32 Trauma to the Craniovertebral Junction



10.1055/b-0034-81409

32 Trauma to the Craniovertebral Junction

Gonzalez, L. Fernando, Webb, K. Michael, Crawford, Neil R., Sonntag, Volker K. H.

The craniovertebral junction (CVJ) is formed by three main bony structures: the basiocciput and the first and second cervical vertebrae. The relations among these bony elements are constant because of the complex array of associated ligaments. The ligaments are the main structures responsible for stability in this region. Maintaining the appropriate alignment is important in protecting the upper spinal cord and medulla and both vertebral arteries.


This chapter covers the traumatic conditions that most often affect the CVJ and related surgical treatments. A thorough understanding of the local anatomy and bio-mechanics of this region is needed to treat the pathology in this location effectively.



Anatomy and Biomechanics


The occipital condyles articulate with the lateral masses of C1 in cup-shaped joints that facilitate motion primarily for flexion and extension. This joint is the largest contributor to flexion and extension from any single motion segment in the cervical spine, but its role during axial rotation and lateral bending is minimal. C1 rests on top of the C2 facets, which resemble relatively flat shoulders with a lateral and inferior slope. The configuration of C1 on top of C2 predicts and facilitates its axial rotation and minimizes lateral bending and flexion-extension at this joint. The second cervical vertebra is also known as the axis. Its vertical portion, the odontoid process, which is an essential part of this complex, forms the pivot point about which the majority of the axial rotation of the head occurs.


This complicated anatomical configuration is kept in functional alignment by the complex associated ligamentous array ( Fig. 32.1 ). The cruciate ligament consists of two portions, one vertical and one horizontal. Its fibers are interwoven in a cross. The vertical segment extends from the posterior aspect of the body of C2 and attaches rostrally on the anterior aspect of the foramen magnum. The horizontal band corresponds to the transverse ligament, which has two lateral insertions on each side of the medial aspect of the lateral mass of C1 and another insertion on its own synovial joint behind the odontoid process. The transverse ligament serves as a “seat belt” to prevent the odontoid process from posterior translation.


The paired alar ligaments run from the lateral and superior aspect of the odontoid process and fan out to the inferior aspect of each occipital condyle. Their main function is to stretch simultaneously when the head rotates to either side, thereby preventing excessive axial rotation at the occiput–C1–C2 complex.1 The apical ligament extends from the tip of the odontoid process and fans out to the anterior rim of the foramen magnum. Its main function is to keep the atlas under compression between the axis and the occipital bone. Some doubt its importance because its presence is inconsistent.2

Fig. 32.1a, b Ligamentous anatomy of the craniovertebral junction (CVJ). (a) Axial and (b) coronal views of the main ligaments involved in the stability of the CVJ. (Used with permission from Barrow Neurological Institute.)

The function of a second group of ligaments is less clear. The tectorial membrane, the most rostral extension of the posterior longitudinal ligament (PLL), inserts on the anterior rim of the foramen magnum. Its main function is to serve as a restraint during flexion of the head. The main function of the cranial extension of the anterior longitudinal ligament (ALL), known as the anterior atlanto-occipital membrane, is to restrain hyperextension of the head.


In summary, the stability of the CVJ primarily is imposed by a complex array of ligaments that apply high tension on the occiput–C2 complex to maintain its integrity while still allowing a wide range of motion.


Traumatic injuries of the CVJ need to be studied and evaluated for treatment as a group. From a methodological perspective, it is useful to organize CVJ injuries as primary bony injuries, primary ligamentous injuries, and combined injuries involving both ligaments and bones.



