Posterior Cervical Arthrodeses: Occiput-C2 and C1-C2

 

 

 

 

DEFINITION

The occipitocervical articulation is formed by the occiput, the atlas (C1), and the axis (C2). This functional unit provides a large degree of mobility and range of motion through the strong ligamentous structures and cup-shaped joints.

Over 50% of the total axial rotation occurs between C1 and C2, whereas flexion-extension movement predominantly occurs at the occipitoatlantal junction which allows approximately 20 degrees of

extension.1

Cervical spine range of motion is significantly decreased in children after posterior occipitocervical arthrodesis. Axial rotation is the most affected, decreasing 30 degrees in each direction, flexion and

extension each decrease by 13 degrees, and lateral bending by 7 degrees in each direction.25

Excessive movement at this junction due to either bony or ligamentous abnormalities causes instability. A wide variety of pathologies such as genetic and congenital developmental abnormalities, trauma, tumors, and inflammatory and degenerative conditions can lead to upper cervical spine instability.

Depending on the degree of displacement and spinal canal compromise, cord compression and myelopathy may occur.

Major instability is usually addressed with surgical occipitocervical or C1-C2 arthrodesis. Since Foerster first described a technique for occipitocervical arthrodesis using fibular strut graft in 1927, several procedures have been reported with variable rates of fusion and techniques of stabilization.

In this chapter, brief information on upper cervical spine instability is given and general principles of occipitocervical and C1-C2 arthrodesis are discussed. Also, different techniques developed for posterior occipitocervical fusion are described in detail.

Modern instrumentation techniques have dramatically changed the instrumentation potential and wiring strategies are being replaced for screw instrumentation techniques. These techniques can decrease the risk of cord damage from wire passage, increase biomechanically fixation, and improves fusion rates.13,

16

 

ANATOMY

 

It is important to understand that the pediatric upper cervical spine is not a “miniature model” of the adult spine. The cervical spine approaches adult size and shape by ages 8 to 12 years, as growth cartilage fuses and vertebral bodies gradually lose their oval or wedge shape and become more rectangular.

 

The upper cervical spine has unique development, anatomy, and biomechanics.

 

 

The C1 develops from three ossification centers, a body and two neurocentral arches, which become visible by age 1 year (FIG 1A).

 

The neurocentral synchondroses fuse with the body at about 7 years of age and may be mistaken for fractures on radiographs.18

 

The C2 is derived from five primary ossification centers: the two neural arches or lateral masses, the two halves of the dens, and the body.

 

 

There are two secondary centers: the ossiculum terminale and the inferior ring apophysis (FIG 1B).

 

The two halves of the odontoid are generally fused at birth but may persist as two centers, known as the

dens bicornis.19

 

The dentocentral synchondrosis, which separates the dens from the body, closes between the ages of 5 and 7 years (FIG 1B). Until the ossification is complete, it gives the appearance of a “cork in a bottle” on an open mouth odontoid view.

 

 

 

The tip of the dens appears at age 3 years and is fused by age 12 years. Occasionally, it remains as a separate ossiculum.11, 19

 

After skeletal maturity, the C1 does not have a body as such and is shaped like a ring. The flat, cup-shaped articular surface under the occipital condyle allows for flexion, extension, and some bending. The dens

articulates with C1 through the dorsal facet of the anterior arch. C1-C2 articulation allows for rotation.1 The vertebral artery passes through the foramen that is located in the transverse processes.

 

The ligamentous structure allows for a wide range of motion of the upper cervical spine while maintaining stability. The short ligaments at the base of the skull are as follows (FIG 1C):

 

 

The tectorial membrane anteriorly from the foramen magnum is a continuation of the posterior longitudinal ligament that provides considerable support.

 

Cruciate ligament, which includes transfer ligaments, restrains against atlantoaxial anteroposterior (AP) translation.

 

Alar and apical ligaments, which run from foramen magnum to the odontoid, act as secondary stabilizers.

 

Posterior atlanto-occipital ligaments are the continuation of the ligamentum flavum.

 

Vertebral artery anatomy can also be different in children compared to adults. Younger patients have vertebral arteries that are significantly closer to the midline over the superior border of C1 than older patients. Ninety-

seven percent of vertebral arteries are greater than a centimeter lateral to the midline.12

 

PATHOGENESIS

 

Fundamentally, upper cervical spine instability can develop from osseous or ligamentous abnormalities resulting from acquired or congenital disorders. As a result of instability, excessive motion and spinal cord compression may occur at the occipitoatlantal and/or atlantoaxial joint.

 

 

P.702

 

 

 

FIG 1 • A-C. Anatomy ossification centers of the C1, C2, and cervical vertebrae during development.

 

 

In nontraumatic conditions, ligamentous laxity (particularly in the transverse ligament) or abnormalities of the odontoid cause instability.

 

In Grisel syndrome, a type of atlantoaxial rotatory displacement, inflammation of the retropharyngeal space, caused by upper respiratory tract infections or by adenotonsillectomy, spreads through the pharyngovertebral veins to the ligaments of the upper cervical spine. This results in impairment of the transverse atlantal ligament

and instability.27

 

In Down syndrome, the main cause of atlantoaxial instability is the laxity of the transverse ligament, which holds the dens against the posterior border of the anterior arch. Also, malformation of the odontoid can be

observed in this condition.8

 

Klippel-Feil syndrome is associated with congenital cervical anomalies, such as occipitocervical synostosis, basilar impression, and anomalies of the odontoid, and can be associated with instability, stenosis, or both.

 

Odontoid anomalies include aplasia, hypoplasia, duplication, third condyle, persisting ossiculum terminale, and os odontoideum.

 

NATURAL HISTORY

 

Patients with upper cervical instability frequently have other associated pathologic conditions in the occipitocervical region such as spinal stenosis, basilar impression, cervical fusions, occipitalization, or congenital anomalies of the C1 or C2 (dens), and central nervous system abnormalities.

