Abstract
Aims
Developmental cervical spinal stenosis (DcSS) is a well-known predisposing factor for degenerative cervical myelopathy (DCM) but there is a lack of consensus on its definition. This study aims to define DcSS based on MRI, and its multilevel characteristics, to assess the prevalence of DcSS in the general population, and to evaluate the presence of DcSS in the prediction of developing DCM.
Methods
This cross-sectional study analyzed MRI spine morphological parameters at C3 to C7 (including anteroposterior (AP) diameter of spinal canal, spinal cord, and vertebral body) from DCM patients (n = 95) and individuals recruited from the general population (n = 2,019). Level-specific median AP spinal canal diameter from DCM patients was used to screen for stenotic levels in the population-based cohort. An individual with multilevel (≥ 3 vertebral levels) AP canal diameter smaller than the DCM median values was considered as having DcSS. The most optimal cut-off canal diameter per level for DcSS was determined by receiver operating characteristic analyses, and multivariable logistic regression was performed for the prediction of developing DCM that required surgery.
Results
A total of 2,114 individuals aged 64.6 years (SD 11.9) who underwent surgery from March 2009 to December 2016 were studied. The most optimal cut-off canal diameters for DcSS are: C3 < 12.9 mm, C4 < 11.8 mm, C5 < 11.9 mm, C6 < 12.3 mm, and C7 < 13.3 mm. Overall, 13.0% (262 of 2,019) of the population-based cohort had multilevel DcSS. Multilevel DcSS (odds ratio (OR) 6.12 (95% CI 3.97 to 9.42); p < 0.001) and male sex (OR 4.06 (95% CI 2.55 to 6.45); p < 0.001) were predictors of developing DCM.
Conclusion
This is the first MRI-based study for defining DcSS with multilevel canal narrowing. Level-specific cut-off canal diameters for DcSS can be used for early identification of individuals at risk of developing DCM. Individuals with DcSS at ≥ three levels and male sex are recommended for close monitoring or early intervention to avoid traumatic spinal cord injuries from stenosis.
Cite this article: Bone Joint J 2024;106-B(11):1333–1341.
Take home message
Clinically recommended cut-off anteroposterior spinal canal diameters for developmental cervical spinal stenosis (DcSS) are: C3 < 13 mm, C4 < 12 mm, C5 < 12 mm, C6 < 12.5 mm, C7 < 13.5 mm. These can be used as the criteria for detecting DcSS in individuals at risk of developing degenerative cervical myelopathy (DCM).
Individuals with multilevel (≥ 3 vertebral levels) cervical spinal canal narrowing and of male sex are at high risk of developing DCM.
Early intervention or prophylactic surgery can be considered for these individuals to avoid traumatic spinal cord injuries.
Introduction
Developmental cervical spinal stenosis (DcSS) is a congenital condition associated with genetic disturbance with pre-existing narrowing of the cervical spinal canal.1-3 DcSS is considered as an important risk factor for degenerative cervical myelopathy (DCM),1,2,4-9 which can result from various forms of degenerative pathologies including cervical spondylosis, posterior longitudinal ligament ossification, and ligamentum flavum hypertrophy.10 Patients with DcSS are more susceptible to DCM, as the narrowed dimension of cervical spinal canal predisposes patients to spinal cord compression even with a mild degree of degeneration, whereas individuals without canal narrowing may not develop DCM with degeneration of the same extent.1,2,8,11 Cervical canal narrowing of DcSS can be found on medical imaging even before the onset of symptoms. Therefore, it is ideal to identify these patients with a high risk of spinal cord compression early, for consideration of prophylactic surgery or posterior canal widening in order to prevent severe cord injury.12
Currently, there is a lack of image-based definition of DcSS representing the multilevel characteristics of developmental spinal stenosis. Generalized narrowing of the spinal canal from developmental spinal stenosis has implications for the likelihood of developing symptoms and early surgery.13,14 Lai et al13,15-17 showed that in the lumbar spine, developmental spinal stenosis led to multilevel stenosis because the abnormal spinal canal development affected multiple vertebral levels rather than a single level. Given that these anomalies occur during neural tube development, these phenotypes should also be identified in the cervical spine, but existing DcSS definitions have not taken into consideration the nature of multilevel involvement of vertebral stenosis.1,8,18-20 MRI is the gold-standard imaging modality for DCM, as both osseous structures and soft-tissues are clearly seen, thereby enabling direct sagittal diameter measurement of the cervical spinal canal, vertebral body, and spinal cord, as well as assessment of ligamentum hypertrophy and any associated intervertebral disc herniation.2,6,10 Several studies have reported normal MRI cervical spine morphology in population cohorts,2,21-25 while some attempted to define DcSS based on generalized but not level-specific measurement at individual vertebral levels.2,8 Among those studies proposing DcSS cut-off values, study populations were heterogeneous with various methodological and imaging methods, ranging from cadaveric studies to lateral radiograph assessment, and from CT to MRI.4,8,18,19,23,26-30 There is, however, a lack of consensus on the definition of DcSS, with clinical implications for the management of patients with DCM.31-33
This study aims to: define the developmental narrowing of the cervical spinal canal based on MRI, and its multilevel characteristics; assess how common cervical developmental canal narrowing is in the local general population; and evaluate the presence of multilevel DcSS in predicting the development of DCM which requires surgical intervention. The study provides population-based cut-off values of spinal canal diameter for detecting DcSS in the Chinese population.
