hindfoot deformity and any necessary soft tissue reconstruction such as FDL transfer to the navicular or deltospring ligament repair, plus ankle replacement
There are no useful outcome figures available for these procedures.
Management guidelines
The available evidence is not sufficient to produce conclusive practice guidelines. In addition, many issues are not addressed in the existing literature at all. However, we need some reasonable guidelines and the following are proposed:
- in the first instance, most patients are treated symptomatically, with analgesia and physiotherapy for tendon and joint pain and a corrective in-shoe orthosis for flexible planovalgus deformity
- patients with severe or stiff deformities need accommodative orthoses, AFOs or braces
- patients may need accommodative shoewear
- patients who fail the above treatment will have deformity series radiographs and an MR scan
- if these confirm tendonopathy they may be offered surgery using the following guidelines:
- tendonopathy without deformity, Conti type 1 MR changes – debridement
- tendonopathy without deformity, Conti type 2/3 MR changes – debridement and FDL transfer
- tendonopathy with fully flexible mild/moderate planovalgus deformity – medial displacement calcaneal osteotomy, debridement and FDL transfer
- tendonopathy with fully flexible severe planovalgus deformity – Evans calcaneal osteotomy, debridement and FDL transfer
- tendonopathy with planovalgus deformity with subtalar but not midfoot stiffness – corrective subtalar fusion, debridement and FDL transfer
- tendonopathy with midfoot but not subtalar stiffness – talonavicular or double fusion
- tendonopathy with generally stiff planovalgus deformity – triple fusion
- patients with deformity but no tendonopathy may be managed similarly except
- omitting soft tissue reconstructions
- their pain is more likely to be due to OA so they are more likely to need fusions
- patients with tarsometatarsal instability or OA may need realignment arthrodesis and their other surgery adjusted accordingly
|