 

 

When one encounters one of these conditions, others should be sought also.

 

Instability of the upper cervical spine and stenosis often are two major factors in the development of myelopathy.

 

Patients who are symptomatic at initial presentation are often at risk for progressive neurologic symptoms. Once cervical myelopathy develops, it rarely resolves entirely.

 

Paralysis and death are rare but may be encountered in patients with upper cervical spine instability.

 

PATIENT HISTORY AND PHYSICAL FINDINGS

 

Upper cervical spine instability is rare in patients without predisposing conditions or trauma.

 

The instability is usually determined in radiographic examination of the children with syndromes or conditions

known to have frequent involvement of the musculoskeletal system.24 An orthopaedic surgeon is usually consulted for children with such conditions.

 

Clinical presentation can vary because of the associated syndromes and anomalies.

 

Patients may present with symptoms such as loss of range of motion, stiffness, mechanical pain of the head or neck, and torticollis.

 

It is not uncommon to see patients presenting with neurologic symptoms, which can vary from minor sensory or motor disturbances to established myelopathy. Neurologic symptoms or signs result from mechanical compression of the spinal cord or nerve roots.

 

Torticollis may be the presenting symptom of rotatory or postinfectious atlantoaxial instability.

 

According to the degree of compression and the affected site of the spinal cord, signs and symptoms can vary. They may include loss of physical endurance, difficulty walking, weakness, and upper motor neuron signs (spasticity, hyperreflexia, clonus, Babinski sign), which can be seen with anterior spinal column involvement.

 

 

P.703

 

Pain deficits and proprioception and vibratory sense deficits can be seen with posterior spinal column involvement.

 

Increased nasal resonance may also be observed. It may occur because of the decreased size of the nasopharynx resulting from anterior displacement of the C1.

 

Vertebral artery distortion and insufficiency may lead to bizarre symptoms such as syncopal episodes, sudden postural collapse without unconsciousness, change in behavior, dizziness, and seizures.

 

In cerebellar involvement, nystagmus, ataxia, and incoordination are the common findings.

 

Neurogenic bladder and bowel, cranial nerve involvement, paraplegia, hemiplegia, and quadriplegia should be kept in mind; sometimes, the patient presents with only one of these findings.

 

IMAGING AND OTHER DIAGNOSTIC STUDIES

 

Standard radiographs include AP, open mouth odontoid, and lateral (neutral and flexion-extension) cervical spine views.

 

Instability can be identified on the lateral flexion-extension view. Atlantoaxial instability is diagnosed on the basis of an increased atlantodental interval (ADI).

 

 

The ADI is measured from the anterior aspect of the dens to the posterior aspect of the anterior ring of the C1 (FIG 2A).

 

In children older than 8 years and in adults, the ADI should be 3 mm or less, whereas in younger children, the ADI should be 4 mm or less (some consider 4.5 to 5 mm acceptable).14

 

In children, we consider an ADI of 4 mm or more as evidence of atlantoaxial instability. This measurement does not always correlate with the degree of brainstem or cord compression (as seen on magnetic resonance imaging [MRI]), however. An asymptomatic patient may have instability.

 

Space available for the spinal cord (SAC) is measured from the posterior border of the dens to the anterior

border of the posterior tubercle. According to Steel's rule of thirds,22 SAC should be about one-third of the diameter of the ring of C1 (see FIG 2A).

 

 

 

 

FIG 2 • A. Lateral craniometry of the craniocervical junction with landmarks, commonly used lines, and methods for examining the relationship between the C1, odontoid, and foramen magnum and measuring the SAC. The ADI is measured from the anterior aspect of the dens to the posterior aspect of the anterior ring of the C1. The McRae line connects the anterior rim of the foramen magnum to the posterior rim. The Chamberlain line is drawn from the posterior margin of the hard palate to the posterior margin of the foramen magnum. The McGregor line is drawn from the most caudal point of the occipital projection to the posterior edge of the hard palate. The Wackenheim line is drawn parallel to the posterior surface of the clivus. B. Method for calculating the Wiesel-Rothman line for atlanto-occipital instability. A line is drawn connecting the anterior and posterior arches of the C1 (line 1 to 2). Two lines are drawn perpendicular to this line, one through the basion and the other through the posterior margin of the anterior arch of the C1 (line 3). A change in the distance (x) between these lines of more than 1 mm in flexion and extension indicates increased abnormal translational motion. C. Lines used to calculate the Power ratio. A line is drawn from the basion (B) to the posterior arch of the C1 (C) and a second line from the opisthion (O) to the anterior arch (A) of the C1. The length of the first line is divided by the length of the second.

 

 

This safe zone allows for some degree of pathologic displacement. Displacement of more than one-third of the diameter causes cord compression.

 

 

 

This measurement directly describes the SAC, which is highly associated with the neurologic involvement. The relationship between the foramen magnum, C1, and odontoid can be determined in lateral radiographs. The line of McRae connects the anterior rim of the foramen magnum to the posterior rim (see FIG 2A).

 

The upper tip of the odontoid should normally be 1 cm below the anterior margin of the foramen magnum.

 

If the effective sagittal diameter of the canal (length of the line) is less than 19 mm, neurologic symptoms occur.

 

The line of Chamberlain is drawn from the posterior margin of the hard palate to the posterior margin of the foramen magnum (see FIG 2).

 

 

The tip of the odontoid should be 6 mm below this line. It bisects the line in basilar invagination. However, determination of the landmarks can be difficult on plain radiographs.

 

The McGregor line is drawn from the most caudal point of the occipital projection to the posterior edge of the hard plate (see FIG 2A).

 

This line is one of the best for detecting basilar impression because the osseous landmarks can usually be seen at all ages. If the tip of the odontoid process lies more than 4.5 mm above this line, the finding is consistent with basilar impression.

