Annals ofthe Rheumatic Diseases 1990; 49: 851-857

The rheumatoid foot: a sideways look

Eric G Anderson

As with many other aspects of the foot, therheumatoid foot, historically, has been treatedsymptomatically. If there is a painful teno-synovitis, synovectomy; a painful joint notresponding to drug treatment, fusion; bonyprominences (including metatarsal heads),excision; if the structure collapses, a valgus'insole'; and if all else fails, bespoke footwearand discharge.

But is this good enough for 1990 when our

knowledge of orthopaedic biomechanics hasdeveloped to a degree unthought of 20 yearsago, our surgical expertise has been refined toan extent that was not possible 40 years ago, andthe consequences of our treatment are appreci-ated as they could not have been when rheuma-toid arthritis was first described? The answer

must surely be an emphatic 'No'.Before considering our approach to the ortho-

paedic management of foot disease we mustfirmly establish our objectives. We must con-

sider the foot as a mechanism, both in itsfunctional anatomy and pathology, for unlesswe understand what the disease process does tothe functioning of the foot our managementplan is liable to be misdirected and the gain tothe patient less than optimum.Of course, the foot does not function in

isolation. It is at the end of a kinetic chain and atthe interface between ground and body. Itsrelation to the more proximal joints must alwaysbe at the forefront of one's thinking (fig 1).The possibilities and limitations of surgery

must be defined, the contributions of orthoticsand footwear explored, and then, and only then,can the place of these modalities be evaluated as

an integrated treatment plan for any one

subject.

Table I Objectives intreatment

Maintenance of independenceRelief of painPreservation of foot functionProtection of feet

Footwear provisionChiropodical care

Department ofOrthopaedic Surgery,Western Infirmary,Glasgow Gll 6NTE G Anderson

ObjectivesIt is easy to look for local objectives in the foot:freedom from, or reduction in pain, for example.But feet are patients' most direct interface withtheir environment. With the common duplicityof joint dysfunction, continued function of thefeet becomes a major factor, and thus main-tenance of the independence of the patient mustbe a first priority. This, of course, does notimply pain free locomotion without the use ofaids. The ideal is not always the best objective.Consideration must be given to the subject'sparticular needs (table 1).

Maintenance of independence implies thatfoot function is of paramount importance andthat treatment, however well intentioned, doesnot compromise it. This sideways look will bemuch concerned with this aspect, rather than

I

Figure I The kineticchain.

I

A

\1 C)

I

851

on 18 May 2018 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.49.Suppl_2.851 on 1 O

ctober 1990. Dow

nloaded from

852

delving into the niceties of the exact origins ofpain in the rheumatoid foot. Pain, after all, is asubjective phenomenon, and its alleviation mayrequire more than a surgeon or orthotist canoffer. Nevertheless, the relief of pain or theattempted relief of pain, must feature highly asan objective.

Last ofour primary objectives is the protectionof the foot. Within the shoe the rheumatoidfoot, liable to complications of dysvascularity aswell as deformity, needs chiropodical care ofskin and of nails. How often these are forgotten.And how much unnecessary suffering they cancause.These stated objectives also imply the ability

of the patient to cope with footwear. Howeverwell the foot functions, however long it remainspain free, the independence of that patient alsodepends on his/her ability to find suitablefootwear. Shape is not the only problem patientsmay have. Footwear can also be adapted to easethe gait pattern and other joints affected by thedisease process.

