Keywords
os odontoideum - etiology - treatment options - controversies of management
Introduction
Os odontoideum (OO) is an anomaly of the body of the axis in a particular odontoid,
identified as a smooth, independent ossicle of variable size and shape separated from
the base of a shortened odontoid process by an obvious gap, with no osseous connection
to the body of C2.
There are various hypotheses on the genesis of OO, including traumatic and developmental.[1]
[2]
[3]
[4]
[5]
[6] Orthotopic OO lies in the normal position on the odontoid process, moving with the
atlas anterior arch, whereas the dystopic morphology describes an ossicle fused to
the basion.[3]
The etiology and some of the aspects of the management remain controversial till date.
The authors analyzed the cases of the OO treated at their institution to look at the
etiologic aspects, common clinical presentations, treatment strategies adopted, evolution
of management patterns, and the long-term results to derive conclusions regarding
the appropriate management strategies.
Materials and Methods
Between 2004 and 2017, the authors treated 18 OO patients at Nizam’s Institute of
Medical Sciences, Hyderabad, India. Retrospectively, the data were obtained from the
hospital records and the patients were followed up.
All patients underwent neutral, flexion, and extension radiographs of the craniovertebral
junction (CVJ), and computed tomography (CT) and magnetic resonance imaging (MRI)
of the CVJ. CT angiographies of the vertebral arteries were done as part of standard
workup to look for vertebral artery anomalies since 2012.
Preoperative clinical evaluation was done, and Nurick’s grading was used for comparison
of the functional outcomes.
The main concern in OO is atlantoaxial dislocation (AAD) and the attending cord compression
from it. The basic philosophy of treatment of OO is addressing the instability at
atlantoaxial joint by fixation methods. The choice of surgical procedure was based
on the surgeon’s expertise and the comfort with a particular fixation method. In the
initial years, the choice was mainly C1C2 wiring methods or occipitocervical fusions;
however, the later years saw growing expertise in C1C2 fixation methods. Therefore,
the C1C2 fixation methods have become the standard of treatment for OO. In patients
with no instability and associated Chiari’s malformations, only foramen magnum decompression
was performed to address the symptoms arising out of the tonsillar decent. C1C2 fusion
procedure was performed even in asymptomatic cases.
Postoperative follow-up was done at 3, 6, 12 months, and then once every year.
Results
Most patients presented from the first to third decades of their life. There was almost
equal incidence in both male and female ([Table 1]).
Table 1
Patient characteristics of 18 patients
|
Total no of patients
|
18
|
|
Male-to-female ratio
|
10:8
|
|
Age (mean)
|
25.93 ± 16.05 (6–65)
|
|
Preoperative Nurick’s grade
|
1.64 ± 0.99
|
|
Postoperative Nurick’s grade
|
1.25 ± 0.55
|
|
Clinical symptoms
|
|
|
Asymptomatic
|
1
|
|
Neck pain
|
4
|
|
Quadriparesis following trivial trauma
|
6
|
|
Progressive quadriparesis
|
7
|
|
Associated radiologic features
|
|
|
Atlantoaxial dislocation (AAD)
|
17
|
|
BI
|
1
|
|
Chiari’s malformation with syrinx
|
1
|
|
Other associated syndromes
|
|
|
Down’s syndromes
|
1
|
|
Scoliosis
|
1
|
|
Surgical procedures
|
|
|
C1C2 wiring
|
2
|
|
C1C2 transarticular screws
|
2
|
|
C1 lateral mass and C2 pedicle screw fixation
|
11
|
|
Occipitocervical fusion
|
2
|
|
Foramen magnum decompression
|
1
|
|
Average follow-up in months
|
20.82 ± 20.57
|
Presentation
Though the common presentation was quadriparesis (13/18), the important observation
is that 6 out of 18 patients were diagnosed with OO when they developed quadriparesis
following trivial trauma. Six patients had history of significant injury such as fall
from height during their childhood before they were diagnosed with OO. The significance
of this trauma cannot be assessed as none of these patients had imaging prior to the
significant trauma.
One patient had Chiari’s malformation with syrinx with no AAD. The patient underwent
foramen magnum decompression.
Until 2014 different surgeons of our hospital used different surgical techniques,
including OCF (occipitocervical fusion), C1C2 wiring, and C1C2 transarticular screw
placement. The surgical technique was based on the surgeon’s expertise and comfortable
with a particular technique.
After 2014, C1 lateral mass and C2 pedicle screw technique has become the standard
technique for OO with AAD treatment ([Figs. 1] and [2]). In two cases of OO, the authors implanted cages in the C1C2 joint to achieve distraction
and further manipulation of joint to achieve complete reduction in the AAD, thereby
proper alignment at the CVJ.
Fig. 1 A 38-year-old woman presented with quadriparesis after a trivial fall while walking.
(A) CT sagittal film of showing orthotopic os odontoideum (OO). The fusion of the OO
to the ring of the C1 arch is noted. (B, C) Lateral view flexion and extension of X-ray of CVJ showing odontoideum fragment moving
on flexion and extension. In flexion, because of an incompetent odontoid, the posterior
ring of the C1 arch moves forward and encroaches on to the spinal canal compressing
the spinal cord. In extension, the fused C1 arch and the OO move posteriorly and cause
compression over the spinal cord. (D) Postoperative X-ray after C1 lateral mass and C2 pedicle screw fixation. Restoration
of alignment appreciated.
Fig. 2 A 17-year-old boy presenting with quadriparesis following fall while playing. He
as well had neck pain at the time of presentation. (A) CT sagittal section showing dystopic os odontoideum (OO). The OO is out of alignment
with the odontoid. (B) MRI sagittal section showing atlantoaxial dislocation (AAD) and compression of the
cervicomedullary junction (CMJ). The flexion and extension X-rays of the (not shown
in the picture) craniovertebral junction (CVJ) not showing any reduction in the AAD.
