Injury to the spinal cord without radiological abnormality (SCIWORA) in adults

Abstract

Injury to the spinal cord without radiological abnormality often occurs in the skeletally immature cervical and thoracic spine. We describe four adult patients with this diagnosis involving the cervical spine with resultant quadriparesis. The relevant literature is reviewed. The implications for initial management of the injury, the role of MRI and the need for a high index of suspicion are highlighted.

Full text

1034 THE JOURNAL OF BONE AND JOINT SURGERY
P. Kothari, MS(Orth), Specialist Orthopaedic Registrar
B. Freeman, FRCS Orth, Senior Spinal Fellow
M. Grevitt, FRCS Orth, Consultant Spinal Surgeon
Centre for Spinal Studies and Surgery
R. Kerslake, FRCR, Consultant Radiologist
Department of Diagnostic Radiology
University Hospital, Queen’s Medical Centre, Nottingham NG7 2UH,
UK.
Correspondence should be sent to Mr M. Grevitt.
©2000 British Editorial Society of Bone and Joint Surgery
0301-620X/00/710641 $2.00
Injury to the spinal cord without radiological
abnormality (SCIWORA) in adults
P. Kothari, B. Freeman, M. Grevitt, R. Kerslake
From the University Hospital, Nottingham, England
I
njury to the spinal cord without radiological
abnormality often occurs in the skeletally immature
cervical and thoracic spine. We describe four adult
patients with this diagnosis involving the cervical
spine with resultant quadriparesis. The relevant
literature is reviewed. The implications for initial
management of the injury, the role of MRI and the
need for a high index of suspicion are highlighted.
J Bone Joint Surg [Br] 2000;82-B:1034-7.
Received 12 October 1999; Accepted after revision 19 April 2000
Since its first description by Pang and Wilberger,
1
injury to
the spinal cord without radiological abnormality (SCI-
WORA) has been well documented in the paediatric lit-
erature.
2,3
The relatively large size of the head and the
greater inherent mobility in the immature axial skeleton,
combined with ligamentous laxity or disruption, render the
spinal cord vulnerable to damage in high-energy trauma. In
the absence of osseous injury on plain radiographs or
tomography, MRI should demonstrate abnormalities in all
cases. SCIWORA in adults is rare, but is of considerable
importance because of the potential problems of manage-
ment inherent in the diagnosis. We report four such cases,
highlighting common clinical features, and consider their
relevance in general trauma practice.
Case Reports
Case 1. A 24-year-old male motorcyclist came off the road
at high speed. He wore no helmet and had a severe head
injury. When initially assessed at the receiving hospital the
plain radiographs of the neck were normal apart from a
narrow spinal canal (Fig. 1a). On regaining consciousness
he was noted to be quadriparetic with MRC grade-II power
in most muscle groups. Apart from some dysaesthesia in
the upper limbs there were no other neurological symptoms
or signs. He was transferred to our unit 48 hours later, by
which time power in the upper limbs had recovered to
MRC grade III and in the lower limbs to grade IV. MR
images showed mild, focal swelling of the cord and oedema
at C3/4, with prevertebral soft-tissue swelling and disrup-
tion of the anterior longitudinal ligament. In addition, there
was low signal intensity in the C3/4 disc with a shallow,
posterior bulge without focal compression of the cord.
There was oedema in the posterior ligamentous complex at
the same level and at C4/5. Abnormalities were seen in the
horizontal linear signal in the bodies of T2 to T6 indicative
of undisplaced compression fractures (Fig. 1b). A diagnosis
of central cord syndrome was made. The patient made
further neurological recovery, and repeat MRI two months
after the injury showed resolution of the oedema in the cord
but persistent signal changes which were suggestive of
myelomalacia.
