Pantalar and Tibiotalocalcaneal Arthrodesis for ... - CiteSeerX

(opvright I 992 h� i�/ti� Journal of Biiii ig,ii! Joint Ss�rgerv. Iili-i’rtiiir�I1i.il

1042 THE JOURNAL OF BONE AND JOINT SURGERY

Pantalar and Tibiotalocalcaneal Arthrodesis

for Post-Traumatic Osteoarthrosis of the Ankle and Hindfoot*BY JOHN A. PAPA. M.D.t. AND MARK S. MYERSON. MDI. BALTIMORE. MARYLAND

In vc’stigatioll J)t’rforlne(l at Union Me,norial Hospital, Tile Johns Hopkins Universit%� Baltintore

ABSTRACT: Twenty-one patients had a unilateral

extended arthrodesis of the ankle and hindfoot (a tib-

iotalocalcaneal procedure in thirteen patients and a

pantalar procedure in eight) for post-traumatic os-

teoarthrosis or deformity, or both. The operation was

performed through a transfibular extended lateral ap-

proach, and autogenous bone graft and rigid internal

fixation was used. A final alignment of 0 to S degrees of

valgus, 0 to S degrees ofcalcaneus, and external rotation

equal to that of the contralateral side was sought.

Subjective and objective evaluation, including a per-

sonal interview, physical examination, and radiographic

and dynamic pedobarographic analysis, was performed

at a mean interval of thirty-two months (range, twenty-

four to fifty-four months) after the operation. A solid

fusion was achieved in eighteen (86 per cent) of the

twenty-one patients. There were five malunions (24 per

cent) and two superficial wound problems (10 per cent).

Ofthe seventeen patients who were not retired from

work, eleven returned to work: nine to an occupation

that involved standing and two to a sedentary occupa-

lion. Although seventeen (81 per cent) of the twenty-

one patients reported that they were much improved,

twenty (95 per cent) had some pain, and most benefited

from modifications in shoe-wear. Patients who had had

a tibiotalocalcaneal arthrodesis were more mobile and

functioned at a higher level than those who had had apantalar arthrodesis.

Extended arthrodesis of the ankle and hindfoot is a

complex, technically demanding procedure, and should

be regarded as a salvage operation capable of producing

a satisfactory result and usually providing a reasonable

alternative to amputation.

Patients who have post-traumatic osteoanthnosis in-

volving both the ankle and the subtalar joints pose a

difficult therapeutic challenge. In addition to debilitat-

ing pain. these patients frequently have diffuse stiffness

of the forefoot, midfoot. hindfoot, and ankle second-

*No benefits in any form have been received or will be received

from a commercial party related directly or indirectly to the subject ofthis article. No funds were received in support of this study.

tMatthews Orthopaedic Clinic. 1315 South Orange Avenue, Or-

lando, Florida 32856.

IFoot and Ankle Center, Union Memorial Hospital. 201 East Uni-versity Parkway. Baltimore, Maryland 21218.

any to associated injuries, prolonged immobilization, and

previous operations. Operative tneatment is limited to

extended arthnodesis or to Syme or below-the-knee am-

putation. The anticipated outcome after amputation is

well known27’”’3’7’525. After a post-traumatic Syme on

below-the-knee amputation, patients can be expected to

walk rather well, with only a minimum increase in expen-

diture of energy compared with normal individuals”326.

In addition, the pain that is often perceived as coming

from the amputated part is, in most patients, eliminated.

However, patients who have had a Syme on below-the-

knee amputation need a prosthesis for walking, with the

exception of limited distances for patients who have had

a Syme amputation.

The expected functional results in patients who have

had an extended arthrodesis are not well established.

Most of the previous reports of pantalar arthrodesis

have dealt primarily with treatment of deformities of the

paralytic lower extnemity’3469’2”�24�27. Debate continues

concerning the merits of a one-stage procedure’3’2’624

compared with a two-stage procedure”2’22, different op-

enative approaches (anterolateral”, lateral2’25, or poste-

non2524), type of fixation (external23 or internal6’4), and

need for bone-grafting. We performed this retrospective

study of our experience with extended arthrodesis of the

ankle and hindfoot to determine the associated morbid-

ity and the expected functional nesults and limitations.

Materials and Methods

Between 1986 and 1989, twenty-one patients (four-

teen men and seven women) had a unilateral extended

anthrodesis of the ankle and hindfoot (a tibiotalocal-

caneal procedure in thirteen patients and a pantalan

procedure in eight) performed by the senior one of us

(M. S. M.). The term pantalar is used if the tibiotalan,

talocalcaneal, talonavicular, and calcaneocuboid joints

are included in the arthrodesis. The average age of the

patients at the time of the operation was forty-five years

(range, twenty-one to sixty-eight years). The operation

was done on the right foot in ten patients and on the left

in eleven.

All of the operations were done as salvage proce-

dures for the treatment of painful, disabling osteoar-

throsis on deformity. or both, involving both the ankle

and subtalar joints. An operation was done only after

failure of non-operative management. which generally

F. 49 Closed fracts., Secondary avascularankle and necrosis, talus

DATA ON THE TWENTY.ONE PATIENTS WHo HAD AN EXTENDED ARmR0DEsI5 OF THE HINDFOOT

Initial Treatment andAssociated Injuries Other Previous Indication for Type of Duration of

Case Sex.Age Initial Injuries and Conditions Procedures Extended Arthrodesis Arthrodesis Follow-up Complications(Yr.s.) (Mos.)

