11: Posterior C1-C2 Fusion: Harms and Magerl Techniques

Published on 21/04/2015 by admin

Filed under Surgery

Last modified 22/04/2025

Print this page

This article have been viewed 6313 times

Procedure 11 Posterior C1-C2 Fusion

Harms and Magerl Techniques

Steven K. Leckie, Joseph M. Zavatsky, Ishaq Syed, Joon Y. Lee

Technique A: Posterior C1-2 Polyaxial Screw and Rod Fixation (Harms Technique) (Harms and Melcher, 2001)

Indications

FIGURE 11-1

FIGURE 11-2

Indications Pearls

• Similar risk of vertebral artery injury compared to transarticular screws (Yoshida et al, 2006).

• Does not require the use of sublaminar wires, thus decreasing the risk of neural injury.

• Screws can assist in the C1-2 reduction.

• Integrity of the posterior arch of C1 is not required.

• Can be incorporated as part of fusions to the occiput and/or the subaxial spine.

Indications Pitfalls

• Contraindicated for

• Comminuted fractures of the lateral mass of C1 or pedicle of C2.

• A large C2 transverse foramen obstructing the pedicle of C2.

• Anatomic limitations of the lateral mass of C1 could prevent the use of a 3.5-mm screw (Tan et al, 2003)

• Potential risk of irritation of the C2 ganglion causing occipital neuralgia

Examination/Imaging

Neurologic and musculoskeletal examination.

Preoperative imaging should include plain radiographs (Figure 11-3, A), computed tomography (CT) (Figure 11-3, B), CT angiography, and magnetic resonance imaging (MRI) (Figure 11-3, C) of the cervical spine.

• Radiographs should include anteroposterior (AP), lateral, and open mouth. Combined lateral mass displacement in excess of 7 mm or an atlantodens interval (ADI) greater than 3 mm suggests transverse ligament disruption.

• A CT scan with axial, sagittal, and coronal thin-cut (1-mm) reconstruction images through the upper cervical spine is an important part of preoperative planning. First, it provides accurate detail of the bony anatomy (associated ligamentous injury are often found on MRI). Second, it delineates the position of the foramen transversarium through which the vertebral artery runs. Third, it allows measurement of the length of the screws that will be utilized in the C1 lateral mass and C2 pars.

• Approximately 20% of patients requiring atlantoaxial fusion show anatomic variations in the path of the vertebral artery and osseous anatomy that would preclude screw placement (Jun, 1998; Madawi et al, 1997). In addition to evaluating vertebral artery dominance, CT angiography can delineate the spatial relationship of the vertebral artery relative to the C1 lateral mass and C2 pars.

• MRI allows enhanced visualization of any soft tissue injuries, including injury to the transverse atlantal ligament, as well as visualization of the spinal cord.

♦ Odontoid fractures with transverse atlantal ligament injury can be addressed with a posterior approach.

♦ MRI is also useful in rheumatoid patients to give a more accurate assessment of the space available for the cord, which can be underestimated in patients with soft tissue pannus that can be a source of cord compression not visualized on CT or a radiograph.

Noninvasive magnetic resonance angiography (MRA) can be utilized to evaluate vertebral artery injury, patency, and/or dominance (in lieu of CT angiography that requires administration of dye contrast).

FIGURE 11-3, A-C

Examination/Imaging Pearls

• Unless specifically indicated, the authors do not routinely use MRA to evaluate the vertebral artery anatomy because of its inability to define the spatial relationship between the artery and surrounding bony architecture.

Treatment Options

• Techniques for posterior C1-C2 fusion include

• Posterior C1-2 polyaxial screw and rod fixation (Harms technique)

• C2 translaminar screws

• C1-2 transarticular facet screws (Magerl technique)

• Gallie “bone block” graft placed posteriorly between the arches of C1 and C2 secured with sublaminar wire

• Brooks “wedge” bone graft secured to the posterior laminae with sublaminar wire

• Halifax interlaminar clamp

Surgical Anatomy

The posterior arch of C1 and the C1-2 facet joint are key anatomic landmarks for the placement of C1 lateral mass screws. The dorsal root ganglion of C2 lies just posterior to the starting point of the C1 screw and must be gently retracted caudally for adequate exposure (Figure 11-4, A). The starting point for the C1 screw is at the midpoint of the inferior portion of the C1 lateral mass at its junction with the posterior arch. The more superior and medial trajectory of the screws, when compared with transarticular screws, decreases the risk of vertebral artery injury (Figure 11-4, B and C)

The ponticulus posticus or congenital arcuate foramen is a common bony anomaly of the atlas (Young et al, 2005) (Figure 11-4, D). It is a bony arch on the cephalad aspect of the C1 lamina that contains the vertebral artery. If present, it can easily be confused with the lamina of C1 and must be identified during the posterior dissection and placement of C1 lateral mass screws to prevent vertebral artery injury.

FIGURE 11-4, A-D

Positioning

After an awake fiberoptic nasotracheal intubation is performed, a nasogastric tube is inserted for intraoperative gastric drainage.

If the patient is immobilized in a halo vest preoperatively, either the halo can be left in place and attached directly to the Mayfield headholder using an adapter or it can be removed. If the halo ring is removed, the patient is placed in Mayfield tongs and a hard cervical collar before being turned into the prone position. In coordination with anesthesia, the surgeon stands at the head of the hospital bed and stabilizes the patient’s neck. The patient is cautiously turned in the prone position on the operating table with the torso on bolsters or a four-poster frame. The Mayfield tongs or the halo ring is fixed to the operating table using a Mayfield headholder with the neck in a neutral position (Figure 11-5, A and B).

All bony prominences are well padded, and the patient’s arms are secured by their side using a folded sheet that is tucked beneath them.

Using fluoroscopic C-arm, proper alignment of the atlantoaxial bony structures is confirmed with the radiograph centered at C1-2. The lateral fluoroscopic image must not be oblique at C1-2; otherwise, malpositioning of the drill can result in erroneous screw placement (Figure 11-6).

• The appropriate C1 and C2 radiographic landmarks are visualized. The lateral wall of the C1 lateral mass and medial wall of the C2 pars interarticularis are important landmarks defined on the open-mouth view. The C-arm gantry is canted in the cephalad or caudad direction until all bony landmarks are clearly identified.

