Download NEUROMUSCULAR SCOLIOSIS TECHNICAL

Neuromuscular Scoliosis
Technical Monograph
Intra-operative Techniques
for Complex Deformities in a
Diverse Patient Population
Contents
Author
Title of Chapter
1. Betz, Randal
Neuromuscular Scoliosis: Introduction
2. Shah, Suken
Sacropelvic Fixation Techniques
Page
1-4
5-16
3. Betz, RandalUnique Challenges with Scoliosis and
Dislocated Hips
17-24
4. Dabney, Kirk
The Correction of Pelvic Obliquity Using
Cantilever Correction
25-32
5. Shah, Suken
Correction of Pelvic Obliquity with an
All Screw Construct
33-44
6. Dabney, Kirk
Scoliosis in Cerebral Palsy
45-56
7.Gabos, Peter
Spinal Fusion in Myelomeningocele
57-64
8. Samdani, Amer
Spinal Cord Injury: Surgical Considerations
Particularly with Respect to Sagittal Plane
65-72
9. Mackenzie, William
Surgical Management of Neuromuscular
Spinal Deformity – Muscular Dystrophy &
Spinal Muscular Atrophy
73-80
10. Cahill, Patrick
Complications in Neuromuscular Scoliosis Surgery
81-88
Neuromuscular Scoliosis: Introduction
Randal R. Betz, MD
Neuromuscular scoliosis is a very common condition, but
it presents the orthopaedic surgeon or spine surgeon with
many treatment challenges. The patient population is diverse,
and the deformities can be complex. Most surgeons have
developed their surgical treatment plans and intraoperative
techniques through experience. While many different
treatments work in a given surgeon’s hands, there is no one
place that a younger surgeon or an experienced surgeon
presented with a previously unseen deformity can go to read
and gain knowledge from the vast experience of experts in
the field.
1
Neuromuscular scoliosis is a very common condition, but
it presents the orthopaedic surgeon or spine surgeon
with many treatment challenges. The patient population
is diverse, and the deformities can be complex. Most
surgeons have developed their surgical treatment plans
and intraoperative techniques through experience. While
many different treatments work in a given surgeon’s
hands, there is no one place that a younger surgeon or an
experienced surgeon presented with a previously unseen
deformity can go to read and gain knowledge from the
vast experience of experts in the field.
The purpose of this monograph is to present in a
short, concise fashion some of the experiences of
master surgeons, including key aspects of their surgical
technique, ways to avoid problems, and how to deal with
complications. The authors were instructed to assume
that the reader has a significant knowledge base and
some experience treating paralytic spine deformities using
basic instrumentation. This monograph will highlight the
key components of the surgeons’ techniques in treating
neuromuscular scoliosis arising from different diagnoses.
A key component of correction of neuromuscular scoliosis
is that the instrumentation in the majority of cases extends
down into the sacrum and the ileum. Multiple options for
fixation will be presented in a chapter by Dr. Suken Shah,
including the use of the S2 screw (S2AI screw fixation)
developed by Drs. Sponseller and Kebaish. Additional
chapters will expand on the use of the sacropelvic fixation
to include correction of pelvic obliquity through the use
of cantilever maneuvers via unit rods (by Dr. Kirk Dabney)
and using an all screw construct (Dr. Shah). The new
EXPEDIUM® Sacropelvic Collection provides a considerable
assortment of fixation options into the sacrum and
ileum, along with lateral connectors (closed, open, and
polyaxial) to help facilitate fixation into this region. Drs.
Dabney and Miller have extensive experience correcting
pelvic obliquity purely by coronal cantilever maneuvers.
In one chapter, Dr. Dabney will discuss the use of the
unit rod as well as the precontoured rods and proximal
connectors from the new EXPEDIUM Neuromuscular
System. These pre-contoured rods and connectors mimic
a u-rod to help correct the coronal deformity and pelvic
obliquity. The use of distraction and compression alone
with an all screw construct is not enough to correct pelvic
obliquity; correction may require derotating the lumbar
and thoracolumbar spine with the use of pedicle screws as
described in the chapter by Dr. Amer Samdani.
2
While not discussed specifically in this monograph, many
surgeons are now using halofemoral traction to correct
some of the pelvic obliquity before a surgical incision is
made. The surgeon should keep in mind that the use of
halofemoral traction may be counterproductive in cases
of hyperlordosis because the hips must be kept flexed to
correct the hyperlordosis, and pulling with femoral traction
is counterproductive.
The presence of subluxed or dislocated hips in the
patient with neuromuscular scoliosis presents unique
challenges. The majority of neuromuscular disorders
that include subluxed or dislocated hips usually fall into
one of three diagnoses, which include 1) spinal cord
injury; 2) myelomeningocele; and 3) cerebral palsy.
However, the highest prevalence of dislocation is probably
caused by sepsis, and it is essential that any relatively
acute, radiographically identified dislocated hip have
an aspiration and adequate work-up to rule out sepsis.
Generally, the spine deformity is treated first followed
by the hip, for reasons explained in Chapter 3 by Dr.
Randal Betz which include the high incidence of deep vein
thrombosis if the hip surgery were to be performed first.
Perioperative medical management, for prevention both of
deep vein thrombosis and heterotopic ossification about
the hip following surgery, presents additional challenges
for the spine surgeon.
Spinal deformity secondary to cerebral palsy presents
unique issues with regards to a hyperlordotic spine. Dr.
Dabney will present some of his correction techniques and
how he capitalizes on some of the technical advances such
as reduction pedicle screws.
Patients with myelodysplasia and spine deformity present
their own challenges. Dr. Peter Gabos will provide
suggestions on three major things that have changed his
practice and resulted in significant improvements for these
patients: 1) He includes PLIFs (posterior lumbar interbody
fusions) where there are no lamina; 2) he uses S2 screws
(S2AI screw fixation) to provide a low profile but good
fixation in the pelvis; and 3) he enlists the help of plastic
surgeons to complete the closure to help reduce the high
infection rate.
Dr. Samdani will describe treatment of patients with spinal
cord injury and matters to consider during preoperative
planning. Many of these patients use compensatory
spinal motion to assist with getting their hand to their
mouth/face for feeding/grooming. Placing them in a
standard prefixed sagittal profile (unit rod) may eradicate
this compensatory motion and render these patients
completely dependent, whereas they were independent
prior to the spinal fusion. These patients may need a
different sagittal profile emulating their normal sitting
posture in a wheelchair which requires a reduced lumbar
lordosis and enhanced thoracolumbar kyphosis. Dr.
Samdani will also describe how to assess the sagittal
profile preoperatively and execute it intraoperatively.
Dr. William MacKenzie will discuss the needs of patients
with muscular dystrophy. Many of these patients are much
smaller at the time of spinal correction and fusion and are
less tolerant of the massive blood loss. The necessity of
fusing or not fusing to the pelvis is discussed.
Finally, a chapter by Dr. Patrick Cahill will discuss some of
the unique complications that can occur in patients with
neuromuscular scoliosis. The most common and more
devastating for the patient is wound infection, and he
discusses suggestions for prevention and treatment based
on evidence from the literature.
The authors are very appreciative of the opportunity
to collectively incorporate our experience into this
monograph. We thank Ines Troconis Merolli for her
support and DePuy Synthes Spine for financially backing
this important educational contribution. We also thank
Carolyn Hendrix for her editorial assistance with this
project.
3
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-60-000 9/2012 JC
CA #8780 / DJ10051A
Sacropelvic Fixation Techniques
Suken A. Shah MD
Indications for sacropelvic fixation include long spinal fusions
for scoliosis, high grade spondylolisthesis, pelvic obliquity
correction, sagittal plane deformity correction, sacral
fractures and lumbosacral fusions in patients with poor bone
quality and osteoporosis. Each of these indications requires
strong distal fixation to resist significant flexion moments
and cantilever forces in this transitional area. Despite many
advances and developments in spinal instrumentation
techniques, fixation failure at the lumbosacral junction
continues to be a challenge.
Due to the significant biomechanical forces across this
junction and relatively poor sacral cancellous bone quality,
fusion at the lumbosacral junction has been associated with
rates of pseudarthrosis as high as 33-39%,1-2 loss of lordosis1-2
and instrumentation failure, especially with historical methods
like body casting, Harrington and Luque instrumentation.
Cotrel-Dubousset (CD) instrumentation and the Galveston
technique were introduced in the 1980’s and although
pseudarthrosis rates were lower, there were instrumentationrelated complications with the CD system and technical
difficulties with rod contouring, insertion in the ilium and rod
loosening with the Galveston technique.
5
Biomechanics
Techniques of Sacropelvic Fixation
McCord, et al pointed out the importance of the
lumbosacral pivot point, which was defined as the point
at the middle osteoligamentous column between the last
lumbar vertebra and the sacrum. The further the implants
extend anterior to this point, the greater the stiffness
of the construct. Various instrumentation models (iliac
fixation, S1 fixation, and S2 fixation) were tested. The
two constructs that withstood the greatest load before
failure had caudad iliac (rod or screw) fixation and they
concluded that crossing the sacroiliac joint is warranted if
instrumentation extends anterior to the pivot point.3
Various contemporary techniques of sacropelvic fixation
have been described to achieve better fusion rates,
improve bone purchase, neutralize forces, avoid pull-out,
ease difficulty and reduce complications. In this chapter,
I will cover the key components of surgical techniques
for the majority of the sacral and iliac fixation options.
This will include the following: 1) Galveston and unit rod
technique, 2) Iliac bolts/screws with and without offset
connections to the longitudinal rods, 3) Double iliac bolts
and 4) S2AI screw fixation.
O’Brien identified three distinct zones of the sacropelvic
region. Zone 1 consists of the S1 vertebral body and the
cephalad margins of the sacral alae; Zone 2, the inferior
margins of the sacral alae, S2, and the area extending
to the tip of the coccyx; and Zone 3, both ilia. Fixation
strength improves progressively from Zone 1 to Zone 3.
Zone 3 offers the greatest biomechanical fixation strength
to counter the pull-out forces and bending moments
at the lumbosacral junction. Also, in agreement with
McCord, iliac fixation with pelvic screws or iliac rods
allows placement of implants more anteriorly beyond the
lumbosacral pivot point than any other implant type.4
Lebwohl and colleagues performed a biomechanical
comparison of lumbosacral fixation in a calf spine model
and noted that supplementary fixation distal to S1 pedicle
screws provides a benefit over S1 fixation alone, and iliac
fixation was superior to a second point of fixation in
the sacrum.5
Cunningham, et al studied ex vivo porcine spines
biomechanically to ascertain the value of anterior column
support compared with that of iliac fixation in lumbosacral
fusions. When tested to failure, the authors found that
iliac screws significantly reduced lumbosacral motion,
particularly with axial rotation, flexion-extension, and
lateral bending. Iliac fixation was found to be more
protective of S1 screws and more resistive of motion than
anterior interbody cages.6
However, other investigators have argued in favor of
anterior column support, particularly at L4-L5 and L5S1, in long fusions to the sacrum.7,8 Anterior lumbar
interbody fusions place the bone graft ventral to the
instrumentation and the lumbosacral pivot point, and
the graft is placed in compression to optimize fusion and
stability.8 This procedure improves the overall chance of
fusion, decreases the strain on caudal pedicle screws and
has a definite role in long fusions to the sacrum, especially
in adults and other patients at risk for pseudarthrosis.
6
1. Galveston and Unit Rod Technique
The Galveston technique allows for the incorporation
of the ilium into the foundation of the construct via the
insertion of rods between the inner and outer tables of
the ilium, which provides a broader base and a more
biomechanically advantageous position.1,3 The transverse
portions of the rods are inserted under a large muscle
flap and enter the ilium at the posterior superior iliac
crest. The orientation is approximately 30° to 35° caudally
and 20° to 25° laterally. The rods cross the SI joint, and
contouring can be difficult.9 The technique lowers the
pseudarthrosis rate of long fusions to the sacrum10 but it
is also associated with a moderate incidence of loosening
secondary to micromotion at the rod tips within the
ilium, despite lumbosacral fusion.11 Radiographically,
this is described as a windshield-wiper effect and may be
associated with pain and the need for implant removal.11,12
However, in our long term experience of unit rod fixation
in cerebral palsy patients, this has not been a common,
symptomatic issue requiring re-operation.
Lonstein and colleagues recently published their results
and complications of 93 patients with cerebral palsy
and scoliosis who underwent PSF with Luque-Galveston
instrumentation with an average follow up of 3.8
years.12 Coronal curve correction was 50% and pelvic
obliquity correction was 40% at latest follow-up. The late
complication rate was 47% and included the windshield
wiper sign, junctional kyphosis, pseudarthrosis (7.5%) and
implant problems including breakage, dislodgement and
prominence. Seven patients required reoperation, most
commonly for pseudarthrosis and/or failed implants.
The unit rod, by virtue of its pre-contoured unibody
construction, provides rigid control of spinal deformities
involving pelvic obliquity and allows for a cantilever
mechanism to correct the pelvic obliquity and the scoliosis
simultaneously.13 The rods are available in various lengths
with corresponding thoracic, lumbar and pelvic contours.
Our specific technique of unit rod insertion follows.
At the inferior margin of the incision, the outer wing of
the ilium is subperiosteally exposed down to the sciatic
notch and sponges are packed out over the pelvis to
maintain hemostasis. The right and left drill guides for
the unit rod are placed in the respective sciatic notch;
care should be taken to ensure that the drill guide is as
inferior as possible along the posterior superior iliac spine
(PSIS). The handles of the drill guide are the reference
points for alignment: the lateral handle should be parallel
with the pelvis and the axial handle parallel with the
sacrum. Next, the drill hole is made utilizing the guide
using a 3/8” drill to the predetermined depth directed
towards the anterior inferior iliac spine (AIIS); the hole is
then palpated with a ball-tipped feeler to confirm that
there has been no breach of the cortical bone of the inner
or outer pelvic table. Alternatively, after establishing the
landmarks, the pedicle gearshift can be used to cannulate
the cancellous bone of the iliac pathway, either freehand
or with fluoroscopic guidance. Gelfoam should be
inserted into the drill holes to control cancellous bone
bleeding. After the proper length unit rod is selected,
the pelvic limbs of the rod are crossed and inserted into
their respective drill holes. Each limb should be advanced
alternately in 1cm increments with an impactor. Care
must be taken to maintain control of the rod and ensure
that it does not penetrate either table of the pelvis. In the
setting of hyperlordosis the marked anterior inclination of
the pelvis increases the risk of the pelvic limb perforating
the inner cortex during insertion. The pelvic ends of the
rod need to be directed in a more posterior direction to
accommodate this angulation; rod placement is facilitated
by manual correction of the lordosis prior to rod insertion.
In instances of marked lordosis, the pelvic limbs of the rod
may be cut and inserted separately and then attached to
the rod with rod-to-rod connectors.
Figure 1A
Figure 1C
Figure 1B
In our large series of surgical correction of
scoliosis patients with cerebral palsy using unit rod
instrumentation, we reported an average coronal curve
correction of 68% and pelvic obliquity correction of 71%
with a very cost-effective implant system (see Figure 1).
Intraoperative complications with pelvic fixation occurred
in some patients with sagittal plane deformities; lumbar
hyperlordosis was a risk factor. Late postoperative
complications occurred in 12 patients: 3 pseudarthroses, 3
deep infections and 6 prominent proximal implant issues.
Although the unit rod provides excellent correction of
pelvic obliquity and resistance to flexion moments, there
is little resistance to axial pullout and torsion by virtue of
the rods being smooth and immediate micromotion of
the iliac construct after insertion. These concerns can
be mitigated by adding a transverse connector distally
to improve torsional rigidity and adding lumbar pedicle
screws distally at L5 to significantly improve axial pullout,
strength and stiffness.14
Figure 1D
Figure 1. A,B) Preoperative sitting X-rays of a patient with severe thoracolumbar scoliosis and pelvic obliquity. C,D) Postoperative sitting X-rays of the
patient after posterior spinal fusion with the unit rod.
7
2. Iliac Screws
The difficult learning curve associated with rod contouring
and insertion into the ilium using the Galveston technique
has been resolved with the use of iliac screws, which
permits screws of variable length and diameter to be
inserted into each ilium separately. Simpler fixation to
the pelvis with screws may also mitigate the potential
complications of the Galveston technique, which was
up to 62% in one series and 47% in Lonstein’s series.12
The screws can then be connected to the main construct
by various connectors. This technique has simplified the
process of obtaining iliac fixation, especially in patients
with significant pelvic asymmetry or hyperlordotic lumbar
deformities while improving pullout strength through
better interdigitation of the threaded implant within the
iliac cortical and cancellous bone.15
When placing iliac screws in lieu of smooth Galveston
pelvic fixation, the starting point is similar, 1-2cm inferior
to posterior superior iliac spine. To decrease implant
prominence, a notch in the ilium can be used to bury
the head of the screw below the contour of the cortical
bone. A pedicle gearshift or drill is used to cannulate the
cancellous bone between the inner and outer tables of
the ilium, directed toward the anterior inferior iliac spine,
1cm above the sciatic notch. Careful attention should be
made not to pass the probe down the Sacroiliac (SI) joint,
taking the time to ensure that the pathway is between
the cortical walls of the ilium. The pathway is then
palpated with a ball tipped-probe and tapped to increase
purchase and size the diameter and length of the screw.
The exposure of the PSIS can be made through the same
incision, elevating the paraspinal lumbosacral musculature
and soft tissues, with a subperiosteal dissection of the PSIS
and a portion of the outer table or through a separate,
small oblique fascial incision by retracting the skin over
the iliac crest. Alternatively, using a technique described
by Wang and colleagues16 with the VIPER® Screw System,
iliac screws can be inserted with a fluoroscopicallyguided percutaneous technique to avoid the morbidity
and complications associated with such a large muscle
dissection and soft tissue devitalization.
Every effort should be made to insert the largest diameter,
longest screws that can be safely accommodated to
achieve favorable biomechanics. Screws should extend
past McCord’s sacropelvic pivot point at the anterior edge
of the L5/S1 disk at a minimum.3 The further anteriorly
and laterally the screws extend, the better control of
pelvic flexion, extension, obliquity and rotation. Even in
smaller patients, we have been successful in implanting
screws of 7.5-8.0mm in diameter and 65-80mm in length.
Larger patients can accommodate screws of up to 10mm
x 100mm. Large taps are available to enlarge the entry
8
site and size the screws by using insertional torque.
