Abstract
Objective
To establish a preliminary magnetic resonance imaging (MRI) database of whole spine of healthy Chinese adolescents.
Methods
MRI examination of whole spine and hindbrain was performed in 41 enrolled students aged 11–17 years (mean age 13.95; 18 males, 23 females) using a 1.5‐T MR Scanner. Measurements of the ratio of anteroposterior (AP) and transverse (TS) diameters of the cord, cerebellar tonsillar level related to the basion‐opsithion (BO) line, location of conus medullaris, total cord length, total vertebral length, cord/vertebral length ratio, thoracic cord area, thoracic vertebral area, thoracic cord/vertebral area ratio were obtained.
Results
Mean values of cervical AP and TS were 6.63 mm and 12.21 mm, respectively. The mean level of cerebellar tonsillar related to BO line was 3.97 mm. Mean level of conus medullaris located in L1 lower 1/3. Total cord length was 399.34 mm, total vertebral length was 529.49 mm, Cord/vertebral length ratio was 0.75 and thoracic cord/vertebral area ratio was 0.17 on average. Vertebral length was correlated with age (r = 0.352, P = 0.024) whereas cord length and their ratio were not (P > 0.05). Compared with female, male had significantly larger cervical AP and TS, longer cervical cord (P < 0.01), higher position of conus medullaris (P < 0.05).
Conclusion
MRI is a useful tool for assessment of the whole spine. The longitudinal and cross‐sectional morphology of spinal cord in healthy Chinese adolescents may benefit further study of spine cord in adolescent idiopathic scoliosis as well as in other spine diseases.
Keywords: Adolescent, Magnetic resonance imaging, Spine cord, Whole‐spine
Introduction
Magnetic resonance imaging (MRI) is widely used in evaluating the spinal cord in many spine diseases1, 2, 3. In review of the published literature, we found that there have been no MRI data concerning healthy adolescents to date, such as total spine length, total vertebral length, and cross‐sectional morphology of the spine.
Our aim was 1) to determine the cross‐sectional and sagittal‐sectional morphology of the spinal cord and spine, to correlate them with different gender and age groups, 2) to establish our database of whole spine parameters by MRI in Chinese healthy adolescents. Study of craniocervical junction and spine is important to spinal surgeons and may reveal pathogenesis of some diseases4, 5.
Materials and Methods
Subjects
The study included 41 normal adolescents aged 11 to 17 years (18 boys, 23 girls), randomly recruited from local schools. All subjects were carefully assessed clinically to rule out any known abnormalities of the spine and spinal cord. If MRI examination of the volunteer had demonstrated abnormality of the spine or spine cord, this subject would be excluded. Ethical approval and informed consent from all subjects was obtained.
MRI Assessment
Whole‐spine MRI examination of the spine and hindbrain was performed in all subjects using a 1.5‐T MR Scanner (Philips, Holland) with a spine array coil. Sagittal scans of the whole spine from foramen magnum to sacrum and transverse (TS) image of the vertebral column were obtained. Scan parameters as follows: (i) sagittal, T1WI: TR/TE = 400/8 ms, Turbo factor = 5, NSA = 4, T2WI: TR/TE = 3000/120 ms, Turbo factor = 52, NSA = 6; (ii) axial, T2WI: TR/TE 3000/120 ms, Turbo factor = 24, NSA = 4; (iii) MRM: TR/TE = 8000/1000 ms, Turbo factor = 250, NSA = 2. All the measurements were performed on a workstation (EasyVision, Philips Medical Systems, Best, Holland), which allowed simultaneous display of the sagittal, coronal, and axial views of the vertebral column. Cross‐sectional measurement of the cord was made at the midpedicle level on a true axial plane, which was parallel to the superior endplate of that particular vertebra.
Firstly, at the selected cross‐sectional plane of each cervical and thoracic vertebra, the longest axis of the cord was taken for measuring the maximum TS diameter of the cord, while the largest AP diameter of the cord was measured along a perpendicular axis to the TS diameter. The AP/TS ratio was then calculated. The areas of spinal cord and vertebral canal were also measured at the same level using the best fit elliptical shape tool available in the workstation program (Fig. 1).
Figure 1.