Fig. 1a Fig. 1b
Case 1. Figure 1a – Lateral plain radiograph of the cervical spine showing
minimal prevertebral soft-tissue swelling at C3/4 (arrows). No fracture is
apparent. The alignment of the vertebral bodies is normal. Figure 1b –
Sagittal MRI (T2-weighted) showing a small prevertebral haematoma with
elevation of the anterior longitudinal ligament. A shallow posterior bulge
of the C3/4 disc is apparent with focal oedema. The alignment of the
vertebral bodies is normal. There are undisplaced linear fractures in the
upper thoracic spine (arrow). Other images show posterior ligamentous
disruption at C3/4 and C4/5 indicating a predominantly hyperflexion
injury.

Case 2. A 33-year-old woman was admitted with quad-
riparesis after a fall from a ladder. Physical examination
revealed a sensory level at C5, MRC grade-II power in the
upper and lower limbs but no tenderness in the neck. Plain
radiographs showed congenital fusion at C2/3 and C7/T1
(Fig. 2a). No bony injury was apparent. She was given
high-dose intravenous methylprednisolone. MRI showed
marked constitutional narrowing of the cervical spinal
canal from C3 to C7 with oedema of the cord between C4
and C7. There was focal compression of the cord at C5/6
caused by a spondylotic ridge. Posterior ligamentous oede-
ma was apparent at the C4/5 and C5/6 segments with more
diffuse soft-tissue oedema at the upper cervical levels (Fig.
2b).
As there was no early evidence of neurological improve-
ment she underwent laminectomy from C3 to C6; no dural
tears were seen at operation. She was transferred to the
regional spinal-injuries unit where she recovered MRC
grade-IV power in the upper and lower limbs and regained
some sphincter control.
Case 3. An 18-year-old man was physically assaulted
while drunk. He sustained a severe head injury with a
Glasgow Coma Score of 7 on admission. A CT scan of the
brain revealed contusion of the left parietal lobe. The
patient’s level of consciousness returned to normal 48 hours
later but he was quadriparetic. Review of the initial cervical
radiographs was thought to show a minimal anterior com-
pression fracture of C6 with slight rotation (Fig. 3a).
On transfer to our unit he was noted to have a sensory
level at C5 with no active movement in any of his four limbs
but no posterior cervical tenderness. Catheter tug sensation
was present but the anus was patulous. The bulbocavernosus
reflex was present at that stage. MRI showed oedema of the
mid-cervical cord and focal haemorrhage at C5/6 (Fig. 3b).
There was a shallow, diffuse disc bulge at the same level.
Subtle horizontal, linear oedema was seen in the vertebral
bodies of C5 and C6 without evidence of discrete fractures.
There was associated disruption of the posterior ligament at
C4/5, C5/6 and C6/7 with diffuse soft-tissue oedema in the
upper cervical region. He was managed conservatively but
there was no neurological recovery.
Case 4. A 49-year-old unsecured male passenger in a rear
seat was ejected from a car in an accident at high speed. He
sustained serious head, maxillofacial and abdominal inju-
ries. Plain radiographs of the neck showed no bony injury.
After resuscitation he required a laparotomy at which lac-
erations of the liver were repaired.
He was transferred to our neurosurgical unit where he
was observed to have movement in all four limbs. Repeat
CT scans of the brain, and sections through C7/T1 which
had not been adequately demonstrated on the initial views,
were normal. An intracranial pressure transducer was inser-
ted and his maxillofacial fractures were reduced and stabi-
lised. After four days in intensive care, following
extubation and reversal of sedation, he was conscious
enough to complain of neck pain. He remained quad-
riparetic, however, with no useful motor function below
C5. MRI of the spine revealed a constitutionally narrow
canal. There was a small posterior disc protrusion at C5/6
with associated focal oedema of the cord at this level.
