Access Keys:
Skip to content (Access Key - 0)

Odontoid Fractures - To Fuse or Not to Fuse

Scroll down to respond to the OrthopaedicsOne Poll: Because screw fixation is so technically challenging, fusion of an odontoid fracture should be the treatment of choice

To fuse or not to fuse? That has been the ongoing question in the treatment of odontoid fractures. Drs. Albert J.M. Yee and Joel A. Finkelstein will debate the advantages and technical aspects of one versus the other.

Viewpoint 1: Albert J.M. Yee, MD, MSc, FRCSC

Odontoid Fractures - Anterior Odontoid Screw Fixation

Fractures of the odontoid account for approximately 20% of all cervical fractures, with approximately 70% being Type II fractures (ie, fracture crossing the base of the odontoid process at the junction with the axis body).1 The optimal treatment of Type II fractures remains controversial: The evidence suggests no standards or guidelines, but rather presents options in care.2-4 The decision regarding whether to operate and, if so, what type of procedure to perform, is subject to much discussion.

The bimodal pattern of patient age (young or old) presenting with these injuries that result from either high- or low-energy trauma is important to consider when planning treatment.5,6 An overwhelming goal of surgery in spinal fracture care is to restore mechanical stability. In odontoid fractures, surgical stability can be achieved by either fracture fixation of the C2 body to the odontoid process or by arthrodesis of the C1-C2 motion segment.

The desire to maintain cervical motion following an odontoid fracture is attractive, and, thus, odontoid screw fixation has been increasingly reported. Lessening or obviating the need for halo immobilization is desirable; additionally, autologous bone graft harvest is not needed. Fracture healing rates of this procedure has been demonstrated, in observational studies, to be comparable to atlantoaxial fusion rates, with a similar complication rate between procedures, although patient numbers in these studies are small and randomized studies are lacking.7-9


Patient Selection Criteria

Patient selection in odontoid fracture care is critical. Issues relating to fracture configuration, the size of the remaining "peg" in achieving distal fixation, the need for and ease of fracture reduction, and bone density are all variables to consider in the ability to properly place screw(s) with sufficient fixation to encourage fracture healing (Figures 1 and 2).

Figure 1. Computed tomography scan coronal (Figure 1A, top) and sagittal (Figure 1B, bottom) reformatted images of a 26-year-old female with a displaced Type II odontoid fracture following a high-energy motor-vehicle accident.

Figure 2. Plain lateral (Figure 2a, top) and open-mouth odontoid AP (Figure 2b, bottom) view of the same patient 9 months after fracture reduction performed under image guidance and general anaesthesia followed by anterior odontoid screw fixation with two AO small fragment terminally threaded lag screws.

While odontoid screw fixation may be an attractive option to many patients presenting with Type II odontoid fractures, its practical use may be tempered by the aforementioned factors, as the procedure does present technical challenges even in experienced hands. The risk of fracture non-union can be significant, particularly in some displaced Type II odontoid fractures.

The ability to achieve proper fracture realignment, as well as the ability to achieve screw fixation that satisfactorily stabilizes the fracture adhering to AO principles in lag screw fixation, will influence surgical success. C1-C2 fusion is often an easier procedure to perform technically, while acknowledging that the procedure significantly reduces neck motion, particularly in rotation, by about 50%.

Important Considerations

Several critical steps in odontoid fixation warrant discussion. Patient head and neck position need to be considered in the context of anaesthesia, fracture reduction, and steps to optimize spinal precautions. Reductions performed with the patient awake provide the opportunity to clinically monitor neurologic status. Reduction and adjustments to neck position under anaesthesia provide less-immediate feedback while attempting to properly reduce a fracture and facilitate the necessary insertional screw trajectory. The potential expertise and availability of intraoperative neuro-monitoring needs to be considered and coordinated in advance of surgery, as should discussions with anaesthesia regarding intubation strategies.

Often the head and neck need to be positioned in some extension to facilitate exposure to the inferior edge of C2. If fracture reduction is lost, for example with a posteriorly displaced fracture configuration, less extension should be utilized until provisionary fixation can be achieved. Real-time intraoperative imaging is necessary throughout the procedure.

The procedure is often performed with biplanar fluoroscopy, which mandates modifications to the head support of conventional operating room tables. Gardner Well's Tongs or halo traction can maintain in-line C-spine control; using a radiolucent frame extension from the table to the patient's head as well as a low-profile frame extension will facilitate ease in positioning fluoroscopic equipment. The ability to have two C-arms in the operating room can facilitate the procedure, although more advanced intraoperative imaging and 3D technology may be available at certain centers.