These various options are all available in the EXPEDIUM®
Neuromuscular System and the Sacropelvic Collection in
closed and open screw (with the Top Notch® feature)
head options along with offset connections that will allow
modular, rigid connections to the longitudinal rods up to
the spinal implants. Since the typical iliac pathway starting
at the PSIS is 1-2cm lateral to the pedicular line, offset
connectors are frequently needed. Options in the set are
fixed or variable axis, open (with the Top Notch feature)
or closed connectors (see Figure 2) that can be customized
for length or short-throw rod contour and attached to the
rod with slip strengths equal to EXPEDIUM pedicle screw
connections (see Figure 3). Alternatively, the longitudinal
rods can be contoured laterally in the coronal plane distal
to the L5 or S1 screws to connect directly to the iliac
screws either prior to rod insertion, or with in situ benders.
A retrospective, single-center cohort study comparing
two groups of 20 patients (flaccid and spastic paralytic
scoliosis) each with Luque-Galveston constructs and iliac
screws showed similar maintenance of pelvic obliquity and
scoliosis correction, but the iliac screw technique avoids
the complex lumbosacral 3-dimensional rod bends and
had less haloing around the pelvic implants with minimal
implant complications. The Galveston group had 4 broken
rods and two reoperations and the iliac screw group had 1
broken screw and no reoperations.
A larger, multicenter retrospective study of 157 patients
with virtually equal distribution compared the unit rod
to rods bent by surgeons intraoperatively with iliac screw
fixation and found that the unit rod had better correction
of pelvic obliquity. However, the drawbacks included
longer mean surgical time, more blood loss, longer
hospital stay, higher infection rate and more proximal
fixation problems cited in the unit rod patients. Specific
mention was also made of the challenges of unit rod
implantation in a hyperlordotic patient.
Figure 2. Various open and closed offset connections are useful to make the lateral
translation from the longitudinal rods to the traditional iliac screw pathway.
Figure 3A
Figure 3B
Figure 3C
Figure 3D
Figure 3. A,B) Preoperative sitting X-rays of a patient with severe thoracolumbar scoliosis and pelvic obliquity. C,D) Postoperative sitting X-rays of the
patient after posterior spinal fusion with precontoured rods and proximal transverse connector from the Expedium Neuromuscular System and iliac
screws with offset connectors from the EXPEDIUM Sacropelvic Collection.
9
3. Double Iliac Screws
4. S2 Alar Iliac (S2AI) Screws for Pelvic Fixation
Occasionally, double iliac screw fixation is needed to
overcome challenges of osteoporosis, fractures involving
the sacropelvis or tumor resections/reconstructions to
impart better biomechanical stability. An overview of the
technique and experience in patients with neuromuscular
scoliosis is available in the literature. Two screws may
be placed on each side in the ilium (a total of four in the
pelvis), connected by a rod and then subsequently to
the longitudinal members with an offset connector. This
technique seems to decrease complications such as rod
disengagement, distal screw failure and implant failure
cephalad to the pelvis. The technique involves a muscle
splitting approach to the standard PSIS starting point,
exposure of the outer table of the pelvis to the sciatic
notch to identify the trajectory and cannulation of the
cancellous bone between the inner and outer tables,
aiming for the AIIS. The first screw is placed approximately
2cm superior to the PSIS with a second screw placed 2cm
further superiorly.
The S2AI technique, described by Kebaish and
Sponseller,17-20 has become my preferred technique for
various reasons. Since the starting point is 1.5cm deeper
than the traditional iliac entry from the PSIS, decreased
implant prominence is the main advantage. The starting
point and consequently, the screw head, is in line with L5
and S1 pedicle screws, thus avoiding an offset connection
from the longitudinal rod (see Figure 4); this point is
easily found, requires little muscle dissection and can
even be performed in a minimally invasive fashion16,18
(see Figure 5). The starting point for the screw is 2-4mm
lateral and 4-8mm inferior to the S1 foramen; minimal
muscle stripping and dissection is needed (see Figure
6). A sharp awl or burr is used to mark the starting
point and penetrate the cortical bone. Then a drill or
pedicle gearshift is used to enter the cancellous bone
of the sacrum directed towards the dorsal aspect of
the sacro-iliac joint, into the ilium. I find palpation of a
point just above the ipsilateral greater trochanter of the
proximal femur as a valuable virtual target (see Figure
7). The trajectory is lateral (approximately 40 degrees
to the horizontal plane) and 20-30 degrees caudal (this
depends on pelvic tilt) and fluoroscopy is helpful to
guide this trajectory. The AP projection shows the pelvis
and sciatic notch. The teardrop view helps ensure that
the pathway is in the thickest part of the ilium without
a cortical breach (see Figure 8). Once in the ilium, the
pathway is 1-2cm above the sciatic notch directed towards
the anterior-inferior iliac spine (see Figure 9). A polyaxial
screw, typically 80-100mm long, is used and adults can
accommodate diameters of 8-10mm; screw diameters of
7-8mm and lengths of 65-80mm can be used in smaller
patients (see Figure 10).
Figure 4A
Figure 4B
Figure 4. A) The S2AI pathway allows the iliac screw to line up with the lumbar and sacral pedicle screws so no offset connection is needed when
securing the longitudinal rods. B) Intraoperative photo of the screw arrangement and alignment. The iliac screw is noted by the arrowhead.
10
Figure 5A
Figure 5B
Figure 5. A,B) Intraoperative photos illustrating use of a cannulated
gearshift from the VIPER SAI Screw Set to cannulate the S2AI pathway
under fluoroscopy and then use of a guidewire to tap and implant the
SAI screw. Minimal muscle dissection is required and the technique is
very efficient. C) illustration of guidewire placement in the S2AI pathway.
Figure 5C
Figure 6. The starting point of the SAI screw is a point between the S1
and S2 foramen, along the lateral border, inline with an S1 pedicle screw.
Figure 7. From the starting point, one should aim for the AIIS, palpating
the tip of the ipsilateral greater trochanter.
11
Figure 8A
Figure 8B
Figure 8C
Figure 8. A-C) Fluoroscopy can be helpful in confirming the proper trajectory. The C-arm should be
angled 20-30 degrees caudal (depending on the degree of pelvic tilt) and 40-50 degrees rotated in
the horizontal plane to visualize the pelvic pathway, the iliac teardrop and direction of the implant.
Figure 9. The trajectory should be 10-20mm superior
to the sciatic notch aiming towards the AIIS. The iliac
teardrop is formed by the medial iliac wall, lateral
iliac wall and sciatic notch inferiorly. If the implant is
contained in this teardrop view and confirmed, cortical
breach is unlikely.
Figure 10A
Figure 10B
Figure 10. A-C) Axial cross-section, posterior and lateral iliac views of the screw pathway in the ilium.
12
Figure 10C
The new VIPER SAI screw has novel features that are
desirable for use in this capacity: a favored angled
polyaxial head for added tilt and ease of rod attachment, a
smooth shank for enhanced outer diameter and strength
at the proximal part of the screw and minimal disruption
of the SI joint, large threadform for cancellous bone
purchase and a diverse variety of sizes for all types of
patients (see Figure 11). Kebaish and colleagues reported
on a group of 52 patients (adult and pediatric) followed
over two years using this technique and reported lower
complication rates than other techniques with no adverse
effect on the SI joint and only one patient who required
implant removal.20 Excellent results were described in
a cohort study of smaller patients (predominantly with
scoliosis secondary to cerebral palsy) with the S2AI
technique compared to screws inserted into the ilium
from the traditional starting point of the PSIS. Patients
with S2AI screws had better restoration of pelvic obliquity
and fewer complications; no deep infections, prominent
implants or anchor migration compared to 3 patients with
infections and 3 instances of implant prominence, skin
breakdown or anchor migration in the PSIS group (see
Figure 12).
The S2AI screw technique can also be performed
minimally invasively using existing EXPEDIUM instruments
and its cannulated components. This technique saves
operative and fluoroscopy time after placement of
the guidewire down the pathway with either a drill or
cannulated gearshift.
Figure 11. The VIPER SAI screw features a favored angle articulation
and smooth, cannulated shank.
Figure 12A
Figure 12B
Figure 12. A-B) Postoperative sitting X-rays of a patient with
neuromuscular scoliosis after posterior fusion and instrumentation
with precontoured rods and a proximal transverse connector from
the Expedium Neuromuscular System and pelvic fixation using the
VIPER SAI screw via the S2AI pathway.
Complications and Avoidance
Complications associated with sacropelvic fixation include
injuries to the adjacent structures (bladder, colon, iliac or
gluteal vessels) due to misplacement, cortical bone breach,
implant prominence and loosening, wound problems,
infections and implant failure and nonunion.
S1 pedicle screws placed in a convergent manner may
injure the middle sacral artery or veins. Diverging S1
screws or alar screws that are placed bicortically and are
too long may cause injury to the common or internal iliac
artery/vein or lumbosacral trunk (L5 nerve root). S2 screws
that are too long and placed in a convergent trajectory
may injure the inferior hypogastric plexus or colon. Iliac
screws violating the sciatic notch may injure the superior
gluteal artery and those that violate the medial ilium
may injure the internal iliac artery/vein or lumbar plexus,
causing an iliacus hematoma or violation of the hollow
viscera of the pelvis. Iliac screws that are too long
anteriorly may violate the acetabulum. Misplacement can
be avoided by becoming familiar with the anatomy at risk,
cadaver experience, navigation, the use of blunt probes,
confirmation of bony end point at all steps of screw
placement and knowledge about the fluoroscopic X-ray
views to confirm proper screw placement.21-24
13
Implant (screw head) prominence is cited as a frequent
cause of removal as in one series of 36 patients with
iliac screw fixation of whom eight required subsequent
removal,25 but this can be mitigated by resecting the top
of the iliac crest, forming a notch for the screw head to
lie deeper within the ilium or using the S2AI technique,
which allows the screw head to be almost 15mm deeper
than traditional iliac screws.17 Kebaish’s series of 52 adult
patients with S2AI fixation listed only one patient that
needed subsequent removal.20
The infection rate seems lower with this technique as well,
perhaps since less dissection is needed and, combined
with soft tissue preservation and lower profile, seems
ideally suited even for our malnourished neuromuscular
patients. In a report of 32 patients with cerebral palsy
and scoliosis and two-year follow up from surgery, there
were no infections, prominent implants or failures of
the S2AI screws. The infection rate associated with iliac
fixation alone is difficult to compare from the literature
because iliac fixation is performed in conjunction with
other procedures. However, in a thorough two-year
follow-up study of patients treated with iliac fixation, a
group reported an infection rate of 4% (3 of 81 patients).
Those authors also noted iliac screw back-out in three
patients, all of whom had spondylolisthesis; however, no
pseudarthrosis occurred in these patients.
Nonunion and implant failure typically occur together,
and host biology as well as strength of fixation may play a
role. Proper workup to rule out a pseudarthrosis must be
undertaken prior to implant removal to allow for proper
preparation in the operating room. Again, consideration
of anterior interbody support is important, especially in
the face of pseudarthrosis. At L5/S1, an anterior structural
graft is biomechanically favorable since it is loaded in
compression and anterior to the pivot point; this can
optimize fusion and stability.8
14
Summary
Many sacropelvic fixation techniques have been described
historically to aid surgeons in long constructs to the
sacrum, but only a few are still widely used, including
sacral screws, the Galveston technique, iliac screws, and
S2AI screws. S1 pedicle screws are most effective when
placed with “tricortical” fixation, angled medially and
upward towards the sacral promontory. Sacral ala and S2
pedicle screws have been used to improve the strength
of the construct, but studies have shown no significant
biomechanical alteration or improved clinical results. Iliac
screws (bolts) have been shown to protect S1 screws
and have superior pullout strength as well as high fusion
rates. Offset connectors are used to connect the screws
to the longitudinal rods with the traditional PSIS starting
point. S2 alar iliac (S2AI) screws have the advantages of
decreased implant prominence since the insertion point
is up to 15mm lower than the prominent part of the
posterior iliac crest; and, the use of an offset connector is
avoided since the point of insertion is in line with the S1
screw and longitudinal rods. Some of these techniques are
associated with complications, which can be minimized by
paying close attention to the regional anatomy, minimizing
soft tissue dissection, and choosing low-profile implants
and techniques. To add anterior structural support,
anterior interbody fusions or TLIFs should be considered in
long fusions that extend to the thoracic spine, which will
offload some of the stresses from the posterior implants
and allow early bony union.
References
1. Devlin VJ, Boachie-Adjei O, Bradford DS, et al. Treatment
of adult spinal deformity with fusion to the sacrum using
CD instrumentation. J Spinal Disord 1991;4:1–14.
14. Erickson MA, Oliver T, Baldini T, et al. Biomechanical
assessment of conventional unit rod fixation versus a
unit rod pedicle screw construct: a human cadaver study.
Spine (Phila Pa 1976) 2004; 29: 1314-9.
2. Balderston RA, Winter RB, Moe JH, et al. Fusion to the
sacrum for nonparalytic scoliosis in the adult. Spine (Phila
Pa 1976) 1986; 11: 824-9.
15. Schwend RM, Sluyters R,Najdzionek J. The pylon concept
of pelvic anchorage for spinal instrumentation in the
human cadaver. Spine (Phila Pa 1976) 2003; 28: 542-7.
3. McCord DH, Cunningham BW, Shono Y, et al.
Biomechanical analysis of lumbosacral fixation. Spine
(Phila Pa 1976) 1992;17:S235-43.
16. Wang MY, Ludwig SC, Anderson DG, et al. Percutaneous
iliac screw placement: description of a new minimally
invasive technique. Neurosurg Focus 2008; 25: E17.
4. O’Brien MF. Sacropelvic fixation in spinal deformity. In:
DeWald RL, editor. Spinal deformities: the comprehensive
text. New York: Thieme; 2003. p 601-14.
17. Chang TL, Sponseller PD, Kebaish KM, et al. Low profile
pelvic fixation: anatomic parameters for sacral alar-iliac
fixation versus traditional iliac fixation. Spine (Phila Pa
1976) 2009; 34: 436-40.
5. Lebwohl NH, Cunningham BW, Dmitriev A, et al.
Biomechanical comparison of lumbosacral fixation
techniques in a calf spine model. Spine (Phila Pa 1976)
2002;27:2312-20.
18. Kebaish KM. Sacropelvic fixation: techniques and
complications. Spine (Phila Pa 1976) 2010: 35: 2245.
6. Cunningham BW, Lewis SJ, Long J, et al. Bio-mechanical
evaluation of lumbosacral reconstruction techniques for
spondylolisthesis: an in vitro porcine model. Spine (Phila
Pa 1976) 2002;27:2321-7.
7. Ogilvie JW, Schendel M. Comparison of lumbosacral
fixation devices. Clin Orthop Relat Res 1986;203:120-5.
8. Crock HV. Anterior lumbar interbody fusion: indications
for its use and notes on surgical technique. Clin Orthop
Relat Res 1982;165:157-63.
9. Allen BL Jr, Ferguson RL. The Galveston technique of
pelvic fixation with L-rod instrumentation of the spine.
Spine (Phila Pa 1976) 1984; 9: 388-94.
10. Allen BL Jr, Ferguson RL. L rod instrumentation for
scoliosis in cerebral palsy. J Pediatr Orthop 1982;2:87–96.
11. Broom MJ, Banta JV, Renshaw TS. Spinal fusion
augmented by Luque-rod segmental instrumentation
for neuromuscular scoliosis. J Bone Joint Surg Am
1989;71:32–44.
12. Lonstein JE, Koop SE, Novachek TF, et al. Results and
complications after spinal fusion for neuromuscular
scoliosis in cerebral palsy and static encephalopathy using
Luque-Galveston instrumentation. Spine (Phila Pa 1976)
2012; 37: 583-91.
13. Bell DF, Moseley CF, Koreska J. Unit rod segmental spinal
instrumentation in the management of patients with
progressive neuromuscular spinal deformity. Spine (Phila
Pa 1976) 1989;14:1301.
19. O’Brien JR, Matteini L, Yu WD, et al. Feasibility of
minimally invasive sacropelvic fixation: percutaneous S2
alar iliac fixation. Spine (Phila Pa 1976) 2010; 35: 460-4.
20. Kebaish KM, Pull ter Gunne AF, Mohamed AS, et al. A
new low profile sacropelvic fixation using S2 alar iliac
(S2AI) screws in adult deformity fusion to the sacrum:
a prospective study with minimum 2-year follow-up.
Presented at: the North American Spine Society Annual
Meeting; November 10–14, 2009; San Francisco, CA.
21. Lehman RA Jr, Kuklo TR, Belmont PJ, et al. Advantage of
pedicle screw fixation directed into the apex of the sacral
promontory over bicortical fixation: a biomechanical
analysis. Spine (Phila Pa 1976) 2002; 27: 806-11.
22. Zindrick MR, Wiltse LL,Widell EH, et al. A biomechanical
study of intrapeduncular screw fixation in the
lumbosacral spine. Clin Orthop Relat Res 1986; 203: 99
–112.
23. Mirkovic S, Abitbol JJ, Steinman J, et al. Anatomic
consideration for sacral screw placement. Spine (Phila Pa
1976) 1991; 16: S289-94.
24. Leong JC, Lu WW, Zheng Y, et al. Comparison of
the strengths of lumbo-sacral fixation achieved with
techniques using one and two triangulated sacral screws.
Spine (Phila Pa 1976) 1998; 23: 2289 –94.
25. Emami A, Deviren V, Berven S, et al. Outcome and
complications of long fusions to the sacrum in adult spine
deformity: Luque-Galveston, combined iliac and sacral
screws, and sacral fixation. Spine (Phila Pa 1976) 2002;
27:776-86.
15
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-68-000 9/2012 JC
CA #8988A / DJ10057A
Unique Challenges with Scoliosis and Dislocated Hips
Randal R. Betz, MD
The majority of neuromuscular disorders that include dislocated
hips fall into one of three diagnoses, which include 1) spinal cord
injury; 2) myelomeningocele; and 3) cerebral palsy. The highest
prevalence of dislocation is probably caused by sepsis, and it
is essential that any relatively acute, radiographically identified
dislocated hip have an aspiration and adequate work-up to rule
out sepsis. The prevalence for non-septic dislocated hips in these
populations is as follows:
In my experience with patients with SCI, there is about a 40%
prevalence of hip instability. The prevalence increases with the
length of time they are followed post injury. In patients with
spina bifida, the prevalence varies according to the level of
motor function. In cerebral palsy, hip instability almost always
occurs in the severely involved patients (Gross Motor Function
Classification System [GMFCS] V).
