11 Atlantoaxial Transarticular Fixation Techniques
The posterior upper cervical spine C1–C2 transarticular screw fixation technique was introduced by Magerl in the 1980s and has been widely published.1–4 It was developed because of partially unsatisfactory results with wiring techniques.5,6 Magerl’s transarticular posterior technique was conceptually a further development of the lateral transarticular screw fixation, which had been developed in 1971 by Barbour, who used a bilateral neck approach instead.7–9 Furthermore, this lateral approach to the C1–C2 joints was anatomically more demanding and more prone to complications than a simple posterior midline approach due to the anatomical dissection involved.
In the 1990s, we developed an anterior approach for transarticular C1–C2 screw fixation, which again represented progress in terms of surgical access, as it avoids posterior paravertebral muscle dissection, a possible source of postoperative neck pain.7,10,11 For this anterior transarticular screw fixation, the simple, well-established unilateral anterior approach is chosen in the virtual space between the trachea, esophagus, and long muscles on the medial side and the sternocleidomastoid muscle with the underlying neurovascular bundle on the lateral side. When we first published our data, we realized that others had come to the same idea and published their experiences even earlier in the neurosurgical literature.12
This approach is extremely atraumatic and is recommended in elderly patients (supine position, very little blood loss) and in combination with an anterior odontoid screw fixation or anterior plating of the middle cervical spine.
Posterior Techniques
Surgical Principles
The original transarticular screw fixation as described by Magerl1 allows rigid stabilization of the C1–C2 segment in all directions. The screw enters from the posterior aspect of the massa lateralis of C2, passes through the subchondral bone of the superior articular surface of C2, and ideally crosses the joint gap of C1–C2 to enter the subchondral bone of the inferior C1 joint surface. Consequently, the fixation point of the transarticular screws is located exactly in the center of the facet joints of C1–C2, exposing this screw only minimally to flexion, extension, and rotational forces.
This screw is not a compression screw but a positioning screw, thus avoiding compression of the roots within the C1–C2 foramen by narrowing the foramen. The screw is under minimal adverse load, and breakage is extremely rare even when formal fusion is lacking.
Biomechanics
There have been several studies investigating the biomechanical behavior of C1–C2 screw fixation in comparison with other fixation systems.2,13–22 The C1–C2 transarticular screw with interspinous graft is a standard technique for atlantoaxial fixation15 and can be considered equivalent to Goel and Laheri’s technique of screw-plate/rod systems that are anchored in the lateral mass of C1 and C2.16 The transarticular fixation of C1–C2 prevents lateral bending and axial rotation better than current posterior cable-secured graft constructs, which are somewhat better in terms of flexion and extension. This is true only when the transarticular screw fixation is not combined with an interspinous graft.18 Similar in vitro biomechanical tests for five different C1–C2 constructs confirm the above findings and demonstrate clearly that the Gallie or Brooks-Jenkins cable fixation alone may not be adequate for atlantoaxial arthrodesis. The combination of the Brooks-Jenkins fixation with one transarticular screw is superior to the construct of a Gallie fusion with one transarticular screw fixation.19 Richter et al. tested six fixation systems for the atlantoaxial segment and came to the conclusion that simple transarticular screws from C2 to C1 are best for stabilization, lateral bending, and axial rotation.20 When a Gallie fusion or claw was used, additional stability was obtained. The authors concluded that the combination of transarticular screw fixation with a Gallie fusion and Goel and Laheri’s technique of lateral mass fixation are the best fixation techniques of all those tested.20 They also mentioned that lateral mass screws (Goel and Laheri’s technique) or claw fixation together with isthmic screws is a valid alternative in case transarticular screws are not feasible.20
Surgical Technique
Surgical Approach
The patient is positioned in the prone position, and the position of C1–C2 is checked using lateral image intensifier control. The alignment of C1–C2 is facilitated by traction either with Gardner-Wells tongs/halo or a Mayfield clamp, the latter of which controls translation more easily. The neck is flexed as much as possible to facilitate insertion of the screws, and an image intensifier is used to exclude redislocation in case of a fracture dislocation ( Fig. 11.1 ).
A midline incision is performed from the occiput to the tip of the spinous process of C3. The arch of C1, spinous process, laminae, and inferior articular processes of C2 are exposed subperiosteally. In case of a persistent anterior dislocation of C1 over C2, a reduction is done by pushing on the spinous process of C2 and/or by pulling gently on the posterior arch of C1, either with a towel clamp or with a sublaminar wire or suture sling. Persistent posterior dislocation would require opposing forces.
It is recommended that the surgeon gently expose the inner side of the isthmus and pedicle of C2 and stay medially as close as possible to this landmark. A small dissector is used to expose the cranial surface of the lamina and isthmus of C2 by careful subperiosteal dissection up to the posterior capsule of the atlantoaxial joint ( Fig. 11.2a ). Medial to the isthmus, the atlantoaxial membrane is visible. The laterally situated vertebral artery is not exposed. It is further recommended to get a computed tomography (CT) scan of the C1–C2 segment before a surgical intervention to rule out an anomalous course of the vertebral artery.
Standard Technique
Using lateral image intensifier control, a 2.5-mm-long drill bit is inserted in a strictly sagittal direction. The oscillating attachment prevents soft tissues from being wrapped around the drill bit. The entry point of the drill bit is at the lower edge of the caudal articular process of C2 ( Fig. 11.2b ). The drill bit goes through the isthmus near its posterior and medial surface. It then enters the lateral mass of the atlas close to its posteroinferior edge. Anteriorly, the drill perforates the cortex of the lateral mass of C1. The screw length is measured, and the direction of the screw canal is checked using an image intensifier ( Fig. 11.2c ).