The anterior neck needs to be widely draped, with the exposure through a standard left or right anterior Smith-Robinson approach to the cervical spine, but with a skin incision near C5-6. Thus, the appropriate trajectory for the drill and screw can be obtained allowing placement in the anterior inferior portion of the C2 vertebral body. This will require cephalad retractors that expose the retropharyngeal space up to the level of C2-3. Exposure of the inferior bony portion of C2 is performed with efforts to minimize disruption to the C2-3 disc.

Positioning of the Screw

It is important to facilitate seating of a screw so that it is almost countersunk at C2/3 to minimize prominence of hardware that may cause post-surgical dysphagia. Starting 2-3 mm lateral to the midline of C2 also helps in this regard and will allow placement of either one or two 3.5-mm screws that can engage bone on the far side of the fracture line. Outcomes do not appear to be that different when using one or two screws.10,11

Attention to screw position - engaging as much of the bone in the remaining odontoid peg - and efforts to facilitate fracture compression using lag techniques are important to achieve. There does not appear to be significant differences biomechanically between one versus two screws in load-to failure stability. Internal fixation with one or two screws appears to provide approximately 50% of the initial strength of an unfractured odontoid11. Two screws may help control rotational stability, although it may be a challenge getting one, let alone two, screws in good position. Nine millimeters appears to be a critical diameter for the placement of two 3.5 cortical screws.12

A cannulated screw system can facilitate insertion, although there is some risk to inadvertent advancement of the wire. This can be addressed by the insertion of two 1.25-mm Kirschner wires, removing one and overdrilling the path of the removed wire using a 2.5-mm drill bit. The near fragment can be overdrilled with a 3.5-mm drill bit, tapping, and subsequent insertion of a partially threaded screw with or without a washer. In relatively osteopenic bone, tapping to the far fragment may not be necessary, although in healthier bone, this may be important to facilitate engagement of the far fragment and achieve fracture compression.

There are limited opportunities to ensure good screw fixation into the odontoid process. If not performed properly, there may also be difficulty in achieving fracture compression, as there is a tendency in tap and/or screw advancement to push the process away in a distractive manner given the limited size of the fragment. Smaller 2.7-mm screws could be considered in anatomically smaller odontoids. The use of Herbert screws has been reported, although some consideration needs to be given as to how much torque should be applied to obtain adequate fracture compression and avoid stripping the thread while attempting to seat the terminal threads into the body of C2.

Complications and Contraindications

Complications such as neurologic sequelae and loss of fracture fixation have been described, although many of these relate to early usage and inappropriate indications for this technique.

Relative and absolute contraindications to consider in this technique include:

  • Irreducible fracture pattern
  • Fracture with involvement and comminution of the atlantoaxial joints
  • Long oblique fracture line, particularly if positioned from anterior-caudal to posterior-cranial
  • Fracture involvement of the anterior-caudal C2 region with comminution
  • Concomitant unstable C1 ring fractures
  • Pathological fractures
  • Osteopenia

The use of screw fixation in established odontoid fracture non-unions is also controversial.


In summary, the main advantage to odontoid screw fixation is the preservation of cervical motion, which fusion surgery does not permit. In experienced hands, this technique can facilitate fracture healing and provide early spinal stability, thereby enhancing the recovery and rehabilitation for select patients.

Reprinted with permission from the Summer 2008 issue of COA Bulletin


  1. Anderson LD, D'Alonzo RT, 1974. "Fractures of the odontoid process of the axis." J Bone Joint Surg Am 56 (8): 1663-74
  2. Ochoa G, 2005. "Surgical management of odontoid fractures." Injury 36 Suppl 2: B54-64
  3. Ziai WC, Hurlbert RJ, 2000. "A six year review of odontoid fractures: the emerging role of surgical intervention." Can J Neurol Sci 27 (4): 297-301
  4. Julien TD, Frankel B, Traynelis VC, Ryken TC, 2000. "Evidence-based analysis of odontoid fracture management." Neurosurg Focus 8 (6): e1
  5. Bednar DA, Parikh J, Hummel J, 1995. "Management of type II odontoid process fractures in geriatric patients; a prospective study of sequential cohorts with attention to survivorship." J Spinal Disord 8 (2): 166-9
  6. Koivikko MP, Kiuru MJ, Koskinen SK, Myllynen P, Santavirta S, Kivisaari L, 2004. "Factors associated with nonunion in conservatively-treated type-II fractures of the odontoid process." J Bone Joint Surg Br 86 (8): 1146-51
  7. Marchesi DG, 1997. "Management of odontoid fractures." Orthopedics 20 (10): 911-6
  8. McCullen GM, Garfin SR, 2000. "Spine update: cervical spine internal fixation using screw and screw-plate constructs." Spine (Phila Pa 1976) 25 (5): 643-52
  9. Chang KW, Liu YW, Cheng PG, Chang L, Suen KL, Chung WL, Chen UL, Liang PL, 1994. "One Herbert double-threaded compression screw fixation of displaced type II odontoid fractures." J Spinal Disord 7 (1): 62-9
  10. Jenkins JD, Coric D, Branch CL, 1998. "A clinical comparison of one- and two-screw odontoid fixation." J Neurosurg 89 (3): 366-70
  11. Sasso R, Doherty BJ, Crawford MJ, Heggeness MH, 1993. "Biomechanics of odontoid fracture fixation. Comparison of the one- and two-screw technique." Spine (Phila Pa 1976) 18 (14): 1950-3
  12. Nucci RC, Seigal S, Merola AA, Gorup J, Mroczek KJ, Dryer J, Zipnick RI, Haher TR, 1995. "Computed tomographic evaluation of the normal adult odontoid. Implications for internal fixation." Spine (Phila Pa 1976) 20 (3): 264-70

