Obtaining Iliac Crest Bone Graft
A very similar operative procedure has been used for all of our patients
who have been managed with an intramedullary rod for the treatment of
congenital pseudarthrosis of the tibia. The patient is placed in a lateral
position on a radiolucent operating table. The entire lower extremity and the
ipsilateral pelvis are prepped and draped. A posterior iliac crest bone graft
is first harvested through a curved incision centered over the posterior
superior iliac spine. Sharp dissection is carried down through the
subcutaneous tissues to the gluteus fascia overlying the posterior superior
iliac spine. Subperiosteal exposure of the outer wall of the posterolateral
surface of the ilium is obtained. In doing so, the iliac apophyseal cartilage
is elevated off the ilium from lateral to medial. This enhances exposure of
the proximal outer table and facilitates the process of obtaining an adequate
bone graft in a younger child. Corticocancellous strips and cancellous bone
graft are obtained. The wound is irrigated, and a suction drain is placed. The
apophyseal cartilage and the gluteal musculature are then securely reattached
to the ilium. The subcutaneous tissue and skin are closed, and the soft
supports about the torso are removed so that the patient can be turned to the
supine position.
Exposure
A sterile tourniquet is placed about the thigh, the leg is elevated, and
the tourniquet is inflated. The pseudarthrosis site is approached directly
anteriorly with a longitudinal incision of several centimeters. Dissection is
sharply carried down to the tibia, and subperiosteal exposure of the
pseudarthrosis site is obtained. Soft tissue is excised from the
pseudarthrosis site, and the two bone fragments are exposed. A small (2.5 to
3.2-mm) drill-bit with power is used to reestablish the medullary canal in
both the proximal and distal fragments. Use of the C-arm ensures that the
drill-bit is optimally centered to prepare the medullary canals for later rod
placement. If the fibula is intact and is distracting the pseudarthrosis site,
it must be osteotomized at this time. The fibula is approached through a
separate longitudinal incision, typically distally in the leg and anterior to
the peroneal musculature. Subperiosteal exposure of the fibula is obtained,
the bone is cut with a small power saw, and the ends are allowed to
overlap.
Rod Selection
The tibia is then stabilized with a Williams rod
(Fig. 1, A and
B). This device, which is available commercially
(Zimmer, Warsaw, Indiana), consists of an indwelling rod and an insertion rod
that are coupled together (Fig.
2). The rod is inserted as a single unit and, when the indwelling
portion of the rod is optimally positioned, the insertional segment is
removed. Preoperative planning is essential in order to select a rod of
appropriate length and diameter. The length of the rod that is needed is
calculated on the basis of the expected length of the tibia after the
pseudarthrosis has been excised and angular deformity has been corrected as
well as on the basis of the level of fixation. Typically, the rod abuts the
proximal tibial physis and transfixes the ankle joint; however, the age of the
child and anticipated remaining growth of the distal part of the tibia are
used to determine the appropriate final location of the distal end of the rod
at the time of surgery. With longitudinal growth, the distal part of the
tibia, the ankle, and the foot migrate distally. Our current strategy is to
minimize the duration of ankle immobilization with the rod. The recommended
ideal rod placements are as follows.
Less than five years old. The distal end of the rod should extend
well into the body of the calcaneus (Fig.
3). The proximal end abuts the tibial physis to aid in providing
long-term stability.Five to eight years old. The distal end of the rod should be more
proximal in the hindfoot, transfixing the ankle but not the subtalar joint.
The proximal end may be left 1 cm distal to the proximal tibial physis to
allow later advancement of the rod.More than eight years old. The ankle does not require
transfixation to adequately stabilize the pseudarthrosis. Therefore, the rod
is placed with the proximal portion abutting the proximal tibial physis and
the distal portion extending to within 1 to 2 cm of the subchondral plate of
the distal tibial epiphysis.
Less than five years old. The distal end of the rod should extend
well into the body of the calcaneus (Fig.
3). The proximal end abuts the tibial physis to aid in providing
long-term stability.
Five to eight years old. The distal end of the rod should be more
proximal in the hindfoot, transfixing the ankle but not the subtalar joint.
The proximal end may be left 1 cm distal to the proximal tibial physis to
allow later advancement of the rod.
More than eight years old. The ankle does not require
transfixation to adequately stabilize the pseudarthrosis. Therefore, the rod
is placed with the proximal portion abutting the proximal tibial physis and
the distal portion extending to within 1 to 2 cm of the subchondral plate of
the distal tibial epiphysis.