Functional anatomyOver the years much myth has been perpetuatedin the anatomy textbooks, particularly about thestructural function of the foot. Who was notbrought up on functioning arches and theprimary loading function of the first metatarsalhead? This, despite much thoughtful andpioneering work-for example, by Wood Jonesin the 1940s' and Hicks in the 1950s.2 Our ownwork, supporting that of Cavanagh, dispels theanatomists' concepts of forefoot loading withsignificant consequences for forefoot surgery.Space does not admit of a detailed treatise onthis subject, but we shall highlight the moreimportant features that directly influencemanagement.We now consider the basic foot structure in

the way Wood Jones did as two segmentedbeams, each segment bound firmly to the nextby strong plantar ligaments which allow flexionat the bone interfaces but prevent extension (fig2). These beams cross at the subtaloid joint, ajoint whose axis lies in an oblique plane at 42°plus or minus to the horizontal and 150 plus orminus medial to the saggital plane. The plusesand minuses are important: they indicate a widerange of normality and, like the ranges of actualnormal joint movement in the foot, these jointsvary more than almost any other joint in thebody. The system is complicated by the presenceof the midtarsal joints, usually working incombination, which permits rotation betweenthe forefoot and hindfoot. This combination of

Subtalar Midtarsal jointjoint

joints works effectively as a torque transmitterwhereby foot rotation results in lower legrotation and, of course, vice versa. This hassignificant consequences in the diseased hind-foot.The foot and its ligaments are intrinsically

stable at rest; the muscles of the leg are largelyinactive.3 4 They come into their own duringgait and are indispensable in the maintenance ofthe structure of the foot in action. Activity doesimply stress, but stress can be applied to thefoot without activity. Obesity, fortunately notsuch a common problem in the rheumatoidpatient, increases stress and must be tackledseriously when in the presence of foot symp-tomatology.

Arguably, the most important single musclein maintaining dynamic foot stability is theposterior tibial muscle, particularly in the rheu-matoid patient. Why should this be moreimportant than the peroneal muscles? Simplybecause of the natural tendency of the stressedfoot to collapse into hyperpronation (or valgus)when the tibialis posterior becomes attenuatedor ruptures. Loss of peroneal function is notaccompanied by such dramatic alteration in footstructure.

In the normal foot, when the forefoot is flaton the floor, the heel is straight and thesubtaloid joint is in its neutral position, neitherinverted nor everted. It is common to findnormal variations whereby the forefoot liesinverted or everted to the transverse planerelative to the hindfoot owing to the dispositionof the midfoot joint axes, but compensated forby available midfoot rotation or subtaloid move-ment, particularly eversion. It is clear, therefore,that loss of such compensatory movements inthese joints, for whatever reason, may result inabnormal rotational stresses being set up in themidfoot and the hindfoot on weight bearingwith the production of a typical symptomatologyand clinical picture (fig 3).The function of the toes has been the subject

of discussion and investigation.5 They areprimarily used as stabilisers when the footweight bears on the forefoot, and not forpropulsion, at least at normal walking speeds.Feet which tend to supinate need to maintainbalance, and the lateral toes claw desperately

Figure 3 Inverted (varus) forefoot: (A) uncompensated;(B) compensated.

Figure 2 Normalfootstructure.

Anderson

on 18 May 2018 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.49.Suppl_2.851 on 1 O

ctober 1990. Dow

nloaded from

The rheumatoidfoot

trying to maintain it. Pronating feet, on theother hand, find their forefeet and their toestaking up a relatively valgus posture, whichforward motion can only accentuate, especiallyin the hallux. Thus hallux valgus is an inevitableconsequence of posterior tibial division orrupture.Normal gait consists not only of antero-

posterior rocking of the foot but also of rotation.Three anteroposterior rockers are described:the body rolling over the heel as the calcaneumrolls inside the heel's skin like the wheel of atank inside its track; the tibia rocking over thetalus; and the metatarsal heads rolling over theplantar skin in a like fashion to the calcaneum.'Roll over' is therefore a more appropriatedescription for this latter phase of gait than'push off (fig 4).The high pressures seen under the hallux

simply reflect the maintenance of body posturerather than any active propulsion into space.Although mention has been made of metatarsalhead roll over and the pressures generatedthereunder, it has been shown that in thenormal population 48% have maximum forefootpressure under the second and third metatarsalheads, 17% under the first head alone, 11%evenly under the first, second, and third heads,and 24% have a large lateral component ofpressure under the fourth or fifth heads at leastequal to the pressure under the first, second, orthird heads (fig 5).6To return to our normal gait pattern: on heel