If necessary, adjustments can be made while the patient is in the Mayfield headholder, to obtain reduction. Reduction should be confirmed on fluoroscopic radiograph. If possible, extreme positions of the neck should be avoided.

Somatosensory evoked potential (SSEP) and transcranial motor evoked potential (MEP) monitoring are neurophysiologic spinal cord monitoring methods that can be utilized intraoperatively. Baseline readings can be obtained before and after placing the patient in the prone position.

FIGURE 11-5, A-B

FIGURE 11-6

Positioning Pearls

• An open-mouth view is obtained by placing an appropriate-size roll of sterile gauze in the patient’s mouth to facilitate a clear open-mouth view.

• In very osteopenic bone, inverse (negative) radiologic images can be utilized for better bony visualization.

• On the lateral C-arm image, the posterior occiput should be flexed off the posterior arch of C1 to facilitate screw placement at C1.

Positioning Equipment

• C-arm fluoroscopic radiograph should be positioned at the head of the operating table.

• Mayfield headholder

• Bolsters or a four-poster frame

Portals/Exposures

An electric razor is used to remove all hair from the patient’s occipital, suboccipital, and neck regions. If a definitive fusion is being performed, the posterior iliac crest is also shaved for bone graft harvesting.

The skin surfaces of the neck and posterior iliac crest are prepared and draped in a sterile fashion.

Using the inion of the occiput cranially, and the protuberance of the vertebral prominens caudally, the midline is identified and marked from the occiput to C3-4 with a sterile marker.

The subcutaneous skin of the planned skin incision can be infiltrated with 0.5% lidocaine containing epinephrine diluted 1:100,000.

A 10-blade scalpel is used to sharply incise the skin in the midline from the occiput to C3-4.

Bovie electrocautery is used for the subcutaneous dissection down to and through the underlying ligamentum nuchae. Midline dissection of the nuchal ligament provides a relatively avascular dissection and decreases the risk of injury to the greater and third occipital nerves. Self-retaining retractors are inserted for adequate visualization.

At the cephalad end of the incision, a 1.5-cm fascial cuff of trapezius, along the nuchal ridge, can be elevated to facilitate lateral exposure of C1-2, but this is not usually necessary. Subperiosteal dissection of the paraspinous muscular insertions from the suboccipital bone is completed.

The midline tubercle of the arch of C1 and the larger spinous process of C2 are used as palpable landmarks during dissection. Starting at the midline, the periosteum of C1 and the tip of the spinous processes of C2 and C3 are incised sharply.

Careful subperiosteal dissection is continued from C3 to C1, starting in the midline and proceeding laterally. Periosteal elevators can facilitate the subperiosteal dissection of the paraspinous muscles as they are swept laterally. The lateral masses and pedicles of C3 and C2 are exposed with care not to disturb the C2-3 facet capsules.

The C1-2 joint can be exposed with dissection over the superior surface of the C2 pars. Significant venous bleeding can be encountered with dissection around the venous plexus of the C2 nerve. This can effectively be controlled with bipolar electrocautery, thrombin-soaked Gelfoam, cotton pledgets, and various commercial gelatin thrombin preparations.

To decrease the risk of injuring the vertebral artery on the cephalic surface of the C1 lamina, identify the lamina and follow the caudal edge of the posterior arch during exposure of C1. If present, the ponticulus posticus or congenital arcuate foramen must be identified during the posterior dissection, because it can easily be confused with the lamina of C1 (Young et al, 2005).

The dissection is complete with exposure of the suboccipital rim of the foramen magnum.

Portals/Exposures Pearls

• The C2 spinous process is an easily identifiable landmark. The C2 spinous process sits more posterior relative to the arch of C1 and can be used to orient your dissection.

• The cephalad orientation of the C2 pars necessitates exposure down to C3. This facilitates the placement of the C2 pars screw.

• The lateral dissection should not be carried past the lateral border of the C1-2 articulation to avoid iatrogenic injury of the vertebral artery.

Portals/Exposures Pitfalls

• In the setting of a C1 fracture, when palpating the bony landmarks for dissection, take care to avoid pushing posterior fragments into the cord.

Procedure

Step 1

The dorsal root ganglion of C2 must be carefully retracted caudally to expose the starting point for the C1 lateral mass screw. The starting point for the C1 screw is at the midpoint of the inferior portion of the C1 lateral mass at its junction with the posterior arch.

C-arm imaging can be used to verify the midpoint and trajectory of the C1 lateral mass screw.

A 2-mm high-speed burr is used to mark the starting point for the drill and prevent the drill from walking off the convex surface of the posterior inferior lateral mass of C1.

With the tip of the drill pointing anterior through the lateral mass of C1, a 2-mm drill bit is used to drill a bicortical pilot hole in a straight to slightly convergent trajectory in the anteroposterior plane, and parallel to the posterior arch of C1 in the sagittal plane (Seal et al, 2009). Drill position is confirmed on AP and lateral C-arm fluoroscopic images (Figures 11-7 and 11-8).

A depth gauge can be used to confirm the measurement obtained from the preoperative CT scan of the appropriate length screw and can be checked on lateral fluoroscopic radiograph.

The drill hole is tapped and the 3.5-mm polyaxial screw is placed into the C1 lateral mass. An 8-mm unthreaded portion of the C1 polyaxial screw sits proud above the bony surface of the lateral mass, allowing the polyaxial portion of the screw to sit above the posterior arch of C1 so that the rod can be linked to the C2 screw head. The proud segment of the screw is unthreaded and theoretically minimizes the risk of irritation of the greater occipital nerve.

Step 1 is repeated for the contralateral C1 lateral mass.

FIGURE 11-7

FIGURE 11-8

Step 1 Instrumentation/Implantation

• Penfield elevators

• Blunt pedicle probe

• High-speed burr

• Pneumatic drill and 2-mm drill bit

Step 2

At C2, a no. 4 Penfield is used to define the medial border of the C2 pars. The starting point for the C2 pars interarticularis screw is in the superior and medial quadrant of the C2 pars. The entry point for placement of the C2 pars screw is marked with the 2-mm high-speed burr (Figure 11-12).

The starting hole and drill bit trajectory can be confirmed under C-arm guidance with open-mouth and lateral views.

A pilot hole is made using a 2-mm drill bit in a 20- to 30-degree convergent and cephalad trajectory, using the superior and medial aspect of the C2 pars as a guide. The integrity of the walls of the pilot drill hole is confirmed with a blunt pedicle probe.