Methods
The study included a total of 2,114 individuals, consisting of 95 patients who underwent surgery for DCM from March 2009 to December 2016 at a single tertiary spine centre and 2,019 individuals openly recruited from the local general population (Figure 1). All individuals included had to be aged 18 years or above, and of Chinese ethnicity. The patients with DCM were diagnosed by spinal surgeons based on clinical assessment of symptoms along with MRI evidence. Clinical information and MRI of the DCM patients were retrieved from the local hospital database. Patients with cervical spine fractures or tumour, infection, rheumatoid arthritis, congenital cervical spinal deformities, previous cervical spine surgery, spinal cord injury due to other causes including trauma or traffic accidents, hyperflexion, or hyperextension were excluded. For the population-based cohort, individuals were openly recruited via news articles and emails, with the invitation to undergo MRI and clinical assessment at the time of recruitment for the screening of genomics in relation to degenerative skeletal disorders (Theme-based Research Scheme (T12-708/12 N) and Area of Excellence Scheme (AoE/M-04/04)). Individuals with a diagnosis or history of cervical myelopathy, cervical spine fractures, tumours, infections, congenital pathologies, or previous cervical spine surgeries were excluded. Exclusion criteria also included MRI images of poor quality. Ethical approval from the local institutional review board and informed consent from participants were obtained.
Fig. 1
Flowchart of patient recruitment.
MRI protocol
For both DCM patients and population-based cohort participants, 3-Tesla HD MRI machines were used for imaging. For DCM patients, only preoperative MRI scans were used for measurement. These scans were taken with slice thickness and slice spacing of 3 mm and 0 mm, respectively. Field of view was 24 cm × 24 cm. The imaging matrix was 512 × 512. The repetition time and echo time were 3,000 to 4,000 ms and 80 ms to 90 ms, respectively. For MRI of the general population cohort, the field of view was 30 cm × 30 cm, with 4 mm slice thickness and 0.4 mm slice spacing. The imaging matrix was 448 × 336. The repetition time and echo time were 3,000 to 4,000 ms and 80 ms to 120 ms, respectively.
MRI measurement
A standardized method of measurement was used by the two investigators (JHML, VWYL) who performed the measurement independently, and they were blinded to all clinical information. T2-weighted sagittal slices of the cervical spine were selected for measurements. All measurements were performed on the median slice with the apparent largest spinous process, using eUnity Diagnostic Viewer v. 6.10.2 (Canada) and Philips DICOM Viewer R3.0-SP15 (Philips, Netherlands).
The anteroposterior (AP) diameter of the cervical spinal canal, vertebral body, and spinal cord was measured at mid-vertebral level from C3 to C7. The AP diameter of the cervical spinal canal was measured by a line perpendicular to the anterior surface of the spinal cord connecting the midpoint of the superior and inferior endplate of the posterior surface of the vertebral body to the spinolaminar line. The AP diameter of the cervical spinal cord was measured by the same line used for measuring the AP diameter of the cervical canal. The AP diameter of the vertebral body was measured by a line from the midpoint of the anterior and posterior margin of the vertebral body (Figure 2).34 To represent the relationships between the cervical spinal canal, vertebral body, and spinal cord, the canal-body ratio (AP diameter of cervical spinal canal divided by AP diameter of cervical vertebral body), space available for cord (AP diameter of cervical spinal canal minus AP diameter of cervical spinal cord), cord-canal ratio (AP diameter of cervical spinal cord divided by AP diameter of cervical spinal canal), and cord-vertebral body ratio (AP diameter of cervical spinal cord divided by AP diameter of cervical vertebral body) were calculated.8,21
Fig. 2
Measurement on T2-weighted sagittal image: a) anteroposterior (AP) diameter of cervical spinal canal; b) mid-vertebral AP diameter of vertebral body; c) AP diameter of cervical spinal cord (measured on the same line for a)).
Inter- and intrarater reliabilities were assessed by Cronbach α analysis. The first and second rounds of measurement of randomly selected MRI slices with 50 measurements per parameter were performed four weeks apart. An α value of 0.90 to 1.00 and 0.80 to 0.90 indicated excellent and good reliability, respectively.35 Good to excellent inter-rater reliability (α = 0.84 to 0.97) and excellent intrarater reliability (α = 0.91 to 0.99) were found.
Statistical analysis
Descriptive and frequency statistics were presented for demographics and spine morphological measurement (AP diameter of cervical spinal canal, spinal cord, vertebral body, canal-body ratio, space available for the cord, cord-canal ratio, cord-vertebral body ratio). Data normality was tested.
In order to represent the extent of cervical canal narrowing resulting in the development of DCM, spinal canal sizes of the cohort of DCM patients (n = 95) were used as reference for assessing whether stenotic vertebral levels were present in an individual. At each vertebral level of C3 to C7, the median AP spinal canal diameter of DCM patients was first found, and this was considered appropriate for defining a vertebral level as stenotic, since the median value reflected the narrowest 50% AP spinal canal diameter of DCM patients. Each individual in the general population cohort (n = 2,019) was then examined if their spinal canal diameter was below the level-specific median value at each vertebral level. To address the nature of multilevel involvement, a composite score for DcSS was established for each population-based individual with a full score of 5, consisting of a score of 1 for each stenotic level. A composite score of ≥ 3, meaning the AP spinal canal diameter was narrower than the DCM median values at three or more vertebral levels, indicated the individual was likely to have multilevel pre-existing bony canal narrowing. This definition also relates to the surgical relevance of a two-level decompression corresponding to three vertebral levels. This stratified the general population cohort into DcSS versus non-DcSS, with prevalence rate reported in percentage.
This differentiation of population-based cohort between DcSS and non-DcSS was examined through intergroup comparison of cervical spine morphologies using Mann-Whitney U tests. Receiver operating characteristic (ROC) analysis was performed to determine the optimal cut-off AP spinal canal diameter at each level from C3 to C7 for detecting DcSS, with the highest Youden’s Index, sensitivity, and specificity at each vertebral level.36 An area under the curve (AUC) of 0.7 to 0.8, 0.8 to 0.9, or ≥ 0.9 was considered acceptable, excellent, or outstanding, respectively, while 0.5 indicated no discriminating power.37 Multivariable logistic regression was used for the prediction of developing DCM requiring surgical intervention with the presence of DcSS (Yes/No, based on canal diameter less than the newly defined cut-off values at ≥ 3 levels) as a factor, together with patient demographic data.
