Extraoral anterior approach for the treatment of os odontoideum: illustrative case

Abstract

BACKGROUND

Os odontoideum is typically treated with instrumented fusion through a posterior cervical approach. When this approach fails, limited options for revision are available. Occipitocervical fusion and transoral anterior fusions have been utilized in the past but are associated with high morbidity and complications.

OBSERVATIONS

Here the authors report a case of os odontoideum that was treated with an anterior cervical extraoral approach after failed posterior instrumented fusion. They discuss the challenges that can be encountered with the failure of fusion and the limited options when it comes to approach and fixation of os odontoideum.

LESSONS

To the authors’ knowledge and based on a review of the literature, this case represents the first use of an anterior extraoral prevascular approach to the high cervical spine to address os odontoideum. They demonstrate that this approach can be utilized as a reasonable alternative to transoral surgery and should be considered in cases where additional or alternative fixation is desired without the morbidity and complications associated with occipitocervical fusion or a transoral approach, especially in a younger patient population.

Os odontoideum is a deviation from normal anatomy of the cervical spine in which there is a division of the dens into two separate segments. This division occurs near the base of the dens forming a small hypoplastic dens and a fully corticated ossicle cranially. The exact etiology is unknown and remains controversial, with support for both a developmental/congenital theory and a traumatic theory.^1,2 This phenomenon is clinically significant, as it can be associated with cervical instability and neurological decline.³

Asymptomatic os odontoideum without neurological deficit can be managed conservatively with observation and follow-up imaging.^4,5 When symptomatic, particularly if presenting with neurological decline, myelopathic signs and symptoms, and/or craniocervical pain due to instability, surgical intervention is warranted and is typically successful.⁶

The goals of surgical management are to reduce the os, decompress the cervical spinal canal, and provide stability to the atlantoaxial joint. This is typically accomplished by either atlantoaxial fusion or occipitocervical fusion.⁷ Posterior instrumented fusion with C1–2 screw and rod fixation is a common technique, with large case series reporting good outcomes.⁶ While posterior approaches provide easier access for reduction and fusion, occasionally an anterior approach is needed for irreducible dislocations or in the revision setting. Because of the local anatomy, a transoral approach is typically utilized.^4,6–8 However, this approach is not without complications, including wound-healing issues, frequent infections, velopharyngeal insufficiency, and difficulty with nutrition postsurgery.⁹ Here, we present a case in which an extraoral anterior approach was used to gain access to the atlantoaxial articulation for decompression and fusion, maximizing the benefits of anterior access to the high cervical spine while minimizing complications associated with a transoral approach.

Illustrative Case

A 22-year-old male with history of os odontoidium 6 years status post–C1–2 posterior spinal fusion and instrumentation presented to the clinic with 2 weeks of left-sided neck pain and headaches. On examination, no neurological deficits were noted; however, plain radiographs were obtained, which demonstrated lucency surrounding his hardware (Fig. 1A). A computed tomography (CT) scan was obtained, which demonstrated nonunion of his fusion with lucencies around his hardware (Fig. 1B). The patient decided to undergo revision posterior spinal fusion with a Fiber Wire (Arthrex) cable construct and iliac crest allograft for treatment of his nonunion. This construct is demonstrated in Fig. 2A. A follow-up CT scan was obtained at 3 months postoperatively for routine evaluation of healing, which unfortunately redemonstrated a pseudoarthrosis/nonunion (Fig. 2B). At this point, with two prior failed attempts at posterior atlantoaxial fusion, the patient was offered several options. These included occipitocervical fusion, with the downside of significantly limiting his motion; transoral options; and revision atlantoaxial fusion with utilization of an anterior approach in addition to revision posterior fixation. Ultimately, he elected to proceed with revision anterior/posterior C1–2 fusion for preservation of motion.

FIG. 1.

A: Preoperative plain radiograph showing failed posterior lateral mass and pedicle screw construct with lucency surrounding the screws. B: Preoperative sagittal CT scan again showing lucency surrounding C1 lateral mass screws and C2 pedicle screws. Lack of bridging bone at the fusion site is also apparent, suggesting pseudoarthrosis/nonunion.

FIG. 2.

A: Postoperative radiograph after the first revision surgery utilizing a cable construct with iliac crest allograft. B: CT scan 3 months postoperatively showing cable fixation with iliac crest allograft. Pseudoarthrosis/nonunion is noted with no bridging bone at the fusion site.

This was completed in a staged manner. Initially, a high anterior approach was utilized with a C1–2 plate construct with structural allograft. This was achieved via an extraoral approach, which was advantageous given the known relatively high risk of complications and morbidity with transoral approaches.⁹ The construct is demonstrated in Fig. 3.

FIG. 3.

A: Postoperative CT after extraoral anterior fusion with plate construct and structural allograft. B: Final 1-month postoperative plain radiograph demonstrating stable C1–2 combined anterior and posterior fusion. C: Two-month postoperative radiograph demonstrates stable C1–2 combined anterior and posterior fusion. There appears to be some bridging bone and evidence of fusion.

His postoperative course was complicated by a hypoglossal nerve neuropraxia due to retraction of the nerve, which was necessary during the high anterior cervical approach. However, he was doing well at his 2-month follow-up with only mild hoarseness, and his difficulty swallowing and other symptoms had resolved.