The longest axis of the cord is taken for measuring the maximum transverse (TS) diameter of the cord, while the largest anteroposterior (AP) diameter of the cord is measured along a perpendicular axis to the transverse diameter. The AP/TS ratio is then calculated. (A) Thoracic cord area (B) Thoracic vertebral area, configuration and position of subarachnoid signal (C) Location of conus medullaris (D) Cerebellar tonsillar level related to the basion‐opsithion (BO) line, odontoid process level related to the BO line, angle between axis of medulla oblongata and cervical cord (α), angle between axis of medulla oblongata and BO line (β).
The total vertebral length and total cord length were measured by using a distance‐measuring tool available in the workstation program. In brief, using the straightened best mid sagittal section of the vertebral column, length of vertebral column was measured from the tip of odontoid process (C2) down to the inferior endplate of L5 while length of cord was measured from the level of the tip of odontoid process (C2) down to the conus medullaris (Fig. 2). The tip of conus medullaris was identified and its location was recorded in relation to the upper, middle, or lower third of the adjacent vertebral body and disc, by the same method as described by Saifuddin et al.6 T12 upper 1/3 is marked as “0”, T12 middle 1/3 is marked as “1”, T12 lower 1/3 is marked as “2”, and T12 ‐L1 disc is marked as “3” accordingly (Fig. 1).
Figure 2.
At the straightened best mid sagittal section of the vertebral column, length of vertebral column was measured from the tip of odontoid process (C2) down to the inferior endplate of L5 while length of cord was measured from the endplate of L5 while length of cord was measured from the level of the tip of odontoid process (C2) down to the conus medullaris.
The level of the cerebellar tonsil relative to a reference line connecting the basion and opsithion (BO) line was measured according to the method described by Aboulezz et al.7 Distance measured from above the BO line was conventionally assigned as a positive value while distance measured below the BO line was assigned as a negative value. Similarly, distance of atlas odontoid process related to the BO line was measured. Additionally, the angle between the axis of the medulla oblongata and the cervical vertebra (α), the angle between the axis of the medulla oblongata and the BO line (β) was obtained (Fig. 1).
All the above measurements were taken on both T1WI and T2WI of the exact same scanning plane by two different investigators within an interval of 14 days. The interclass correlation coefficients of different parameters range from 0.943 to 0.976, indicating a very high interobserver reliability. Measurement of a phantom with known dimensions revealed a linear measurement precision of up to 0.1 mm to check validity.
Statistics Analysis
All the MRI measurements were expressed as mean value with standard deviation. Linear regression was performed to test the effect of age on various parameters. Analysis of variance and Bonferroni test was used to explore the association of different MRI parameters among different age groups. The χ2 test and t test were used to compare the different MR parameters between male and female groups. SPSS version 14.0 was used for all statistical analysis. Statistical significance was considered with P < 0.05.
Results
Spinal Cord Length, Vertebral Column Length, and Their Ratio
Spine cord length was 399.34 ± 28.29 mm, vertebral length was 529.49 ± 32.46 mm, and cord /vertebral length ratio was 0.75 ± 0.03, on average. Vertebral length was correlated with age (r = 0.352, P = 0.024) whereas cord length and their ratio were not (P > 0.05). After adjustment for age, there was no significant difference in spinal cord length, vertebral column length and their ratio between the male and female groups. Total cord length was correlated with standing height and seating height (P < 0.01). Total vertebral length was correlated with standing height (r = 0.774, P = 0.000) and seating height (r = 0.382, P = 0.014). Cord /vertebral length ratio was not correlated with standing or seating height (Tables 1 and 2).
Table 1.