A p-value < 0.05 was considered statistically significant. All statistical analyses were performed by SPSS Statistics v.28 (IBM, USA) and G*Power v.3.1.9.7. (Heinrich-Heine-Universität Düsseldorf, Germany).
Results
A total of 2,114 individuals were studied, comprising 95 DCM patients (68 males and 27 females) and 2,019 cohort participants (780 males and 1,239 females), with a mean age of 64.6 years (SD 11.9) and 50.1 years (SD 9.6) for males and females, respectively. Baseline characteristics of the DCM patients and population-based cohorts are presented in Table I. The causes of DCM were: ossification of the posterior longitudinal ligament (n = 18), yellow ligament thickening with and without disc-osteophyte complex (n = 2), and cervical spondylotic myelopathy (n = 75, which consisted of soft disc with or without disc bulging (n = 2), ligamentum flavum hypertrophy (n = 17), and disc bulging/extrusion/protrusion, disc herniation, disc prolapse, or disc-osteophyte complex (n = 56)). In Table II, the median AP canal diameters for DCM patients were: 12.0 mm for C3, 11.2 mm for C4, 11.8 mm for C5, 12.7 mm for C6, and 14.1 mm for C7. By using these median values for screening, 12.2% (247 of 2,019) had a composite score of ≥ 3. At the individual level, 27.9% of the general population cohort had canal narrowing at C7, and 18.5% at C6. C5 to C7 was the cervical segment with highest percentage of consecutive multilevel canal narrowing (11.5%, 233 of 2,019) (Table II).
Table I.
Characteristics of the general population and degenerative cervical myelopathy cohorts.
| Parameters | General population cohort | DCM cohort | ||||
|---|---|---|---|---|---|---|
| Total | Male | Female | Total | Male | Female | |
| Sex, n (%) | 2,019 | 780 (38.6) | 1,239 (61.4) | 95 | 68 (71.2) | 27 (28.4) |
| Mean age at recruitment, yrs (SD) | 50.1 (9.6) | 64.6 (11.9) | ||||
| Mean height, m (SD) | 1.70 (0.06) | 1.57 (0.06) | 1.66 (0.06) | 1.54 (0.04) | ||
| Mean age at surgery, yrs (SD) | N/A | 64.9 (11.9) | ||||
| Operated levels, n (%) | ||||||
| C3 to 4 | N/A | 12 (12.6) | ||||
| C3 to 5 | N/A | 5 (5.3) | ||||
| C3 to 6 | N/A | 39 (41.1) | ||||
| C3 to 7 | N/A | 5 (5.3) | ||||
| C4 to 5 | N/A | 13 (13.7) | ||||
| C4 to 6 | N/A | 6 (6.3) | ||||
| C4 to 7 | N/A | 1 (1.1) | ||||
| C5 to 6 | N/A | 13 (13.7) | ||||
| C6 to 7 | N/A | 1 (1.1) | ||||
| Mean AP diameter of cervical spinal canal, mm (SD) | ||||||
| Vertebral level | ||||||
| C3 | 13.7 (1.3) | 13.6 (1.2) | 13.8 (1.4) | 12.0 (1.2) | 11.7 (1.2) | 12.1 (1.1) |
| C4 | 13.3 (1.4) | 13.3 (1.4) | 13.3 (1.4) | 11.4 (1.1) | 11.2 (1.1) | 11.4 (1.1) |
| C5 | 13.6 (1.5) | 13.5 (1.5) | 13.6 (1.5) | 11.8 (1.2) | 12.0 (1.1) | 11.8 (1.3) |
| C6 | 14.1 (1.6) | 14.0 (1.6) | 14.1 (1.6) | 12.7 (1.6) | 12.9 (1.5) | 12.7 (1.6) |
| C7 | 14.9 (1.5) | 14.9 (1.5) | 15.0 (1.5) | 14.2 (1.2) | 14.2 (1.2) | 14.2 (1.2) |
| Mean mid-vertebral AP diameter of vertebral body, mm (SD) | ||||||
| C3 | 14.7 (1.5) | 16.2 (1.6) | ||||
| C4 | 14.5 (1.7) | 16.3 (1.7) | ||||
| C5 | 14.2 (1.9) | 16.3 (1.7) | ||||
| C6 | 14.4 (1.8) | 16.5 (1.8) | ||||
| C7 | 14.4 (1.6) | 16.1 (1.7) | ||||
| Mean AP diameter of cervical spinal cord, mm (SD) | ||||||
| C3 | 7.5 (0.6) | 6.2 (0.8) | ||||
| C4 | 7.2 (0.6) | 6.1 (1.0) | ||||
| C5 | 7.0 (0.6) | 5.9 (0.9) | ||||
| C6 | 6.8 (0.5) | 5.8 (0.8) | ||||
| C7 | 6.4 (0.5) | 5.5 (0.6) | ||||
| Mean canal-body ratio (SD) | ||||||
| C3 | 0.94 (0.13) | 0.75 (0.12) | ||||
| C4 | 0.92 (0.11) | 0.71 (0.11) | ||||
| C5 | 0.96 (0.13) | 0.73 (0.12) | ||||
| C6 | 0.99 (0.13) | 0.78 (0.14) | ||||
| C7 | 1.05 (0.12) | 0.89 (0.11) | ||||
| Mean space available for cord, mm (SD) | ||||||
| C3 | 6.1 (1.4) | 5.8 (1.2) | ||||
| C4 | 6.1 (1.4) | 5.3 (1.2) | ||||
| C5 | 6.5 (1.6) | 5.9 (1.3) | ||||
| C6 | 7.2 (1.6) | 7.0 (1.6) | ||||
| C7 | 8.6 (1.5) | 8.8 (1.2) | ||||
| Mean cord-canal ratio (SD) | ||||||
| C3 | 0.56 (0.07) | 0.52 (0.07) | ||||
| C4 | 0.55 (0.07) | 0.54 (0.09) | ||||
| C5 | 0.53 (0.07) | 0.50 (0.08) | ||||
| C6 | 0.49 (0.07) | 0.46 (0.08) | ||||
| C7 | 0.43 (0.05) | 0.39 (0.04) | ||||
| Mean cord-vertebral body ratio (SD) | ||||||
| C3 | 0.52 (0.07) | 0.38 (0.06) | ||||
| C4 | 0.50 (0.07) | 0.38 (0.07) | ||||
| C5 | 0.50 (0.08) | 0.36 (0.07) | ||||
| C6 | 0.48 (0.07) | 0.35 (0.06) | ||||
| C7 | 0.45 (0.06) | 0.34 (0.05) | ||||
-
AP, anteroposterior; DCM, degenerative cervical myelopathy; N/A, not applicable.
