One hundred and ninety patients ranging from seventeen to seventy years of
age who had stage-2 or 3 posttraumatic ankle arthritis
(Table I) were managed between
1994 and 2004.
Surgical strategies were decided according to the stage of arthritis, the
age of the patient, the quality of joint alignment, and the range of motion of
the adjacent foot joints (the midtarsal, Lisfranc, and subtalar joints)
according to the algorithm described in
Table II.
All patients were examined clinically and radiographically, and the
American Orthopaedic Foot and Ankle Society (AOFAS) clinical rating
score7 was
calculated both preoperatively and at the time of follow-up. The AOFAS score
was graded as excellent (80 to 100), good (70 to 79), fair (60 to 69), or poor
(=59). Patients who were managed with allograft were also studied with
computed tomography and magnetic resonance imaging. A cartilage biopsy was
performed during hardware removal at 1.5 years of follow-up.
Surgical Strategies
Stage-2 Arthritis with Preserved Ankle Anatomy: Arthroscopic
Débridement and Ankle Arthrodiastasis
(Fig. 1)
Twelve patients ranging from twenty-one to forty-eight years of age were
managed with arthroscopic débridement and arthrodiastasis with an
external fixator8.
Standard arthroscopic approaches were used, and débridement of
degenerated cartilage and fibrous tissue was performed. In seven cases,
chondral damage of the talus was treated with
microfracture9.
A monolateral hinged ankle external fixator (Orthofix, Bussolengo, Italy)
was applied to maintain the joint in distraction
(Figs. 2-A, 2-B, and 2-C).
Postoperatively, the external fixator was maintained for four weeks, during
which time partial weight-bearing was allowed and active and passive motion of
the ankle joint was encouraged. The external fixator was removed at four
weeks, and weight-bearing and physiotherapy progressed over the next four
weeks.
Stage-2 Arthritis with Supramalleolar Malalignment: Supramalleolar
Osteotomy (Fig. 3)
Eight patients ranging in age from seventeen to forty-five years were
managed with supramalleolar
osteotomy10.
Through a standard approach to the distal part of the tibia, a corrective
osteotomy of the deformity was performed, usually 4 cm proximal to the
articular surface. In six patients the osteotomy gap was filled with a wedge
of allograft bone to restore alignment
(Figs. 4-A, 4-B, and 4-C), and
in the other two patients a closing-wedge osteotomy was performed. Fixation
was achieved either with a plate and screws or with two Kirschner wires.
Postoperative treatment consisted of plaster cast immobilization for six
weeks without weight-bearing. This was followed by partial weight-bearing in a
fracture boot that permitted the patient to perform active and passive ankle
motion exercises. When there was radiographic evidence of bone healing (at an
average of twelve weeks), full weight-bearing was permitted.
Stage-2 Arthritis with Intra-Articular Malalignment: Joint
Reconstruction (Fig.
5)
Forty-two patients ranging from nineteen to fifty-six years of age were
managed with ankle
reconstruction11
with use of a lateral approach to the ankle joint. In twenty-five cases,
malunion of the lateral malleolus was corrected by creating a fibular
suprasyndesmotic osteotomy and lengthening the fibula. A bone graft, harvested
directly from the lateral side of the tibia, was interposed in the fibular
defect and was stabilized with a plate and screws
(Figs. 6-A, 6-B, and 6-C). In
seventeen patients, the fibular shortening was associated with lateral tibial
plafond depression and the fibular lengthening was combined with elevation and
bone-grafting of the articular surface
(Figs. 7-A, 7-B, and 7-C).
Associated procedures included medial malleolar revision to treat a malunion
(eleven patients) and repair of a chronic tear of the deltoid ligament (nine
patients). In twenty-five patients, Achilles tendon lengthening was also
performed to correct tendon contraction produced by the prolonged state of
deformity and to obtain at least a neutral position of the ankle once the
deformity was corrected.
Postoperative treatment consisted of plaster cast immobilization for three
weeks without weight-bearing, followed by non-weight-bearing in a fracture
brace for six weeks to permit daily active and passive range-of-motion
exercises. Partial weight-bearing was then permitted for the next four weeks.
After there was radiographic evidence of bone-healing (at an average of twelve
weeks), full weight-bearing was permitted.
Stage-3 Arthritis with Preserved or Restored Ankle Anatomy in
Patients Less Than Fifty Years of Age and with <25° of Motion in Other
Foot Joints and Arthritis in Other Foot Joints or in Patients Who Refused
Arthrodesis: Allograft (Figs. 8-A and
8-B
Eighteen patients ranging from nineteen to fifty years of age were managed
with a fresh bipolar shell osteochondral
allograft12-14.
The allograft was harvested from a cadaver donor and was preserved in a
chondroprotective antibiotic solution at 4°C for as long as fourteen days.
A patient was selected from a special waiting list when an appropriate graft
became available. The choice of the patient was based on a
patient-to-allograft match obtained by means of computed tomographic scanning.
Surgery involved preparation of the allograft by using designed jigs to obtain
articular surfaces with a 1-cm thickness of subchondral bone for both the
tibial and talar components. With use of a 2-cm incision technique, the
patient's ankle was exposed and 1 cm was resected from both articular surfaces
with the same jig that was used for the graft. Then the graft was implanted
and was fixed with twist-off screws (Figs.
9-A through 9-F).
Postoperative treatment consisted of non-weight-bearing in a cast-boot for
four months. Active and passive range of ankle motion was performed after two
weeks. Partial weight-bearing was permitted after four months, while full
weight-bearing was permitted after there was evidence of bone-healing (at an
average of eight months).