Variant characteristics of spine and spine cord on MRI in healthy adolescents
| Characteristics | Maximum | Minimum | Mean | SD |
|---|---|---|---|---|
| Age (years) | 17.00 | 11.00 | 13.95 | 1.30 |
| Standing height (mm) | 170.00 | 145.00 | 159.98 | 6.21 |
| Seating height (mm) | 94.50 | 45.00 | 82.66 | 9.04 |
| Cervical AP (mm) | 8.33 | 4.42 | 6.63 | 0.64 |
| Cervical TS (mm) | 17.27 | 8.66 | 12.21 | 1.62 |
| Cervical AP/TS | 0.72 | 0.37 | 0.55 | 0.07 |
| Position of conus medullaris | T11 | L2 | T6 | 2.39 |
| Cerebellar tonsillar to BO line (mm) | 15.56 | −2.04 | 3.97 | 3.63 |
| Odontoid process to BO line (mm) | 13.33 | 1.56 | 5.39 | 1.94 |
| Medulla oblongata/cervical vertebra (α, °) | 35.00 | 4.00 | 19.88 | 6.13 |
| Medulla oblongata/BO line (β, °) | 84.00 | 58.00 | 68.66 | 6.21 |
| Cervical length (mm) | 143.33 | 97.78 | 117.28 | 8.79 |
| Thoracic length (mm) | 310.00 | 226.67 | 260.73 | 17.38 |
| Lumbar length (mm) | 200.45 | 127.78 | 151.47 | 12.85 |
| Total cord length (mm) | 469.99 | 347.78 | 399.34 | 28.29 |
| Total vertebral length (mm) | 612.22 | 474.43 | 529.49 | 32.46 |
| Cord/vertebral length ratio | 0.83 | 0.67 | 0.75 | 0.03 |
| Thoracic cord area (mm2) | 50.79 | 16.97 | 31.25 | 6.79 |
| Thoracic vertebral area (mm2) | 480.34 | 135.21 | 227.62 | 67.27 |
| Thoracic cord/vertebral area ratio | 0.98 | 0.08 | 0.17 | 0.14 |
Note: AP, anteroposterior; BO, basion‐opsithion line; TS, transverse.
Table 2.
Comparison between characteristics of MRI in male and female healthy adolescents (mean ± SD)
| Characteristics | Male (18 cases) | Female (23 cases) | t value | P‐value |
|---|---|---|---|---|
| Age (years) | 14.06 ± 1.16 | 13.87 ± 1.42 | 0.63 | 0.54 |
| Standing height (mm) | 160.20 ± 7.50 | 159.78 ± 5.16 | 0.39 | 0.70 |
| Seating height (mm) | 83.08 ± 10.74 | 82.33 ± 7.71 | 0.47 | 0.64 |
| Cervical AP (mm) | 6.82 ± 0.47 | 5.77 ± 0.73 | 5.32 | 0.00 |
| Cervical TS (mm) | 12.81 ± 1.71 | 11.75 ± 1.41 | 3.63 | 0.00 |
| Cervical AP/TS | 0.60 ± 0.07 | 0.56 ± 0.07 | 2.90 | 0.01 |
| Position of conus medullaris (Thoracic) | 5.28 ± 1.99 | 6.61 ± 2.55 | 2.50 | 0.02 |
| Cerebellar tonsillar to BO line (mm) | 4.48 ± 3.57 | 3.58 ± 3.70 | 1.17 | 0.25 |
| Odontoid process to BO line (mm) | 4.80 ± 1.53 | 5.85 ± 2.12 | 2.37 | 0.03 |
| Medulla oblongata/cervical vertebra (α, °) | 19.90 ± 6.52 | 19.83 ± 5.90 | 0.06 | 0.95 |
| Medulla oblongata/BO line (β, °) | 69.40 ± 6.87 | 68.09 ± 5.74 | 1.10 | 0.28 |
| Cervical length (mm) | 118.41 ± 11.57 | 116.40 ± 5.94 | 1.62 | 0.12 |
| Thoracic length (mm) | 259.07 ± 24.17 | 262.03 ± 9.66 | 1.47 | 0.16 |
| Lumbar length (mm) | 148.83 ± 11.28 | 153.54 ± 13.85 | 1.63 | 0.12 |
| Total cord length (mm) | 394.12 ± 34.51 | 403.43 ± 22.25 | 2.01 | 0.06 |
| Total vertebral length (mm) | 526.31 ± 42.57 | 531.97 ± 22.37 | 1.21 | 0.24 |
| Cord/vertebral length ratio | 0.75 ± 0.02 | 0.76 ± 0.04 | 1.14 | 0.27 |
| Thoracic cord area (mm2) | 32.33 ± 4.38 | 30.41 ± 8.20 | 1.12 | 0.27 |
| Thoracic vertebral area (mm2) | 206.48 ± 52.00 | 232.91 ± 77.92 | 1.63 | 0.12 |
| Thoracic cord/vertebral area ratio | 0.20 ± 0.04 | 0.14 ± 0.04 | 8.70 | 0.00 |
Note: AP, anteroposterior; BO, basion‐opsithion line; TS, transverse.