A depth gauge can be used to confirm the measurement obtained from the preoperative CT scan of the appropriate length screw and can be checked on lateral fluoroscopic radiograph.

The drill hole is tapped, and the 3.5-mm polyaxial screw is placed into the C2 pars (Wait et al, 2009) (Figure 11-13).

Step 1 is repeated for the contralateral C2 pars.

FIGURE 11-12

FIGURE 11-13

Step 2 Pearls

• Intraoperative landmarks, the preoperative thin-cut (1-mm) axial CT scan, and lateral and open-mouth fluoroscopic imaging can all aid in the accurate placement of the C1 lateral mass and C2 pars interarticularis screws.

• Alternative procedures for patients with unilateral vertebral artery anomalies at C2

• Polyaxial screw placement at C1 and C3 on the side of the anomalous vertebral artery.

• Polyaxial screw placement at C1, C2, and C3 on the side with normal anatomy. Perform a posterior fusion from C1 to C3.

• If an anomalous vertebral artery exists, you can also place a crossed-intralaminar C2 screw if the lamina is wide enough on CT, as an alternative point of fixation.

Step 2 Instrumentation/Implantation

• Penfield elevators

• Blunt pedicle probe

• High-speed burr

• Pneumatic drill and 2-mm drill bit

Step 3

If reduction of C1 is necessary, the patient’s head can be repositioned before fixation of the rods to the screws. When performed, the reduction is visualized under fluoroscopy.

Step 3 Pearls

• C1-2 reduction can be accomplished by direct manipulation of the C1 and C2 screws. The authors’ recommendation is to obtain a reduction preoperatively, if possible while the patient is awake, to assess neurologic status. Alternatively, reduction can be performed after the patient is positioned prone on the operating table before preparation and draping.

Step 4

The interconnecting rods are measured and secured with locking nuts (Figure 11-14).

Distraction or compression of the construct can be accomplished at this time.

The locking nuts are tightened with a torque wrench (Figure 11-15).

FIGURE 11-14

FIGURE 11-15

Step 4 Pearls

• With this technique, one can avoid damage to the C1-2 facet joints, and the rods and screws can be used as temporary fracture fixation without definitive fusion (i.e., type II and III odontoid fractures) (Harms and Melcher, 2001). Eventual removal of the hardware can allow the patient to regain atlantoaxial motion after fracture healing has occurred.

• The integrity of the posterior arch of C1 is not necessary for stable fixation.

• Patients with rheumatoid arthritis often have instability adjacent to the atlantoaxial region requiring a more extensive fusion. This technique can be incorporated as part of fusions to the occiput and/or the subaxial spine.

Step 4 Instrumentation/Implantation

• Polyaxial screw and rod system, reduction tools

Step 5

For definitive fusion, posterior iliac crest bone graft is harvested. The posterior superior iliac crest is palpated, and an 8-cm line centered over the crest is marked with a sterile marking pen. A 10-blade scalpel is used to incise the skin. Self-retaining retractors are inserted.

Bovie electrocautery is used to dissect the subcutaneous tissue down to the junction of the lumbodorsal and gluteus maximus fascia.

The posterior superior iliac crest is palpated, and the fascia is initially reflected using Bovie electrocautery. A subperiosteal dissection is performed using a Cobb elevator at the superior and lateral margins of the crest. Osteotomes are used to make a cortical window in the crest that is reflected medially. The window is made within 8 cm of the posterior superior iliac crest to avoid injury to the superior cluneal nerves.

Small gouges are used to harvest cancellous bone graft through the cortical window.

The graft is placed in a sterile cup mixed with the patient’s blood and covered.

Hemostasis is obtained with bone wax. The wound is irrigated and the graft site is packed with Marcaine-soaked Gelfoam, and the cortical window is closed.

The retractors are removed and the incision is closed in layers.

Step 5 Instrumentation/Implantation

• Cobb elevators

Step 6

The cervical operative field is irrigated with antibiotic solution, and the irrigation fluid is suctioned from the field.

If definitive fusion is being performed, the posterior surfaces of C1 and C2 are decorticated with the high-speed burr. Alternatively, decortication of the posterior aspects of C1 and C2 can be performed before insertion of the screws and rods.

The cancellous bone graft is placed over the decorticated surfaces of C1 and C2.

The C1-2 joint surfaces can also be decorticated and packed with bone graft for an intraarticular fusion.

Alternatively, a piece of iliac crest bone graft can be fashioned and secured between the C1 ring and C2 spinous process.

Postoperative Care and Expected Outcomes

The patient should be taken to the recovery room or the surgical intensive care unit (SICU) for postoperative recovery.

Supine and upright lateral cervical radiographs should be obtained in the cervical collar to assess stability on postoperative day 1. If atlantoaxial stability has been obtained, the patient can be mobilized (see Figure 11-16).

On postoperative day 1, or when medically stable, the patient can be transferred to a standard surgical floor.

A postoperative CT scan can be obtained if there is any question concerning screw placement.

The patient can be discharged from the hospital when medically stable.

Rigid cervical collar immobilization is used postoperatively.

Routine outpatient static lateral and supervised dynamic lateral flexion and extension radiographs can be obtained at approximately 4 weeks to ascertain stability. If stability is obtained, the cervical collar can be weaned from use. Radiographs are obtained at 4- to 6-week intervals to assess stability and fusion.

Additionally, a CT scan can be obtained 3 to 6 months postoperatively to assess fusion and fracture healing.

Postoperative Pitfalls

• Screw malposition can result in

• Inadequate purchase, resulting in a potentially unstable construct. Rigid external immobilization in a halo vest for 10 to 12 weeks should provide enough atlantoaxial stability to achieve fusion if instability is present.

• Dural tear and cerebrospinal fluid (CSF) leak. The authors advocate primary repair of all dural tears, as well as placement of Gelfoam over the repair.

• Violation of the foramen transversarium can result in vertebral artery rupture, dissection, pseudoaneurysm, or occlusion. Even without direct violation of the arterial wall, the screw threads can contact the artery and result in injury because of normal pulsatile flow. If this is discovered intraoperatively or on postoperative imaging, the screw should be surgically removed to decrease the risk of vertebral artery injury.

