Spinal Infection

Spinal Infections

Dr Harry Sunaryo SpOT

Before the introduction of modern
antibiotic therapy, mortality in
patients with vertebral osteomyelitis
was as high as 25%.1 Antibiotic
therapy combined with surgical
débridement and stabilization has
decreased mortality to less than 5%
to 15%.2-4 Early diagnosis also has
improved outcomes by facilitating
rapid initiation of antibiotic treatment
and preventing abscess formation,
structural instability, and neurologic
deterioration.
Spinal infections are evaluated
according to their location, the pathogen
or pathogens involved, route of
the infection, age of the patient, and
immune status of the host. The location
of the infection may involve the
osseous vertebra, the intervertebral
disk, the epidural space, or the surrounding
soft tissues. The pathogens
are usually either bacterial or fungal;
however, the widespread use of
broad-spectrum antibiotics and the
increasing number of immunocompromised
patients have led to infections
with unusual organisms.
A systematic approach must be
taken in the diagnosis and treatment
of each type of spinal infection.
The presentation and efficacy
of the various elements of the initial
evaluation differ markedly for acute
hematogenous infection, granulomatous
spinal infection, pediatric
hematogenous diskitis, epidural
abscess, and postoperative spinal
infection.
Pathophysiology of Spinal
Infection
Pyogenic vertebral osteomyelitis is a
bacterial infection that can arise
from a number of sources—direct
inoculation, contiguous spread from
an adjacent infection, or hematogenous
seeding. Direct inoculation
can result from penetrating injuries
or from percutaneous or open spinal
procedures (eg, chemonucleolysis,
diskography, diskectomy) done on
the intervertebral disk. Local spread
of bacteria or fungi can occur following
intra-abdominal and retroperitoneal
abscesses. Although
local spread from direct inoculation
of bacteria into the spinal canal is
likely to become more prevalent as
the number of spinal procedures
increases, hematogenous seeding of
infection is still by far the most common
mechanism of spinal infection.
Potential sources of pathogenic
organisms include skin and softtissue
infections, infected vascular
access sites, and the urinary tract.
The two major theories for hematogenous
dissemination are the
venous theory and the arteriolar
theory. Batson5 developed the
venous theory using both live animal
and human cadaveric models.
He demonstrated retrograde flow
from the pelvic venous plexus to the
perivertebral venous plexus via
valveless meningorrhachidian veins.
In the arteriolar theory, Wiley and
Trueta6 proposed that bacteria can
become lodged in the end-arteriolar

Spinal infections can occur in a variety of clinical situations. Their presenta-
tion ranges from the infant with diskitis who is unwilling to crawl or walk to
the adult who develops an infection after a spinal procedure. The most common
types of spinal infections are hematogenous bacterial or fungal infections, pedi-
atric diskitis, epidural abscess, and postoperative infections. Prompt and accu-
rate diagnosis of spinal infections, the cornerstone of treatment, requires a high
index of suspicion in at-risk patients and the appropriate evaluation to identify
the organism and determine the extent of infection. Neurologic function and
spinal stability also should be carefully evaluated. The goals of therapy should
include eradicating the infection, relieving pain, preserving or restoring neuro-
logic function, improving nutrition, and maintaining spinal stability.

Both mechanisms are likely
significant in the establishment of
an infectious focus in the spinal column.
In the cervical spine, an extensive
prevertebral pharyngeal
venous plexus also may act as a
conduit for the spread of bacteria.7
Local spread of infection can
occur in a number of ways. Once
the infection is established adjacent
to the end plate of one vertebral
body, it can rupture through that
structure into the adjoining disk and
infect the next vertebral body. The
disk material is relatively avascular
and is rapidly destroyed by the bacterial
enzymes . In the cervical
spine, if the infection penetrates
the prevertebral fascia, it can extend
into the mediastinum or into the
supraclavicular fossa, markedly
increasing the extent and severity of
the process. From the lumbar spine,
abscess formation may track along
the psoas muscle and into the buttock
(piriformis fossa), the perianal
region, the groin, or even the popliteal
fossa. The extension of infection
from the vertebral body or disk
into the spinal canal may result in an
epidural abscess or even bacterial
meningitis. Destruction of the vertebral
body and intervertebral disk
can potentially lead to instability
and collapse. In addition, with collapse
of the vertebral body, infected
bone or granulation tissue may be
retropulsed into the spinal canal,
causing neural compression or vascular
occlusion. With pyogenic
osteomyelitis, the lumbar spine is
more commonly affected than the
thoracic or cervical spine.8
The pathogenesis of spinal infection
differs markedly between children
and adults because of anatomic
differences in the vascular anatomy
of the vertebrae. In children, vascular
channels cross the cartilaginous
growth plate and end within the
nucleus pulposus. These channels
provide pathways for direct inoculation
of organisms into the avascular
nucleus pulposus. Since these
vascular channels are not present in
adults, the direct seeding of the disk
does not occur, but rather spreading
occurs by direct extension with rupture
of the infective focus through
the end plate into the disk.
Neurologic deterioration can be a
devastating consequence of spinal
infection. A number of different
factors can cause neural deficit.
Direct spread of infected material
into the spinal canal can produce an
epidural abscess that may compress
the neural elements or cause thrombosis
or infarction of the regional
vascular supply to the spinal cord.
Direct hematogenous spread rarely
results in epidural abscess without
the presence of associated diskitis or
osteomyelitis. Pathologic fracture
can occur, with associated extrusion
of either infected material or bony
elements into the spinal canal.
Kyphosis and/or spinal instability
resulting from destruction of the
disk, vertebral bone, and posterior
stabilizing structures can cause
neural impingement. Eismont et al4
reported several additional risk factors
that predispose to neurologic
deterioration: diabetes, rheumatoid
arthritis, steroid use, advanced age,
a more cephalad level of infection
(ie, high thoracic or cervical), and
infection with Staphylococcus species.
The pathophysiology of granulomatous
spinal infection differs from
that of pyogenic infections. The
most common form of granulomatous
disease of the spine is caused
by Mycobacterium tuberculosis (Pott’s
disease). Although endemic in
many developing countries, tuberculosis
(TB) was nearly eradicated in
A B
56-year-old man presented with severe back pain following a urologic procedure.
He had an elevated ESR but no leukocytosis. A, T1-weighted sagittal MR image of
the lumbar spine shows severe edema of the L3-4 disk and adjacent soft tissues. B, T2-
weighted sagittal MR image shows high signal intensity in the L3-4 disk and adjacent vertebral
bodies, consistent with pyogenic diskitis and osteomyelitis. Cultures obtained from
a CT-guided biopsy of the disk space grew Staphylococcus aureus.
Spinal Infections; however, there
has been a recent resurgence of TB
with resistant strains and in patients
with human immunodeficiency
virus (HIV). Although less than 10%
of patients with TB have skeletal
involvement, 50% of the skeletal involvement
occurs in the spine.
Depending on the series, between
10% and 61% of patients present
with or develop a neurologic
deficit.9
With TB, the primary route of
infection to the spine is hematogenous
from a pulmonary or genitourinary
source, although direct
spread from adjacent structures can
occur. Three major patterns of spinal
vertebral body involvement
have been documented: peridiscal,
central, and anterior.10 The most
common form, peridiscal, occurs
adjacent to the vertebral end plate
and spreads around a single intervertebral
disk. Extension to the adjacent
vertebra occurs as the granulomatous
abscess material tracks
beneath the anterior longitudinal
ligament. Unlike the situation in
pyogenic infections, the intervertebral
disk is usually spared. Central
involvement occurs in the middle
of the vertebral body and can be
mistaken for a tumor. Destruction
of the vertebral body will then lead
to spinal deformity. Anterior
involvement begins beneath the
anterior longitudinal ligament,
causing scalloping of the vertebral
body . In contrast with
peridiscal involvement, which
affects a single motion segment,
anterior involvement can produce
a spinal abscess that extends over
multiple levels. Primary involvement
of the posterior structures is
uncommon. Regionally, the thoracic
spine is most often involved,
followed by the lumbar spine and
cervical spine. Paraspinal extension
with abscess formation is common
and can occur at any level.
Spinal infections can be classified
as acute, subacute, or chronic depending
on the duration of symptoms.
Symptoms that have persisted
for <3 weeks are acute; those lasting
from 3 weeks to 3 months are subacute.
Chronic infections last >3
months and either are caused by
indolent organisms, are granulomatous
in nature, or are incompletely
treated (eg, infections with resistant
organisms, or the presence of foreign
material in the area of infection).
Clinical Evaluation
Pyogenic Vertebral
Osteomyelitis
Pyogenic vertebral osteomyelitis
is more common in males than in
females and also more common in
elderly populations.2,11 However,
the incidence of infection is increasing
in younger age groups in populations
with intravenous drug abuse
or immunocompromise after organ
transplantation or chemotherapy.
Accordingly, spinal infection should
be considered in the differential
diagnosis of acute-onset spinal pain
in patients older than 50 years or
with diabetes, rheumatoid arthritis,
immunocompromise (from medical
illness or pharmacologic immunosuppression),
or a history of intravenous
drug abuse.
The clinical presentation of vertebral
osteomyelitis depends on the
location of the infection, the virulence
of the organism, and the immune
status of the host. Back or
neck pain is the most consistent
symptom of pyogenic infection.
Observed in >90% of patients, the
pain is often quite severe and is
associated with notable paraspinal
muscle spasm. The pain may occur
A B C D
Figure 2 A 33-year-old woman presented with back pain of several months’ duration. A, Anteroposterior radiograph shows collapse of
the vertebral body and paraspinal soft-tissue shadow (arrowheads). B, Lateral radiograph also shows collapse and interior scalloping
(arrow). C, Sagittal T1-weighted MR image shows a large anterior abscess, extensive vertebral body involvement, and relative sparing of
disk spaces. D, The patient underwent CT-guided biopsy and aspiration with placement of a pigtail catheter for 1 week to drain this
abscess. She underwent anti-TB treatment for 1 year, with resolution of pain and no development of deformity.
at night and is usually present regardless
of activity level. Radicular
leg or arm pain is less common but
may be present with neurologic
involvement, which occurs in less
then 10% of patients. Fevers are
documented in approximately
50% of the affected population.12
Weight loss is common but may
not be easily recognized by patients
because it may occur slowly over a
period of weeks to months before
the infection is diagnosed and
treated.
The presence of other signs or
symptoms depends on the extent of
the infectious process. A patient
with a psoas abscess may have pain
with hip extension. Cervical abscess
formation may lead to torticollis or
dysphagia. Radiculopathy, myelopathy,
or even complete paralysis can
occur with neural compression as a
result of abscess, instability, or
spinal deformity. Direct spread of
the infection into the epidural space
can cause meningitis.
Gram-positive organisms are responsible
for the majority of vertebral
column infections in both adults
and children, with Staphylococcus
aureus accounting for >50%. Infection
with gram-negative organisms
such as Escherichia coli, Pseudomonas,
and Proteus may occur following
genitourinary infections or procedures.
Intravenous drug abusers are
also prone to Pseudomonas infections.
Anaerobic infections may be encountered
in patients with diabetes
or following penetrating trauma.
Low-virulence organisms such as
coagulase-negative staphylococci
and Streptococcus viridans may cause
indolent infections. These organisms
may not be detected unless
blood cultures are held for more
than 10 days and should not be disregarded
as contaminants in the
presence of clinical infection. Salmo-
nella, presumably from an intestinal
source, can cause vertebral osteomyelitis
in children with sickle cell
anemia.
Laboratory Studies
Laboratory studies may be useful
but are usually nonspecific. The
white blood cell count will be elevated
in approximately half the
cases of acute pyogenic osteomyelitis
but typically is normal in the
presence of subacute or chronic
infection. The erythrocyte sedimentation
rate (ESR) is a more sensitive
test and is elevated in >90% of patients.
The C-reactive protein (CRP)
level, an acute-phase reactant with a
much quicker normalization time,
may be more helpful in following
the course of treatment than the
ESR. A rapid decrease in the CRP
level indicates an adequate response
to treatment and can help determine
when to switch from intravenous to
oral antibiotics. Blood cultures may
be negative in up to 75% of patients,
particularly if the infection involves
a low-virulence organism. It is
extremely important to delay antibiotic
therapy until appropriate cultures
have been obtained unless the
patient is septic and critically ill.
Even then, blood and urine cultures
should be obtained before the administration
of antibiotics.
Evaluation of laboratory measurements
for malnutrition is as important
as the diagnostic tests that detect
the presence of infection. Weight
loss >30% of ideal body weight during
the course of the infection indicates
severe malnutrition. Other
laboratory measurements that are
associated with severe malnutrition
include a serum albumin level of <3
g/dL, serum transferrin measurement
of <150 μg/dL, and an absolute
lymphocyte count of <800/mL. Although
it is a measurement less commonly
used in orthopaedics, a 24-
hour urinary creatinine excretion of
<10.5 mg in men or <5.8 mg in
women indicates a negative nitrogen
balance associated with malnutrition.
Biopsy
The definitive diagnosis of spinal
pyogenic osteomyelitis requires
identification of the organism
through a positive blood culture or
from a biopsy and culture of the
infected site. Blood cultures may be
diagnostic in as few as 25% to 33%
of cases.2 Cultures taken during
fever spikes may provide better
diagnostic results. Biopsy of the
infected area is often necessary to
initiate the appropriate antibiotic
regimen. Other sources of obvious
infection, such as the urine, must
also be cultured. Spinal biopsies
may be done percutaneously, using
computed tomography (CT) or fluoroscopy
to localize the focus of infection.
The accuracy of closed biopsy
techniques varies and has been
reported to be about 70%.13 Key factors
may be insufficient tissue
retrieval or administration of antibiotics
prior to biopsy. A core sample
obtained from a Craig biopsy needle
for bone or a TruCut (Baxter Travenol,
Deerfield, IL) or similar needle
for soft tissue is preferable to fineneedle
aspiration except when an
abscess cavity is present. Antibiotics
must not be started until the biopsy
is done and sufficient tissue is obtained
for culture, gram stain, and
histology. If a diagnosis is not confirmed
on the first attempt, a second
closed biopsy should be considered
before open biopsy is done.
An open biopsy is indicated
when needle biopsy fails to identify
an organism, when the infection is
inaccessible by standard closed
techniques, or when there is marked
structural damage with neurologic
compromise. Open biopsies are
diagnostic in >80% of cases.14 Minimally
invasive techniques, such as a
laparoscopic or thoracoscopic approach,
may be considered when that
approach is appropriate to decrease
the morbidity of the procedure.
Biopsies should be sent for gram
stain, acid-fast stain, and aerobic,
anaerobic, fungal, and TB cultures.
Bacterial cultures should be maintained
for 10 days to detect low-virulence
organisms. Histologic studSpinal
Infections

