Unique paradoxical atlantoaxial dislocation with C1-C2 facet diastases and isolated ligamentous injury to the craniovertebral junction without neurological deficits: A case report.

STUDY
DESIGN: Retrospective review of the case file.
OBJECTIVE: The primary objective was to report this rare case and discuss the mechanism of dislocation and technique of manual closed reduction of C1-C2 vertebrae in such scenarios. SUMMARY OF BACKGROUND DATA: Posterior atlantoaxial dislocation (AAD) is extremely rare and a few cases have been reported in English literature. This young man sustained a high speed car accident and survived an extreme hyperextension injury to the craniovertebral junction (CVJ) without any neurological deficits. On evaluation for neck pain he was noted with a dislocated odontoid lying in front of Atlas. There was C1-C2 facet diastases. No bony injury was noted at CVJ. Transverse axial ligament (TAL) was intact. He underwent a successful awake reduction of the dislocation. The joint had to be manually distracted, realigned, and released under the guidance of fluoroscopy. This was followed by single stage C1-C2 Goel's fusion with awake prone positioning. This patient was able to go back to work at the end of 3 months (GOS 5).
CONCLUSIONS: This condition is extremely rare, can be carefully reduced manually under adequate neuromonitoring, and requires C1-C2 fusion in the same sitting.

INTRODUCTION

Posterior atlantoaxial dislocation (AAD) is rare and mostly follows trauma to the craniovertebral junction (CVJ).[1] Biomechanics include a significant force with severe hyperextension and cranial traction at CVJ. This is known to be associated with extensive facial injuries.[12] Most of the patients are neurologically intact.[13] Closed reduction and collar immobilization has been reported to be successful in a few cases though intact transverse axial ligament (TAL) is a prerequisite for conservative management.[14] Most of the previously reported cases have been managed with C1-C2 fusion.[1] We describe a patient with isolated ligamentous injury, without any fractures at CVJ, managed with closed reduction and fusion. Though similar cases have been reported in English literature (12 cases), it is rare and warrants a discussion on mechanism and management.[13]

CASE REPORT

This 32-year-old gentleman was involved in a high speed car accident without a fastened seatbelt. He survived a whiplash injury, lost consciousness, and sustained severe facial injuries with soft tissue swelling. On examination he had stridor, airway compromise, and hypoxia for which he underwent an emergency tracheostomy. He had no neurological deficits. He had a fractured mandible and his neck was tilted to the right with chin to the left.

His X-ray cervical spine showed a displaced odontoid, anterior to the anterior arch of atlas. Computed tomography (CT) scan of the CVJ showed a dislocated odontoid, anterior to C1 arch in front of right lateral mass of atlas [Figure 1]. The C1-C2 facets were seen diastased bilaterally [Figure 2]. Magnetic resonance imaging (MRI) of CVJ [Figure 3] showed disrupted posterior longitudinal ligament with severe kinking of the spinal cord. TAL was seen intact on T2 axial MRI. MR angiography showed a flow cutoff in the left vertebral artery at the level of C2-C3 and a thin streak of flow was noted in the right vertebral artery at the level of CVJ [Figure 4].

Figure 1

Showing preoperative computed tomography (CT) scan of craniovertebral junction (CVJ). (a) Sagittal view with posterior dislocation (b) Coronal view with odontoid in front of C1 arch (c and d) Axial views showing odontoid tip

Figure 2

Showing preoperative CT scan of CVJ with reconstruction. (a) Showing the odontoid in front of C1 arch (b) Showing the posterior view with bilateral C1-C2 joint diastases

Figure 3

Showing Magnetic resonance imaging (MRI) of CVJ. (a) Sagittal view pinching of the cord (b) Axial view — displaced odontoid (down arrow) and intact transverse axial ligament (arrow to right)

Figure 4

Preoperative MR angiography showing flow cutoff at the level of C2-C3 in the left vertebral artery (arrow to the left) and thin streak of flow in the right vertebral artery opposite C1-C2 (arrow down)

Management

He underwent an awake reduction of the dislocation in the OR. Using Gardner-Wells skull traction, slow distraction with gradual rotation to the left was applied. A continuous fluoroscopy revealed progressive realignment of the odontoid [Figure 5]. This also corrected the neck tilt. He was awake when positioned prone and after confirming the position of odontoid and intact neurological status, he underwent C1 lateral mass and C2 pedicle screw fixation with fusion across the C1-C2 joint. The joint capsule was torn with a large gap between C1 and C2 facets. It was filled with bone graft after debridement to facilitate fusion. The vertebral arteries were looking perfectly normal and were seen positioned normal in relation to the CVJ.