• Intraoperative vertebral artery injury is the feared complication that can cause serious clinical sequela, including brainstem stroke. If this occurs, the screw can be left in place to tamponade bleeding, and then vascular studies should be obtained.

• A direct microvascular repair by a vascular surgeon after the surrounding bone is skeletonized or after endovascular repair are also options. Irrespective of the method used to tamponade the bleeding, a postoperative angiogram should be obtained to evaluate the vertebral artery.

• Potential risk for internal carotid artery injury with bicortical screw fixation (Currier et al, 2008) anterior to the ring of C1.

Technique B: C1-2 Transarticular Facet Screws (Magerl Technique)

Indications

Indications Pearls

• Biomechanically superior to posterior wiring techniques (Henriques et al, 2000)

• Postoperative halo-vest immobilization not usually required

Indications Pitfalls

• Potential vertebral artery injury

• Technically demanding

• Maximum biomechanical stability achieved when combined with posterior wiring for three-point fixation (Henriques et al, 2000)

• Requires intact posterior arches of C1 and C2

• Sublaminar wires increases risk of neurologic injury

• Contraindicated for

• Comminuted fractures or erosive lesions of the lateral mass of C1

• Bilateral anomalous vertebral artery pathways through the C1 lateral mass

• Inadequate reduction and alignment of the atlantoaxial complex

• Relative contraindication

• A unilateral anomalous vertebral artery pathway through the lateral mass of C1 or C2 on one side, with a concomitant contralateral dominant vertebral artery. If injury occurs to the dominant artery with screw placement, blood supply to the brainstem and the cervicomedullary junction can be compromised.

Examination and Imaging

Complete neurologic and musculoskeletal examination

Preoperative imaging should include plain radiographs (see Figure 11-3, A), CT (see Figure 11-3, B), CT angiography, and MRI (see Figure 11-3, C) of the cervical spine.

• Radiographs should include AP, lateral, and open mouth to evaluate the alignment of C1 and C2. Supervised dynamic lateral flexion and extension views helps determine the reducibility of atlantoaxial subluxation when present.

• A CT scan with axial, sagittal, and coronal thin-cut (1-mm) reconstruction images through the upper cervical spine is an important part of preoperative planning. A CT can provide anatomic detail of the lateral mass of C1, and the C2 facet for optimal placement of transarticular facet screws. It delineates the position of the foramen transversarium through which the vertebral artery runs. Additionally, it allows for the evaluation of adequate bone stock of the atlas for sufficient fixation of the screws.

• Approximately 20% of patients requiring atlantoaxial fusion show anatomic variations in the path of the vertebral artery and osseous anatomy, which would preclude screw placement (Jun, 1998; Madawi et al, 1997). In addition to evaluating vertebral artery dominance, CT angiography can delineate the course of the vertebral artery through the foramen transversarium and its spatial relationship to the surrounding bony architecture. This can help determine if a screw can be placed safely with minimal risk to the vertebral artery.

• MRI permits visualization of any soft tissue injuries including the transverse atlantal ligament, as well as visualization of the spinal cord.

♦ MRI can help identify injury to the transverse atlantal ligament with concomitant odontoid fracture, which may help dictate anterior versus posterior operative approach. Even with union, anterior odontoid screw fixation alone will not restore atlantoaxial stability secondary to ligamentous disruption.

♦ When present in rheumatoid patients, identification of pannus posterior to the odontoid can give a more accurate measurement for the space available for the cord.

Noninvasive magnetic resonance angiography (MRA) can be utilized to evaluate vertebral artery injury, patency, and/or dominance.

Examination/Imaging Pearls

• Unless specifically indicated, the authors do not routinely use MRA to evaluate the vertebral artery anatomy because of its inability to define the spatial relationship between the artery and surrounding bony architecture.

Treatment Options

• Techniques for posterior C1-C2 fusion include

• C1-2 transarticular facet screws (Magerl technique) with and without bone graft and sublaminar wiring

• Posterior C1-2 polyaxial screw and rod fixation (Harms technique)

• C2 translaminar screw

• Gallie “bone block” graft placed posteriorly between the arches of C1 and C2 secured with sublaminar wire

• Brooks “wedge” bone graft secured to the posterior lamina with sublaminar wire

• Halifax interlaminar clamp

Surgical Anatomy

The cephalad orientation of the C1-2 transarticular screw and the final position of the neck necessary for adequate atlantoaxial alignment may require percutaneous placement of the transarticular facet screws (Figure 11-17).

The ponticulus posticus or congenital arcuate foramen is a common bony anomaly of the atlas (Young et al, 2005). It is a bony arch on the cephalad aspect of the C1 lamina that contains the vertebral artery. If present, it can easily be confused with the lamina of C1 and must be identified during the posterior dissection to prevent vertebral artery injury.

The gray ramus communicans of the C2 nerve is a reliable landmark for locating the entry point for a screw on the C2 pars (Cavalcanti et al, 2010).

FIGURE 11-17

Positioning

After an awake fiberoptic nasotracheal intubation is performed, a nasogastric tube is inserted for intraoperative gastric drainage.

If the patient is immobilized in a halo vest, a rigid Philadelphia collar is placed on the patient before halo-vest removal. In coordination with anesthesia, the surgeon stands at the head of the hospital bed and stabilizes the patient’s neck. The patient is cautiously repositioned in the prone position on the operating table with the torso on bolsters or a four-poster frame. The halo ring can be attached to the Mayfield adapter. If the halo ring is removed, the patient is placed in Mayfield tongs before prone positioning. After repositioning the patient prone, the Mayfield tongs are fixed to the operating table using a Mayfield headholder with the neck in a neutral position.

All bony prominences are well padded, and the patient’s arms are secured by their side using a folded sheet that is tucked beneath them.

Using fluoroscopic C-arm, proper alignment of the atlantoaxial bony structures is confirmed with the radiograph centered at C1-2. The lateral fluoroscopic image must not be oblique at C1-2; otherwise incorrect screw trajectory can result.

If necessary, adjustments can be made while the patient is in the Mayfield headholder to obtain reduction and should be confirmed on fluoroscopic radiograph. If possible, extreme positions of the neck should be avoided.

Somatosensory evoked potential and transcranial motor evoked potential monitoring are neurophysiologic spinal cord monitoring methods that can be utilized intraoperatively. Baseline readings can be obtained before and after placing the patient in the prone position.