Viewpoint 2: Joel A. Finkelstein, MD, FRCSC

Odontoid Fractures - Surgical Management by Fusion

Approximately 50% of axial rotation of the cervical spine occurs at the C1-C2 articulation. Instability at this level occurs most commonly as a result of fractures of the odontoid or bursting injuries of the atlas with disruption of the transverse ligament. The goals of treatment are to:

  • Reestablish the normal anatomic relationships at the level or the ring of C1
  • Maintain these relationships either through osteosynthesis of the dens or through a solid fusion of C1 and C2

The theoretical advantage of osteosynthesis of the dens is to maintain motion. However, loss of C1-C2 motion may occur, despite one's best efforts, due to scarring of the transverse ligament related to the injury itself or less-than-anatomic alignment of the odontoid process. Inability to achieve anatomic alignment also precludes direct fixation of the dens. This technique is also contraindicated in patients with osteoporosis, nonunion, os odontoideum, or thoracic kyphosis. The role of fusion for odontoid fractures is well established, and a number of techniques are available to achieve the goals of treatment in a safe and effective manner.


The mechanism of injury may be from forces causing extreme flexion, extension, or rotation. Fractures of the odontoid can be difficult to identify in the osteopenic skeleton of the older patient. In most cases, fractures can be identified on either the lateral cervical spine radiograph or open-mouth odontoid view. CT scans with sagittal and coronal reformations are useful for a better understanding of the fracture pattern and for surgical planning.

Fracture classification

Classification of odontoid fractures is based on the level of the fracture. The system of Anderson and D'Alonzo is most widely used. This classification has the greatest clinical applicability and prognostic implications.1

  • Type I – Type I injuries represent an avulsion fracture of the alar ligament from the dens. The importance of this injury is its potential association with an atlantooccipital dissociation. In the absence of this, conservative treatment is indicated.
  • Type II – Type II injuries are the most common odontoid fractures. They are transverse or oblique fractures through the waist of the odontoid. This type of fracture occurs in the cortical bone caudad to the transverse ligament and above the cancellous body of the axis. Displacement can be either anterior or posterior. Type II fractures have the highest incidence of nonunion. Various factors contribute to the poor prognosis associated with these unstable injuries. It is believed, however, that Type II fractures cause damage to the vascularity of the dens. The remaining dens — a hard cortical bone — has a small surface area for healing. Other factors have been suggested that may affect union. These include initial displacement of more than 5 mm, angulation greater than 10 degrees, posterior displacement and comminution, age greater than 40 years, delay in diagnosis, and smoking. A Type II fracture that cannot be reduced anatomically or maintained in a halo requires fusion.
  • Type III – Type III injuries are fractures into the body of the axis and usually have a well-vascularized, broad cancellous surface area. Treatment is generally with a hard collar or a halo orthosis.
C1-C2 fusion options
  • Wiring techniques – Gallie or Brooks fusion techniques utilize a sublaminar C1 wire with either a C2 spinous process or a C2 laminar wire. A tricortical iliac crest bone graft is placed on the C1 arch and over the C2 lamina, and it is maintained by the wire. Wires are less commonly used today as they are less biomechanically stable and do not control lateral bending, axial rotation, or anteroposterior translation compared to the newer Magerl or Harms techniques.2 Wiring techniques cannot be used in patients with posterior arch fractures of C1, and they require the use of postoperative halo-vest immobilization.
  • C1-C2 fusion using Magerl transarticular screws – The preoperative plan initially involves determining the feasibility of safely6 passing a screw across the C1-C2 joint without causing injury to the vertebral artery or the spinal canal. CT axial and sagittal reconstruction views are invaluable in visualizing the path of the vertebral artery and measuring the relationship between the artery, the canal, and the joint (Figure 1). The danger to the vertebral artery lies in the C2 body as it ascends from a more medial position laterally. If the location is initially too medial, the pathway takes it very close to the C1-C2 joint; there may be little room for a screw to pass. The decision to pass a unilateral screw or to defer to another technique will depend on the location of the vertebral artery. A further technical issue with this procedure mandates that C1 be reduced fully on C2 or the vertebral artery will be at risk for injury. The technique can be supplemented with a Gallie wiring technique for added stability. Cancellous bone graft is placed into the C1-C2 facet joint after decorticating the posterior portion of the joint. Postoperatively, a hard collar is worn for 6 weeks.3