There was also injury of the mid-cervical posterior liga-
ment and horizontal, linear signal abnormalities in the
vertebral bodies of T3 and T4 (Fig. 4). Subsequent CT of
1035INJURY TO THE SPINAL CORD WITHOUT RADIOLOGICAL ABNORMALITY (SCIWORA) IN ADULTS
VOL. 82-B, N
O
. 7, SEPTEMBER 2000
Fig. 2a Fig. 2b
Case 2. Figure 2a – A lateral plain radiograph of the cervical spine
showing congenital fusion at C2/3 and C7/T1. Minor spondylotic changes
are apparent at C5/6 but no fracture is evident. The central canal is narrow.
Figure 2b – Sagittal MRI showing congenital fusion at C2/3 and C7/T1
and constitutional stenosis of the central canal. There are spondylotic
changes at the mobile mid-cervical levels and a shallow traumatic protru-
sion of the disc at C5/6, with associated oedema of the cord from C3 to
C7. There is diffuse soft-tissue oedema (arrow) indicative of a hyper-
flexion injury.
Fig. 3a Fig. 3b
Case 3. Figure 3a – Lateral plain radiograph of the cervical spine showing
a small anterosuperior corner fracture of C6 (arrow). Figure 3b – Sagittal
T2-weighted MRI. The images are degraded by movement artefact. The
area of low-signal intensity centrally within the cord at C5/6 is indicative
of haemorrhage and there is surrounding oedema. There is a shallow
posterior bulge of the C5/6 disc although the small anterior fracture of C6
cannot be readily appreciated. Other images show posterior ligamentous
injury indicative of a hyperflexion injury.

the entire cervical spine showed no fracture. There were
spondylotic changes at C4/5 and C5/6. Eight days after
injury supervised dynamic radiographs with the patient
awake, showed no evidence of instability; 15 days after
injury, there was neurological improvement to grade IV/V,
from the C7 myotome with some active movement of the
lower limbs (grade-II ankle dorsiflexion). He was later
transferred to the regional spinal rehabilitation unit.
Discussion
SCIWORA in the paediatric spine is a reflection of the
inherent elasticity of the soft tissues which ensures immedi-
ate spontaneous reduction after considerable intersegmental
displacement. The flexibility of the spine is reduced with
increasing age and skeletal maturation, and with it, the
likelihood of bone injury becomes greater.
SCIWORA in adults is not well documented. Chen et al
4
described five cases with traumatic central cord syndrome
and an abnormal cord signal on MRI but without definitive
compression of the cord. There is only one case of SCI-
WORA reported in the surgical literature affecting the adult
spine.
5
In this patient, myelography demonstrated gross
leakage of contrast and a subsequent CT scan suggested
rupture of the cord, confirmed on MRI which also showed
an interspinous ligamentous injury.
In the conscious patient, pain or neurological symptoms
will suggest the possibility of injury to the cervical spine.
Cervical injuries are likely to be missed in multiple trauma,
craniofacial injuries and drug- or alcohol-related accidents;
three of our cases illustrate the difficulties of clinical
assessment in such situations. Adequate spinal radiographs
are essential.
6
The cross-table lateral radiograph will dem-
onstrate 75% of fractures, with a sensitivity of 82% to 85%.
The diagnostic accuracy approaches 92% to 99% when all
cervical views (lateral, anteroposterior, open mouth and
oblique) are obtained.
7
CT will demonstrate subtle injuries
to the posterior arch or lateral masses and injuries in the
atlantoaxial region.
Despite the above data, a recent survey of UK practice
by Lockey, Handley and Willett
8
showed wide variation in
the protocols for reviewing the cervical spine in the uncon-
scious patient. Some units relied only on a cross-table
lateral radiograph. Where the C1/2 or C7/T1 areas were not
demonstrated on plain radiographs, there was often no
protocol for CT of the neck at the same time as examina-
tion of the brain. Lockey’s own departmental policy was to
take dynamic fluoroscopic views in the intubated or uncon-
scious patient to exclude instability. Bedside fluoroscopic
examination has been reported in other centres where the
incidence of cervical instability was 5% in patients with
otherwise normal radiological appearances.