 

The line of Wackenheim is drawn parallel to the posterior surface of the clivus (see FIG 2A).

 

 

The inferior extension of the line should be in touch with the posterior tip of the odontoid. In basilar invagination, it is over that line.

 

The Wiesel-Rothman line is drawn connecting the anterior and posterior arches of the C1. Two lines are drawn perpendicular to this line, one through the basion and the

 

P.704

other through the posterior margin of the anterior arch of the C1 (FIG 2B).

 

A change in the distance (x) between these lines of more than 1 mm in flexion and extension indicates increased abnormal translational motion.

 

The ratio of Power is calculated from a line drawn from the basion to the posterior arch of the C1 and a

second line from the opisthion to the anterior arch of the C1 (FIG 2C). The length of the first line is divided by the length of the second.

 

 

A ratio of less than 1.0 is normal.

 

A ratio of 1.0 or more is abnormal and is diagnostic of anterior occipitoatlantal dislocation.

 

MRI is useful to identify pathologic changes at the dura mater and spinal cord as well as additional soft tissue pathologies.

 

 

Functional MRI scans performed in flexion and extension can be used to assess dynamic brainstem or cord compression.

 

CT scan is essential for screw trajectory planning and it can also provide additional information regarding the bony anomalies.

 

 

Two-dimensional (2-D) and three-dimensional (3-D) reconstructions can clarify the course of the vertebral artery and careful attention should be paid to the location of the artery through the foramen transversarium of C2, where is most likely to be injured.

 

In atlantoaxial rotational displacement, pathoanatomy is determined by fine-cut dynamic CT scan with left-right rotation of the head.

 

CT or magnetic resonance (MR) angiogram can be useful to evaluate the vertebral artery anatomy prior to instrumentation of C1-C2.

 

DIFFERENTIAL DIAGNOSIS

Pseudosubluxation Os odontoideum

Congenital muscular torticollis Ankylosing spondylitis

 

 

NONOPERATIVE MANAGEMENT

 

Children with known risk of upper cervical instability should be evaluated carefully. Especially patients with congenital syndromes associated with upper cervical spine instability should have periodic clinical and

radiographic examinations until maturity.24

 

Upper cervical spine radiographs including AP, lateral (neutral and flexion-extension views), and open mouth odontoid views are obtained periodically to assess and detect any trends and changes.

 

Patients and parents should be educated about the diagnosis and natural history of the disorder and encouraged to report any symptoms as soon as they occur.

 

Because of bone and ligament abnormalities, patients with upper cervical spine instability have a greater risk of spinal cord injury even with minor trauma and even when they are asymptomatic.

 

As previously described, periodic observation should be done, and if any progression is noticed, the patient should be prepared for appropriate surgical stabilization when indicated.

 

 

In children, we consider an ADI of 4 mm or more as evidence of atlantoaxial instability. Documented significant instability at the atlanto-occipital or atlantoaxial joints is an indication for posterior arthrodesis of occiput-C2 and C1-C2 arthrodesis, respectively.

 

In some congenital disorders such as Morquio syndrome, progression of the instability is frequent; in these cases, prophylactic fusion should be considered before neurologic symptoms occur.24

 

However, in Down syndrome, the patients with instability are usually asymptomatic, and in most cases, signs and symptoms progress slowly. Restriction of high-risk activities usually is appropriate. If the clinical symptoms persist or neurologic symptoms are starting to occur in the setting of significant instability, surgical treatment is

indicated.8, 21

 

Children with congenital fusion of cervical vertebrae (mostly in Klippel-Feil syndrome) should be restricted from high-risk sports. Patients with progressive symptomatic segmental instability or neurologic compromise

are candidates for surgical stabilization.23

 

SURGICAL MANAGEMENT

 

The main indication for the posterior occiput-C1 or C1-C2 arthrodesis is instability of the atlanto-occipital or atlantoaxial joint.

 

Many techniques of atlantoaxial fusion using cables, transarticular screws, plates, and bone graft materials have been described.

 

For isolated atlantoaxial instability, we describe the Gallie technique, the Mah modified Gallie technique, and the Brooks sublaminar wire technique.

 

 

The authors currently preferred instrumentation method which includes occipital plating and Harms (C1 lateral mass with C2 pedicle screw) or Magerl (C1-C2 transarticular screw) techniques are also described.

 

Preoperative Planning

 

 

Plain radiographs and CT scans are reviewed and any osseous findings noted. MRI evaluation for the spinal cord compression is recommended.

 

CT scans and MR images should be reviewed to evaluate the course of the vertebral arteries, the dimensions of the C1 lateral mass, the trajectories and sizes of the C2 pedicles, and the thickness of the keel of the occipital bone.

 

Occipital screws risk dural sinus injury. The safe zone for occipital screw placement is a triangular region created by connecting two dots 2 cm lateral to the midline just distal to the external occipital protuberance (EOP) and a midline point 2 cm inferior to the EOP9, 17( FIG 3).

 

An appropriate halo ring is measured for the patient.

 

Somatosensory evoked potentials, transcortical motor evoked potentials, and electromyography are the preferred neurologic monitoring modalities.

 

Flexible fiberoptic intubation with manual in-line axial stabilization should be considered to minimize cervical motion during intubation maneuvers.

 

Positioning

 

After induction of general anesthesia in the supine position, a halo ring is applied, and the patient is carefully turned to the prone position.

 

 

P.705

 

 

 

 

FIG 3 • Pertinent intracranial (A) and extracranial (B) occipital anatomy. Highlighted are the safe zones for screw instrumentation. (From Roberts DA, Doherty BJ, Heggeness MH. Quantitative anatomy of the occiput and the biomechanics of occipital screw fixation. Spine 1998;23:1100-1108.)