Bony Injuries



Condylar Fractures


Anderson and Montesano3 classified condylar fractures into three types: type I comminuted fractures, type II skull base fractures that extend into the occipital condyle, and type III avulsion fractures ( Fig. 32.2 ). In general, types I and II are considered stable fractures, which may heal with treatment by cervical immobilization (collar) alone. Type III fractures compromise the ligaments. They are usually unstable and require rigid immobilization.4



Atlas Fractures


A wide variety of fractures can occur at C1 ( Fig. 32.3 ). The key factor that determines treatment is the integrity of the transverse ligament. In general, most C1 fractures heal satisfactorily with immobilization. Cervical collars provide minimal immobilization at the CVJ and should be used in cases of nondisplaced fractures ( Fig. 32.4 ). Mildly displaced fractures require sterno-occipital mandibular immobilization (SOMI) or a halo brace, as do severely comminuted fractures, which can render the transverse ligament physiologically incompetent. In such cases, a halo brace or surgical fixation is indicated. Patients treated in a halo brace should be followed closely. If fusion fails to occur, they should undergo surgical stabilization. Figure 32.4 compares the range of motion of various cervical orthoses.



Axis Fractures


C2 fractures can be classified as odontoid fractures, hang-man’s fractures, and miscellaneous fractures ( Fig. 32.5 ).5Type I odontoid fractures occur when the tip of the odontoid process is avulsed. This type of fracture is rare and usually heals well when treated with a semirigid orthosis, such as a Minerva brace. Type II odontoid fractures occur at the base of the dens and involve minimal displacement (< 6 mm). Their fusion rate ranges from 85% to 90% when treated with a halo brace. Fractures associated with > 6 mm of displacement have an 80% chance of nonunion if treated only with an external orthosis.6 Consequently, early surgical fixation is recommended. Type III odontoid fractures compromise the vertebral body of C2 and usually heal with treatment in a halo brace. These patients must be closely followed. If there are no signs of fusion after 3 months of immobilization, surgery should be considered.

Fig. 32.2 Normal ligamentous anatomy and type I comminuted fractures, type II skull base fractures that extend into the occipital condyle, and type III avulsion fractures. (Used with permission from Barrow Neurological Institute.)
Fig. 32.3a–g There are a variety of atlas fractures. a The four-part ring or burst fracture is classically referred to as a Jefferson fracture. b The comminuted lateral mass fracture is extremely common. It creates C1–C2 instability by rendering the transverse ligament incompetent because it detaches the tubercle for insertion of the transverse ligament. More stable patterns of injury include (c) the unilateral ring fracture, (d) linear lateral mass fracture, (e) posterior ring fracture, (f) anterior arch fracture, and (g) contralateral ring fracture. The major determinant of stability of these injuries is whether the transverse atlantal ligament is anatomically and physiologically intact. (Used with permission from Barrow Neurological Institute.)
Fig. 32.4 Kinematic comparisons showing different ranges of motion provided by various types of cervical orthoses (based on data from Johnson et al.).33 (Used with permission from Barrow Neurological Institute.)

Displaced acute type II odontoid fractures can be treated with odontoid screw fixation. This technique requires the transverse ligament to be intact. If the ligament is disrupted, posterior C1–C2 fusion is indicated. This procedure requires visualization of the odontoid process in lateral and anteroposterior projections, which is facilitated by using two C-arms ( Fig. 32.6 ). When available, frameless stereotactic image guidance can be useful.


Odontoid screw fixation includes a standard anterior cervical exposure of the C2–C3 interspace from the anterior cervical incision at C4–C5 ( Fig. 32.7 ). A pilot hole is drilled at the base of C2 ( Fig. 32.8 ), which is cannulated across the fracture with a Kirschner (K) wire. Next, a cannulated drill is used to insert a partially threaded screw to obtain a lag effect. Sasso et al.7 found that the stability of fixation with either one or two screws is surprisingly equivalent.