Technique: Anterior Extraoral Approach to the Upper Cervical Spine

The patient was positioned supine on a radiolucent table with a gel donut head positioner. Our skin incision was made on the left side of the neck approximately 1 cm distal to the mandible. This incision was transverse beginning at the midline and ending at the anterior border of the sternocleidomastoid muscle. Dissection was then carried through the subcutaneous layer down to the platysma, creating flaps superiorly and inferiorly. We then incised the platysma with bipolar electrocautery. The submandibular gland was mobilized about its inferior border to retract it cranially and laterally. At this point, the sternocleidomastoid fascia was identified and incised. We dissected along the medial border of this muscle, palpating the carotid artery to find the plane between this and the trachea medially. At this point, the tendon of the digastric muscle was identified and taken down, tagging it for later repair. We were then able to identify and mobilize the hypoglossal nerve superiorly in the wound with an accompanying artery, which was a branch of the lingual artery (Figs. 4 and 5). Both of these structures were dissected and mobilized far medially and laterally to allow for retraction for more proximal exposure. Then, with blunt dissection, we palpated the vertebral column and proceeded to place a disc space marker. Throughout this time, the carotid artery was retracted laterally, the trachea medially, and the hypoglossal nerve cranially. We used intraoperative fluoroscopy to confirm we were at the C1–2 level and then proceeded to expose the anterior arch of C1 and the C2 vertebral body using standard technique with a subperiosteal dissection out to the longus colli muscles laterally. The os odontoideum defect was identified, which was the uppermost portion of the C2 vertebral body where the odontoid process would normally be attached. The inferior arch of C1 and the superior portion of the C2 body were decorticated with a high-speed burr. Next, a structural iliac crest allograft was fashioned and gently tamped into the space between C1 and C2. An anterior cervical plate was applied with two screws in the C1 arch and 2 screws in the vertebral body of C2.

FIG. 4.

Diagram demonstrating the anterior extraoral retropharyngeal approach to C1–2 (dashed arrow).

FIG. 5.

Intraoperative photographs of anterior extraoral retropharyngeal approach to C1–2 demonstrating the anatomical course of the hypoglossal nerve in the proximal aspect of the incision. On the right is a zoomed-in photograph of the left for better visualization. The star represents the trachea and esophagus retracted medially. The triangle represents the hypoglossal nerve. The square represents the sternocleidomastoid and carotid sheath.

Fluoroscopy was utilized to confirm proper placement of our plate and screw construct with structural allograft. The wound was then closed in a layered fashion, reattaching the digastric tendon followed by 2–0 Vicryl (Ethicon) sutures for the platysma layer and staples for skin. A Penrose drain was left deep in the wound to allow drainage.

The second stage consisted of revision posterior lateral mass screws at C1 and pars screw fixation at C2 with allograft. The C1 and C2 tracts were packed with allograft morselized bone, and the screw diameter was upsized. All screws had satisfactory purchase. Additionally, the C1–2 facet joints were decorticated with a high-speed burr and allograft bone was packed into the joints. The posterior arch of C1 and the lamina and spinous process of C2 were decorticated and bone grafted as well.

Discussion

Observations

To our knowledge, this represents the first case of os odontoideum treated with anterior plate and screw fixation via an extraoral approach. The construct choice for treatment of os odontoideum typically depends on its reducibility.¹⁰ Posterior stabilization with atlantoaxial fusion is the preferred treatment method due to its relative ease and preservation of motion compared with occipitocervical fusion.¹¹ Posterior stabilization is classically performed with lateral mass screw fixation in C1 and pars/pedicle screws at C2.^6,8,10,11 When reduction and stabilization are not possible via this posterior approach and construct, an anterior transoral approach can be performed or occipitocervical fusion is advocated.¹⁰

Several anterior approaches and constructs have been described. Most commonly, a transoral approach is utilized, which can be advantageous in achieving reduction of a significantly displaced os odontoideum. The downside to this approach is a high rate of complications, with an increased risk of infection, wound-healing complications, velopharyngeal dysfunction, and difficulty maintaining nutrition postoperatively.^9,12 These complications make an extraoral anterior approach to the high cervical spine more favorable.

Historically, two main retropharyngeal extraoral approaches to the upper cervical spine have been applied for a variety of pathologies with good outcomes. The first approach, originally described by Whitesides and Kelly,¹³ is a retrovascular approach that allows relatively simple access to the lateral C1–2 articulations; however, it does not allow for anterior decompression or the use of strut grafts. This also requires a bilateral approach to perform C1–2 fusion. In a case series by Laus et al.,¹² one patient with os odontoideum was treated via this bilateral Whitesides approach using transarticular screw fixation. The second approach is a prevascular approach originally described by McAfee et al.,¹⁴ which allows for extensive anterior decompression and strut graft placement.

Lessons

In our case, we demonstrate that an extraoral approach is possible when anterior access to C1–2 is necessary and allows for plate stabilization with allograft strut placement. Good clinical outcomes were obtained in our patient while minimizing the risk of infection attributed to the transoral anterior approaches. Additionally, we were able to avoid the significant debilitating loss of motion encountered with occipitocervical fusions in this young patient. The patient did have a hypoglossal neurapraxia that had mostly resolved at our final, although limited, follow-up. Unfortunately, our patient was lost to follow-up after only 2 months. Further research into the use of an extraoral anterior approach to the high cervical spine would be advantageous so that we might add this to our armamentarium of choices for approach and fusion constructs in these patients.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: all authors. Analysis and interpretation of data: Placide. Drafting of the article: Farrar, Grasso. Critically revising the article: all authors. Reviewed submitted version of the manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Farrar.