ies also should be done, if possible,
to detect metabolic or neoplastic
processes. If tissue is available,
pathologic examination should be
conducted to differentiate between
acute and chronic infection and to
help detect the presence of acid-fast
bacilli and fungal elements. The
development of polymerase chain
reaction as a diagnostic tool has facilitated
rapid detection of the infecting
agent, especially when indolent
and low-virulence organisms are
involved.15,16 However, technical
problems with cross-contamination
can lead to false-positive results.
Tuberculosis
The clinical presentation of a
patient with a tuberculous spinal
infection is highly variable. As with
pyogenic infections, back pain is the
most common symptom; however,
it is usually less severe than in a
pyogenic infection. Patients with
chronic infection also may experience
weight loss, malaise, fevers,
and night sweats. Kyphotic deformities,
neurologic deficits, or cutaneous
sinuses may occur after prolonged
or very severe infections.
Neurologic deficit can occur from
epidural extension of the tuberculous
infection, from destruction of
bone with retropulsion of infected
material into the spinal canal, or
from progressive kyphotic deformity.
Elderly patients appear to be
at higher risk for developing a neurologic
deficit. The differential diagnosis
of spinal infection includes
primary and metastatic tumors; infections
with atypical bacteria such
as Actinomyces, Nocardia, and Bru-
cella; infections with atypical mycobacteria;
and fungal infections such
as coccidioidomycosis, blastomycosis,
cryptomycosis, candidiasis, and
aspergillosis. Immunocompromised
patients are at risk for developing
infections with atypical mycobacteria.
Fungal infections also have
become more common with the
increasing use of broad-spectrum
antibiotics, especially in combination
with central venous catheters for
parenteral nutrition .
Suspicion of a mycobacterial
infection is the basis for establishing
the diagnosis. Patients from Southeast
Asia or South America, prison
populations, and frequenters of
homeless shelters are at high risk for
contracting TB. A patient with a
family member or household contact
with TB also should be considered
as at high risk. Laboratory tests
are usually nonspecific. A leukocytosis
may or may not be present.
The ESR may be normal in up to
25% of cases. Although the purified
protein derivative skin test can help
detect active infection or past exposure
to TB, the test is not fully reliable
because of false-negative results
that can occur in the malnourished
and the immunocompromised.
Polymerase chain reaction for detection
of tuberculous infection holds
great promise for a faster diagnosis.
Pediatric Diskitis
The highly variable clinical presentation
of a child with diskitis may
lead to delays in recognition and
diagnosis. Active children may
often associate the onset of pain with
some activity or minor trauma. In
the absence of systemic symptoms
of infection, further workup is necessary
if the pain does not resolve in
1 to 2 weeks. In general, however,
vertebral infection should be suspected
when the child has a lowgrade
fever and pain, refuses to bear
weight, or assumes a flexed position
of the spine. The patient also may
complain of abdominal pain. These
nonspecific findings are more common
in children over the age of 5
years.17-19 In contrast, infants are
more likely to be systemically ill.
Older children are more likely to be
able to identify the spine as the
source of pain. Although uncommon,
these same symptoms can be
observed with spinal tumors in children,
such as Ewing’s sarcoma.
The white blood cell count may
or may not be elevated, but the ESR
is usually mildly elevated and the
CRP level, markedly elevated.
Infants typically will demonstrate a
leukocytosis and elevated ESR.20
Blood cultures can be positive in
up to 50% of cases.19
Acute infections are more likely
to yield positive blood cultures.19
Certainly the child who appears ill
and febrile should have all possible
sources of infection cultured. If a
biopsy is needed, it can be done
under CT guidance; a 60% to 70%
yield rate for infectious lesions can
be expected.21 If a trial of antibiotics
was initiated prior to biopsy without
response, antibiotics should be
suspended for 3 to 4 days before the
procedure to ensure greater accuracy
from the cultures.
A B
A 40-year-old woman with
rheumatoid arthritis and chronic steroid
use developed severe back pain and paraplegia
after treatment with broad-spectrum
antibiotics for necrotizing fasciitis. A,
Lateral radiograph of the lumbar spine
shows bony destruction of the end plates
of L2 and L3. B, T2-weighted sagittal MR
image of the lumbar spine demonstrates
diskitis and vertebral osteomyelitis at L2-3,
with severe canal stenosis from an epidural
collection (arrowhead). Cultures taken at
the time of anterior débridement were consistent
with a Candida infection. The patient
obtained pain relief and improvement
in motor function after aggressive anterior
débridement and reconstruction with an
autogenous tricortical iliac graft and 6
weeks’ administration of intravenous liposomal
amphotericin B.
Epidural Abscess
The presence of a spinal epidural
abscess is usually associated with
the occurrence of diskitis or vertebral
osteomyelitis. Rarely does an
epidural abscess occur hematogenously
without spinal involvement.
This condition is caused by direct
seeding of bacteria into the epidural
venous plexus, in contrast with the
more common route of local extension
from adjacent disk or bone. In
the absence of diskitis or vertebral
osteomyelitis, an epidural abscess
can be difficult to diagnose and can
progress rapidly, with devastating
consequences; prompt diagnosis
and early treatment are critical in
these rare cases. Risk factors for the
development of epidural abscess
include history of intravenous drug
use, diabetes, trauma, obesity, percutaneous
or open procedures (eg,
spinal surgery, nerve or epidural
block, or diskography), HIV, and
renal failure.22-26 Patients may present
with back pain, progressive
neurologic deficit, or fever. Although
leukocytosis may not be
present, the ESR is almost always
elevated.
Radiographic Evaluation
Imaging studies are crucial to localize
the infection, assess the extent of
involvement, and determine the
response to treatment. Radiographs
may demonstrate progressive osteolysis
and end plate destruction,
often best seen on the anteroposterior
view (Fig. 2, A). As the disease progresses,
the disk space narrows and
eventually collapses (Fig. 3). Plain
radiographs, however, may not
demonstrate abnormal findings for
up to several weeks after the process
has begun. Soft-tissue extension
must be suspected in the presence
of an abnormal psoas shadow,
widening of the mediastinum , or enlargement of the retropharyngeal
soft-tissue shadow. The
presence of gas in the soft tissues
suggests an infection with an anaerobic
organism.
In contrast with pyogenic infections,
skeletal radiographs in a
tuberculous infection often demonstrate
vertebral destruction with
relative preservation of the disk
spaces. As the infection progresses,
the disk is also destroyed and a
kyphotic deformity may be present,
especially in the thoracic spine. A
chest radiograph always should be
obtained to assess for active pulmonary
disease.
In pediatric diskitis, radiographs
of the spine should be assessed for
disk space narrowing, end plate erosions,
bony destruction, and paravertebral
soft-tissue swelling. These
changes may not occur for several
days or weeks after onset of symptoms.
They usually persist, eventually
leading to disk space narrowing
or autofusion.18,27 Although late
kyphosis is rarely seen in pediatric
spinal infections, a notable exception
is infantile osteomyelitis, which generally
is associated with more initial
bony destruction and resembles congenital
kyphosis in late stages.20
Radionuclide studies can be
much more sensitive than radiographs
in detecting early infections.
Technetium 99m bone scintigraphy
is sensitive (~90%) but nonspecific,
especially in adults with degenerative
joint disease.28 Because the
study is dependent on local blood
flow, false-negative results have
occurred in areas of relative ischemia
in very young and elderly patients.
In pediatric vertebral osteomyelitis,
the technetium 99m bone
scan is positive in 74% to 100% of
cases,17,19 facilitating earlier diagnosis
of diskitis in children. Wenger et
al19 showed that use of bone scans
allowed diskitis to be diagnosed an
average of 8.3 days earlier than
without.
When used in conjunction with
technetium 99m scans, gallium 67
citrate scans have high sensitivity
and specificity in detecting foci of
infection. The tracer, an analog of
ferritin, is secreted by leukocytes at
sites of infection. Gallium scans also
normalize during the recovery phase
and may be used to follow treatment
response. This test, however, may
not be effective in leukopenic patients
and may not detect low-virulence
organisms. Indium 111-labeled
scans have a poor sensitivity in vertebral
osteomyelitis (17%) and are
not recommended.29
CT is useful in delineating the
extent of bony destruction and softtissue
extension and is helpful in preoperative
planning. However, the
status of the neural elements cannot
be accurately assessed without the
use of myelographic dye, which is
contraindicated in suspected infection
because it places the patient at
risk for developing meningitis or
arachnoiditis. Although the CT scan
with intravenous contrast also can
demonstrate soft-tissue extension,
distinction between abscess and
granulation tissue may be difficult.
Magnetic resonance imaging
(MRI) is the modality of choice in
the diagnosis and evaluation of
spinal infections because it provides
excellent imaging of the soft tissue,
neural elements, and inflammatory
changes in the bone (Figs. 2, B and
3, C). MRI has an extremely high
sensitivity (96%) and specificity
(93%) in detecting infections of the
vertebral column.28 It is noninvasive,
allows detection of paravertebral
and epidural extension, and
clearly visualizes neurologic structures.
T1-weighted sequences
demonstrate decreased signal intensity
in both the vertebral body
and disk from edema. T2-weighted
images show increased signal intensity
in both the vertebral body and
disk with loss of the normal intranuclear
cleft .
The administration of gadolinium
in combination with MRI improves
resolution and allows an infectious
process to be distinguished from
Spinal Infections

degenerative changes of the end
plate and intervertebral disk .
The vascular-based enhancement
also allows differentiation of an
epidural granulation from an epidural
abscess. An epidural mass
may be isointense or hypointense
on T1-weighted images, shows high
signal on T2-weighted images, and
may show peripheral enhancement
visible with gadolinium.30 Short T1
inversion recovery sequences often
can help to differentiate an infection
from other pathologic entities. Even
with MRI, however, granulomatous
infections can be difficult to distinguish
from tumors of the spine.
Thus, a biopsy is often required to
make a definitive diagnosis.

Treatment
Pyogenic Infections
The goals for treatment of spinal
infections should be to establish a
diagnosis and identify the pathogen,
eradicate the infection, prevent or
minimize neurologic involvement,
maintain spinal stability, and provide
an adequate nutritional state
to combat infection. Establishing
a diagnosis and identifying the
pathogen is of primary importance.
Once the organism has been identified,
intravenous antibiotic therapy
should be initiated according to the
culture results and sensitivities. A
course of 2 to 6 weeks of parenteral
antibiotics is usually recommended.
This is followed by a course of oral
antibiotics, depending on the virulence
of the organism, susceptibility
of the host, and other factors, such as
retained hardware. Conversion to
oral antibiotics should be made only
with clinical improvement, normalization
of the ESR and CRP level, or
resolution of the infection as demonstrated
in imaging studies.
In addition to antibiotic therapy,
immobilization, rest, and proper nutrition
are recommended. Molded
contact braces are effective in the
lumbar region, whereas a halo or a
rigid cervicothoracic orthosis may
be required for cervical osteomyelitis.
Immobilization of the affected
area aids in pain relief and helps
prevent deformity.
Surgery is indicated in five circumstances:
to obtain a tissue diagnosis
after a failed closed needle
biopsy or from a location inaccessible
by closed methods; for drainage
of an abscess that is causing sepsis
or neurologic deficit; to treat neurologic
deficit secondary to compression
either by the infection (abscess
or granulation) or structural destruction;
for structural instability or
deformity; or for failure of medical
management to reduce persistent
symptoms or elevated laboratory
measurements.
The location of the infection and
the intended purpose of the surgery
often dictate the surgical approach.
Because the majority of these infections
involve the vertebral body and
the disk, an anterior approach is
most commonly used to maximize
access to the infected tissue. A posterolateral
approach to the thoracic
spine may be considered in certain
instances, or a costotransversectomy
if only culture, biopsy, or abscess
drainage is necessary. Because these
and endoscopic approaches avoid
thoracotomy, they may cause less
morbidity in the medically fragile
patient.
If an anterior approach is used
for débridement and decompression
of the spinal canal, reconstruction
should be done with an autogenous
structural graft, such as tricortical
iliac crest or middle third of the
fibula. Iliac crest is preferable because
of the abundant amount of
cancellous bone. Fresh-frozen allografts
in combination with autogenous
bone may be considered for
structural support, but structural
autogenous bone grafts are preferred.
Vascularized bone grafts
have gained popularity during the
last decade because of their intrinsic
blood supply and faster rate of incorporation.
In the thoracolumbar
junction, a vascularized rib graft
may be used, and in the lumbar
spine, vascularized rib or iliac
grafts.31-34 Recently, titanium surgical
mesh filled with autogenous
bone has been used as an alternative
to structural autogenous graft.
Depending on the degree of preoperative
kyphosis and length of the
reconstruction, a posterior fusion
A B C
A 38-year-old man with HIV and a CD4 cell count of 20 presented with back
pain of several weeks’ duration and no radiculopathy. A, T1-weighted sagittal MR image
shows edema at the L5-S1 disk space and adjacent end plates. The asterisk (*) indicates an
epidural collection consistent with an epidural abscess. B, T1-weighted gadoliniumenhanced
sagittal MR image shows uptake at the L5-S1 disk space and the epidural collection.
C, T2-weighted sagittal MR image shows no notable canal compromise by the anterior
collection. However, there is severe destruction of the adjacent bone of L5 and S1.
*
with instrumentation may be required
to adequately stabilize the
spine. This is usually undertaken 1
to 2 weeks after the initial surgical
débridement. The staging of the
procedures allows for an interval of
intravenous antibiotics and optimization
of medical and nutritional
parameters before placement of the
instrumentation.
Hyperalimentation is an effective
way to maximize the patient’s nutritional
status before and after surgery
and between stages. The infection
places the patient in a catabolic state
because of metabolic losses that have
occurred before the diagnosis of infection
is made. The goal of nutritional
supplementation is to restore
the patient to the premorbid nutritional
status. Nutrition consultation
and monitoring of laboratory measurements
are helpful in reaching a
positive nitrogen balance. These
include achieving a serum albumin
level >3 g/dL, an absolute lymphocyte
count >800/mL, and a 24-hour
urine creatinine excretion >10.5 mg
in men and >5.8 mg in women.
Tuberculosis
Once the diagnosis of a tuberculous
infection is established, aggressive
treatment is necessary to
eradicate the infection. A four-drug
regimen of isoniazid, rifampin,
ethambutol, and pyrazinamide is
used as first-line therapy for 6
months. The response to treatment
is assessed by routine clinical examinations
and radiographs. The
emergence of multidrug-resistant
mycobacteria will provide further
challenges in the treatment of these
infections in the future.
Indications for surgery in tubercular
infections are the same as for
pyogenic infections. The most common
surgical technique, the Hong
Kong procedure, involves débridement
of infected bone, decompression
of the spinal canal, and correction
of the kyphotic deformity using
structural grafting35 . Additional
posterior fusion with instrumentation
also may be required.
The second procedure can be either
staged or done on the same day,
depending on the tolerance of the
patient. Autogenous iliac crest or
fibula is ideal for structural grafting.
Rib graft alone has been shown to
be inadequate unless a vascularized
rib is used to accelerate the rate of
incorporation. The Hong Kong procedure
is preferred over anterior
débridement alone because the
addition of an anterior strut corrects
and prevents progressive kyphotic
deformity. Laminectomy without
adjunctive stabilization is contraindicated
because damage to the
posterior structures in the presence
of weakened anterior structures will
lead to progressive kyphosis and
neurologic injury.
Failure of medical treatment or
development of neurologic deficit is
A B
C D
A 22-year-old woman presented with a long history of back pain.
Anteroposterior (A) and lateral (B) radiographs show erosion and partial collapse of the
T12 vertebral body (arrow). C, T1-weighted MR image demonstrates extensive anterior
and posterior column involvement. Because of atypical MR image findings, a posterior
biopsy was performed, which revealed TB. D, Postoperative lateral radiograph. Because
of partial collapse and extensive involvement, the patient underwent anterior reconstruction
using autogenous rib graft.