 

 

Head is adjusted with the halo device supported by a head positioner, which is securely fixed to the operating table. A Mayfield positioning device can also be used (FIG 4).

 

The table is placed on slight reverse Trendelenburg position to facilitate venous drainage and reduce facial swelling.

 

The shoulders are taped to improve radiographic visualization, and the patient's hair is shaved 2 cm above the EOP.

 

Lateral fluoroscopy or radiography should be done before starting the procedure to confirm the alignment of the occiput and cervical spine.

 

Donor sites (rib or iliac crest) are also prepared for graft harvesting.

Approach

 

A posterior midline incision is made from the occiput to the intended distal fusion site (usually C2) for occipitocervical arthrodesis.

 

 

 

 

FIG 4 • A. Patient's head positioning with halo device supported by a head positioner, which is securely fixed to the operating table. B. Prone positioning on radiolucent table, shoulders are taped distally to improve x-ray visualization.

 

 

Proximally, midline dissection up to the EOP is performed for occipitocervical instrumentation.

 

Through an exact midline split of the nuchal ligament, the paraspinal muscles are elevated with electrocautery and Cobb elevators and retracted laterally. Excessive lateral dissection should be avoided to prevent injury to the vertebral artery, which runs in a serpentine course in relationship to the C2 vertebra.

 

For atlantoaxial arthrodesis, the dissection is started from the lower occiput and the surgeon identifies the posterior arch of the C1, the bifid spinous processes, and the lamina of the C2. The C2-C3 interval should not be exposed.

 

Care should be taken during exposure to avoid dissection immediately superior to the arch of C1 where the vertebral artery can be found in children 1 cm lateral to the midline.

 

The surgical exposure of vertebra is limited up to the intended level of fusion to prevent unintentional inclusion of the adjacent level in the fusion mass.

 

 

P.706

 

TECHNIQUES

• Occipital Plating and C1-C2 Harms Screws Instrumentation

Subperiosteal exposure proximally to the level of the EOP is performed; bone wax can be used to control bony bleeding.

A 4.5-mm screw placement is planned over the occipital safe zone as described earlier. A precontoured plate is chosen and appropriately placed below the EOP.

Occipital drill holes are hand drilled through the plate with an appropriate drill guide with fixed depths (starting at 6 mm depth) (TECH FIG 1A).

4.5-mm bone screws can be kept unicortical, and preoperative CT scan measurement can aid in the length calculation because pediatric occiput width is variable. The midline keel will allow the surgeon to place the longer and stronger screws.

If bicortical purchase is required (in cases of poor bone quality or thin occiput), progressive hand drilling

 

should be made in 2-mm increments, with frequent blunt palpation until the inner table is breached, avoiding dural penetration and cerebrospinal fluid leak.

 

Screw holes should be tapped because of the hard nature of the occipital tables.

 

If a foramen magnum decompression is to be performed in conjunction with instrumentation, care should be taken to leave enough occipital safe zone bone for screw placement.

 

The occipital plate is secured with uni- or bicortical 4.5-mm bone screws (TECH FIG 1B).

C1 Lateral Mass Screw

 

The lateral mass of C1 is covered by the C2 dorsal root ganglion.

 

C1 lateral mass is approached with careful detachment of the paraganglious tissue from the underside of the posterior arch of C1, this can be performed with fine-tipped bipolar cautery.

 

Bleeding is frequently encountered from the C1 to C2 epidural venous plexus (TECH FIG 2A) and should be controlled with bipolar cautery and thrombin-soaked Gelfoam sterile sponge (Pfizer). A Penfield no. 4 retractor can be used to gently retract the C2 dorsal root ganglion caudally exposing the inferior border of the lateral mass (TECH FIG 2B).

 

C1 screw entry point is located in the center of the C1 lateral mass (TECH FIG 2C). Overhanging bone form the posterior arch of C2 sometimes requires removal with a Kerrison rongeur without disrupting the integrity of the C1 arch (TECH FIG 2D).

 

 

 

TECH FIG 1 • A. Occipital drill holes are hand drilled through the plate with an appropriate drill guide with fixed depths (starting at 6 mm depth). B. Appropriate occipital plate positioning. In this case, a foramen magnum decompression was performed distal to the plate location.

 

 

A 2-mm high-speed burr can be used to mark the entry point and avoid slippage with the hand drill. The screw on the coronal plane is directed 0 to 5 degrees medially and parallel to the inferior border of the C2 lateral mass (20 to 30 degrees cranial) on the sagittal plane (TECH FIG 2E,F).

 

The screw is tapped, and we recommend bicortical placement of a 3.5-mm polyaxial of the appropriate length. Although the lateral mass will be 10 to 15 mm deep, the C1 screw should be a longer partially threaded screw that will allow for the unthreaded segment to be proud posteriorly and in contact with the C2 nerve root avoiding irritation of the greater occipital nerve.

 

The proud polyaxial head of the C2 screw will also allow for symmetric alignment with the C2 screw in order to permit rod fixation between the two segments.

C2 Pedicle Screw

 

Various C2 instrumentation techniques are valid and available but the surgeon should evaluate individual

anatomy for proper instrumentation technique. C1 pars interarticularis can be performed, but in children, the pars is usually small and we prefer the use of a pedicle-type screw.

 

A transarticular screw C1-C2 can be performed, but distal exposure or percutaneous incision need to be done to achieve the steep angle for insertion of such a screw. For these reasons, we favor the use of C2 pedicle screws.

 

The exposure of C2 is carried out to the lateral edge of the lateral mass and not past this point because the vertebral arteries lie lateral to it. Complete visualization of the dorsal isthmus as well as the medial and lateral borders is possible and necessary in pediatric patients. Given the smaller anatomy, it is necessary to ensure safety of screw placement via thorough exposure.

