STUDY DESIGN: Literature review. OBJECTIVES: Metastatic disease affecting the craniovertebral junction (CVJ) only accounts for 0.5% of all spine metastases. The management of these disease processes is complex, which involves multimodality radiological studies and various surgical approaches. We aimed to review the available evidence and summarize the findings in this review. METHODS: The authors conducted search of PubMed and Google Scholar with the following search terms: metastasis, craniovertebral junction (CVJ), occipitocervical, approaches, stability, and radiotherapy. Articles were reviewed by the authors and determined for inclusion based on relevance and level of evidence. RESULTS: The majority of relevant research reviewed composed of literature reviews of particular aspects regarding metastatic disease affecting the craniovertebral junction, including diagnosis, surgical approach, and radiotherapy. CONCLUSIONS: Prompt evaluation of rotational neck pain with or without occipital neuralgia may reveal early metastatic disease within a stable CVJ. Magnetic resonance imaging appears to be the gold standard imaging modality in detecting this pathology, with nuclear bone scan playing a role in distinguishing benign and malignant processes. Unfortunately, no level 1 evidence exists for use of either radiotherapy or surgery in these cases; however, from the available literature, spinal instability and evidence of progressive neurology are relative indications for operative intervention.
STUDY DESIGN: Literature review. OBJECTIVES: Metastatic disease affecting the craniovertebral junction (CVJ) only accounts for 0.5% of all spine metastases. The management of these disease processes is complex, which involves multimodality radiological studies and various surgical approaches. We aimed to review the available evidence and summarize the findings in this review. METHODS: The authors conducted search of PubMed and Google Scholar with the following search terms: metastasis, craniovertebral junction (CVJ), occipitocervical, approaches, stability, and radiotherapy. Articles were reviewed by the authors and determined for inclusion based on relevance and level of evidence. RESULTS: The majority of relevant research reviewed composed of literature reviews of particular aspects regarding metastatic disease affecting the craniovertebral junction, including diagnosis, surgical approach, and radiotherapy. CONCLUSIONS: Prompt evaluation of rotational neck pain with or without occipital neuralgia may reveal early metastatic disease within a stable CVJ. Magnetic resonance imaging appears to be the gold standard imaging modality in detecting this pathology, with nuclear bone scan playing a role in distinguishing benign and malignant processes. Unfortunately, no level 1 evidence exists for use of either radiotherapy or surgery in these cases; however, from the available literature, spinal instability and evidence of progressive neurology are relative indications for operative intervention.
The skeletal system is the third most common site of metastasis in the body, with spine
metastasis the most common within this system,[1,2] and of these the majority are found within the thoracic spine, with up to 15% of
cases found with the cervical spine.[3,4] The definition of metastasis is the spread of a disease-producing agency (as cancer
cells) from the initial or primary site of disease to another part of the body.[5] Common primary lesions included thyroid, lung, breast, renal, and prostate carcinoma,
and the literature would suggest breast carcinoma being the most common primary site,
attributing to just over a third of all cases.[6] Metastatic disease affecting the craniovertebral junction (CVJ) is less common and
only accounts for 0.5% of all spine metastases.[7] The cervical spine itself can be divided anatomically into 3 regions: the
occipitocervical region or craniocervical junction, the subaxial region, and the
cervicothoracic junction. The occipitocervical spine encompasses the occiput to C2, and CVJ
tumors are defined by anatomic involvement of the occipital condyles and/or atlantoaxial
spine. The median survival rate of patients with spinal metastasis is 10 months.[8] In this context, the main goals of the management of these patients is local tumor
control, reducing analgesic requirement, salvage or improve neurological function, and
maintain stability of the junction operatively if necessary. We present a case treated in
our spinal unit as well as the available literature on this rare clinical scenario. The
authors conducted search of PubMed and Google Scholar with the following search terms:
metastasis, craniovertebral junction (CVJ), occipitocervical, approaches, stability, and
radiotherapy. Articles were reviewed by the authors and determined for inclusion based on
relevance to topic of review and level of evidence. Institutional review board approval was
not sought for this study, as patients were not recruited for the study.
Methods
To construct this review article, the authors conducted search of PubMed, Google Scholar,
and the Cochrane database, with the following search terms: metastasis, craniovertebral
junction (CVJ), occipitocervical, approaches, stability, and radiotherapy. The combined
number of articles in the completed search was 165.The inclusion criteria were the following,Articles pertaining to clinical presentation of pathologyArticles pertaining to management of pathologySurgical approaches to CVJAn article in EnglishThe exclusion criteria were the following:Case reportsArticle pertaining to subaxial metastatic disease of cervical spineNot an article in EnglishUltimately through his process, 44 articles were included in this review.