a clear indication for surgical
débridement, decompression, and
stabilization. Early decompression
will maximize the patient’s functional
recovery. A more chronic
neurologic deficit due to cord compression
over structural deformity
also may be treated with decompression
and stabilization. However,
the prognosis for neurologic
recovery in the face of chronic
deficits is not as optimistic.
Pediatric Diskitis
Whether diskitis in children is
infectious or inflammatory in origin
remains controversial. Although
the recommended treatment will
vary depending on the suspected
origin, immobilization with casting
or bracing is uniformly recommended.
The use of antibiotics has been
controversial, with satisfactory
results reported in several studies
regardless whether a patient
received antibiotics. Scoles and
Quinn18 reported that all patients
were asymptomatic at the time of
hospital discharge, whether or not
antibiotics were administered. In
addition, none of these patients had
a relapse. In contrast, Ring and
Wenger36 observed that patients
treated with intravenous antibiotics
for at least 6 days had a more rapid
resolution of symptoms and the
lowest likelihood of developing
recurrent symptoms. Oral antibiotics
or no treatment were more
likely to lead to prolonged or recurrent
symptoms. Based on their
experience, they felt that a short
course of parenteral antibiotics was
more likely to result in rapid relief
of symptoms and a lower incidence
of recurrent symptoms. Crawford et
al17 reserved antibiotics for patients
who failed to respond to immobilization,
bed rest, traction, or casting.
Epidural Abscess
Surgical drainage is almost universally
recommended for treatment
of an epidural abscess (Fig. 4). Conservative
management of epidural
abscesses, however, may be appropriate
if the patient has no neurologic
deficit, if the involvement is extensive,
if the patient is not expected to
survive surgery, or if paralysis has
been present for >48 hours so that
neurologic improvement would be
unlikely.26,30 For example, patients
with lumbar involvement, no neural
compromise, and diagnostic cultures
can be effectively treated with
intravenous antibiotics. As with
osteomyelitis, from 2 to 6 weeks of
intravenous antibiotics is usually
recommended. An extended period
of oral antibiotics may be necessary
depending on the immunocompetency
of the patient and the sensitivity
of the organism.
Patients with neurologic deterioration
are best managed with surgical
decompression and débridement
in addition to antibiotic therapy.
Anterior abscesses, particularly
with vertebral body involvement,
should have anterior débridement.
This can be done using either an
open or endoscopic approach. Posteriorly
located infections can be
adequately treated by a laminectomy.
Patients with extensive involvement
can be treated through
multilevel laminectomies. However,
care should be taken not to remove
more bone than is indicated for
decompression because of the risk of
postlaminectomy deformity. Prompt
and aggressive treatment of neurologic
compression appears to favorably
affect neurologic recovery.26
Summary
The most common types of vertebral
osteomyelitis are hematogenous
bacterial or fungal infections
(pyogenic or granulomatous), pediatric
diskitis, epidural abscess, and
postoperative infections. Successful
diagnosis and treatment depend on
an appropriate index of suspicion.
The optimal management of patients
with spinal infection requires
understanding the circumstances
that resulted in the infection, the
organism involved, and the degree
of bony and neurologic compromise.
Early detection and medical
treatment may obviate the need for
surgical intervention. When surgical
débridement is indicated, its
prompt initiation appears to result
in good clinical outcomes. In addition,
maximizing the patient’s nutritional
status with hyperalimentation
improves the outcomes of both medical
and surgical treatment.
References
1. Guri JP: Pyogenic osteomyelitis of the
spine: Differential diagnosis through
clinical and radiographic observations.
J Bone Joint Surg Am 1946;28:29-39.
2. Carragee EJ: Pyogenic vertebral osteomyelitis.
J Bone Joint Surg Am 1997;79:
874-880.
3. Garcia A Jr, Grantham SA: Hematogenous
pyogenic vertebral osteomyelitis.
J Bone Joint Surg Am 1960;42:429-436.
4. Eismont FJ, Bohlman HH, Soni PL,
Goldberg VM, Freehafer AA: Pyogenic
and fungal vertebral osteomyelitis with
paralysis. J Bone Joint Surg Am 1983;
65:19-29.
5. Batson OV: The vertebral system of
veins as a means for cancer dissemination.
Prog Clin Cancer 1967;3:1-18.
6. Wiley AM, Trueta J: The vascular anatomy
of the spine and its relationship to
pyogenic vertebral osteomyelitis. J Bone
Joint Surg Br 1959;41:796-809.
7. Parke WW, Rothman RH, Brown MD:
The pharyngovertebral veins: An
anatomical rationale for Grisel’s syndrome.
J Bone Joint Surg Am 1984;66:
568-574.
8. Sapico FL, Montgomerie JZ: Vertebral
osteomyelitis. Infect Dis Clin North Am
1990;4:539-550.
Bobby K-B Tay, MD, et al
Vol 10, No 3, May/June 2002 197
9. Boachie-Adjei O, Squillante RG:
Tuberculosis of the spine. Orthop Clin
North Am 1996;27:95-103.
10. Doub HP, Badgley CE: The roentgen
signs of tuberculosis of the vertebral
body. AJR Am J Roentgenol 1932;27:
827-837.
11. Krogsgaard MR, Wagn P, Bengtsson J:
Epidemiology of acute vertebral
osteomyelitis in Denmark: 137 cases in
Denmark 1978-1982, compared to
cases reported to the National Patient
Register 1991-1993. Acta Orthop Scand
1998;69:513-517.
12. Torda AJ, Gottlieb T, Bradbury R:
Pyogenic vertebral osteomyelitis:
Analysis of 20 cases and review. Clin
Infect Dis 1995;20:320-328.
13. Kornblum MB, Wesolowski DP,
Fischgrund JS, Herkowitz HN: Computed
tomography-guided biopsy of
the spine: A review of 103 patients.
Spine 1998;23:81-85.
14. Sapico FL, Montgomerie JZ: Pyogenic
vertebral osteomyelitis: Report of nine
cases and review of the literature. Rev
Infect Dis 1979;1:754-776.
15. Berk RH, Yazici M, Atabey N, Ozdamar
OS, Pabuccuoglu U, Alici E:
Detection of Mycobacterium tuberculosis
in formaldehyde solution-fixed, paraffin-
embedded tissue by polymerase
chain reaction in Pott’s disease. Spine
1996;21:1991-1995.
16. Meier A, Persing DH, Finken M,
Bottger EC: Elimination of contaminating
DNA within polymerase chain
reaction reagents: Implications for a
general approach to detection of
uncultured pathogens. J Clin Microbiol
1993;31:646-652.
17. Crawford AH, Kucharzyk DW, Ruda
R, Smitherman HC Jr: Diskitis in children.
Clin Orthop 1991;266:70-79.
18. Scoles PV, Quinn TP: Intervertebral
discitis in children and adolescents.
Clin Orthop 1982;162:31-36.
19. Wenger DR, Bobechko WP, Gilday DL:
The spectrum of intervertebral discspace
infection in children. J Bone Joint
Surg Am 1978;60:100-108.
20. Eismont FJ, Bohlman HH, Soni PL,
Goldberg VM, Freehafer AA: Vertebral
osteomyelitis in infants. J Bone
Joint Surg Br 1982;64:32-35.
21. Omarini LP, Garcia J: CT-guided percutaneous
puncture-biopsy of the
spine: Review of 104 cases [French].
Schweiz Med Wochenschr 1993;123:2191-
2197.
22. Junila J, Niinimaki T, Tervonen O:
Epidural abscess after lumbar discography:
A case report. Spine 1997;22:
2191-2193.
23. Kindler CH, Seeberger MD, Staender
SE: Epidural abscess complicating
epidural anesthesia and analgesia.
Acta Anaesthesiol Scand 1998;42:
614-620.
24. Knight JW, Cordingley JJ, Palazzo MG:
Epidural abscess following epidural
steroid and local anaesthetic injection.
Anaesthesia 1997;52:576-578.
25. Prendergast H, Jerrard D, O’Connell J:
Atypical presentations of epidural
abscess in intravenous drug abusers.
Am J Emerg Med 1997;15:158-160.
26. Sampath P, Rigamonti D: Spinal
epidural abscess: A review of epidemiology,
diagnosis, and treatment. J Spinal
Disord 1999;12:89-93.
27. Song KS, Ogden JA, Ganey T, Guidera
KJ: Contiguous discitis and osteomyelitis
in children. J Pediatr Orthop 1997;
17:470-477.
28. Modic MT, Feiglin DH, Piraino DW, et
al: Vertebral osteomyelitis: Assessment
using MR. Radiology 1985;157:157-166.
29. Whalen JL, Brown ML, McLeod R,
Fitzgerald RH Jr: Limitations of indium
leukocyte imaging for the diagnosis
of spine infections. Spine 1991;16:
193-197.
30. Lang IM, Hughes DG, Jenkins JP, St
Clair Forbes W, McKenna F: MR imaging
appearances of cervical epidural
abscess. Clin Radiol 1995;50:466-471.
31. Ikeda K, Yokoyama M, Okada K,
Tomita K, Yoshimura M: Long-term
follow-up of the vascularized iliac bone
graft. Microsurgery 1998;18:419-423.
32. Hayashi A, Maruyama Y, Okajima Y,
Motegi M: Vascularized iliac bone
graft based on a pedicle of upper lumbar
vessels for anterior fusion of the
thoraco-lumbar spine. Br J Plast Surg
1994;47:425-430.
33. Mosheiff R, Meyer S, Floman Y, Kaplan
L, Eid A, Cohen I: Anterior vascularized
rib strut graft in the treatment of
Pott’s disease in the young child. Bull
Hosp Jt Dis 1993;53:61-65.
34. Lascombes P, Grosdidier G, Olry R,
Thomas C: Anatomical basis of the anterior
vertebral graft using a pediculated
rib. Surg Radiol Anat 1991;13:259-263.
35. A controlled trial of anterior spinal
fusion and debridement in the surgical
management of tuberculosis of the
spine in patients on standard chemotherapy:
A study in Hong Kong. Br J
Surg 1974;61:853-866.
36. Ring D, Wenger DR: Pyogenic infectious
spondylitis in children: The evolution
to current thought. Am J Orthop
1996;25:342-348.