Stage-3 Arthritis with Preserved or Restored Ankle Anatomy in
Patients More Than Fifty Years of Age or with <25° of Motion in Other
Foot Joints or in Patients Who Refused Arthrodesis: Ankle Prosthesis
(Fig. 10)
Fifty-two patients ranging in age from fifty-one to seventy years were
managed with three different designs of total ankle replacement: twenty
patients received an LCS implant (New Jersey Low Contact Stress; Endotec,
South Orange, New Jersey), twenty received a STAR implant (Scandinavian Total
Ankle Replacement; Waldemar-Link, Hamburg, Germany), and twelve received a BOX
implant (BOX Total Ankle Replacement; Finsbury Orthopaedics, Leatherhead,
United
Kingdom)15,16.
A standard anterolateral or anteromedial approach was used. Tibial and
talar surfaces were prepared with use of specific instrumentation, ensuring
correct evaluation of alignment of the axis of the tibial component in both
the frontal and sagittal planes and minimizing the amount of bone resection.
During preparation of the talus, special care was taken not to damage the
subchondral bone. After prosthesis implantation, other procedures that were
performed included Achilles tendon lengthening (thirty-eight patients),
ligament reconstruction (six), correction of deformity of the foot (eight),
and reduction of previous mortise widening with a syndesmotic screw
(five).
Postoperative treatment consisted of plaster cast immobilization without
weight-bearing for two weeks, followed by the use of a fracture brace without
weight-bearing for two weeks. Active and passive ankle motion was begun at two
weeks. After four weeks, partial weight-bearing was permitted in the fracture
brace for four weeks, and full weight-bearing was begun after eight weeks
(Figs. 11-A and 11-B).
Stage-3 Arthritis with Preserved or Restored Ankle Anatomy and
>25° Range of Motion in Foot Joints, No Arthritis in Other Foot Joints,
or Nonrestorable Ankle Anatomy: Arthrodesis
(Fig. 12)
Fifty-eight patients ranging from eighteen to fifty-eight years of age were
managed with ankle arthrodesis according to the method of Merle
d'Aubigné17.
Through a lateral approach, a fibular osteotomy was performed 2 cm proximal to
the ankle joint. Two parallel cuts were performed, one through the distal part
of the tibia and one through the talar dome, with resection of the arthritic
articular surfaces and exposure of the cancellous bone. The foot was
positioned in 5° to 10° of valgus and 5° to 10° of external
rotation. In thirty-six patients with good range of motion of the midtarsal
joint, the ankle was positioned in 90°, whereas in twenty-two patients
with midtarsal joint arthritis, the ankle was positioned in 10° of
dorsiflexion. Fixation was obtained primarily with one hydroxyapatite-coated
6.5-mm cancellous
screw18
(Fig. 13) inserted from the
anterolateral aspect of the tibia into the body of the talus
(Figs. 14-A and 14-B).
Postoperative treatment consisted of plaster cast immobilization for six
weeks without weight-bearing, followed by an additional six weeks in the cast
with complete weight-bearing. Full unprotected weight-bearing was allowed
after bone-healing was evident radiographically, usually by twelve weeks.
Statistical Analysis
All continuous data are expressed in terms of the mean and the standard
deviation of the mean. The data in the groups were compared with use of the t
test (in groups with at least twenty patients) or the Wilcoxon test evaluated
with the Monte Carlo method for small samples. The level of significance was
set at p < 0.05.
After a mean duration of follow-up of five years (range, two to ten years),
190 ankle joints had been fused or salvaged; among these joints were 108
excellent, fifty-seven good, and thirteen fair results according to the AOFAS
score. The remaining twelve joints had a poor result and underwent reoperation
(Fig. 15).
The sixty-two patients with stage-2 arthritis presented with a mean
preoperative AOFAS score of 35.5 ± 8.9. At the time of the latest
available follow-up (mean, five years), the mean score was 80.5 ± 11.4
(p < 0.05). Fifty-five patients had a satisfactory clinical result with no
need for additional surgical treatment, despite radiographic evidence of
slight progression of arthritis over time. Of the seven ankles that had a poor
result in this group, three had undergone reconstruction and four had
undergone unsuccessful arthrodiastasis. These ankles were treated with
arthrodesis or total ankle replacement at an average time of four years after
the first intervention.
The eighteen patients with stage-3 arthritis who were managed with an
allograft had a mean AOFAS score of 28.3 ± 12.3 preoperatively and 83.3
± 11.2 at the time of the most recent available follow-up (minimum, two
years) (p < 0.05). Two patients had a poor result, necessitating an
arthrodesis in one patient and revision surgery in the other. A cartilage
biopsy at the time of hardware removal confirmed >90% chondrocyte viability
in the cases of sixteen patients. All of the sixteen successful allografts
demonstrated mild to moderate arthritis of the transplanted surfaces
radiographically at the time of follow-up, although these changes could not be
correlated with pain, reduced motion, or loss of function.
In the group of fifty-two patients with stage-3 arthritis who were managed
with total ankle arthroplasty, the mean AOFAS score improved from 30.3
± 14.3 preoperatively to 84.4 ± 10.4 at the time of the most
recent available follow-up (p < 0.05). Three patients with a poor result
underwent revision surgery (arthrodesis).
Among the fifty-eight patients with stage-3 arthritis who were managed with
arthrodesis, the mean AOFAS score was 28.8 ± 11 preoperatively and 77.5
± 8 at a mean of seven years postoperatively (p < 0.05). No revision
surgery was performed. A review of the radiographs showed progression of
arthritis in the ipsilateral midtarsal and/or subtalar joint in twenty of
these fifty-eight patients.