Prolonged immobil. Osteoarth., ankle and tibiotalo-in cast subtalar joints calcaneal

External fix.: latissi- Malunion after Pantalar

mus dorsi free arthrodesis of ankleflap: split-thickness (rigid equinovarusskin graft; arthro- deform.)

desis, ankleExternal fix.: rectus Malunion after Pantalar

abdominis free arthrodesis of ankleflap (failed): split- (rigid equinovarusthickness skin graft: deform.)

arthrodesis. ankle4 M. 67 Chronic in- Arthrodesis, ankle

stahil., ankle

and subtalarjoints

5 F. 21 Open fracts.. Open fracts., ipsilat. Internal fix.: external Osteoarth. and in- Tihiotalo-

ankle and 4th and 5th fix.: latissimus stabil.. ankle and calcanealcalcaneus: metatarsals: soft- dorsi free flap: split- subtalar jointscrush injury. tissue defect, foot thickness skin graft:foot and and ankle resect.. 4th and 5th

ankle raysft M. 36 Open soft- Open fracts.. ipsilat. External fix. Osteoarth., peritalar Pantalar

tissue injury, tibia and femur joints: equinovarus

ankle and deform.foot

7 M. 54 Open fract.. Open fract., ipsilat. External and internal Osteoarth., ankle and Tihiotalo-ankle; crush tibia fix. subtalar joints: calcanealinjury. ankle varus deform..and foot hindfoot

8 M. 55 Closed fract.- Fract., contralat. Internal fix.: arthro- Osteoarth., subtalar Tihiotalo-disloc.. ankle ankle desis, ankle joints calcaneal

9 F. 6() Closed fract.- Secondary avascular Prolonged immobil. Osteoarth., ankle and Tibiotalo-

disloc.. talus necrosis, talus in cast subtalar joints calcaneal10 M. 36 Open fract.- Fracts.. contralat. tibia, Internal fix. Osteoarth.. ankle and Tihiotalo-

disloc., ankle ankle. and calcaneus subtalar joints calcaneal

I I M. 45 Open fract.. Compart. syndrome. External fix.; latissi- Osteoarth., peritalar Pantalarankle: crush ipsilat. leg and foot: mus dorsi free jointsinjury. ankle soft-tissue defect, flap: split-thicknessand foot ankle and foot skin graft: arthro-

desis, ankle12 M. 64 Closed fract.- Internal fix.; arthro-

disloc.. ankle desis. ankle

35 Varus malunion

31 Talonavicular non-

union (successfullyrevised); mar-ginal woundslough

30

34

25

25 Calcaneal malunion

24

29

25 Equinus malunion

41 Equinus malunion

Tihiotalo-

calcaneal

54

‘U) Tibiotalar non.

union (stable andasymptomatic):

equinus malunion28

31

24

24

42 Talonavicular non-union (stable andasymptomatic):sural neuroma

2620 M. 35 Closed fract.- Internal fix.: arthro-disloc.. ankle desis, ankle

Internal fix.

Internal fix. 46

PANTALAR AND TIBIOTALOCALCANEAL ARTHRODESIS FOR POST-TRAUMATIC OSTEOARTHROSIS

TABLE I

1043

talus2 M. 25 Open fract..

ankle

3 M. 24 Open crushinjury. ankleand foot

13 F. 64 Closed pilonfract.

14 F. 42 Open fract.-disloc.. talus

15 F. 6() Open soft-

tissue injury.ankle and

foot16 M. 50 Closed pilon

fract.: soft-tissue injury,

foot

17 M. 35 Closed fract.-disloc., ankle:soft-tissueinjury. foot

18 M.2I Openpilon

fract.

19 M. 30 Open pilonfract.

21 F. 68 Open fract.-disloc.. ankle

Deep post.-compart.syndrome. ipsilat.leg; soft-tissue

defect, foot

Soft-tissue defect.

ankle and foot

External fix.: arthro-desis, ankle

Fract.-disloc.. ipsilat. External fix.Lisfranc joint

Prolonged immobil.in cast: brace

Arthrodesis, ankle

Arthrodesis, ankle

Fracts.. contralat. tibia Internal fix.: rectusand femur: rupture. abdominis freecontralat. Achilles flap: external fix.tendon: compart.syndrome. soft-tissue defect. ipsilat.leg

Fracts., ipsilat. External fix.: arthro-humerus and femur desis, ankle

Osteoarth., subtalar Tihiotalo-joints calcaneal

Non-union after Tihiotalo-arthrodesis of ankle: calcanealosteoarth., subtalarjoints

Osteoarth.. subtalar Tibiotalo-joints calcaneal

Osteoarth.. peritalar Pantalarjoints

Instabil. and osteo-

arth., ankle andsubtalar joints

Osteoarth.. subtalar Pantalarjoints: rigid varusdeform., forefoot

Malunion (equinus) Pantalar

after arthrodesis ofankle; osteoarih..pentalar joints

Rigid equinus varus tibiotalo-deform., ankle and calcanealhindfoot; osteoarth.,subtalar joints