Positioning Pearls

• In very osteopenic bone, inverse (negative) radiologic images can be utilized for better bony visualization.

Positioning Pearls

• It is imperative that you confirm that the C1-2 joint is reduced or capable of reduction preoperatively; otherwise the accurate placement of the transarticular screws can be difficult and dangerous.

Positioning Equipment

• C-arm fluoroscopic radiograph should be positioned at the head of the operating table.

• Mayfield headholder

• Four-poster frame

Portals/Exposures

An electric razor is used to remove all hair from the patient’s occipital, suboccipital, and neck regions. If transarticular screw fixation with bone graft and sublaminar wiring is being performed, the posterior iliac crest is also shaved for bone graft harvesting.

The skin surfaces of the neck and posterior iliac crest are prepared and draped in sterile fashion.

Using the inion of the occiput cranially, and the protuberance of the vertebral prominens caudally, the midline is identified and marked from the occiput to C3-4 with a sterile marker.

The subcutaneous skin of the planned skin incision can be infiltrated with 0.5% lidocaine containing epinephrine diluted 1:100,000.

A 10-blade scalpel is used to sharply incise the skin in the midline from the occiput to C3-4.

Bovie electrocautery is used for the subcutaneous dissection down to and through the underlying ligamentum nuchae. Midline dissection of the nuchal ligament provides a relatively avascular dissection and decreases the risk of injury to the greater and third occipital nerves. Self-retaining retractors are inserted for adequate visualization.

At the cephalad end of the incision, a 1.5-cm fascial cuff of trapezius, along the nuchal ridge, can be elevated to facilitate lateral exposure of C1-2, but this is not usually necessary. Subperiosteal dissection of the paraspinous muscular insertions from the suboccipital bone is completed.

The midline tubercle of the arch of C1 and the larger spinous process of C2 are used as palpable landmarks during dissection. Starting at the midline, the periosteum of C1 and the tip of the spinous processes of C2 and C3 are incised sharply.

Careful subperiosteal dissection is continued from C3 to C1, starting in the midline and proceeding laterally. Periosteal elevators can facilitate the subperiosteal dissection of the paraspinous muscles as they are swept laterally. C2 and C3 are exposed laterally with care not to disturb the C2-3 facet capsules.

The C1-2 joint can be exposed with dissection over the superior surface of the C2 pars. The capsule of the C1-2 facet is reflected from caudal to cephalad, using caution not to injure the C2 nerve and surrounding vessels. Significant venous bleeding can be encountered with dissection around the venous plexus of the C2 nerve. This can effectively be controlled with bipolar electrocautery, thrombin-soaked Gelfoam, and cotton pledgets.

To decrease the risk of injuring the vertebral artery on the cephalic surface of the C1 lamina, identify the lamina and follow the caudal edge of the posterior arch during exposure of C1.

The dissection is complete with exposure of the suboccipital rim of the foramen magnum.

Portals/Exposures Pearls

• The C2 spinous process is an easily identifiable landmark. The C2 spinous process sits more posterior relative to the arch of C1 and can be used to orient your dissection.

• If present, the ponticulus posticus or congenital arcuate foramen can easily be confused with the lamina of C1. It must be identified during the posterior dissection to prevent vertebral artery injury.

• The lateral dissection should not be carried past the lateral border of the C1-2 articulation, to avoid iatrogenic injury of the vertebral artery.

Portals/Exposures Pitfalls

• When palpating the bony landmarks for dissection, take care to avoid translocation of the atlas and its potentially serious consequences.

Procedure

Step 1

If posterior bone graft and sublaminar wiring are going to be utilized with transarticular screw fixation, bone graft harvesting and passage of the C1 sublaminar wire should be completed before the insertion of the transarticular screws.

After the dissection, the C1-2 posterior arch interspace and graft size is approximated. A moist sponge is placed in the wound to prevent tissue desiccation while harvesting the bone graft.

The posterior superior iliac crest is palpated, and an 8-cm line centered over the posterior superior iliac crest is marked with a sterile marking pen. A 10-blade scalpel is used to incise the skin. Self-retaining retractors are inserted.

Bovie electrocautery is used to dissect the subcutaneous tissue down to the junction of the lumbodorsal and gluteus maximus fascia.

The posterior superior iliac crest is palpated, and the fascia is initially reflected using Bovie electrocautery. A subperiosteal dissection is performed using a Cobb elevator at the superior and lateral margins of the crest. Osteotomes and an oscillating saw are used to obtain a tricortical strut graft within 8 cm of the posterior superior iliac crest to avoid injury to the superior cluneal nerves.

A 1.5 × 4 cm tricortical strut graft is obtained. Small gouges/curettes are used to then harvest additional cancellous bone graft.

Hemostasis is obtained with bone wax. The wound is irrigated, and the graft site is packed with thrombin-soaked Gelfoam. The retractors are removed, the wound is irrigated, and the incision is closed in layers.

Any soft tissue on the graft is removed with a small Cobb or curette. The graft is placed in a sterile cup mixed with the patient’s blood and covered.

Step 1 Pearls

• Stabilization of the C1-2 segment with transarticular screws can make the passage of the sublaminar wire more difficult and dangerous. Therefore it is the authors’ preference to pass the sublaminar wire before screw fixation.

• The C1 sublaminar wire can aid in reduction of the atlantoaxial segment, thus facilitating placement of the transarticular screws.

Step 1 Instrumentation/Implantation

• Cobb elevators

• Oscillating saw

Step 3

A no. 4 Penfield elevator is used to define the medial border of the C2 pars. Care must be taken not to violate the dura with the Penfield. The Penfield can be used to protect the C2 nerve root and venous plexus with gentle rostral retraction. The removal of any ligamentum flavum adjacent to the C2 lamina will improve visualization of the C2 pars and atlantoaxial joint and help with orientation for transarticular screw placement.

Step 3 is repeated for the contralateral C2 pars.

Step 3 Instrumentation/Implantation

• Penfield elevators

Step 4

Using the Gallie technique, a C1 sublaminar wire is passed before the placement of the transarticular screws. A smooth arch is bent into the end of 16- or 18-gauge double-looped wire and conformed to approximate both the length and thickness of the C1 lamina.