Table II.
Composite score distribution and multilevel canal narrowing in the general population-based cohort.
| Vertebral level | C3 | C4 | C5 | C6 | C7 |
|---|---|---|---|---|---|
| DCM cohort | |||||
| Mean spinal canal diameter, mm (SD) | 12.0 (1.2) | 11.4 (1.1) | 11.8 (1.2) | 12.7 (1.6) | 14.2 (1.2) |
| Median spinal canal diameter, mm (IQR) | 12.0 (11.3 to 12.7) | 11.2 (10.6 to 12.1) | 11.8 (10.9 to 12.6) | 12.7 (11.4 to 13.9) | 14.1 (13.5 to 15.1) |
| General population cohort below median DCM values, n (%) | 176 (8.7) | 119 (5.9) | 233 (11.5) | 374 (18.5) | 563 (27.9) |
| Distribution of composite scores in general population cohort, n (%)* | |||||
| 0 | 1,345 (66.6 | ||||
| 1 | 279 (13.8) | ||||
| 2 | 148 (7.3) | ||||
| 3 | 117 (5.8) | ||||
| 4 | 111 (5.5) | ||||
| 5 | 19 (0.9) | ||||
| Location of consecutive multilevel narrowing, n (%) | |||||
| C5 to C7 | 233 (11.5) | ||||
| C4 to C6 | 119 (5.9) | ||||
| C3 to C5 | 19 (0.9) | ||||
-
*
i.e. number of narrow canals.
-
DCM, degenerative cervical myelopathy.
Given the comparable age and sex of individuals with and without DcSS, intergroup comparison revealed that the DcSS group had a significantly smaller AP canal diameter than the non-DcSS group at all measured levels (all p < 0.001) (Table III). The DcSS group had a smaller canal-body ratio at C4 to C7 (all p < 0.001) and less space available for the cord at C3 to C7 (all p < 0.01), while the spinal cord AP diameters were similar, at C5 to C7 (all p > 0.05) in particular. DcSS had a significantly larger cord-canal ratio (C3: p = 0.041; C4 to C7: p < 0.001). Cut-off values of AP canal diameters were defined for DcSS as follows: C3 12.9 mm, C4 11.8 mm, C5 11.9 mm, C6 12.3 mm, C7 13.3 mm (Table IV). The C3 cut-off value had the worst sensitivity (73.8%) and specificity (41.7%), whereas each level of C4 to C7 cut-off values had excellent sensitivity (98.5% to 98.6%) and specificity (96.8%).
Table III.
Comparisons of imaging parameters between developmental cervical spinal stenosis (DcSS; composite score ≥ 3) and non-DcSS (composite score 2 or less) individuals in the general population-based cohort.
| Parameters | DcSS group | Non-DcSS group | Post-hoc power (effect size) | p-value |
|---|---|---|---|---|
| Patients, n | 247 | 1,772 | ||
| Mean age, yrs (SD) | 51.4 (9.0) | 49.9 (9.6) | 0.58 (0.16) | 0.087* |
| Sex, n | 0.79 (0.06) | 0.370† | ||
| Male | 89 | 691 | ||
| Female | 158 | 1,081 | ||
| Mean AP diameter of cervical spinal canal, mm (SD) | ||||
| C3 | 13.3 (1.5) | 13.7 (1.3) | 0.98 (0.30) | < 0.001* |
| C4 | 11.1 (0.6) | 13.6 (1.1) | > 0.99 (2.38) | < 0.001* |
| C5 | 11.0 (0.7) | 13.9 (1.2) | > 0.99 (2.52) | < 0.001* |
| C6 | 11.4 (0.7) | 14.4 (1.3) | > 0.99 (2.41) | < 0.001* |
| C7 | 12.6 (0.6) | 15.3 (1.2) | > 0.99 (2.36) | < 0.001* |
| Mean mid-vertebral AP diameter of vertebral body, mm (SD) | ||||
| C3 | 13.7 (1.1) | 14.8 (1.5) | > 0.99 (0.75) | < 0.001* |
| C4 | 13.5 (1.2) | 14.6 (1.7) | > 0.99 (0.67) | < 0.001* |
| C5 | 13.1 (1.3) | 14.4 (1.9) | > 0.99 (0.71) | < 0.001* |
| C6 | 13.3 (1.4) | 14.6 (1.8) | > 0.99 (0.74) | < 0.001* |
| C7 | 13.3 (1.3) | 14.5 (1.6) | > 0.99 (0.77) | < 0.001* |
| Mean AP diameter of cervical spinal cord, mm (SD) | ||||
| C3 | 7.5 (0.6) | 7.6 (0.6) | 0.63 (0.17) | 0.039* |
| C4 | 7.1 (0.6) | 7.2 (0.6) | 0.63 (0.17) | 0.027* |
| C5 | 7.0 (0.5) | 7.0 (0.6) | 0.05 (0.00) | 0.076* |
| C6 | 6.8 (0.5) | 6.8 (0.5) | 0.05 (0.00) | 0.393* |
| C7 | 6.3 (0.4) | 6.4 (0.5) | 0.80 (0.20) | 0.076* |
| Mean canal-body ratio (SD) | ||||
| C3 | 0.98 (0.14) | 00.93 (0.13) | 0.99 (0.38) | < 0.001* |
| C4 | 0.83 (0.09) | 0.94 (0.11) | > 0.99 (1.02) | < 0.001* |
| C5 | 0.85 (0.10) | 0.98 (0.13) | > 0.99 (1.03) | < 0.001* |
| C6 | 00.87 (0.11) | 01.00 (0.12) | > 0.99 (1.09) | < 0.001* |
| C7 | 0.95 (0.10) | 1.06 (0.12) | > 0.99 (0.93) | < 0.001* |
| Mean space available for cord, mm (SD) | ||||
| C3 | 5.8 (1.6) | 6.2 (1.4) | 0.97 (0.28) | 0.004* |
| C4 | 4.0 (0.8) | 6.4 (1.3) | > 0.99 (1.92) | < 0.001* |
| C5 | 4.0 (0.8) | 6.9 (1.3) | > 0.99 (2.32) | < 0.001* |
| C6 | 4.6 (0.8) | 7.6 (1.3) | > 0.99 (2.40) | < 0.001* |
| C7 | 6.2 (0.7) | 8.9 (1.3) | > 0.99 (2.17) | < 0.001* |