17
With the high prevalence of hip subluxation and
dislocation in neuromuscular disorders, patients fall
into one of three treatment groups with regards to the
challenges of scoliosis and dislocated hips, to include:
Scenario #1: Development of a scoliosis needing
treatment with an early subluxed hip secondary to pelvic
obliquity where the hip itself does not need treatment.
This assumes that the scoliosis and pelvic obliquity are
corrected prior to developing an actual unstable hip.
Scenario #2: There is scoliosis needing treatment and the
subluxed hip is now unstable and both the hip and spine
need treatment.
Scenario #3: There is severe scoliosis needing treatment
and a longstanding dislocated hip or hips.
18
Treatment
Scenario #1:
Development of a scoliosis needing treatment with an
early subluxed hip secondary to pelvic obliquity where
the hip does not need treatment, provided the scoliosis
and pelvic obliquity are corrected prior to developing
an actual unstable hip. Most commonly, this occurs in
a patient whose hips have a migration index < 40%.
The hip at risk is always on the high side of the pelvic
obliquity. Generally, complete correction via fusion and
instrumentation into the pelvis to correct the pelvic
obliquity helps to correct the early subluxation of the hip.
Sometimes what were thought to be mild or moderate
hip contractures preoperatively now cause problems
with balance postoperatively and may require soft tissue
releases (Figure 1A-G). Occasionally, one sees a more mild
scoliosis that potentially could be braced, but the obliquity
is being caused not by a severe scoliosis but by a pseudo
hip flexion contracture (abduction contracture of the
tensor fascia lata muscle and the gluteus medius) of the
opposite hip from the one that is showing subluxation.
Pseudo hip flexion contracture can be identified by taking
a hip that clearly shows a hip flexion contracture > 45
degrees and then abducting the hip maximally such
that the hip flexion contracture almost disappears. This
demonstrates tightness of the abductors and the tensor
fascia femoris muscles and causes a pelvic obliquity in
both sitting and supine positions. It needs to be corrected
so the scoliosis can be braced. The release of the hip
muscle contractures will prevent further subluxation of the
contralateral hip. The author of this chapter recommends
a modified Soutter release.1 This is a standard anterolateral
approach to the hip as one would do for osteotomies
such as a Salter. The apophysis over the iliac wing is split
anteriorly from approximately the mid axillary line down
through the anterior inferior iliac spine. The apophysis
and gluteus muscles are stripped from the outer table of
the ilium posteriorly enough that the muscle alignment
is straight off the midline of the torso (no future flexion
vector pull of the muscles). The iliac crest then is trimmed
such that reattachment to bone cannot cause recurrence
of the flexion contracture. This muscle release corrects the
pelvic obliquity. The mild scoliosis is now braceable, and
the risk of the opposite subluxed hip progressing to full
subluxation or dislocation is reduced.
Figure 1A (AP) and 1B (lateral). This is an 18-year-old patient with a T1
paraplegia secondary to a motor vehicle accident. He has no trunk control
and needs to utilize his upper extremities for balance. Therefore, the
indications for surgery are prevention of further curve progression and
restoring independent balance for bilateral upper extremity use.
Figure 1C. Demonstrates pelvic
obliquity with C7 falling far lateral
to the center sacral line.
Figure 1D. Shows an intraoperative
radiograph with one rod in
place, demonstrating incomplete
correction of the pelvic obliquity.
Further in situ rod bending and
compression distraction forces are
utilized, resulting in an aligned
spine to the pelvis in Figure 1E.
Figure 1E. Despite good correction
of C7 to the center sacral line and
balanced shoulders to the top
of the pelvis, the patient is still
sitting with a tilt to the left. Note
the adducted position of the right
hip, which is a fixed adduction
contracture.
Figure 1F (final AP) and 1G (final lateral). The patient underwent a right
adductor tenotomy which resulted in correction of the sitting imbalance.
19
Scenario #2:
There is scoliosis needing surgical treatment, and the
subluxed hip is now unstable and also needs treatment.
The overall principle here is to correct the spinal deformity
and pelvic obliquity completely prior to the surgery on
the hip. The first reason is that without correcting the
pelvic obliquity first, the hip will not stay in the socket
for long because of the persistent adducted position.
In this author’s experience, if the spine is not corrected
immediately after the hip is operated on, the hip usually
resubluxes or dislocates.
The second reason for doing the spine first is that
there is very low prevalence of deep venous thrombosis
(DVT) or pulmonary embolus after spine surgery in the
neuromuscular population. In contrast, surgery on the hip
has a high prevalence of DVTs, even with prophylaxis. If
prophylaxis is not continued for three to six months after
the hip surgery, there is risk of developing a late onset
DVT. This makes it extremely difficult to operate on the
spine because of the perioperative risk created by DVT
prophylaxis and/or treatment, especially if a DVT does
occur.
Generally, the hip is treated a few weeks after the spine
is corrected. Following spine surgery, the patient is put
on Lovenox. Doppler studies to rule out DVTs are done
at least twice. If the spinal wound is completely dry and
without any evidence of draining hematoma or infection,
then approximately 2-3 weeks later it is generally
appropriate to operate on the subluxed hip that needs
treatment. Stopping the Lovenox and restarting at the
time of the hip surgery is common practice.
20
Treatment of the Hip
Generally, the patients who are having their spine fused
are older and are therefore not candidates for just an
isolated varus osteotomy to treat the subluxed hip. If the
hip is mildly subluxed, then in our experience a Dega
osteotomy is the treatment of choice. Generally, Chiari
osteotomies are not adequate with regards to posterior
coverage, as most of these subluxed and dislocated hips
in these paralytic populations are going out posteriorly.
Procedures that do seem to be able to correct the
posterior subluxation include a reverse triple innominate
osteotomy or a Ganz osteotomy (Figure 2A-B), and a
posterior shelf osteotomy. In patients who are flaccid, it
appears that a muscle transfer is needed to help maintain
reduction of the hip. The author currently recommends an
external oblique transfer. Whereas it may not be an active
transfer, it may act like a tenodesis.
In patients who have spastic spinal cord injury or cerebral
palsy, it is essential to control the spasticity first before
attempting to relocate the hip. Oral medication is tried
first; however, it may be necessary to insert a baclofen
pump. This can be coordinated prior to or during the
spinal fusion. If the baclofen pump is present before the
spinal surgery, make sure to have a repair kit on hand (see
Complications in NM Scoliosis Chapter for more details).
Complications unique to this scenario include
development of heterotopic ossification (HO) about the
operated hip. Prophylaxis is essential, to include Indocin
(25 mg three times a day). In addition, the use of Didronel
is recommended. Both should be continued for 3-6
months. Whereas medication is counterproductive with
regards to obtaining a fused spine, if the patient gets HO
about the hip, this is disastrous in that he or she then
cannot sit because of lack of hip mobility. If the patient
does get HO and it needs to be resected, then the only
treatment is radiation. The HO can be prevented most
times with medication. In the author’s experience, the
healing of the spine fusion in these patients is so robust
that nonunions have not been seen despite prophylaxis
for HO.
Figure 2A. Preoperative AP of hips showing
subluxation of the left hip in a 15 year-old skeletally
mature patient with thoracic level SCI.
Figure 2B. Postoperative film showing excellent
posterior coverage with a Ganz osteotomy (Case
courtesy of Dr. Harold Van Bosse).
21
Scenario #3:
There is severe scoliosis needing treatment and a
longstanding dislocated hip or hips. In this scenario,
generally the dislocated hips are left alone…the risk
of these patients ending up with stiff hips because of
fibrous tissue or HO is high. Therefore, relocation is rarely
recommended.
The majority of these patients do not ambulate.
Correction of the spine to include correction of the pelvic
obliquity is important. Intraoperative techniques to provide
them with good sitting balance include fixing the spine
in slight thoracolumbar kyphosis and reduced lumbar
lordosis based on emulating a sitting x-ray (see Chapter 8).
In these patients with dislocated hips, it is essential in the
immediate postoperative period prior to prolonged sitting
in a wheelchair that they get seating pressures measured.
22
Summary
In the presence of subluxed or dislocated hips, the patient
with neuromuscular scoliosis presents unique challenges.
A clear plan needs to be identified early on to address
both orthopaedic problems in a sequence that is safe.
As a general principle, if both the hip instability and
the spine deformity require surgery, the author strongly
recommends doing the spine first.
References
1. Smith H. Ankylosis and deformity. In Edmonson AS,
Crenshaw AH eds. Campbell’s Operative Orthopaedics.
6th ed. St. Louis: The CV Mosby Co., 1980:1153.
23
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-64-000 9/2012 JC
CA #8782 / DJ10053A
The Correction of Pelvic Obliquity
Using Cantilever Correction
Kirk Dabney, MD
Spinal deformity in neuromuscular disorders is common and
is frequently accompanied by pelvic obliquity. Severe pelvic
obliquity affects sitting and standing balance, can lead to
pressure sores, and can contribute to hip dislocation. In severe
cases, the patient may lose the ability to sit upright. Surgical
correction must bring the pelvis level to the sitting surface and
perpendicular to the longitudinal axis of the spine, and restore
pelvic sagittal alignment.
25
Patient Assessment
In addition to the medical history, other important
historical information should include asking the patient
and/or caretaker questions regarding sitting and standing
tolerance, history of pressure sores, and the presence or
absence of pain. Physical examination should assess: 1)
sitting and standing balance including whether the head
and torso are centered over the pelvis in both the frontal
and sagittal planes, 2) pelvic obliquity in both the frontal
and sagittal planes, 3) rotational alignment of the spine
and pelvis, and 4) stiffness of the spine and
pelvic obliquity.
The latter is best assessed
by performing the bending
test (Figure 1). As shown
in the illustration, sagittal
plane pelvic stiffness can
be assessed by placing
the patient supine and: 1)
flexing the hips above 90
degrees to see if anterior
Figure 1. Side-Bending Test. The
pelvic tilt corrects, and 2)
patient is bent over the examiner’s
allowing the pelvis and
thigh at the apex of the curve. If
limbs to hang over the
the patient’s curve reverses and
the pelvis levels out becoming
end of the exam table to
perpendicular to the trunk, the
see if posterior pelvic tilt
curve remains flexible enough to
corrects. An anterior release correct through posterior fusion
alone. If not, an anterior release
(anterior discectomy) in a
is performed.
stiff deformity should be
considered if the pelvis
does not correct to neutral in the frontal plane or to
within the normal anatomic pelvic tilt in the sagittal plane.
Anterior release is performed one to two levels above
and below the apex of the stiff scoliotic, hyperlordotic,or
hyperkyphotic curve to close down the convexity of the
deformity. Radiographic assessment should include AP
and lateral views, either sitting (if the patient can sit) or
standing (if the patient can stand independently). The
Cobb angle and pelvic obliquity (using the horizontal
method) are measured in the coronal plane. Bending
x-rays can also be helpful in assessing whether an anterior
release should be performed. Pelvic tilt is assessed on the
lateral radiograph.
26
The overall goal after surgical correction is to shift weight
bearing during seating to the posterior thigh mass.
Excessive anterior pelvic tilt and lumbar hyperlordosis
causes the patient to sit with the abdomen resting on
the anterior thighs and can often result in pain. On the
other hand, excessive posterior pelvic tilt and lumbar
hyperkyphosis causes the patient to sit with excessive
pressure on the sacrum which can lead to pressure sores.
Surgical Treatment and Instrumentation
Cantilever correction is a powerful method to correct
pelvic obliquity. It requires instrumentation that can
firmly anchor into the pelvis and can then be used as a
lever arm to swing the pelvis into a corrected position
that is perpendicular to the longitudinal axis of the
spine. Traditionally, the unit rod is ideal for this purpose;
however, it can be difficult to place the pelvic limbs
of the rod in cases of severe pelvic obliquity and/or
lumbar hyperlordosis ( > 70 degrees), because it must
be placed into the pelvis in one unit. The ideal system
provides a lever arm that is as powerful as the unit rod
and can allow for easier fixation to the pelvis through
modularity. Cantilever correction using the EXPEDIUM®
Neuromuscular System can accomplish this. The pelvis
is exposed by dissecting up over the sacroiliac joint onto
the lumbar muscle attachment on the inner table of the
pelvis. It is important not to dissect into the sacroiliac
joint subperiosteally, as significant bleeding can be a
problem. By dissecting over the joint itself, little bleeding
is encountered. The muscle is then sharply and bluntly
dissected up over the iliac crest apophysis. The overlying
fascia is then divided and the outer wing of the ilium is
then subperiosteally exposed from the posterior superior
iliac spine (PSIS) forward along the posterior one-third of
the pelvis and down to the sciatic notch. A guide which
hooks into the sciatic notch is then utilized to aim a drill
hole (or alternatively a pedicle probe can be utilized)
from the PSIS start point to just anterior and superior to
the sciatic notch. If a guide is not available, the pedicle
probe alone can be used and aimed just above the sciatic
notch by direct palpation or by using intraoperative
fluoroscopy. This is the region where the pelvis is most
dense for pelvic rod or screw fixation1. An intraoperative
AP and oblique fluoroscopic view (Figure 2) is taken to
confirm the trajectory of the drill or probe to make sure
there is no penetration of the inner or outer pelvic table.
Pelvic screw fixation of largest diameter possible (7, 7.75,
8, 9 and 10mm diameter are available) is placed in this
trajectory and should be of sufficient length to pass the
sciatic notch by at least 1cm (Figure 2). The author prefers
to use a closed polyaxial screw head to maximize the
rigidity of the final rod-pelvic screw construct. Typically,
pelvic screws alone are used; however, when additional
fixation is needed to improve the rigidity of pelvic fixation,
S1 screws are added. This is preferred over sacral screw
fixation alone, because pelvic screw fixation provides
a better lever arm to correct both pelvic obliquity and
sagittal plane pelvic deformity. Alternatively, pelvic screws
can be placed using the medial portal as described by
Chang et al2. Advocates for this method state that there
is less exposure time and less bleeding, and that the screw
head is less prominent. While we have not found bleeding
or exposure time to be less in our hands, the screw is less
prominent using this approach. We have obviated the
screw head being prominent at the PSIS start point by
notching the ilium at the entrance point with a rongeur
and countersinking the screw. The fixed lateral rodded
connectors (usually the 10 or 20mm) are then provisionally
connected to the pelvic screws.
Figure 2. Intraoperative AP and oblique views showing proper placement of pelvic screw. A) Note the AP View
shows the trajectory of the pedicle probe from the PSIS to just superior and adjacent to the sciatic notch and B) the
final screw position at least 1cm lateral to the notch. The oblique view shows C) the probe and D) the final screw
position between the inner and outer cortex just superior to the sciatic notch which appears as a “teardrop”.
Figure 2A
Figure 2B
Figure 2C
Figure 2D
27
After the iliac or sacral fixation is completed, sublaminar
wires are passed at every level of the spine to be fused. If
hyperlordosis ( > 60 degrees) in the lumbar spine exists,
then pedicle screws with reduction posts are used in the
lumbar spine (See Chapter 6 on Cerebral Palsy Scoliosis).
The EXPEDIUM® pre-contoured (sagittal plane) stainless
steel 6.35-inch rods are connected to the previously placed
iliac screws via the lateral rodded connectors (Figure
3A). The 6.35-inch diameter rods are used to provide
maximum rigidity. Critical to the correction is attaching
and securing each of the pre-contoured rods to the iliac
screws with the fixed lateral connectors so that each of
the rods is perfectly perpendicular to the horizontal axis
of the pelvis and that the sagittal contour of the rods
is aligned with the sacrum (Figure 3A). This is done by
adjusting the rod rotationally so that the line on the rod
is directly posterior and in line with the sacrum (Figure 3B
and 3D). The sagittal bend should be identical on each rod
and should also be aligned so that the contour matches
from proximal to distal. If these steps are not meticulously
followed, the pelvic obliquity will not be fully corrected
with the cantilever maneuver. Once this is done, the set
screws on both the iliac screws are tightened and torqued
down onto the rod. Next, a drop entry cross connector
is attached so as to lie in the lumbar region of the spine
to add further rigidity to the system and to prevent
rotation and shift of the rods in any direction. Similarly, a
proximal closed connector is also placed at the top of the
construct and tightened which secures the rigidity of the
rectangular construct (Figure 3A and 3C). The rods are not
cut until it is secured proximally. After the rod construct is
“built” from distal to proximal, cantilever correction can
Figure 3A
Figure 3. A-B) Shows AP profile of the prebent rods connected via the
lateral connectors to the iliac screws. Note that the connection should
make the rods perpendicular to the horizontal axis of the pelvis to
maximize cantilever correction. A distal cross-connector and proximal
closed connector make the construct rigid. The rods should match in
both length and contour with rotation matched before final tightening of
the connectors. C) Shows the lateral contour of the pre-bent rods in line
with the sagittal profile of the sacrum. D) There is a line on the posterior
contour of the rod to help line it up with the sacrum.
Figure 3B
Figure 3C
Figure 3D
Line
on rod
28
be started. This begins with the surgeon pushing the rod
down to the L5 lamina only and then tightening the wire
to the rod (a jet wire twister is helpful). The wires must
never be used to pull the rod to the spine. This is repeated
in a step-by-step manner: the rod is pushed down to the
L4 lamina and then the wire tightened. Next, the rod is
pushed down to the L3 lamina and the wire tightened.
The process is repeated one level at a time up to the T1
lamina (Figure 4). The wires are cut 1cm in length and
then bent and impacted down to the midline so as not to
remain prominent. Just prior to tightening the wires
at the proximal levels, the rod can be cut at the T1 level
and the closed connector brought up to the proximal end
of the rod and retightened and torqued, and then the
wires tightened.
Figure 4A
Alternatively, if one is using reduction pedicle screws in
the lumbar spine then the same push the rod to the spine
technique is used. This will help avoid screw pullout. If a
screw does begin to pull out, then add a sublaminar wire
at that same level and use in conjunction with the screw.
Figure 4. A) The rod “construct” is manually pushed down at each level with
a rod pusher at each level before tightening wires (if sublaminar wires are
to be used). This is important to prevent wire breakage or cutout through
the lamina. It is important to keep the spinous process centered between
the rods. If pedicle screws are being used, it is still important to push the
rod down at each level gradually so as to prevent plowing out of the screws.
Reduction posts can be helpful, however, rod pushers help so that all of
the reduction force is not placed on the screw alone. B) The rod is gradually
pushed to the spine at each level.
