Craniovertebral junction evaluation by computed tomography in asymptomatic individuals in the Indian population
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.232288
Source of Support: None, Conflict of Interest: None
Keywords: Computed tomographic scan, craniovertebral junction, Indian, normal parameters, osteometry
Most of the literature on craniovertebral junction (CVJ) anomalies focuses on correction of an established deformity. Normal CVJ parameters are still poorly understood and historically have been defined based on measurements made on plain X-ray films. There is paucity of data defining the metrics of normal CVJ osteology in the Indian population. Moreover, X-ray measurements have become obsolete in the computed tomography (CT) era, which enables much more accurate and objective determination of the osteometric data. Most studies in which CVJ parameters have been recorded include patients with known congenital bony CVJ anomalies or patients with Chiari malformation More Details.,,, The importance of knowing the normal CVJ parameters cannot be overstated when rapid advancements are being made in CVJ instrumentation and correction of CVJ deformities.
Accordingly, the present study is an attempt to quantitate the normal osteometric indices for anatomically normal CVJ in an Indian population.
This retrospective observational study included 49 consecutive adult patients who underwent a CT angiogram for a suspected vascular condition unrelated to the CVJ. Patients with a previously undiscovered asymptomatic CVJ anomaly, pathologic involvement of the CVJ, or an incomplete study of the region were excluded.
Computed tomography measurements
All the measurements were performed independently by two observers (CD, SM), and the recordings were made upon agreement with a maximum permissible interobserver difference of 1 mm for a particular measurement. In the case of dispute between observers, the opinion of an expert radiologist was considered final. The studies were performed on a 64-row multidetector CT scanner (MDCT), and images were reconstructed into 0.5-mm slice thickness. All the scans were analyzed on the 'bone window' setting. All the measurements were made using the picture archiving and communication system (PACS). The following measurements were made:
The data was analyzed using the statistical programming language R. The means were compared using the unpaired Student's t-test.
The CT measurement of CVJ morphometry was made in 49 asymptomatic Indian individuals. Out of the total study population, 31 were male subjects whereas 18 were female subjects with a mean age of 50.9 years (standard deviation [SD] 14.91; minimum age 18 years; maximum age 81 years). All the patients underwent CT angiography for a cerebrovascular condition. All the measurements are summarized in [Table 1]. In comparison between males and female subjects in this series, statistically significant difference was found only in the C1 lateral mass anterior-posterior diameter, odontoid length, and odontoid screw trajectory length, which were smaller in female subjects when compared to the male ones. [Table 2] compares the measurements between male and female subjects.
In our study, we evaluated the CT parameters of CVJ in 49 patients with imaging indications unrelated to the CVJ pathology. Most of the existing literature focuses on patients with existing congenital anomaly and deformity correction. The normal range depicted in [Table 1] is important not only in planning deformity correction but also in designing spinal implants of appropriate dimensions in an Indian population.
Rojas et al., assessed normal anatomic relationships of craniocervical articulations on MDCT among 200 patients who underwent imaging as per the trauma protocol and did not demonstrate any soft tissue or bony abnormality on imaging. They assessed only 6 parameters – basion-axial interval (BAI), basion-dens interval (BDI), powers ratio, ADI, and atlanto-occipital interval (AOI) in each patient. They compared these values with previously accepted data on plain radiographs after appropriate statistical analysis. They reported that 95% of their patients had an ADI less than 2 mm, which was smaller than the previously accepted value of 3 mm based on studies reported in 1960s using plain radiographs. We found that only 3 patients had an ADI >2 mm (6.1% of the study group) and no patient had an ADI >2.5 mm. The findings are similar to those of Batista et al., who did not find any patient with an ADI >2 mm. Based on our findings and those of the studies quoted above, we can infer that 2 mm should be considered as the upper limit of normal ADI for adult patients on CT scan sagittal images.
The mean basal angle in our observations was 121.65° (SD, 5.12°; range, 109.7–133.7°). This agrees well with the findings of other series. Koenigsberg et al., using the same anatomical parameters (from the top of the dorsum sellae to the nasion and the basion) of 200 normal adults using MRI, obtained a mean value of 129 ± 6°. Botelho and Ferreira  in their series of 33 asymptomatic patients reported a range of 107–132°(mean 119° ± 7.1°) as the normal basal angle. They suggested that a diagnosis of platybasia should be made when the basal angle is greater than 133°. Smoker et al., measured BA as the angle formed by the intersection of the nasion–tuberculum sellae line and the tuberculum sellae–basion line and suggested that BA should always be less than 140°. Understandably, when the BA is measured with respect to the centre of the sella instead of the tip of dorsum sella, a higher value is observed.