Figure 1. Lateral and open mouth odontoid view of a patient undergoing C1-C2 fusion by the Magerl technique.

  • Atlantoaxial fusion using C1 lateral mass and C2 pedicle fixation (Harms technique) – This technique can be used preferentially, or as an alternative to the Magerl technique where the vertebral artery precludes safe passage of a C1-C2 transarticular screw. Polyaxial screws and rods connect the lateral masses of C1 to the pedicles of C2. This technique is biomechanically equivalent to the Magerl technique. The location of the C1 lateral mass is anterior to the overhang of the C1 posterior arch and the screw head must sit proud to allow connection to the rod. The entry point is the midpoint of the lateral mass directed straight anteriorly to the anterior arch of C1. The C2 pedicle is cannulated and its orientation is 20 degrees cephalad to the transverse plane and 30 degrees medial to the sagittal plane. The safest technique for cannulating the C2 pedicle is by direct exposure and palpation of the superior and medial edges of the pedicle. A lateral fluoroscopic image can help to determine the proper cephalad angulation. Cancellous bone graft is placed into the C1-C2 facet joint after decorticating the posterior portion of the joint. Postoperatively, a hard collar is worn for 6 weeks (Figure 2).4

Figure 2. AP and lateral radiograph following fixation using the Harm’s technique (lateral mass C1 screws and C2 pedicle screws).

  • C2 intralaminar fixation – Bilateral crossing of C2 laminar screws can be used as an alternative to C2 pedicle fixation. This technique has the advantage of having less risk to the vertebral artery and may be advantageous if there is difficulty with pedicle fixation. As described by Harms, the C1 fixation is as above. This technique is not as biomechanically rigid as the Magerl or Harms procedures, but early case series report satisfactory results.5

In summary, understanding of the anatomy and appropriate preoperative planning are mandatory for safe execution of any of the fusion techniques described above for C1-C2 instability. With the numerous tools now available in the surgeon's armamentarium, when surgery is required, fusion can be performed safely with good clinical outcome.

Reprinted with permission from the Summer 2008 issue of COA Bulletin


  1. Anderson LD, D'Alonzo RT, 1974. "Fractures of the odontoid process of the axis." J Bone Joint Surg Am 56 (8): 1663-74
  2. Grob D, Crisco JJ, Panjabi MM, Wang P, Dvorak J, 1992. "Biomechanical evaluation of four different posterior atlantoaxial fixation techniques." Spine (Phila Pa 1976) 17 (5): 480-90
  3. Jeanneret B, Magerl F, 1992. "Primary posterior fusion C1/2 in odontoid fractures: indications, technique, and results of transarticular screw fixation." J Spinal Disord 5 (4): 464-75
  4. Harms J, Melcher RP, 2001. "Posterior C1-C2 fusion with polyaxial screw and rod fixation." Spine (Phila Pa 1976) 26 (22): 2467-71
  5. Wright NM, 2004. "Posterior C2 fixation using bilateral, crossing C2 laminar screws: case series and technical note." J Spinal Disord Tech 17 (2): 158-62

Reprinted with permission from the Summer 2008 issue of COA Bulletin


Added By

The license could not be verified: License Certificate has expired!

Log In or Sign Up to follow this author.

More From
The license could not be verified: License Certificate has expired!
The license could not be verified: License Certificate has expired!
Content Partner

Learn about becoming an OrthopaedicsOne Content Partner.

Academic Resources

Resources on Odontoid Fractures - To Fuse or Not to Fuse from Pubget.

The license could not be verified: License Certificate has expired!
Related Content

Resources on Odontoid Fractures - To Fuse or Not to Fuse and related topics in OrthopaedicsOne spaces.

Showing first 5 of 217 results