Figure 4B
29
Outcomes
The outcomes of cantilever correction in neuromuscular
scoliosis are documented in the literature primarily
through the experience of using the unit rod. Our
experience shows pelvic obliquity can be corrected by
approximately 70-80%. Intraoperative complications
with pelvic fixation are primarily due to the difficulty with
pelvic rod placement in patients with preoperative lumbar
hyperlordosis. This problem is obviated by placing pelvic
screws separately and then connecting the pre-contoured
rods, distal crosslink, and then proximal connector as
described earlier in this chapter.
It has been our experience that fusion to the pelvis does
not cause a loss of ambulatory function. Restoration of
sagittal alignment is key in preserving ambulatory function
which can be easily done through the use of the precontoured rods from the EXPEDIUM® Neuromuscular
System. Through the use of the pre-contoured rods, the
lumbar spine is reduced to the proper sagittal alignment
with either sublaminar wires or pedicle screws.
30
References
1. Miller F, Moseley C, Koreska J. Pelvic anatomy relative
to lumbosacral instrumentation. J Spinal Disord
1990;3:169-73.
2. Chang TL, Sponseller PD, Kebaish KM, Fishman EK.
Low profile pelvic fixation: anatomic parameters for
sacral alar-iliac fixation versus traditional iliac fixation.
Spine (Phila Pa 1976) 2009;34:436-440.
31
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-65-000 9/2012 JC
CA #8785 / DJ10059A
Correction of Pelvic Obliquity
with an All-Screw Construct
Suken A. Shah MD
Neuromuscular scoliosis may be associated with significant
pain due to sitting difficulties and truncal imbalance, especially
with pelvic obliquity and pressure sores; however, many of
the patients with static encephalopathy or TBI are unable to
articulate their symptoms. Progressive deformity is believed
to be the result of both muscle imbalance and anatomic
deformity. Curve progression leads to subsequent deformity
and trunk imbalance with associated loss of function.
33
Background
Historically, there has been debate regarding when to
extend the posterior spinal fusion to the pelvis. Pelvic
obliquity has been noted to progress in neuromuscular
scoliosis if the pelvis is not fused,1,2 and traditionally many
authors have recommended fusion to the pelvis in nonambulatory patients. In the ambulatory patient with
pelvic obliquity, fusion to the pelvis has been traditionally
avoided due to the belief that it will adversely affect
ambulatory function.3,4 Our experience demonstrated
preserved ambulatory function in ambulatory patients
with cerebral palsy who were fused with unit rod
instrumentation, as determined by gait analysis. We
hypothesized that the conventional assumption that
ambulatory potential is limited by fusion to the pelvis arose
from early attempts at pelvic fixation with Harrington rods
that removed lumbar lordosis, and possibly confounded
by variable of prolonged immobilization and bed rest in
earlier segmental systems that did not provide sufficient
rigidity for immediate postoperative mobilization and
ambulation therapy.
The Galveston technique to extend the fusion across
the pelvis by placing each Luque rod between the pelvic
tables4,7 has demonstrated acceptable fusion rates across
the L5-S1 segment, and appears to provide good control
of pelvic obliquity. While the impaction of two Luque
rods into the pelvis with associated segmental fusion via
sublaminar wires provides a reasonable construct in the
sagittal plane with resistance of flexion/extension, there
exists a moment arm of rotation about the two rods
allowing for rod translation with respect to one another,
loss of torsional control and subsequent progression
of pelvic obliquity, pseudarthrosis, and implant failure.6
The use of Luque rods smaller than 6.35mm diameter
may increase the incidence of implant failure2,5, but
the intraoperative bending of 6.35mm diameter steel
rods to the optimal 3-dimensional geometry for pelvic
implantation in a patient with a severe lumbopelvic
deformity presents a technical challenge. Immediate
and durable rigid fixation is essential for surgical success.
Sanders, et al2 found, in their retrospective study of Luque
rod instrumentation, that a postoperative curve greater
than 35 degrees, preoperative curves greater than 60
degrees, crankshaft deformity, and not fusing to the
pelvis were factors associated with postoperative
curve progression.
34
The unit rod developed by Bell, Mosley, et al5 addresses
some of the potential limitations of dual Luque rod
instrumentation. The implant design of a proximally
connected, precontoured rod provides for better rotational
control, as the degree of rotational freedom between two
independent Luque rods is eliminated. Initial mean curve
corrections with the unit rod were reported to be 54.6%,
with a mean loss of correction of 6.5% at 2 year followup5; experience at our institution documented a mean
scoliosis correction with the unit rod of 66% and 75%
correction of pelvic obliquity.
However, Luque-Galveston constructs and unit rods do
have significant complications related to distal fixation,
and treatment of significant sagittal plane deformities
are a challenge. Patients with lumbar hyperlordosis are
at risk for implant pullout, rod breakage, migration and
pseudoarthrosis, and those with thoracic hyperkyphosis
are at risk for proximal wire cutout and breakage, proximal
junctional kyphosis and instrumentation prominence.
Subsequently, many of these patients require reoperations,
which are costly and not without further risk.8
Lonstein and his colleagues recently published their
results and complications of 93 patients with cerebral
palsy and scoliosis who underwent PSF with LuqueGalveston instrumentation with an average follow up of
3.8 years.7 Coronal curve correction was 50% and pelvic
obliquity correction was 40% at latest follow up. The late
complication rate was 47% and included the windshield
washer sign, junctional kyphosis, pseudarthrosis (7.5%)
and implant problems including breakage, dislodgement
and prominence. Seven patients required reoperation,
most commonly for pseudarthrosis and/or failed implants.
Long iliac anchors provide superior control of pelvic
obliquity, while necessitating more dissection of the outer
table musculature for insertion. The standardized shape
of the unit rod produces a relatively uniform coronal and
sagittal profile. However, there are limitations to its use:
a) it is particularly challenging to implant in a hyperlordotic
patient, b) it does not lend itself to use of hooks or
screws as originally designed, c) estimation of correct
implant length is difficult in patients with large curves, d)
dissection at the cranial end of the spine to pass wires may
produce a tendency for junctional kyphosis and finally, e)
the unit rod does not perform well in curves with an apex
in the high thoracic spine.
Advantages of Screw Fixation
Iliac screws perform better in pull out strength than
smooth Galveston rods for pelvic fixation, and the use
of segmental pedicle screw constructs have shown
substantial correction and high fusion rates, while
accomplishing the goals of leveling pelvic obliquity and
addressing seating problems. These newer, modular
systems can navigate some of the substantial challenges
in these patients, such as abnormal pelvic anatomy,
osteoporotic bone and hyperlordosis and avoid some of
the risk in early instrumentation failure, but come at a
substantial monetary expense. Our experience shows
that in cases of marked lordotic deformity, better fixation
and correction was achieved with pedicle screws and
frequently reduction-tab pedicle screws in the lordotic
segment. A cadaveric biomechanical study has confirmed
that the use of L5 pedicle screws significantly increases
the lateral and oblique stiffness of the unit rod construct.9
McCall et al10 retrospectively examined a cohort of
patients with neuromuscular scoliosis in whom those
with a stable lumbosacral articulation were instrumented
with a “U-rod” (unit-rod without the pelvic limbs) with L5
pedicle screw fixation. The L5-S1 interspace mobility was
assessed on the basis of L5 tilt; patients with more than 15
degrees of L5 tilt were instrumented with a standard unit
rod construct. McCall et al found that in follow-up the
patients that were instrumented to L5 with the U-rod had
similar results to those fused with the standard unit rod
construct. However, it has been our institution’s practice
to fuse all patients with spastic conditions to the pelvis to
avoid late pelvic obliquity below an instrumented fusion.
The difficult learning curve associated with rod contouring
and insertion into the ilium using the Galveston technique
has been resolved with the use of iliac screws, which
permits screws of variable length and diameter to be
inserted into each ilium separately. Simpler fixation to
the pelvis with screws may also mitigate the potential
complications of the Galveston technique, which was up
to 62% in Gau’s series11 and 47% in Lonstein’s series.7
The screws can then be connected to the main construct
by various connectors. This technique has simplified the
process of obtaining iliac fixation, especially in patients
with significant pelvic asymmetry or hyperlordotic lumbar
deformities while improving pullout strength through
better interdigitation of the threaded implant within the
iliac cortical and cancellous bone.12,13
A retrospective, single-center cohort study comparing
two groups of 20 patients (flaccid and spastic paralytic
scoliosis) each with Luque-Galveston constructs and iliac
screws showed similar maintenance of pelvic obliquity and
scoliosis correction, but the iliac screw technique avoids
the complex lumbosacral 3-dimensional rod bends and
had less haloing around the pelvic implants with minimal
implant complications. The Galveston group had 4 broken
rods and two reoperations and the iliac screw group had 1
broken screw and no reoperations.14
We reported in a larger, multicenter retrospective study
of 157 patients, with virtually equal distribution, that
compared the unit rod to “surgeon-bent” rods with
iliac screw fixation and found that the unit rod group
had better pelvic obliquity correction. However, the
disadvantages seen in cases with the unit rod include
increased mean surgical time, blood loss, hospital stay,
infection rate and proximal fixation problems.8 The
increased blood loss may be a result of the multiple
laminotomies, wire passage, and wire manipulation,
although this is speculative. We hypothesized that the
increased incidence of infection after unit rod surgery may
be due to the dissection of the subcutaneous tissue and
muscle needed to make the transition from the iliac wings
to the spine.
Other concerns with these constructs were the lucencies
that surrounded the iliac rods and the rod bend, which
could become prominent with any migration. In addition,
dissection of the paraspinous muscle required for the unit
rod transition from the spine to the pelvis compromised
the muscle flaps covering the rods. For insertion of a
unit rod, subcutaneous flaps must be elevated bilaterally
to the iliac crests, and the paraspinous muscle over the
midline is incised or stretched to where the rod joins the
iliac posts at the entrance to the pelvis. These steps are
not needed for insertion of sacral or iliac screws. For iliac
screws, although the lateral dissection maybe the same for
a traditional PSIS starting point, a connector may be used
to join the screw to the rod with less tissue trauma and
minimally invasive techniques are available. Iliac screws
have decreased, but not eliminated, the rate of implant
lucency and the need for revision seen with unit rods.
The increased incidence of clinically significant proximal
implant prominence may be a result of the shape of the
rod, which has a dorsally-directed bend at the top, as well
as the supralaminar dissection needed to pass the wires
proximally.8,14,15
35
There are substantial data in the literature that
pedicle screws in the lumbar and thoracic spine, by
nature of their 3-dimensional correction control of
the vertebral body, allow surgeons to obtain better
correction of spinal deformities in all three planes and
subsequently very little loss of correction over time and
an insignificant pseudarthrosis rate. Pedicle screws,
combined with posterior-based releases, have decreased
the need and frequency of an anterior release for many
of the severely involved neuromuscular patients and
decreased the morbidity associated with combined
anterior/posterior surgery.
Combined with halo / femoral or cranial tong / boot
traction, screws are a powerful correction tool that
avoids the morbidity of an anterior procedure even for
large, stiff curves (Figure 1). Traction helps tremendously
in obtaining intraoperative balance of the patient’s
trunk and pelvis. Sometimes the obstacles of severe
hyperlordosis and/or hip and knee flexion contractures
with or without hip dislocation will challenge the
spinal deformity surgeon with positioning the patient
in traction, but various strategies are available to
ameliorate these problems. The patient’s hips can be
flexed to assist with correction of lumbar hyperlordosis,
unilateral leg traction may be used on the high side of
the pelvis to assist with pelvic obliquity correction, and
padding can be used to accommodate contractures.
For stiff curves, or those over 90 degrees, anterior release
of the apical levels of the curve may be indicated, since
it is necessary to gain flexibility to obtain adequate curve
correction, which then indirectly helps correction of some
of the pelvic obliquity. Anterior surgery increases the
complication rate and morbidity of spinal surgery in these
patients, and it is unclear whether to stage the anterior
and posterior procedures separately (one week apart)
or to do both procedures on the same day. Evidence
exists to support both strategies, and it is our practice
to stage surgeries for patients with severe involvement
and multiple medical co-morbidities and problems. For
relatively healthy patients, we usually perform both
stages on the same day, provided that the time under
anesthesia or blood loss is not too substantial after the
anterior release.7,8 Screw fixation with custom-bent
rods or pre-contoured rods (available in the Expedium®
Neuromuscular System) allow the surgeon to tailor
fixation and correction options for a preferred technique
and particular patient. Review of the literature available
indicates that pedicle screw fixation offers better major
curve correction, improved derotation, lower proximal
implant complications and may avoid the need for an
anterior release. Accuracy and neurological safety with
freehand pedicle screw placement is similar to that of
idiopathic scoliosis.
Figure 1. Intraoperative photo of a patient in cranial tong and distal boot traction, frequently used to manage
large curves with pelvic obliquity. Intraoperative traction is a powerful tool to obtain passive correction of large
deformities, but high quality TcMEP monitoring is important during the procedure.
36
Surgical Technique
After intubation, appropriate monitoring leads and
establishment of large bore IV access and arterial and
central venous catheterization, the patient should
be placed prone on a radiolucent table or four post
frame. Care should be taken to ensure that all bony
prominences are well padded to avoid skin breakdown,
especially in thin patients, and that the abdomen
hangs free. The hips can be allowed to gently flex with
knee and thigh support to passively correct lumbar
hyperlordosis. Intraoperative traction has been described
to correct pelvic obliquity.
A standard posterior exposure of the spine from T1
to the sacrum is performed. An aggressive posterior
release with radical facetectomy and ligamentum flavum
resection is important in creating flexibility in the rigid
apical portion of the curve and in all but the largest, stiff
curves makes a posterior-only approach sufficient for
correction of the scoliosis.
Figure 2A
Figure 2B
Figure 2. A,B) Preoperative sitting X-rays of a patient with spastic
quadriplegic cerebral palsy and severe thoracolumbar scoliosis and
pelvic obliquity. The patient had significant difficulties sitting, pain
and skin problems.
At the inferior margin of the incision, the outer wing of
the ilium can be subperiosteally exposed to the sciatic
notch in the open version of the technique. When placing
iliac screws in lieu of smooth Galveston pelvic fixation, the
starting point is similar: 1-2cm inferior to the posterior
superior iliac spine. To decrease implant prominence, a
notch in the ilium can be used to bury the head of the
screw below the contour of the cortical bone. A pedicle
gearshift or drill is used to cannulate the cancellous
bone between the inner and outer tables of the ilium,
directed toward the anterior inferior ilica spine. I now
prefer the S2AI screw technique for pelvic fixation and
that is described in Chapter 2 on Sacropelvic Fixation
Techniques. This technique allows a lower profile, favored
angle implant that lines up with L5 and S1 screws, thus
obviating the need for an offset connector. This preferred
technique for pelvic fixation along with an Expedium
Neuromuscular System modular rigid, closed transverse
connector for at least the proximal end (distal end
optional) results in a strong construct which allows control
of pelvic obliquity, low profile and, to date after 3-4 year
follow up, no implant failures (Figure 2). I have adopted a
less invasive technique over a guide wire using the screw’s
cannulated feature to preserve the soft tissues and limit
fluoroscopy time.
Figure 2C
Figure 2D
Figure 2. C,D) Postoperative sitting X-rays of the patient 2 years
after posterior spinal fusion with the Expedium Spine System and
components of the EXPEDIUM Sacropelvic Collection and Neuromuscular
Systems. Note the profound correction of pelvic obliquity and restoration
of coronal decompensation and sagittal sitting balance.
37
Due to the challenges of obtaining vertebral fixation
in non-ambulatory patients with severe, rigid
thoracolumbar curves, when using pedicle screw
fixation in the thoracolumbar spine, use of polyaxial
screws with reduction tabs will ease rod approximation
and provide stable fixation. Alternatively, new Viper®
Cortical Fix Screws are available, specifically designed
to increase pull-out strength in bone with a cortical,
short-pitch thread form for the pedicular bone and a
long, dual lead thread form for cancellous purchase.
Proper sizing of the screws should ensure good cortical
fill and length in the vertebral body. In osteopenic
bone, with careful technique, bicortical fixation through
the anterior vertebral body (except L4), upsizing the
diameter to achieve cortical pedicle fixation or other
techniques may improve bony purchase.
I prefer Expedium polyaxial screws with reduction
tabs on the apical concavity to assist with translation
and derotation of a severe curve; these screws are
especially effective in pulling the spine dorsally to a
prebent rod to correct lordosis (Figure 3). The reduction
tabs allow a slow, controlled force to be applied to the
spine and limits the risk of screw pullout; the polyaxial
head allows accommodation of severe deformity
and anatomic constraints of full correction. On the
convex apex, Expedium uniplanar screws are helpful
for derotation and cantilever forces to be applied for
3-dimensional correction.
38
Figure 3. Reduction tab
polyaxial pedicle screws
in the concave, lordotic
portion of the curve facilitate
correction of the spine to
the rod in a slow, controlled
manner without specialized
instruments. The tabs are
removed after final tightening
of the set screws.
The length of the rod is measured from T1/T2 to the
pelvis. Note that the correction of a kyphotic deformity
will shorten the spine and the correction of a lordotic
deformity will lengthen the spine. Special attention
should be paid to selection of the rod material (SS,
Ti or CoCr), diameter and strength of the rod; the
biomechanical properties of the rod and the force it
can impart to the spine are important considerations
when treating patients with osteopenic changes found
in the vertebrae of neuromuscular scoliotic spines. The
Expedium Neuromuscular System contains precontoured
rods to save time in the operating room. The proper rod
contour (hypokyphotic, normal or hyperkyphotic) and the
length (short, standard or long) is selected for implantation
and the rod lengths are tailored for the patient with a
provisional look at how the rod will lay proximally and
whether sufficient length is present for attachment to the
iliac screws. The contours are checked and customized.
Then, a specially designed modular rigid, closed transverse
connector from the Expedium Neuromuscular System is
selected for the proximal end and distal end (optional);
they are available in 4mm increments and need to be slid
over the top of the rods and secured prior to implantation.
With this step, a modular, precontoured construct is
quickly and easily made, and when connected to the iliac
screws and distal lumbar pedicle screws is transformed
into a powerful a cantilever correction tool to level the
pelvis and correct severe thoracolumbar scoliosis like the
unit rod (Figure 4 and Figure 5 - Clinical Example).
Figure 4. Once the lumbo-pelvic foundation has been established by
connecting the pre-bent rods to the iliac screws and lumbar pedicle
screws, the two rods are secured proximally by a closed, transverse
connector and the construct is now transformed into a powerful
cantilever correction tool, like the unit rod, to level the pelvis and correct
severe thoracolumbar scoliosis.