AP, TS, AP/TS Cord Ratio, Cord Area, Vertebral Canal Area and Their Ratio
In the cervical cord, the mean AP was 6.63 ± 0.64 mm, mean TS was 12.21 ± 1.62 mm and the AP/TS Cord Ratio was 0.55 ± 0.07, which were significantly different between the male and female groups. The thoracic cord area was 31.25 ± 6.79 mm2, the thoracic vertebral canal area was 227.62 ± 67.27 mm2 and the ratio of spinal cord area to vertebral canal area was 0.17 ± 0.14. There was no significant difference in cord area and vertebral canal area between the male and female groups, but their ratio was significantly different (P < 0.01, Tables 1 and 2).
Cerebellar Tonsillar Level from BO Line, Odontoid Process from BO Line, α and β Angle
Cerebellar tonsillar level was not correlated with age (P > 0.05). Forty students (97.56%) had the cerebellar tonsillar level above the BO line (median tonsillar position 3.73 mm above BO line) and one student had cerebellar tonsillar level below the BO line(‐2.044 mm). Odontoid process from BO line was 5.386 mm on average and there was significant difference between male and female groups. The mean value of α and β was 19.88° ± 6.13° and 68.66° ± 6.21°, respectively, there was no difference between the male and female groups, and they were not correlated with age.
Position of Conus Medullaris
The conus medullaris was found to terminate at the level of L1 lower 1/3 on average, ranging from T12 lower 1/3 to L2 disc. Details of the location of conus are given in Table 4. There was a significant difference in the position of the conus medullaris between male and female groups (t = 2.499, P = 0.020). The position of the conus medullaris was not correlated with age, but there was significant difference between one group (13 ≤ age ≤ 15) and another group (age > 15) (One‐Way ANOVA, P < 0.05; Tables 3 and 4).
Table 4.
Position of conus medullaris on MRI in healthy adolescents with different age [cases (%)]
| Positions | Total (41 cases) | Age < 13 (15 cases) | 13 ≤ Age ≤ 15 (15 cases) | Age > 15 (11 cases) |
|---|---|---|---|---|
| T12 lower 1/3 | 3 (7.3) | 1 (6.7) | 2 (13.3) | 0 (0.0) |
| T12L1 disc | 4 (9.8) | 2 (13.3) | 2 (13.3) | 0 (0.0) |
| L1 upper 1/3 | 5 (12.2) | 1 (6.7) | 3 (20.0) | 1 (9.1) |
| L1 middle 1/3 | 5 (12.2) | 2 (13.3) | 1 (6.7) | 2 (18.2) |
| L1 lower 1/3 | 5 (12.2) | 1 (6.7) | 3 (20.0) | 1 (9.1) |
| L1–2 disc | 10 (24.4) | 4 (26.7) | 3 (20.0) | 3 (27.3) |
| L2 upper 1/3 | 2 (4.9) | 1 (6.7) | 0 (0.0) | 1 (9.1) |
| L2 middle 1/3 | 4 (9.8) | 1 (6.7) | 1 (6.7) | 2 (18.2) |
| L2 lower 1/3 | 1 (2.4) | 0 (0.0) | 0 (0.0) | 1 (9.1) |
| L2–3 disc | 2 (4.9) | 2 (13.3) | 0 (0.0) | 0 (0.0) |
| Mean level | L1 lower 1/3 | L1 lower 1/3 | L1 middle 1/3* | L1–2 disc* |
*LSC (ANOVA) test (P < 0.05).
Table 3.