Figure 5

Showing intraoperative images in the different stages of reduction. (a) Prior to distraction b) after distraction, and (c) after realignment and release

Postoperative course

He had quadriparesis in the immediate postoperative period. It was predominantly proximal in involvement and was symmetrical. CT scan of the CVJ revealed normal alignment of atlas and axis and implants in situ [Figure 6]. MRI of the CVJ revealed normal spinal cord with normal alignment. MRA of the vertebral artery showed occlusion of the right vertebral artery beginning at the origin and extending up to the origin of the basilar artery intracranially, suggesting complete thrombosis secondary to a possible traumatic dissection. The possibility of worsening secondary to surgical manipulation could not be ruled out. However, the left vertebral flow pattern was the same as preoperative MRA with segmental loss of flow at the distal V1 segment [Figure 7]. The patient was treated with anticoagulants and he made gradual recovery without any deficits at 2 weeks after surgery. He was able to go back to work after 3 months and neurologically intact at 1 year after surgery. Repeat CT scan showed good alignment of implants.

Figure 6

Postoperative CT scan of CVJ. (a) Screws in C1 lateral mass (b) screws in C2 pedicle (c) coronal view with implants in situ and normally placed odontoid, and (d) sagittal normal alignment

Figure 7

Postoperative images. (a) CT tomography with implants in situ and normal alignment (b) Sagittal T2 MRI with normal cord intensity (c and d) MR angiography showing complete occlusion of the right vertebral artery and segmental narrowing of the left vertebral artery

DISCUSSION

OC-C1-C2 joint is an osseoligamentous complex and extremely mobile; a component of the cervical spine. OC-C1 joint is primarily a “yes-yes” joint (flexion-extension of 23-24.5 degrees) and C1-C2 joint is primarily a “no-no” joint (axial-rotation of 23.3-38.9 degrees to each side) with atlas acting as a washer between occiput and the rest of the cervical spine. It is largely believed that flexion at OC-C1 is limited by impingement of odontoid tip and extension by the tectorial membrane. There occurs a small flexion-extension movement at C1-C2 joint (of 10.1-22.4 degrees) with transverse ligament limiting flexion and tectorial membrane along with C1-C2 articular joint limiting extension.[5]

As our patient had not fastened the seatbelt, whiplash hyperextension at the CVJ followed by flexion caused the facial injuries. The force was severe enough to fracture mandible. The possible mechanism has been put forth by Haralson and Boyd and it seems to be the case here.[12] However, probably the angle and direction of the distraction force was along the horizontal axis of lateral mass of atlas and not below it, resulting in the C1 arch being pulled superiorly and posterior along the tip of the odontoid due to hyperextension, tearing all the surrounding ligaments and finally dislocating it posterior to anterior arch of atlas. This would have torn the facet joint capsule and secondarily causing diastases of C1-C2 facets. Also the hyperextension would have tilted the angle of C1 arch superiorly and allowing it to slip beyond odontoid. This type of dislocation has been reported in cases of rheumatoid arthritis with AAD (6.7%) where the connective tissues get eroded by the disease causing the posterior glide of C1 arch over odontoid.[67]

In the previous cases reported in English literature, facial soft tissue injury was common and seen in 57% of the cases. Most of them did not have any neurological deficits and this feature reflects the fact that the space available for the cord is abundant at this level. Though closed reduction was successful in most of the cases, 36% of them required odontoidectomy.[13] An intact TAL is a necessity for conservative therapy as odontoid may slip posterior if TAL is ruptured. In our case too, as reported by Chaudhary et al., intact TAL could be demonstrated radiologically.[4] The direction of traction, requirement of neuromonitoring, and the role of fluoroscopy has been well-emphasized previously.[16] As the risk of redislocation and cord compression exists during positioning under general anesthesia, awake positioning is advised as a safety measure especially if spinal cord monitoring facilities are unavailable. Including our patient, closed reduction has been successful in 62% of cases reported. What factors facilitate a closed reduction needs to be defined yet.[13] Though conservative approach was successful in four of the 12 cases (33%), we believe that these dislocations are unstable as most of the surrounding ligaments are ruptured and require fusion at the onset.[5] Also a distracted C1-C2 joint is common in these cases and a considerable sized graft is required to fill up the joint space to achieve satisfactory fusion. Varied types of posterior constructs have been used in such cases and it is for the surgeon to choose from the available options.[13] Our patient's postoperative deterioration was probably secondary to traumatic vertebral artery dissection-occlusion, which was initiated by trauma and progressed with surgical manipulation. However, he improved with probable compensation from the opposite vertebrals and anticoagulants over a period of 2 weeks. In 12-34% of the cases, trivial trauma (including chiropractic manipulations) is considered as an important risk factor for cervical artery dissections.[8] We believe our patient had the same reason for vertebral artery occlusion.

To conclude:

Posterior AAD is rare and is mainly caused by traumatic hyperextension and cranial traction at CVJ.

Significant force is involved in disrupting the ligaments. Though C1-C2 facet diastases are common, vertebral bone injury is uncommon.

Neurological deficits are rare as spinal canal is wider at the involved level.

Closed reduction can be successful, but needs to be done under fluoroscopic control with neuromonitoring.

It is unstable and C1-C2 fusion is necessary.