C1 lateral mass and C2 pedicle screw fixation with cage placement in the joints and
compression of the screw heads posteriorly achieved the reduction in the AAD. (C, D) Postoperative X-ray of CVJ and the CT sagittal section showing screws and the final
alignment achieved.
There were no major surgical complications in the immediate postoperative period.
Average follow-up of the patients was 20.82 ± 20.57. Bone fusion was good in all cases
of OCF. C1C2 fixation did not show well evident bone formation, but there was no screw
pull-out or implant failure observed during follow-up. In one patient who underwent
C1C2 wiring, there was no evident bone fusion on X-rays. However, there was no implant
failure or breakage of wires.
Discussion
Os odontoideum is a rare anomaly of the odontoid process first described by Giacomini
in 1886. There is considerable controversy regarding the etiology of the OO about
whether it is congenital or traumatic. The proponents of the congenital etiology hypothesize
that the OO is the result of the failure of the dens to fuse with the body of the
C2.[3] This hypothesis is based on the observations that OO is present in twins,[2]
–
[4] in cases with congenital syndromes and other associated anomalies at the CVJ.[2]
[7]
[8]
[9]
[10]
[11]
[12]
[13] It was believed that OO is caused by a failure of fusion between the first and second
sclerotomes. However, this theory has been questioned because the neurocentral synchondrosis
is located below the level of the superior articulating facet, whereas the gap in
OO is frequently located above the plane of the superior articulating facet.
This gave rise to the traumatic origin hypothesis. Fielding et al suggested that with
a fracture or disruption through the neurocentral synchondrosis, the alar ligaments
that attach to the apex of the odontoid may gradually distract the fragment away from
the base. The apex and base of the odontoid continue to have adequate perfusion, but
the midportion suffers from lack of blood supply and thus contributes to poor healing.[5] This is supported by the observation that many OO patients had a remote trauma in
their childhood.[14] The “traumatic cause” hypothesis is supported by case reports of patients with a
previously documented intact C2 who later were found to have OO after remote trauma.
Schuler et al reported on a 2-year-old patient who fell out of her crib and complained
of neck pain; her initial cervical X-ray was normal. After continued neck pain, repeat
cervical X-rays were obtained 13 months after her injury, which demonstrated OO with
atlantoaxial instability.[16] Zygourakis et al, in their report of 2-year-old girl who had C1–2 ligamentous injury
demonstrated with subsequent development of the OO, proposed that the development
of the OO is as a result of the culmination of the trauma and vascular compromise.[7]
In this series, there were six patients who had a history of significant trauma in
their childhood; however, none of them was investigated in the period following it.
Therefore, it is difficult to derive any conclusion out of the aforementioned observations
about the etiology of OO.
Clinical Presentation
The presentation of OO can be an asymptomatic patients in whom OO is an incidental
discovery, neck pain, headache, compressive myelopathy, presentation with quadriparesis,
or quadriplegia after a traumatic event.[16]
[17] In this series, one patient was incidentally diagnosed to have OO, when she had
cervical X-rays as part of workup for scoliosis. Investigation of occasional neck
pain revealed OO in some. Development of quadriparesis following trivial trauma in
six patients in this series supports the argument that even asymptomatic patients
may require surgical treatment, particularly in this setup.
Os odontoideum can be clearly visualized using plain radiographs with the open mouth,
anteroposterior, and lateral views. In addition, plain dynamic lateral radiographs
(performed in flexion and extension) can further evaluate atlantoaxial instability.
However, CT of the CVJ, CT angiography, and MRI of the CVJ are important to better
understand the bony anatomy and soft tissue compression. Though various parameters
such as space available for the cord and instability index have been proposed, they
may not reflect the true instability at the CVJ.[18]
[19]
Management
All patients who are symptomatic and have neck pain or compressive myelopathy need
surgical treatment. There is controversy regarding the management of asymptomatic
OO. Some propose follow-up whereas others prefer treatment on case-to-case basis.
We believe asymptomatic patients stand significant risk of deterioration even on minor
trauma as is evident from the high number of patients in this series presenting for
the first time after a trivial injury. In addition, considering the reports of sudden
deaths, deterioration after incidents of minor trauma, gives the impression that surgery
should be strongly considered even in asymptomatic patients.[20]
[21] Factors such as age, activity level, and radiographic findings, including evidence
of atlantoaxial instability and anatomy favorable for surgical instrumentation need
to be considered before opting for a conservative approach.
C1C2 fixations have emerged as the procedure of choice because they offer the advantage
of retaining neck motion except neck rotation. As the authors perform the C1 lateral
mass, C2 pedicle screws placement with joint distraction techniques for the CVJ with
greater ease, these techniques have become the favorable for treatment of OO. The
C1C2 distraction techniques with cage placement were especially considered when the
patient has AAD, which is not completely reducible ([Fig. 2]). Foramen magnum decompression alone can be considered for patients of OO with no
obvious instability at C1C2. However, the C1C2 joint distraction techniques for treatment
of Chiari’s malformations as proposed by Goel can also be considered.[21]
[22]
Conclusion
Os odontoideum etiology is controversial, but vascular insult either before or after
birth seems to cause it. Majority of OO are associated with significant instability
at atlantoaxial joint. Surgical treatment of even asymptomatic patients may be a better
option in Indian setting. C1C2 fixations techniques seem to be superior to other techniques
in terms of offering stability as well as minimal restriction of movements.