Malunion (equino- Pantalarvarus) after arthro-desis of ankle;osteoarth., subtalarjoints

Non-union afterar- Tihiotalo-throdesis of ankle: calcanealosteoarth., subtalarjoints

Osteoarth.. ankle and Tibiotalo-subtalar joints calcaneal

34

27 Superficial woundinfect.

I

FI;. I-A Fui. 1-B

Figs. I-A through l-D: Case 7. This patient had had a previous O�Cfl trimalleolar fracture of the ankle. which had been treated operatively,

and an open fracture of the ipsilateral distal tihial diaphysis. Clinical examination and diagnostic injections confirmed involvement of both theankle and subtalar joints.

Figs. I-A and 1-B: Preoperative anteroposterior and lateral radiographs showing post-traumatic osteoarthrosis involving both the ankle and

subtalar joints. Non-operative treatment had produced no response.

1044 J. A. PAPA AND M. 5. MYERSON

TIlE JOURNAL OF BONE AND JOINT SURGERY

included a polypropylene ankle-foot orthosis and non-

steroidal anti-inflammatory medication. A below-the-

knee weight-bearing cast was used for five patients to

help determine the value of the proposed arthrodesis.

When possible, an effort was made to avoid inclusion of

the transverse tarsal joint (the talonavicular and calca-

neocuhoidjoints) in the arthrodesis. However. if osteoar-

throsis or deformity was present at that level, a pantalar

arthrodesis was performed.

Physical examination and evaluation with plain ra-

diognaphs were used routinely as the primary determi-

nants in the selection of joints that needed anthrodesis.

Diagnostic sequential blocks with a local anesthetic

( 1 per cent plain lidocaine) were employed in seventeen

of the patients to localize the underlying source of the

pain specifically. Computerized tomography or hone-

scanning with technetium. or both. were performed in six

patients when additional clarification was needed.

All patients had a history of major trauma: ten pa-

tients had had a fracture on fracture-dislocation of the

ankle (five open and five closed): four. a pilon fracture

(two open and two closed): three. a severe soft-tissue

injury ofthe ankle and hindfoot;one,a unilateral fracture

of the ankle and calcaneus: two. a fracture-dislocation of

the talus: and one, recurrent traumatic instability of the

ankle and subtalar joints. Multiple associated injuries

were present in thirteen patients (62 per cent) and in-

cluded a variety of fractures and fracture-dislocations

(Table I). Five patients had also sustained major soft-

tissue defects of the lower extremity. for which free

microvascular tissue transfer was performed. Three pa-

tients had sequelae that we believe were secondary to a

deep posterior-compartment syndrome of the leg, mani-

fested by an equinovarus contracture and weak plantar

flexion. One of these patients had probably sustained a

synchronous compartment syndrome of the foot also,

evidenced by intrinsic atrophy and claw-toe deformities.

Two patients had had development of post-traumatic

avascular necrosis of the talus by the time the extended

anthrodesis was performed.

All patients had good protective plantan sensation.

The absence of adequate protective sensation was con-

sidered a contraindication to the creation of a rigid foot

through the extended arthrodesis of the hindfoot. When

the procedure would have otherwise been indicated in a

patient who lacked sensation. consideration was given to

amputation.

Most patients had had various treatments before the

extended anthrodesis (Table I).Ten patients initially had

had stabilization with an external fixator. and an addi-

tional five had had internal fixation. Arthnodesis of the

ankle had been attempted in eleven patients, five of

whom had a solid union with acceptable alignment. In

these five patients, conversion to a tibiotalocalcaneal on

pantalar arthrodesis was obtained with the addition of

an arthrodesis of the hindfoot (subtalan or triple). Four

FIG. I-C Fl;. 1-D

Follow-up anteroposterior and lateral radiographs, showing good alignment after successful healing of a tihiotalocalcaneal arthrodesis.

Although not readily apparent. the shorter of the two large cancellous-bone screws is engaging the anterior tihial cortex. The other large

cancellous-bone screw has gained excellent purchase in the sclerotic hone at the site of the healed fracture of the tihial shaft.

PANTALAR AND TIBIOTALOCALCANEA1. ARTHRODESIS FOR POST-TRAUMATIC OSTEOARTHROSIS 1045

VOL. 74-A, NO. 7. AUGUST 1992

of the eleven patients had a fusion in poor alignment and

two had a non-union.

In addition to the arthrodesis of the hindfoot, a con-

rective osteotomy was needed at the level of the ankle

for the four patients who had a malunion of a fracture in

the region of the ankle. The two patients who had a

non-union after an attempted arthrodesis had a standard

one-stage anthrodesis of the ankle and hindfoot.

Twenty of the twenty-one patients had had prolonged

immobilization of the affected foot before being seen by

us. The average interval between the injury and the most

recent anthrodesis was six years (range, one to twenty-

four years). Due to the severity of the injuries, the fre-

quent presence of associated injuries, and the numerous

treatments that had been used previously, the foot was

generally atrophied and stiff. Six patients had been of-

fered amputation as a treatment option. but all had

refused.