Comparison between characteristics of MRI in healthy adolescents with different age (mean ± SD)
| Characteristics | Age < 13 (15 cases) | 13 ≤ Age ≤ 15 (15 cases) | Age > 15 (11 cases) |
|---|---|---|---|
| Standing height (mm) | 159.33 ± 7.40 | 159.60 ± 5.30 | 161.36 ± 5.95 |
| Seating height (mm) | 81.63 ± 9.67 | 81.33 ± 10.63 | 85.86 ± 4.78 |
| Cervical AP (mm) | 6.45 ± 0.45* | 6.91 ± 0.65* | 6.48 ± 0.77 |
| Cervical TS (mm) | 11.93 ± 0.66 | 12.37 ± 1.50 | 12.38 ± 2.55 |
| Cervical AP/TS | 0.54 ± 0.03 | 0.56 ± 0.08 | 0.54 ± 0.10 |
| Position of medullary conus (Thoracic) | 6.33 ± 2.74 | 5.00 ± 2.07* | 7.00 ± 1.90* |
| Cerebellar tonsillar to BO line (mm) | 4.67 ± 4.47 | 3.78 ± 3.03 | 3.29 ± 3.26 |
| Odontoid process to BO line (mm) | 5.63 ± 2.54 | 5.43 ± 1.78 | 4.99 ± 1.11 |
| Medulla oblongata/cervical vertebra (α, °) | 19.40 ± 6.31 | 19.73 ± 5.75 | 20.73 ± 6.84 |
| Medulla oblongata/BO line (β, °) | 68.80 ± 7.20 | 67.40 ± 4.78 | 70.18 ± 6.69 |
| Total cervical length (mm) | 109.18 ± 8.30* | 111.03 ± 7.86 | 116.78 ± 9.69* |
| Total thoracic length (mm) | 257.33 ± 11.18 | 258.00 ± 20.29 | 269.09 ± 18.73 |
| Total lumbar length (mm) | 145.99 ± 8.52 | 154.25 ± 15.68 | 155.15 ± 11.98 |
| Total cord length (mm) | 390.01 ± 27.32 | 385.76 ± 21.35* | 410.52 ± 32.03* |
| Total vertebral length (mm) | 512.51 ± 23.37* | 523.28 ± 33.28 | 541.02 ± 37.19* |
| Cord/vertebral length ratio | 0.76 ± 0.03* | 0.74 ± 0.03* | 0.76 ± 0.03 |
| Thoracic cord area (mm2) | 32.90 ± 8.55 | 30.11 ± 6.68 | 30.55 ± 3.59 |
| Thoracic vertebral area (mm2) | 249.35 ± 86.77 | 208.62 ± 41.15 | 223.90 ± 62.67 |
| Thoracic cord/vertebral area ratio | 0.14 ± 0.03 | 0.15 ± 0.05 | 0.15 ± 0.04 |
*LSC (ANOVA) test (P < 0.05).
Discussion
Despite extensive research over decades of intraspinal abnormalities, there has been no comprehensive MRI study of healthy Chinese students. Recently, MR imaging of the whole spine makes it feasible to measure spinal cord length in scoliosis patients. Chu et al. reported a significantly reduced cord to vertebral length ratio was found in 14 adolescent idiopathic scoliosis (AIS) subjects with a severe curve when compared with 14 age and sex‐matched controls2. The above was contributed mainly by the lengthening of thoracic segment of the vertebral column whereas the absolute spinal cord length did not show equal proportion of lengthening in AIS. So, to establish our database of whole spine parameters by MRI in Chinese healthy adolescents is really necessary.
The position of the conus medullaris is important in the diagnosis of tethered spinal cord8. In our study, the position of the conus medullaris was found to terminate at the level of L1 lower 1/3 on average, ranging from T12 lower 1/3 to L2 disc. There was a significant difference in the position of the conus among different age groups and gender groups. Saifuddin6 reported that the conus medullaris ranged from T12 middle 1/3 to L3 upper1/3, and mean level is L1 lower 1/3 in 690 healthy adults. Thomson9 and Needles10 found the position of the conus medullaris was lower in females than in males, according to their study on corpses. However, Soleiman et al.11 found the mean level of conus medullaris is L1 middle 1/3 (main age, 20–80 years).