| Mean cord-canal ratio (SD) | ||||
| C3 | 0.57 (0.08) | 0.56 (0.07) | 0.48 (0.14) | 0.041* |
| C4 | 0.64 (0.06) | 0.53 (0.06) | > 0.99 (1.83) | < 0.001* |
| C5 | 0.64 (0.06) | 0.51 (0.06) | > 0.99 (2.17) | < 0.001* |
| C6 | 0.60 (0.06) | 0.48 (0.05) | > 0.99 (2.34) | < 0.001* |
| C7 | 0.50 (0.04) | 0.42 (0.04) | > 0.99 (2.00) | < 0.001* |
| Mean cord-vertebral body ratio (SD) | ||||
| C3 | 0.55 (0.07) | 0.52 (0.07) | > 0.99 (0.43) | < 0.001* |
| C4 | 0.53 (0.07) | 0.50 (0.07) | > 0.99 (0.43) | < 0.001* |
| C5 | 0.54 (0.07) | 0.50 (0.08) | > 0.99 (0.51) | < 0.001* |
| C6 | 0.52 (0.07) | 0.48 (0.07) | > 0.99 (0.57) | < 0.001* |
| C7 | 0.48 (0.06) | 0.44 (0.06) | > 0.99 (0.67) | < 0.001* |
-
*
Mann-Whitney U test.
-
†
Chi-squared test.
-
DcSS, developmental cervical spinal stenosis.
Table IV.
Receiver operating characteristics analysis of developmental cervical spinal stenosis cut-off values of anteroposterior spinal canal diameter per cervical vertebral level.
| Vertebral level | Cut-off value of AP spinal canal diameter for DcSS, mm | AUC (95% CI) | Sensitivity % | Specificity % | Youden’s index | p-value |
|---|---|---|---|---|---|---|
| C3 | 12.9 | 0.573 (0.533 to 0.614) | 73.8 | 41.7 | 0.155 | < 0.001 |
| C4 | 11.8 | 0.998 (0.997 to 0.999) | 98.6 | 96.8 | 0.954 | < 0.001 |
| C5 | 11.9 | 0.998 (0.997 to 0.999) | 98.5 | 96.8 | 0.953 | < 0.001 |
| C6 | 12.3 | 0.998 (0.997 to 0.999) | 98.6 | 96.8 | 0.954 | < 0.001 |
| C7 | 13.3 | 0.998 (0.997 to 0.999) | 98.6 | 96.8 | 0.954 | < 0.001 |
-
AP, anteroposterior; AUC, area under the curve; DcSS, developmental cervical spinal stenosis.
Based on the newly defined cut-off values of spinal canal diameter and narrowing at ≥ three levels, 13.0% of the population-based cohort (262 of 2,019) and 47.4% of the DCM patients (45 of 95) had DcSS. Logistic regression reveals that the presence of DcSS at ≥ three levels was a significant predictor of developing DCM requiring surgical intervention, with an odds ratio (OR) of 6.12 (95% CI 3.97 to 9.42; p < 0.001), together with sex (OR 4.06 (95% CI 2.55 to 6.45); p < 0.001) (Table V). We can interpret this as showing that an individual with multilevel cervical canal narrowing is six times more likely to develop DCM compared to those with canal narrowing at less than three levels or without narrowing, and that males are four times more likely to have DCM than females.
Table V.
Logistic regression of developing degenerative cervical myelopathy requiring surgery. X2(2) = 101.403, p < 0.001, correct prediction = 95.5%
| Parameter | Coefficient, β (SE) | Ward X2 | Odds ratio (95% CI) | p-value |
|---|---|---|---|---|
| Presence of multilevel DcSS* | 1.811 (0.220) | 67.592 | 6.12 (3.97 to 9.42) | < 0.001 |
| Male sex† | 1.401 (0.236) | 35.087 | 4.06 (2.55 to 6.45) | < 0.001 |
-
*
Reference: no multilevel DcSS.
-
†
Reference: female.
-
DcSS, developmental cervical spinal stenosis.
Discussion
By using MRI, this study has defined level-specific cut-off values of AP cervical spinal canal diameter for DcSS, importantly incorporating the multilevel characteristic of pre-existing canal narrowing in developmental spinal stenosis. Based on the proposed criteria of defining DcSS (i.e. with canal narrowing detected at three or more levels), individuals with DcSS were successfully differentiated from those without DcSS as DcSS was demonstrated to have a narrower canal diameter, less space available for the cord, and a higher cord-canal mismatch, while spinal cord sizes were similar. The relationship between spinal cord size and canal space is crucial, as the space available for the neural bundles is the amount of leeway available for developmental spinal stenosis. Given the similar spinal cord size, less canal space available poses a risk for neural compression. In addition, the presence of such DcSS was found to be predictive of the development of DCM that warrants surgery. This provides a set of canal size criteria for identifying patients potentially at risk of developing DCM, making image screening at an early stage possible prior to further deterioration and traumatic spinal cord compression.