9
Between 8% and 15% of cervical fractures are associated
with herniation of a disc which may compress the cord.
Benzel et al,
10
using MRI, found an incidence of 16% of
abnormality of a disc following trauma. In two of our cases
SCIWORA occurred in association with a constitutionally
narrow canal. We question the wisdom of performing
dynamic radiographs in this situation when demonstrable
instability may cause injury to the cord. Furthermore, as
our cases illustrate, there may not be documented move-
ment of the limbs before admission to hospital. If the
patient recovers consciousness with obvious quadriparesis,
the question arises as to whether the deficit was the result
of the initial accident or of the examination. There are
obvious medicolegal implications.
Prolonged cervical immobilisation with hard collars may
cause complications.
11
Increased intracranial pressure, care
of the skin and difficulties with nursing may demand early
removal of a hard collar, in a patient who is ventilated after
a head injury. Our experience leads us not to rely on plain
radiographs or CT to exclude cervical injury. In such
circumstances, it is safer to presume the presence of an
unstable spinal injury. MRI is useful in demonstrating
ligamentous injury and is being advocated for routine
review of the cervical spine.
12
The logistics of MRI in the
ventilated patient and the limited availability of this inves-
tigation throughout the UK prevents its use from being
obligatory. It is our policy that when MRI is not practical,
the collar is removed when the patient is sedated and
intubated. The neck is assumed to be unstable and sup-
ported with sand bags in neutral alignment. The collar is
reapplied when the patient is being weaned from the ven-
tilator, until formal neurological examination is performed.
Depending on the circumstances, dynamic radiography or
MRI is then performed.
MRI has been useful in this series as there were focal
1036 P. KOTHARI, B. FREEMAN, M. GREVITT, R. KERSLAKE
THE JOURNAL OF BONE AND JOINT SURGERY
Fig. 4
Case 4. Sagittal T2-weighted MRI showing a
shallow posterior protrusion of the disc at C5/6
with associated oedema of the cord. There are
minimally displaced linear fractures at T3 and
T4 (other images showed posterior ligamentous
injury indicative of hyperflexion injury at C5/
6).

changes in the cord, and ligamentous and disc injury. In
addition, relative narrowing of the spinal canal is well
demonstrated. Plain radiography and the use of the Pavlov
ratio have not been helpful in estimating relative dimen-
sions of the canal or the space available for the cord.
13
The
importance of narrowing of the canal in the context of
SCIWORA is that the mechanism of inducing the central
cord syndrome probably differs from that seen in the older
patient. In the latter case, the cord is compressed between
the enfolded ligamentum flavum and anterior vertebral
osteophyte in a hyperextension injury.
14
Our cases have
shown significant signal changes on T2-weighted images,
particularly in the posterior ligament complex. This implies
a hyperflexion-compression force, as described by Allen et
al.
15
In support of this is the linear high-signal changes in
the upper thoracic vertebrae indicative of axial loads trans-
mitted caudally at the moment of impact. The damage to
the cord may then occur as a consequence of traction,
forward translation and compression between a modest disc
bulge and the laminae. This concept has been supported by
the observations of Chen et al
4
and may be similar to that
described by Braakman and Penning,
16
who refer to it as a
‘hyperflexion sprain’.
The demonstration of changes in the cord has also been
useful in predicting recovery. The neurological outcome in
patients with focal oedema of the cord has been better than
in the patient with overt haematoma of the cord as seen on
MRI. These observations appear to confirm the findings of
Schaefer et al
17
in which focal oedema was associated with
motor recovery of 70%.
In conclusion, SCIWORA must be suspected in all
patients with a neurological deficit and apparently normal
radiographs. A high index of suspicion is necessary in
injured patients in whom movement of all limbs is not seen.
The presence of constitutional narrowing of the canal
would seem to predispose to lesions of the cord especially
in young adults with a compression-flexion type of injury.