The relative position of the cerebellar tonsillar level to the BO line is very important in diagnosis of Chiari malformation12. In our study, 40 students (97.56%) had a cerebellar tonsillar level above the BO line. Kim et al. had a similar finding. The cerebellar tonsillar level was not correlated with age (linear regression, P > 0.05)13. Ishikawa et al.14 reported that in all of 50 healthy Japanese subjects, the cerebellar tonsils were located at or above the level of the foramen magnum. Samuelsson et al.15 performed spine MRI in 26 patients with idiopathic scoliosis. In all four with tonsillar ectopia, the tonsils were less than 5 mm below the foramen magnum, but two were associated with syringomyelia. Mikulis et al.16 described the variation of the level of normal tonsils with age and noted that in younger people the level is significantly higher than in older adults.
Disproportionate neuro‐osseous growth might explain the pathogenesis of adolescent scoliosis17. Lord et al.18 reported that the vertebral epiphyses close earlier in the upper region than in the lower thoracic spine and the major component of adolescent longitudinal spinal growth takes place in the lower thoracic spine between T5 and T10. The establishment of a normal adolescent database is beneficial to such studies.
Although the sample capacity is limited, to the best of our knowledge, this is the first report of whole spine and spine column examined by MRI in healthy Chinese adolescents. In the literature, MRI studies of the normal spine mainly investigate adult populations and focus on part of the spine19, 20, 21, 22, 23, 24. In the future, we will enlarge the sample of healthy students and try the kinetic MRI for more data, which will definitely be beneficial to the abnormal MRI study of the juvenile spine.
MRI is a useful tool for assessment of the whole spine. To the best of our knowledge, this is the first report of whole spine and spine column examined by MRI in healthy Chinese adolescents.
The study of longitudinal and cross‐sectional morphology of spinal cord in healthy Chinese adolescents may benefit further study of the spinal cord in adolescent idiopathic scoliosis as well as in other spine diseases.
Disclosure: No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.
References
- 1. Liu T, Chu WC, Young G, et al MR analysis of regional brain volume in adolescent idiopathic scoliosis: neurological manifestation of a systemic disease. J Magn Reson Imaging, 2008, 27: 732–736. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Chu WC, Lam WW, Chan YL, et al Relative shortening and functional tethering of spinal cord in adolescent idiopathic scoliosis?: study with multiplanar reformat magnetic resonance imaging and somatosensory evoked potential. Spine (Phila Pa 1976), 2006, 31: E19–E25. [DOI] [PubMed] [Google Scholar]
- 3. Macdonald A, Chatrath P, Spector T, Ellis H. Level of termination of the spinal cord and the dural sac: a magnetic resonance study. Clin Anat, 1999, 12: 149–152. [DOI] [PubMed] [Google Scholar]
- 4. Chu WC, Man GC, Lam WW, et al A detailed morphologic and functional magnetic resonance imaging study of the craniocervical junction in adolescent idiopathic scoliosis. Spine (Phila Pa 1976), 2007, 32: 1667–1674. [DOI] [PubMed] [Google Scholar]
- 5. Chu WC, Man GC, Lam WW, et al Morphological and functional electrophysiological evidence of relative spinal cord tethering in adolescent idiopathic scoliosis. Spine (Phila Pa 1976), 2008, 33: 673–680. [DOI] [PubMed] [Google Scholar]
- 6. Saifuddin A, Burnett SJ, White J. The variation of position of the conus medullaris in an adult population. Spine (Phila Pa 1976), 1998, 23: 1452–1456. [DOI] [PubMed] [Google Scholar]
- 7. Aboulezz AO, Sartor K, Geyer CA, Gado MH. Position of cerebellar tonsils in the normal population and in patients with Chiari malformation: a quantitative approach with MR imaging. J Comput Assist Tomogr, 1985, 9: 1033–1036. [DOI] [PubMed] [Google Scholar]