With the lack of consensus on a clear definition of DcSS, it remains unclear under what criteria an individual is considered at risk of DCM. Our study has first used the local DCM cohort as a guiding reference and incorporated the characteristics of multilevel involvement of canal narrowing at three vertebral levels or more for initial screening of stenotic levels. This has revealed that cervical canal narrowing is not uncommon (12.2% of the population-based cohort, with consecutive levels of canal narrowing most common at the lower cervical segment at C5 to C7), however more customized values for DcSS screening in the general population are required. This is because we found that when the screening criteria are too stringent, as demonstrated by the conventional use of a generalized spinal canal diameter of 11 mm for screening a narrowed canal,38 many individuals with narrowed canals as well as the multilevel characteristics are missed in comparison with using the DCM cohort level-specific median canal diameters (Figure 3). Hence, there is a need to refine the set of diagnostic criteria of cervical canal sizes, and we have derived the cut-off spinal canal diameters from the population-based cohort.
Fig. 3
a) Number of individuals with canal narrowing screened by conventional versus degenerative cervical myelopathy (DCM) median anteroposterior (AP) canal diameters. b) Composite score distributions representing the number of levels with canal narrowing.
We recommend clinicians detect the presence of DcSS by examining the AP spinal canal diameter, particularly at C4 to C7 based on our level-specific cut-offs defined with high sensitivity and specificity. The recommended AP spinal canal diameters for clinical use are: C3 < 13 mm, C4 < 12 mm, C5 < 12 mm, C6 < 12.5 mm, and C7 < 13.5 mm. Early identification of individuals with DcSS and early action are extremely important, as the traumatic spinal cord injuries from stenosis at the cervical spine can cause irreversible changes in the cord.39 As cervical canal narrowing is a risk factor for cervical myelopathy,24 once a patient is identified with cervical canal narrowing at any level, closer monitoring and vigilance of DCM development are mandatory. When multilevel (≥ 3) spinal canal narrowing is detected, the presence of such DcSS indicates a six-times increased likelihood of developing DCM requiring surgical intervention, compared to without DcSS. This coincides with studies by Hukuda et al27 and Chen et al,4 who found a smaller canal diameter (transverse and sagittal, respectively) to be a predisposing factor for cervical spondylotic myelopathy. Male sex is also associated with higher risk. These prompt the consideration of not only early intervention but also pre-emptive posterior canal enlargement during the silent period of stenosis or when symptoms first arise.40 With early intervention, patients can benefit from better prognosis and faster neurological recovery.12 Furthermore, the defined cut-off AP canal diameters can be used for screening potential DCM in cases of degeneration already detected in other segments of the spine. In order to avoid over-diagnosis of DcSS by these defined criteria, validation and long-term longitudinal follow-up studies are required to evaluate the effectiveness of applying these cut-offs in early DCM identification, and any clinical evidence for early or prophylactic decompression or canal enlargement surgery for asymptomatic DcSS.38,41
This study has limitations. The number of individuals in the DCM and population-based cohorts differed significantly due to the relatively lower number of DCM patients, which can affect the robustness of comparative analysis. However, the ratio of DCM and general population individuals is comparable with the prevalence (5.3%) of cervical canal compression in asymptomatic individuals in Japan.7 Additionally, slight differences of MRI in slice thickness and spacing for DCM patient and population-based cohorts might cause small measurement errors. There is a lack of long-term follow-up to confirm the predictive value of the MRI cut-off values for DCM development, and further investigation of long-term follow-up data of the general population cohort is necessary. Moreover, the defined DcSS criteria is only applicable to the Chinese population. Future evaluation of the applicability of these cut-off values and the differences between populations is required. Similar studies need to be conducted in different countries in order to determine the prevalence of DcSS worldwide.
Based on the analyses of the MRI cervical spine morphologies of DCM patients and the general population, DcSS is defined with level-specific cervical canal diameter cut-off values with the incorporation of the characteristics of multilevel involvement. Clinical use of these DcSS canal diameter cut-offs (C3 < 13 mm, C4 < 12 mm, C5 < 12 mm, C6 < 12.5 mm, C7 < 13.5 mm) can aid in the early identification of individuals with developmental spinal stenosis at the cervical spine. Individuals with DcSS at ≥ 3 levels and of male sex are at high risk of developing DCM requiring surgery, and they are recommended for consideration of early intervention during silent stenosis or when symptoms first arise.