We prefer a single cannulated lag odontoid screw for the fixation of type II odontoid fractures. This technique has two primary advantages: it provides immediate reduction, and fusion of the odontoid process does not affect the axial rotation of C1–C2. During exposure, a tubular retraction system can be used to retract soft tissue.8

Fig. 32.5 Classification of fractures of the second cervical vertebrae. Type I, II, and III odontoid fractures are, respectively, through the apex of the dens, across the base of the dens, and into the body of C2. Hangman’s fractures involve spondylitic fractures bilaterally across the pars interarticularis. Miscellaneous C2 fractures include the laminae, facets, spinous process, and body of C2 (i.e., nonodontoid, nonhangman’s fractures). (Used with permission from Barrow Neurological Institute.)


Hangman’s Fracture


A bilateral fracture of the pars interarticularis of C2 is known as a hangman’s fracture. The mechanism underlying this type of fracture is hyperextension, which also can be associated with fractures of the posterior arch of C1. When the C2 pars fractures, the spinal canal widens. Usually, no neurological deficit occurs. In patients with no spondylolisthesis, a cervical collar is usually enough to treat a hangman’s fracture. When C2–C3 subluxation is > 4 mm or C2–C3 is angulated > 11°, halo fixation or surgical fixation is recommended.9 In patients with locked facets at C2–C3 or in patients whose alignment cannot be maintained, surgical fixation is indicated.


Unstable hangman’s fractures can be treated with anterior cervical fusion of C2–C3, posterior fusion of C1–C3, or direct fixation of the pars fracture with lag screws. The latter avoids compromising adjacent levels and does not limit the range of axial rotation at C1–C2.10



Combined C1–C2 Fractures


Combined fractures should be analyzed individually. In general, however, combined injuries can be managed with a halo brace except when the transverse ligament is disrupted or when a type II odontoid fracture with > 6 mm of displacement is present. For such cases, early surgical fixation should be considered.



Ligamentous Injuries


Occipitoatlantal dislocation (OAD) and atlantoaxial dislocation (AAD) are well-documented, devastating injuries that are usually fatal. In both cases, enormous distractive energy is applied and pulls the head vertically. In the case of OAD, this energy separates the skull base from C1. In the case of AAD, C1 is separated from C2. In the worst-case scenario, a combined OAD and AAD injury, C1 is isolated from the occiput and from C2 ( Fig. 32.9 ). How the CVJ is distracted at two different levels by the same type of mechanism is unknown, as is the site at which the energy of the injurious force is absorbed.


The complex ligamentous array at the CVJ keeps C1 sandwiched between the occiput and C2. During pure axial distraction of the CVJ, different types of injury will manifest, depending on where the vertical portion of the cruciate ligament is sectioned ( Fig. 32.9 ). In both OAD and AAD, the transverse ligament is likely intact because the injurious force is distractive along the spinal axis. These patients seldom show evidence of anteroposterior subluxation. If all the vertically compressive ligaments (apical, alar, and tectorial membrane) are sectioned, and the vertical portion of the cruciate ligament is ruptured above the transverse ligament, the result is OAD ( Fig. 32.9 ). If the articular capsules between C1 and C2 and the vertical segment of the cruciate ligament inferior to the transverse ligament fail, distraction would occur at C1–C2. The result would be AAD instead of OAD.

Fig. 32.6 Organization of the operating room showing the relative positions of the patient, personnel, and dual-image intensifier. (Used with permission from Barrow Neurological Institute.)
Fig. 32.7a, b a The patient is positioned supine with the head and neck extended to facilitate placing the screw into the tip of the dens. A trans-verse incision is made over the C4–C5 interspace to provide a horizontal trajectory, parallel to the anterior surface of the spine. b The appropriate screw angle with respect to the patient’s chest, as shown, cannot be achieved in a patient with a short neck or barrel chest. (Used with permission from Barrow Neurological Institute.)
Fig. 32.8 Cannulated screws use the Kirschner (K) wire to direct the insertion of the hollow tools and hollow screws into the bone. (Left) The K wire is drilled into the dens to fixate the fracture and to provide a guide for the screw. (Middle) A self-tapping screw is inserted directly over the K wire. (Right) The K wire is removed. (Used with permission from Barrow Neurological Institute.)