 

The starting point for the C2 pedicle screw is the intersection of a horizontal line through the midline of the lamina and a vertical line through the center of the pars interarticularis (see TECH FIG 2C).

 

The medial border of the pedicle as well as the superior articular facet of C2 can be palpated with a Penfield no. 4 elevator and the starting point marked with a 2-mm burr. A hand drill is used, and the screw hole trajectory should be 20 to 30 degrees medially and cephalad; this angulation should be individualized for each patient with evaluation of the CT scan (TECH FIG 3A).

 

P.707

 

 

 

TECH FIG 2 • A. Diagram demonstrating the C1-C2 epidural venous plexus. Bleeding is frequently encountered and bipolar cautery should be used for control. B. Penfield no. 4 retractor can be used to gently retract the C2 dorsal root ganglion caudally exposing the inferior border of the lateral mass. C. The posterior entry points for C1 and C2 screws. D. Overhanging bone from the posterior arch of C2 sometimes requires removal with a Kerrison rongeur without disrupting the integrity of the C1 arch. E,F. Orientation of C1 lateral mass screw. The screw on the coronal plane is directed 0 to 5 degrees medial and parallel to the inferior border of the C2 lateral mass (20 to 30 degrees cranial) on the sagittal plane. (C,D: From Melcher RP, Harms J. C1-C2 posterior screw rod fixation In: Bradford DS, Zdeblick T, eds.

Master Techniques in Orthopaedic Surgery: The Spine, ed 2. Philadelphia: Lippincott Williams & Wilkins, 2004:129-145.)

 

 

The screw hole should be bicortical, and a ball-tipped probe can be used to palpate anterior breech. The hole is tapped, and measurement of the appropriate length with a depth gauge is performed and checked on fluoroscopy.

 

Decortication of the lamina and pars of C2 is done followed by placement of a 3.5-mm screw of

appropriate length. The polyaxial head of the C2 screw should be in line with the C1 screw head.

 

After C1-C2 instrumentation is completed, a 3.5- or 4.5-mm rod is contoured to provide neutral sagittal occipitocervical alignment. The surgeon must contour the rod carefully to avoid kyphosis. Radiographic assessment of positioning should be performed with fluoroscopy or final x-rays (TECH FIG 3B).

 

If further reduction is required (occasionally, reduction is obtained with positioning), the screw heads can be used to generate anterior translation of the posteriorly dislocated facet. The bony stock of C1 and C2 will not allow for major manipulation in a young child and the surgeon should be careful to not plow a screw. Traction and gentle manipulation can also be helpful when reduction is required. The rod can be placed loosely and tightened once successful reduction is achieved.

 

Biomechanically plating with six occipital screws connected to a C2 pedicle or a C1-C2 transarticular construct has proven to be the most rigid occipitocervical fixation (TECH FIG 3C).20

 

Occiput, C1, and C2 decortication is performed.

 

 

P.708

 

 

 

TECH FIG 3 • A. C2 pars screw direction, the medial border of the pedicle as well as the superior articular facet of C2 can be palpated with a Penfield no. 4 elevator and the starting point marked with a 2-mm burr. A hand drill is used and the screw hole trajectory should be 20 to 30 degrees medially and cephalad. B. Postoperative radiograph demonstrating final occiput-C1-C2 construct. C. Final occiput-C1-C2 construct. A foramen magnum decompression was performed in this case for a Chiari decompression in conjunction with instrumentation.

 

 

Morcellized posterior superior iliac spine (PSIS) iliac crest autograft and allograft are packed into the fusion area to add additional support.

 

The surgical area is irrigated, hemostasis is obtained, and the incision is closed in three layers: The supraspinous ligament can be sutured to the spinous process of C2, then watertight fascial closure is performed with 1-0 Vicryl; subcutaneous tissue is closed with interrupted 3-0 Vicryl; and a 3-0 running Monocryl subcuticular is used for the skin.

 

Additional halo immobilization is only necessary if the surgeon encounters poor bone quality or if there is concern for wound complications secondary to prominent occipital hardware.

C1-C2 Transarticular Screws and C2 Translaminar Screws

 

The technique is similar to a pedicle screw. A cannulated system can be used to help with the increased precision required for this instrumentation technique.

 

A threaded guidewire is used to cannulate the pars, making sure that the borders remain well visualized and defined by the assistant.

 

The lateral fluoroscopic view is frequently checked, as the inclination of the pars screws in the cephalocaudal direction is paramount. The most frequent error in trajectory would be aiming to anterior and either missing C1 or only catching the anteroinferior edge of C1.

 

A Penfield in the C1-C2 joint can help confirm position both by direct feel and radiographically. When reached with the guidewire, the Penfield is removed and the wire advanced into C1 (TECH FIG 4A,B).

 

After confirmation of the appropriate positioning, drill and tap over the guidewire. Decortication of the lamina and pars of C2 is done followed by placement of a 4.0-mm screw ranging from 32 to 40 mm (TECH FIG 4C,D).

 

Translaminar C2 screws are not described in detail but can be a salvage option when other fixation methods are not possible (TECH FIG 4E,F).

 

 

 

TECH FIG 4 • C1-C2 transarticular screw. A. Penfield in the C1-C2 joint can help confirm position both by direct feel and radiographically. B. When reached with the guidewire, the Penfield is removed and the wire advanced into C1. (continued)

 

 

P.709

 

 

 

TECH FIG 4 • (continued) C. C1-C2 transarticular screw AP direction. D. The polyaxial head of the C2 screw should be in line with the C1 screw head on sagittal plane and coronal plane as well. E,F. Placement of translaminar screws.

• Occipitocervical Arthrodesis with Iliac Graft

   

  At the level below the transverse sinus, four transverse-oriented holes are drilled through both cortices of the occiput with a highspeed diamond drill.6

   

  The holes are aligned transversely with two on each side of the midline. At least 1 cm of intact bone should be left between the holes to prevent wire pullout through the skull (TECH FIG 5A).