The wire is carefully passed beneath the C1 lamina. This maneuver can be facilitated by the use of 00-silk suture or a wire-passer. Using both hands, the wire is pulled beneath the C1 lamina. One hand pulls and advances the sublaminar wire, while the other hand provides constant gentle tension on the other end of the wire. This keeps the wire flush against the underside of the lamina, decreasing the risk of impinging the wire upon the cord.

Step 4 Instrumentation/Implantation

• 16- or 18-gauge double-looped wire

• A wire-passer or 00-silk suture

Step 5

Atlantoaxial alignment is necessary for accurate placement of C1-2 transarticular screws. This is confirmed using lateral C-arm fluoroscopy (see Figure 11-17).

If necessary, anterior atlantoaxial subluxation can be manually realigned intraoperatively with gentle traction on the C1 sublaminar wire and pressure on the axis.

Posterior atlantoaxial subluxation can be realigned with flexion repositioning of the head in the Mayfield tongs. Alternatively, posterior atlantoaxial subluxation can be reduced with the placement of the bone block graft between C1 and C2. After placement, the graft can be manually manipulated against the posterior arch of C1 to obtain reduction. Reduction can be maintained after the C1 sublaminar wire is secured around the spinous process of C2 and the bone block graft.

The cephalad angle of the C1-2 transarticular screw can make placement difficult. To facilitate screw placement the axis can be gently displaced rostrally toward the occiput. This presents the C2 pars at a better angle for accurate placement of the transarticular screws.

The position of the neck can also influence the trajectory necessary to obtain optimal placement of the transarticular screws. Percutaneous placement of the C1-2 facet screws may be necessary if intraoperative atlantoaxial alignment precludes the placement of the screws through the wound created by the posterior cervical dissection.

Step 6

If the trajectory requires a percutaneous approach, a stab incision is made with a 15-blade scalpel through the skin into the paraspinal region of T2-3.

A pointer tunneler is placed into the soft tissue protection sheath and inserted through the stab incision made at T2-3. The soft tissue protection sheath has a large handle that can be used to manipulate and guide your trajectory through the soft tissues.

In the approximate trajectory necessary for transarticular screw placement, a tunnel can carefully be made through the soft tissues of the neck with the instruments exiting at the inferior aspect of the occipitocervical wound.

The tip of the tissue sheath is placed against the inferior articular process of C2, and the pointed tunneler is removed from the tissue sheath.

Step 6 Instrumentation/Implantation

• Pointer tunneler

• Soft tissue protection sheath

Step 7

The K-wire drill guide is inserted into the soft tissue sheath. A 1.2-mm diameter K-wire is then inserted into the drill guide and secured to a reversible pneumatic drill.

If the percutaneous technique is not required, the K-wire drill guide can be inserted directly into the soft tissue sheath and placed in the C2 fossa.

The starting point for screw entry is in the C2 fossa, 2 to 3 mm lateral to the junction of the lamina and lateral mass (Figures 11-18 and 11-19).

C-arm fluoroscopy is used to identify all bony atlantoaxial structures and guide the trajectory of the 1.2-mm K-wire. The K-wire is aimed toward the superior border of the lateral mass of C1, just lateral to the C2 pars interarticularis, and parallel to the medial border of the C2 pars interarticularis.

The K-wire is advanced through the pars interarticularis of C2, along the central axis of the C2 pedicle in a 0- to 10-degree convergent trajectory in the AP plane.

A no. 4 Penfield can be used to protect the C2 nerve root and venous plexus with gentle rostral retraction. This allows visualization of the K-wire as it passes across the posterior edge of the atlantoaxial facet joint into the lateral mass of C1. The tip of the K-wire should engage the anterosuperior margin of the C1 lateral mass.

C-arm fluoroscopy is used to confirm K-wire placement.

FIGURE 11-18

FIGURE 11-19

Step 7 Pearls

• To decrease the risk of vertebral artery injury, the K-wire must not be directed laterally, and must follow the medial wall of the C2 pars.

• Intraoperative bony landmarks, the preoperative thin-cut (1-mm) axial CT scan, and AP and lateral fluoroscopic imaging can all aid in the accurate placement of the K-wires and C1-2 transarticular screws (Weidner et al, 2000).

• Screw fixation with a cannulated 3.5-mm screw system is the authors’ preferred method for transarticular screw placement.

Step 7 Instrumentation/Implantation

• 1.2 × 50 mm K-wires

• K-wire drill guide

• Reversible pneumatic drill

• Penfield elevators

Step 8

A cannulated drill bit is placed and drilled over the K-wire under C-arm fluoroscopy. Care must be taken as the drill is advanced over the K-wire. Because there is a risk of bending the tip of the K-wire as it crosses the C1-2 facet joint and penetrates the C1 cortical surface, impingement of the K-wire in the cannulated drill can occur. This can result in subsequent advancement of the K-wire into the posterior oropharyngeal fossa.

The drill guide is removed from the tissue sheath and a second K-wire of identical length is placed in the tissue sheath until it rests on the posterior surface of C2 adjacent to the other K-wire. The difference in length of the K-wires determines the length of the screw, and it is measured with a ruler.

The near cortex is tapped with a cannulated tap through the tissue sheath.

A cannulated fully threaded 3.5-mm cortical screw of appropriate length (usually 40 to 45 mm) is inserted with a cannulated screwdriver over the K-wire under fluoroscopic visualization. The anterior cortex of the C1 lateral mass should be purchased with the screw. Unicortical screws may be considered to avoid neurovascular injury in cases with satisfactory bone quality (Cyr et al, 2008) (Figure 11-20).

The K-wire is removed and screw placement can be confirmed with C-arm fluoroscopic imaging.

Steps 5 to 7 are repeated for the contralateral C1-2 transarticular screw.

FIGURE 11-20

Step 8 Instrumentation/Implantation

• 1.2 × 50 mm K-wire

• Cannulated drill bit

• Cannulated tap

• Cannulated fully threaded 3.5-mm cortical screws (Figure 11-21)

FIGURE 11-21

Step 9

The future areas of contact of the posterior arches of C1 and C2 with the strut graft are decorticated with the high-speed burr. The inferior cortical margin of the posterior arch of C1 and superior cortical margin of the posterior arch and spinous process of C2 are decorticated with the burr.

A notch is made on the inferior margin of the spinolaminar junctions of C2 with an angled Kerrison rongeur for future seating of the sublaminar wire.