strike the foot usually hits the ground supinated,tends to pronate in stance phase, and supinateagain on roll over. Clearly any inability of theleg to rotate will affect the ability of the foot torotate within a functioning hindfoot complex,and contrariwise, a diseased hindfoot complexwill affect the natural rotational movements ofthe foot or leg, or both. Thus the foot isinextricably linked to knee and hip function aspart of the kinetic chain.

1 Extrinsic disease elsewhere in the leg whichalters the function of the foot. A clear exampleof this is the varus knee, which results in thefoot hyperpronating to produce a foot flat in thestance phase of gait as well as the pronatedposition adopted by simple standing (fig 6). Butthere are knock on effects too. The subtaloidand midtarsal joints may be stressed at theextremes of their ranges, resulting in permanentalteration of the joint axes and secondarydamage to the joint. Of more immediate impor-tance, the posterior tibial tendon is put underimmense strain and may attenuate or ultimatelyrupture.2 Intrinsic disease-the joints. There is atendency for either forefoot or hindfoot to beaffected, unusually both. In the former themetatarsophalangeal joints bear the brunt, butin the latter the midtarsal, subtaloid, and anklejoints may all be involved initially with asynovitis before the degenerative changesdevelop. The talonavicular joint is often affectedearly, and synovitis of this joint should belooked for routinely.3 Intrinsic disease-tendons and sheaths.Although any one of the tendons passing over oraround the ankle may become diseased, theconsequences are, as one might expect from theforegoing, much more severe for the foot if theposterior tibial tendon is affected. Early syno-vectomy is worthwhile to stem the diseaseprogress in an attempt to reduce or delay tendoninfiltration, incompetence, and ultimaterupture, with all the consequences this has forfoot function.7

PathologyThree basic pathologicalfoot:

problems affect the

Figure 4 The plantar articulations.

0O0G 000 00(d 00

\Q% % 24% >8%

FgureS Distribution ofpressure under normalfeet-normal patterns.

Normal Varus kneeFigure 6 Effect ofvarus knee onfoot alignment.

853

on 18 May 2018 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.49.Suppl_2.851 on 1 O

ctober 1990. Dow

nloaded from

Anderson

Eamination and investigationFrom discussion so far it should be evident thatexamination should include the whole leg.Exposure to include the knee is mandatory;more if required. Examination must includeactive and passive movements-active motionagainst resistance being useful in determiningthe integrity of tendons. The function of thehindfoot can be shown by standing the patienton tiptoe and looking at the heels from the rear(Do they invert?) and the shin from the front(Does it rotate externally?). For those patientswho cannot stand on tiptoe, passive dorsiflexionof the relaxed weightbearing hallux producesthe same results.

Radiographs should always be taken weightbearing, and a lateral view should be routine.Specialised views, such as the weightbearingmetatarsal profile, do not contribute much inthis instance, as it is often impossible to positionthe patient satisfactorily on account of pain. Aswith any surgery in the rheumatoid patient theanaesthetist will appreciate radiographs of thecervical spine.Much interest is now being shown in foot

pressure measuring devices. The accuracy andrepeatability of some is in doubt, and althoughmany multicoloured pictures and a mass of datalook impressive and can be readily accumulated,interpretation is difficult and much basic workremains to be done before this information canbe interpreted with any validity. To date thedynamic pedobarograph with the Sheffieldcomputerised analysis system is the mostreliable.8More helpful are simple video gait studies,

where the video picture can be analysed at alater date frame by frame. It is, however, timeconsuming and needs to be viewed by theclinician personally. Typed reports can bemisleading and are next to useless. Photographybefore operation is a neglected method ofrecording, and more consideration might begiven to obtaining good clinical photographs atstrategic points during treatment.