The 3.5-mm cortex screws are inserted after checking with a 3.5-mm tap across the C1–C2 joint; the anterior cortex of C1 must not be tapped ( Fig. 11.2d ). Proper caudocephalad drilling may be difficult because the neck muscles and the upper torso prevent the correct placement of the drill bit. Gently pulling the spinous process of C2 cranially with a towel clamp facilitates drilling. It is often necessary to drill through a distal percutaneous stab wound to place the drill in the correct angle of ~60°. Prepping and draping of the upper thoracic spine are necessary to be able to enter the drill bit through a stab incision.23–25
Drilling in a horizontal direction must be avoided because
At the level of C2, the vertebral artery runs upward anteriorly to the C1–C2 joint and could easily be damaged.
The screw could exit C2 anteriorly and not enter the atlas.
Postoperative Care
Patients are immobilized in a firm collar for a period of 6 to 8 weeks but are allowed to remove the collar for daily care. After 6 to 8 weeks, the collar can be discarded when resting. If additional posterior wiring has been used, a soft collar can be worn instead of a firm collar.
Cannulated Screw Technique
As in direct anterior fixation of the odontoid,26 we used the cannulated screw technique for both posterior and anterior transarticular screw fixation.7,10 This technique also has been used and published by others.26,27 With Kirschner (K) wires as drill guides, we performed the surgery in a minimally invasive manner. Through bilateral stab wounds, a K wire can be placed under anteroposterior and lateral simultaneous image intensifiers from the described entry points toward the lateral projection of the C1 arch. The interspinous graft can then be placed between the C1 and C2 through a small midline incision. This technique has been further elaborated by the Ulm group.23–25 More recently, an image-guided technique was presented; however, this technique is far from being used on a routine basis.28
Using lateral image intensifier control, a 1.2-mm K wire is inserted with a surgical drill guide in a strictly sagittal direction into each hole ( Fig. 11.3a, b ). The entry point of the K wire is at the lower edge of the caudal articular process of C2 ( Fig. 11.3a ). The length of the screw is established with a special ruler by measuring the protruding part of the guide-wire ( Fig. 11.3c ). Before inserting the cannulated screw over the K wire, the entrance for the screw is prepared by a special 3.5-mm cannulated countersink ( Fig. 11.3d ) to facilitate the starting purchase of the screw. The appropriate 3.5-mm self-tapping, cannulated cancellous bone screw (same as for odontoid fixation, usually ~40–45 mm long) is inserted over the guidewire ( Fig. 11.3e ). The progress of the screw must be observed under lateral image intensifier control to ensure that the K wire does not migrate proximally beyond the C1 arch and the screw does not push away (anteriorly and upward) the lateral mass of C1, consequently dilating the C1–C2 joint.
In severe degenerative diseases, the sclerotic subcortical bone of the C2 joint may prevent insertion of the self-drilling screw. In this case, a 2.7-mm cannulated drill bit is used to cross the joint, and a cannulated fully threaded 3.5-mm screw can be inserted after regular tapping of the predrilled screw hole.
Postoperative Care
Patients are immobilized in a firm collar for a period of 6 to 8 weeks but are allowed to remove the collar for daily care. After 6 to 8 weeks, the collar can be discarded when resting. If additional posterior wiring has been used, a soft collar can be worn instead of a firm collar.
Fusion
Following bilateral screw fixation, according to the above described techniques, a posterior C1–C2 fusion is performed, preferentially Brooks-Jenkins or Gallie fusion.5,6 According to Magerl’s technique, it is preferable to supplement the graft with a posterior wire or cable, as this increases the stability of the fixation and hence the fusion rate ( Fig. 11.3f, g ). Instead of a wire, a nonresorbable suture can be used.
When there is a defect or fracture of the posterior arch of C1, a fusion of the atlantoaxial joint must be performed. For visualization of the atlantoaxial joint, K wires are drilled into the posterior aspect of the lateral mass of the atlas. For this purpose, the greater occipital nerve is retracted cranially. The soft tissues containing the greater occipital nerve and its accompanying venous plexus are retracted simultaneously. The atlantoaxial joints are exposed by opening the posterior capsule, thus making the C1–C2 joint visible ( Fig. 11.4 ). The articular cartilage of the posterior half of the facet joint is removed with either a small chisel or a sharp curet, after which the joints are packed with cancellous bone, and the screws are inserted.
Advantages and Disadvantages
The advantages of transarticular C1–C2 screw fixation are biomechanical superiority to the wiring techniques, the possibility of maintaining reduction, and the lack of need of integrity of the posterior arch of C1 for the fixation technique. Furthermore, this technique can be easily combined with a Gallie fusion, thus obviating the need for relevant external immobilization.
The disadvantages include the fact that this is a demanding technique and that there is a potential risk of bleeding due to injury to the venous plexus located lateral to the entry point of the screws or due to injury of the vertebral artery.29–33 This is particularly true in anomalies of the vertebral artery course in relation to the massa lateralis of C1 and C2.34–36 An aiming device may facilitate insertion of the screws.37,38 It may be difficult to apply this technique in patients with an upper thoracic kyphosis because the necessary inclination of the screws may be in conflict with the kyphosis, and the insertion of the screw therefore may be limited.
Depending on the obliquity of the screw insertion, a complete reduction of the C1–C2 joint may be necessary. When there is erosion of the transverse process of C2, the vertebral artery may change direction; this aberrant trajectory of the vertebral artery may make this technique impossible. In such a case, bilateral transarticular screw fixation may not be possible. An option may be unilateral transarticular screw fixation combined with posterior C1–C2 wiring.
Case Illustration
Transarticular screw fixation of C1–C2 with Gallie-type titanium cable fixation in a 55-year-old rheumatoid patient is shown in Fig. 11.5.