   

  Surgical loupes and a headlamp are recommended for this procedure.

   

  Using a high-speed diamond burr, the surgeon makes a transverse-oriented trough into the base of the occiput to fit the rectangular superior part of the iliac autograft.

   

  A single corticocancellous autograft (3 × 4 cm) is harvested through an oblique incision over the PSIS.

   

  A rectangular graft is taken. The surgeon creates a notch in the inferior base of the graft to be suitable for the base of the spinous process of the second or third vertebra (TECH FIG 5B).

   

  A 16- or 18-gauge wire is passed through the burr holes on each side of the midline and the wire is looped on itself (TECH FIG 5C,D).

   

  A sublaminar wire is placed under the ring of C2 or C3 (or passed through the base of the spinous process, if structurally sufficient, or if there is canal stenosis).

   

  The left side of the graft accepts the left end of the wire and the right end of the graft accepts the right end of the wire (TECH FIG 5E).

   

  The edges of the graft are contoured to fit appropriately into the occipital trough and around the base of the spinous process (TECH FIG 5F).

   

  The wires are tightened over the graft in figure-8 shape. After satisfactory tightening, the edges of the wire are cut and bent away from skin (TECH FIG 5G,H).

   

  Intraoperative fluoroscopy or radiography is used to confirm the alignment of the occiput and cervical spine, stability, and the position of the graft and wires.

   

  The graft should be structurally stable at the end of the procedure.

   

  Flexion-extension of the halo frame, better contouring of the graft, and appropriate tightening of the wires can be used to make adjustments in reduction and alignment.

   

  P.710

   

   

   

  TECH FIG 5 • A. Four transverse-oriented occipital burr holes and rectangular trough. B. Corticocancellous rectangular graft with a notch at the inferior base of the graft. C. A Luque wire is passed through the occipital burr hole and another wire is passed sublaminarly under the arch of C2 or through the base of the C2 spinous process. D. Occipital wire is looped on itself. E. Schematic drawing showing occipital wire looped on itself and the wire passed through the base of the C2 spinous process.

  F. Graft (arrow) placed between the occiput and C2. G. The wires are tightened over the graft in a figure-

  8 shape, twisted, and cut. H. Schematic drawing showing the graft placement and securing with wires.

   

   

  P.711

• Occipitocervical Arthrodesis with Rib Graft

   

  An oblique incision overlying the posterior rib allows for adequate exposure.3

   

  The muscle fibers are separated, and dissection is carried down to the periosteum of the rib.

   

  Adequate rib is exposed and cut.

   

  The size of the rib graft is greater than the area to be fused because part of the rib is used as morcellized graft.

   

  Using a rib cutter, the graft is cut distally and proximally and removed (TECH FIG 6A).

   

  Irrigation fluid is placed in the surgical site and positive pressure applied to check for pleural leaks.

   

  If a pleural tear is detected, air can be removed from the chest cavity by using a red rubber tube and suction.

   

  A larger leak may require placement of a thoracostomy drainage tube.

   

  In all patients, a chest radiograph should be taken after rib harvest to rule out pneumothorax.

   

  Two full-thickness structural grafts are prepared to fit the arthrodesis site.

   

  The rib grafts can span large defects and fit nicely into large or abnormally shaped skull, and we find this best for young infants.

   

   

   

  TECH FIG 6 • A. Adequate rib is exposed and harvested. B. Rib graft is placed and fixed with braided cables and no. 5 Mersilene suture. C. Schematic drawing showing the rib graft fixed with wire. D,E. A 5-year-old-boy immobilized with a halo vest postoperatively. (C: Adapted from Cohen MW, Drummond DS, Flynn JM, et al. A technique of occipitocervical arthrodesis in children using autologous rib grafts. Spine 2001;26:825-829.)

   

   

  A 16- or 18-gauge wire is looped through the burr holes on each side of the midline (see TECH FIG 5C).

   

   

  The burr holes are drilled and aligned similarly to the ones described for the iliac graft technique. There is no need to create a groove at the base of the occiput.

   

  Braided cable or no. 5 Mersilene sutures may be used instead of wire.

   

  With Mersilene sutures, there is a reduced risk of cutting out in thin bone of poor quality.

   

  After this, purchase of two wires is made to the posterior elements of most caudal vertebra on each side of the midline by sublaminar wiring.

   

  Suitable grafts on either side are then secured to the occiput and lamina of the most caudal vertebra by wires.

   

  The stability of the grafts is checked under radiographic control and the wires are then crimped and cut (TECH FIG 6B,C).

   

  Adjustments are made by flexion-extension of the halo frame, contouring of the graft, and appropriate tightening of the wire.

   

  Intraoperative radiographs are obtained to confirm acceptable reduction, alignment, and placement of the graft.

   

  For both techniques, morcellized autograft is packed into the arthrodesis site. The wound is closed in layers.

   

   

  The halo vest is worn for 8 to 12 weeks after both techniques to maintain postoperative stability (TECH FIG 6D,E).

   

• Posterior C1-C2 Arthrodesis

  P.712

   

  The preoperative planning is similar to that for the occipitocervical arthrodesis described earlier in the section.

  Gallie Technique

   

  After exposing the posterior arch of the C1 and spinous process of the C2, the two free ends of a single 16- or 18-gauge wire are passed beneath the posterior arch of the C1 from a superior to inferior

  direction.10

   

  The free ends are passed beneath the posterior arch and are brought around superiorly to loop on themselves.

   

  A rectangular corticocancellous autograft is harvested from the posterior iliac spine.

   

  A notch is created at the distal part of the graft. This part will be placed across the spinous process of C2.

   

  The notched graft is placed between the posterior portion of the arch of C1 and the posterior spinous process of C2.