Acute Management Of Spinal Cord Injury

Acute Management of Spinal Cord Injury

Dr Harry Sunaryo SpOT

Each year in the United States,
between 7,600 and 10,000 individuals
sustain and survive a spinal
cord injury. A complex interplay
of regulatory developments and
social issues has influenced trends
in spinal injury. Improvements in
emergency medical services systems,
the development of safer
automobiles, more occupational
safety standards, and better regulation
of contact sports have had a
positive impact on demographic
trends. However, while the overall
incidence of traumatic spinal cord
injury is decreasing nationally, the
percentage due to acts of domestic
violence is sharply on the rise. In
general, more patients are surviving
the initial traumatic injury, and
trends over time indicate an increase
in the proportion of persons
with incomplete paraplegia and a
decrease in the proportion of persons
with complete tetraplegia.1
A number of postinjury trends
have developed: Advances in the
rehabilitation of patients with
spinal cord injuries have resulted
in shorter hospital stays. Between
1974 and 1994, average acute and
rehabilitation hospital stays following
injury declined from 122 days
to 53 days for paraplegic patients
and from 150 days to 75 days for
quadriplegic patients.1 According
to a 1996 study,1 92% of patients
with spinal cord injury are discharged
to independent living or
residential living situations with
assistance. The average life expectancy
for an individual with a
spinal cord injury remains below
normal, but continues to increase.
These positive trends notwithstanding,
the overall impact of
spinal cord injury on society and
on the individual patients and their
families is staggering. It has been
estimated that there are between
183,000 and 203,000 persons living
with spinal cord injuries in the
United States. Estimates of lifetime
costs for health care and living
expenses vary depending on severity
of injury and age at the time of
injury. For example, lifetime costs
for a 25-year-old individual with
high quadriplegia are estimated to
be $1,350,000, whereas costs for a
50-year-old paraplegic patient are
estimated to be $326,000.1
Moreover, each person who sustains
a spinal cord injury undergoes
a devastating transformation
in quality of life, with a loss of
independence and a profound
impact on lifestyle, personal goals,
economic security, and interpersonal
relationships. For example,
in a study from the National Spinal
Cord Injury Statistical Center,1 only


