Developmental canal stenosis was thought to be caused mainly by a decrease in the sagittal diameter of cervical canal. Its diagnosis was achieved if the Pavlov ratio, the sagittal diameter of the middle spinal canal to the sagittal diameter of the middle vertebral body, was 0.82 or less on plain radiographs.1 For Chinese population, the standard value was set to be 0.75.2
However, this diagnostic criterion was established in an era when the methods of radiographic examination were quite limited. With the development of magnetic resonance imaging (MRI), problems of this criterion seemed to emerge. For example, there is still reserving space between the spinal cord and walls of spinal canal in T2-weighted MRI image, and it has been arguable whether the existence of the reserving space for the spinal cord was correlated with the stenosis of the bony canal. Besides, the preferable surgical approach for patients with developmental cervical stenosis (DCS) is not yet clear.3
In this study, we measured and calculated the values in MRI images in patients with or without stenosis of cervical spinal canal and tried to reveal the influence of the intradural reserving space on the spinal cord, as well as the difference in their clinical manifestation and surgical approaches.
A cohort of 264 (164 males, 100 females) patients with cervical spondylotic myelopathy (CSM) who underwent operation at Peking University Third Hospital between February 2006 and July 2010 were recruited retrospectively for the study. The average age was 56.89 years (ranging from 34 to 84).
Patients with CSM with or without DCS were included in our research. Patients with the following manifestations were excluded: any type of ossification of posterior longitudinal ligament (continuous, limited, or mixed), 1disc herniation or osteophyte with occupation rate > 50%,4 ossification of ligamentum flavum, cervical kyphosis deformity (the angle from C2 to C7 <0° or local Cobb angle >5°),5,6 and trauma history. The average age of onset was 54.58 years (ranging from 23 to 84 ) and the average duration was 17.7 months (1–120 months). The average preoperative modified Japanese Orthopaedic Association (mJOA) score was 11.9 (4–16.5).
The radiographic data included the lateral images on X-ray and axial and sagittal T2-weighted magnetic resonance images in neutral position of the cervical spine from C3 to C7 in 264 patients at 1 320 levels. The measurements were recorded as minimum 0.1 mm using the length measurement device of Centricity Enterprise Web 3.0 on plain radiographs. The MR images were produced on a 1.5 T Scanner (Siemens Sonata; Siemens Medical Systems, South Iselin, NJ, USA). The sequences used were sagittal T2 weighted (TE 130 ms and TR 4 481 ms) and axial T2 weighted (TE 12 ms and TR 587 ms) and flip angle 6° with MTC. The parameters were measured independently by two senior spine surgeons, and the average value of each parameter at each level was taken for independent calculations. The definitions of the parameters were as follows:
Pavlov ratio was defined as the ratio of the sagittal diameter of the spinal canal to the sagittal diameter of the vertebral body (Figure 1A). All cases were categorized into two groups: DCS group (stenosis group, the ratio at any level from C3 to C7 was 0.75 or smaller) and non-DCS group (non-stenosis group, the ratios at C3-C7 were all over 0.75).
MRI Pavlov ratio was defined as the ratio of the sagittal diameter of the dural sac to the sagittal diameter of the vertebral body (Figure 1B).
Occupation ratio was defined as the ratio of the transverse area of the spinal cord to the transverse area of the dural sac in axial image in T2-weighted MRI (Figure 1C).
Reserving space was defined as the space surrounding the spinal cord filled with cerebrospinal fluid (CSF) in axial image in MRI.
Segments with or without reserving space: the average of the occupation ratio in each segment of the non-stenosis group was considered as the standard. The segment for which the occupation ratio was no more than the standard value was defined as with reserving space, otherwise without reserving space.
Space-reserving/non-space-reserving subgroup: the case who had three or more segments with “reserving space” was classified as space-reserving subgroup; otherwise classified as non-space-reserving subgroup.
