Unusual Fracture Dislocation of the Craniovertebral Junction
Eric E. Awwad, David S. Martin, K. Charles Cheung, and Robert J. Bernardi
Summary: We present a rare case of C-1 fracture–posterior
atlantoaxial subluxation in which neither fracture of the dens nor
preexisting abnormality of C-1 or C-2 was present.
Index term: Atlas and axis
Bilateral posterior translatory atlantoaxial
subluxation is an extremely rare hyperextension
injury formerly described in the presence of a
dens fracture or preexisting lesion of C-1 or C-2.
A case of this rare subluxation associated with a
fracture of the anterior arch of C-1 is reported.
Case Report
A 47-year-old man was struck on the back of the lower
part of the neck by a falling tree, and became quadriplegic
and lost tactile sensibility on falling to the ground. Some
mild improvement in symptoms occurred during the next
10 to 20 minutes. No significant facial injuries were
present. Neurosurgical evaluation diagnosed central cord
syndrome, with injury at the C5-6 level. Radiographs
showed spondylosis at C5-6, a fracture of the C-7 spinous
process at the point of impact, and an unusual fracture
dislocation of the craniovertebral junction.
Additional studies were done. Complex motion tomo-
grams (Fig 1A) showed posterior dislocation of the basion
relative to the dens. Wackenheim’s line (1) (Fig 1B) was
separated from the posterior tip of the dens. The dens-
basion distance was increased. The Powers ratio (2) (Fig
1B) was normal and was compatible with posterior atlan-
tooccipital disassociation. The X-line method of Lee et al
(3) (Fig 1C) was not positive for atlantooccipital disloca-
tion in that the C-2/opisthion line was not displaced from
its intersection with the posterior arch of C-1. It was ab-
normal in that the basion/C-2 spinolaminar line was a
minimum of 8 mm from the dens. There was no dens
fracture. Axial computed tomography (Fig 2A) showed a
fracture of the anterior ring of C-1. The relationship of the
occipital condyles and lateral masses of C-1 was ana-
tomic. Subluxation of the C1-2 facet joints was present
(Fig 2B). Subsequent sagittal magnetic resonance images
Received October 18, 1993; accepted after revision February 8, 1994.
From the Departments of Radiology (E.E.A., D.S.M.) and Surgery, Division of Neurosurgery (R.J.B.), St Louis (Mo) University Health Sciences Center,
and Section of Neurosurgery, Southeast Missouri Hospital, Cape Girardeau, Mo (K.C.C.).
Address reprint requests to Eric E. Awwad, MD, Department of Radiology, St. Louis University Health Sciences Center, 3635 at Grand, PO Box 15250,
St Louis, MO 63110-0250.
AJNR 16:1155–1157, May 1995 0195-6108/95/1605–1155 q American Society of Neuroradiology
1155
after halo fixation showed stripping of the posterior longi-
tudinal ligament from the dorsal aspects of C-2 and C-3
with minimal extension of the tear into the posterior C3-4
interspace. The findings were considered most compatible
with a translatory bilateral posterior subluxation at the
atlantoaxial joint with a fracture of the anterior arch of C-1.
Intraoperatively, the ring of C-1 was found to be stable
relative to the base of the skull, but the ring of C-2 was
hypermobile. An occiput–to–C-2 fusion, performed by us-
ing a threaded Steinmann pin and Songer cables, caused
reduction of the subluxation. One month later, a partial
corpectomy and diskectomy was performed at C5-6.
At 6-month follow-up, marked improvement was
shown in the central cord syndrome. The patient was able
to walk without assistance and feed himself, and clinical
improvement continued. The craniovertebral junction re-
mained in good alignment with the patient wearing a cer-
vical collar.
Discussion
Previous authors have concluded that for a
bilateral translatory posterior subluxation at
C1-2 to occur, either a dens fracture or a pre-
existing abnormality of the dens or anterior arch
of C-1 is required (4 – 6). This includes destruc-
tion by tumor or infection, rheumatoid arthritis,
and congenital absence or maldevelopment.
This dens of our patient was normal. The
anterior fracture of the arch of the atlas was not
pathologic nor was there evidence for a preex-
isting defect. We believe this represents a force-
ful hyperextension injury in which the atlas frac-
tured instead of the dens. The stripping of the
posterior longitudinal ligament and a central
cord syndrome at the C5-6 level are compatible
with this mechanism (7). Most likely, the injury
was caused by hyperextension with pressure of
the anterior surface of the dens against the an-
terior arch. The facet joints at C1-2 disrupted,
allowing movement between the two vertebral
1156 AWWAD AJNR: 16, May 1995