39
Facetectomies and decortication are performed and
preliminary grafting of the area under the rod is
performed. If the standard PSIS starting point for the iliac
screws was used, then offset connectors are needed, but if
the S2AI pathway is used, no offset connection is needed.
The iliac screws have the Top Notch® feature to facilitate
rod engagement with the Expedium Flex Clip Rod
Reducer, or the set screws can be inserted freehand. S1
and L5 screws are captured and reduced; now the distal
foundation is established and with the precontoured rods
linked at the top with the closed transverse connector,
superior control of the pelvis is obtained. The rods are
cantilevered to simultaneously level the pelvis and correct
the scoliosis and sagittal plane deformity. The screws
are sequentially captured from distal to proximal and
segmental fixation ensures that no single screw is prone
to pull out. As previously stated, reduction tab polyaxial
screws on the concave apex are important for a slow,
guided correction of the spine, and the rod is directed to
the spine and held in place while the screws are reduced
to the rod. Sufficient time for viscoelastic creep and
careful visualization of the bone-screw interface will avoid
pull out, but occasionally in situ bending may be needed;
coronal and sagittal benders are available in the set.
40
Pedicle screw fixation allows 3-dimensional control of the
vertebrae and facilitates coronal, sagittal and axial plane
correction of these complex deformities. The Expedium
Spine System vertebral body derotators, Quick Sticks,
and available racks to connect multiple derotators are
important adjuncts in the correction of the axial rotation.
Insertion of pedicle screws in this application is similar to
the standard technique and can be used in conjunction
with posterior-based osteotomies and vertebral column
resection. Compression/distraction with simultaneous
derotation optimizes correction before final set screw
torque-limited tightening. Hemostasis is confirmed,
decortication is performed and additional bone graft
with antibiotics is placed over the posterior elements with
special attention at the lumbosacral junction. A watertight
closure with absorbable suture is performed (without a
drain is our preference) and a sterile dressing with a water
tight barrier to prevent soiling is applied.
Correction of neuromuscular scoliosis with pedicle and
iliac screw fixation is typically 63-70% with leveling
of the pelvis and excellent sagittal alignment.8,15 With
contemporary techniques and cantilever correction of
the pelvis to neutral, it is unusual to have residual pelvic
obliquity greater than 10 degrees. The patient’s personal
wheelchair should be readjusted to accommodate his new
trunk proportions and pelvic alignment (Figure 5).
Figure 5A
Figure 5B
Figure 5. A,B) Preoperative sitting X-rays of a patient with
spastic quadriplegic cerebral palsy and thoracolumbar scoliosis
with pelvic obliquity. The patient had significant difficulties
sitting, pain and hyperlordosis.
Figure 5C
Figure 5D
Figure 5E
Figure 5F
Figure 5. E,F) Sitting X-rays of the patient demonstrating
maintenance of erect alignment and sitting posture.
Figure 5. C,D) Immediate postoperative X-rays showing
restoration coronal, sagittal and pelvic alignment with an all-screw
construct. Iliac screws were inserted via the S2AI pathway.
41
Summary
Correction of neuromuscular scoliosis and pelvic obliquity
with a screw construct offers many advantages including
improved pelvic fixation with a low incidence of failure,
infection, prominence or soft tissue complications,
especially when using the S2AI technique. Furthermore,
3-dimensional correction of the severe apical deformity
especially through derotation at the thoracolumbar
junction results in better truncal balance and chest wall
restoration important for pulmonary function and sitting.
Finally, proximal screw constructs with custom bent rods
result in less failure, pullout and prominence requiring
revision than the unit rod.
42
References
1. Bradford D. Neuromuscular spinal deformity. In:
Bradford D, Lonstein J, Ogilvie J et al (eds) Moe’s
textbook of scoliosis and other spinal deformities. WB
Saunders, Philadelphia: 1987; 271-305.
9. Erickson MA, Oliver T, Baldini T, et al. Biomechanical
assessment of conventional unit rod fixation versus
a unit rod pedicle screw construct: a human cadaver
study. Spine (Phila Pa 1976) 2004; 29: 1314-9.
2. Sanders JO, Evert M, Stanley EA et al. Mechanisms of
curve progression following sublaminar (Luque) spinal
instrumentation. Spine (Phila Pa 1976) 1992;17:781789.
10. McCall RE, Hayes B (2005) Long-term outcome in
neuromuscular scoliosis fused only to lumbar 5. Spine
(Phila Pa 1976) 2005;30: 2056-2060.
3. Banta JV, Drummond DS, Ferguson RL. The treatment
of neuromuscular scoliosis. Instr Course Lect 1999;48:
551-562.
4. Allen BL Jr, Ferguson RL. The Galveston technique
for L rod instrumentation of the scoliotic spine. Spine
(Phila Pa 1976) 1982;7:276-284.
5. Bell DF, Moseley CF, Koreska J. Unit rod segmental
spinal instrumentation in the management of patients
with progressive neuromuscular deformity. Spine
(Phila Pa 1976) 1989;23:2308-2317.
6. Broom MJ, Banta JV, Renshaw TS. Spinal fusion
augmented by Luque-rod segmental instrumentation
for neuromuscular scoliosis. J Bone Joint Surg Am
1989;71:32–44.
7. Lonstein JE, Koop SE, Novachek TF, et al. Results and
complications after spinal fusion for neuromuscular
scoliosis in cerebral palsy and static encephalopathy
using Luque-Galveston instrumentation. Spine (Phila
Pa 1976) 2012; 37: 583-91.
8. Sponseller PD, Shah SA, Abel MF, et al. Scoliosis
surgery in cerebral palsy: differences between unit
rod and custom rods. Spine (Phila Pa 1976) 2009; 34:
840-4.
11. Gau YL, Lonstein JE, Winter RB, et al. LuqueGalveston procedure for correction and stabilization
of neuromuscular scoliosis and pelvic obliquity: a
review of 68 patients. J Spinal Disord 1991; 4: 399410.
12. Kuklo TR, Bridwell KH, Lewis SJ, et al. Minimum
2-year analysis of sacro-pelvic fixation and L5–S1
fusion using S1 and iliac screws. Spine (Phila Pa 1976)
2001; 26: 1976-83.
13. Schwend RM, Sluyters R,Najdzionek J. The
pylon concept of pelvic anchorage for spinal
instrumentation in the human cadaver. Spine (Phila Pa
1976) 2003; 28: 542-7.
14. Peelle MW, Lenke LG, Bridwell KH, et al. Comparison
of pelvic fixation techniques in neuromuscular spinal
deformity correction: Galveston rod versus iliac and
lumbosacral screws. Spine (Phila Pa 1976) 2006; 31:
2392-2398.
15. Modi HN, Hong JY, Mehta SS, et al. Surgical
correction and fusion using posterior-only pedicle
screw construct for neuropathic scoliosis in patients
with cerebral palsy: a three year follow up study.
Spine (Phila Pa 1976) 2009; 34: 1167-75.
43
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-67-000 9/2012 JC
CA #8980A / DJ10056A
Scoliosis in Cerebral Palsy
Kirk Dabney, MD
Cerebral palsy (CP) is a heterogeneous disorder that is
caused by a static lesion to the immature motor cortex.
The prevalence of scoliosis in the cerebral palsy population
is proportional to the severity of neurologic impairment. It
ranges from 5% in spastic diplegia to 64-74% in spastic
quadriplegia. Progression often also occurs during adulthood
especially in lumbar and thoracolumbar curves > 40 degrees.
Curvatures > 60 degrees begin to affect sitting and standing
balance as well as head control. Pelvic obliquity is commonly
associated with scoliosis. Surgical management is indicated in
patients not tolerating seating (in non-ambulators) or standing
(in ambulators). Surgery may also be indicated to alleviate pain
or to correct pelvic obliquity prior to correcting a concomitant
hip dislocation. Sagittal plane deformities can occur in
combination with scoliosis or in isolation in the patient
with CP. Hyperlordosis and hyperkyphosis can occur with or
without scoliosis and in isolation are much less common than
scoliosis in the patient with CP. In our experience, sagittal
plane deformity alone can also cause seating problems and
pain, especially when > 70 degrees.
45
Problems with Luque / Galveston Instrumentation
The gold standard for surgical instrumentation in the
treatment of neuromuscular scoliosis in the past has been
Luque rod instrumentation (with Galveston extension to
the pelvis and cross-linkage [to prevent rod rotation and
shift] with sublaminar wiring). The unit rod incorporated
these concepts into one unit and serves as a very powerful
cantilever to correct scoliosis (especially lumbar and
thoracolumbar curves) and pelvic obliquity. Technical
difficulties with the unit rod include difficulty with pelvic
limb placement (especially with hyperlordosis), problems
judging final rod length proximally, inadequate correction
of severe rotational deformity, and having sufficient lever
to correct upper thoracic scoliotic and kyphotic curves.
Due to the modularity of the EXPEDIUM® Neuromuscular
System and the unique implant additions such as proximal
rod connectors, reduction screws and an assortment
of sacral iliac fixation options, these difficulties are
significantly lessened.
Patient Assessment
Patients with cerebral palsy often have associated
medical comorbidities which correlate strongly with
postoperative complications, including gastroesophageal
reflux, aspiration pneumonia, poor nutrition, the presence
of a seizure disorder, and low bone mineral density.
These should be identified preoperatively and
co-managed medically.
Physical examination should assess sitting and standing
balance, pelvic obliquity, curve stiffness, and the
curvature’s coronal, sagittal and rotational components.
The orthopedic surgeon should also evaluate for the
coexistence of hip subluxation or dislocation. The patient
should also have a detailed neurological examination.
Curves that are stiff in both the coronal (scoliosis) and
sagittal planes (severe hyperlordosis and hyperkyphosis)
may require anterior release or preoperative traction. The
side-bending test is helpful for assessing coronal plane
stiffness (Figure 1). Sagittal
plane stiffness should also
be assessed by seeing if
kyphosis will reduce in
the supine position over a
bolster and assessing the
flexibility of hyperlordosis
by hyperflexing both the
hips and pelvis in the supine
position. If the deformity
Figure 1. Side-Bending Test. The
cannot be reduced in
patient is bent over the examiner’s
thigh at the apex of the curve. If
either plane, anterior
the patient’s curve reverses and
discectomy should be
the pelvis levels out becoming
considered. It is important
perpendicular to the trunk, the
to assess and distinguish the curve remains flexible enough to
correct through posterior fusion
coexistence of a hip flexion
alone. If not an anterior release is
contracture and adduction
performed.
contracture. This can be
done by stabilizing the pelvis in a neutral position (flexing
the opposite hip to level the pelvis to detect hip flexion
contracture and assessing hip abduction with the pelvis
in neutral obliquity). If these contractures are present, the
parents should be warned that a staged hip contracture
release may be necessary 4 to 6 months after spinal
surgery (see Figure 2 in Chapter 3, Unique Challenges
with Scoliosis and Dislocated Hips). The surgeon must
also assess the need for fusing to the pelvis. In the patient
with CP, this is almost always necessary to prevent distal
extension (late pelvic obliquity) of a curvature fused too
short. Patients with a poor “righting reflex” should be
fused to the pelvis and up to T1.
The righting reflex is noted on physical exam in that the
patient demonstrates a balancing reflex/ability to center
the head over the body center or pelvis.
46
Standing or sitting AP and lateral radiographs are
important not only to measure curve magnitude but to
look at balance. Coronal and sagittal bending films are
operator dependent and, if done correctly, can provide
important information about curve flexibility. This author
no longer obtains traction films, since the amount of
correction that can be obtained using cantilever correction
at the time of surgery is better than traction films would
predict. The author does not routinely use bending films
to determine anterior release; the clinical bending test is
preferred because bending films are also very operator
dependent and may be misleading. The author uses the
amount of stiffness of the pelvic obliquity on physical
exam to see whether or not the pelvis can correct back
into a physiologic range by bending and adding traction to
the patient and looking at visually. Others surgeons have
stated that if the main curve correction is not better than
40 degrees on coronal bending, then anterior discectomy
is warranted. It is reasonable to consider anterior release if
residual kyphosis and lordosis is > 60 degrees on sagittal
bending films.
The anesthesiologist and surgeon should be prepared
for the possibility of major intraoperative blood loss.
Type and cross-matched blood (up to twice the patient’s
blood volume) should be available as well as fresh frozen
plasma and platelets. A cell saver machine is also helpful.
Antifibrinolytics (transemic acid) should be considered
to help reduce intraoperative bleeding. Spinal cord
monitoring should be incorporated. Any loss of sensation,
muscle spasticity, or loss of bladder or bowel control
would be considered a poor outcome.
Surgical Treatment and Instrumentation
The principles of spinal deformity correction in cerebral
palsy are to: 1) level the pelvis (by correcting pelvic
obliquity) with sitting and/or standing; 2) center the head
over the trunk and pelvis via C7 to CSL (center sacral line);
3) restore trunk balance (rib cage centered over the pelvis);
4) restore sagittal balance (lumbar lordosis and thoracic
kyphosis, including the correction of anterior or posterior
pelvic tilt, respectively); 5) maximize segmental fixation
in the face of what is often osteoporotic bone; and 6)
minimize operative time in these patients, who often have
multiple comorbidities and who are at greater risk for
wound infection. Rotational correction is less important
unless the rotational deformity is affecting pelvic or
trunk balance.
Correcting the Pelvic Obliquity
The correction of pelvic obliquity is detailed in Chapter
4 (The Correction of Pelvic Obliquity Using Cantilever
Correction). Fixation and fusion to the pelvis should be
considered in every patient with CP to prevent late pelvic
obliquity. Only if the ambulatory patient has a level pelvis
and adequate balance (“righting reflex”) should the
surgeon consider ending fixation more proximal at L4
or L5. If fixation to the pelvis is not done, fixation in the
lumbar spine at least 3 levels (6 screws) with pedicle screw
fixation and complete derotation is essential. Cantilever
correction and fixation to the remainder of the spine using
sublaminar wires is then to be utilized as described.
47
Correction of Main Thoracic Curves
Problems with the head being off center are most
problematic in thoracic curvatures. Spinal curvatures in the
thoracic region are more difficult to correct with cantilever
correction. In such cases, starting distally at the pelvis or in
the lumbar spine leaves a short lever arm at the proximal
end of the rod by the time the surgeon reaches fixation
in the thoracic spine. This makes it very difficult (if not
impossible) to center the head over the remainder of the
thorax and pelvis. This type of curvature is very difficult to
correct with the traditional unit rod since it requires distal
fixation into the pelvis first (Figure 2A). With this type of
curvature, fixation using the Expedium Neuromuscular
System can be started proximally first due to its modularity
(Figure 2B). After exposing the spine and pelvis (in cases
where pelvic fixation is required), pelvic screws and
sublaminar wires are placed as previously described. The
rod construct is then preassembled by first connecting
the pre-contoured rods with the proximal closed rod
connector at the top and placing a cross connector in the
lumbar region. The rods should be parallel from proximal
to distal with respect to their contour. Next, the top of the
rod construct is secured using sublaminar wires, hooks,
or screws from T1 down to the apex of the curvature.
After the apical vertebra is secured to the rod, cantilever
correction can be performed by gradually pushing the
rod down to the next more distal vertebra, tightening the
sublaminar wire or reducing into the implant, pushing
the rod down to the next more distal vertebra, and
then tightening the wires/implants, performing the same
maneuver progressively down the spine until the pelvic
screws are reached (Figure 2C). The lateral rod connectors
are then utilized to connect the rod to the pelvic screws.
Using this “proximal to distal” cantilever technique for
cantilever correction allows for a better lever arm to
correct thoracic scoliosis as well as thoracic kyphosis
(Figure 2C).
48
Figure 2A. It is difficult to cantilever this type of thoracic curvature using
the unit rod due to insufficient lever arm.
Figure 2A
Figure 2B. This diagram shows proximal to distal cantilever technique
that can be used for thoracic scoliosis. Preoperative and postoperative
radiographs are shown.
Figure 2B
Figure 2C. Similarly, proximal to distal cantilever technique can be used for
thoracic kyphosis. Preoperative and postoperative radiographs are shown.
Figure 2C
49
Correction of Lumbar and Thoracolumbar Curves
Trunk imbalance is most problematic in lumbar and
thoracolumbar curves. Trunk Balance is achieved with
cantilever correction from distal to proximal as described
for the correction of pelvic obliquity. In cases where
severe rotation is present (in the trunk or pelvis) and it is
significantly impacting sitting or standing balance, pedicle
screws should be used around the apex of the curve
instead of sublaminar wires. Depending on the severity of
the deformity, an Expedium uniplanar screw provides easy
sagittal accommodation of the rod to screw interface and
gives maximum axial derotation transmission of forces. If
the curve is too severe, then polyaxial screws are used to
accommodate the coronal and sagittal plane alignments
but sacrifice axial correction force. Reduction screws
can make the rod insertion easier and allow even more
slow, controlled reduction of the deformity over multiple
levels, at once lessening screw pull out (especially in these
patients who already have reduced bone mineral density)
(Figure 3A and 3B).
Cantilever correction is then performed first, securing the
rod to the spine from distal to proximal with sublaminar
wires or pedicle screws. A derotation maneuver at the
apical level of the curve is also performed if pedicle screws
were placed after the cantilever correction. The multiple
reduction instruments in the Expedium Neuromuscular
System allow slow, careful axial correction to minimize
screw loosening in the bone. In addition, it is important to
have screws placed in at least 4-5 vertebral levels around
the apex in order to distribute the derotational force and
prevent plowing of the screws within bone that is often
osteoporotic. Larger diameter screws filling the pedicle to
its cortical wall and/or the VIPER Cortical Fix Screw (Figure
4) can be helpful to reduce pullout problems of the screws
when there is poor bone quality.
50
Figure 3A. Reduction of hyperlordosis
can be achieved using pedicle screws
with reduction posts.
Figure 3B. Preoperative and postoperative photographs are shown.
Correction of Hyperlordosis or Hyperkyphosis
Restoring sagittal balance is critical. Excessive kyphosis
or lordosis may occur with or without the presence of
scoliosis. Reduction pedicle screws placed in the region of
the hyperlordotic lumbar spine can significantly facilitate
slow, controlled correction over multiple levels at the same
time to minimize the risk of loss of fixation. After securing
the rods to the pelvis with iliac screws as described in
Chapter 4 (The Correction of Pelvic Obliquity Using
Cantilever Correction), the pre-contoured rods are pushed
down into the reduction post and secured with the set
screw. Reduction of the hyperlordosis is achieved by
gradually screwing down the set screws (Figure 3). Great
care is taken to notice any evidence of posterior pullout
of the screws. Cortical fixation screws (Figure 4) may be
used to help prevent pullout. An additional suggestion is
the use of Clements pedicle dilators which dial up the size
of the pedicle to allow a bigger diameter screw and better
pedicle cortex fixation. If pullout starts to occur despite all
of the above-mentioned techniques, then a sublaminar
wire can be placed at the same level as the screw for
additional fixation, acting as belt and suspenders. Once
the hyperlordosis is reduced, the remainder of the spinal
instrumentation is completed.