Routine blood examinations should not needfurther discussion.

Treatment armamentariumSURGERYFirst in the list, only because it is a surgeon whowrites. The options in surgery are limited andbasic (table 2).Osteotomy is helpful in altering the alignment

of structures, particularly in situations of fixeddeformity, but care must be exercised in ordernot to replace one problem with another.

Excision of part or all of one of the bones ofthe small toes still has a useful place. Amputationof all toes does not.

Joint replacement arthroplasty has been moresuccessful in the ankle joint than in the footitself; its use in the first metatarsophalangealjoint is debatable, and in the second to fiftheven more so. Excision arthroplasty is still themainstay of forefoot rheumatoid surgery in oneform or another."'

Fusion is a much underused and under-estimated procedure which deserves more atten-

tion in rheumatoid disease. It has a significantpart to play in the prevention of the develop-ment of structural deformity.Tendon sheaths have been subjected to syno-

vectomy for a long time in the history of thedisease. Repair of the acute rupture of theposterior tibial tendon is a very worthwhileprocedure, but too often it is seen too late forprimary repair to be successful. Then, or ifdiseased beyond salvation, tendon transfer maygive acceptable results. Tendon transfers alsofind a place in the treatment of the retractedsmall toe, giving a cosmetically acceptable,though functionally equally worthless, digit.

Simple tenotomy has little to offer, exceptperhaps in combination with other procedures.

ORTHOTICOrthoses are external appliances which influencethe function of a part in one way or another.Three groups should be considered:1 Orthodigital splinting with a silicone rubberputty, which can be moulded directly to thepatient's toes to protect or correct mobiledeformity. Latex protective covers also fall intothis category (fig 7).2 Foot orthoses. It seems rather obvious tostate that all feet are different, hence thereshould be an insistence on all foot orthosesbeing made to foot casts. Much is possible nowwith new techniques of fabrication, newmaterials and, most importantly, new under-standing of the biomechanical function of theseorthoses (fig 8).

Figure 7 Orthodigital splint.

Figure 8 Functional castedfoot orthosis.

Tabk 2 Surgwcal optons

BoneOsteotomyExcision

JointArthroplasty

ExcisionReplacement

RealignmentFusion

TendonSynovectomyTransferDivisionRepair

854

on 18 May 2018 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.49.Suppl_2.851 on 1 O

ctober 1990. Dow

nloaded from

The rheumatoidfoot

3 Leg orthoses. The ankle foot orthosis ismost used, sometimes in conjunction with aseparate foot orthosis. Although the 'drop footsplint' is the commonest in general orthopaedicpractice, the rigid ankle foot orthosis is probablythe most useful in rheumatoid disease.

FOOTWEARThe requirements of rheumatoid patients arefor footwear which is comfortable (no pressure,no rubbing), lightweight, durable, stylish, and,preferably, waterproof. This, of course, is analmost impossible collection of demands.

Patients' own footwear, as well as stockfootwear, can have simple useful adaptationsmade to the soles and the heels that help theirgait immeasurably.

CHIROPODYAn understanding and skilled chiropodist is anabsolutely invaluable member of a caring team,just as is the orthotist and shoemaker. Regularattention to nails and callosities and orthodigitalsplinting all fall within his/her ambit.

WALKING AIDSThese must be included for completeness if forno other reason. Two feet may be insufficienton their own for safe ambulation and theassurance of a stick or other aid can meanthe difference between independence andimmobility through fear.