Clinical Presentation and Investigations
The most common presenting complaint in this cohort of patients is cervical spine pain,
more specifically mechanical pain and occipital neuralgia, which presents as headaches in
the suboccipital region and retroauricular areas,[9] It has been noted by Bilsky et al[10] that rotational pain is present in 90% of patients with CVJ involvement. Of note,
myelopathy is an uncommon presenting complaint, and it is only present in 0% to 22% of cases.[11] Sixty years of age is the mean age of presentation of patients with upper cervical
spine metastasis.[12] The first imaging modality typically used to evaluate neck pain is a plain
radiograph of the cervical spine, to assess the alignment. It is not sensitive in
revealing metastatic lesions in bone as over half of the patients will have normal
findings on the radiograph.[13] Computed tomography (CT) is useful in this setting, first, in quantifying the
volume of lytic bone related to the metastatic tumor, and second, characterizing
suboccipital keel, which is important for preoperative planning.[12] The gold standard imaging modality in characterizing cervical spine tumors is
magnetic resonance imaging (MRI), for both soft tissue and bone tumors.[12] Nuclear medicine can be employed to help distinguish between begin and metastatic
tumors, which is particularly useful if the MRI findings are equivocal. Laufer et al[14] have shown that benign and malignant pathologies can be differentiated by using
18FDG positron emission tomography scans. Lesions with standardized uptake values
generated by this scan of over 2 warrant further investigation for a metastatic source,
and those with a SUV under could be followed for advancement with serial MRIs.CT-guided biopsy has been utilized to derive tissue from the CVJ but are considered
technically challenging, with transoral biopsy routes avoided due to seeding concerns.
Typically, the other metastatic lesions are more amenable to biopsy as a case with a
solitary metastasis to the cervical spine is rare.[15] Knowledge of the histology of the tumor is important as it can prompt preoperative
embolization and reduces potential intraoperative bleeding. Candidate histologies
typically reserved for this treatment include metastasis originating from renal cell
carcinoma, thyroid, and hepatocellular carcinoma respectively.[15]
Decision Making: Radiation or Surgery
Bilsky et al[10] preformed a retrospective review of a prospectively maintained spine database.
Thirty-three patients were identified from the database that had metastatic tumor
involving the atlantoaxial spine. Of these 25 underwent external beam radiation therapy
(EBRT) and were selected based on normal or near-normal alignment with radiosensitive
tumors. Eight patients underwent operative intervention with indications including
atlantoaxial displacement more than 5 mm or angulation exceeding 11° with displacement
more than 3.5 mm, and persistent pain after nonoperative therapy. The difference in median
survival rate between the groups in this study has been attributed to selection bias
related to disease burden and medical fitness to undergo operative intervention. The
median survival rate for the group undergoing operative intervention was 16 months
compared with 5 months for receiving EBRT alone. Stereotactic radiosurgery (SRS) can
achieve high-energy radiation to isolated lesions in the vertebral body and it has shown
favorable results when compared with EBRT in cervical spinal metastatic disease, not
uniquely located at the CVJ. The local control rates appear to be significantly higher.[16-20] It has been shown in 9 patients with metastatic disease to CVJ that SRS is a
feasible option, if patients are selected with similar criteria employed by Bilsky et al.[10] Tuchman et al[20] showed a median survival of 4 months with none of this cohort requiring subsequent
surgery. In the pursuit of evidence for decision making in cervical spine with metastatic
disease, Fehlings et al[21] preformed a literature review. They found no level 1 evidence, but general
guidelines for clinical management. With reference to the CVJ, metastatic disease
involving the CVJ positively influences the decision to surgically stabilize the spine,
with the posterior approach being favored, with radiation therapy selected for stable
cervical spines. Azad et al[22] preformed a retrospective review of 25 patients with a mean SINS (Spinal
Instability Neoplastic Score) of 7.3, with metastatic disease affecting the CVJ. In this
cohort, the most common primary tumors were breast and non–small cell carcinoma. All
patient in this cohort received SRS, with no tumor determined to be unstable based on the
SINS, and no patient treated with either radiotherapy or surgery prior to SRS. In this
cohort only 2 patients subsequently required post-SRS surgery to stabilize the cervical
spine.
Determining Stability of CVJ and Importance of Vascular Anatomy
In determining spinal stability, the Denis 3-column system has been employed, with
involvement of 2 columns indicating instability.[23] In recent times, the SINS has been validated with a sensitivity rate of 96% and
specificity of 80%. It derives a score form of 6 components: location, the presence of
pain, radiographic alignment, posterolateral involvement of spinal elements, degree of
vertebral body collapse, and bone lesion morphology.[24]Embarking on any surgical approach in this anatomical region warrants careful
consideration of vascular anatomy. Vertebral artery injury is catastrophic iatrogenic
complication of cervical spine surgery, as it may cause cerebrovascular accidents. Gluf et al[25] retrospectively reviewed 191 consecutive patients who underwent atlantoaxial
transarticular screw fixation, for a spectrum of pathology affecting the CVJ. In this
series, 5 patients suffered vertebral artery injury. In a similar retrospective study,
Wright et al[26] concluded that the overall risk of vertebral artery injury was 4.1%, with risk of
neurological sequelae 0.2%, during C1 and C2 transarticular screw placement. Peng et al[27] conducted a literature review of the factors related to vertebral injury.