Demographic trends in the occurrence of injury and improvements in the early
management of spinal trauma are changing the long-term profile of patients
with spinal cord injuries. More patients are surviving the initial injury, and
proportionately fewer patients are sustaining complete injuries. While preventive
efforts to reduce the overall incidence of spinal cord injury are important, a
number of steps can be taken to minimize secondary injury once the initial trauma
has occurred. Recent efforts have focused on understanding the biochemical
basis of secondary injury and developing pharmacologic agents to intervene in
the progression of neurologic deterioration. The Third National Acute Spinal
Cord Injury Study investigators concluded that methylprednisolone improves
neurologic recovery after acute spinal cord injury and recommended that
patients who receive methylprednisolone within 3 hours of injury should be
maintained on the treatment regimen for 24 hours. When methylprednisolone
therapy is initiated 3 to 8 hours after injury, it should continue for 48 hours. In
addition to the adoption of the guidelines of that study, rapid reduction and stabilization
of injuries causing spinal cord compression are critical steps in optimizing
patientsÔ long-term neurologic and functional outcomes.
about a third of persons with paraplegia
and about a fourth of those
with quadriplegia were employed
at postinjury year 8. The likelihood
of a marriage remaining intact or of
getting married is far lower than in
the noninjured population.
Most recent successes have been
the result of efforts to decrease the
incidence of primary spinal cord
injury and advances in the rehabilitation
phase of care. This article
focuses on measures to reduce the
potential for secondary mechanical
injury and to address the physiologic
process that ensues once the
primary spinal cord injury has
occurred.
Pathophysiology of Spinal
Cord Injury
Mechanism of Injury
The initial traumatic injury typically
involves impact, compression
and contusion of the spinal cord,
and resultant immediate damage to
nerve cells, axonal tracts, and blood
vessels. Complete severance of the
spinal cord following cervical trauma
is rare; however, as a
result of the primary mechanical
insult, the secondary physiologic
processes, including hemorrhage,
edema, and ischemia, rapidly extend
to contiguous areas in the
cord. Residual pressure on the
cord from bone, ligaments, and
disk material can also exacerbate
the mechanical damage to the cord
after the primary injury.
The secondary injury process is a
complex cascade of biochemical
events, the exact mechanism and
sequence of which are only partially
understood. After the initial impact,
hemorrhage and inflammation
occur in the central gray matter of
the cord. On a systemic level, autonomic
nervous system dysfunction,
hypotension, and bradycardia contribute
to impaired spinal cord perfusion,
which further compounds
the ischemia. Experimental studies
in animal models of spinal cord
injury have shown increases in tissue
water content and sodium and
lactate levels, along with decreases
in extracellular calcium levels, tissue
oxygenation, and pyruvate and
adenosine triphosphate concentrations.
2 Taken together, these observations
are consistent with an overall
scenario of ischemia, hypoxia,
uncoupling of oxidative phosphorylation,
and aerobic glycolysis.
A number of theories have been
proposed to explain the pathophysiology
of secondary injury. Each
theory provides a piece of this complex
puzzle, and there is evidence
of close synergism between the various
mechanisms of secondary
injury. The free-radical theory suggests
that due to rapid depletion of
antioxidants, oxygen free radicals
accumulate in injured central nervous
system tissue and attack membrane
lipids, proteins, and nucleic
acids. As a result, lipid peroxides
are produced, causing the cell
membrane to fail.
The calcium theory implicates
the influx of extracellular calcium
ions into nerve cells in the propagation
of secondary injury. Calcium
ions activate phospholipases,
proteases, and phosphatases, resulting
in both interruption of
mitochondrial activity and disruption
of the cell membrane.
The opiate receptor theory is
based on evidence that endogenous
opioids may be involved in the
propagation of secondary spinal
cord injury. There is evidence that
opiate antagonists, such as naloxone,
may improve neurologic recovery
in experimental models of
spinal cord injury. However, different
studies have reported conflicting
results, and it may be that
the beneficial effect of opiate antagonists
is dose-responsive.
The inflammatory theory is based
on the hypothesis that inflammatory
substances (e.g., prostaglandins,
leukotrienes, platelet-activating factor,
and serotonin) accumulate in
acutely injured spinal cord tissue
and are mediators of secondary tissue
damage.3 Anti-inflammatory
agents have been tested extensively
in spinal cord injury.
Histologic manifestations of
acute spinal cord injury include
necrosis of central cord gray matter
in the first hours after injury, followed
by cystic degeneration. Over
the ensuing several weeks, the
development of scar tissue extends
into the axonal long tracts, with disruption
of axonal continuity.
Effect of Timing of
Decompression
In a 1995 in vivo animal study,
Delamarter et al4 evaluated the
Fig. 1 Complete severance of the spinal
cord after a severe C6 fracture-subluxation.
The 18-year-old male patient sustained a
diving injury and immediate C6 quadriplegia.
This magnetic resonance image obtained
90 minutes after the injury depicts
complete severance of the cord at the base
of the C6 vertebra and hemorrhage into the
cord cephalad to the C6 level (arrow).
Acute Management of Spinal Cord Injury
168 Journal of the American Academy of Orthopaedic Surgeons
effect of timing of decompression of
the spinal cord after acute experimental
spinal cord compression
injury . In their canine
model, 50% spinal cord compression
was surgically obtained with a
constriction band. Decompression
was then performed immediately in
6 dogs and at 1 hour, 6 hours, 24
hours, and 1 week, respectively, in
the other four groups of 6 dogs
each. Data from somatosensory
evoked potential monitoring, daily
neurologic examinations, and histologic
and electron-microscopic studies
performed at autopsy were
available for all animals. Initially,
all 30 dogs were paraplegic. The
dogs that underwent immediate
decompression or decompression
after 1 hour recovered the ability to
walk as well as control of the bowels
and bladder. When compression
lasted 6 hours or more, there was no
neurologic recovery, and progressive
necrosis of the spinal cord was
noted on histologic examination
This research suggests that
not all damage to the spinal cord
occurs at the time of initial trauma
and that the extent and persistence
of damage depend in part on the
duration of compression.
Pharmacologic
Intervention
The development of pharmacologic
agents to halt progression of secondary
neurologic damage after a
primary injury has been based on a
growing understanding of the
sequence of biochemical events.
There are ongoing research efforts
at the basic and preclinical levels,
as well as several major clinical
studies. A number of agents,
including corticosteroids, 21-
aminosteroids, free-radical scavengers,
opiate antagonists, calciumchannel
blockers, and neurotrophic
factors, are being investigated.
Table 1 lists a number of these
agents by class. Methylprednisolone,
tirilazad, and GM1 ganglioside
are each currently being evaluated
in ongoing clinical trials.
Methylprednisolone
The initial rationale for use of glucocorticoids
in the treatment of acute
spinal cord injury was based on
their efficacy in treatment of cerebral
edema in patients with closed head
injury and brain tumors. Subsequently,
additional mechanisms
have been proposed for the beneficial
effects of methylprednisolone,
including reduction of excitatory
amino acid neurotoxicity, inhibition
of lipid peroxidation, increases in
spinal-tissue blood perfusion, and
slowing of traumatic ion shifts.5
The Second National Acute Spinal
Cord Injury Study (NASCIS-II),
which was a prospective, randomized,
placebo-controlled, doubleblinded
clinical trial, demonstrated
that intravenous administration of
high-dose methylprednisolone improved
clinical outcomes.6 Completed
in January 1990, NASCIS-II
was the first clinical trial to demonstrate
statistically significant neurologic
recovery from, or reversal of,
neurologic injury. The NASCIS-II investigators
evaluated the efficacy
and safety of methylprednisolone
and naloxone in a placebo-controlled
multicenter study of 487 patients
with acute spinal cord injury.
Ninety-five percent of the patients
were treated within 14 hours of
injury. Methylprednisolone was
given to 162 patients in a bolus dose
of 30 mg per kilogram of body
weight, followed by an infusion at
the rate of 5.4 mg/kg per hour for 23
hours. Naloxone was given to 154
patients as a 5.4-mg/kg bolus injection,
followed by an infusion at the
Preoperative
SEP
SEP After
Compression
SEP
6 Weeks After
Decompression
% of Tibial Amplitude
20
10
0
30
40
50
60
70
80
90
100
Time of Decompression
Zero
1 Hour
6 Hours
24 Hours
1 Week
Fig. 2 Somatosensory evoked potential (SEP) recovery after decompression of experimental
spinal cord injury in 30 dogs. Note the mean deterioration of the amplitude of posterior
tibial SEPs, compared with preoperative values, after compression of the spinal cord and
the subsequent recovery in amplitude 6 weeks after decompression. Six weeks after
decompression, only the dogs in group 1 (immediate decompression) and group 2 (decompression
at 1 hour) showed significant improvement (P<0.05) in amplitude. (Reproduced
with permission from Delamarter RB, Sherman J, Carr JB: Pathophysiology of spinal cord
injury: Recovery after immediate and delayed compression. rate of 4.0 mg/kg per hour for 23
hours. Placebo was given to 171
patients.
The NASCIS-II data demonstrated
that patients who received a
high-dose methylprednisolone
infusion within 8 hours of spinal
cord injury had better recovery of
neurologic function at 6 weeks, 6
months, and 1 year after injury,
compared with patients treated
with placebo or naloxone.6 Although
the degree of neurologic
recovery was strongly related to
the completeness of injury, patients
with complete injuries as well as
those with incomplete injuries
improved more after treatment
with methylprednisolone than after
placebo administration.
There were no statistically significant
differences in mortality and
morbidity in the methylprednisolone
group in comparison to
the placebo group. However, patients
with incomplete spinal cord
injuries treated with methylprednisolone
beyond 8 hours postinjury
had significantly less neurologic
recovery than similar patients treated
with placebo, indicating that
there may be a detrimental effect to
late administration of methylprednisolone.
Treatment with naloxone
in the doses used in NASCIS-II did
not significantly improve neurologic
recovery in comparison to placebo.
6
The NASCIS-II study has been
criticized for deficiencies in experimental
design and incomplete data.
Detailed medical and surgical protocols,
as well as radiologic descriptions
of the injuries, were not
reported. Description of the initial
severity of neurologic injuries within
each of the treatment groups was
not provided in detail. The scheme
for grading neurologic improvement
in NASCIS-II did not employ
functional measures of outcome;
therefore, it was not possible to
assess clinically useful degrees of
recovery.7,8
The Third National Acute Spinal
Cord Injury Study (NASCIS-III)
was a multicenter, randomized,
double-blinded prospective study
reported in May 1997.9 Because
NASCIS-II showed greater neurologic
recovery with methylprednisolone,
the investigators felt an
obligation to include methylprednisolone
in the treatment of all
A B
C D
Fig. 3 Histologic findings in an experimental model of spinal cord injury in dogs. A,
Section of spinal cord approximately 1 cm cephalad to spinal cord injury after immediate
decompression. Note the mild deformity of the spinal cord but only minimal histologic
damage (hematoxylin-eosin staining). B, Higher-power view of a similar section from a
dog after 1 hour of constriction. Note the mild to moderate cord deformity, the early
degeneration in the central cord, and mild peripheral destructive changes. C, Spinal cord
section from a dog with decompression after 6 hours of compression (hematoxylin-eosin,
original magnification ´6. Note the severe degeneration in the central cord (arrows) and
the posterior columns. Spinal cord damage was significantly related to the duration of
compression. D, Electron-microscopic view showing neural tissue and exiting dendrite.
Section was taken 5 mm caudad to the level of compression from a dog after 6 hours of
compression. Note the severe degenerative changes in the mitochondria (arrows) and disorganization
on both sides of the exiting dendrite (arrowheads) (original magnification
´6,000). (Parts C and D reproduced with permission from Delamarter RB, Sherman J, Carr
JB: Pathophysiology of spinal cord injury: Recovery after immediate and delayed compression.
patients in NASCIS-III and all subsequent
clinical trials. Therefore,
the three groups of patients in
NASCIS-III all received an initial
30-mg/kg bolus dose of methylprednisolone
before randomization.
The first group of NASCIS-III
patients (n = 166) received an infusion
of methylprednisolone at a
rate of 5.4 mg/kg per hour for 23
hours after the bolus dose. The
second group (n = 166) received the
methylprednisolone infusion for a
total of 48 hours after the bolus
dose. The third group (n = 167)
received a bolus dose of methylprednisolone,
followed by a 2.5-
mg/kg bolus of tirilazad every 6
hours for 48 hours.
Neurologic function was assessed
at the time of initial presentation
and at 6 weeks and 6 months
after spinal cord injury. At the time
of the 6-month follow-up, 94.7% of
surviving patients were available
for evaluation. Examinations were
conducted by NASCIS-trained
physicians and nurses and included
quantitative scoring of motor and
sensory function, as well as functional
independence measures.
In patients who were treated less
than 3 hours after injury, essentially
identical rates of motor recovery
were observed in all three treatment
groups. In patients in whom
treatment was initiated between 3
and 8 hours after injury, the 48-
hour methylprednisolone group
recovered significantly more motor
function than the 24-hour methylprednisolone
group. The 48-hour
tirilazad group recovered at a rate
slightly faster than the 24-hour
methylprednisolone group, but the
difference was not statistically significant.
Patterns of recovery of
sensory function paralleled those
for recovery of motor function.
However, differences in sensory
function improvement between the
groups were smaller. Greater improvement
in functional independence
measures at 6 months was
observed in the 48-hour methylprednisolone
group than in the 24-
hour group. The 48-hour tirilazad
group improved at rates between
those for the two methylprednisolone
groups.
Small differences in complication
rates were noted between the
groups, with higher rates of severe
sepsis and severe pneumonia in the
48-hour methylprednisolone group.
These complications did not affect
overall mortality. Although the
NASCIS-II investigators did not
report a statistically significant difference
in mortality and morbidity
between treatment and control
groups, the first NASCIS study
demonstrated that 10 days of glucocorticoid
treatment was associated
with an increased risk of complications.
7 Other authors have associated
the use of high-dose glucocorticoids
in the treatment of acute
spinal cord injury with increased
risk of pneumonia and wound infections
and prolongation of hospital
stay.10
On the basis of the results of the
NASCIS-III trial, the investigators
recommended that patients with
acute spinal cord injury who receive
methylprednisolone within 3
hours of injury should be maintained
on the treatment regimen for
24 hours. They further recommended
that when methylprednisolone
therapy is initiated 3 to 8
hours after injury, it should be continued
for 48 hours.9
Tirilazad
Tirilazad is a lazeroid (synthetic
21-aminosteroid). Lazeroids are
extremely potent antioxidants and
exhibit neuroprotective effects by a
variety of other mechanisms as
well, such as improving spinal cord
blood flow and membrane stabilization.
Because lazeroids have
none of the glucocorticoid properties
of methylprednisolone, tirilazad
may have fewer side effects.
GM1 Ganglioside
Gangliosides are complex acidic
glycolipids found in high concentrations
in central nervous system
tissue as a major component of the
cell membrane. In animal studies,
gangliosides have been shown to
stimulate the growth of nerve cells
in damaged tissue.11 Their mechanism
of action involves enhancing
survival of residual axonal tracts
passing through the site of injury,
thereby facilitating the recovery of
useful motor function distally.
Gangliosides also act to limit cell
destruction by excitatory amino
acids.
In a 1991 randomized, prospective
clinical trial, Geisler et al12
demonstrated statistically significant
neurologic improvement in
patients given a parenteral GM1
Table 1
Pharmacologic Agents Under Investigation for Use in Treatment of Acute
Spinal Cord Injury
Agent Class
Naloxone m-Opiate receptor antagonist
Methylprednisolone Corticosteroid
Nimodipine Calcium-channel blocker
4-Aminopyridine Potassium-channel blocker
GM1 Ganglioside Glycolipid (neurotrophic factor)
Tirilazad (lazeroid) Lipid peroxidase inhibitor
Vitamin E Free-radical scavenger
ganglioside sodium salt, compared
with patients given placebo. At
follow-up 1 year after injury, significant
improvement was noted
on the basis of both the American
Spinal Injury Association motor
score and the Frankel classification
grade. Analysis of the data indicated
that improved function in
patients treated with GM1 ganglioside
occurred in initially paralyzed,
rather than paretic, muscles.
Currently, a large multicenter
study is in progress to validate the
initial clinical results seen with
GM1 ganglioside treatment.13 The
study also seeks to establish the
safety and efficacy of two dose regimens
of GM1 ganglioside.
4-Aminopyridine
4-Aminopyridine is a fast potassium-
channel blocker, which has
been shown in experimental models
of spinal cord injury to enhance
nerve conduction through demyelinated
nerve fibers by prolonging
the duration of action potentials.
When 4-aminopyridine was given
in limited clinical trials to patients
with incomplete injuries, it produced
temporary neurologic improvements,
which persisted for as
long as several days after administration
of the drug.14
Spinal Cord Regeneration
A number of studies to investigate
the regeneration of axonal tracts
after traumatic spinal cord injury
are currently underway. For example,
researchers at the University of
Zurich administered antibodies to
neutralize myelin-associated neurite
growth inhibitory factor to
young adult rats that had undergone
partial transection of the
midthoracic spinal cord. The treatment
resulted in growth of corticospinal
axons around the site of
injury and into spinal cord levels
caudal to the injury.15
Recently, Cheng et al16 reported
on a study in which they completely
transected a 5-mm section of
spinal cord at the T8 level in adult
rats. This was followed by grafting
of peripheral nerve implants from
individual axonal tracts to areas of
neuronal cell bodies to bridge the
gap. Acidic fibroblast growth factor,
a constituent of normal spinal
cord tissue, was mixed with fibrin
glue and then used to stabilize the
grafts. Rat hind-limb function
improved progressively over a 6-
month period, compared with controls.
Although this study is far
removed from clinical application
to traumatic spinal cord injury in
humans, it represents the first evidence
that regeneration can occur
in a completely transected spinal
cord of an adult animal and suggests
that therapies will eventually
be discovered for regeneration of
the spinal cord after traumatic
injury.
Management of Acute
