Intramedullar nail arthrodesis of the
talocrural and subtalar joint in patients with rheumatoid arthritis

K.A. GIANNIKAS
Harlow Wood Orthopaedic Hospital, Near Mansfield, Nottighamshire, UK
King's Mill Centre, Sutton in Ashfield, Nottinghamshire, UK

 

ABSTRACT
Background: Rheumatoid arthritis of the talocrural and subtalar joint present relative rare, but can cause considerable disablement to the affected patients. We present a series of patients with severe rheumatoid arthritis, who underwent fusion of the talocrural and subtalar joint with the assistance of an intramedullary rod.
Methods: Eight patients were identified from the theatre records of two hospitals during a period of 18 years. Clinical notes and radiographs were reviewed. The patients were further invited for a clinical evaluation, and if this was not possible, they were contacted by telephone. The details of the surgical technique are presented, which is dependable on the deformity encountered.
Results: Follow-up ranged between 18 and 83 months (mean 31.2 months). There were no documented early complications apart from a case in which the nail was misplaced during the operation. Patients were encouraged to start weight-bearing from the first postoperative day. In the long term all patients reported significant improvement in their pain.
Conclusions: Retrograde intramedullar arthrodesis can be considered an efficient method in dealing with concomitant pathology of the ankle and subtalar joint. The procedure can be tailored depending to the deformity confronted, adopting a less aggressive attitude to the already described in the literature.
Key words: Rheumatoid arthritis, talocrural joint, ankle joint, subtalar joint, intramedullary nailing.

INTRODUCTION
Over 30 different surgical procedures have been reported in the literature for arthrodesis of the talocrural joint and several more for the subtalar joint. The results of all these procedures are, generally speaking, similar, but not all of them are suitable for the rheumatoid patient because of the compromised tissue quality and the severity of the deformities that are encountered[4]. Furthermore, the alternative of ankle joint replacement has not as yet been proven to be efficient, and its use can be restricted further by a diseased subtalar joint[5].
Retrograde intramedullar nailing for ankle and subtalar arthrodesis using the transcalcaneal approach was first described in 1991 by Stone and Helal9. In 1995, Moore et al[6] published their own experience, and later, from the same institute, Pochtko et al[7] described a cadaveric study to show the ideal nail entrance point. Each of these series included rheumatoid patients but did not attempt to adapt their procedures to the specific needs and pathology of the rheumatoid patient. Nevertheless, the results achieved were similar, if not better, than the other methods of arthrodesis of the two joints[2], establishing the usefulness of the technique.
The scope of this article is to present the results of a surgical technique that has been adapted to treat specifically the needs, the expectations, and the pathology of the patient with rheumatoid arthritis.

PATIENTS AND METHODS
A retrospective review was carried out of all patients with rheumatoid arthritis who had undergone intramedullary nailing for talocrural and subtalar arthrodesis from 1980 until 1997. Ten rheumatoid feet belonging to 10 patients were identified from the theatre records of Harlow Wood Orthopaedic Hospital and King's Mill Centre. Records and radiographs were retrieved and examined. Two of the patients had to be excluded from the study because no sufficient records and/or radiographs could be found. Of the remaining patients the indications, the clinical and radiographic severity of their disease, the operative procedure, and the postoperative course was studied. Six of the patients were clinically reviewed. A further patient was unable to attend and was therefore interviewed on the telephone. The remaining patient deceased three years following her operation. Her last clinical review (two years) was deemed as her final outcome. Follow up ranged between 18 and 83 months (mean 31.2 months).
All eight patients were female. Five of the operations were on the left side and three on the right. Seven of the patients were diagnosed with seropositive erosive rheumatoid arthritis and one patient had seronegative erosive rheumatoid arthritis. Two of the seropositive patients had associated Sjogrens disease, a third had an overlapping syndrome with chronic active hepatitis and primary biliary cirrhosis and a fourth had an overlap with systemic lupus erythematous.
Radiographic appearance in all cases was consistent with stages 3 and 4 as graded by the American Rheumatism Associations classification system[1]. The duration of their disease ranged between 13 to 33 years (mean: 20,4 years).

1a.

1b. 1c.
Figure 1, 2. Preoperative, 1st day postoperatively and 10 months after the operation. The patient was full weight bearing completely painfree one month after the operation.