Figure 4. This diagram shows the screw design of VIPER
Cortical Fix Pedicle Screw.
Cortical
Fix Thread
Hyperkyphosis can also be corrected using the Expedium
Neuromuscular System using either a “distal to proximal”
(in cases of lumbar or thoracolumbar kyphosis) or
“proximal to distal” cantilever technique (in the case of
a thoracic kyphosis) (Figure 2). It is critical that fixation
is completed up to at least T1 to prevent “drop-off”
proximal junctional kyphosis at the cervicothoracic
junction. Firm fixation at the proximal most end is
recommended. The author usually does not use laminar
hooks at T1, as they can result in unreliable fixation in
soft bone as well as tending to pull out in patients with
spasticity who do a lot of head posturing. Pedicle screws
in T1 or T2 can be another alternative as can transverse
process hooks on T1 if pedicle screws cannot be inserted.
Maximize Fixation in Face of Osteoporotic Bone
In patients with quadriplegic CP who have severe
osteoporosis, a full evaluation of the patient’s bone density
should be performed using DEXA scanning. In addition,
evaluation of the patient’s vitamin D metabolism should
be performed by obtaining a full vitamin D panel as
well as calcium, phosphorous, and alkaline phosphotase
levels. Preoperative treatment may be necessary with
several courses of intravenous pamidronate (in cases of
osteoporosis) or with vitamin D (in cases of osteomalacia,
usually secondary to seizure medication). Sublaminar
fixation is preferred since laminar fixation has been shown
to be biomechanically stronger than pedicle screws in
the presence of osteoporotic bone1. If screw fixation is
necessary, such as in cases of hyperlordosis or severe
rotation, sublaminar wire fixation should be added at
each level of screw fixation. In the author’s experience,
this is most commonly necessary at the L5 level. As
described previously in this chapter, cortical fixation screws
(Figure 4) and/or pedicle dilatation techniques can be
effective adjuncts.
Standard
EXPEDIUM/VIPER
Dual Lead Thread
51
Minimizing Operative Time
Sublaminar Wire Passage
Minimizing operative time is crucial, because the longer
the spine is exposed, the higher the risk of bleeding
and infection. Generally, in surgeons well trained in
both techniques, passing sublaminar wires is faster than
pedicle screw fixation. Most routine spinal deformity
corrections can be performed with segmental sublaminar
wire fixation alone. It is also more cost effective than
screw fixation. When pedicle screw fixation is performed,
it is limited only to the levels that are needed, usually
around the apex of the scoliotic curve or in the region
of hyperlordosis. In cases where epidural bleeding
is problematic, pedicle screw fixation can also be
considered to avoid further epidural bleeding. In the
author’s experience, however, epidural bleeding can
be minimized by limiting removal of the ligamentum
flavum to the central portion and avoiding going too
lateral where the veins are more prominent (Figure 5).
When epidural bleeding does occur, the application
of Surgifoam and thrombin usually stops epidural
bleeding. Other considerations for minimizing the
operative time include careful preoperative assessment
of curve stiffness. A posterior only approach is usually
all that is needed for curve correction when the curve is
< 70 degrees and still flexible. In curves > 70 degrees,
it may be safest to perform anterior release in a stiff
curve to minimize the technical difficulty and risk
of loss of fixation while attempting to correct a stiff
deformity from a posterior approach alone. Although
not used by this author, some surgeons prefer using
anterior instrumentation to derotate the spine after
anterior release which then makes posterior correction
even easier. Some surgeons prefer a posterior only
approach utilizing extensive posterior osteotomies and
even vertebral body resections in the case of severely
stiff curves. Other surgeons may prefer some form of
preoperative or perioperative traction.
After the spine is completely exposed, the spinous
processes are completely removed and the ligamentum
flavum is carefully removed to expose the sublaminar
spaces. Double Luque wires are bent (prebent wires are
also available) and passed under the lamina of L5 up to
and including the T1 lamina. The radius of curvature for
the wire bend must approximate the width of the lamina
to allow safe passage of the wire. Two double wires are
passed at the L5 and the T1 lamina only while single wires
are passed at each of the other levels. The wires are pulled
up to equal length and bent next, with the midline bent
flat down onto the lamina and the beaded end flat down
lateral onto the paraspinous muscles. This helps prevent
the wires from getting inadvertently pushed into the
spinal canal and allows for easier wire organization
(Figure 5 A-G).
52
Figure 5. Luque Wire Passage-when passing wires, it is important to roll the wires under the lamina (A-D), being
careful not to catch the tip under the lamina (E), which will lever the wire into the canal and place pressure on
the spinal cord. Wires are bent down to the midline in the middle and the ends are bent down flat against the
paraspinous muscles (F-G).
Figure 5A
Figure 5F
Figure 5B
Figure 5C
Figure 5G
Figure 5D
Figure 5E
53
Outcomes
The long-term results of cantilever correction of spinal
deformity have been with the unit rod. In our experience,
approximately 70% curve correction is achieved (in both
posterior only or when anterior release is indicated).
Improved appearance, quality of life, and ease of care is
achieved in the majority of patients. Difficulties remain,
however, using the unit rod when correcting hyperlordosis,
judging rod length, correcting severe rotation, and
having a sufficient lever arm to correct more proximal
thoracic curves. The Expedium Neuromuscular System
can produce similar results through cantilever correction
of both scoliotic and sagittal plane deformities in patients
with CP. In addition, pelvic fixation with iliac screws is
technically easier than with the unit rod, especially in the
face of severe lordosis. Problems with judging rod length
is obviated with the Expedium Neuromuscular System
as the rod is merely cut at either the top or bottom end
of the fixation level. Severe rotational deformity can be
addressed by adding pedicle screw fixation, especially
Expedium uniplanar screws. Finally, thoracic curves can
be more easily corrected using a ”proximal to distal”
cantilever maneuver which cannot be performed with the
unit rod, because the pelvic limbs must be placed first.
54
References :
1. Coe JD, Warden KE, Herzig MA, et al. Influence of
bone mineral density on the fixation of thoracolumbar
implants. A comparative study of transpedicular screws,
laminar hooks, and spinous process wires. Spine (Phila
Pa 1976) 1990;15:902-7.
55
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-63-000 9/2012 JC
CA #8786 / DJ10058A
Spinal Fusion in Myelomeningocele
Peter G. Gabos, MD
Surgical treatment of spinal deformity in patients with
myelomeningocele presents a group of challenges unique
to this patient population. The combined effects of severe
multiplanar spinal deformity that can include scoliosis,
kyphosis and lordosis with elements of congenital vertebral
malformations, absence of posterior elements, markedly
abnormal pedicle trajectory, compromised skin, subcutaneous
dura, marked truncal obesity, and disuse osteopenia can
challenge all aspects of surgical care. Confounding medical
conditions that can have a major impact on postoperative
morbidity include shunted hydrocephalus, Chiari malformation,
tethered cord, bladder augmentation procedures, bowel
incontinence, thoracic insufficiency syndrome, and marked
lower extremity contractures.
57
Preoperative Evaluation
In general, scoliotic curves < 40 to 50 degrees can be
managed nonoperatively. Bracing can be utilized in
select cases to assist in truncal stabilization and handsfree sitting but would not typically be used for curve
management. Poor skin and anterior-placed genitourinary
and bowel appliances can render bracing difficult at
best. In addition to curve magnitude, these patients have
other functional problems that may make correction and
fusion of their spine desirable. The impact of curvature on
seating and, where applicable, on the use of long-leg and
pelvic-based lower extremity orthoses may be important
to avoid pressure ulceration over insensate skin. Spinal
stabilization may be advantageous if progressive curvature
makes hands-free sitting difficult or impossible. Signs of
impaired pulmonary function, including the presence of
a “Marionette sign” or increasing loss of thoracic volume
due to increasing spinal deformity or diaphragmatic
intrusion into the thorax, should be recognized. In the
presence of kyphosis, pressure ulceration over the apex
of the kyphus can be a lifelong struggle, and chronic
ulceration and vertebral osteomyelitis can occur (Figure 1).
Figure 1. Preoperative photograph of a chronically infected
myelomeningocele patient who sustained breakdown over an area of
untreated severe lumbar kyphosis. There is a draining wound, loss of
skin coverage and exposed, necrotic bone.
58
Preoperative evaluation should involve comprehensive
characterization of all aspects of the spinal deformity.
In addition, the exact location of bifid or open bony
segments is critical to avoid dural tearing or further
neurologic injury, since posterior deficiencies cephalad
to the frankly obvious level of myelomeningocele
can also exist. Pelvic and sacral anatomy must also
be rigorously defined, as the majority of patients will
require stabilization and fusion to the pelvis. Flexibility
evaluation can include supine bending, fulcrum bending,
and/or spinal traction films to assess the need for
spinal osteotomies and/or vertebral column resection.
Computed tomography (CT) scan and the use of 3D CT
technology can facilitate preoperative planning of surgical
approach, correction techniques, and spinal and pelvic
anchor placements, especially in cases involving multiple
congenital malformations where plain radiography alone
may not be sufficient. Magnetic Resonance Scanning (MRI)
scanning can evaluate for subcutaneous dura, spinal cord
tethering, Chiari malformation, and syrinx.
If the patient has a pre-existing ventriculoperitoneal (VP)
shunt, a preoperative shunt evaluation should include
radiographic confirmation of the structural integrity
of any shunts from point of exit from the skull, with
particular attention to the cervical region where shunt
fracture is most common. Preoperative head CT or MRI
can be invaluable as a comparison tool for assessing for
postoperative hydrocephalus should shunt failure after
deformity correction occur. This is of particular importance
if a structural failure (separation) of a shunt is noted
preoperatively, since arrested hydrocephalus should not
be assumed. Postoperatively, radiographic confirmation
of shunt integrity should be obtained. Symptoms of
hydrocephalus due to shunt failure can include headache,
nausea, emesis, lethargy, extra-ocular movement
abnormalities, cognitive changes, neurologic deterioration,
and even respiratory arrest and death.
Whenever feasible, pulmonary function testing may
be useful to characterize components of pulmonary
insufficiency. A multidisciplinary approach, including
consultation with colleagues from neurosurgery (for
shunt management and consideration for detethering),
plastic surgery (for assistance with closure and wound
management), urologic surgery (for decreasing the risk
of postoperative urinary tract infections, placement of
preoperative urinary catheters if bladder reconstruction
or diversion procedures have been performed), and
rehabilitative medicine (for enhancing postoperative
rehabilitation and functional outcome) can optimize
decisions regarding timing of surgery and plans for
postoperative recovery. The environment to which the
patient will be returned after surgery can also greatly
impact the overall success of the procedure, and
evaluation of family structure and dynamics by a team
from Social Services is essential. Functional assessments
should include detailed evaluation of motor and sensory
level, upper and lower extremity use, ambulatory function
and adaptive equipment where applicable, method of
catheterization, and overall level of desired postoperative
independence.
Preoperative nutritional parameters including serum
albumin and white blood cell count, urinalysis, and urinary
cultures should be obtained, and treatment of urinary
tract infection should be completed prior to surgery.
Aggressive postoperative nutrition via hyperalimentation
may be required to support wound healing. Complete
assessment of skin integrity, truncal obesity, dentition, and
overall hygiene should be performed.
Preoperative Considerations
Surgical approach will vary based on curve stiffness
and severity and may include single-stage or two-stage
anterior and posterior surgery. Preoperative halo traction
may be indicated for markedly stiff and/or severe scoliosis
or kyphosis. If a halo is to be placed, the location of
pre-existing VP shunts must be considered. Single-stage,
posterior-only approaches can be performed in the
majority of cases but may require the use of advanced
osteotomy techniques, including vertebral column
(apical) resection. Neuromonitoring is considered on a
case-by-case basis depending on underlying neurologic
deficit. In the non-ambulatory patient with little or no
lower extremity function, upper extremity monitoring
in the surgical position for a period of time may be
useful to prevent brachioplexopathy or peripheral nerve
compression, especially in cases where prolonged surgical
time is anticipated. Consultation with a plastic surgeon for
incision placement may be helpful in optimizing wound
coverage, including the use of preoperatively placed soft
tissue expanders.
Surgical Techniques
Skin incisions must take into account pre-existing skin
scarring, and the skin quality must be assessed over the
ilia bilaterally if pelvic fixation will incorporate fixation
over the posterior superior iliac spines (PSIS). Careful
attention to body positioning on the spinal OR table must
incorporate adequate padding and positioning of all
soft tissue and bony prominences, and lower extremity
positioning may be difficult in the face of significant lower
extremity deformity or contracture. Extending or flexing
the hips can be helpful in achieving more or less lumbar
lordosis, respectively. If fluoroscopic imaging will be
utilized for optimizing pelvic fixation, taking the necessary
images prior to prepping and draping will confirm
visualization of the iliac wing teardrop for placement
of PSIS or second-sacral (S2) to iliac screws to assure
unobstructed radiographic access to the pelvis (Figure 2).
Figure 2. An intra-operative FlouroScan image of the iliac teardrop is
taken bilaterally to assure an unobstructed radiographic view during
pelvic screw fixation. The images are obtained after initial positioning
of the patient on the operating room table, just prior to prepping
and draping.
59
As the skin and soft tissue dissection proceeds, definitive
knowledge of the location of potentially exposed dural
elements and posterior bony deficiencies must be taken
into account to avoid dural tearing or spinal cord injury. In
select cases, spinal cord transection and durotomy may be
performed at this point.
compromised and adherent areas. Iliac osteotomy should
be performed with caution or not at all if PSIS screw
placement is desired, as it may compromise fixation.
Fixation options can vary based on surgeon experience
and corrective goals. For the majority of patients, surgical
goals will include restoration of spinopelvic balance for
wheelchair use, so strategizing for reduction of pelvic
obliquity and pelvic-to-shoulder (truncal) rotation is
important. Restoring the center sacral line (CSL) to a more
neutral position to bring the head back over the pelvis in
both the sagittal and coronal planes also allows for better
weight distribution from the ischial tuberosities to the
posterior thighs to avoid skin breakdown and pressure
ulceration. Coccygeal morphology should also be noted,
as excessive surgical lordosis may impact ulceration over
this prominence in thinner patients.
Anchors based within the posterior elements, such as
sublaminar wires or laminar hooks, cannot be utilized
in regions of frank posterior element deficiency. Pedicle
screws can offer improved versatility and strength of
fixation in the spine and are essential at areas where
vertebral column resection or osteotomies are planned.
Temporary stabilization rods placed into pedicle screws
around large osteotomies, and areas of vertebral column
resection, can help avoid abrupt spinal translation,
typically that which occurs from the distal spinal segment
translating anteriorly. Reduction screws and polyaxial
screws allow for easier and more gradual rod capture into
the screw heads, especially in osteopenic bone. Uniplanar
screws can enhance spinal derotation, typically at curve
apices and at upper and lower construct foundations.
In regions of posterior element absence, disc excision
and placement of structural interbody supports from
an anterior or posterior approach can add stability and
surface area for fusion. When performed posteriorly, these
supports are typically inserted prior to rod placement.
The pedicles in the lower vertebral segments may be
difficult to access secondary to marked lateral-to-medial
trajectory from the embryologic failure of bony closure.
Obese body habitus combined with increased lumbar
lordosis can pose a particularly difficult problem to
cannulating the pedicles. Osteotomy of the iliac wings
can allow better access and trajectory for pedicle screw
placement into the lower lumbar vertebral bodies and
sacral promontory (Figure 3). In regions where dense
scarring and compromised skin overlies the posterior iliac
spines, osteotomizing the posterior iliac crest and leaving
a thin rim of crest cartilage (if present) or bone attached to
the overlying skin can also serve to prevent skin tearing in
60
Figure 3. Osteotomies of the iliac wings can allow
better access and trajectory for pedicle screw
placement into the lower lumbar vertebral bodies
and sacral promontory (top Image).
Second-sacral (S2) to iliac screw fixation allows deeper
seating of the iliac screws to avoid skin breakdown over
the screw heads and lower spinal rods (see Chapter 2:
Sacropelvic Fixation Options), and is not compromised
by iliac crest osteotomy. In many cases, the S2 to iliac
screw allows rod seating directly from the spine to pelvis
without the need for offsets or other lateral connectors to
link up the rod, which may eliminate an area of potential
fixation failure. Placement of caudal lumbar screws and/
or first sacral (S1) screws (if utilized) prior to selection of
the start point for S2 to iliac screw insertion is helpful if
direct rod seating from the spine is desired (Figure 4). If
PSIS screw use is desired, recession of the screw head into
a surgically-created iliac slot should be strongly considered
to avoid implant prominence (Figure 5). Other screw
options would include sacral fixation utilizing S1 and S2
sacral screws.
Figure 5. A surgically created iliac slot will help avoid
implant prominence when PSIS screw is used.
Figure 4. Image shows the straight alignment of L5
pedicle screw, S1 promontory screw and S2 to iliac
pelvic screws. The screws are placed in that sequence
to allow rod seating directly from the pelvis to spine
without the need for any type of lateral or offset
connector from the pelvic fixation to the rod.
Techniques of rod placement for deformity correction can
vary depending on surgical goals and surgeon preference.
Principles of load-sharing across multiple implants and
gradual rod reduction should be adhered to in order to
prevent pull-out of implants in osteopenic bone. Rod
reduction techniques can vary from sequential versus
simultaneous dual rod placement. When placing dual rods
simultaneously, a surgeon-contoured or pre-contoured set
of rods can be linked proximally with a cross-connector
similar to unit rod instrumentation utilizing cantilever
correction. In this case, the pelvic screws are captured
first, followed by caudal to cephalad gradual rod capture.
En bloc versus segmental rotational correction can be
achieved with uniplanar screw fixation, typically required
at curve apices. Other corrective techniques that can be
utilized include translational correction, distraction (to
restore kyphosis) and compression (to restore lordosis) and
in situ rod bending. After initial correction is completed,
intraoperative fluoroscopic images allow for assessment
of any residual pelvic obliquity, which can be further
improved utilizing compression or distraction across the
lumbosacral and pelvic fixation. A full-length lateral spinal
radiograph is also obtained to assess overall sagittal
contour and screw length prior to wound closure
(Figure 6 A-I).