Spinal Cord Injury
Evaluation and Medical
Management
Although current understanding
of the pathophysiology of acute
spinal cord injury is limited, the
recommended treatment protocol
is based on three major
objectives. First is prevention of
secondary injury by pharmacologic
intervention, such as administration
of methylprednisolone within
8 hours after injury, in accordance
with the guidelines established in
NASCIS-III. Patients should be
given a 30-mg/kg bolus dose of
methylprednisolone, followed by
either a 23-hour or a 48-hour infusion
at the rate of 5.4 mg/kg per
hour.6
Second, hypoxia and ischemia at
the local site of spinal cord injury
should be minimized by controlling
hemodynamic status and oxygenation.
All patients should receive
supplemental oxygen sufficient to
achieve an oxygen saturation approaching
100%. This should be
initiated as soon as the diagnosis of
spinal cord injury is made. Patients
with high cervical injuries may
require intubation to reach this
level.
Neurogenic shock results from
the disruption of sympathetic outflow
by cord injury. It is clinically
manifested by hypotension due to
vasodilatation and bradycardia
secondary to unopposed vagal
influence on the heart. Patients in
neurogenic shock typically have a
heart rate between 50 and 70 beats
per minute and a systolic pressure
30 to 50 mm Hg below normal.
Neurogenic shock must be differentiated
from hypovolemic shock,
Table 2
Acute Management of Cervical Spinal Cord Injury
1. Maintenance of perfusion systolic blood pressure >90 mm Hg
2. 100% O2 saturation via nasal cannula
3. Early diagnosis by plain radiography
4. Methylprednisolone therapy (loading dose of 30 mg/kg followed by
infusion at rate of 5.4 mg/kg per hour for 23 or 48 hours)
5. Immediate traction reduction for cervical fracture and dislocation
6. Spinal imaging (MR imaging and/or computed tomography)
7. Surgery if indicated for residual cord compression or fracture instability
Acute Management of Spinal Cord Injury
172 Journal of the American Academy of Orthopaedic Surgeons
which presents with a combination
of tachycardia and hypotension,
generally due to blood loss
from abdominal or pelvic injury.17
Treatment of neurogenic shock
includes an initial fluid challenge,
Trendelenburg positioning (10 to
20 degrees), vasopressors (e.g.,
dopamine and phenylephrine hydrochloride)
after central line
placement, and atropine for treatment
of bradyarrhythmias. Systolic
blood pressure should be
restored to normal as quickly as
possible.
Third, once a spinal cord injury
is suspected, the spine should be
immobilized to prevent further
neurologic injury. Currently, most
spinal cord injury patients are
transported to trauma centers by
emergency medical services personnel
and arrive immobilized on a
trauma board with a collar. Effective
management requires the assumption
that every polytraumatized
or unconscious patient has a
spinal cord injury until proven otherwise.
Early recognition and appropriate
acute management of spinal
cord injuries is critical to improving
overall patient outcome. For
example, the incidence of complete
neurologic injury in patients with
traumatic spinal insults admitted to
one regional spinal cord injury system
in 1972 was 81%; by 1992, this
had dropped to 57%.18 In another
study,19 the proportion of complete
spinal cord injuries decreased from
64% to 46% after the establishment
of a regional spinal cord injury
unit.
Spinal cord injury is frequently
accompanied by other injuries,
many of which can be life-threatening.
For example, of patients with
spinal cord injury secondary to
motor-vehicle accidents, 40% have
associated fractures, 42.5% experience
loss of consciousness, and
16.6% have a traumatic pneumothorax
or hemothorax.20 The initiation
of evaluation and treatment of
acute spinal cord injuries may be
delayed by the need to treat more
life-threatening injuries. Nevertheless,
during the acute resuscitation
and evaluation of the polytrauma
patient, the spine should be stabilized
and protected from further
injury at all times.
Accurate radiologic (Fig. 4) and
neurologic assessment of the patient
with a spinal cord injury
should be part of the secondary
trauma survey. When feasible,
malaligned vertebral fractures or
dislocations should be reduced concurrently
with ongoing trauma
resuscitation measures. Early intervention
is essential to limit the secondary
spinal cord injury. If the
patient survives the life-threatening
injuries, the outcome of the spinal
injury will be a predominant factor
influencing the future quality of
life.
Patients presenting with either a
neurologic deficit or evidence of
cervical spine instability should be
placed in cervical traction with
tongs or a halo ring. Contraindications
to cervical traction include
distraction injuries at any level in
the cervical spine and type IIA
hangmanÕs fractures. The objectives
of application of halo or tong
traction are spinal stabilization
and, when possible, rapid decompression
through realignment of
the spinal canal.
A lateral cervical spine film
showing C1 to T1 should be available
before the application of traction
and should be repeated after
the initial application of 10 to 15 lb.
Weight can then be added in 5- and
10-lb increments, followed by serial
neurologic evaluations and repeat
radiographs until evidence of
alignment is seen. Intravenous
administration of 1 to 4 mg of
midazolam hydrochloride as an
adjunct to achieve muscle relaxation
and use of fluoroscopy can
facilitate a more rapid, controlled
reduction of cervical facet dislocations.
Contraindications to continued
attempts at reduction using
traction include worsening neurologic
deficits and evidence of distraction
by more than 1.0 cm in a
disk space. Reduction is typically
obtained with 40 to 70 lb of traction,
although use of more than 100
lb has been reported.21
For initial immobilization, cervical
tongs and the halo ring each
have advantages. In some centers,
cervical tongs are preferred because
of the rapidity and ease with which
they can be applied by one person
in an emergency room. Halo application
takes somewhat longer and
generally requires two persons, but
has the advantage of control of
alignment in three planes and can
facilitate the reduction of unilateral
and bilateral facet dislocations.
Availability of traction equipment
is important; delays in application
of traction are common due to the
necessity of obtaining a halo from
another location or due to ongoing
radiologic or trauma evaluation.
Ideally, the halo or tongs should be
compatible with magnetic resonance
(MR) imaging. However, the
application of cervical traction
should not be delayed in order to
first obtain a diagnostic study, such
as MR imaging or computed tomography/
myelography.
Slucky and Eismont19 recommend
MR imaging for assessment
of the degree of spinal cord compression
in patients with complete
or incomplete neurologic deficit, as
well as in patients whose neurologic
status has deteriorated and
those in whom disk retropulsion
with canal compromise or posterior
ligament injury is suspected. The
MR images should be obtained
after application of traction; reduction
of a dislocation in a patient
with a severe incomplete or complete
neurologic deficit should not
be delayed for completion of an
MR study.
A B C
D E
F G
Images of a 26-year-old woman who fell while
rollerblading and sustained a severe C5 fracture-subluxation
(teardrop fracture). Twenty minutes after the
injury she was urgently transported to the emergency
room, and complete C5 quadriplegia was identified.
A, Initial MR image shows severe spinal cord compression
by the C5 vertebral body, illustrated by the marked
signal change in the cord directly above the fractured
vertebra. B, The initial computed tomographic (CT)
reconstruction illustrates the severe fracture-subluxation
of the C5 vertebral body. The initial MR imaging and
CT studies were obtained within 1 hour after injury.
C, Axial MR image demonstrates severe damage to the
spinal cord (arrows) with what appears to be midline
separation of the cord, probably representing hematoma
into both sides of the cord. D, Axial CT scan depicts a
midline fracture through the C5 vertebral body as well
as posterior laminar fractures bilaterally and severe
spinal canal compression. E, Lateral cervical spine radiograph
taken after application of Gardner-Wells tongs
and 30 lb of traction demonstrates restoration of the normal
cervical alignment and partial reduction of the C5
vertebral fracture-subluxation. F, Approximately 2
hours after the injury, the patient underwent a C5 vertebrectomy
with spinal cord decompression and anterior
fusion with an iliac-crest strut graft and anterior plate
fixation. A Philadelphia collar was worn for 6 weeks.
The patient was transferred to a spinal cord rehabilitation
unit 4 days after surgery. G, At the 6-month followup
examination, the patient demonstrated complete root
recovery to the C7 level on the right side and single-root
recovery to the C6 level on the left side. An MR image
obtained at that time depicts significant signal changes
in the spinal canal at the level of the cord injury.
Serial Examinations
The objectives of the initial neurologic
examination conducted
during the secondary trauma survey
are to establish the level and
type of neurologic deficit and to
determine whether there is any
motor or sensory sparing distal to
the level of injury. The initial evaluation
is the most valuable from a
prognostic standpoint, as it guides
treatment decisions and serves as a
baseline for subsequent evaluations.
Follow-up examinations
should be performed at regular
intervals and also whenever the
patient is transferred or undergoes
traction adjustments or surgical
procedures. In a multicenter study
of deterioration of neurologic status
after spinal cord injury, Marshall
et al22 prospectively evaluated
283 patients admitted to five trauma
centers. Fourteen of these patients
deteriorated neurologically
during acute hospital management.
In 12 of the patients, deterioration
could be specifically associated
with a management intervention,
such as traction or halo-vest application,
surgery, or Stryker frame or
rotating bed rotation.
The use of the American Spinal
Injury Association scoring diagram
for spinal cord injury helps examiners
obtain accurate, complete,
and reproducible neurologic assessments.
If examinations are
recorded each time in the same format
and with use of the same data
points, they can be easily compared
with one another.
Timing of Operative Treatment
The timing of surgery remains a
controversial issue. There is little
debate that emergency surgical
decompression is indicated for a
progressive neurologic deficit in the
presence of persistent spinal cord
compression. Operative intervention
in other clinical circumstances
can be done on an acute or urgent
basis or can be delayed. Ducker et
al23 advocated acute operative
intervention for patients with cervical
spinal cord injury who require
open reduction or decompression
for persistent spinal cord compression,
instability at the occipital cervical
junction, or atlantoaxial instability.
Other authors recommend
treating nonprogressive neurologic
deficits on a semiurgent basis,
when the patient is medically stable.
24
In a multicenter study, Marshall
et al22 had three patients with cervical
spinal cord injuries whose
neurologic condition deteriorated
after surgery. Each patient had
been operated on within 5 days of
injury. No such deterioration was
noted when surgery was performed
after 5 days. On the basis
of these observations in a very
small sample of patients, they recommended
that early surgical
intervention should be performed
only to avoid further deterioration
in neurologic function.
There have been other reports of
marked neurologic recovery in
patients who presented initially
with complete deficits and canal
compromise and were treated with
rapid closed reduction and restoration
of alignment. In one of the
earliest retrospective reviews,
Frankel et al25 evaluated the data
on 682 patients who underwent
postural reduction at the National
Spinal Injuries Centre in England
between 1951 and 1968. On detailed
analysis of the neurologic
results, the authors noted that a
small number of patients with complete
neurologic lesions initially
and a larger number of patients
with incomplete lesions improved.
No mention was made of a correlation
between timing of the reduction
and degree of recovery. Furthermore,
the authors could not
correlate the severity of the neurologic
lesion or the degree of reduction
achieved with the neurologic
recovery.
Hadley et al26 presented the data
on a series of 68 patients with acute
traumatic cervical-facet fracturedislocations.
One patient, who presented
initially with a unilateral
dislocation and a complete deficit,
improved neurologically after reduction
to the point that he could
ambulate with arm braces. Another
patient, who presented with
a complete neurologic deficit due
to a bilateral facet dislocation,
underwent closed reduction with
cervical traction within 4 hours of
injury and was neurologically
intact at last follow-up (54 months
after injury).
In patients with incomplete neurologic
function, the results of very
rapid reduction are more promising.
In a series of 100 surgically
treated cervical spine injuries, Aebi
et al27 noted neurologic improvement
after manual or surgical
reduction in 31 patients. Of these
patients, 75% underwent reduction
within 6 hours of the injury. In
contrast, 85% of the 69 patients
who had no neurologic recovery
underwent reduction more than 6
hours after injury.
These clinical observations are
consistent with the previously cited
experimental conclusions drawn by
Delamarter et al4 regarding the
effect of timing of decompression
of the spinal cord after acute experimental
spinal cord compression
injury. The findings in that study
suggest that not all damage to the
spinal cord occurs at the time of
initial trauma and that the extent
and persistence of damage depend
in part on the duration of compression.
It therefore appears that a
window of opportunity may exist
in many spinal cord injuries. Although
the time available for intervention
is short, there is a period
when complete injury may be partially
reversible.
Other authors have considered
both the force of the initial injury
and the timing of decompression in
the prognosis for recovery.28 Although
the force of the initial injury
may be the predominant factor, the
timing of decompression or reduction
and medical management are
the only factors over which the
spine surgeon has control.
Summary
Recent advances in understanding
of the pathogenesis of spinal cord
injury hold promise for future
improvement in clinical outcomes.
In the meantime, early management
in accordance with the
NASCIS-III protocol, along with
rapid reduction and stabilization,
affords the best opportunity for
optimization of the long-term outcome
in patients with spinal cord
injuries.
References
1. National Spinal Cord Injury Statistical
Center: Spinal Cord Injury Facts and
Figures at a Glance. Birmingham, Ala:
University of Alabama, 1999.
2. Janssen L, Hansebout RR: Pathogenesis
of spinal cord injury and newer treatments:
A review. Spine 1989;14:23-32.
3. Young W, Huang P, Kume-Kick J:
Cellular ionic and biomolecular mechanisms
of the injury process, in Benzel
EC, Tator CH (eds): Contemporary
Management of Spinal Cord Injury. Park
Ridge, Ill: American Association of
Neurological Surgeons, 1995, pp 28-31.
4. Delamarter RB, Sherman J, Carr JB:
Pathophysiology of spinal cord injury:
Recovery after immediate and delayed
compression. J Bone Joint Surg Am
1995;77:1042-1049.
5. Greene KA, Marciano FF, Sonntag VK:
Pharmacological management of
spinal cord injury: Current status of
drugs designed to augment functional
recovery of the injured human spinal
cord. J Spinal Disord 1996;9:355-366.
6. Bracken MB, Shepard MJ, Collins WF,
et al: A randomized, controlled trial of
methylprednisolone or naloxone in the
treatment of acute spinal-cord injury:
Results of the Second National Acute
Spinal Cord Injury Study. N Engl J
Med 1990;322:1405-1411.
7. Zeidman SM, Ling GS, Ducker TB,
Ellenbogen RG: Clinical applications of
pharmacologic therapies for spinal cord
injury. J Spinal Disord 1996;9:367-380.
8. Geisler F: Prevention of spinal cord
injury, in Benzel EC, Tator CH (eds):
Contemporary Management of Spinal
Cord Injury. Park Ridge, Ill: American
Association of Neurological Surgeons,
1995, p 262.
9. Bracken MB, Shepard MJ, Holford TR,
et al: Administration of methylprednisolone
for 24 or 48 hours or tirilazad
mesylate for 48 hours in the treatment
of acute spinal cord injury: Results of
the Third National Acute Spinal Cord
Injury Randomized Controlled TrialÑ
National Acute Spinal Cord Injury
Study. JAMA 1997;277:1597-1604.
10. Galandiuk S, Raque G, Appel S, Polk
HC Jr: The two-edged sword of largedose
steroids for spinal cord trauma.
Ann Surg 1993;218:419-425.
11. Rengachary S, Alton S: Resuscitation
and early medical management of the
spinal cord injury patient, in Benzel
EC, Tator CH (eds): Contemporary
Management of Spinal Cord Injury. Park
Ridge, Ill: American Association of
Neurological Surgeons, 1995, pp 60-61.
12. Geisler FH, Dorsey FC, Coleman WP:
Recovery of motor function after
spinal-cord injury: A randomized,
placebo-controlled trial with GM-1
ganglioside. N Engl J Med 1991;324:
1829-1838.
13. Geisler FH, Dorsey FC, Coleman WP:
Past and current clinical studies with
GM-1 ganglioside in acute spinal cord injury.
Ann Emerg Med 1993;22:1041-1047.
14. Hansebout RR, Blight AR, Fawcett S,
Reddy K: 4-Aminopyridine in chronic
spinal cord injury: A controlled,
double-blind, crossover study in eight
patients. J Neurotrauma 1993;10:1-18.
15. Bregman BS, Kunkel-Bagden E, Schnell
L, Dai HN, Gao D, Schwab ME: Recovery
from spinal cord injury mediated
by antibodies to neurite growth
inhibitors. Nature 1995;378:498-501.
16. Cheng H, Cao Y, Olson L: Spinal cord
repair in adult paraplegic rats: Partial
restoration of hind limb function.
Science 1996;273:510-513.
17. Soderstrom CA, McArdle DQ, Ducker
TB, Militello PR: The diagnosis of
intra-abdominal injury in patients
with cervical cord trauma. J Trauma
1983;23:1061-1065.
18. Waters RL, Apple DF Jr, Meyer PR Jr,
Cotler JM, Adkins RH: Emergency and
acute management of spine trauma, in
Stover SL, DeLisa JA, Whiteneck GG
(eds): Spinal Cord Injury: Clinical Outcomes
From the Model Systems. Gaithersburg,
Md: Aspen Publishers, 1995, p 57.
19. Slucky AV, Eismont FJ: Treatment of
acute injury of the cervical spine. Instr
Course Lect 1995;44:67-80.
20. Go BK, DeVivo MJ, Richards JS: The
epidemiology of spinal cord injury, in
Stover SL, DeLisa JA, Whiteneck GG
(eds): Spinal Cord Injury: Clinical
Outcomes From the Model Systems.
Gaithersburg, Md: Aspen Publishers,
1995, pp 44-45.
21. Cotler JM, Herbison GJ, Nasuti JF,
Ditunno JF Jr, An H, Wolff BE: Closed
reduction of traumatic cervical spine
dislocation using traction weights up
to 140 pounds. Spine 1993;18:386-390.
22. Marshall LF, Knowlton S, Garfin SR, et
al: Deterioration following spinal cord
injury: A multicenter study. J Neurosurg
1987;66:400-404.
23. Ducker TB, Bellegarrigue R, Salcman
M, Walleck C: Timing of operative
care in cervical spinal cord injury.
Spine 1984;9:525-531.
24. Connolly PJ, Abitbol JJ, Martin RJ,
Yuan HA: Spine: Trauma, in Garfin
SR, Vaccaro AR (eds): Orthopaedic
Knowledge Update: Spine. Rosemont,
Ill: American Academy of Orthopaedic
Surgeons, 1997, p 202.
25. Frankel HL, Hancock DO, Hyslop G,
et al: The value of postural reduction
in the initial management of closed
injuries of the spine with paraplegia
and tetraplegia: I. Paraplegia 1969;7:
179-192.
26. Hadley MN, Fitzpatrick BC, Sonntag
VK, Browner CM: Facet fracturedislocation
injuries of the cervical
spine. Neurosurgery 1992;30:661-666.
27. Aebi M, Mohler J, Zach GA, Morscher E:
Indication, surgical technique, and
results of 100 surgically-treated fractures
and fracture-dislocations of the cervical
spine. Clin Orthop 1986;203:244-257.
28. Guha A, Tator CH, Endrenyi L, Piper
I: Decompression of the spinal cord
improves recovery after acute experimental
spinal cord compression injury.
Paraplegia 1987;25:324-339.

