T. Sagiuchia, S. Tachibanaa, K. Satoa, S.
Shimizua, I. Kobayashia, H. Okaa, K. Fujiia, S.
Kanb
Department of Neurosurgery,
Kitasato University School of Medicine,
Sagamihara, Kanagawa, Japan
SUMMARY: We describe the case of a
26-year-old man who presented with symptoms
compatible with Lhermitte sign that occurred
during yawning. It was associated with
congenital partial aplasia of the posterior arch
of the atlas. Cervical multisection-detector CT
myelography during yawning showed compression of
the upper cervical cord due to the inward
mobility of the isolated posterior tubercle. The
symptoms completely disappeared following
removal of the isolated posterior tubercle.
Congenital partial aplasia of the posterior
arch of the atlas (C1), a developmental failure
of chondrogenesis, is a rare anomaly. It is
clinically important because it causes acute
neurologic deficits, which are dosely associated
with neck extension.
We report a case of congenital partial
aplasia of the posterior arch of the C1 with a
transient Lhermitte
sign caused by
yawning,
which worsened after minor head trauma. We
discuss the etiologic factors of the myelopathy
on the basis of our radiologic and operative
findings.
Case Report
A 26-year-old man with other medical history
had experienced years of an electric-shock-like
pain running from his spine to all 4 extremities
during
yawning. He
also noted the strong electric-shock-like pain
and slight weakness in all extremities and a
giddy feeling immediately after a minor frontal
head trauma 1.5 years before admission. His
motor functions spontaneously improved after 18
months with a conservative treatment, but he
noticed that the electric-shock-like pain during
yawning was
worsening. He was admitted to our hospital for
investigation and treatment. He was alert, and
no abnormalities were found on cranial nerve and
motor examination. Deep tendon reflexes were
slightly increased bilaterally. Pinprick,
temperature, light touch, and vibratory
sensations were intact. A Romberg test was
positive, when dosing his eyes. The
electric-shocklike pain was transiently induced
by
yawning.
Cervical radiography, CT, and bone 3D-CT of
the upper cervical spine showed bilateral bony
defects of the posterior arch of the Cl with an
isolated posterior bone, diagnosed as a type D
of congenital partial aplasia ofthe posterior
arch of the C1. There was no cervical canal
stenosis or fractures. Dynamic cervical
radiography during neck extension and during
neck extension with the mouth opening to
simulate a posture during
yawning
showed slight inward mobility of the posterior
bone fragment, though Lhersnitte sign was not
induced by these examinations. Cervical MR
imaging showed a small intramedullary
T2-weighted hyperintensity with slight contrast
enhancement in the posterior column at a level
slightly below the posterior tubercle of the
C1.. Sagittal MR imaging during neck extension
showed inward mobility of the posterior tuberde,
affecting the posterior column of the cord.
Dynamic cervical multisection-detector CT (MDCr)
myelography with sagittal reconstruction was
obtained during neck extension with opening of
the mouth and during
yawning.
When the patient was in a neutral position, no
abnormalities were evident. During maximal neck
extension with the mouth open, reduction in
distance between the occiput and the
spinousprocess ofthe axis and inward mobility of
the posterior bone fragment were shown. Although
the dorsal dural sac was slightly depressed by
the posterior bone fragment, the spinal cord was
intact. MDCT myelography during
yawning
dearly showed strong compression of the cord due
to the inward mobility of the posterior bone
fragment (Fig 1G), and thermitte sign was
transiently induced. Follow-up MR imaging after
6 months showed persistent contrast enhancement
and slight reduction of the intramedullary
hyperintensity, which indicated serial changes
of a cord injury.
Total removal of the isolated posterior bone
fragment was performed in the prone position.
During surgery, the bone fragment was confirmed
to be a part of the posterior tubercle of the
Cl. It was covered with loose and fibrous
connective tissue, and attachment of the ream
capitus posterior minor was observed in the
center. The isolated bone fragment was gently
removed by resection of 2 elastic connective
tissue bands extending from the lateral mass to
the 2 ends. Thus, the final diagnosis was type D
of congenital partial aplasia of the posterior
arch of the Cl.' Immediately after the surgery,
the patient's symptoms during
yawning
completely disappeared, and he was discharged
without any complications on the seventh
postoperative day.
Discussion
Three primary ossification centers of the
C1-one at the anterior center and forming the
anterior tubercle and the other 2 at lateral
centers and forming the posterior arch and
lateral mass-appear during the embryonic period.