The patients were evaluated with respect to level of

satisfaction with the procedure, pain. walking and stand-

ing endurance, capacity to climb hills and stairs. and

ability to return to work. The presence and location of

calluses, as well as the sagittal and coronal alignment of

the hindfoot, were documented. Clinical examination

included determination of the residual range of motion

in the sagittal and coronal planes of the midfoot and

forefoot. In addition, the range of motion was measured

at the first metatarsophalangeal and knee joints. Pal-

pation was routinely performed in search of areas of

tenderness. Limb-length inequality was determined by

leveling of the pelvis with blocks of known height.

Anteropostenior. lateral, and axial radiographs of the

hindfoot were made to allow assessment of alignment in

the coronal and sagittal planes and determination of the

presence of fusion. An arthnodesis was considered sue-

cessful when tnabeculations were noted to traverse each

operative site.

Dynamic pedobanognaphic data (Biokinetics, Be-

thesda, Maryland) were obtained for each patient, when

the patient was barefoot, and an attempt was made to

correlate these data with the clinical and radiographic

findings.

Operative Technique

The procedure is performed with the patient in the

supine position and a bolster placed under the ipsilateral

buttock. A pneumatic tourniquet is placed on the proxi-

mal part of the thigh. and the ipsilateral iliac crest and

extremity, from toes to groin, are prepared and draped.

The limb is draped free at a level proximal to the knee

to allow adequate visualization for the accurate intnaop-

erative determination of alignment.

An extended lateral approach is employed through

an incision measuring approximately eighteen centime-

tens. Proximally.the incision is made in line with the distal

part of the fibula, and distally. it is curved over the sinus

tansi.This incision is modified ifeithen a previous incision

or a free flap is present. If there has been a previous

lateral incision, it is used and extended proximally or

distally as necessary. The incision is not made across a

free flap. Otherwise, previous incisions are not used and

an attempt is made to avoid crossing them, if possible.

The osteotomy is made in an oblique fashion in the

distal pant of the fibula, approximately seven centimeters

1046 J. A. PAPA AND M. S. MYERSON

TIlE JOURNAL OF BONE AND JOINT SURGERY

proximal to its tip. with an oscillating saw. That portion

of the fibula is then removed. A second incision is made

anteromedially. directly over the medial malleolus for a

tibiotalocalcaneal arthnodesis and slightly anterior to the

medial malleolus, with extension distally over the dorsal

aspect of the talonavicular joint, for a pantalan arthno-

desis. The medial malleolus is removed after osteotomy

is performed obliquely with a chisel. The tibiotalan antic-

ular surfaces are removed with an oscillating saw held

perpendicular to the long axis of the tibia, with cane being

taken to preserve as much bone stock as possible. Sharp

chisels are used to remove all articulan cartilage from the

subtalar joints and, when appropriate, from the tab-

naviculan and calcaneocuboid joints.

Because most patients who have this type of salvage

procedure have underlying deformities of the foot, an in

situ arthrodesis is rarely performed. Therefore, wedges

are removed, as necessary, to allow connection of deform-

ities in an effort to obtain a plantigrade foot. An equino-

varus deformity is corrected with a dorsolatenal biplane

wedge resection. preferably from the apex of the defon-

mity. which is typically at the level of the subtalar joint.

Final alignment of the hindfoot in 0 to 5 degrees of valgus,

0 to 5 degrees of calcaneus, external rotation equal to that

of the contralateral side, and one-half to one centimeter

of posterior translation of the talus under the tibia is

sought. The effects of wedge resection of the hindfoot on

the positions of the midfoot and forefoot are closely

monitored. In the absence of a supple midfoot and fore-

foot, correction of deformities of the hindfoot frequently

necessitates rotational correction and arthnodesis at the

level of the transverse tarsal joint.

Initial fixation is obtained with 2.0-millimeter guide-

pins. and the position is checked with intraoperative

radiographs to verify the alignment and placement of the

wires. An autogenous graft composed ofcancellous bone

always is packed into available spaces before final in-

ternal fixation is achieved with 7.0-millimeter, partially

threaded, cannulated cancelbous-bone screws (Synthes

USA. Monument, Colorado). Adequate cancellous bone

is usually obtained from the removed malleoli. Occasion-

ally. the malleoli do not provide sufficient cancelbous

bone for the graft. and in such instances, additional can-

cellous bone is obtained from the anterior iliac crest.

Two screws are inserted through separate stab mci-

sions in the skin and directed from posterior and distal

to anterior and proximal, passing through the calcaneus

and the body of the talus and into the distal part of the

tibia. An attempt is made to gain purchase, with at least

one screw, on the anterior tibial cortex. An additional

screw is occasionally placed from the distal pant of the

tibia into the calcaneus (Figs. 1-A through 1-D). Excel-

lent compression usually is obtained across the operative

sites in both the ankle and subtalar joints. A pantalar

arthrodesis also includes fixation of the operative sites in

the talonavicubar and calcaneocuboid joints with screws

or power staples.