For the clivus–canal angle (CCA), a range of 150° to 180° is generally accepted as normal on plain radiographs. Ventral spinal cord compression has been found with measurements less than 150° on plain radiographs. Various studies based on MR and CT scan imaging have showed a broad range of values with mean CCA values being lower than 150° in these studies. The mean CCA in our study was 150.69° (SD, 12.05°; range, 118.20–173.70°). This is in concordance with the studies by Botelho and Ferreira  (148° ±9.8°; range, 129–179°) and Batista et al., (153.6° ±7.6°; range, 132.3–173.9°).
The clivus length was measured as the distance from the top of the dorsum sellae to the basion. The mean length obtained in our study was 43.05 mm (SD, 3.33; range, 34.84–51.18). This is almost similar to the study by Heiss et al., who reported a mean clivus length of 43.2 mm and 44.7 mm, respectively. In the studies by Dufton et al., and Batista et al., comparing the clival length in patients with Chiari malformation (CM) and normal individuals, it was found that patients with CM had a shorter clivus length (4.02 ± 0.45 cm) than in the controls (4.23 ± 0.42 cm, P= 0.009).
The thickness of EOP is important during the planning of occipito-cervical instrumentation. The mean EOP thickness was measured to be 12.15 mm (6.8–25.52 mm). Morita et al., in their study involving 105 individuals, found that the maximum thickness of the occipital bone was consistently found at the level of the EOP. Areas with thickness >8 mm were more frequent at the EOP and up to 2 cm in all directions, as well as up to 1 cm in all directions at a height of 1 cm inferiorly, and up to 3 cm from the EOP inferiorly. Males had a thicker occipital bone around the EOP compared to females. These findings suggest that implant sizes for anchoring the occipital bone should be individualized based on preoperative CT studies.
The occipital condyle connects the cranium with the vertebral column. The average occipital condyle height in the study by Saralaya et al., was 10.2 mm. They measured the height of 140 occipital condyles from 70 dry skulls. The mean occipital condyle height on the right and left sides in the present study was 10.07 and 10.16 mm, respectively, which is in close agreement with their findings. This is similar to the findings in the CT morphometric study by Batista et al., (height of 10.48 mm for the right condyle and 10.74 mm for the left condyle). Naderi et al., found the occipital condyle height of 9.2 mm, length of 23.4 mm, and width of 10.6 mm. The anatomical knowledge of occipital condyle may help in planning occipital condyle screw insertion for transcondylar occipito-cervical fixation, as well as for planning lateral approach to the skull base, such as the transcondylar approach.
The mean distance of the odontoid tip from the McRae line in this series was 5.11 mm and no patient had the odontoid tip above the McRae line. Cronin et al., in their study regarding the evaluation of McRae skull base line, found the mean distance of the odontoid tip to McRae line to be of 5 mm. Our observations are in agreement with the existing literature that in normal individuals, the odontoid tip should be inferior to this line by about 5 mm.
The mean AP length of C1 lateral mass was 17.22 mm on the right side and 17.09 mm on the left side. The mean height of the C1 lateral mass was 12.52 mm and 12.50 mm on the right and left sides, respectively. Christensen et al., in their anatomic surface osteometric analysis of cadaveric cervical spine, stated that the minimum dimensions found from 240 C1 lateral masses were 13.15 mm in anterior-posterior, 4.22 mm in medial-lateral, and 4.73 mm in cephalocaudal directions. Goel and Laheri  followed by Harms and Melcher  described the posterior C1-C2 fixation technique for atlantoaxial stabilization, and since then this technique has been extensively used worldwide for atlantoaxial stabilization for various indications. Knowing the dimension of C1 would help in refining the surgical strategy in such patients.