We hypothesize that if lesions occur both above and below the transverse ligament, a combined OAD/AAD injury or OAD alone will occur ( Fig. 32.9 ).



Occipitoatlantal Dislocation


OAD is a devastating injury that very few individuals survive.11,12 Multiple landmarks are used to establish this diagnosis accurately. A Power’s ratio > 1 is one of the most sensitive, objective measurements used to identify this type of injury.11,13 Unfortunately, no radiographic method is completely accurate for detecting OAD.14


The diagnosis of OAD in children is more challenging than in adults. On sagittal and coronal CT reconstructions, Pang et al. measured the distance between the occipital condyle and the lateral mass of C1 in normal children. A separation > 2 mm was highly suggestive of OAD.15,16 If the distance is > 4 mm, OAD is confirmed.15,16



Treatment

OAD implies a severe ligamentous disruption that will not heal by itself;17 immediate reduction and internal fixation are required. The ideal fixation system would provide enough support while compromising only the involved segments (occipitoatlantal joint). Fixation should be immediate, and no hardware would be inserted into the spinal canal, which can itself be abnormally compressed.18


Different systems and techniques have been described for fixation of the CVJ. These systems can be classified into two broad groups: techniques that use wires and those that involve plates and screws. Wiring and screw techniques imply anchorage to the sub-occipital surface connected to a rod and plate secured to the spine, typically with sublaminar wiring. A contoured braided loop is a simple, inexpensive, and useful resource for fixating the CVJ to the desired level ( Fig. 32.10 ).19 However, the laminae of the involved levels must be intact so that the wires can be tightened around them. Several techniques incorporate plates and screws. Screws can be inserted into the occipital bone, or epidural “washers” can be connected to a frame anchored to the spine with transarticular (C1–C2) screws.20 Occipital screws risk penetration of the inner table of the calvarium or the dura.


Biomechanically, screw fixation is superior to fixation with wire or cable. After numerous repetitive cycles, wire fixation tends to lose strength from fatigue.20 Unlike cables or wires, which fixate the bone through a cerclage effect, screws rigidly engage bone. However, their efficacy depends on the quality of the bone in which they are inserted because the bone–screw interface is the weakest point of screw fixation.

Fig. 32.9a–d Schematic representation of the different ligamentous injuries that involve the CVJ. a Intact cruciform ligament. b Rupture of the superior band of the cruciform ligament causes a wide separation between C1 and the occipital condyles. c Rupture of the inferior band of the cruciform ligament causes a wide opening between C1 and C2. d Complete isolation of C1 due to rupture of the cruciform ligament above and below the transverse ligament. Due to the distractive mechanism of injury, the transverse ligament is preserved in occipitoatlantal dislocation (OAD) and atlantoaxial dislocation (AAD). (From Gonzalez LF, Klopfenstein JD, Crawford NR, Dickman CA, Sonntag VK. Use of dual transarticular screws to fixate simultaneous occipitoatlantal and atlantoaxial dislocations. J Neurosurg Spine 2005;3:318–323. Used with permission from Journal of Neurosurgery.)
Fig. 32.10a–e a Contoured threaded Steinmann pin, shaped like a U, the width of which depends on the patient’s size and on the curvature of the CVJ. The bend typically resembles a hockey stick. b Bender used to create a uniform U shape. It is also used to bend the rod to the desired curvature to contour to the shape of the CVJ. c Close-up of the rod bender shows the three different aperture orifices that can be rotated to bend the contoured rod into the U shape. d Contoured Steinmann pin affixed to the CVJ with laminar wires and to the skull through burr holes. e Example of the bone graft placed under compression between the occiput and C2 and wired to secure its position. (a–c From Apostolides PJ, Sonntag VKH, Dickman CA. Occipitocervical wiring techniques. In Dickman CA, Spetzler RF, Sonntag VKH, eds. Surgery of the Craniovertebral Junction. New York: Thieme; 1998:800. d and e Used with permission from Barrow Neurological Institute.)