   

  Now the free ends of the looped wire are brought down over the graft and passed below the spinous process.

   

   

  The free ends of the wire are tightened and twisted over the graft (TECH FIG 7). Morcellized bone grafts may be packed into the fusion area to add additional support.

   

  Intraoperative fluoroscopy is necessary to check for satisfactory reduction and alignment of C1-C2.

  Mah Modified Gallie Technique

   

  In 1989, Mah described a modification of the Gallie technique.15

   

  All the steps of the Mah technique are similar to the Gallie technique, except that a threaded Kirschner wire is placed through the spinous process of the C2 and both ends of the Kirschner wire are cut, leaving about 2.5 cm of the total wire.

   

  The free ends of the looped wire are brought down below the free ends of the threaded Kirschner wire (TECH FIG 8).

   

  The free ends of the wire are tightened, secured, and crimped over the graft.

  Brooks Technique

   

   

   

  A standard posterior midline incision is used to expose the posterior arch of C1 and the lamina of C2.2 Two sublaminar wires are passed under both C1 and C2 laminas, one on each side of the midline.

   

   

   

  TECH FIG 7 • Gallie technique for atlantoaxial arthrodesis.

   

   

   

  TECH FIG 8 • Mah modified Gallie technique for atlantoaxial arthrodesis.

   

   

  Unlike the Gallie technique, two separate corticocancellous grafts are required in this technique. A single rectangular iliac crest graft is harvested; it can be separated into two equal parts.

   

  Each iliac crest graft is cut into a trapezoid-like shape (one end is narrower than the other) so that they can be wedged between the C1 and C2 posterior arches (TECH FIG 9A).

   

  The grafts are snugly wedged into place. The wires are then tightened around the grafts, twisted, and cut (TECH FIG 9B).

   

   

   

  TECH FIG 9 • Brooks arthrodesis. A. Lateral view demonstrating a wedge-shaped graft between the spinous processes to prevent hyperextension. The graft is shaped so that one end is narrower than the other to achieve a good fit. B. The grafts are snugly wedged between the C1 and C2 posterior arches, and the wires are tightened around the grafts.

   

   

   

  P.713

   

  PEARLS AND PITFALLS
  	Approach ▪ Excessive lateral dissection should be avoided so as not to damage the vertebral arteries and major venous junctions.

  • In the case of an open posterior arch, meticulous and blunt dissection should be used to keep from injuring the dura mater and the cord.

   

  • Excessive dissection and extended dissection time are avoided.

  • This may increase the risk of inadvertent extension of the fusion mass.

 

Occipital plating and C1-C2 Harms screw instrumentation

• CT imaging and evaluation of vertebral artery anatomy is mandatory.

• Instrumentation is possible in majority of pediatric patients with manageable complications.7

• Reports have shown that this technique is safe and feasible to use

  in children older than 6 years of age.

• Clinical union rates of 90%-100% have been reported.26

 

Occipitocervical arthrodesis with iliac graft technique

• This technique is associated with stable internal fixation and high fusion rates.3,11

   

  Occipitocervical arthrodesis with rib graft technique

  • A large occipitocervical segment can be spanned.

  • Rib grafts can be shaped easily because of their elastic structure.

  • This technique is best for infants, small children, or patients with an abnormally shaped skull.

     

    C1-C2 arthrodesis with Gallie technique

    • This procedure is more suitable in older children who have a competent spinous process.

    • It is not always necessary to pass wire or cable underneath the lamina of C2.

    • This technique provides good stability in flexion and extension but may be insufficient in rotational maneuvers.

       

      C1-C2 arthrodesis with Mah modification

    • The wire can securely hold behind an insufficiently ossified spinous process with the help of a transverse Kirschner wire.

       

      C1-C2 arthrodesis with Brooks technique

    • This technique provides more rotational stability than the Gallie technique.

    • Disadvantages include the need to pass bilateral sublaminar cables beneath both C1 and C2.

 

POSTOPERATIVE CARE

 

Postoperative management can include halo vest immobilization for 8 to 12 weeks. In cases of screw fixation, the halo is not necessary. If there are concerns regarding stability of the construct, poor bone quality or hardware prominence immobilization with halo can be maintained postoperatively.

 

Using a standardized method of halo application reduces the rate of complications associated with halo use in children.4

 

The rate of pin tract infection with prolonged use of a halo device in children is similar to that in adults.

Particular care should be taken to keep the pin tract sites clean.5

 

 

Short-term antibiotic treatment is usually satisfactory in decreasing inflammation at the pin site.

 

 

When a bony union is documented radiographically, the halo device is removed.

 

 

Patients can gradually return to their daily activities.

 

Special care should be taken to avoid excessive flexion or extension of the neck.

 

Long-term follow-up is necessary for evaluation of potentially progressing junctional instability below the level of fusion.

 

 

The additional stress placed on the adjacent vertebrae below the level of fusion may result in instability with time.7

 

 

OUTCOMES

Screw fixation techniques in children have demonstrated excellent outcomes.

Several reports have demonstrated a better than 90% fusion rates and normal alignment on postoperative imaging studies.26

Occipitocervical fusions with plates and screws have proven to be relatively safe and effective in treating pediatric patients.26

Reports of wiring techniques for occipitocervical arthrodesis in 38 children with more than 2 years of follow-up have been reported.

Thirty-eight patients were treated with autograft and posterior wiring.

Thirty-four patients had bony union, three patients had fibrous union, and one patient had nonunion.

Ninety-seven percent of the patients (37 children) showed baseline or improved neurologic status at the most recent follow-up.

Complications

Superficial infection, postoperative pneumonia (1 patient), and pin tract infections from halo pins

In 11 patients (29%), we had a distal extension of the fusion mass, 7 patients had fusion at one additional level and 4 patients had fusion at two additional levels.