References
1. Sasaki T , Kadoya S , Iizuka H . Roentgenological study of the sagittal diameter of the cervical spinal canal in normal adult Japanese . Neurol Med Chir . 1998 ; 38 ( 2 ): 83 – 88 . Crossref PubMed Google Scholar
2. Kato F , Yukawa Y , Suda K , Yamagata M , Ueta T . Normal morphology, age-related changes and abnormal findings of the cervical spine. Part II: Magnetic resonance imaging of over 1,200 asymptomatic subjects . Eur Spine J . 2012 ; 21 ( 8 ): 1499 – 1507 . Crossref PubMed Google Scholar
3. Cheung JPY , Kao PYP , Sham P , Cheah KSE , Chan D , Cheung KMC , et al. Etiology of developmental spinal stenosis: a genome-wide association study: etiology of developmental spinal stenosis . J Orthop Res . 2017 . Crossref PubMed Google Scholar
4. Chen IH , Liao KK , Shen WY . Measurement of cervical canal sagittal diameter in Chinese males with cervical spondylotic myelopathy . Zhonghua Yi Xue Za Zhi . 1994 ; 54 ( 2 ): 105 – 110 . PubMed Google Scholar
5. Hayashi H , Okada K , Hamada M , Tada K , Ueno R . Etiologic factors of myelopathy. a radiographic evaluation of the aging changes in the cervical spine . Clin Orthop Relat Res . 1987 ; 214 : 200 – 209 . PubMed Google Scholar
6. Morishita Y , Naito M , Hymanson H , Miyazaki M , Wu G , Wang JC . The relationship between the cervical spinal canal diameter and the pathological changes in the cervical spine . Eur Spine J . 2009 ; 18 ( 6 ): 877 – 883 . Crossref PubMed Google Scholar
7. Nakashima H , Yukawa Y , Suda K , Yamagata M , Ueta T , Kato F . Narrow cervical canal in 1211 asymptomatic healthy subjects: the relationship with spinal cord compression on MRI . Eur Spine J . 2016 ; 25 ( 7 ): 2149 – 2154 . Crossref PubMed Google Scholar
8. Nouri A , Tetreault L , Nori S , Martin AR , Nater A , Fehlings MG . Congenital cervical spine stenosis in a multicenter global cohort of patients with degenerative cervical myelopathy: an ambispective report based on a magnetic resonance imaging diagnostic criterion . Neurosurgery . 2018 ; 83 ( 3 ): 521 – 528 . Crossref PubMed Google Scholar
9. Torg JS , Naranja RJ Jr , Pavlov H , Galinat BJ , Warren R , Stine RA . The relationship of developmental narrowing of the cervical spinal canal to reversible and irreversible injury of the cervical spinal cord in football players . J Bone Joint Surg Am . 1996 ; 78-A ( 9 ): 1308 – 1314 . Crossref PubMed Google Scholar
10. Nouri A , Tetreault L , Singh A , Karadimas SK , Fehlings MG . Degenerative cervical myelopathy: epidemiology, genetics, and pathogenesis . Spine . 2015 ; 40 ( 12 ): E675 – 93 . Crossref PubMed Google Scholar
11. Countee RW , Vijayanathan T . Congenital stenosis of the cervical spine: diagnosis and management . J Natl Med Assoc . 1979 ; 71 ( 3 ): 257 – 264 . PubMed Google Scholar
12. OSCIS investigators , Chikuda H , Koyama Y , et al. Effect of early vs delayed surgical treatment on motor recovery in incomplete cervical spinal cord injury with preexisting cervical stenosis: a randomized clinical trial . JAMA Netw Open . 2021 ; 4 ( 11 ): e2133604 . Crossref PubMed Google Scholar
13. Lai MKL , Cheung PWH , Samartzis D , Karppinen J , Cheung KMC , Cheung JPY . Clinical implications of lumbar developmental spinal stenosis on back pain, radicular leg pain, and disability . Bone Joint J . 2021 ; 103-B ( 1 ): 131 – 140 . Crossref PubMed Google Scholar
14. Cheung PWH , Fong HK , Wong CS , Cheung JPY . The influence of developmental spinal stenosis on the risk of re-operation on an adjacent segment after decompression-only surgery for lumbar spinal stenosis . Bone Joint J . 2019 ; 101-B ( 2 ): 154 – 161 . Crossref PubMed Google Scholar
15. Lai MKL , Cheung PWH , Cheung JPY . A systematic review of developmental lumbar spinal stenosis . Eur Spine J . 2020 ; 29 ( 9 ): 2173 – 2187 . Crossref PubMed Google Scholar
16. Lai MKL , Cheung PWH , Samartzis D , Cheung JPY . Prevalence and definition of multilevel lumbar developmental spinal stenosis . Global Spine J . 2022 ; 12 ( 6 ): 1084 – 1090 . Crossref PubMed Google Scholar
17. Lai MKL , Cheung PWH , Song Y-Q , Samartzis D , Cheung JPY . Pedigree analysis of lumbar developmental spinal stenosis: determination of potential inheritance patterns . J Orthop Res . 2021 ; 39 ( 8 ): 1763 – 1776 . Crossref PubMed Google Scholar
18. Inoue H , Ohmori K , Takatsu T , Teramoto T , Ishida Y , Suzuki K . Morphological analysis of the cervical spinal canal, dural tube and spinal cord in normal individuals using CT myelography . Neuroradiology . 1996 ; 38 ( 2 ): 148 – 151 . Crossref PubMed Google Scholar
19. Bajwa NS , Toy JO , Young EY , Ahn NU . Establishment of parameters for congenital stenosis of the cervical spine: an anatomic descriptive analysis of 1,066 cadaveric specimens . Eur Spine J . 2012 ; 21 ( 12 ): 2467 – 2474 . Crossref PubMed Google Scholar
20. Pavlov H , Torg JS , Robie B , Jahre C . Cervical spinal stenosis: determination with vertebral body ratio method . Radiology . 1987 ; 164 ( 3 ): 771 – 775 . Crossref PubMed Google Scholar
21. Ishikawa M , Matsumoto M , Fujimura Y , Chiba K , Toyama Y . Changes of cervical spinal cord and cervical spinal canal with age in asymptomatic subjects . Spinal Cord . 2003 ; 41 ( 3 ): 159 – 163 . Crossref PubMed Google Scholar
22. Kar M , Bhaumik D , Ishore K , Saha PK . MRI study on spinal canal morphometry: an Indian study . J Clin Diagn Res . 2017 ; 11 ( 5 ): AC08 – AC11 . Crossref PubMed Google Scholar
23. Kovalova I , Kerkovsky M , Kadanka Z , et al. Prevalence and imaging characteristics of nonmyelopathic and myelopathic spondylotic cervical cord compression . Spine . 2016 ; 41 ( 24 ): 1908 – 1916 . Crossref PubMed Google Scholar
24. Nagata K , Yoshimura N , Hashizume H , et al. The prevalence of cervical myelopathy among subjects with narrow cervical spinal canal in a population-based magnetic resonance imaging study: the Wakayama spine study . Spine J . 2014 ; 14 ( 12 ): 2811 – 2817 . Crossref PubMed Google Scholar