The wounds are then closed in layers oven suction

drains, and a bulky dressing, reinforced with posterior

and u-shaped plaster splints, is applied. The drains are

removed at twenty-four to forty-eight hours. The patient

is usually discharged from the hospital by the third or

fourth postoperative day.

At ten to fourteen days postoperatively, the beg is

placed in a non-weight-bearing below-the-knee fiber-

glass cast. Six weeks postoperatively. weight-beaning is

begun in a below-the-knee weight-beaning cast. Changes

of the cast, clinical evaluation, and radiographic exami-

nation are performed at approximately three to four-

week intervals until union. The patient is then allowed to

walk as tolerated, wearing a shoe with a cushioned heel

and rocker sole. A shoe-lift is used to compensate for

leg-length inequality. The lift is selectedso that the side

that was operated on will remain approximately 0.5 cen-

timeter shorter than the contralateral side.

Results

The twenty-one patients were followed for an aver-

age of thirty-two months (range, twenty-four to fifty-four

months) after the operation. Seventeen patients (81 pen

cent) reported that they were much improved after the

operation. Twenty patients (95 pen cent) noted some pain

with walking; this varied in severity and threshold. Re-

sidual pain was generally described as mild, intermittent,

and not incapacitating. Of the twenty patients who had

pain, six (30 per cent) did not use any medication for it.

Eleven patients (55 pen cent) took non-narcotic analge-

sic medication occasionally and rarely or never had pain

at rest. Two patients (10 per cent) reported daily use of

a short-acting non-steroidal anti-inflammatory medica-

tion. One patient used a narcotic analgesic persistently,

although intermittently.

The patients’ standing endurance at the most recent

follow-up averaged three hours (range, one-half hour to

eight hours), and the mean walking endurance was nine-

teen blocks (range. three to forty-five blocks), compared

with the preoperative values of fifteen minutes (maxi-

mum, one hour) and one and one-half blocks (range,

one-half to three blocks). All patients reported some

difficulty on uneven ground or when ascending or de-

scending a hill. Most patients compensated for the lost

motion with external rotation of the limb to negotiate

inclines. Two patients reported going up and down hills

in a sideways fashion.

No patient used a cane, crutches, or a walker because

of the foot. One patient, who was seventy-two years old

at the latest follow-up. had unrelated low-back pain and

stated that she often used a cane to minimize the risk of

falling. All but two patients had some difficulty with

climbing and descending stairs; eighteen used a banister

if one was available, and ten negotiated one step at a time.

No patient could run effectively.

In general, the patients were found to benefit subjec-

tively from modifications in shoe-wean. Optimum results

PANTALAR AND TIBIOTALOCALCANEAL ARTHRODESIS FOR POST-TRAtJMATIC OSTEOARTHROSIS 1047

VOL. 74-A, NO. 7. AUGUST 1992

were noted with a combination of a cushioned heel, a

tapered rocker-sole, and an appropriate heel-and-sole

lift. In addition, soft, accommodating onthotic inserts

were helpful for several patients who had well localized

problems of the forefoot.

At the latest follow-up examination, all patients had

an excellent range of motion of the ipsilateral knee

(average flexion, 132 degrees and average extension.

0 degree) without recurvatum, except for one patient

who had sustained severe intra-articular trauma to the

ipsilateral knee at the time of the original injury. This

knee lacked 13 degrees of full extension and flexed

to only 70 degrees. Motion of the metatarsophalangeal

joint of the ipsilateral great toe was frequently noted

to be limited, with mean dorsiflexion of only 34 degrees

(range, 20 to 76 degrees). The remaining motion of the

mid-tarsal joints (joints between the metatarsophalan-

geal and transverse tarsal joints) and of the transverse

tarsal joint (if fusion had not occurred) of the involved

foot averaged 1 1 degrees in the sagittal plane (range, 5 to

25 degrees) and 5 degrees in the coronal plane (range,

0 to 12 degrees).

No clinically notable rotational malunions were seen.

Each involved foot was found to be in external rotation,

within 5 degrees of the angle on the contralatenal side.

Alignment in the sagittal plane was best determined

nadiographically; the mean alignment, as measured by

the angle subtended by the long axis of the tibia and

the plantar plane of support of the foot during weight-

bearing, was 1 degree of calcaneus (range, 6 degrees of

equinus to 8 degrees of calcaneus). Alignment in the

coronal plane was measured clinically; the mean align-

ment was 1 degree of valgus (range, 5 degrees of vanus to

5 degrees of valgus).

The mean amount of shortening of the ipsilateral leg

was 1.5 centimeters (range, 1.0 to 2.5 centimeters).

A solid union was noted clinically and radiognaphi-

cally at a mean of fourteen weeks (range, eight to twenty-

four weeks) postoperatively in eighteen (86 per cent) of

the twenty-one patients. There were three non-unions,

two involving the tabonaviculan articulation and the

other, the tibiotalar joint. Failure occurred, therefore,

in only S per cent of the fifty-eight fusion sites. Neither

of the two patients who had avascular necrosis of the

talus had a non-union. One of the patients who had a

tabonaviculan non-union was symptomatic, and a success-

ful revision operation was performed with an autoge-

nous bone graft from the iliac crest. This patient had

sustained a crush injury to the foot, with loss of skin, and

the dorsomedial portion of the midfoot was relatively

avascular. The remaining two patients who had a non-

union were clinically stable and asymptomatic with ne-

spect to those sites.

