Eur Spine J DOI 10.1007/s00586-015-3818-0

ORIGINAL ARTICLE

A diagnostic study of thoracic myelopathy due to ossification of ligamentum flavum Fabo Feng • Chuiguo Sun • ZhongQiang Chen

Received: 30 April 2014 / Revised: 14 February 2015 / Accepted: 14 February 2015 Ó Springer-Verlag Berlin Heidelberg 2015

Abstract Purpose We set out to establish a magnetic resonance imaging (MRI) and computed tomography (CT)-based diagnostic method for determining the responsible segments in thoracic myelopathy due to ossification of the ligamentum flavum (OLF). Methods Forty-four patients who underwent surgery for treatment of myelopathy due to OLF between June 2005 and May 2013 were enrolled in this study as the myelopathy group. Forty-four patients who were identified through CT and MRI scans to have OLF but had no definite neurologic deficits prior to the examination were included as the control group. MRI and CT examination were reviewed, and the degree of spinal canal compromise was graded on axial T2-weighted MRI. Anteroposterior spinal canal diameter was measured at the maximally stenosed level on axial and sagittal CT. The canal grade and the cross-section area-occupying ratio were measured and calculated on the CT scans. The diagnostic coincidence rates for the indices were then compared. Results Cases of Grade IV were all in the myelopathy group while cases of Grade II were all in the control group. The canal grade (paramedian) was the most relevant continuous variable with the largest JOA score (r = 0.685, P \ 0.005). A canal grade (paramedian) of \60 % can be used as a critical value for determining OLF-induced myelopathy (sensitivity and specificity, 95.5 %). F. Feng  C. Sun  Z. Chen (&) Department of Orthopedics, Peking University Third Hospital, Beijing, China e-mail: [email protected] F. Feng Department of Orthopedics, Zhejiang Provincial People’s Hospital, Hangzhou, China

Conclusion Spinal canal compromise is relevant to spinal cord deficits in patients with OLF, and a canal grade (paramedian) can be used to quantify spinal cord deficits. Additionally, a canal grade (paramedian) of \60 % on axial CT scan can serve as a critical value for diagnosing OLF-induced myelopathy, especially for Grade III compression on T2-weighted MRI. Keywords Magnetic resonance imaging  Computed tomography  Diagnostic method  Thoracic myelopathy  Ossification of the ligamentum flavum

Introduction Thoracic ossification of ligamentum flavum (OLF) is a common cause of thoracic myeloradiculopathy, and has been reported almost exclusively in East Asian countries [1–8]. Studies of thoracic OLF have also increasingly been reported from other regions in recent years [9, 10]. In China, there are also more and more reports concerning the epidemiological investigations and surgery of TOLF [11– 14]. It is well known that OLF is often accompanied with other spinal degenerative diseases [15], leading to complex clinical manifestations and difficulty in clinical diagnosis. Meanwhile, it is difficult to determine the exact responsible segment in multi-segmental stenosis cases. These diagnostic problems are related to the specific region and level of surgical decompression. If the extent of decompression is insufficient, the treatment will be incomplete, resulting in residual symptoms. However, unnecessary trauma and a series of complications will occur if the extent of decompression is too large. Therefore, it is necessary to establish a simple and accurate method of identifying the responsible segment of OLF that is associated with neurologic deficit.

123

Eur Spine J

Magnetic resonance imaging (MRI) and computed tomography (CT) are the most commonly employed methods for radiological diagnosis of thoracic myelopathy. In this study, we reviewed the MRI and CT manifestations of 44 cases to determine the critical value of radiological diagnosis of OLF-induced thoracic myelopathy, and established a new diagnostic method for discriminating responsible segment in thoracic myelopathy.