The pB-C2 measurement had a mean length of 8.25 mm, and no patient had a pB-C2 length less than 5 mm. The maximum pB-C2 measurement was 11.8 mm. This is similar to the findings of a study of 125 patients by Khaleel et al., who measured the pB-C2 line ranging from 0 to 11.2 mm (mean, 6.5 ± 2.1 mm) in length. Adopting a 9 mm  cut-off for predicting ventral brainstem compression seems to be inappropriate in the light of these discordant findings. The mean odontoid retroflexion angulation was 80.38° in our measurements, whereas the odontoid retroversion had an average angulation of 87.41°. Khaleel et al., found that the odontoid retroflexion ranged from 70 to 89°, and the odontoid retroversion angle ranged from 57° to 87° (mean, 71.9° ±5.3°). The range and mean of the odontoid height were 17–27 mm and 22 ± 1.8 mm, respectively. The mean pB-C2 line was 6.5 ± 2.1 mm with a range of 0–11.2 mm. The authors concluded that the odontoid process in adults was longer, more posteriorly inclined, and had a greater pB-C2 measurement when compared to the pediatric population data available in the literature. The posterior odontoid process inclination has been associated with Chiari I malformation (CM-I) in the pediatric population. Besachio et al., found a significant difference between adults with CM-I and 150 sex-matched controls, in the mean clivus–canal angle and the mean pB-C2 line. Goel  in his study of 65 patients with CM who were treated by atlantoaxial fixation postulated that the pathogenesis of CM with or without basilar invagination or syringomyelia was primarily related to the instability in the atlantoaxial region. These parameters are especially important in treating CM patients with CVJ anomalies.
The mean dens length was 21.4 mm; and, when the length of the body was measured along with the dens, the mean length was 32.42 mm. The mean length of the expected odontoid screw trajectory was 35.76 mm. Korres et al., in their study involving 115 patients, found that the mean distance from the tip of the odontoid to its base was 17.25 and 17.28 mm in female and male subjects, respectively; whereas the mean distance from the tip of the dens to the anterior-inferior corner of the body of axis was 39.2 mm. Interestingly, male subjects had a longer odontoid length and odontoid screw trajectory compared to the female ones. The mean odontoid screw trajectory was shorter in Indian studies compared with those reported from the west. Women had a significantly shorter trajectory length than males, which is similar to the observations made in a study on Western subjects  and in the Indian population. Consequently, odontoid screws for implantation in Indian patients must be fabricated with the length ranges customized for Indian population. The screws designed on the basis of Western data  may be excessively long and may not achieve the lag effect essential for osteosynthesis along the fracture line of type II odontoid fractures.
The mean C2 pedicle thickness was 4.33 mm on the right side and 4.12 mm on the left side. The C2 pedicle transverse angle was 44.43° on the right side and 41.71 mm on the left side. The mean estimated C2 pedicle screw length was 24.59 mm on the right side and 24.83 mm on the left side. Ould-Slimane et al., analyzed the morphometric properties of 200 C2 pedicles. They found the average length to be 26.18 mm and the average diameter as 5.18 mm. The average pedicle transverse angle of 36.6° was demonstrated. Bhatnagar et al., in their study of 50 patients, found the mean pedicle length and width being 15.5 ± 3.5 and 4.7 ± 1.7 mm, respectively. Thus, the C2 mean pedicle diameter and its length in Indian patients are slightly lesser when compared to their assessment in the West; this fact should be kept in mind during C2 pedicle cannulation. Important differences between studies from other geographical regions and our observations have been summarized in [Table 3].
The study population comprised only adult patients over the age of 18 years; hence, pediatric and adolescent age groups are not represented, and the developmental aspects of normal CVJ morphometry have not been captured.
The study population may be more representative of the north Indian population because of the geographical location of the tertiary hospital where this study was conducted. However, because significant racial differences do not exist among Indians, we expect the observations of this study to be translatable to other regions of the country.
As this was a retrospective, non-funded study based on the available data, the sample size was limited. Hence, the CVJ parameters could not be meaningfully compared among the different age groups.
Due to lack of CT-based osteometric studies on the CVJ in the Indian population, clinicians have depended upon descriptions offered by studies performed among Western populations. However, as our study highlighted, significant differences do exist between Indian and western population with respect to several parameters in CVJ osteometry, which are relevant to surgical procedures commonly performed upon this region of the spine. Recognizing these differences may augment our understanding of the CVJ and lead to improved design of spinal implants suited to the Indian population.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]