Operative technique
Indication for the operation in all patients was the failure of conservative treatment for painful talocrural joint movement with an ankylosed or painful subtalar joint.
The operations involved patients under the care of four different orthopaedic consultants, but the same principles were adopted by all. The operations undertaken can be grouped in three different types, depending on the degree of destruction of the ankle and hindfoot:
Type A. In three cases the alignment of the hindfoot and the ankle joint could be achieved with closed means. The operative procedure involved a small incision on the plantar surface of the foot and after manipulating the hindfoot at the desired position, a guide wire was introduced under radiographic control through the os calcis to the tibial medulla. After reaming, an intramedullary nail was introduced and the wound closed. A below knee POP back-slab was fashioned at the end of the operation. Two days after, the plaster was completed and the patient was allowed to partially weightbear. Full weightbearing was encouraged 2 weeks postoperatively, pain permitting. In 6-8 weeks the plaster was removed and the patient allowed to return to her normal activities.
Type B. In two cases where the valgus deformity could not be corrected by closed means, through a lateral approach, a wedge was removed at the talocalcaneal junction, removing, therefore, the subtalar joint and correcting the deformity. After that an intramedullary nail was introduced as previously described. Postoperative management was the same.
Type C. In the three remaining cases the degree of destruction of the hindfoot was such, that complex reconstructive surgery was required. The operative procedure was customised according to the individual deformity but as a principle both the ankle and the subtalar joints were exposed, and after the necessary soft tissue release and debridment of the articular cartilage remains, the hindfoot was positioned at the desired position. An intramedullary nail was then introduced to act as a splint to maintain the alignment of the hindfoot. In all cases bone graft was placed in the defects. Postoperative rehabilitation was prolonged depending on individual circumstances and ranged from 8-12 weeks. Partial weight bearing was generally allowed during this period to allow the patient independent mobilisation.
In all cases, a short Kuntscher nail was used, apart from the earliest case in this series where a Smith Petersen nail was inserted. A minor degree of plantarflexion and up to 100 of valgus of the hindfoot was generally accepted in all types of operations, while varus deformity was avoided.

3.
Figure 3. Twenty months after "type A" arthrodesis using a Kuntscher nail. There is evidence on the lateral view of a conic osseous defect at os calcis, indicating movement at the ankle joint. This movement was also evident clinically, but the patient never complained of any pain.

 

RESULTS
No wound infections were documented, and in all cases the immediate postoperative period was uncomplicated. At their latest review, all patients reported either minimal or no hindfoot pain, and considerable improvement of their general mobility.
In the group that had the first and the second type of operation, the only complication was in a lady whose hindfoot drifted to 15o valgus and the nail migrated distally without causing sufficient pain to warrant its removal.
In the third type of operation, that complex primary reconstructive surgery was required, one of the patients had the nail introduced in the soft tissues of the lower leg instead of the tibial medullary canal. In the specific patient the guide wire was not introduced under radiographic control as most of the operation was performed under direct vision. As a result the foot returned to the pre-operational deformity, without, though, any painful movement at her ankle and subtalar joint. The patient although dissatisfied with the cosmetic outcome of the operation and the fact that she has to mobilise wearing an AFO, claims to be pain free and can walk without any aid. Another patient in the same group was mobilising without any discomfort and was discharged from follow up 16 months after her operation. Four months after the discharge, she sustained an injury at her arthrodesed hindfoot, fracturing the nail at the level of the ankle joint. The broken distal part of the nail was removed and a below knee weight bearing plaster was applied for three months. As the patient continued to experience pain, a bone scan was performed that showed high uptake at the ankle region. The ankle joint was explored through a posterior approach and the subtalar joint was bone-grafted again. She remained for a further three months in a weight bearing plaster and finally progressed to full union.