Surgical options were determined by symptoms, physical signs, and radiologic measurements, and patients received anterior cervical decompression and fusion or posterior “opening-door” laminoplasty accordingly. The number of compression levels, direction of compression, the maximal compressed level, surgical trauma, and general health status were also taken into consideration to facilitate the surgical option.
The cases were followed up for 29 months (24–60 months) on average. Patients with complications were ruled out at the final follow-up, such as spinal cord injury, postoperative epidural tear, CSF fistula, or epidural hematoma. The severity of myelopathy was evaluated by the mJOA score and recovery rate was determined with Hirabayashi's formula ((24-month postoperative score - preoperative score)/(17 - preoperative score) × 100%).7
The MRI Pavlov ratio and the occupation ratio were compared between the stenosis and non-stenosis groups before surgery. The age of onset, duration of disease, pre- and postoperative mJOA score, and recovery rate were also compared between reserving and non-reserving space group.
Data were analyzed with SPSS 15.0 software (SPSS Inc., Chicago, IL, USA). Statistical significance was set at 0.05. Descriptive statistics in the form of mean ± standard deviation (SD) for all spine parameters were obtained from all patients. All populations were confirmed to approximate the normal distribution by probability plots. Pearson's correlation coefficient was used to calculate the correlation between MRI and clinical parameters. Equal variance criterion of all groups was upheld by Mauchly's test. Then paired or group Student t-test was utilized to evaluate the difference among different groups.
There were 152 patients in stenosis group and 112 patients in non-stenosis group. Comparisons of MRI Pavlov ratio and occupation ratio between stenosis and non-stenosis groups are shown in Table 1. The MRI Pavlov ratio is significantly smaller in stenosis group than non-stenosis group at all levels (P <0.05). However, the occupation ratio was only significantly different at C7 between the two groups (P <0.05). The occupation ratio had the largest value at C5 and the smallest value at C7 in both groups.
The comparisons of the age of onset, duration of disease, and pre- and postoperative mJOA score in 24 months between space-reserving and non-space-reserving subgroups are shown in Table 2. Patients with shorter duration of the disease for clinics had a better recovery rate in non-space-reserving subgroup (P <0.05) and the patients in non-space-reserving subgroup had an earlier onset of the symptoms than the space-reserving subgroup.
The pre- and postoperative mJOA score and the recovery rate in space-reserving and non-space-reserving subgroups are shown in Tables 3 and 4, respectively. The surgical approach of the anterior decompression and fusion was significantly more effective than posterior “openingdoor” laminoplasty in space-reserving subgroup (P <0.05). However, the results of the two approaches were similar in non-space-reserving subgroup.
The bony cervical canal stenosis was thought to be one of the precipitating factors that cause CSM.8,9 Pavlov et al used the ratio of the sagittal diameter of the spinal canal to the sagittal diameter of the vertebral body from plain radiographs (Pavlov ratio) as a reliable determinant instead of the true diameter of the cervical spinal canal. They concluded that a value ≤0.82 represented stenosis in 92% of cases with a 6% false-positive rate.1 Dang et al found a value ≤0.75 of the Pavlov ratio that indicated cervical canal stenosis for Chinese population.2
However, evidences concerning the correlation between the Pavlov ratio and the true sagittal diameter of the cervical spinal canal are still controversial. Herzog et al10 reported a good correlation between the sagittal diameter of the spinal canal on plain films and sagittal CT images. Blackley et al11 showed that the canal-to-body ratio was of limited value in the assessment of the true diameter of the canal. With the wide use of magnetic resonance images, Hulcelle et al12 made a direct measurement of the dural sac, spinal cord, and other soft tissues and found that an individual cervical spine bony measurement was unable to accurately assess the degree of compression and the reserving space for the spinal cord. After the measurement of the sagittal diameter of the dural sac and the spinal cord directly on T2-weighted sagittal MRI images, we utilized the MRI Pavlov ratio, a parameter reflecting the relationship between the vertebral body, and the dural sac and its surrounding soft tissues, to evaluate the exact structure in the spinal canal in order to prevent the phenomenon that the dural sac still had adequate space around it despite the canal being relatively narrow. Moreover, in this study, we found that the MRI Pavlov ratio was significantly smaller in patients with DCS in every segment, which suggested that the vertebral canal bony stenosis was probably accompanied by dural stenosis, and this could be a pathogenic factor influencing spinal cord compression.