Treatment principlesBefore discussing the methods of achievingtreatment aims it is worth enumerating somebasic principles:

1 Synovectomy of tendon sheaths should beperformed at an early stage of disease as soon asthere is evidence of a hypertrophic synovitis.2 The weightbearing heel should be keptunder the load line of the leg.3 Mobile joints should be kept mobilewherever possible, unless their mobility isoutside the normal range, allowing collapse ofthe foot architecture.4 During gait the body load line should passforward through the foot without producingabnormal rotational forces in the foot.5 Foot movement should maintain the threerockers and if not this should be compensatedfor by adaptation of footwear.6 Loss offorefoot to hindfoot alignment shouldbe compensated orthotically.7 Forefoot surgery should consider theproblem of pain rather than function. By thetime surgery is indicated it is destructive ratherthan constructive.

TreatmentRheumatoid disease in the foot tends to presentin one of three ways: talonavicular arthritis,peroneal or posterior tibial tenosynovitis, or asmetatarsalgia. Although forefoot and hindfoot

disease can coexist, commonly the patient isinflicted with one or the other.

HINDFOOTThe role of the posterior tibial tendon cannot beemphasised too strongly as it is the key to themost difficult to manage of all the rheumatoidfoot problems-the collapsed pronated foot.Early synovectomy of the tendon sheath notonly relieves discomfort but will possibly delayor prevent attenuation and rupture. Steroidinfiltrations always run the risk ofintratendinousinjection, which may hasten the day of rupture.When the diagnosis of ruptured posterior

tibial tendon has been made, in the presence ofa correctable hyperpronation, transfer of theflexor digitorum longus tendon to the distalposterior tibial tendon stump has been found togive acceptable results.7 12Even with an affected posterior tibial tendon

the midfoot may start to collapse owing todisease in these joints. Talonavicular jointdisease is a not uncommon presenting form inthe foot. Our experience with these feet, tend-ing to collapse but still correctable, is to fuse thetalonavicular joint to provide the foot with astable, medial beam and prevent the calcaneumtaking up a fixed valgus position. This needs tobe carried out at the first signs of midfoothyperpronation, and we routinely use cancellousiliac bone graft to ensure sound union.

Hyperpronation, by virtue of its mechanics,inevitably results in the hallux being forced intovalgus even in the absence of rheumatoiddisease in the first metatarsophalangeal joint. Inthis situation fusion of the first metatarsophal-angeal joint is the preferred option. This doesnot apply when there is severe forefoot diseasepresent.Where the foot has collapsed into a pronated

valgoid position, the requirements are multiple.The heel is no longer under the leg, the midfootis twisted, often to the extent that the talar headstarts to bear weight medially, and the gaitbecomes flat footed with the loss of the anteriorrocker. A simple medial translational calcanealosteotomy can help put the loadbearing area ofthe heel back under the loadbearing line of theleg (fig 9) and relieve much pain due to turningmoments generated in the ankle/subtaloid com-plex on weight bearing.

Figure 9 Translational cakaneal osteotomy.

855

on 18 May 2018 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.49.Suppl_2.851 on 1 O

ctober 1990. Dow

nloaded from

Anderson

Total medial _Support I E

Medial buttress

_ Lateral 'choc' toprevent heel slippinglaterally

Figure II Proximal partialphalangectomy (afterStainsby).

Figure 10 Orthotic support ofthe collapsed (valgus) foot.

The collapsed midfoot, however, is notamenable to simple surgery. Its accommodationwithin bespoke footwear is required but the'overhanging' midfoot must be supported by anappropriate orthotic insole (fig 10). This in-evitably leads to a bulky shoe, and all the skillsof the shoemaker and orthotist are required tokeep the weight down to manageable proportionsfor the patient.

Subtalar fusion is rarely indicated on theseprinciples but if carried out the need to ensureaccurate alignment of hindfoot and forefoot isessential.

FOREFOOTWith the forefoot intermetatarsal head pain is acommon symptom. It has been suggested thatrheumatoid disease of the intermetatarsal bursaeis sufficient to cause either stretching or com-pression of the interdigital nerves. The symp-toms have been relieved by infiltration of thebursae with steroid. We now recognise, how-ever, that forefoot malalignment, in particularthe inverted forefoot, can present in non-rheumatoid patients with similar symptoms,and there is no reason to expect that therheumatoid foot is any different. Carefulexamination of the foot is, as always, essentialand any malalignment with symptoms whichdoes not have any other evidence of inter-metatarsal bursitis or fails to respond to steroidsshould be compensated orthotically. Orthoticinsoles for these patients need to be made withmore pliable, shock-absorbing materials other-wise they will not be tolerated.