Pertaining to anterior cervical approaches, lateral dissection puts the vertebral artery
at most risk, with the authors suggesting constant reference to the midline prudent. With
the posterior approach, safe screw placement is critical to avoid vertebral artery injury,
with preoperative imaging of vascular structure, either by CT angiogram or MRI, is
important in a considered surgical approach. Variability in the anatomy of the vertebral
artery can be described as both extraosseous and intraosseous. Extraosseous anatomical
anomalies of the vertebral artery include fenestration and persistent first intersegmental
artery. With fenestration, the vertebral artery gives 2 branches following its exit from
the C2 transverse foramen, with one branch entering the C1 foramen and the second entering
the spinal canal between C1 and C2. In the case of the persistent first intersegmental
artery the vertebral artery enters the posterior arch of C1, from the C2 transverse
foramen directly. Yamazaki et al[28] conducted a retrospective study looking at 100 consecutive patients who underwent
CVJ instrumentation surgery and had vertebral arteries 3-dimensional computed tomographic
angiography (3D CTA) preformed preoperatively to establish the existence of potential
anatomical anomalies. They report a rate of 10% extraosseous anomalies, 2% were
fenestrated and 8% were persistent first intersegmental artery, which were more common in
patients with atlantoaxial subluxation and congenital pathology of the CVJ. Anomaly of
intraosseous vertebral artery was noted in 31% of cases (see Figures 1 and 2).
Figure 1.
Posterior view of Atlas (C1) and Axis (C2) with related vascular structures.
Figure 2.
Axial view of (C1) displaying proximity of pedicel screw placement to vertebral
artery.
Posterior view of Atlas (C1) and Axis (C2) with related vascular structures.Axial view of (C1) displaying proximity of pedicel screw placement to vertebral
artery.
Indications for Surgery
In determining the degree of surgery required for a given patient with metastatic disease
of the CVJ, the modified Tokuhashi score can prove helpful.[24] The score has 6 components encompassing performance status, neurological status,
and burden of metastatic disease. With a low score on the modified Tokuhashi, palliative
surgery or radiotherapy can be considered. Of note vertebroplasty has been employed
successfully in pathologic cervical compression fractures of the upper cervical spine.[29,30] Ultimately, the risk and benefit of an operative intervention should be discussed
with each patient prior to any potential surgery.
Surgical Approaches
Once the decision to proceed with surgery has been reached, the approach to the
metastatic lesion at the CVJ must be considered, including the stability of the CVJ and
the expected degree of bony resection, Studies have shown that resection of 50% or more of
the occipital condyle produces instability.[31,32] All approaches to the CVJ demand characterization of the vertebral arteries
preoperatively, and vertebral artery angiograms and duplexes are commonly requested.
Intraoperative monitoring of sensory evoked potentials, motor function, and cranial nerve
function may be useful in selected cases.[33,34] As Fehlings et al[21] have discussed, the posterior approach appears to be the most frequently selected
approach. This allows access to the posterior elements of the CVJ and is extensile to
allow for access for instrumentation. Other approaches include transoral and posterior
lateral.
The Transoral Approach[35]
A major goal of this approach is the correction of irreducible anterior compression at
the cervicomedullary junction.[36,37] Of importance in this approach is the management of dental hygiene as to reduce
potential sources of infection due to dental caries. In terms of positioning the patient,
it is typically preformed with the patient awake to determine any alterations in
neurological status. The patient is placed supine with a horseshoe support to help hold
the head in extension, with fiberoptic intubation typically elected in this scenario. This
approach allows an exposure of 2 cm laterally and bilateral from midline from the clivus
to the C3 body. A 2-layer closure of both the pharyngeal mucosa and musculature is
advocated as this area is susceptible to dehiscence.[38] Following resection of the lesion, stability of the CVJ can be addressed through
this approach using the transoral atlantoaxial reduction plate systems.[39] Disadvantages of this approach include a deep working distance and restrained
operative view, and greater exposure in this approach can be achieved via La Fort I or
transmandibular osteotomies.
The Posterolateral Approach[40]
This approach allows good exposure of the anterior and lateral lesion at the CVJ, and
particularly suited for lesions with high vascularity. It can be performed in the prone
position with the head in a Mayfield frame, but other positions have been described. The
skin incision begins at the mastoid and ascends superiorly and continues inferior to the
superior nuchal line before descending the midline to C4. It can be extended caudally and
cephalad as required. Of note, if instrumentation is required, bilateral exposure is a
requirement.