 

COMPLICATIONS

Graft or wire breakage Nonunion, insufficient fusion

Additional fusion levels and loss of motion Junctional instability distal to the fusion mass Infection

Deep wound infection Meningitis

Pin tract infections (halo)

 

Neurologic injury Donor site morbidity

 

 

 

ACKNOWLEDGMENTS

P.714

Thanks to John P. Dormans, Gokce Mik, and Purushottam A. Gholve, who authored this chapter in the first edition, which provided the basis for this revision.

 

REFERENCES

1. Bogduk N, Mercer S. Biomechanics of the cervical spine. I: normal kinematics. Clin Biomech 2000;15:633-648.

    

2. Brooks AL, Jenkins EB. Atlanto-axial arthrodesis by the wedge compression method. J Bone Joint Surg Am 1978;60:279-284.

    

3. Cohen MW, Drummond DS, Flynn JM, et al. A technique of occipitocervical arthrodesis in children using autologous rib grafts. Spine 2001;26:825-829.

    

4. Copley LA, Dormans JP, Pepe MD, et al. Accuracy and reliability of torque wrenches used for halo application in children. J Bone Joint Surg Am 2003;85-A(11):2199-2204.

    

5. Dormans JP, Criscitiello AA, Drummond DS, et al. Complications in children managed with immobilization in a halo vest. J Bone Joint Surg Am 1995;77:1370-1373.

    

6. Dormans JP, Drummond DS, Sutton LN, et al. Occipitocervical arthrodesis in children: a new technique and analysis of results. J Bone Joint Surg Am 1995;77:1234-1240.

    

7. Dormans JP, Wills B. Junctional instability and extension of fusion mass associated with posterior occipitocervical arthrodesis in children. Presented at POSNA 2004 Annual Meeting. St. Louis, MO, April 2004.

    

8. Doyle JS, Lauerman WC, Wood KB, et al. Complications and long-term outcome of upper cervical spine arthrodesis in patients with Down syndrome. Spine 1996;21:1223-1231.

    

9. Ebraheim NA, Lu J, Biyani A, et al. An anatomic study of the thickness of the occipital bone. Implications for occipitocervical instrumentation. Spine 1996;21:1725-1729; discussion 9-30.

    

10. Gallie WE. Skeletal traction in the treatment of fractures and dislocations of the cervical spine. Ann Surg 1937;106:770-776.

     

11. Ganey TM, Ogden JA. Development and maturation of the axial skeleton. In: Weinstein SL, ed. The Pediatric Spine, ed 2. Philadelphia: Lippincott Williams & Wilkins, 2001:3-54.

     

12. Goldstein R, Sunde C, Assad P, et al. Location of the Vertebral Artery at C1: How Far Out Laterally Can You Safely Dissect? In: POSNA Annual Meeting. Los Angeles, CA, 2013.

     

     

13. Hwang SW, Gressot LV, Rangel-Castilla L, et al. Outcomes of instrumented fusion in the pediatric cervical spine. J Neurosurg Spine 2012;17:397-409.

     

     

14. Locke GR, Gardner JI, Van Epps EF. Atlas-dens interval (ADI) in children: a survey based on 200 normal cervical spines. Am J Roentgenol Radium Ther Nucl Med 1966;97:135-140.

     

     

15. Mah JY, Thometz J, Emans J, et al. Threaded K-wire spinous process fixation of the axis for modified Gallie fusion in children and adolescents. J Pediatr Orthop 1989;9:675-679.

     

     

16. Melcher RP, Harms J. C1-C2 posterior screw rod fixation. In: Bradford DS, Zdeblick T, eds. Master Techniques in Orthopaedic Surgery: The Spine, ed 2. Philadelphia: Lippincott Williams & Wilkins, 2004: 129-145.

     

     

17. Nadim Y, Lu J, Sabry FF, et al. Occipital screws in occipitocervical fusion and their relation to the venous sinuses: an anatomic and radiographic study. Orthopedics 2000;23:717-719.

     

     

18. Ogden JA. Radiology of postnatal skeletal development. XI. The first cervical vertebra. Skeletal Radiol 1984;12:12-20.

     

     

19. Ogden JA. Radiology of postnatal skeletal development. XII. The second cervical vertebra. Skeletal Radiol 1984;12:169-177.

     

     

20. Puttlitz CM, Goel VK, Traynelis VC, et al. A finite element investigation of upper cervical instrumentation. Spine 2001;26:2449-2455.

     

     

21. Segal LS, Drummond DS, Zanotti RM, et al. Complications of posterior arthrodesis of the cervical spine in patients who have Down syndrome. J Bone Joint Surg Am 1991;73:1547-1554.

     

     

22. Steel HH. Anatomical and mechanical considerations of the atlantoaxial articulation. J Bone Joint Surg Am 1968;50:1481-1482.

     

     

23. Tracy MR, Dormans JP, Kusumi K. Klippel-Feil syndrome: clinical features and current understanding of etiology. Clin Orthop Relat Res 2004;(424):183-190.

     

     

24. Wills BP, Dormans JP. Nontraumatic upper cervical spine instability in children. J Am Acad Orthop Surg 2006;14:233-245.

     

     

25. Wills BP, Jencikova-Celerin L, Dormans JP. Cervical spine range of motion in children with posterior occipitocervical arthrodesis. J Pediatr Orthop 2006;26:753-757.

     

     

26. Wills BPD, Drummond DS, Schaffer A, et al. Posterior occipitocervical arthrodesis in children:

    intermediate and long-term outcomes. Presented at AAOS 2005 Annual Meeting, Washington, DC, February 2005.

     

     

27. Wilson BC, Jarvis BL, Haydon RC III. Nontraumatic subluxation of the atlantoaxial joint: Grisel's syndrome. Ann Otol Rhinol Laryngol 1987;96:705-708.