It is most uncommon for the full blownpicture of the Morton's neuroma to appear, butif it does, resection of the traumatised segmentof nerve may be necessary.The occasional retracted toe can be best dealt

with by excising the proximal 4/Sths of theproximal phalanx, by dividing the long extensortendon at the level of the metatarsophalangealjoint, and by suturing the distal cut end to thelong flexor tendon in the gap. The cosmeticeffect is excellent, though functionally the toeremains equally useless (fig 11) (Stainsby,unpublished data).

When dealing with mallet toes it is probablyworth remembering to tenotomise the flexortendon even if distal interphalangeal joint fusionis being attempted.

Treatment of the destroyed metatarsophal-angeal articulations with gross retraction of thetoes has been a matter of debate. Stevens'review of forefoot excision arthroplasty versusfootwear suggested that there was no differencein the comfort of the two groups after fiveyears.13 This may be so, but the immediaterelief to those patients after surgery is verygratifying, both for the patient and the surgeon,and we still consider it a worthwhile procedure. 4Our own preference is for the operation

described by Kates, Kessel, and Kay in 1967. "1They advocated the excision of all metatarsalheads through an elliptical plantar incision,excising the skin overlying the metatarsal heads.The first metatarsal head can be excisedthrough a medial longitudinal excision. Muchhas been made of the need to have the metatarsalstumps load bearing equally, and rightly so. Anartistic curve on a non-weightbearing radio-graph, taken on the theatre table does notnecessarily equate with functional comfort. Ourpractice is at operation to push up against themetatarsal tread with the back of the hand,when any undue prominence can be easily feltand readily dealt with; perhaps still not veryscientific, but a little more practical. What isnot acceptable is excision of the second to fifthmetatarsal heads combined with the Keller'sprocedure. The first metatarsal head is no moreable to cope with the extra loading than anyother prominent metatarsal head.Treatment of the ankle is discussed by

Kirkup (pp 837-44, this issue), but it is worthnoting that the stiffened ankle, whether bydisease, surgical fusion, or disappointingarthroplasty, reduces or eliminates the secondrocker ofthe gait cycle. This can be compensatedby a technique used by the prosthetists in theirartificial SACH foot-solid ankle cushion heel.The incorporation of a spongy wedge into theheel of the shoe aids the forward propulsion ofthe body on heel strike and contributes signifi-cantly to an improved gait in these patients (fig12).15S

Over the years the advised indications for theHelal oblique metatarsal osteotomies havechanged, but many have been disappointed

856

on 18 May 2018 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.49.Suppl_2.851 on 1 O

ctober 1990. Dow

nloaded from

The rheumatidfoot

Figure 12 Cushion wedge heel.

with the results, particularly when used as

originally described for midfoot metatarsalgia,and we have all but abandoned it. The biggest

W problems have been with high pressure transferto the adjacent metatarsal heads after theoperation, and to non-union of the osteotomies,although it should be said, the latter were rarelytroublesome to the patient.At an earlier stage, where the hallux is drifting

into valgus, the foot pronating, but where thelesser metatarsophalangeal joints are not sub-luxed, fusion of the first metatarsophalangealjoint may be indicated. 6 The loss of firstmetatarsophalangeal movement may be com-

pensated with a rocker sole modification to thepatient's footwear. The extra weight of this maybe as disadvantageous as the loss of the forefootrocker to the patient, but there is no certain wayof predicting this.