Step 9 Instrumentation/Implantation

• High-speed burr

• Kerrison rongeur

Step 10

Using a Leksell rongeur, the tricortical strut graft is converted into a bicortical graft by removing the rounded cortical edge.

The graft is placed between the posterior arches of C1 and C2 for final sizing and to approximate the midline of the graft. The Leksell rongeur is used to make any final modifications for optimal sizing of the graft.

A Kerrison rongeur is used to notch the midline on the inferior aspect of the graft to accommodate the spinous process of C2 for a secure fit.

Step 10 Instrumentation/Implantation

• Leksell rongeur

• Kerrison rongeur

Step 12

The posterior cortical surfaces of the C1 posterior arch, C2 lamina, and strut graft are decorticated with the high-speed burr.

The bone dust and shavings from the burr are left in the operative field, and the cancellous bone graft is placed over the posterior decorticated surfaces of C1, C2, and strut graft.

Step 12 Instrumentation/Implantation

• High-speed burr

Step 13

The wound is closed securely in a layered fashion, obliterating any dead space.

Steri-Strips are placed perpendicular to the incision and covered by sterile 4 × 4 gauze and a clear Tegaderm dressing.

Final AP and lateral cervical radiographs are obtained to assess placement of the hardware and alignment of the atlantoaxial region (Figure 11-23).

A rigid cervical collar (i.e., Philadelphia or Miami-J) is secured in place.

The patient is removed from the Mayfield headholder. The surgeon stands at the head of the operating table and is responsible for stabilizing the neck when the patient is repositioned onto the hospital bed in the supine position.

FIGURE 11-23

Postoperative Care and Expected Outcomes

The patient should be taken to the recovery room or the surgical intensive care unit (SICU) for postoperative recovery.

Supine and upright lateral cervical radiographs should be obtained in the cervical collar to assess stability on postoperative day one. If atlantoaxial stability has been obtained, the patient can be mobilized.

On postoperative day 1, or when medically stable, the patient can be transferred to a standard surgical floor.

A postoperative CT scan can be obtained if there is any question concerning screw placement.

The patient can be discharged from the hospital when medically stable.

Rigid cervical collar immobilization is used for approximately 8 to 12 weeks postoperatively.

Routine outpatient static lateral radiographs can be obtained 4 weeks postoperatively to ascertain stability. Radiographs can be obtained at 4-week intervals to assess stability and fusion. Supervised dynamic lateral flexion and extension radiographs can be obtained at the end of the 8- to 12-week period to further assess atlantoaxial stability. If stability is obtained, the cervical collar can be weaned from use.

Additionally, a CT scan can be obtained 3 to 6 months postoperatively to assess fusion and fracture healing.

Fusion can be achieved in nearly 100% of cases, although 16.7% of patients may experience complications (Finn and Apfelbaum, 2010).

Postoperative Pitfalls

• Screw malposition can result in:

• Inadequate purchase, resulting in a potentially unstable construct. Rigid external immobilization in a halo vest for 10 to 12 weeks should provide enough atlantoaxial stability to achieve fusion if instability is present.

• Dural tear and CSF leak. The authors advocate the primary repair of all dural tears, as well as placement of Gelfoam over the repair.

• Violation of the foramen transversarium can result in vertebral artery rupture, dissection, pseudoaneurysm, or occlusion. Even without direct violation of the arterial wall, the screw threads can contact the artery and result in injury because of normal pulsatile flow. If this is discovered intraoperatively or postoperatively, the screw should be left in place to tamponade bleeding, and then vascular studies should be obtained.

• Intraoperative vertebral artery injury is the most feared complication that can cause serious clinical sequelae, including brainstem stroke. If this occurs, the screw should be removed immediately, and the site should be packed with pieces of Gelfoam large enough not to be a source of emboli. Bone wax can also be used to help tamponade the bleeding. A direct microvascular repair by a vascular surgeon, after the surrounding bone is skeletonized, is also an option. Irrespective of the method used to tamponade the bleeding, a postoperative angiogram should be obtained to evaluate the vertebral artery.

Cavalcanti D, Agrawal A, Garcia-Gonzalez U, et al. Anterolateral C1-C2 transarticular fixation for atlantoaxial arthrodesis: landmarks, working area, and angles of approach. Operative Neurosurg. 2010;67:38-42.

Five cadaver necks were dissected bilaterally to study anatomic landmarks, and then 10 CT scans were analyzed to quantify working area and optimal angles of approach. The C2 transverse process was a landmark for dissecting posterior to the carotid sheath, and gray ramus communicans from the superior cervical ganglion to the C2 nerve was a landmark for locating the C2 pars. The mean working area was 71.2 mm², and the ideal angle for screw placement was 22.9 degrees medial to the sagittal plane and 25.3 degrees posterior to the coronal plane.

Currier B, Maus T, Eck J, et al. Relationship of the internal carotid artery to the anterior aspect of the C1 vertebra. Spine. 2008;33:635-639.

The authors retrospectively reviewed 50 head and neck CT scans, performed with contrast, to study the relationship between the anterior aspect of the C1 vertebra and the internal carotid artery. The mean shortest distance between the artery and C1 was 2.88 mm on the left and 2.89 mm on the right. The lumen of the artery was medial to the foramen transversarium in 84% of cases. The authors conclude that the proximity of C1 to the internal carotid artery poses moderate risk in 46% of cases and high risk in 12% of cases on at least one side. They therefore recommend preoperative contrast imaging in all cases in which a screw is to be placed in C1. If the artery is in close proximity to the anterior border of C1, unicortical fixation or a different fusion technique should be considered.

Cyr S, Currier B, Eck J, et al. Fixation strength of unicortical versus bicoritcal C1-C2 transarticular screws. Spine J. 2008;8:661-665.

The internal carotid artery and hypoglossal nerve lie in close proximity to the anterior aspect of C1. The authors performed a biomechanical study of pullout strength in 15 cadaver specimens. They found no statistically significant difference in pullout strength between unicortical and bicortical C1-2 transarticular screws. In cases with adequate bone stock, the authors recommend unicortical screws to avoid neurovascular injury.

Finn M, Apfelbaum R. Atlantoaxial transarticular screw fixation: update on techniques and outcomes in 269 patients. Neurosurgery. 2010;66A:184-192.