Ten mild. asymptomatic plantar calluses were noted

in seven of the patients. The calluses were under the

interphalangeal joint of the great toe in three patients;

under the first, second, and third metatarsal heads in

two patients each: and under the second metatarsal head

in one.

Return to work was achieved by eleven of the seven-

teen patients who had not yet retired. Of these seventeen

patients. five were involved in Workers’ Compensation

cases and one, in ongoing litigation; only one of those six

had returned to work. Of the eleven patients who were

not involved in litigation or a Workers’ Compensation

case, only one did not return to work. That patient had

remained out of work due to an ongoing problem with

the ipsilateral knee,which was treated operatively twelve

months after the pantalar anthrodesis. He was receiving

therapy for the knee at the latest follow-up and was

expected to return to work within three months.

The pedobarographic investigation confirmed clini-

cal impressions derived from the evaluation of alignment

of the hindfoot, the flexibility of the midfoot. and the

range of motion of the metatarsophalangeal joint of the

great toe.

One patient, who was noted, clinically and radio-

graphically. to have 5 degrees of varus of the hindfoot

with a fused but incompletely pronated transtarsal joint,

was found to have increased pressure of prolonged du-

ration under the fifth metatarsal head. Five patients who

had a markedly restricted range of motion (30 degrees

or less) of the metatarsophalangeal joint of the great toe

had abnormally high pressures in the region of the distal

pant of the great toe. This finding was accentuated in

the subset of patients who also had a limited range of

motion ofthe mid-tarsaljoint in the sagittal plane. All ten

sites that had plantan calluses had corresponding evi-

dence of prolonged increases in local pressure. recorded

pedobanognaphically.

Complications

There was one superficial wound infection hut no

deep wound infections (Table I). One patient had a

minor wound slough, which healed after d#{233}hnidement in

the office. Non-union occurred in three patients, at one

site each (two. at the talonavicular joint and one. at the

ankle). Three patients had a malunion in 2 to 4 degrees

ofequinus:one patient,in 8 degrees ofcalcaneus:and one

patient, in 5 degrees of varus. In one patient. a sural

neuroma developed; this was excised, and the stump of

the nerve was transposed into the peroneus brevis mus-

dc. Calluses were frequently associated with malunion,

but no patient had a plantar ulcer. We attribute the lack

of ulceration largely to the fact that all patients had

normal plantan sensation.

Discussion

Staging of pantalar arthrodesis has been advocated

by numerous authors who have believed that a one-stage

procedure is associated with an increased prevalence

of avascular necrosis, malunion, non-union, and wound

problems’52’22. Avascular necrosis has been reported in

close association with arthnodesis of the hindfoof’�2.

1048 j. A. PAPA AND M. S. MYERSON

THE JOURNAL OF BONE AND JOINT SURGERY

This complication. however. was not found in our pa-

tients. and probably it is relevant clinically only when

techniques associated with the temporary removal of the

talus are employed, such as that originally described by

Lorthioir.

Non-union is a frequent complication of extended

arthrodesis of the hindfoot. with reported pnevalences of

0 to 28 per cent (mean. 15 per cent)�4’’5272t7.The ankle

has been the most frequently reported site of pseudar-

thnosis,folbowed by the talonavicularjoint. In the current

series. three patients (14 per cent) had a non-union: one.

at the ankle.and the other two,at the tabonaviculan joint.

If the total number of sites of attempted arthnodesis

is considered. failure of healing was noted in only three

(5 per cent) of fifty-eight sites. It has been our impression

that. when using lag-screw fixation. superior purchase

with increased compression is achieved by crossing both

the talocalcanealjoint and the tibiotalar joint. compared

with crossing either joint alone. This is particularly cvi-

dent when one or both of the two screws engages the

anterior tihial cortex.

Theonetically. the prevalence of wound problems,

including infection. would be expected to he higher with

the more extensive single-stage arthrodesis than with

the two-stage procedure. This might be anticipated if

there was marked deformity that needed correction. In

the current series. there was one superficial and no deep

wound infection. The superficial infection responded

rapidly to standard measures. including local care of the

wound and a short course of a broad-spectrum antibiotic.

One minor wound slough was noted at the postero-

inferior aspect of the incision: it resolved readily with

routine care of the wound. This I 0 per cent prevalence of

local wound problems compares favorably with the rates

reported previously for extensive on isolated arthrodesis

of the hindfoot’�’222�.

The technical demands of a single-stage pantalar or

tibiotalocalcaneal arthnodesis are great. Unlike an iso-

lated anthrodesis of the ankle or hindfoot. after which

motion of either the subtalar joints on the ankle is possi-

ble. an extended anthrodesis eliminates all local compen-

satory motion except minimum residual motion in the

midfoot and forefoot. Therefore, when performing these

extensive procedures. the need to obtain precise final

alignment in multiple planes is critical.