25. Nell C , Bülow R , Hosten N , Schmidt CO , Hegenscheid K . Reference values for the cervical spinal canal and the vertebral bodies by MRI in a general population . PLoS One . 2019 ; 14 ( 9 ): e0222682 . Crossref PubMed Google Scholar
26. Freedman BA , Hoffler CE , Cameron BM , et al. A comparison of computed tomography measures for diagnosing cervical spinal stenosis associated with myelopathy: a case-control study . Asian Spine J . 2015 ; 9 ( 1 ): 22 – 29 . Crossref PubMed Google Scholar
27. Hukuda S , Xiang LF , Imai S , Katsuura A , Imanaka T . Large vertebral body, in addition to narrow spinal canal, are risk factors for cervical myelopathy . J Spinal Disord . 1996 ; 9 ( 3 ): 177 – 186 . PubMed Google Scholar
28. Murakami K , Nagata K , Hashizume H , et al. Prevalence of cervical anterior and posterior spondylolisthesis and its association with degenerative cervical myelopathy in a general population . Sci Rep . 2020 ; 10 ( 1 ): 10455 . Crossref PubMed Google Scholar
29. Nagata K , Kiyonaga K , Ohashi T , Sagara M , Miyazaki S , Inoue A . Clinical value of magnetic resonance imaging for cervical myelopathy . Spine . 1990 ; 15 ( 11 ): 1088 – 1096 . Crossref PubMed Google Scholar
30. Yue WM , Tan SB , Tan MH , Koh DC , Tan CT . The torg--pavlov ratio in cervical spondylotic myelopathy: a comparative study between patients with cervical spondylotic myelopathy and a nonspondylotic, nonmyelopathic population . Spine . 2001 ; 26 ( 16 ): 1760 – 1764 . Crossref PubMed Google Scholar
31. Blackley HR , Plank LD , Robertson PA . Determining the sagittal dimensions of the canal of the cervical spine. The reliability of ratios of anatomical measurements . J Bone Joint Surg Br . 1999 ; 81-B ( 1 ): 110 – 112 . Crossref PubMed Google Scholar
32. Moskovich R , Shott S , Zhang ZH . Does the cervical canal to body ratio predict spinal stenosis? Bull Hosp Jt Dis . 1996 ; 55 ( 2 ): 61 – 71 . PubMed Google Scholar
33. Prasad SS , O’Malley M , Caplan M , Shackleford IM , Pydisetty RK . MRI measurements of the cervical spine and their correlation to pavlov’s ratio . Spine . 2003 ; 28 ( 12 ): 1263 – 1268 . Crossref PubMed Google Scholar
34. Ulbrich EJ , Schraner C , Boesch C , et al. Normative MR cervical spinal canal dimensions . Radiology . 2014 ; 271 ( 1 ): 172 – 182 . Crossref PubMed Google Scholar
35. Vangeneugden T , Laenen A , Geys H , Renard D , Molenberghs G . Applying concepts of generalizability theory on clinical trial data to investigate sources of variation and their impact on reliability . Biometrics . 2005 ; 61 ( 1 ): 295 – 304 . Crossref PubMed Google Scholar
36. Fluss R , Faraggi D , Reiser B . Estimation of the youden index and its associated cutoff point . Biom J . 2005 ; 47 ( 4 ): 458 – 472 . Crossref PubMed Google Scholar
37. Mandrekar JN . Receiver operating characteristic curve in diagnostic test assessment . J Thorac Oncol . 2010 ; 5 ( 9 ): 1315 – 1316 . Crossref PubMed Google Scholar
38. Shigematsu H , Cheung JPY , Mak K-C , Bruzzone M , Luk KDK . Cervical spinal canal stenosis first presenting after spinal cord injury due to minor trauma: an insight into the value of preventive decompression . J Orthop Sci . 2017 ; 22 ( 1 ): 22 – 26 . Crossref PubMed Google Scholar
39. Yoo D-S , Lee S-B , Huh P-W , Kang S-G , Cho K-S . Spinal cord injury in cervical spinal stenosis by minor trauma . World Neurosurg . 2010 ; 73 ( 1 ): 50 – 52 . Crossref PubMed Google Scholar
40. Milligan J , Ryan K , Fehlings M , Bauman C . Degenerative cervical myelopathy: diagnosis and management in primary care . Can Fam Physician . 2019 ; 65 ( 9 ): 619 – 624 . PubMed Google Scholar
41. Cheung JPY , Cheung PWH , Chiu CK , Chan CYW , Kwan MK . Variations in practice among asia-pacific surgeons and recommendations for managing cervical myelopathy: the first Asia-Pacific Spine Society collaborative study . Asian Spine J . 2019 ; 13 ( 1 ): 45 – 55 . Crossref PubMed Google Scholar
Author contributions
P. W. H. Cheung: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing, Visualization
J. H. M. Leung: Data curation, Formal analysis, Investigation, Visualization, Writing – original draft
V. W. Y. Lee: Investigation, Validation, Writing – review & editing, Data curation
J. P. Y. Cheung: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing, Data curation, Visualization, Writing – original draft
Funding statement
The authors disclose receipt of the following financial or material support for the research, authorship, and/or publication of this article: Theme-based Research Scheme (T12-708/12 N), Area of Excellence Scheme (AoE/M-04/04), and HKU MRes Programme.
ICMJE COI statement
J. P. Y. Cheung reports institutional payments from the Theme-based Research Scheme (T12-708/12 N), Area of Excellence Scheme (AoE/M-04/04), and HKU MRes Programme for this study. J. H. M. Leung also reports institutional payments from HKU MRes Programme for this study.
Data sharing
The datasets generated and analyzed in the current study are not publicly available due to data protection regulations. Access to data is limited to the researchers who have obtained permission for data processing. Further inquiries can be made to the corresponding author.
Ethical review statement
Ethics approval was obtained from the institutional review board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (IRB reference number: UW15-598, UW 10-172).
Open access funding
The open access fee for this article was funded by the Department of Orthopaedics and Traumatology, The University of Hong Kong.
Open access statement
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
Social media
Follow J. P. Y. Cheung on X @jasonpycheung
This article was primary edited by S. P. F. Hughes.