MRI is a sensitive method of demonstrating ligamentous
damage and protrusion of the discs, which are important
features in the mechanics of injury to the cord. The use of
dynamic fluoroscopy in these patients may be dangerous.
No benefits in any form have been received or will be received from a
commercial party related directly or indirectly to the subject of this
article.
References
1. Pang D, Wilberger JE. Spinal cord injury without radiographic
abnormalities in children. J Neurosurg 1982;57:114-29.
2. Pang D, Pollack IF. Spinal cord injury without radiographic abnor-
mality in children: the SCIWORA syndrome. J Trauma
1989;29:654-64.
3. Dickman CA, Zabramski JM, Hadley MN, Rekate HL, Sonntag
VK. Paediatric spinal cord injury without radiographic abnormalities:
report of 26 cases and review of the literature. J Spinal Disord
1991;4:296-305.
4. Chen TY, Lee ST, Lui TN, et al. Efficacy of surgical treatment in
traumatic central cord syndrome. Surg Neurol 1997;48:435-40.
5. Hirsh LF, Duarte L, Wolfson EH. Thoracic spinal cord injury
without spine fracture in an adult: case report and literature review.
Surg Neurol 1993;40:35-8.
6. Davis JW, Phreaner DL, Hoyt DB, Mackersie RC. The etiology of
missed cervical spine injuries. J Trauma 1993;34;342-6.
7. West OC, Anbari MM, Pilgram K, Wilson AJ. Acute cervical spine
trauma: diagnostic performance of single view versus three-view
radiographic screening. Radiology 1997;204:819-23.
8. Lockey AS, Handley R, Willett K. ‘Clearance’ of cervical spine
injury in the obtunded patient. Injury 1998;29:493-7.
9. Sees DW, Rodriguez Cruz LR, Flaherty SF, Ci Ceri DP. The use of
bedside fluoroscopy to evaluate the cervical spine in obtunded trauma
patients. J Trauma 1998;45:768-71.
10. Benzel EC, Hart BL, Ball PA, et al. Magnetic resonance imaging for
the evaluation of patients with occult cervical spine injury. J Neu-
rosurg 1996;85:824-9.
11. Ajani AE, Cooper DJ, Scheinkestel CD, Laidlaw J, Tuxen DV.
Optimal assessment of cervical spine trauma in critically ill patients: a
prospective evaluation. Anaesth Intensive Care 1998;26:487-91.
12. Goldberg AL, Rothfus WE, Deeb ZL, et al. The impact of magnetic
resonance on the diagnostic evaluation of acute cervico-thoracic spinal
trauma. Skeletal Radiol 1988;17:89-95.
13. Blackley HR, Plank LD, Robertson PA. Determining the sagittal
dimensions of the canal of the cervical spine: the reliability of ratios of
anatomical measurements. J Bone Joint Surg [Br] 1999;81-B:110-2.
14. Northrup BE. Evaluation and early treatment of acute injuries to the
spine and spinal cord. In: The Cervical Spine Research Society
Editorial Committee. Cervical Spine, 3rd ed. Philadelphia: Lippin-
cott-Raven Publishers, 1998:545.
15. Allen BL Jr, Ferguson RL, Lehman TR, O’Brien RP. A mechanistic
classification of closed, indirect fractures and dislocations of the lower
cervical spine. Spine 1982;7:1-27.
16. Braakman R, Penning L. The hyperflexion sprain of the cervical
spine. Radiol Clin Biol 1968;37:309-20.
17. Schaefer DM, Flanders E, Osterholm JL, Northrup BE. Prognostic
significance of magnetic resonance imaging in the acute phase of
cervical spine injury. J Neurosurg 1992;76:218-23.
1037INJURY TO THE SPINAL CORD WITHOUT RADIOLOGICAL ABNORMALITY (SCIWORA) IN ADULTS
VOL. 82-B, N
O
. 7, SEPTEMBER 2000