Fig 1. A, Sagittal tomogram shows posterior displacement of the basion relative to the dens.
B, Line drawing of A. Wackenheim’s line (broken line) is drawn as a caudal extension of the posterior surface of the clivus. The line
should be tangent to the posterior tip of the dens. The skull base is displaced posteriorly relative to the cervical spine, so the line is
separated from the dens. If the displacement were anterior, the line would intersect the dens. The Powers ratio (solid lines) is useful in
assessing anterior occipitoatlantal dislocation. It is defined as the ratio of the lines BC/OA, where B is the basion, C is the posterior arch
of C-1, O is the opisthion, and A is the anterior arch of C-1. The normal ratio is less than 1. If the ratio is greater than 1, anterior
occipitoatlantal dislocation exists. The ratio may be normal with longitudinal distraction or posterior occipitoatlantal dislocation. An
associated atlas fracture invalidates the ratio.
C, With the method described by Lee, a line drawn from the basion to the midpoint of the spinolaminar line of C-2 should contact the
dens or be within 5 mm of it. A line from the posteroinferior corner of the body of C-2 to the opisthion should pass through the
spinolaminar line of C-1. Anterior atlantooccipital dislocation displaces both lines forward from their reference point; posterior dislocation
displaces them posteriorly; with longitudinal distraction, the descending limb is displaced posteriorly and the ascending line is displaced
anteriorly.

Fig 2. A, Axial computed tomogram

shows an anteriorly displaced fragment of
the anterior ring C-1 (C1) in anatomic posi-
tion relative to the odontoid (C2). There is a
good anatomic relationship between the oc-
cipital condyles (OC) and the lateral masses
of C-1 (C1LM). A few small fragments are
present posterior to the dens (arrows).
B, Axial computed tomogram shows an-
terior subluxation of C-2 relative to C-1.

segments and displacement. In this instance,
the facet joints remained subluxed after injury.
Such a mechanism implies that when a dis-
placed segment of the anterior arch is identified,
disruption of the facet joints at C1-2 should be
considered.
The most satisfactory treatment for this rare
and complex injury is not clearly defined. Arth-
rodesis between the opisthion, C-1, and C-2
after skeletal traction has been advocated, al-
though it is not known if the alignment will re-
main corrected (4, 5).
In summary, bilateral posterior translatory at-
lantoaxial subluxation is an extremely rare hy-
perextension injury. Formerly, it was thought
that the presence of a dens fracture or preexist-
ing lesion of C-1 or C-2 was required. This case
indicates that the lesion can also be produced if
the anterior arch of C-1 is fractured. The iden-
tification of a similar fracture without interfacet
subluxation should imply instability because of
capsular disruption of the interfacet joints with
spontaneous reduction.
References
1. Thiebaut F, Wackenheim A, Vrousos C. Un nouveau repère pour
le diagnostic des déplacements de la dent de l’axis dans les
traumatismes et les malformations de la charnière cervicooccipi-
tale. Rev Otoneuroophtalmol 1960;32:410
2. Powers B, Miller MD, Kramer RS, et al. Traumatic anterior atlanto-
occipital dislocation. Neurosurgery 1979;4:12–17
AJNR: 16, May 1995 FRACTURE DISLOCATION 1157

3. Lee C, Woodring JH, Goldstein SJ, Daniel TL, Young AB, Tibbs 6. Isdale IC, Corrigan AB. Backward luxation of the atlas: two cases
PA. Evaluation of traumatic atlantooccipital dislocations. AJNR of an uncommon condition. Ann Rheum Dis 1970;29:6 –9
Am J Neuroradiol 1987;8:19 –26 7. Schneider RC, Cherry G, Pantek H. The syndrome of acute central
4. White A, Panjabi M. Clinical Biomechanics of the Spine. 2nd ed. cervical spinal cord injury. J Neurosurg 1954;11:546 –577
Philadelphia: Lippincott, 1990:296
5. Lipson SJ. Cervical myelopathy and posterior atlanto-axial sub-
luxation in patients with rheumatoid arthritis. J Bone Joint Surg
Am 1985;67-A:593–597