Legg Calve Perthes Disease

Legg-Calve-Perthes Disease

Dr.Harry Sunaryo SpOT

Legg-Calve-Perthes disease was originally described independentlyin 1910 by Legg of the UnitedStates, Calve of France, and Perthesof Germany. The condition is a diseaseof children in which the essentiallesion is not simply ischemia,but also includes the resultingprocess of resorption, collapse, andrepair, which may result in apainful, poorly functioning hip.The majority of patients do wellinto their fifth decade; however, inone long-term study of untreatedpatients, 50% had disabling arthritisby age 55.1 Treatment is largelya matter of the physicianÕs personalpreference, as the literature offers little scientific evidence to suggestsuperiority of one treatment over another or even to conclusivelyestablish the efficacy of any treatment over the natural history of thedisease.PathogenesisBiopsy studies have established thatvarious stages of bone necrosis andrepair are present in the femoralhead in Legg-CalvŽ-Perthes disease.However, numerous experiments inwhich the blood supply to thefemoral head was interrupted in asingle event failed to produce thecharacteristic lesion of Legg-CalvŽ-Perthes disease. Other researchershave found that recurrent injury tothe circumflex arteries in dogs mimicsthe appearance of the disease,suggesting that Legg-CalvŽ-Perthesdisease occurs over time with repetitiveinjury. The dependence of theblood supply to the femoral head onretinacular vessels that coursethrough a nondistensible intracapsularspace makes the theory ofischemia secondary to tamponadepersuasive. Nevertheless, althoughartificially elevated intracapsularhip pressure and increased intraosseousvenous pressure have beenshown to produce femoral headischemia in animals experimentally,they have not been directly linked toLegg-CalvŽ-Perthes disease inhumans. Furthermore, ultrasoundevaluation of 4- to 8-year-old boyswith early Legg-CalvŽ-Perthes diseasemanifested by groin pain andlimited hip motion also failed to confirmthe presence of capsular distention.2More recently, clotting factorsand increased blood viscosity havebeen implicated as potential causes.Gregosiewicz et al2 examined theserum of 26 boys with early Legg-CalvŽ-Perthes disease and found asignificantly greater level of a1-antitrypsinthan in control subjects, indicatinga decrease in fibrinolyticactivity and enhanced intravascularclotting.A possible relationship betweenendocrine abnormalities, especiallythyroid disorders, and Legg-CalvŽ-
AbstractLegg-CalvŽ-Perthes disease is a self-limited disease of the femoral head that presentsin the first decade. The pathogenesis is thought to involve bone necrosis,collapse, and repair. The presenting complaint is often a painless limp or hippain, with decreased abduction and internal rotation of the hip. Factors that arebelieved to correlate with a poor prognosis are onset of symptoms after age 8years, lateral head subluxation, involvement of over 50% of the femoral head withcollapse of the lateral pillar, and the combination of an aspherical femoral headand an incongruent joint. The current cornerstones of treatment are maintenanceof hip motion, relief of symptoms, and containment. Containment may beachieved by bracing or surgical means. The literature remains inconclusive onthe indications for and effects of treatment. A long-term study has suggested thatdisabling arthritis of the hip develops in the sixth decade of life in 50% ofuntreated patients.J Am Acad Orthop Surg 1996;4:9-16Perthes disease has also been studied.In a large recent study,3 it wasfound that although children withLegg-CalvŽ-Perthes disease areeuthyroid, they have significantlyhigher levels of free thyroxine andfree triiodothyronine than controlsubjects. In addition, there was a statisticallysignificant increase in freethyroxine levels with an increasingdegree of femoral-head involvement.Earlier work by the sameinvestigators had shown that plasmalevels of insulin-like growth factor Iwere reduced early in Legg-CalvŽ-Perthes disease.Histologic evaluation of femoralheadspecimens from patients withLegg-CalvŽ-Perthes disease showsnecrosis at different stages of repair.In specimens from 11 children,Catterall et al4 found three stages ofthe disease process. In the initialstage, bone necrosis predominatesfrom the physis to the subchondralcartilage-bone interface. Femoralheadossification is stopped, whilethe articular cartilage, nourished bysynovial fluid, continues to grow,resulting in the radiographicappearance of a small ossific nucleuswith a widened cartilage space. Inthe second stage, resorption ofnecrotic bone and creeping substitutionof the necrotic bone by vascularconnective tissue occur slowly overa 1- to 3-year period. The third stageis characterized by the appearance ofosteoblasts, followed by new-boneformation and healing. This stage isprobably concurrent with revascularization.Collapse of the necrotic subchondralbone leads to loss of femoralheadheight, as well as hip pain andsynovitis. After collapse, a renewedcycle of bone resorption and repairbegins. Some authors believe thereare no symptoms until a subchondralpathologic fracture occurs andthat, without collapse, ÒsilentÓinfarcts may result in nothing morethan a growth-arrest line or a Òheadwithin-a-headÓ radiographic appearance.However, in their study,Salter and Thompson5 found that77% of children with Legg-CalvŽ-Perthes disease did not have evidenceof subchondral fractures onradiographs obtained at the time ofdiagnosis.The collapse of a subchondral fracturecan lead to flattening of the ossificregion of the femoral head. The articularcartilage remains intact initially.Healing usually results, with coxamagna. Growth disturbances willoccasionally occur if isolated areas ofthe physis are affected. Physealinvolvement is variable, but the anteriorportion of the physis is most commonlyaffected. As the disease issecondary to multiple infarcts of varyingage, multiple stages of infarctionor repair may be present at any onetime. It is interesting that Catterall etal4 also found that the unaffected hipdemonstrates thickening of the articularcartilage with irregular staining, aswell as thinner physeal cartilage withirregular cell columns, suggestingthat some preexisting condition contributesto susceptibility to Legg-CalvŽ-Perthes disease.We do not recommend routinelaboratory tests for endocrine orhematopoietic disorders in otherwisehealthy children with Legg-CalvŽ-Perthes disease.Clinical PresentationThe usual age at presentation is 4 to10 years. The condition is more frequentin boys than in girls, andaffected children are often small fortheir age. The child usually haslimped for a few weeks or months.There is often no pain; when present,pain is usually mild and can affectthe knee. Because the hip is innervatedby three nerves, the pain maybe referred to the suprapatellarregion (femoral nerve), the medialthigh (obturator nerve), or the buttock(sciatic nerve). In many cases,diagnosis is delayed due to failure toexamine the hip when the patientcomplains of knee pain. Somepatients report having sufferedacute trauma when the pain started.Limited internal rotation andabduction of the hip are the most consistentfindings. Limitation of internalrotation is best tested with the hipin extension. Limited hip motionearly in the disease is due to musclespasm and synovitis; late in the disease,it may be due to bone impingementof the femoral head on theacetabulum. Gait may be antalgicand may exhibit a gluteus mediuslurch. The Trendelenburg test isoften positive, and quadriceps atrophyin the affected leg is common.The affected leg may appear significantlyshorter due to an adductioncontracture; however, significantshortening is not common unlesssevere coxa plana has developed.Radiologic Evaluation andStagingThe diagnosis is generally made andthe course of the disease is followedwith anteroposterior and frog-leglateral radiographs of the pelvis.The child often has a delayed boneage. When the condition is bilateral,which occurs in 10% to 20% of cases,the hips may be at different stages.The first radiograph should beobtained without the use of a shield,to exclude the presence of pelvicabnormalities.The radiographic appearance ofLegg-CalvŽ-Perthes disease may bemimicked by other conditions,including multiple epiphyseal dysplasia,spondyloepiphyseal dysplasia,thyroid disease, GaucherÕsdisease, and trichorhinophalangealsyndrome, as well as by corticosteroidusage. Numerous radiologicsigns and classification systems havebeen described.10 Journal of the American Academy of Orthopaedic SurgeonsLegg-Calvé-Perthes DiseaseWaldenstrom classified radiologicfindings on the basis of the evolutionaryphase of the disease. Initially, thefemoral head is radiodense andsmaller, while the cartilage space ofthe hip is wider. The increased radiodensityoccurs because the surroundingbone has a normal blood supply,thus appearing osteopenic comparedwith the avascular segment. After asubchondral fracture, the fragmentationstage follows. The radiographicappearance of lateral fragmentationof the femoral epiphysis is caused byongoing necrotic bone resorption andnew-bone formation. This leads intothe healing phase, when furtherreossification occurs and radiodensitybecomes normal. The residualdeformity may be coxa magna, coxaplana, or coxa breva. TheWaldenstrom classification demonstrateswhich stage of the disease ispresent, but has no predictive valuefor long-term outcome or treatment.The Catterall classification ,which is the most commonly used,defines four groups, primarily on thebasis of the amount of involvementof the femoral head and the presenceof radiographic Òat riskÓ signs.6Group I shows anterior centralinvolvement of the head and nometaphyseal reaction. Group II shows about 50% involvementanterolaterally, with the medial andlateral-pillar portions of the femoralhead intact, plus anterolateral metaphyseallesions. In group III, about75% of the head is involved, includingthe lateral column, with diffusemetaphyseal reaction. In group IV, 1 The Catterall classification of Legg-CalvŽ-Perthes disease. In group I there is involvement (hatched areas) of the anterior head only,no sequestrum, and no collapse of the epiphysis. In group II, only the anterior head is involved, and there is a sequestrum with a clear junction.In group III only a small part of the epiphysis is not involved. In group IV there is total head involvement.the entire head is involved, with diffuseor central metaphyseal reaction.Catterall suggested that groups I andII have favorable outcomes withouttreatment, while groups III and IVhave a poorer prognosis and requiretreatment. Initially established froma retrospective review, the Catterallclassification has the disadvantagethat the appropriate group designationmay appear to change as the diseaseproceeds.After reviewing the radiographs of1,264 children with Legg-CalvŽ-Perthes disease, Salter and Thompson5concluded that the extent of the subchondralfracture correlated preciselywith the subsequent extent of maximumresorption, potentially providingan early means of predicting theeventual extent of femoral-headinvolvement. The subchondral fracture,or Òcrescent sign,Ó is a transientphenomenon in the early stages of thedisease, lasting 2 to 9 months.Radiographic visualization of such afracture may herald the onset of symptomsin a previously asymptomaticprocess.The Salter-Thompson classificationis based on the extent of subchondralfracture. In group A, less than half ofthe femoral head is involved; in groupB, more than half of the femoral headis involved. The advantages of thisclassification system are early applicabilityand simplicity. The major disadvantageof the Salter-Thompsonclassification is that 77% of the patientsin their series did not have radiographsshowing a subchondral fracture.As the study was retrospective,it was not possible to obtain replacementsfor poor-quality radiographs,which might have been more likely toshow a crescent sign. Despite theirdrawbacks, both the Catterall andSalter-Thompson classifications seemto have prognostic value.7A relatively new classification hasbeen proposed by Herring et al8(Fig. 2). On an anteroposteriorradiograph of the pelvis obtained inthe early fragmentation phase, theheight of the lateral epiphyseal pillaris compared with the height of thenormal contralateral epiphysis. Ingroup A, there is no collapse of thelateral pillar. In group B, the lateralpillar maintains at least 50% of itsoriginal height. In group C, there iscollapse of the lateral pillar, withloss of more than 50% of its originalheight. In a retrospective analysis,no hip in group A had progressivefemoral-head flattening, while onlythe oldest patients in group B (meanage at onset of disease, 10 years) hada flattened femoral head. Patients ingroup C were most likely to exhibitprogressive flattening during reossification(incidence, 17%). This studyalso showed that, contrary to previousbelief, the femoral head maycontinue to deform during the 3- to4-year period of reossification.Ritterbusch et al9 compared theHerring lateral-pillar classificationand the Catterall classification andfound that the lateral-pillar classificationwas a better predictor of longtermoutcome and had greaterinterobserver reliability. We currentlyfavor the lateral-pillar classification.Other Imaging StudiesThe initial radiographic changes ofLegg-CalvŽ-Perthes disease may beabsent at the onset of symptoms.Technetium-99m bone scanningwith use of a pin-hole collimatormay allow earlier diagnosis thanradiography. Early in the course ofthe disease, a bone scan displaysÒcold spots,Ó representing avascularareas that involve a significant partof the femoral head, while thefemoral physis and acetabular rimshow normal radioisotope uptake.During revascularization, thefemoral head demonstrates increasedisotope uptake.Revascularization of bone can occurby recanalization of existing vesselsor by neovascularization throughthe development of new vessels.Recanalization occurs rapidly (minutesto weeks), whereas neovascularizationis a prolonged process(months to years). The bone-scan12 Journal of the American Academy of Orthopaedic SurgeonsLegg-Calvé-Perthes DiseaseGroup A Group B Group CFig. 2 The Herring classification of Legg-CalvŽ-Perthes disease. In group A, the lateral pillarretains its original height and shows slight radiographic changes. In group B, the lateralpillar may show density changes and height loss, but retains at least 50% of its originalheight. In group C, the lateral pillar is characterized by radiolucencies and collapse to lessthan 50% of its original height.pattern of revascularization thatbegins with visualization of the lateralepiphyseal column appears torepresent recanalization of themedial circumflex artery. Thisprocess occurs relatively rapidly andcarries a good prognosis. A scintigraphicappearance of base filling,indicating extension of radioactivitythrough the growth plate into thebase of the epiphysis by new vessels,represents neovascularization,which carries a poorer prognosis.10Magnetic resonance (MR) imagingmay depict bone infarctionbefore radiography does. In addition,MR imaging provides directvisualization of articular and physealcartilage and can be used to estimatefemoral-head sphericity betterthan radiography. However, MRimaging has not proved advantageousover radiography for seriallyfollowing the disease process and iscertainly more expensive. As allstudies to date are based on radiographiccriteria, it is unclear at thistime whether bone-scan or MRchanges warrant treatment.Arthrography is an excellentdynamic study in that it allows visualizationof femoral-head shape andhip-joint congruency through thefull range of motion of the hip.Arthrography can demonstrate thebest position for containment if thefemoral head is congruent and candepict the presence of hinge abductionif the femoral head is deformed.Natural HistoryMost reviews of patients with Legg-CalvŽ-Perthes disease have theinherent problems of retrospectivereviews, including grouping of varioustreatments and degrees of diseaseseverity. However, in follow-upstudies into late middle age ofpatients with Catterall group II, III,or IV disease, 70% to 90% wereactive and free of significant pain,with good range of motion, althoughtheir radiographs were often abnormal.Typical of these studies is thatby Gower and Johnston.11 Theyevaluated 36 patients (average age,45 years) 36 years after the onset ofsymptoms of Legg-CalvŽ-Perthesdisease. Of the 30 patients who weretreated nonoperatively, 86% hadscores on the Iowa Hip Rating Scaleof more than 80 points, and 8% hadbeen treated with arthroplasty.This same group of patients waslater reexamined by McAndrew andWeinstein,1 when the patients had anaverage age of 55 years. The numberof patients with Iowa hip scores over80 points had fallen to 40%. Forty percentof the patients had undergonetotal hip arthroplasty, and an additional10% had disabling osteoarthritis.A strong correlation between lossin width of the joint space (as seenradiographically) and the change inthe Iowa hip score was noted. Othermeasurements of femoral-head andacetabular deformity and congruencyhad not changed over this follow-upperiod. It was concluded that theobserved deterioration in functioncould be attributed solely to the onsetof osteoarthritis.It should be noted that these twostudies evaluated the results of noncontainmenttreatment. No similarlong-term studies after containmenttreatment have been published.PrognosisA number of radiographic at-risksigns that have been discusseddeserve brief mention. Catterall proposedthat