CRITICAL CONCEPTSINDICATIONS:A child of any age with an unstable congenital pseudarthrosis of the tibia,
regardless of previous treatment, may be managed with this technique. The
procedure is indicated for most patients who have a congenital pseudarthrosis
of the tibia as long as excessive excision of the pseudarthrosis and/or
pathologic tissues is not necessary.CONTRAINDICATIONS:An infected pseudarthrosis site from previous surgery or a lesion in the
tibia that is so extensive that reduction will produce an undesirable amount
of shortening of the tibia. Amputation may be more appropriate than
reconstruction under these circumstances.PITFALLS:Inadequate preparation ahead of time such that a rod of the optimal
diameter and/or length is unavailable at the time of surgeryEccentric placement of the rod in the canal and/or fragments that are not
anatomically alignedTransfixion of the ankle with the foot in dorsiflexionFailure to adequately immobilize the patient postoperatively, allowing
early motion at the pseudarthrosis site that prevents consolidation from
occurringFailure to recognize that consolidation has not occurred and that the
construct needs to be redoneFailure to reinsert a rod after it has been observed that the distal part
of the tibia has not migratedFailure to address valgus and leg-length inequalities in a timely
fashionAUTHOR UPDATE:The technique has changed since the original article was
published4 in that
we now pay more attention to the location of the rod within the medullary
canal, the alignment of the fragments in both the frontal and sagittal planes,
and the position of the hindfoot when passing the rod across the ankle. We do
not use circumferential wire but rather circumferential absorbable suture
material to secure the barrel-stave bone graft construct. We are more likely
to osteotomize an intact fibula than we are to leave it alone. We now consider
unlocking the ankle earlier than was the case in the past by pushing the rod
across the ankle joint, and we no longer use electrical stimulation in
conjunction with the surgical treatment.
CRITICAL CONCEPTS
INDICATIONS:
A child of any age with an unstable congenital pseudarthrosis of the tibia,
regardless of previous treatment, may be managed with this technique. The
procedure is indicated for most patients who have a congenital pseudarthrosis
of the tibia as long as excessive excision of the pseudarthrosis and/or
pathologic tissues is not necessary.
CONTRAINDICATIONS:
An infected pseudarthrosis site from previous surgery or a lesion in the
tibia that is so extensive that reduction will produce an undesirable amount
of shortening of the tibia. Amputation may be more appropriate than
reconstruction under these circumstances.
PITFALLS:
Inadequate preparation ahead of time such that a rod of the optimal
diameter and/or length is unavailable at the time of surgeryEccentric placement of the rod in the canal and/or fragments that are not
anatomically alignedTransfixion of the ankle with the foot in dorsiflexionFailure to adequately immobilize the patient postoperatively, allowing
early motion at the pseudarthrosis site that prevents consolidation from
occurringFailure to recognize that consolidation has not occurred and that the
construct needs to be redoneFailure to reinsert a rod after it has been observed that the distal part
of the tibia has not migratedFailure to address valgus and leg-length inequalities in a timely
fashion
Inadequate preparation ahead of time such that a rod of the optimal
diameter and/or length is unavailable at the time of surgery
Eccentric placement of the rod in the canal and/or fragments that are not
anatomically aligned
Transfixion of the ankle with the foot in dorsiflexion
Failure to adequately immobilize the patient postoperatively, allowing
early motion at the pseudarthrosis site that prevents consolidation from
occurring
Failure to recognize that consolidation has not occurred and that the
construct needs to be redone
Failure to reinsert a rod after it has been observed that the distal part
of the tibia has not migrated
Failure to address valgus and leg-length inequalities in a timely
fashion
AUTHOR UPDATE:
The technique has changed since the original article was
published4 in that
we now pay more attention to the location of the rod within the medullary
canal, the alignment of the fragments in both the frontal and sagittal planes,
and the position of the hindfoot when passing the rod across the ankle. We do
not use circumferential wire but rather circumferential absorbable suture
material to secure the barrel-stave bone graft construct. We are more likely
to osteotomize an intact fibula than we are to leave it alone. We now consider
unlocking the ankle earlier than was the case in the past by pushing the rod
across the ankle joint, and we no longer use electrical stimulation in
conjunction with the surgical treatment.
Rod Insertion
Once the proximal and distal tibial fragments have been prepared, the
entire rod assembly is inserted into the distal tibial segment and is advanced
antegrade with use of a power source through the medullary canal; across the
distal tibial physis, the ankle joint, and the subtalar joint; and out through
the heel pad (Fig. 4). As the
rod is driven distally, it is imperative that it is centered in the medullary
canal. Monitoring with a C-arm is essential. As the rod is driven across the
ankle joint, it is essential to minimize the tendency for calcaneal and/or
valgus angulation of the foot. The desired position is neutral
dorsiflexion-plantar flexion of the foot and neutral varus-valgus angulation
of the ankle. This position should be verified clinically and
fluoroscopically. The tibial segments are then anatomically aligned at the
pseudarthrosis site, and the rod is driven retrograde into the proximal
fragment (Fig. 5). Again,
direct visualization and the use of fluoroscopy are imperative in order to
confirm that the rod is centralized within the medullary canal of the proximal
segment and the tibia is anatomically aligned in both the coronal and sagittal
planes. Occasionally, because of extensive deformity, an additional osteotomy
of the proximal tibial fragment is necessary in order to ensure optimal
intramedullary rod placement and anatomic alignment of the tibia. The proximal
part of the tibia is osteotomized with a power saw at the point of maximal
angular deformity as determined with fluoroscopy.
