Feasibility of Double Anterior Odontoid Screw: A CT-Based Morphometric Analysis of the Axis in Adult Indian Population
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.304125
Source of Support: None, Conflict of Interest: None
Keywords: Coronal plane, CT morphometry, double odontoid screws, odontoid fracture, Herbert screw, sagittal plane
Among the spectrum of odontoid fractures, the management of the type-II fracture is unique in various aspects. External immobilization is associated with very high nonunion rates as well as the complications of the halo itself. Internal fixation using C1-2 arthrodesis compromises both rotation and flexion movements at the neck significantly.,,,,, Hence, in the presence of intact transverse atlantal ligament, insertion of the anterior odontoid screw is the ideal procedure in younger patients, as it provides better functionality in terms of neck movements. Majority of the initial reports have emphasized on double odontoid screws, as it provides better stability and stiffness to the odontoid.,,, We analyzed the CT-based dimensions of the odontoid process in both the transverse and sagittal planes and assessed the feasibility of placing double odontoid screws.
This was a prospective, morphometric analysis of retrospective data from a tertiary care center. Two hundred and fifty patients undergoing CT angiography of the cervical spine were included in this study. Patients with craniovertebral junction anomaly, presence of movement artifacts during the study, and any accompanying cervical spine disease like traumatic, infective, neoplastic, or degenerative etiology were excluded from the study.
The CT scans were performed on the Somatom Definition edge 128 slice 64-row detector Siemens CT scanner. Two specialists, neuroradiologist and a neurosurgeon, reviewed the data of each patient independently and the following findings were noted after consensus:
Patient data were sourced from the medical record section of our hospital. CT angiography images were retrieved from the Picture Archiving and Communication System (PACS).
CT angiograms of the neck (0.6 mm slice thickness) were obtained using helical acquisitions from the aortic arch through the skull base. A 20–22-G cannula was inserted through the right antecubital vein following which intravenous administration of 50 mL iodinated contrast (Iohexol) at 5 mL/s flow was done. The images were reconstructed by a standard algorithm.
Source images were retrieved from PACS onto the syngovia image-processing software, which is a multimodality reading solution, built on a client-server platform. Multi-planar reconstructions (sagittal, coronal, and axial reconstructions) were analyzed to select the best images for measurement. The images were magnified for accurate measurement.
A total of 250 patients were evaluated with age ranging from 18 to 80 years. 174 were (69.6%) and 76 (30.4%) were females. The odontoid dimensions in transverse and sagittal planes were measured and tabulated [Table 1].
The mean TD at the odontoid waist was 8.844 mm in males and 8.471 mm in females (P = 0.016).
The mean TD at the widest point of odontoid was 9.930 mm in males and 9.421 mm in females (P = 0.002).
The mean TD at C2 at the level of inferior endplate was 15.068 mm in males and 13.639 mm in females (P < 0.001).
Mean AP diameter 2.5 mm away from the midline on the left side at the odontoid waist was 9.870 mm in males and 9.154 mm in females (P < 0.001).
Mean AP diameter 2.5 mm away from the midline on the right side at the odontoid waist was 9.381 mm in males and 9.012 mm in females.
Mean AP diameter 2.5 mm away from the midline on the left side of C2 at the level of inferior endplate was 15.824 mm in males and 14.833 mm in females (P < 0.001).
Mean AP diameter 2.5 mm away from the midline on the right side of C2 at the level of inferior endplate was 15.781 mm in males and 14.673 mm in females (P < 0.001)
Mean AP diameter 2.5 mm away from the midline in the left side at C2 body superior endplate was 11.819 mm in males and 11.199 mm in females (P < 0.001).
Mean AP diameter 2.5 mm away from the midline on the right side at C2 body superior endplate was 11.62 mm in males and 11.23 mm in females (P < 0.001).
The craniovertebral junction is the site of maximum range of movements in the spine. Rotation of the neck occurs between the atlas and the axis. This is possible due to the special design of the axis along with its appendage, the dens arising from the body. The rotation of the neck takes place around the dens, which is secured to C1 with the transverse atlantal ligament (TAL). Fracture of the dens constitutes approximately 10% of all cervical spine injuries. Anderson and D Alonzo distinctly classified the odontoid fractures into three types in 1974. Type II constitutes approximately two-thirds of all the fractures of the odontoid.,,,,,
Literature is replete with a variety of management strategies for the odontoid fractures. The available evidence supports the use of external immobilization in the majority of the type-I and type-III fractures. However, the management of type-II fractures is challenging and both external immobilization and internal stabilization techniques are recommended. External immobilization as a stand alone treatment, in the form of a cervical collar or occipito-cervical halo has been associated with high rates of nonunion (80%), neck stiffness, atlantoaxial instability, even myelopathy., Type-II odontoid fracture is usually located at the waist of the dens and is further classified into three types as; anterior oblique, posterior oblique, and comminuted. Stabilization of this fracture involves either anterior or posterior approaches. Posterior stabilization involves fixing C1 and C2 using the Goel-Harms technique. This procedure leads to the loss of nearly 50% rotatory and 10% flexion-extension movement of the neck., Placement of anterior odontoid screw spanning across the fracture line is more physiological and leads to the preservation of the motion. Hence, in indicated cases, although technically challenging, odontoid screw insertion is the procedure of choice.