In the case of pure OAD with no other superimposed injuries, the best available treatment from a biomechanical perspective is fixation and fusion of the compromised segment. That is, the occipitoatlantal joint could be fixated with transarticular screws crossing the occipitoatlantal joint (C1–occiput) ( Fig. 32.11 ) or with a C1 lateral mass screw and interconnected with a rod fixated to a plate affixed to the skull with screws ( Fig. 32.12 ). Because the thickest part of the suboccipital bone is involved, we prefer systems that use keel screws. These techniques preserve the normal range of motion during flexion- extension and axial rotation at C1–C2.


Occipitoatlantal transarticular screws were first reported by Grob in 200118. These screws provide direct rigid screw fixation of the occipitoatlantal joints bilaterally. This technique is based on a principle similar to that underlying the Magerl technique for C1–C2 transarticular screw fixation.1,21


At our institution, the anatomy and biomechanics of occipitoatlantal fixation with transarticular screws were investigated in terms of the feasibility and safety of inserting screws in bony structures in relation to the surrounding vascular structures (vertebral arteries), hypoglossal nerves, and spinal cord.22


The occipitoatlantal joints must be realigned intraoperatively. The gap should be reduced. This reduction can be performed preoperatively by placing the halo under compression. The entry point of the screws is the midpoint of the C1 lateral mass posteriorly under the sulcus arteriosus of C1 (the same entry point for C1 lateral mass screws). The trajectory of the screw is oriented rostrally in the sagittal plane through the atlanto-occipital joint into the occipital condyle. The target on a lateral radiograph projection is an imaginary point 1 cm rostral to the anterior arch of C1. The trajectory of the screw is positioned to avoid the hypoglossal canal, spinal canal, and vertebral arteries. The fixation is performed with a cannulated screw technique using a K wire to anchor the dislocated bone followed by drilling and placement of the screw to fixate the bones directly. The technique is performed under fluoroscopic guidance. Intraoperative navigation and intraoperative computed tomography (CT) can also be used. A piece of autologous iliac crest bone graft is wired to the surface of the occiput into the posterior arch of C1 as in a C1–C2 fusion.


The disadvantages of transarticular screw fixation include the potential to injure the vertebral artery in the sulcus arteriosus.


An alternative to fusing the occiput–C1 complex is a technique that uses C1 lateral mass screws interconnected with rods to a plate affixed to the skull through keel screws. This construct provides good stability but to a lesser extent than transarticular screws, especially during extension and lateral bending.23


Appropriate preoperative measurement of the occipital keel is key to preventing cerebellar hemorrhage, epidural hematomas, and leakage of cerebrospinal fluid.

Fig. 32.11a–c The C1–occipital transarticular screw fixation. a Lateral projection shows the trajectory and purchase into the occipital condyle. b Coronal view shows the medially tilted trajectory with an entry point just under the sulcus arteriosus. The bone graft is left under compression between C1 and the suboccipital surface. c Sagittal CT reconstruction of a patient with an isolated OAD shows the screw trajectory and its position just below the hypoglossal foramen. (Used with permission from Barrow Neurological Institute.)
Fig. 32.12a–c The C1–occipital keel fixation. a Sagittal view shows the C1 lateral mass screw connected to a rod that is anchored rostrally to the suboccipital region through keel screws. b Coronal view shows the bent rods with a medial insertion at the occipital keel and anchored inferiorly with C1 lateral mass screws c Lateral radiograph shows the screws under the sulcus arteriosus of C1 and the rostral anchorage at the keel. (Used with permission from Barrow Neurological Institute.)

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 14, 2020 | Posted by in NEUROSURGERY | Comments Off on 32 Trauma to the Craniovertebral Junction

Full access? Get Clinical Tree

Get Clinical Tree app for offline access