Despite meticulous attention to detail. there were

five malunions (four in the sagittal plane and one in the

coronal plane) in this series. Three patients had fusion in

mild equinus ranging from 2 to 4 degrees. Although.

subjectively. these three patients did not differ markedly

from the other eighteen patients at the latest follow-up.

none had yet returned to work. and two of the three had

greatly increased peak pressures ofthe forefoot on pedo-

barographic analysis. Because dynamic gait analysis was

not performed as part of this study. and no gross devia-

tion in observable gait was noted in these three patients.

no specific data regarding subtle alterations in gait can

be provided. We believe that an equinus position will

lead to increased loading of the forefoot and, possibly, to

metatarsalgia. In addition, equinus may be associated

with an awkward gait and the possible development of

genu recurvatum. Therefore. we think that no attempt

should be made to compensate for leg-length inequality

by placement of the foot in an equinus position: rather, a

shoe-lift should be employed for this purpose. One pa-

tient had fusion in 8 degrees of calcaneus but did not

complain of pain in the heel. On the basis of this series,

as much as 8 degrees of mild calcaneus is well tolerated.

without the development of pain in the heel.

Malunion in the coronal plane was observed in one

patient who had had a pantalar arthrodesis and in whom

the hindfoot was fused in 5 degrees of varus. Pronation

at the site of the transverse tarsal arthrodesis provided

partial compensation: the patient returned to demanding

work in a warehouse and was quite satisfied with the

result of the operation. He did, however. have formation

of a mild callus under the fifth metatarsal head, as well

as pedobarographic evidence of increased pressures in

the lateral part of the forefoot. Therefore. when a true

pantalar arthrodesis is performed, care must be taken to

position the midfoot and forefoot appropriately in the

coronal plane. in order to obtain a plantigrade foot. This

is particularly true for patients who have minimum mo-

tion of the forefoot and little ability to compensate for

the fusion of the hindfoot during toe-off.

No rotational malunions were observed in this series.

We believe that avoidance of internal rotation combined

with mild posterior subluxation of the talus on the tibia

will enhance gait by minimizing the length of the rela-

tively rigid appendage. which must be stepped oven.

When indicated, it appears that the transverse tarsal

joint should be spared arthrodesis. Patients who had had

a tihiotabocalcaneal arthnodesis maintained greater aver-

age motion in the sagittal and coronal planes than those

who had had a pantalar arthrodesis (13 and 6 degrees,

compared with 7 and 3 degrees). These differences were

significant when they were analyzed with an unpaired

Student t test (p < 0.008 and p < 0.025). The over-all

degree of motion at the latest follow-up, however, was

less than half of what would be expected in normal feet

that had had a similar procedure, according to labo-

ratory data reported by Gellman et a!. These findings

are. nonetheless, not surprising. considering that most

of our patients had sustained severe injuries and many

had had multiple operations and prolonged immobili-

zation. Even if it is minimum, residual range of motion

does provide some functional compensatory ability and,

therefore, should be preserved if it is not painful. No

severe degenerative changes were noted in the thirteen

patients who had had preservation of the transverse

tarsal joint (a tibiotalocalcaneal arthrodesis), at an aver-

age follow-up of thirty-two months (range. twenty-four

to fifty-four months).

All but one of the patients in this study reported

PANTALAR AND TIBIOTALOCALCANEAL ARTHRODESIS FOR POST.TRAt MATIC OSTEOARTII ROSIS 1049

VOL. 74-A, NO. 7. AtJGtJST 1992

some persistent pain in the affected foot at the most

recent follow-up. Residual pain was generally described

as mild, intermittent, and not incapacitating. Although

most patients were very satisfied with the result of the

operation and had marked improvement in the oven-all

level of function, the effects of the initial trauma, pre-

vious operations. and prolonged immobilization proba-

bly were responsible for the persistent symptoms.

In summary, although extended arthnodesis of the

ankle and hindfoot in patients who have post-traumatic

osteoarthrosis usually provides an effective alternative

to amputation. it should be regarded as a salvage proce-

dune. It is a complex. technically demanding operation in

which attainment of precise final alignment is critical.

The described single-stage technique. employing a trans-

fibular extended lateral approach with an autogenous

bone graft and rigid internal fixation. provided satis-

factory results. with most of the patients who had not

yet retired being able to return to work. The decision

between extended arthrodesis or amputation must he

guided. in part. by the wishes of the individual patient

and the philosophy of the treating physician. It is hoped

that the results reported here will help with this often-

difficult decision.

References

1. Ansart, M. B.: Pan-arthrodesis for paralytic flail foot.J. Bone andJoint Surg.. 33-B(4): 503-507. 1951.

2. Baker, G. C. W., and Stableforth, P. G.: Syme’s amputation. A review of sixty-seven cases. J. Bone andJouit Siirg.. 5l-B(3): 482-487. 1969.

3. Barrett, C. R.; Meyer, L. C.; Bray, E. W., Ill; Taylor, R. G.,Jr.; and Koib, F. J.: Pantalar arthrodesis: a long-term follow-up. hotu,ulflnkle,

1:279-283, 1981.

4. Bingold, A. C.: Ankle and subtalar fusion by a transarticular graft. J. Bone and Joint Siirg., 38-B(4): 862-870. 1956.