Single versus double odontoid screw
Bohler and Nakanishi were the first to report the fixation of type-II odontoid fractures using odontoid screws, independently. Bohler used double stainless steel odontoid screws for fixation. Since then, there has been debate over placing single versus double screws in this type of fracture. Double screws theoretically provide greater stability and strength during the period of fracture healing compared to a single screw., Jenkins et al. found a 10% screw breakage in patients undergoing a single screw compared to double screw although the nonunion and malunion rates were comparable. Sasso and colleagues studied the biomechanics of the type-II odontoid fractures in cadavers using single and double screws. Overall they did not find any difference in loading failure between single and double screw. They did document better stiffness in extension loading with double screws. They also concluded that inter-digitaton of the fracture site is the most important factor responsible for fracture stability than the number of screws.
We measured the TD at two locations along its length of dens; first at the level waist (narrowest part) and second at the level of the widest part. Among the 250 patients, the TD of the dens at the waist was <9 mm in 64% patients and was >8 mm in 78% of cases. There was a statistically significant difference between the gender (P < 0.016), with 40% of the total male population and 24% of total females having diameters of >9 mm at the waist. The TD of dens at the level of the widest part was >8 mm in 95% of cases. Among these, 80% of patients had a diameter of >9 mm with 85% males and 70% of total females [Table 1] and [Table 3].
The C2 vertebra along with the dens has been studied morphometrically in diverse populations across the globe. Nucci and colleagues analyzed CT-based dimensions of the dens in the normal population (n = 92) and found that, 95% of the patients had TD of the dens to be more than 9 mm. They found no statistically significant difference between the male and female populations. This could have been due to the inclusion of a heterogeneous patient population, with 51% Caucasians, 32% Blacks, and 17% Asians. Kulkarni et al. found at least one diameter less than 9 mm in the transverse plane of dens in 55% of South Indian patients. Malaysians (34%), South Americans (35%), and Arabians (61% ) similarly had atleast one of the diametres lesser than 9 mm in the transverse plane. Thereby, rendering them unsuitable for double screw placement in the transverse plane.,,,
The mean anteroposterior dimensions (APD) at the waist of the dens was 12.4 mm and 11.6 mm in both male and female populations, respectively (P = 0.016). None of the male and two (2.6%) female patients had an APD of less than 9 mm, with 74 (97.4%) female subjects having APD of more than 9 mm at the waist of the dens in the midline. APD was measured approximately 2.5 mm on either side of the midline in the parasagittal plane at the level of the odontoid waist. The diameter was > 9 mm in 98% of patients on the left parasagittal plane and in 70% of patients on the right [Table 2] and [Table 3]. This indicates that the odontoid process is asymmetric in the sagittal plane and is larger on the left side compared to the right, in 28% of our patient population. In the Malaysian population, the mean APD was 11.3 mm and 10.9 mm in men and women, respectively, 10.5 mm in males and 10 mm in females in the Arabian population.,
Feasibility of double odontoid screws
The most commonly used odontoid screws are available as 3.5 mm diameters cannulated lag screws. The placement of two such screws requires an odontoid process with a minimum diameter of 9 mm in the coronal plane. [Table 4] depicts the diameter of the odontoid in various patient populations across the world. Asians showed the least sizable odontoids ranging between 39 and 55% feasible for double screws placement.,, 95% of the odontoids in the Caucasians were feasible for double screws and 57% in the South Americans., Alternatively, in patients with odontoid diameters of less than 9 mm, placement of two screws of 2.7 mm or a single 4.5 mm Herbert screw has been suggested with comparable efficacy.,,, Most studies have reported odontoid double screw placement in the coronal plane. Going by the transverse dimensions of the odontoid, only 50% of the Asians will be eligible for the double-screw placement.,,
In the current study, we also examined the APD of the odontoid, which was more than 9 mm in 98.4% of cases. Similarly, this dimension was studied by Daher and Yousef et al. and found 97% and 100% patients with a sagittal diameter of more than 9 mm, respectively., Hence, it can be postulated that although double screw placement can be limited by the smaller diameter in the coronal plane, it is feasible in almost all the patients, when considering the sagittal plane. Just by changing the orientation of the screws in the sagittal plane, majority of the type-II odontoid fractures can be managed by double screws. This can be a technically challenging maneuver, but with the help of O-arm and navigation guidance, this can be accomplished. However, further model based and cadaveric studies in future may substantiate the feasibility of this technique.
The odontoid process in the 50% of the Asian population has a smaller diameter in the transverse plane compared to their western counterparts, thereby ample enough to house just a single 3.5 mm/4.5 mm Herbert screw. Nearly 98% of the odontoids in the Asians as well as the western counterparts have adequate diameters in the sagittal plane so as to accommodate double screws. Hence, careful study of the dimensions of the odontoid is mandatory prior to planning double screw insertion. Nearly all the cases of type-II odontoid fractures can be managed with double screws by changing the orientation and inserting them in the sagittal plane, of-course, utilizing the O-arm CT and intraoperative navigation.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]