The authors retrospectively reviewed 269 patients who underwent transarticular screw fixation for a mean follow-up of 15.7 months. Fusion was achieved in 99% of cases. Complications occurred in 16.7% of cases (including five vertebral artery injuries, one of which was bilateral and fatal). The technique could not be applied in 13.3% of cases because of anatomic constraints.

Harms J, Melcher R. Posterior C1-C2 fusion with polyaxial screw and rod fixation. Spine. 2001;26:2467-2471.

The authors describe bilateral insertion of 3.5-mm polyaxial screws into the lateral masses of C1 and the pars of C2, followed by reduction (if needed) and fixation with 3-mm rods. Unlike transarticular screw and posterior wiring techniques, this does not rely on an intact posterior arch, decreases the risk of vertebral artery injury, and can be used to correct fixed C1-2 subluxation. Because the facet joint surfaces remain intact, the patient can regain motion after removal of hardware if indicated. The authors describe successful fusion of 37 patients without neural or vascular injury.

Henriques T, Cunningham B, Olerud C, et al. Biomechanical comparison of five different atlantoaxial posterior fixation techniques. Spine. 2000;25:2877-2883.

Eight cadaver spines were loaded in 3 degrees of freedom after instrumentation with bilateral transarticular screws (1), posterior wiring (2), both (2), or control (3). The authors found that three-point fixation (bilateral transarticular screws in combination with posterior wiring) provides superior durability when biomechanically loaded.

Jeanneret B, Magerl F. Primary posterior fusion C1/2 in odontoid fractures: indications, techniques, and results of transarticular screw fixation. J Spinal Disord. 1992;5:464-475.

The authors present 12 acute odontoid fractures that were fixed with transarticular screws. At follow-up, all were united and had maintained reduction. The authors discuss the unstable fracture patterns that are more appropriately treated with posterior fixation rather than anterior screw fixation.

Jun BY. Anatomic study for ideal and safe posterior C1-C2 transarticular screw fixation. Spine. 1998;23:1703-1707.

Reconstructed CT scans of 64 normal cervical spines were digitally implanted with transarticular screws in multiple trajectories and analyzed with navigation software. One patient had inadequate space available for the screw because of the location of the transverse foramen and vertebral artery, and four others had nearly insufficient space. Lateral fluoroscopy is imperative for safe screw insertion.

Madawi A, Solanki G, Casey AT, et al. Variation of the groove in the axis vertebra for the vertebral artery: implications for instrumentation. J Bone Joint Surg Br. 1997;79:820-823.

The groove for the vertebral artery in C2 was investigated for 50 dry cadaver specimens. The authors found that 11 specimens had either a pedicle width or lateral mass height of less than 2 mm, which would put the vertebral artery at risk or provide inadequate bone stock for transarticular C1-C2 fixation. The authors contend that fine-cut CT scans are crucial for preoperative planning.

Magerl F, Seemann P-S. Stable posterior fusion of the atlas and axis by transarticular screw fixation. In: Kehr P, Weidner A, editors. Cervical Spine I. New York: Springer Wien; 1986:322-327.

Seal C, Zarro C, Gelb D, et al. C1 lateral mass anatomy: proper placement of lateral mass screws. J Spinal Disord Tech. 2009;22:516-523.

The authors dissected 15 cadaver spines to study C1 anatomy with caliper measurements followed by CT scans. Additional specimens were instrumented, and guidelines were established for C1 screw fixation. Fifty clinical cases were retrospectively reviewed. The authors concluded that 10 degrees medial and 22 degrees cephalad was the preferred trajectory for C1 screw fixation.

Tan M, Wang H, Wang Y, et al. Morphometric evaluation of screw fixation in atlas via posterior arch and lateral mass. Spine. 2003;28:888-895.

Fifty atlas specimens were studied with calipers, protractors, and CT to determine the optimal size and trajectory for transarticular screw fixation, and parameters were applied to 5 patients without incident. The longest trajectory distance of the screw path was about 30 mm. The outer thickness at the thinnest part of groove was 4.58 mm, and it was found to be less than 4 mm in four cases (8%). The entry point is 18 to 20 mm lateral to the midline and 2 mm superior to the inferior border of posterior arch. The direction of screw placement is perpendicular to the coronal plane and about 5 degrees cephalad to the transverse plane.

Wait S, Ponce F, Colle K, et al. Importance of the C1 anterior tubercle depth and lateral mass geometry when placing C1 lateral mass screws. Neurosurgery. 2009;65:952-957.

The authors reviewed 100 consecutive cervical CT scans. The mean depth of the C1 tubercle was 6.9 mm (range 2.7 to 11.2 mm). Preoperative planning and lateral fluoroscopy are essential to guide the depth of C1 lateral screw placement.

Weidner A, Wahler M, Chiu T, Ullrich C. Modification of C1-C2 transarticular screw fixation by image-guided surgery. Spine. 2000;25:2668-2674.

CT scan data for 37 prospectively assigned patients was uploaded into a surgical planning computer program that generated an optimal screw trajectory. The surgical field was matched to the virtual computer field, and C2 was drilled according to plan. The historical control group included retrospective analysis of 78 patients who had a similar surgery performed under fluoroscopic guidance. Image-guided surgery reduced, but did not eliminate, the risk of screw misplacement. Surgical time was not increased.

Yoshida M, Feo M, Fujibayashi S, Nakamura T. Comparison of the anatomical risk for vertebral artery injury associated with the C2-pars interarticularis screw and atlantoaxial transarticular screw. Spine. 2006;31:E513-E517.

Three-dimensional reconstructed CT scans of 62 consecutive patients with cervical lesions were retrospectively evaluated to compare the maximum possible diameter of the atlantoaxial transarticular screw and C2-pars screw trajectories. Both techniques had a similar anatomic risk for vertebral artery injury.

Young JP, Young PH, Ackermann MJ, et al. The ponticulus posticus: implications for screw insertion into the first cervical lateral mass. J Bone Joint Surg Am. 2005;87:2495-2498.

The ponticulus posticus is an osseous anomaly of the atlas. Through a retrospective review of 464 lateral radiographs of the neck, the authors found a prevalence of 15.5%. Surgeons should avoid using the ponticulus posticus as a starting point for lateral mass screws in order to protect the vertebral artery.

Related

Operative Techniques Spine Surgery Expert Consult Online