61
Figure 6A. Preoperative sitting
posteroanterior radiograph of a
mid-lumbar level, non-ambulatory
adolescent with progressive
kyphoscoliosis and increasing
difficulty with hands-free sitting in
his wheelchair.
Figure 6B. Preoperative sitting
lateral radiograph.
Figure 6D. Intraoperative photo after spinal exposure,
detethering and dural patch reconstruction. A
posterior lumbar interbody fusion is being performed
to enhance curve flexibility, surface area for fusion
and spinal stability.
62
Figure 6C. Preoperative skin over the region of
previous myelomeningocele repair demonstrates
markedly scarred, paper-thin skin that was adhered to
the underlying posterior iliac crests.
Figure 6E. Intraoperative photo taken after segmental
pedicle screw placement. Sacral-2 (S2) to iliac screws
have been placed to avoid screw prominence at the
posterior superior iliac spines (PSIS), and the iliac
crests were osteotomized to protect the overlying
skin and to facilitate screw placement in the lower
lumbar spine and sacral promontory.
Wound Management
Figure 6F. Intraoperative photo taken after dual
rod placement using techniques of en bloc spinal
translation and direct vertebral derotation.
Figure 6G. Intraoperative appearance after skin closure utilizing wide
mobilization of the subcutaneous tissue and drain placement to decrease
skin tension from hematoma formation.
Operative success is critically impacted by postoperative
wound management. Skin closure in these patients can be
extremely complex, and inability to close the skin in cases
of maximal scoliosis correction can lead to catastrophic
deep infection, especially over regions of exposed dura.
The use of preoperative skin expanders may be considered
in some cases. Careful skin and soft tissue handling during
the entire procedure is critical to success. Mobilization
of large soft-tissue flaps, muscle rotation flaps from the
latissimus dorsi or abdominal obliques, and skin grafting
may be necessary. Careful preoperative assessment of
dependence on upper extremity and shoulder girdle
function for mobility is important in counseling the
patient or family regarding the potential effects of the
use of muscle transfer flaps on crutch or walker use or
wheelchair propulsion. Use of perioperative deep and
superficial drains may help to alleviate swelling and its
subsequent effect on skin tension. Specialized mattresses
for use during postoperative recovery and frequent
inspection of the wound for dehiscence, signs of infection,
or soiling should be utilized, and the nursing staff should
be carefully instructed in the proper replacement of
occlusive dressings.
Postoperative Care
The majority of patients will recover in an ICU setting until
they are medically stabilized. Upon successful extubation,
sometimes up to several days postoperatively, patients are
moved out of bed and into a chair. Their own wheelchair
may need modifications and should be evaluated. Daily
inspections of the incision are performed, and the lower
portion is always kept resealed with a waterproof occlusive
dressing to prevent soiling. Any surgical drains are kept
in place until output totals less than approximately 50 to
100 cc of fluid. Weight bearing, if applicable, and transfer
training are started immediately as tolerated.
Figure 6H. Postoperative sitting
posteroanterior radiograph one
year after surgery. There were
no wound complications, and
the patient could enjoy handsfree sitting and unlimited
sitting tolerance.
Figure 6I. Postoperative sitting
lateral radiograph one year
after surgery.
63
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-64-000 9/2012 JC
CA #8781 / DJ10052A
The Patient with Spinal Cord Injury:
Surgical Considerations Particularly with
Respect to the Sagittal Plane
Amer F. Samdani, MD
Patients with spinal cord injury (SCI) fall under the
spectrum of neuromuscular scoliosis, and many of the
treatment principles outlined elsewhere in this monograph
apply. However, distinct differences exist in patients with
SCI, the knowledge of which may minimize the occurrence
of complications.
65
Preoperative Considerations
As for all patients with neuromuscular scoliosis, the
preoperative evaluation includes a rigorous medical
and anesthesia work-up. At our institution, we perform
a 30 minute Hibiclens back scrub starting three days
preoperatively. Many of these patients are on an every
other day bowel program, and we ensure a bowel
movement the day prior to surgery. Patients with SCI
may develop deep venous thrombosis, and preoperative
ultrasound of all four extremities is performed. These
usually reveal clinically insignificant, chronic superficial
vein thrombosis which then provide a baseline for
comparison postoperatively. However, there have been
reports of patients having clots in deep veins which have
delayed surgery.1
We obtain a preoperative evaluation with physical and
occupational therapy. The purpose of this evaluation is
multifold: First, the therapist outlines to the patient and
family the potential positive and negative functional
consequences of a straighter, yet stiffer, spine. They
emphasize the effect on activities of daily living that the
patient currently performs. Part of this evaluation may
include placing a rigid thoracolumbosacral orthosis (TLSO)
on the patient to emulate spine fusion. This is often
performed on marginally ambulatory patients to predict
their ability to ambulate postoperatively. In addition, the
TLSO trial on patients with high tetraplegia can predict
what happens with loss of compensatory function
patterns that may affect upper extremity function
following a spinal fusion. The preoperative physical
therapy evaluation also reinforces the postoperative
restrictions. Generally, for a period of six months, the
patients will not self-propel their wheelchair, assist in
transfers, or flex their hips past 90 degrees. Factors which
influence the duration of these restrictions include patient
body habitus and bone quality.
66
The preoperative evaluation includes standard PA/lateral
sitting full spine radiographs and supine full spine bend
or traction films. In particular, a high quality sitting
lateral radiograph is imperative, as these patients rely
on an exaggerated kyphosis to permit the performance
of activities of daily living, including self-catheterization
and feeding. We analyzed the sagittal profile in 30
patients with SCI at our center and found the usually
neutral T10-L2 region measured a kyphosis of 19.8
degrees. Similarly, the lumbar lordosis averaged 9.8
degrees. Thus, fusing these patients with a ‘normal’
sagittal thoracic kyphosis and lumbar lordosis may prevent
them from performing at their preoperative level. This is
particularly true of patients with limited upper extremity
functioning. It is very instructive for the surgeon to
observe these patients performing ADLs to truly appreciate
the importance of maintaining their kyphosis. When a
patient with tetraplegia cannot sit up, the authors will
make a temporary TLSO and see how the patient can
function. If the patient does well in the TLSO, then a
lateral radiograph in the TLSO is obtained and the rods
bent to match. Sometimes a malleable rod can be bent
preoperatively against the film and then sterilized.
Intraoperative Considerations
Similar to other patients with neuromuscular
scoliosis, these patients may benefit from the use of
antifibrinolytics, although this should be weighed against
the increased risk of deep venous thrombosis. The surgeon
should also consider the use of a central line, as patients
with cervical level injuries may demonstrate hemodynamic
instability secondary to autonomic dysfunction.2
Furthermore, the anesthesia team should refrain from
the use of succinylcholine, as it may cause release of
potassium and sudden cardiac arrest.3 Intraoperative
antibiotics includes strong coverage of gram positive and
gram negative bacteria, as previously reported by others.4
We prefer cefazolin and gentamicin, with vancomycin
substituting when there is a penicillin allergy. In addition,
we mix vancomycin with the bone graft and also use
vancomycin powder spread on the wound surfaces prior
to closure, as demonstrated by others, to decrease the
incidence of infection.5
Positioning
Proper positioning is important in these patients. Although
we prefer the Jackson table, often these patients have hip
flexion contractures and are better suited to a regular OR
table which allows greater customization by altering the
size of gel rolls. Patients with severe flexion contractures
(> 70 degrees) may require the four poster frame. On a
rare occasion, these patients may need a hip contracture
release a couple of weeks before the spine fusion.
Hip contracture releases should be performed with a
longitudinal incision, since horizontal incisions commonly
dehisce, and this may delay the subsequent spine fusion.
In addition to appropriate pressure point distribution,
Figure 1. A 14 year old patient
with a large, stiff deformity with
marked pelvic obliquity. A) Pre-op
sitting PA, and B) pre-op sitting
lateral. C) She underwent a T2
to pelvis posterior spine fusion
utilizing temporary internal
distraction to maximize correction
of the pelvic obliquity. Image
shows Post-op PA x-ray.
Figure 1A
the surgeon should ensure the maintenance of optimal
sagittal profile. In particular, the Jackson table will lordose
the thoracic spine unless the chest roll is positioned
caudal to the nipple line. Similarly, in some patients it
is easy to overextend the hips, which imparts a nice
lumbar lordosis well suited for an ambulatory patient but
likely very different than the preoperative sitting sagittal
profile of the patient. Overlordosing the lumbar spine in
combination with decreasing thoracic kyphosis will result
in significant negative sagittal balance with a decrease
in the ability to perform daily tasks. In addition, a severe
junctional proximal cervical thoracic kyphosis may result.6
Neuromonitoring
We always perform upper extremity monitoring to prevent
any loss of function from brachial plexus palsy secondary
to prolonged positioning. Any patient with incomplete
injury, including ASIA B (sensation only), should have
neuromonitoring customized to their neurological status.
Traction
As a general rule of thumb, if the preoperative supine films
demonstrate a residual pelvic obliquity of > 20 degrees,
the authors consider utilizing intraoperative traction. This
could be done with halo femoral traction (which provides
excellent results with respect to correction of pelvic
obliquity) or with a temporary device placed between the
pelvis and two ribs above the end vertebra. The spine is
distracted every 15-30 minutes, allowing for viscoelastic
relaxation (Figure 1). In patients with an incomplete
SCI and functional lower extremities, close attention
to neuromonitoring is imperative with relaxing of the
distraction for any significant change.
Figure 1B
Figure 1C
67
Construct
At a minimum, the construct must include large iliac
bolts with a strong foundation of screws at the base (iliac
bolts and S1 screws are preferred by the authors), pedicle
screws across the thoracolumbar junction, and pedicle
screws at the top of the construct. The authors’ preference
remains erring on the side of more screws, particularly in
patients with poor bone quality (Figure 2). Strategically
placed reduction screws at the apex of the concavity assist
with rod reduction. Wherever possible, rod material should
be a titanium or cobalt chrome alloy. The higher incidence
of infections seen with stainless steel discourages its use
in this high-risk population. Rod diameter is often 5.5mm,
but in heavier patients (>70 kg), 6.35mm (¼ inch) rods
should be considered, bone quality permitting. Metabolic
syndrome may develop in these patients over the years,
with a consequent substantial increase in BMI.
Figure 2. The VIPER® Cortical Fix Screw maximizes pullout strength
with its single to dual lead design. It is an excellent option in patients
with poor bone quality, particularly in high stress areas.
68
Several features of the EXPEDIUM® Neuromuscular System
facilitate the operation in these patients (Figure 3). In
addition to the availability of a variety of screws and rod
sizes/materials, the iliac bolts are particularly useful. The
profile of the bolt is minimal, which is further marginalized
with the use of a medial/inferior entry point for the iliac
bolt (See Sacropelvic Fixation Techniques Chapter for more
details on insertion techniques). This entry point will often
obviate the need for a connector, but a variety of both
open and closed multiaxial head connectors are available
to ease connection to the rod (Figure 1). In addition, the
iliac bolts are cannulated and thus can be placed with
minimal dissection.
Figure 3. (Left) EXPEDIUM iliac screw with a low profile head and
a variable angle head allowing for easier rod connection. (Right)
Cantilever correction of pelvic obliquity by introducing rods from distal
to proximal unitized with a cross connector. This allows for powerful
cantilever force to correct the deformity.
Correction
The rod is contoured to match the preoperative sitting
sagittal profile in nonambulatory patients. In ambulatory
patients, the EXPEDIUM Neuromuscular System’s prebent rods offer a good option. The authors will often add
more thoracic or thoracolumbar kyphosis, anticipating
loss of kyphosis from positioning prone. The rods are
introduced simultaneously from the iliac bolts moving
proximally. A cross connector linking the rods distally and
proximally allows for strong cantilever forces to correct
pelvic obliquity (Figure 4). In smaller, more flexible curves,
a single rod introduced from the bottom will suffice.
The majority of correction is obtained via cantilever
forces, following which multiple rounds of compression/
distraction optimize correction.
Although derotating the spine is considered a maneuver
utilized in patients with adolescent idiopathic scoliosis, the
authors have found it to be useful in attaining maximal
coronal balance in patients with stiff, severely rotated
lumbar curves. When looking at an intraoperative PA
radiograph, sometimes the C7 to CSL is more than 2cm
lateral. Instead of a coronal bend in the rods, one should
assess the thoracolumbar rotation status. If still rotated,
en bloc and segmental derotation which will secondarily
correct the C7 to CSL line should be considered. Having
uniplanar pedicle screws in the thoracolumbar spine
allows this rotation correction. Sublaminar wires would
require in situ bending which really is just compensating
for not doing a good 3-D correction.
Figure 4. EXPEDIUM® SFX® Cross Connector Options for spine constructs.
Fixed (top two) and Variable (bottom) options. These allow for efficient
attachment while maintaining a low profile.
69
Postoperative Care
Summary
In the postoperative period, we usually stop the
gentamicin after 48 hours and continue IV gram positive
coverage while the drains are in place. After the IV
antibiotics are discontinued, the patient is switched to
an antibiotic by mouth for six weeks after discharge to
prevent seeded infection from infecting or contaminating
urine. Since many of these patients are n ot continent
of urine or stool, we strictly enforce covering the inferior
one-third of the wound with a bioclusive dressing at all
times for the first month after surgery. Postoperative
ultrasounds of all four extremities are obtained to rule out
DVT at one week. Lovenox is started after the drains are
removed and used for six weeks to prevent DVT. Activity
restrictions for six months include obtaining assistance
with transfers, not flexing the hips past 90 degrees, and
not self-propelling the wheelchair. Spine deformity secondary to spinal cord injury presents
some unique challenges for the patient and surgeon.
Careful attention to the preoperative evaluation
and surgical planning is very important. We strongly
recommend use of the advanced implant fixation options
provided by the EXPEDIUM® Neuromuscular System to
optimize the correction in all three planes. Utilizing 3-D
correction techniques to restore the patient’s best sagittal
balance (standing or sitting) is necessary to maximize the
positive functional outcome and minimize the negative
effects of a spine fusion.
70
References
1. Jones T, Ugalde V, Franks P, et al. Venous
thromboembolism after spinal cord injury: incidence,
time course, and associated risk factors in 16,240
adults and children. Arch Phys Med Rehabil
2005;86:2240-7.
2. Krassioukov A, Claydon VE. The clinical problems in
cardiovascular control following spinal cord injury: an
overview. Prog Brain Res 2006;152:223-9.
3. Nash CL Jr, Haller R, Brown RH. Succinylcholine,
paraplegia, and intraoperative cardiac arrest. A case
report. J Bone Joint Surg Am 1981;63:1010-2.
4. Vitale MG, Mackenzie WG, Matsumoto H, et al.
Surgical site infection following spinal instrumentation
for scoliosis: lessons learned from a multi-center
analysis of 1,352 spinal instrumentation procedures for
scoliosis. Scoliosis Research Society annual meeting,
Louisville, KY, September 14-17, 2011.
5. Borkhuu B, Borowski A, Shah SA, et al. Antibioticloaded allograft decreases the rate of acute deep
wound infection after spinal fusion in cerebral palsy.
Spine (Phila Pa 1976) 2008;33:2300-4.
6. Sink EL, Newton PO, Mubarak SJ, et al. Maintenance
of sagittal plane alignment after surgical correction of
spinal deformity in patients with cerebral palsy. Spine
(Phila Pa 1976) 2003;28:1396-403.
71
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-64-000 9/2012 JC
CA #8784 / DJ9932A
Surgical Management of Neuromuscular
Spinal Deformity – Muscular Dystrophy &
Spinal Muscular Atrophy
William G. Mackenzie, M.D.
Management of spinal deformity in patients with progressive
neuromuscular disorders such as Duchenne’s muscular
dystrophy (DMD) and spinal muscular atrophy (SMA) can be
very challenging. Although neuromuscular disorders share
many common features that result in the pathogenesis of
neuromuscular spinal deformity, these disorders can vary
greatly. Typically, the spinal deformity is a long, sweeping
curve in the thoracolumbar region which is associated with
pelvic obliquity and usually with kyphosis. These large spinal
deformities progress through skeletal maturity and can cause
back pain, sitting intolerance, loss of independent mobility,
and loss of hand function, and can aggravate restrictive
lung disease. Typically, bracing does not play a role in these
deformities; it may slow progression but there is no evidence
that it prevents it. Surgical management can be highly
effective for these spinal deformities. The aim of surgery is
to provide a well-balanced spine which will result in more
comfortable seating and a better quality of life than that
possible with the untreated scoliosis.
73
In the last 20 years, there has been a dramatic change
in the management of patients with DMD. In the era
before steroid therapy, progressive scoliosis occurred in
more than 90 percent of these cases. Most were nonambulatory when the scoliosis was diagnosed. Although
the curve often presents earlier, the peak progression
occurred between 13 and 15 years of age and progressed
at 1 to 3 degrees per month. Although these patients
have a natural history of progressive, restrictive lung
disease, the lung function is related to curve magnitude.
Forced vital capacity (FVC) is usually > 35% of normal
in more than 90% of patients with a scoliosis ≤ 30
degrees. However, with larger curves there is significant
reduction of the FVC. There are increased postoperative
respiratory complications when the FVC goes below 35%.
My experience and that of others suggests that there is
no difference in pulmonary deterioration and long-term
survival following spine surgery. Non-surgical techniques
such as bracing, lordotic positioning, and wheelchair
modification may delay but do not prevent progression
and are not indicated in these patients. Deflazacort,
a steroid, has been shown to slow the progression of
weakness and allow continued ambulation longer than
possible prior to steroid use and to dramatically reduce the
incidence of progressive scoliosis.
Patients with SMA are classified by the International SMA
Consortium according to age at presentation as follows:
Type I: severe onset at birth to 6 months, never able to
sit without support; Type II: intermediate onset before
18 months and able to sit but unable to stand or walk
unaided; and Type III: mild onset after 18 months and able
to stand and walk. Typically, Type I has the more severe
muscle weakness and the more progressive deformity.
Patients with SMA usually have thoracolumbar curves
associated with kyphosis and pelvic obliquity. Despite
the large magnitude, the curves are often flexible but
can continue to progress. A soft or a polypropylene
lined thoracolumbosacral orthosis (TLSO) can be used to
improve sitting balance. Caution with braces should be
exercised, as they can result in significant rib deformity
and must be monitored closely.