The 2 lateral centers form the posterior arches
with chondrogenesis, extending dorsally at 3-5
years of age, and the anterior center unites
these at 5-9 years of age. It is speculated that
the cause of congenital aplasia of the posterior
arch is a failure of chondrogenesis. Currarino
et al classified the anomalous Cl arch into 5
types on the basis of CT and radiographs.' In
this classification, type C and D, manifesting
as absence of the bilateral posterior arches and
an isolated posterior tubercle, are clinically
very important because these anomalies often
cause acute neurologic deficits such as
transient quadriparesis, paraparesis, Lhermitte
sign, chronic neck pain, and headache. To the
best of our knowledge, 18 case reports have been
published in the literature. Patient
backgrounds, including ours, were 9 women and 10
men ranging in age from 8.5 to 57 years (mean,
32.7 years). Regarding manifestation of these
neurologic deficits, Richardson et al and Sharma
et al suggested the cause was the secondary cord
compression caused by inward mobility of the
isolated posterior tubercle and surrounding
fragile soft tissue due to reduction in distance
between the occiput and the spinous process of
the axis during neck extension. In the
diagnosis, cervical lateral radiograph and CT
are very useful and important tools for the
initial diagnosis. Dynamic cervical radiographs
are helpful for evaluating mobility of the
isolated posterior tubercle.
Lhermitte sign in 3 patients, including
ours, was related to neck extension. Lhermitte
sign caused by
yawning is a
rare phenomenon. Although our dynamic studies on
cervical radiography, MR imaging, and MDCT
myelography clearly identified inward mobility
of the isolated posterior tubercle related to
the voluntary neck extension. These radiologic
findings precisely corresponded to the theories
of Richardson et and Sharma et al but Lhermitte
sign was not induced by these dynamic studies
because the voluntary neck extension with or
without the mouth open could not obtain mobility
of the posterior tubercle enough to press the
cord. On the other hand, only dynamic MDCT
myelography obtained during
yawning
dearly demonstrated direct evidence of cord
compression occurring transiently. It provided
very useful information for explaining
manifestation of Lhermitte sign. Although the
significant influence of
yawning on
the neck movement is unclear, it was confirmed
in our studies that
yawning as
an involuntary movement induced maximal neck
extension. This movement was greater than the
neck extension with voluntary mouth opening.
Neck extension due to
yawning
invited maximal inward mobility of the posterior
tubercle enough to press the cord, leading to
transient cord compression. In addition, we
consider that the aggravation of the neurologic
symptoms after minor frontal head trauma was
caused by spinal cord injury. it is assumed that
sudden neck extension following head movement
due to injury and associated involuntary mouth
opening (as a defensive reaction) leads to
transient and maximal inward mobility of the
posterior tubercle. As a result, the spinal cord
is strongly compressed and injured from the
dorsal side. In 5-including ours-of 6 patients
investigated by MR imaging, the intramedullary
T2-weighted hyperintensity was observed at the
level slightly below the posterior tubercle.
This change might reflect cord contusion,
myelomalacia, cord edema, or a presyrinx state.
In our case, the affected cord might have been
injured every time the patient yawned.
The treatment is removal of isolated
posterior tuberde, after which a good outcome
can be expected. We especially consider that
patients with type C or D should undergo surgery
at an early stage to prevent cumulative damage
to the cord.
References
1. Brown CE. Completeabsence of the
posterior arch of the atlas. Anat Rec
1941;81:499-503
2. Hamblen DL. Occipito-cervical fusion
indications, technique and results. J Bone Joint
Surg (Br) 1967;49:33-44
3. HolstenDR. Eine besondere Form von
Defektbildng ln hinteren Atlasbogen. Fortschr
Rontgenstr 1968;108:541-43
4. Dalinlca MY, Rosenbaum AE, van Houten F.
Congenital absence of the posterior ardi of the
atlas. Radiology 1972;103:581-83
5. Richardson EG, Boone SC, Reid RL.
intermittent quadriparesis associated with a
congenital anomaly of the posterior arch of the
atlas. J Bone Joint Surg (Am) 1975;57853-54
6. Spadaro A, Rotondo M, Conforti R, et al.
Aplasia of the posterior arch of the atlas
associated with isolated posterior tuberde. Acta
Neurol Napoli 1987;9:19-25
7. Torreman M, Verhagen ITHJ, Sluzewski M,
et al. Recurrent transient quadriparesia after
minor cervical trauma associated with bilateral
partial agenesia of the posterior arch of the
atlas. J Neurosurg 1996;84:663-65
8. Currarino G, Rollins N, Diehl JT.
Congenital defects of the posterior arch of the
atlas: a report of seven cases including an
affected mother and son. AJNR Am JNeuroradiol
1997;15249-54
9. Sharma A, Gaikwad SB, Deol PS, et ai
Partial aplasia of the posterior arch of the
atlas remnant: findings in three cases. AIM Am J
Neuroradiol 2000;21: 1167-71
10. Klimo P Jr, Blumenthal DT, Couldwell WT.
Congenital partial aplasia of the posterior arch
of the atlas causing myelopathy case report and
review of the literature. Spine
2003;28:E224-28