DISCUSSION
The talocrural joint is not commonly involved in rheumatoid arthritis, as the disease affects more the distal than the proximal foot joints[10]. On the other hand, the subtalar joint tends to be implicated more frequently and earlier than the talocrural joint, and although its destruction may be acutely painful, usually it occurs silently and the patient may not be seen until it is completly destroyed, drifting the hindfoot into valgus. Failure of conservative treatment is the usual indication for surgical correction of the rheumatoid ankle and hindfoot, which can present many challenging problems due to the compromised bone stock and the often superimposed osteoarthrosis.
The transcalcaneal route for ankle and subtalar stabilisation has not gained popularity until relatively recently. As a route, it was first described in 1958 by Sokolowski in Poland[8] for primary stabilisation of severe fracture-dislocations of the ankle joint. Later Childress in 19653, tried to popularise the method as a last resort in severe ankle fractures. Both surgeons used a single Steinmann pin, and reported good ankle and subtalar motion after completion of the treatment.
The route was not mentioned further until 1991, when Carrier and Harris reported a series of 5 patients with Rheumatoid Arthritis that had their ankle joint arthrodesed by inserting two Steinmann's pins transcalcanealy[2]. Their technique also involved decortication of the ankle and subtalar joint surfaces and excision of a small segment of the distal fibula to allow compression.
At the same year Stone and Helal reported a series of 20 operations, 11 of which were in rheumatoid patients. In this series, they described fusion of the ankle and subtalar joint by using a long trifin nail after excising all the remaining articular cartilage through an anterolateral approach[9]. Through the same approach they harvested bone graft from the exposed distal tibial metaphysis that was later deposited at the arthrodesis sites. They reported good results in 19 of their cases.
A further report by Moore et al in 1995, described 19 fusions, 5 of which in rheumatoid ankles[6]. Their operative technique involved removal of a distal fibular segment and excision of the talocrural articular cartilage through two separate incisions. Patients that had subtalar joint movement preoperatively, had the subtalar articular surface denuded, and one rheumatoid patient, had a distal tibial metaphysial osteotomy to correct a varus malalignment. A variety of intramedullary nails were used, most of which were Alta femoral rods. In the majority of the cases the nails were locked proximally or distally. Bone graft from the reaming was impacted in the arthrodesis sites. Results in the rheumatoid patients were satisfactory and only one patient complained of midfoot pain while ambulating and required an AFO.
In all the above mentioned articles emphasis is given in the need of meticulous dissection and removal of all articular surfaces, so as to assure an optimum result. Ambulation of the patient was almost immediate, an advantage that comes in contradiction with most of the other described methods of ankle arthrodesis.
In our experience, a minimally invasive attitude was adopted, specially tailored to suit the expectations and the pathology of the rheumatoid patient. In the event that complex reconstructive surgery for gross subluxation was not required, no attempt was made to remove a fibular segment or to denude the articular surfaces from the ankle, or-if the deformity of the hindfoot permitted- from the subtalar joint. This thought to be unnecessary as, the natural history of the rheumatoid disease is progression to ankylosis once movement is abolished. Furthermore limited bone grafting is inherent to the procedure during the process of reaming. The additional advantages of this procedure are clear, as the rheumatoid tissues are minimally disturbed, thus reducing the possibility of perioperative infection, and, at the same time, diminishing postoperative pain permitting early ambulation of the patient. It was difficult to confirm the presence of osseous union from the radiographic appearance. Indeed, in one case there was definite radiolucency around the distal nail, indicative of persistant movement. Nevertheless, this did not appear to adversely affect the outcome. All patients improved their mobility significantly immediately postoperatively, and achieved a cosmetically acceptable outcome.
With regard to the ankles that required complex reconstructive surgery, our surgical technique was similar to that already described in the literature. In these cases the intramedullary nail served as a valuable splint, which made realignment and bone grafting possible. Apart from the case where the nail was introduced in the soft tissues a functional and cosmetically acceptable outcome was achieved.
The only disadvantage of the transcalcaneal route is that it can only be used if there is concomitant pathology in both talocrural and subtalar joints not responsive to conservative management. This is a rare combination usually occuring late on in the course of severe erosive rheumatoid arthritis[10].

CONCLUSIONS
Retrograde intramedullary ankle arthrodesis is a highly efficient method that can be used for the surgical treatment of the combined talocrural and subtalar rheumatic joints. The procedure can be tailored depending to the deformity confronted, adopting a less aggressive attitude to that already described in the literature, ideal suited for minimally interference with the fragile environment of the rheumatoid foot. Furthermore the patient can start ambulating almost immediately after the operation, optimising independence and reducing the complications that are associated with long term immobilisation.

REFERENCES
1. American Rheumatism Association. Diagnostic criteria for rheumatoid arthritis. Ann Rheum Dis. 1959; 18, 49-54.
2. Carrier D.A., Harris C.M. Ankle arthrodesis with vertical Steinmann's pins in rheumatoid arthritis. Clin Orthop and Rel Res. 1991; 268, 10-14.
3. Childress H.M. Vertical transarticular -pin fixation for unstable ankle fractures. Journal of Bone and Joint Surgery. 1965; 47A, 323-34.
4. Crachiolo A. Operative technique of the ankle and hindfoot. In Helal B, Wilson D (eds) The foot. Edinburgh, Churcill Livinstone 1988; 1205-44.
5. Kirkup J. Rheumatoid arthritis and ankle surgery. Ann Rheum Diseas. 1990; 49, 837-44.
6. Moore T.J., Prince R., Potchatko D., Smith J.W. Retrograde intramedullary nailing for ankle arthrodesis. Foot and Ankle Intern 1995; 16(7), 433-6.
7. Pochtko D.J., Smith J.W., Phillips R.A., Prince B.D., Hedrick M.R. Anatomic structures at risk: Combined subtalar and ankle arthrodesis with a retrograde intramedullary rod. Foot and Ankle Intern. 1995; 16(9), 542-7.
8. Sokolowski T. Transfissione percutanea tibiale con chiordo di Steinmann nel trattamento di fratture bimalleolary con dislocazione. Minerva Med. 1958; 49, 2669-7.
9. Stone K.H., Helal B. A method of ankle stabilisation. Clinc Orthop Relat Research 1991; 268, 102-6.
10. Vigigal E., Jacoby R.K., Dixon A.S.J., Ratcliff A.H., Kirkup J. The foot in chronic rheumatoid arthritis. Ann Rheum Dis. 1975; 34, 292-7.

Mailing address:
K.A. Giannikas FRCS
312 Wakefield Road
Stalybridge
SK15 3BY
United Kingdom