5. Blair, H. C.: Comminuted fractures and fracture dislocations of the body of the astragalus. Operative treatment. Am. J. Siirg.. 59: 37-

43. 1943.

6. Danan, J. P., and Tomeno, B.: Panarthrod#{232}se de l’arri#{232}re-pied. Technique. tolerance Ct r#{235}sultats lointains de Ia triple arthrod�sc. Rev. hir.

orthop., 65: 433-439, 1979.

7. Fleurant, F W., and Alexander,Justin: Below knee amputation and rehabilitation of amputees. Surg.. Gvnee. ullil ()hstet.. 151: 41-44. 1980.

8. Cellman, Harris; Lenihan, Michael; Halikis, Nick; Botte, M. J.; Giordani, Mauro; and Perry, Jacquelin: Selective tarsal arthrodesis: an in

vitro analysis of the effect on foot motion. Foot and Ankle. 8: 127-133. 1987.

9. Hamsa, W. R.: Panastragaloid arthrodesis. A study of end results in eighty-five cases. J. Bone (111(1 Joint .ciug.. 18: 732-736. July 1936.

10. Hornby, Roger,and Harris,W.R.:Syme’samputation.Follow-upstudyofweight-hearing in sixty-eight patients.J. Bone and Joint Surg..

57-A: 346-349, April 1975.

I I . Huang, C.-T.; Jackson, J. R.; Moore, N. B.; Fine, P. R.; Kuhlemeier, K. V.; Traugh, C. H.; and Saunders. P. T.: Amputation: energy cost of

ambulation. Arch. Phvs. Med. and Re/iab., 60: 18-24, 1979.

12. Hunt, W. S., Jr., and Thompson, H. A.: Pantalar arthrodesis. A one-stage operation. J. Bone and Joint Siug.. 36-A: 349-361). April 1954.

13. Kegel, Bernice; Carpenter, M. L.; and Burgess, E. M.: Functional capabilities of lower extremity amputees. Arc/i. P/n’s. Med. and Re/lab.

59: 109-120, 1978.

14. Kivilaakso, R.; Langenski#{246}ld, A.; and Sal#{233}nius, P.: Results of talocruralarthrodesis and pantalar arthrodesis of the ankle joint in post-

traumatic conditions labstracti. Acta Orthop. Scandinavica, 36: 339-340, 1965.

15. Liebolt, F. L.: Pantalar arthrodesis in poliomyelitis. Surgery, 6: 31-34, 1939.16. Lorthioir,J.: Huit cas d’arthrod#{232}se du pied avec extirpation temporaire de Uastragale.J. c/iir. unit. -- Soc. BeIge c/hr.. II: 18-1-187. 191 1.

17. McElwain, J. P.; Hunter, G. A.; and English, E.: Syme’s amputation in adults: a long-term review. Canadian J. Surg.. 28: 203-205. 1985.

18. Malone,J. M.; Moore,Wesley; Leal,J. M.;and Childers,S.J.: Rehabilitation forlower extremity amputation. Arc/i. Surg.. I 16: 93-98. 1981.

19. Marek, F. M., and Schein, A. J.: Aseptic necrosis of the astragalus following arthrodesing procedures of the tarsus. J. Born’ an(/Joint Surg..

27: 587-594. Oct. 1945.

20. Mazur, J. M.; Schwartz, Evan; and Simon, S. R.: Ankle arthrodesis. Long-term follow-up with gait analysis. J. Bone (111(1 Joint .curg.. 61 -A:

964-975, Oct. 1979.

21 . Ouzounian, T. J., and Kleiger, Barnard: Arthrodesis in the foot and ankle. In Disorders of the Foot and Ankle. Medical and Surgical

Management, edited by M. H. Jahss. Ed. 2, vol. 3. pp. 2614-2646. Philadelphia. W. B. Saunders. 1991.

22. Patterson, R. L., Jr.; Parrish, F. F.; and Hathaway, E. N.: Stabilizing operations of the foot. A study of the indications. techniques used. and

end results. J. Bone andJoint Stag., 32-A: 1-26, Jan. 1950.23. Russotti, 6. M.; Johnson, K. A.; and Cass, J. R.: Tihiotalocalcaneal arthrodesis for arthritis and deformity of the hind part of the foot.

J. Bone alidioilit Surg., 70-A: 1304-1307. Oct. 1988.

24. Staples, 0. 5.: Posterior arthrodesis of the ankle and subtalar joints. I Bone andJoint Surg., 38-A: 50-58. Jan. 1956.

25. Steindler, Arthur The treatment of the flail ankle; pan-astragaloid arthrodesis. J. Bone and Joint Surg., 5: 284-293. April 1923.

26. Waters, R. L.; Perry, Jacquelin; Antonelli, Daniel; and Hislop, Helen: Energy cost of walking of amputees: the influence of le�el of

amputation. J. Bone and Joint Surg., 58-A: 42-46, Jan. 1976.

27. Waugh, T. R.; Wagner, Jay; and Stinchileld, F. E.: An evaluation of pantalar arthrodesis. A follow-up study of one hundred atid sixteen

operations.J. Bone andfoint Surg., 47-A: 1315-1321,Oct. 1965.