Occipitocervical Fixation
This may be required in patients with radiographic instability of the CVJ. The treatment
goals in this scenario are reduction, immobilization, and instrumented fixation.[40] Complications of occipitocervical fixation include cerebellar trauma from occipital
screw placement, meningitis, vertebral artery injury, and direct trauma to neurological
tissues including the spinal cord. The evidence for the selection of fixation constructs
at CVJ is clear with cranial and cervical screw fixation constructs superior to wire or
cable constructs.[41,42] The screw fixation constructs preformed better on pain relief, correction of
malalignment, and achieved high fusion rates of 94% to 97%. Complication rates in a recent
series range from 12% to 30%.[43,44]
Craniocervical Junction Tumors in Children
Tumors of the craniocervical junction, including malignant tumors, are thankfully rare in
the pediatric population. This clinical issue may enter the scope of practice of the
readership, and as such it warrants inclusion in this review. The pediatric population
typically presents benign pathology in this anatomical region. Menezes,[45] in a series from a single institution, of 38 pediatric patients below the age of 16
years who underwent operative intervention, reports the most common pathology at this
level as chordoma, followed by fibrous dysplasia and aneurysmal bone cysts. In this group
the pain in the distribution of the C2 dermatome was the most frequent presentation,
occurring over two thirds of this cohort, with a third presenting with cranial nerve
palsies. Chordomas are radio-resistant and hence require surgical resection. Factors
considered important in developing a treatment plan include child age, potential for
growth, craniovertebral stability, and benign of malignant pathology.[46]
Discussion
In dealing with this rare clinical entity, making use of all available evidence can help
decision making in the clinical setting. Patients with this pathology typically present with
occipital neuralgia, which presents as headaches in the suboccipital region and
retroauricular areas,[9] with rotational pain present in 90% of patients with CVJ involvement.[10] Of note, myelopathy is an uncommon presenting complaint. Plain radiographs are not
sensitive in revealing metastatic lesions in bone, as over half of patients will have normal
findings on the radiograph. MRI appears to be the gold standard imaging modality in
characterizing cervical spine tumors, for both soft tissue and bone tumors.[12] Decision making in this cohort of patients can prove difficult. Fehlings et al[21] preformed a literature review in 2009 and found no level 1 evidence but provided
guidelines advocating for posterior approach in surgical candidates and radiotherapy for
patients with stable CVJ. Azad et al[22] preformed a retrospective review of 25 patients with a mean SINS of 7.3, with
metastatic disease affecting the CVJ, and found acceptable results for these patient when
treated with SRS. In contrast, the retrospective review by Bilsky et al[10] showed a long median survival rate for the group undergoing operative intervention of
16 months compared to 5 months for receiving EBRT alone. Once surgical intervention is
required, 3 described approaches are most frequently employed: posterior, posteriolateral,
and transoral. The posterolateral approach allows good exposure of the anterior and lateral
lesions and is suited for lesions with high vascularity.[40] The transoral route is useful for the correction of irreducible anterior compression
at the cervicomedullary junction[36,37] but has the disadvantage of an deep working distance. Pertaining to implant
selection, cervical screw fixation constructs are superior to wire or cable constructs,[41,42] with these fixation constructs preforming better on pain relief, correction of
malalignment, and achieved high fusion rates of 94% to 97%. A significant complication rate,
ranging from 12% to 30%, is attached to these surgeries.[43,44]
Conclusion
The CVJ is a complex anatomical structure comprising 4 synovial joints and accounts for 50%
of all head and neck movements. Despite only 0.5% of all metastatic disease to the spine
occurring at the CVJ, it presents a challenging clinical scenario. Prompt evaluation of
rotational neck pain[10] with or without occipital neuralgia may reveal early metastatic disease within a
stable CVJ. MRI appears to be the gold standard imaging modality in this pathology with
nuclear bone scan playing a role in distinguishing benign and malignant processes, where
CT-guided biopsy might prove difficult in execution. Unfortunately, no level 1 evidence
exists for use of either radiotherapy or surgery in these cases; however, from the available
literature, spinal instability and evidence of progressive neurology are relative
indications for operative intervention. Instrumented fixation constructs for the
stabilization of the CVJ appears to be superior to wire or cable constructs in this setting.
Pathologies in the pediatric population present a rarely encountered but uniquely
challenging clinical scenario.
Authors: Mark H Bilsky; Fintan J Shannon; Scott Sheppard; Vikram Prabhu; Patrick J Boland Journal: Spine (Phila Pa 1976) Date: 2002-05-15 Impact factor: 3.468
Authors: Peter C Gerszten; Steven A Burton; Cihat Ozhasoglu; William J Vogel; William C Welch; Joseph Baar; David M Friedland Journal: J Neurosurg Spine Date: 2005-10