- 8. Sun X, Chu WC, Cheng JC, et al Do adolescents with a severe idiopathic scoliosis have higher locations of the conus medullaris than healthy adolescents? J Pediatr Orthop, 2008, 28: 669–673. [DOI] [PubMed] [Google Scholar]
- 9. Thomson A. Fifth annual report of the committee of collective investigation of the Anatomical Society of Great Britain and Ireland for the year 1893–94. J Anat Physiol, 1894, 29: 35–60. [PMC free article] [PubMed] [Google Scholar]
- 10. Needles JH. The caudal level of termination of the spinal cord in American whites and American negroes. Anat Rec, 1935, 63: 417–424. [Google Scholar]
- 11. Soleiman J, Demaerel P, Rocher S, Maes F, Marchal G. Magnetic resonance imaging study of the level of termination of the conus medullaris and the thecal sac: influence of age and gender. Spine (Phila Pa 1976), 2005, 30: 1875–1880. [DOI] [PubMed] [Google Scholar]
- 12. Sun X, Qiu Y, Zhu Z, et al Variations of the position of the cerebellar tonsil in idiopathic scoliotic adolescents with a cobb angle >40 degrees: a magnetic resonance imaging study. Spine (Phila Pa 1976), 2007, 32: 1680–1686. [DOI] [PubMed] [Google Scholar]
- 13. Kim JT, Bahk JH, Sung J. Influence of age and sex on the position of the conus medullaris and Tuffier's line in adults. Anesthesiology, 2003, 99: 1359–1363. [DOI] [PubMed] [Google Scholar]
- 14. Ishikawa M, Kikuchi H, Fujisawa I, Yonekawa Y. Tonsillar herniation on magnetic resonance imaging. Neurosurgery, 1988, 22: 77–81. [DOI] [PubMed] [Google Scholar]
- 15. Samuelsson L, Lindell D, Kogler H. Spinal cord and brain stem anomalies in scoliosis. MR screening of 26 cases. Acta Orthop Scand, 1991, 62: 403–406. [DOI] [PubMed] [Google Scholar]
- 16. Mikulis DJ, Diaz O, Egglin TK, Sanchez R. Variance of the position of the cerebellar tonsils with age: preliminary report. Radiology, 1992, 183: 725–728. [DOI] [PubMed] [Google Scholar]
- 17. Lao LF, Shen JX, Chen ZG, Wang YP, Wen XS, Qiu GX. Uncoupled neuro‐osseous growth in adolescent idiopathic scoliosis? A preliminary study of 90 adolescents with whole‐spine three‐dimensional magnetic resonance imaging. Eur Spine J, 2011, 20: 1081–1086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lord MJ, Ogden JA, Ganey TM. Postnatal development of the thoracic spine. Spine (Phila Pa 1976), 1995, 20: 1692–1698. [DOI] [PubMed] [Google Scholar]
- 19. Shi L, Heng PA, Wong TT, Chu WC, Yeung BH, Cheng JC. Morphometric analysis for pathological abnormality detection in the skull vaults of adolescent idiopathic scoliosis girls. Med Image Comput Comput Assist Interv, 2006, 9: 175–182. [DOI] [PubMed] [Google Scholar]
- 20. Jinkins JR, Dworkin JS, Damadian RV. Upright, weight‐bearing, dynamic‐kinetic MRI of the spine: initial results. Eur Radiol, 2005, 15: 1815–1825. [DOI] [PubMed] [Google Scholar]
- 21. Gupta P, Lenke LG, Bridwell KH. Incidence of neural axis abnormalities in infantile and juvenile patients with spinal deformity. Is a magnetic resonance image screening necessary? Spine (Phila Pa 1976), 1998, 23: 206–210. [DOI] [PubMed] [Google Scholar]
- 22. Liljenqvist UR, Allkemper T, Hackenberg L, Link TM, Steinbeck J, Halm HF. Analysis of vertebral morphology in idiopathic scoliosis with use of magnetic resonance imaging and multiplanar reconstruction. J Bone Joint Surg Am, 2002, 84: 359–368. [DOI] [PubMed] [Google Scholar]
- 23. Liljenqvist UR, Link TM, Halm HF. Morphometric analysis of thoracic and lumbar vertebrae in idiopathic scoliosis. Spine (Phila Pa 1976), 2000, 25: 1247–1253. [DOI] [PubMed] [Google Scholar]
- 24. Demiryürek D, Aydingöz U, Akşit MD, Yener N, Geyik PO. MR imaging determination of the normal level of conus medullaris. Clin Imaging, 2002, 26: 375–377. [DOI] [PubMed] [Google Scholar]