Materials and methods Patient population We retrospectively analyzed 186 consecutive patients who underwent decompressive laminectomy for thoracic OLF between June 2005 and May 2013. Exclusion criteria were set with the intention of evaluating the effect of OLF that may cause myelopathy only and were as follows: concurrent thoracic ventral compressive lesions, such as the ossification of the posterior longitudinal ligament (OPLL) and thoracic lumbar disc herniation; presence of other tandem spinal diseases such as spinal deformity, lumbar spinal stenosis, and cervical myelopathy that needed operative treatment; a history of spinal surgery; more than two segments of OLF. Cases with other spinal disorders and more than two segments of OLF were excluded because Japanese Orthopedic Association (JOA) scores may easily be Table 1 Summary of the JOA scoring system for the assessment of thoracic myelopathy Neurological status

Score

Lower-limb motor dysfunction No dysfunction

4

Lack of stability and smooth reciprocation of gait

3

Able to walk on flat floor with walking aid

2

Able to walk up/downstairs with handrail

1

Unable to walk

0

Lower-limb sensory deficit No deficit

2

Mild sensory deficit

1

Severe sensory loss or pain

0

affected by these factors. Forty-four patients were enrolled in this study as the myelopathy group. Neurological assessment was conducted for each case by using a modified JOA scoring system (Table 1). There were 23 men and 21 women with a mean age of 57.2 years (range 24–76 years) in the myelopathy group. The duration of symptoms varied from 1 to 156 months with a mean of 39.5 months. One hundred and sixty-four patients who were admitted to hospital because of thoracic or lumbar compressive fractures between June 2011 and August 2013 were retrospectively reviewed. Forty-four patients who were identified through CT and MRI scans to have thoracic OLF but no definite neurologic signs (JOA scores = 11) were included in the study as control group. The control group consisted of 24 men and 20 women with a mean age of 59.1 years (range 19–82 years). No significant variation in sex (v2 = 0.014, P = 0.909) and age (t = 1.986, P = 0.068) distribution was observed between the two groups. Radiographic evaluation and measurement The myelopathy group had MRI scans of the whole spine and CT scans of T1-L2 available for review, while the control group only had MRI and CT scans of the thoracic spine. Both groups had detailed medical records including physical examinations. Two observers, with over 10 years of experience in the diagnosis and treatment of spinal degeneration, independently measured all values. Axial and sagittal slices were used for observations. The degree of spinal canal compromise was graded on axial T2-weighted MRI by using the following criteria: Grade I—OLF exists, but the thecal sac has not yet been compressed, or the OLF is only in contact with the thecal sac. Grade II—the thecal sac is compressed and deformed, but the OLF is not in contact with the spinal cord. Grade III—the subarachnoid space is partially occluded and the OLF is in contact with the spinal cord, but the spinal cord is not yet deformed. Grade IV—the spinal cord is obviously compressed and deformed (Fig. 1). The location of the ossified ligamentum flavum was determined as being around the joint capsules of the facet joints and/or the medial lamina. Definitions, measurements, and scales used

Trunk sensory deficit No deficit

2

Mild sensory deficit

1

Severe sensory loss or pain

0

Sphincter dysfunction No dysfunction Minor difficulty in micturition

3 2

Marked difficulty in micturition

1

Unable to void

0

123

The anteroposterior (AP) spinal canal diameters were measured at the maximally stenosed level on axial and sagittal CT (Fig. 2b, e). The AP spinal canal diameter was measured at three sites: the midline of the canal (point E), the boundary of the canal (point C), and midpoint between the midline and boundary (point D, we defined the point as paramedian). The normal AP canal diameter was calculated as an average diameter of the adjacent non-stenosed segments. The canal grade was calculated as:

Eur Spine J

Fig. 1 Degree of spinal canal occupation on axial MRI. a Grade I, b Grade II, c Grade III, and d Grade IV

Canal grade ¼

Canal diameter of stenosed level  100 %: Normal canal diameter

As the imaging data of early cases were not stored in our Picture Archiving and Communication Systems (PACS), we measured the cross-sectional area by using Adobe Photoshop version CS 4.0 (Adobe Systems, San Jose, CA). The CSA of the spinal canal (Fig. 2d) was measured by using the following method: The widest distance between two pedicles as viewed on a CT scan of a transverse section through the pedicle section (usually have no ossifications) of the same vertebrae, was measured as the transverse spinal canal diameter (A1 to B1) (Fig. 2a), equal to the transverse spinal canal diameter at the maximally stenosed level (A2 to B2). A vertical line extending through the endpoints of the transverse diameter (A2 and B2) determined the boundary of spinal canal and was used to measure the CSA (Fig. 2b). A normal CSA was measured on the pedicle section of the same vertebrae (Fig. 2c). The CSA occupying ratio was expressed as:

CSA occupying ratio ¼

CSA of stenosed level  100 %: Normal CSA

Statistical analysis A univariate correlation analysis was used to test correlation between various factors (Pearson’s correlation coefficients were used for continuous variables, while Spearman’s correlation coefficients were used for categorical variable) and preoperative JOA scores. We selected the top three relevant continuous variables as measurement index of spinal canal compromise and drew the ROC curves. The Youden index (sensitivity ? specificity - 1) was calculated for each index to find the critical value distinguishing between case and control groups. The coincidence rates for the three indices were then compared. SPSS version 18.0 was used for the statistical analysis. The difference was statistically significant if P \ 0.05.

123

Eur Spine J

Fig. 2 a Transverse section of the pedicle, the transverse canal diameter is the widest distance between two pedicles (A1B1). b A1 to B1 distance was equal to the transverse canal diameter at the maximum stenosed level (A2B2). The vertical lines (C and F) were drawn through the endpoints (C and F) as the boundaries of the spinal canal. The AP canal diameter was measured at three sites: the midline

123

(E), boundary (C), and midpoint between the midline and the boundary (D). c A normal CSA was measured by using the transverse section through the pedicle of the same vertebrae (the shadow region). d The CSA of the stenosed level (the shadow region). e The canal grade (sagittal) was calculated as the canal diameter of the stenosed level (GH)/normal canal diameter (G1H1 ? G2H2/2) 9 100 %

Eur Spine J

Results The most commonly affected segment was T10/11, with a proportion of 40.9 and 54.5 % in the myelopathy and control group, respectively. In the myelopathy group, there were 0 cases (0 %) of Grade I and Grade II occupation on T2-weighted MRI, 19 cases (43.2 %) of Grade III and 25 cases (56.8 %) of Grade IV. In the control group, however, there were 0 cases (0 %) of Grade I, 16 cases (36.4 %) of Grade II, 28 cases (63.6 %) of Grade III and 0 cases (0 %) of Grade IV in the control group. The difference was statistically significant (v2 = 42.327, P = 0.000). Correlation of all clinical and MRI parameters with JOA scores is shown in Table 2. The canal grade (paramedian), canal grade (sagittal), and CSA occupying ratio were the top three relevant continuous variables. The ROC curves were used to compare the area under the curve (Table 3) and to calculate the Youden index to determine the optimal cutoff value for diagnosis of OLF-induced myelopathy. These values were 60, 50, and 80 % for the three variables, respectively (Tables 4, 5, 6). Results of analysis of preoperative myelopathy severity and canal grade (paramedian) were displayed in Table 7.

Discussion OLF is a type of pathological heterotopic ossification, and it can cause the slow progress of thoracic myelopathy. It is a common cause of thoracic spinal stenosis and can occur in multiple segments [1]. Clinical data showed that singleTable 2 Factors correlated to preoperative neurological status (preoperative JOA scores) Factor

P value

r coefficient

Age

0.904

0.019

Sex

0.513

0.101

Symptom duration

0.223

OLF location

0.017*

-0.354

OLF level

0.849

-0.029

Degree of occupation on T2-weighted MRI

0.000*

-0.545

Canal grade (sagittal)

0.001*

0.503

Canal grade (midline)

0.031*

0.222

Canal grade (boundary)

0.021*

0.312

Canal grade (paramedian)

0.000*

0.685

CSA occupying ratio

0.011*

0.374

0.187

Categorical variables had their levels represented with numbers as follows: male = 1, female = 2; OLF location—joint capsules = 1, median lamina = 2, joint capsules and median lamina = 3; level— T1–T9 = 1, T10–L1 = 2; degree of occupation—Grade I = 1, Grade II = 2, Grade III = 3 and Grade IV = 4 * Statistically significant by using P \ 0.05