In all cases, the spinal cord could be compressed from all directions, and its oval structure depended on both the sagittal and transverse structures. Therefore, the application of MRI to measure the soft tissues directly was meaningful. Okada et al13 measured the areas of the spinal cord, dural sac, and spinal canal on T1-weighted axial images and proposed a ratio that resulted from the area of the cord divided by the spinal canal in order to prevent individual variation. Based on the these previous research results, we established the concept “occupation ratio,” which was defined as the ratio between the area of the cord and the dual sac instead of the spinal canal, so that some soft tissues such as posterior longitudinal ligament or ligamentum flavum would not distract the assessment of real reserving space for the spinal cord. Okada et al13 reported its maximal value at C4 and C5, and the minimal at C7. Our results showed the maximal value at C5 and the minimal at C7 in both stenosis and non-stenosis groups, which suggested that the cerebral fluid in each transverse area at C5 was less than that at C7. Claire et al14 found that the spinal cord deformation was significantly reduced in the presence of CSF compared with the absence of CSF when applied upon transverse impact. These facts suggested that CSF may be involved in the prevalence of CSM in DCS. We found the occupation ratio was larger in stenosis group and only significantly at C7, which suggested that the transverse area of the CSF was smaller than that in non-stenosis group. Therefore, the decrease in the sagittal diameter probably represented a decrease in the whole area of the dural sac and the reserving space for the cord. However, the incidence of such a difference was not as usual as the ones in MRI Pavlov ratio, which presumably indicated an adaptive modification of the area in the spinal cord. Moreover, in stenosis group the transverse area of the dural sac decreased significantly at C7 while that of the spinal cord remained similar compared with non-stenosis group, which could explain the significantly larger occupation ratio at C7. The largest transverse area of the dural sac at this level could not affect the development of the spinal cord, which indicates that the development of the spinal cord may adapt passively to the surrounding structures. A possible phenomenon may occur as “small spinal canal with small spinal cord.”
Meanwhile, the recovery rate was better in non-spacereserving subgroup than that in space-reserving subgroup, proposing that the more compressed the spinal cord was, the better it recovered after sufficient decompression. Nevertheless, for those with reserving space, a comparatively smaller spinal cord was accompanied by a smaller canal, which did not lead to amelioration in nonspacereserving subgroup.
The surgical treatment for CSM can be roughly categorized as anterior or posterior approach. The effective way of anterior decompression and fusion was intended for those who have herniated discs or osteophytes compressing the spinal cord. On the other hand, posterior laminoplasty was thought to be indicated for those who have both anterior and posterior compression as multiple segmental disc herniation or those who have degenerative or developmental canal stenosis in order to enlarge the spinal canal and make the spinal cord shift backward,15–17 Researches on the comparisons between the two surgical approaches have continued during the past four decades.18–23 Based on our study about patients with CSM with DCS, a decrease in the sagittal diameter of the spinal canal was not thought to be the only determinant for the surgical approach. Both anterior and posterior approaches provided satisfying curative results in non-stenosis group in our cohort, while anterior approach brought better outcomes in patients with reserving space in stenosis group, with respect to its postoperative mJOA recovery rate. The possible reasons are as follows: first of all, the direct decompression of the spinal cord and anterior spinal artery anteriorly, removing both the osteophytes and herniating discs, may ameliorate the blood supply of the nerve conduction bundle;24,25 second, a better CSF circulation might improve the inner environment surrounding the reserving space;14 third, the anterior local bone graft fusion provides solid stability to prevent a micro-trauma and dynamic compression on the spinal cord;26 finally, the backward shift of the spinal cord after laminoplasty may be an unpredictable factor for individual patient despite that such a mechanism was proved to be effective for clinical recoveries after posterior approach.27,28
However, it should be noted that for patients with reserving space after posterior approach, the mean recovery rate was negative and the mean postoperative mJOA was larger than the preoperative one. We found there were three patients for whom the postoperative mJOA was exceptionally lower than the preoperative one, which affected the normal distribution of the overall recovery rate; therefore, whether the mean value could still be a good parameter to represent the real recovery rate is in doubt. On the contrary, for those without reserving space, the spinal cord compressed both ventrally and dorsally could be relieved by both subtotal vertebral corpectomy and laminoplasty to achieve a satisfactory clinical outcome. The mJOA recovery rate was similar between the anterior and posterior approach groups, which suggested the local CSF environment was a critical factor for spinal cord recovery after surgery.