Finally, the rheumatoid foot may also be a

dysvascular foot, and this may limit the amountof surgery it is possible to carry out. It alsoresults in atrophy of the fibro-fatty pads underthe heel and the metatarsal heads that permitthe first and third rockers to function painlessly.Under the forefoot the patient complains of'walking on marbles' (compare the purelyrheumatoid diseased forefoot described as

'walking on broken glass'). This tissue loss can

only be partially compensated by the provisionGPrT, LangerPBikmechanical of an insole of one of the new shear-absorbingNew York. materials (PPT*, Spencot). A word of cautiontSpenco; Spenco Medical about using too thick an insole of such materialsCorp, 6301 Imperial Drive, aboWaco, Texas. as, although it may be more comfortable to

stand on, it may create a degree of instabilitythat impairs the patient's proprioceptive feed-back mechanism-they do not know wheretheir feet are.

Dysvascular toes need protection, and the useof orthodigital splints to prevent chaffing, evenin bespoke footwear, should always be con-sidered. Regular foot care is essential and theplace of the chiropodist in the treatment of thistype of foot cannot be overe:nphasised.

EpilogueThe possible ways of management are endlessand some commoner problems have been usedto illustrate the particular principles of treat-ment and their use in fulfilling the general aimsof patient care. The individual surgeon ordoctor may find his way to different solutions; itmatters not as long as these aims and principlesare adhered to, for it is far easier in foot surgeryto render a foot worse off than to cure it.

1 Jones F W. Structure and function as seen in the foot. London:Bailliere Tindall, 1944.

2 Hicks J H. The mechanics of the foot. 1. The joints. 7 Anat1953; 87: 345-57.

3 Basmaiian J V, Stecko G. The role of muscles in the archsupport of the foot. J Bone joint Surg [Am] 1963; 45:1184-90.

4 Inman V T, Ralston H J, Todd F. Human walking.Baltimore: Williams and Wilkins, 1989.

5 Hughes J, Gerber C, Jagoe R, Clark P, Klenerman L. Theimportance of toes in walking. Ann R Coll Surg Engl 1988;70: 177.

6 Anderson E G, Harrison D. Foot pressure patterns in normalfeet. Ann R Coll Surg Engl 1988; 70: 176-7.

7 Johnson K A. Tibialis posterior tendon rupture. Clin Orthop1983; 177: 140-7.

8 Franks C I, Betts R P, Duckworth T. A microprocessorbased image processing system for dynamic foot pressurestudies. Joural of Medical and Biological Engineering andComputing 1983; 21: 566-72.

9 Hoffman P. An operation for severe grades of contracted orclawed toes. American Journal ofOrthopedic Surgery 1912; 9:441.

10 Fowler A W. A method of forefoot reconstruction. J Bonejoint Surg [Br] 1959; 41: 507-13.

11 Kates A, Kessel L, Kay A. Arthroplasty of the forefoot.J Bone joint Surg [Br] 1967; 49: 552-7.

12 Jahss M H. Spontaneous rupture of tibialis posterior tendon:clinical findings, tomographic studies, and a new techniqueof repair. Foot Ankle 1982; 3: 158-66.

13 Craxford A D, Stevens J, Park C. Management of thedeformed rheumatoid forefoot. Clin Orthop 1982; 166:121-6.

14 Betts R P, Stockley, I, Getty C J M, Rowley D I, DuckworthT, Franks C I. Foot pressure studies in the assessment offorefoot arthroplasty in the rheumatoid foot. Foot Ankle1988; 8: 315-26.

15 Gordon E J, Ardizzone J. Clinical experiences with the SACHfoot prosthesis. J Bone joint Surg [Am] 1960; 42: 226-34.

16 Newman R J, Fitton J M. Conservation of metatarsal heads insurgery of rheumatoid arthritis. Acta Orthop Scand 1983;54: 417-22.

857

on 18 May 2018 by guest. P

rotected by copyright.http://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.49.Suppl_2.851 on 1 O

ctober 1990. Dow

nloaded from