The fibula is assessed at this time. If there is a fibular pseudarthrosis,
fixation of the fibula adds stability to the construct. An intact fibula may
or may not distract the tibial reconstruction. If it does not appear to do so,
it should be left intact. If an intact fibula does appear to prevent
apposition of the tibial fragments, it should be osteotomized and fixed if
possible. When stabilization is attempted, the fibula is approached as
described above. We prefer intramedullary fixation with an appropriately sized
Kirschner wire, which is first driven antegrade through the distal fragment
out the tip of the fibula and then retrograde into the proximal fragment. The
end of the wire can be cut and bent so that it rests on the distal part of the
fibula, or it can be countersunk into the fibula after satisfactory clinical
alignment of the leg has been achieved and confirmed fluoroscopically. Fibular
fixation often is not possible because of the small size of the medullary
canal. In this case, the overlapped fibular fragments are temporarily
stabilized with absorbable circumferential sutures.
The tibial insertion rod is removed after appropriate reduction of both the
tibia and the fibula has been achieved. Unscrewing the insertion rod by one
turn and noting the location of the rod junction with use of fluoroscopy
confirms the exact location of the distal end of the indwelling rod. When
satisfactory alignment is obtained, the insertion rod is removed.
Next, the previously harvested cancellous and corticocancellous bone graft
is placed around the tibial pseudarthrosis site and is secured with absorbable
suture as a barrel-stave construct (Fig.
6). A suction drain is placed and is brought out of the wound
through a separate stab incision. The deep fascia of the leg is not closed.
The subcutaneous tissue and skin are closed in separate layers. If a fibular
pseudarthrosis has been repaired, it is also bone-grafted and the wound is
similarly closed in layers.
According to the protocol, children six years of age or less are treated
with a one and one-half spica cast to ensure minimal stress at the
pseudarthrosis site. The spica cast is replaced with an above-the-knee cast at
six to eight weeks postoperatively. Older children are treated with an
above-the-knee cast with the knee flexed at least 30° for approximately
three to four months. Cast immobilization is discontinued at approximately
four to six months postoperatively, once clinical and radiographic examination
indicates that consolidation is occurring.
Once the cast is removed, the involved extremity is protected with a
custom-fabricated clamshell knee-ankle-foot orthosis with a locked ankle joint
and an initially locked knee joint. Adequate circumferential leg contact is an
essential feature of this orthosis, which is worn twenty-four hours a day.
Walking with progressive weight-bearing is initiated at this point. As
consolidation occurs, the knee joint is unlocked. Typically, the orthosis
remains extended to the thigh for a longer period of time for smaller
children. As consolidation progresses, the knee-ankle-foot orthosis can be
converted to a total-contact ankle-foot orthosis.
With longitudinal growth of the tibia, the rod remains in place and the
distal part of the tibia and the foot migrate away from its distal end until
the ankle joint becomes free. This took approximately three years in patients
treated earlier in our study. Currently, to minimize the duration of ankle
joint immobilization and the potential for disruption of the articular
cartilage during the gradual migration of the rod across the joint, we now
consider surgically advancing the rod across the ankle joint with use of a
pusher rod when the distal end of the indwelling rod approaches the articular
surface of the talus. This is done through a small incision on the plantar
surface of the hindfoot under fluoroscopic guidance. Once the pusher rod is
docked on the indwelling rod, a light-weight mallet is used to advance the
indwelling rod out of the talus, across the ankle joint, and into the distal
part of the tibia. Advancing the indwelling rod should not be considered until
one is convinced that the pseudarthrosis site has consolidated. Once the rod
is out of the ankle, the orthosis is unlocked at the ankle. An orthosis is
worn until the patient reaches skeletal maturity and then during sports
activities thereafter.
On occasion, with longitudinal growth of the tibia, the distal part of the
leg does not migrate distally along the rod either because consolidation has
not occurred or because the rod has become "loose" in the proximal
fragment prior to consolidation. If the tibia is not growing away from the end
of the rod, the rod should be removed and replaced. In three cases in the
early part of this series, we simply removed the rod. In all three cases, the
limb went on to refracture. In cases in which the distal part of the leg has
failed to migrate, the rod should be replaced. Bone-grafting may be considered
if consolidation is in question. To help to further secure the rod in the
proximal fragment, methylmethacrylate should be placed around the rod through
a window in the proximal part of the tibia once the rod has been
reinserted.