Due to its retrospective nature and short-time follow-up, the results of this study might be limited. Besides, the mean recovery rate after posterior approach of patients with reserving space was affected by extremely small individual values. A more convincing clinical conclusion would be drawn if a large asymptomatic cohort is included. Moreover, the clinical results might be influenced by individual surgical techniques.
DCS is associated with a smaller sagittal diameter of the dural sac, so called as “small spinal canal with small spinal cord”; however, it does not lead to a significant decrease in intra-dural space available for the cord. In this study, for patients with normal intra-dural space, the recovery after anterior decompression surgery was better than posterior approach; while for patients with insufficient intra-dural space, the recovery after the two surgical decompression approaches was similar.
1. Pavlov H, Torg JS, Robie B, Jahre C. Cervical spinal stenosis: determination with vertebral body ratio method. Radiology 1987; 164: 771-775.
2. Dang GT, Wang C, Chen ZQ, Liu ZJ, Wang ZG. Ratio method and its diagnostic significance in developmental cervical stenosis
(in Chinese). Chin J Spine Spinal Cord 1992; 2: 146-149.
3. 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: 1263-1268.
4. Park HJ, Kim SS, Chung EC, Lee SY, Park NH, Rho MH. Clinical correlation of a new practical MRI method for assessing cervical spinal canal compression. Am J Roentgenol 2012; 199: W197-W201.
5. Rajshekhar V, Arunkumar MJ, Kumar SS. Changes in cervical spine curvature after uninstrumented one- and twolevel corpectomy in patients with spondylotic myelopathy. Neurosurgery 2003; 52: 799-805.
6. Ohara A, Miyamoto K, Naganawa T, Matsumoto K, Shimizu K. Reliabilities of and correlations among five standard methods of assessing the sagittal alignment of the cervical spine. Spine 2006; 31: 2585-2591.
7. Harabayashi K, Miyakawa J, Satomi K, Maruyama T, Wakano K. Operative results and postoperative progression of ossification among patients with ossification of cervical posterior longitudinal ligament. Spine 1981; 6: 354-364.
8. Wolf BS, Khilnani M, Malis L. The sagittal diameter of the bony cervical spinal canal and its significance in cervical spondylosis
. J Mt Sinai Hosp N Y 1956; 23: 283-292.
9. Payne EE, Spillane JD. The cervical spine; An anatomicopathological study of 70 specimens ( using a special technique) with particular reference to the problem of cervical spondylosis
. Brain 1957; 80: 571-596.
10. Herzog RJ, Weins JJ, Dillingham MF, Sontag MJ. Normal cervical spine morphometry and cervical spine stenosis in asymptomatic professional football players.Plain film radiography, multiplanar computed tomography, and magnetic resonance imaging. Spine 1991; 16(6 Suppl): s178-s186.
11. 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: 110-112.
12. Hulcelle P, Dooms G, Demeure R, Mathurin P, Cornelis G. Cervical myelopathy: MRI evaluation of cord compression. J Belge Radiol 1990; 73: 15-19.