patients are at risk for apoor prognosis if the followingradiographic findings are present:GageÕs sign (a radiolucent ÒVÓ onthe lateral side of the epiphysis), calcificationlateral to the epiphysis, lateralsubluxation of the femoral head,a horizontal physis, and metaphysealcysts. Some authors have consideredÒmetaphysealÓ cysts to beepiphyseal changes superimposedon the metaphysis by three-dimensionalradiographic distortion.12While these authors also argue thatthere has never been clinical oranatomic confirmation of metaphysealcysts, Catterall6 found metaphyseallesions consisting of nonossifiedcartilage in autopsy studies.While Òhead at riskÓ signs havebeen thought to be prognostic indicators,two recent reports, eachinvolving 100 patients, concludedthat no head-at-risk sign other thanlateral subluxation was important asa prognostic indicator.7,13 In a longtermfollow-up study, McAndrewand Weinstein1 found that three radiologicmeasurements correlated witha poor clinical outcome: two or moreCatterall head-at-risk signs, coxamagna (in which the femoral head islarger than normal by 10% or more,as described by Stulberg et al14), anda decrease in joint space in later life.1Coxa magna, coxa breva resultingfrom growth arrest, coxa plana, andacetabular deformities are associatedwith healed Legg-CalvŽ-Perthes disease. Stulberg et al14correlated the radiographic deformityof the femoral head at skeletalmaturity with the eventual functionaloutcome. Femoral heads thatwere flat yet congruent with theacetabulum were at risk for arthritisin the sixth decade of life. As thenormal ball-and-socket joint deformsto a flattened cylinder, the hiploses abduction and rotation capability,while retaining flexion andextension potential. If the femoralhead is flat and is not concentric withthe acetabulum, early severe arthritisoccurs. Hinge abduction andanterior impingement are knownsequelae of a flat, incongruentfemoral head.One of the most important prognosticindicators is limitation of hipmotion. In the initial stages of thedisease, synovitis is responsible loss of hip motion. If the hip remainsstiff for a prolonged period, deformationof the hip joint may develop.Long-term studies have shown thatthe earlier the onset of the disease,the better the long-term prognosis.The age of 6 to 8 years appears to bethe dividing line,1,14 perhaps becausethe acetabulum and femoral headlose some remodeling potential asthe child ages.Salter and Thompson5 found nosignificant difference in outcomebetween Catterall groups I and II(satisfactory outcomes in 100% and93% of patients, respectively) butnoted worse results in group III hips(60% to 76% satisfactory) and groupIV hips (41% to 61% satisfactory).This study appears to support thetreatment of Catterall group III andIV hips but only observation ofgroup I and II hips.Another prognostic indicatorinvolves the lateral pillar. Becausethe intact lateral column protects theepiphysis, it has been shown to be afavorable prognostic indicator.5,8,9TreatmentEarly treatment of Legg-CalvŽ-Perthes disease focused on relieffrom weight-bearing until thefemoral head had reossified.Prolonged bed rest, traction, spicacasts, and special frames and slingswere used, but the results of treatmentwere difficult to assess accurately.The results of current Legg-CalvŽ-Perthes treatment are difficultto evaluate. It is still uncertain whoneeds treatment and who does not.The most commonly used classificationsystem, that of Catterall, mayhave high interobserver variation.In the studies published to date,there is a lack of uniformity of criteriafor selection, treatment, and evaluation,and, more important, thereare few control studies.15 In addition,no studies reported to datehave been prospective or randomized,although a multicenter prospectivestudy is currently ongoing.In a recently published in-depthreview of the literature on Legg-CalvŽ-Perthes management,Herring15 cited the inconclusivenessof the findings.Today, maintenance of hip motionand containment of the involvedfemoral head form the bases for treatmentof Legg-CalvŽ-Perthes disease.However, the 80% good results in a22-year follow-up at an institutionwhere noncontainment treatmentwith a weight-relieving sling or harnesswas used16 have been as good asany reported.The initial goal of treatment is torestore hip mobility and relievepain. The next step is to decide whatpercentage of the femoral head isinvolved and whether active treatmentwill be needed. Containmenttreatment should not begin untilabduction and internal rotation havebeen restored to a near-normal state.Several days of bed rest at home or inthe hospital may be needed toreduce the symptoms. Traction maybe used to help keep the child off hisor her feet. Physical therapy often isuseful to help maintain hip motion.Adductor tenotomy or serial Petrieabduction casts may be necessary inmore resistant cases.Once hip mobility has improved,containment treatment can be considered.The theory behind containmentis that development of a congruentjoint is dependent on maximal contactbetween the immature femoral headand acetabulum. This theory hasbeen proved clinically and in the laboratoryin developmental dysplasiaof the hips, although the results forLegg-CalvŽ-Perthes disease remainsomewhat difficult to interpret.Before beginning any form of containmenttreatment, it is essential to makecertain that the hips can be containedcongruently and that there is no hingeabduction due to femoral-head flattening.Arthrography is particularlyuseful if abduction radiographs takenin the position of proposed containmentare equivocal.For the purpose of discussion, wewill divide Legg-CalvŽ-Perthes diseasepatients into two categories onthe basis of the extent of involvementof the femoral head. Childrenwith less femoral-head involvementare those with Catterall group I or II,Salter-Thompson group A, orHerring type A or B disease. Themajority of these patients have excellentresults regardless of the type oftreatment or lack thereof. Childrenwith more femoral-head involvement(i.e., those with Catterall groupIII or IV, Salter-Thompson group B,or Herring type C disease) oftenhave later disability if untreated.Maintenance of hip range of motionis probably all the treatmentrequired for patients with lessinvolvement, especially those under6 years of age. No containmenttreatment is indicated unless thereare poor prognostic indicators, suchas persistent or recurrent synovitis,lateral subluxation, involvement ofthe lateral pillar, or involvement ofnearly 50% of the femoral head. Werecommend some form of containmenttreatment for most childrenaged 6 years or older in whom thereare more than two poor prognosticindicators. Children with moreinvolvement should receive containmenttreatment. Our preference is touse bracing for children under 7years and surgical containment forthose over 7 years.Nonoperative ContainmentContainment methods that prohibithip motion, such as the use ofspica casts, are not recommended.Motion is a prerequisite for containmenttreatment and cannot beoveremphasized during the courseof treatment. Petrie casts (two longlegcasts separated by an abduction14 Journal of the American Academy of Orthopaedic SurgeonsLegg-Calvé-Perthes Diseasebar) have been used as a means ofcontainment, although medialfemoral condyle flattening has beendescribed with prolonged use.These and other long-leg orthosespromote abduction while allowingmotion at the hip and ambulation.Petrie casts have been largelysuperseded by shorter orthoses. Themost popular of these is the Scottish-Rite Hospital orthosis, which consistsof two thigh cuffs connected byan abduction bar, with joints toallow motion at the hips and knees,promote ambulation, and containthe femoral head. It is generally welltolerated by patients under 7 yearsand well accepted by their parents.Deciding when to remove the braceis a problem, as there is no clear evidenceshowing when containmentceases to be beneficial. The patient isgenerally weaned off use of theorthosis when radiographs showthat the disease is in the reparativestage, as shown by reossification ofthe lateral epiphysis.15 Althoughtwo recent reviews of the use of theScottish-Rite Hospital orthosis inpatients with Catterall group III orIV disease (mean age at diagnosis, 6and 7 years) failed to show anyadvantage over either no treatmentor other means of treatment,17,18 thepossibility of avoiding surgery hasled many pediatric orthopaedists tocontinue to use this device inyounger children with Legg-CalvŽ-Perthes disease.Operative ContainmentThe relative advantages of operativeover nonoperative treatment arealso unclear. Many have reportedthat operative treatment provides animproved outcome compared withthe natural history in children withmore extensive femoral-headinvolvement, although rigorouscontrols have been lacking. A comparativestudy of varus-derotationosteotomy and ambulation-abductionbracing in children over 6 yearsrevealed no difference in results.19Surgical containment can beachieved by proximal femoral varusosteotomy, innominate osteotomy,an acetabular-shelf procedure, orsome combination of femoral andpelvic procedures. However, itmust be remembered that in a normalhip the femoral head is only 63%Òcontained,Ó as the femoral head is120% of a hemisphere while theacetabulum is 75% of a hemisphere.20 Furthermore, which 63%of the femoral head is in contact withthe acetabulum is a function of thephase of gait as the joint changesposition. One review of 72 patientstreated with a femoral or innominateosteotomy disclosed no difference inresults at follow-up.21 There are theusual risks of surgery, and a secondoperation is often necessary toremove hardware. The advantagesof surgery include definitive treatment,a short period of limited activity,and the avoidance of psychosocialissues arising from prolonged bracewearing.Salter22 recommends the innominateosteotomy for children 6 yearsor older with Salter-Thompsongroup B involvement and subluxationof the femoral head in a weightbearingposition. His recommendedcriteria for surgery are pain-free fullrange of motion of the hip and nodeformation of the femoral head.While studies of the innominateosteotomy may have slightly betterresults than other surgical series, it islikely due to the above-mentionedselection criteria. The Salter innominateosteotomy provides a practicallimit of 25 degrees of anterior coverageand a range in lateral coverage of5 degrees at heel strike to 15 degreesat midstance, with an increase inanterior coverage at the expense ofless posterior coverage.20 To ensurepreservation of motion, a psoas andpartial-adductor tenotomy shouldaccompany the procedure, a postoperativehip spica cast should beavoided, and physical therapyshould be begun early.Varus-derotation osteotomy ofthe proximal femur may providecontainment if an abduction-internalrotation radiograph of the hip showscontainment. A dynamic arthrogramis recommended to find thebest position, as the osteotomy cancorrect for rotation, varus/valgus,and flexion/extension as needed.Varus correction should be limited to20 degrees, because complicationssuch as a short leg and abductorlurch are more pronounced withgreater varus angulation, althoughboth usually improve with time.One may expect 15 to 20 degrees ofincreased lateral coverage and about10 degrees of improved anterior coveragefrom a 15-degree-varus, 15-degree-derotation osteotomy of theproximal femur. While this osteotomytheoretically doubles theshear force across the proximalfemoral physis,19 adverse clinicaleffects have not been reported.As there are limitations to theamount of femoral-head coveragethat can be obtained with bothinnominate and intertrochantericosteotomies, a combined proceduremay be indicated in severe subluxation.Theoretically, the potentiallengthening and increased joint pressureresulting from an innominateosteotomy may be avoided by performingan intertrochanteric osteotomy,which provides both shorteningand a decrease in joint-surface pressure.Both osteotomies may be performedthrough a single ilioinguinalapproach or through separate incisions.Use of this combination hasbeen reported in patients withCatterall group III or IV involvement;seven of nine patients had clinicallygood results after a mean follow-upperiod of 50 months.23Shelf arthroplasty has been recommendedfor children over the ageof 8 years who have Catterall II, III, or IV disease, to prevent subluxationand increase acetabularcoverage. Compared with an agematchedcontrol group treated nonoperatively,patients who underwenta shelf arthroplasty had better hipmotion and coverage 2 years aftersurgery.24 Hardware removal is notneeded with this procedure.Hinge abduction occurs laterwhen an enlarged femoral head is laterallyextruded and impinges againstthe lateral acetabular rim on abduction,causing pain. A medial dye poolis seen on arthrography. A valgusosteotomy is indicated if arthrographyshows congruence of the medialfemoral head with the acetabulum asthe hip is adducted, provided a functionalamount of adduction wouldremain after the osteotomy. Chiariosteotomies performed as salvageprocedures in the older child with littleremodeling potential serve toincrease the load-bearing area, whichmay offer several years of pain-freehip function before further reconstructivesurgery is needed.SummaryAlthough the literature is inconclusiveon indications for treatment andwhat type of treatment to use, somerecommendations seem warranted.Regaining and preserving hipmotion cannot be overstressed andis certainly indicated before any containmenttreatment. Poor prognosticfactors include the onset ofsymptoms after age 8 years, lateralhip subluxation, involvement ofmore than 50% of the femoral headwith collapse of the lateral pillar,and an aspherical femoral head withan incongruent joint. In the presenceof any of these factors, it is unlikelythat containment treatment willharm the patient, and it certainlymay help. As long-term studieshave shown that in 50% of untreatedpatients disabling arthritis of the hipdevelops in the sixth decade of life, Legg-Calvé-Perthes Disease
References :
McAndrew MP, Weinstein SL: A longtermfollow-up of Legg-CalvŽ-Perthesdisease. J Bone Joint Surg Am 1984;66:860-869.2. Gregosiewicz A, OkonÕski M, Stol/eckaD, et al: Ischemia of the femoral head inPerthesÕ disease: Is the cause intra- orextravascular? J Pediatr Orthop 1989;9:160-162.3. Neidel J, Boddenberg B, Zander D, et al:Thyroid function in Legg-CalvŽ-Perthesdisease: Cross-sectional and longitudinalstudy. J Pediatr Orthop 1993;13:592-597.4. Catterall A, Pringle J, Byers PD, et al: Areview of the morphology of PerthesÕ disease.J Bone Joint Surg Br 1982;64:269-275.5. Salter RB, Thompson GH: Legg-CalvŽ-Perthes disease: The prognostic significanceof the subchondral fracture and atwo-group classification of the femoralhead involvement. J Bone Joint Surg Am1984;66:479-489.6. Catterall A: The natural history ofPerthesÕ disease. J Bone Joint Surg Br1971;53:37-53.7. Mukherjee A, Fabry G: Evaluation of theprognostic indices in Legg-CalvŽ-Perthes disease: Statistical analysis of 116hips. J Pediatr Orthop 1990;10:153-158.8. Herring JA, Williams JJ, Neustadt JN, etal: Evolution of femoral head deformityduring the healing phase of Legg-CalvŽ-Perthes disease. J Pediatr Orthop1993;13:41-45.9. Ritterbusch JF, Shantharam SS, GelinasC: Comparison of lateral pillar classificationand Catterall classification ofLegg-CalvŽ-PerthesÕ disease. J PediatrOrthop 1993;13:200-202.10. Conway JJ: A scintigraphic classificationof Legg-CalvŽ-Perthes disease.Semin Nucl Med 1993;23:274-295.11. Gower WE, Johnston RC: Legg-Perthesdisease: Long-term follow-up of thirtysixpatients. J Bone Joint Surg Am1971;53:759-768.12. Hoffinger SA, Rab GT, Salamon PB:ÒMetaphysealÓ cysts in Legg-Calve-PerthesÕ disease. J Pediatr Orthop1991;11:301-307.13. Yrjšnen T: Prognosis in PerthesÕ diseaseafter noncontainment treatment: 106hips followed for 28Ð47 years ActaOrthop Scand 1992;63:523-526.14. Stulberg SD, Cooperman DR,Wallensten R: The natural history ofLegg-CalvŽ-Perthes disease. J Bone JointSurg Am 1981;63:1095-1108.15. Herring JA: The treatment of Legg-CalvŽ-Perthes disease: A critical reviewof the literature. J Bone Joint Surg Am1994;76:448-458.16. Kelly FB Jr, Canale ST, Jones RR: Legg-CalvŽ-Perthes disease: Long-term evaluationof non-containment treatment. JBone Joint Surg Am 1980; 62:400-407.17. Meehan PL, Angel D, Nelson JM: TheScottish Rite abduction orthosis for thetreatment of Legg-Perthes disease: Aradiographic analysis. J Bone Joint SurgAm 1992;74:2-12.18. Martinez AG, Weinstein SL, Dietz FR:The weight-bearing abduction brace forthe treatment of Legg-Perthes disease. JBone Joint Surg Am 1992;74:12-21.19. Evans IK, Deluca PA, Gage JR: A comparativestudy of ambulation-abductionbracing and varus derotationosteotomy in the treatment of severeLegg-CalvŽ-Perthes disease in childrenover 6 years of age. J Pediatr Orthop1988;8:676-682.20. Rab GT: Containment of the hip: A theoreticalcomparison of osteotomies.Clin Orthop 1981;154:191-196.21. Sponseller PD, Desai SS, Millis MB:Comparison of femoral and innominateosteotomies for the treatment of Legg-CalvŽ-Perthes disease. J Bone Joint SurgAm 1988;70:1131-1139.22. Salter RB: The present status of surgicaltreatment for Legg-Perthes disease. JBone Joint Surg Am 1984;66:961-966.23. Olney BW, Asher MA: Combinedinnominate and femoral osteotomy forthe treatment of severe Legg-CalvŽ-Perthes disease. J Pediatr Orthop 1985;5:645-651.24. Willett K, Hudson I, Catterall A: Lateralshelf acetabuloplasty: An operation forolder children with PerthesÕ disease. JPediatr Orthop 1992;12:563-568.