74
Indications for Spine Surgery
Classical indications for posterior spinal fusion in DMD
include a progressive scoliosis > 30 degrees and preferably
before the FVC declines below 35% of normal. With
current medical management, many of these patients
may not develop scoliosis or will develop scoliosis in late
adolescence. If these patients have a scoliosis > 50 degrees
in late adolescent (after skeletal maturity), they should
be considered candidates for posterior spinal fusion and
instrumentation. It is important to avoid anterior spinal
fusion in these patients, as there is significant long-term
reduction of pulmonary function with these techniques.
Preoperative Evaluation
Careful multidisciplinary preoperative evaluation is critical
for these patients. Frequent upper respiratory tract
infection and cough is very suggestive of underlying
pulmonary disease. In these patients, it is important to
determine lung volumes and do pulmonary function
testing. A cardiology consultation is very important
for patients with muscular dystrophy because of
the associated cardiomyopathy. These patients can
have problems with sudden hemodynamic changes
intraoperatively, and it is important to have a good
understanding of the cardiac function preoperatively.
There can be significant intraoperative blood loss, and
appropriate blood products should be available, as well as
a cell saver.
Levels of Fusion and Instrumentation Options
There has been some controversy over the optimal levels
of fusion for these patients. Conventional wisdom would
indicate that in patients with neuromuscular disorders
of this type the fusion should extend from T3 to the
sacropelvis. In DMD, it is recommended that if the scoliosis
is < 40 degrees and pelvic obliquity is < 10 degrees, fusion
to L5 is adequate. Fusion to L5 or S1 has comparable
results if the apex of the curve is above L1. There is a
greater blood loss and longer hospital stay with pelvic
fixation, but with recent modifications of pelvic fixation
with iliac or S2 iliac screws, outcomes show no difference
in hospital stay, and difference in blood loss is negligible.
In SMA there are different challenges. In some patients,
the pelvis can be very small, with abnormal anatomical
features and poor bone quality complicating the pelvic
fixation. The EXPEDIUM® 4.5mm Spine System can provide
a lower profile system in thin patients.
Neurophysiological monitoring, including motor evoked
and sensory evoked potentials, should be used. In SMA
the motor evoked potentials are not usually present in
the more severely involved cases, but SSEPs are especially
helpful for monitoring the upper extremities.
Positioning can be difficult because of upper and lower
extremity contractures. Severe hip flexion contractures can
be present in patients with SMA and DMD. The author
uses a Jackson table with either a sling to support the
knees or a specially constructed trough. Rarely does
the author need to release contractures prior to the
spine surgery.
There have been many posterior instrumentation
techniques used in these patients, including a Luque
system with Galveston pelvic fixation and sublaminar
wires, two rods with cross links, or a unit rod, which is
very stable and allows immediate immobilization without
external support. Pelvic fixation with screws is ideal, as
the screws can be placed in a perfect anatomic position
and the rods connected to the pelvic fixation. Inserting
the L rods using the Galveston pelvic fixation technique
or the unit rod can be difficult, particularly in the patient
with a small, deformed pelvis and/or excessive lordosis.
The author favors pelvic screws inset into the iliac crest,
but the S2 screw techniques (see Chapter 2, Sacropelvic
Fixation Techniques) can be used as well, especially where
skin and soft tissue coverage over the screw heads may be
an issue. Other techniques involve the use of segmental
fixation with all pedicle screws or a combination of pedicle
screws and wire fixation.
75
Surgical Technique
The surgical incision can be curved slightly into the
concavity of the spinal deformity, as the correction will
straighten out the incision. The posterior exposure should
be thorough, out to the tips of the costotransverse
processes and the tips of the lumbar transverse processes.
When using screw fixation to the pelvis, the author
elevates the erector spinae off the dorsal surface of the
sacrum and exposes the posterior superior iliac spine
under the muscle mass. It is relatively easy to divide the
iliac apophysis and expose the outer table of the ilium
with a Cobb elevator and sponge. An inset is created at
Figure 1A. Lateral Pre-op 14 yrs
Figure 1B. AP Pre-op 14 yrs
Figure 1. A-B) 14 year-old patient with Duchenne’s muscular
dystrophy. AP/lateral spine x-ray demonstrating thoracolumbar
scoliosis. C-D) Technique of insertion of the pelvic screws.
E-F) At 16 years old, two years after posterior spinal fusion
and instrumentation, no complications.
76
the site of the screw insertion with a rongeur to reduce
prominence. A blunt, straight pedicle finder is used to
create the hole for the insertion of the pelvic screw. The
image intensifier is used to monitor the course of the
pedicle finder, which is ideally just above the convexity of
the greater sciatic notch. To confirm the position in the
pelvis, the image intensifier can be obliqued to make sure
that the device is between the inner and outer cortices
(Figure 1). Marrow can be aspirated for mixing with
the allograft. The screw is generally 80mm in length in
patients with DMD and much smaller for those with SMA
(in the 45 to 60mm range).
Figure 1C
Figure 1D
Figure 1E. AP Post-op 16 yrs
Figure 1F. Lateral Post-op 16 yrs
The author usually uses a unit rod and removes the
pelvic extension with a rod cutter. The end is smoothed
over with a metal cutting high speed burr. Alternately,
two EXPEDIUM precontoured rods can be used with an
EXPEDIUM modular connector proximally simulating the
cut off unit rod. The author usually uses smaller rods than
the 1/4” steel rods in patients with SMA because of the
poor bone quality. Maintaining adequate thoracic kyphosis
is important for function.
The sublaminar wires are inserted with standard
techniques and an extensive facetectomy is done. A
thin layer of allograft is placed into the gutter bilaterally
and the unit rod or EXPEDIUM modular construct is
fixed to the pelvic screws. The position of the rod to the
pelvis is checked using the image intensifier. The rod is
then cantilevered to the spine and the sublaminar wires
tightened from L5 to the top of the construct. Fixation
with pedicle screws could be used rather than sublaminar
wires (Figure 2).
Figure 2A. Lateral and AP Pre-Op images of 15 year-old with Nemaline
rod myopathy and neuromuscular scoliosis
Figure 2B. Lateral and AP Post-Op images of 15 year-old with Nemaline
rod myopathy and neuromuscular scoliosis
77
Postoperative Care
Outcomes
These patients are typically managed in the intensive
care unit postoperatively and are intubated for at least
24 hours. The patients are mobilized into wheelchairs as
quickly as possible by Nursing in conjunction with physical
therapy. Often, the power chairs need to be modified. In
addition, older patients with DMD have gained so much
height after surgery that their van transportation may
need to be modified.
Long-term results are usually very good in these patients.
The quality of life is improved in both populations, and
the parents and caregivers are usually very satisfied. In
the author’s experience, the combination of the rigid,
fused spine and upper extremity weakness can make selffeeding and self-hygiene more difficult. In DMD, major
complications have been seen and are usually related
to significant blood loss and postoperative pulmonary
problems. In SMA, the early postoperative problems are
similar and are primarily respiratory with atelectasis and
pneumonia. Pseudarthrosis and instrumentation failure
are uncommon with modern instrumentation systems,
especially with the use of iliac screw fixation.
78
79
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-69-000 9/2012 JC
CA #8981A / DJ10054A
Complications in Neuromuscular Scoliosis Surgery
Patrick J. Cahill, MD
Spinal deformity surgery for scoliosis related to neuromuscular
conditions is a challenging area of clinical care due to the
severity of deformity and influence of comorbidities and thus
is fraught with complications. Patients with neuromuscular
conditions often have larger magnitude deformities at the
time of surgery due to the higher incidence of scoliosis in their
population and the increased rate of progression as compared
to idiopathic scoliosis.1 Also, the physiology of the underlying
conditions contributes to the increased rate of complications.
81
The overall rate of complications associated with surgery
for neuromuscular scoliosis has been variably reported but
may be as high as 48%, and many are medical. Mortality
is also higher in than in other populations that undergo
spinal deformity surgery.
Any discussion of the treatment of neuromuscular scoliosis
complications begins with prevention. Careful preoperative
evaluation and maximization of health status is imperative.
Evaluation of nutritional parameters to ensure wound
healing is important. Evaluation of pulmonary status often
requires input from a pulmonary specialist to optimize
the patient’s pulmonary function prior to surgery. Patients
with myelomeningocele and spinal cord injury (SCI) have
increased infection rates due to a variety of factors such
as bladder incontinence, requiring frequent bladder
catheterizations. Urinary tract infections are best avoided
with routine preoperative screening and prevented with
perioperative broad spectrum gram negative and positive
prophylactic antibiotics. Patients with neuromuscular
deformity are often malnourished.
Infection after scoliosis surgery is a potentially serious
complication that requires vigilance in prevention and
diagnosis. The factors contributing to the rate of infection
include the underlying diagnosis and age at surgery, which
we have noted to have an inverse relationship with the
risk of infection.
In our experience, the rate of infection in patients with
myelomeningocele is about 18% and is similar in patients
with SCI, who share many of the same risk factors.
There are a number of contributing factors to the high
infection rate. Patients without sensation in their lower
extremities, buttocks, and/or lower back are at risk for the
development of decubiti which can lead to an infection
either by direct contamination or hematologic spread.
Those patients who lack bowel and bladder control risk
seeding a wound with feces or urine. Furthermore, these
patients develop frequent urinary tract infections which
can spread to implanted instrumentation or a surgical
wound through Batson’s venous plexus on the anterior
spine. Previous surgery in the same area of the body (ex.
closure of a meningocele) can put a patient at risk for
further infection if the same area of the body undergoes
reoperation. The skin can have deep ridges which
cannot be cleaned with routine washing. For this reason,
scrubbing with Hibiclens daily for three days preoperatively
may be advantageous (Figure 1). Also, the patients with
myelomeningocele do not have the normal paramedian
spinal musculature at the level of their defect. The less
vascular scar and adipose tissue and the relative proximity
of the dural sac to the skin surface may be a predisposing
factor to infection. Patients with myelomeningocele
have a high rate of incidental intraoperative duratomy.
If an anterior augmentation is included, additional risks
are incurred including damage to abdominal structures,
increased blood loss, prolonged anesthesia time, and
perioperative malnutrition if staging occurs. A persistent
cerebrospinal fluid leak postoperatively may interfere with
wound healing and be a favorable medium for bacterial
growth.
Figure 1. Hibiclens scrub brush (Johnson & Johnson). The authors
recommend providing the family with a Hibiclens scrub brush to scrub
the patient the evening before surgery.
82
Emerging techniques seem to have benefits in reducing
the burden in infection related to scoliosis in patients
with neuromuscular scoliosis. We have noted a dramatic
decrease in operative infections in neuromuscular scoliosis
with the addition of intra-wound delivery of gentamycin
within the bone graft. With this technique, there seems
to be an incidence of infection of around 4% compared
to approximately 15% in those treated without antibiotic
laden bone graft. We have also noted a similarly dramatic
reduction in infection rates for a series of spinal surgeries
by placing vancomycin powder along the wound
edges. No differences in other complications (including
pseudarthrosis) and no episodes of red man syndrome
(adverse reaction to vancomycin) have been noted.
The treatment of an infection generally involves
surgical intervention, including multiple debridements
and, in about half of all deep infections, removal
of instrumentation. Negative pressure wound
dressings (VACs) have sharply decreased the need for
instrumentation removal. A large proportion of implants
with acute infection and/or wound dehiscence can be
permanently retained with wound vacs. For late onset
infections, we recommend a staged approach with
removal of all metallic implants, inspection of the fusion
mass, and a subsequent period of antibiotic treatment.
If a pseudarthrosis is noted at the removal procedure or
if the fusion mass is immature, or if curve progression is
seen, then we recommend a reinstrumentation procedure
with titanium instrumentation. Clark and Shufflebarger
reported a similar algorithm in late onset infection in
adolescent idiopathic scoliosis.2 Shufflebarger et al have
also reported success with a single surgery instrumentation
exchange procedure with titanium implants for the
eradication of infection.3
Instrumentation related complications are also frequently
encountered. Immobility in patients with neuromuscular
disorders can lead to poor bone mineral density which
contribute to loss of fixation. In addition, patients with
myelomeningocele have missing or dysplastic posterior
elements yielding a greatly decreased surface area
of bone for fusion. In our experience, pseudarthrosis
occurs in about 5% of patients undergoing posterior
spinal fusion for neuromuscular scoliosis for most
conditions but may be as high as 16% in patients with
myelomeningocele undergoing posterior-only fusion.
Treatment of pseudarthrosis generally requires revision
instrumentation and fusion, although asymptomatic
pseudarthrosis without deformity progression may only
require observation The most common location of
pseudarthrosis is at L5-S1. Thus, we feel that consideration
should be given to anterior interbody support at the
lumbosacral junction in patients with a high BMI or who
will be otherwise likely to put high stress on that level –
e.g. an ambulatory patient with neuromuscular disorder.
Current techniques with secure segmental fixation and
abundant bone graft should ensure a relatively low rate of
pseudarthrosis.
Tsuchiya et al demonstrated a higher rate of
pseudarthrosis in adult patients with scoliosis fused to the
pelvis who did not have interbody support than those who
did.4 There are many factors predisposing to nonunion:
age, diagnosis, BMI, etc. We suggest a low threshold for
consideration of interbody support, especially at L5-S1
with the risk factors.
83
We do not routinely use bone morphogenetic protein
(BMP) in primary surgeries mainly due to cost concerns
and the lack of superiority data in this population. There
are many bone graft substitutes available which may
be utilized to enhance the fusion. These may include
demineralized bone matrix and allograft crushed cortical
or cancellous bone.
Implant prominence is a particular problem in surgery for
patients with neuromuscular scoliosis. Our rate of revision
surgery for prominent iliac screws is around 10%. One
promising technique that may avoid the prominence of
iliac fixation is a transsacral iliac screw trajectory. Described
by Paul Sponseller, this screw takes a trajectory starting on
the dorsal sacrum at S2, traverses through the sacroiliac
joint, and rests just superior to the sciatic notch within
the ilium.5 This technique (described in “Sacropelvic
Fixation Techniques” chapter) results in a much lower
profile implant and also provides fixation in-line with the
other points of fixation, obviating the need for off-set
connection devices.
Neurologic injury is a potentially devastating complication
to an already impaired population. Loss of bladder
control and protective sensation can compound the
risk of decubiti in non-ambulatory patients. Fortunately,
neurologic injury rarely occurs in this population. In our
experience, incidence of neurologic injury in patients
with neuromuscular scoliosis is lower than in those
with idiopathic scoliosis. With the exception of patients
with complete spinal cord injuries, intraoperative
neurophysiologic monitoring is usually feasible.
Ashkenaze et al reported that they were able to obtain
somatosensory evoked potential signals reliably in
72% of their neuromuscular patients.6 They reported
intraoperative signal loss in 40% of the cases that were
able to be monitored. Only three neurologic injuries
were seen in their series of 104 patients. Some surgeons
consider motor evoked potential (MEP) monitoring to be
contraindicated in patients with shunts and/or a history
of seizures. However, other authors have reported no
difficulties or complications in patients with neuromuscular
scoliosis undergoing fusion who had a seizure history and
84
intraoperative MEP monitoring. An additional neurologic
concern is ventricular shunt malfunction. Shunts are
common in patients with CP and myelomeningocele. As
these shunts are gravity dependent, prolonged periods of
time not in an upright position (intra- and postoperatively)
may lead to shunt malfunction. Preoperative evaluation
by the patient’s neurosurgeon is advised. We also keep a
slight tilt in the bed starting immediately postoperatively.
Lastly, in rare cases it may be necessary to perform
intraoperative intracranial pressure monitoring and
periodic intraoperative decompression via removal of
cerebrospinal fluid through the cranial shunt portal.
An additional concern in patients with seizure disorder is
the potential for excessive bleeding due to a coagulopathy
related to antiseizure medication. Divalproex sodium
(Depakote) is associated with increased intraoperative
blood loss. We recommend attempting to bridge
perioperative pharmacologic seizure management with
other medications under the guidance of a neurologist.
Many patients with CP have intrathecal catheters. The
local delivery of intrathecal baclofen reduces the dose
required to decrease spasticity compared to oral baclofen,
thus minimizing unwanted side effects. Unfortunately,
these catheters create difficulties at the time of spinal
fusion surgery. The catheters traverse the surgical field and
are at risk of being damaged or dislodged during surgery.
Also, they are placed through laminotomies, the presence
of which can increase the risk of incidental durotomy
and cerebrospinal fluid leak at the time of surgery. The
likelihood of inadvertently damaging the baclofen pump
catheter is high. Thus, it is important that at the time of
surgery, a baclofen catheter repair kit be made available
for every patient with a baclofen pump.
The authors feel strongly that spinal deformity surgery
for neuromuscular scoliosis can have a positive impact
on duration and quality of life. However, complications
are frequent and require multidisciplinary attention
to detail in perioperative planning, intraoperatively
and postoperatively. A high level of suspicion must be
maintained.
85
86
References
1. Kalen V, Conklin MM, Sherman FC. Untreated scoliosis
in severe cerebral palsy. J Pediatr Orthop 1992;12:33740.
2. Clark CE, Shufflebarger HL. Late-developing infection
in instrumented idiopathic scoliosis. Spine (Phila Pa
1976) 1999;24:1909-12.
3. Shufflebarger HL, Asghar J, Morales DC. Solution for
late arising infection in stainless steel spinal implants
placed for deformity: exchange for titanium implants.
Scoliosis Research Society annual meeting, Louisville,
KY, September 14-17, 2011.
4. Tsuchiya K, Bridwell KH, Kuklo TR, et al. Minimum
5-year analysis of L5-S1 fusion using sacropelvic
fixation (bilateral S1 and iliac screws) for spinal
deformity. Spine (Phila Pa 1976) 2006;31:303-8.
5. Chang TL, Sponseller PD, Kebaish KM, et al. Low
profile pelvic fixation: anatomic parameters for sacral
alar-iliac fixation versus traditional iliac fixation. Spine
(Phila Pa 1976) 2009;34:436-40.
6. Ashkenaze D, Mudiyam R, Boachie-Adjei O, et al.
Efficacy of spinal cord monitoring in neuromuscular
scoliosis. Spine (Phila Pa 1976) 1993;18:1627-33.
87
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-62-000 9/2012 JC
CA #8783 / DJ10055A
DePuy Spine, Inc.
325 Paramount Drive
Raynham, MA 02767
USA
Tel: +1 (800) 227-6633
www.depuysynthes.com
©DePuy Spine, Inc. 2012. All rights reserved.
DF29-60-100 10/2012 JC
CA #9207A / DJ10290A