13. Okada Y, Ikata T, Katoh S, Yamada H. Morphologic analysis of the cervical spinal cord, dural tube, and spinal canal by magnetic resonance imaging in normal adults and patients with cervical spondylotic myelopathy. Spine 1994; 19: 2331-2335.
14. Jones CF, Kroeker SG, Cripton PA, Hall RM. The effect of cerebrospinal fluid on the biomechanics of spinal cord: an ex vivo
bovine model using bovine and physical surrogate spinal cord. Spine 2008; 33: E580-E588.
15. Hirabayashi K, Toyama Y, Chiba K. Expansive laminoplasty for myelopathy in ossification of the longitudinal ligament. Clin Orthop Relat Res 1999; 359: 35-48.
16. Tsuzuki N, Abe R, Saiki K, Zhongshi L. Extradural tethering effect as one mechanism of radiculopathy complicating posterior decompression of the cervical spinal cord. Spine 1996; 21: 203-211.
17. Uematsu Y, Tokuhashi Y, Matsuzaki H. Radiculopathy after laminoplasty of the cervical spine. Spine 1998; 23: 2057-2062.
18. Orr RD, Zdeblick TA. Cervical spondylotic myelopathy approaches to surgical treatment. Clin Orthop Relat Res 1999; (359): 58-66.
19. Yonenobu K, Fuji T, Ono K, Okada Y, Yamamoto T, Harada N. Choice of surgical treatment for multisegmental cervical spondylotic myelopathy. Spine 1985; 10: 710-716.
20. Kimura I, Shingu H, Nasu Y. Long-term follow-up of cervical spondylotic myelopathy treated by canal-expansive laminoplasty. J Bone Joint Surg Br 1995; 77: 956-961.
21. Goto S, Mochizuki M, Watanabe T, Hiramatu K, Tanno T, Kitahara H, et al. Long-term follow-up study of anterior surgery for cervical spondylotic myelopathy with special reference to the magnetic resonance imaging findings in 52 cases. Clin Orthop Relat Res 1993; (291): 142-153.
22. Yamasaki S, Kokubun M, Sato T, Ishii Y. Anterior decompression and fusion for cervical myelopathy due to disc herniation: Long-term evaluation of results (In Japanese). Spine and Spinal Cord 1997; 10: 797-801.
23. Kawakami M, Tamaki T, Iwasaki H, Yoshida M, Ando M, Yamada H. A comparative study of surgical approaches for cervical compressive myelopathy. Clin Orthop Relat Res 2000; (381): 129-136.
24. Breig A, el-Nadi AF. Biomechanics of the cervical spinal cord. Relief of contact pressure on and overstretching of the spinal cord. Acta Radiol Diagn (Stockh) 1966; 4: 602-624.
25. Breig A, Turnbull I, Hassler O. Effects of mechanical stresses on the spinal cord in cervical spondylosis
. A study on fresh cadaver material. J Neurosurg 1966; 25: 45-56.
26. Lawrence BD, Jacobs WB, Norvell DC, Hermsmeyer JT, Chapman JR, Brodke DS. Anterior versus posterior approach for treatment of cervical spondylotic myelopathy: a systematic review. Spine 2013; 38(22 Suppl 1): s173-s182.
27. Baba H, Uchida K, Maezawa Y, Furusawa N, Azuchi M, Imura S. Lordotic alignment and posterior migration of the spinal cord following en bloc open-door laminoplasty for cervical myelopathy: a magnetic resonance imaging study. J Neurol 1996; 243: 626-632.
28. Sodeyama T, Goto S, Mochizuki M, Takahashi J, Moriya H. Effect of decompression enlargement laminoplasty for posterior shifting of the spinal cord. Spine 1999; 24: 1527-1531.
Keywords:© 2014 Chinese Medical Association
cervical spondylosis; developmental cervical stenosis; reserving space for spinal cord; cervical decompression surgery