segment OLF only accounted for 40–60 % of cases [15]. In multi-segmental OLF, not every affected segment will cause spinal cord damage; however, it is difficult to determine which segment is involved in causing the spinal deficits. If the responsible segment of OLF involved in spinal cord damage can be accurately determined, acceptable surgical results can be achieved and the range needing decompression can be limited, decreasing the extent of surgical invasion. Concurrent spinal degenerative diseases such as lumbar spinal stenosis and cervical spondylotic myelopathy (CSM) are the novel characteristics of OLF. Park et al. [16] reported that OLF was found in 21 of 68 cases of cervical OPLL. Cervical myelopathy and OLF can both lead to lower-limb dysfunction, and it is therefore difficult to determine whether OLF can cause neurologic deficits when the two conditions are simultaneously present. Sun and Chen [17] analyzed the clinical manifestations and radiological representation of 35 cases of myelopathy resulting from OLF associated with CSM. He concluded that in CSM cases, when the component ratio of the upper extremities in the JOA score was higher than 36 % the possibility of associated OLF should be considered. This method can be used as a screening method to prevent missing the OLF diagnosis and to determine whether OLF participates in spinal cord damage when combined with cervical myelopathy. Its disadvantage is that it is rather subjective with a low accuracy, and cannot be used to determine the segment responsible for the neurologic deficits. Electrophysiological examination may have supplementary diagnostic value for cervical and lumbar nerve damage; however, it is less effective in localizing diagnosis of thoracic myelopathy. Baba et al. [18] adopted the method of spinal cord evoked potential to locate the segment responsible for the thoracic canal stenosis. However, this method is mainly employed for thoracic longitudinal ligament ossification and invasive manipulation near the spinal cord is very risky. What’s more, spinal cord evoked potential can easily be influenced by many factors such as age, anesthetics. Currently, MRI is the major imaging diagnostic method in determining the segment responsible for OLF-induced myelopathy. In clinical practice, OLF is considered asymptomatic if it only compresses the dural sac but has not touched the spinal cord. Therefore, conservative treatment and follow-up is suitable in these cases; when OLF causes spinal cord compression and deformation, surgery should be considered. Our results also showed that all Grade II cases were in the control group, while all Grade IV cases were in the myelopathy group. MRI provides important information on dural sac and spinal cord compression, but the interpretation depends on the experience and subjective judgment of the doctor. For those moderately compressed (Grade III cases), the judgment can be

123

Eur Spine J Table 3 The area under the curve of the three indices Variables

Area

Std. error

Asymptotic sig

95 % CI Lower bound

Upper bound

CSA occupying ratio

0.971

0.017

0.000

0.932

1.000

Canal grade (sagittal)

0.925

0.026

0.000

0.873

0.977

Canal grade (paramedian)

0.990

0.007

0.000

0.000

1.000

Table 4 Cut-off values for the CSA occupying ratio CSA occupying ratio (%)

Sensitivity (%)

Specificity (%)

Diagnostic coincidence rate (%)

\75

80

95.4

87.6

\80

90.9

93.2

92.0

\85

97.7

65.9

82.0

Table 5 Cut-off values for canal grade (paramedian) Canal grade (paramedian) (%)

Sensitivity (%)

Specificity (%)

Diagnostic coincidence rate (%)

\55

86.6

100

93.2

\60

95.5

95.5

95.5

\65

100

63.6

82.0

Table 6 Cut-off values for canal grade (sagittal) Canal grade (sagittal) (%)

Sensitivity (%)

Specificity (%)

Diagnostic coincidence rate (%)

\45

72.7

93.2

83.0

\50

86.4

79.5

83.0

\55

95.4

61.4

77.2

Table 7 Results of analysis of preoperative myelopathy severity and canal grade (paramedian) Preop myelopathy severity

No. of patients

JOA score

Canal grade (paramedian)

C7

25

7.88 ± 0.83

45.6 ± 10.9 %

4–6

16

5.12 ± 0.81

33.7 ± 10.4 %

B3

3

2.67 ± 0.58

21.0 ± 6.1 %

\0.001*

\0.001*

P value

* Statistically significant by using P \ 0.05

123

rather subjective. In the myelopathy group, there were 19 cases of moderate compression (Grade III) with neurologic deficit, of which 18 experienced manifestation alleviation immediately after OLF-removal surgery; only one case did not improve after surgery. In the control group, 28 cases showed obvious dural sac compression blockage but no neurological signs. These results indicate that MRI grading alone may be not very accurate in determining the segment responsible for OLF-induced thoracic myelopathy. Measurement of the residual spinal canal area and sagittal diameter of the ossification segment by performing MRI was adopted earlier in the prognosis study of OLFinduced myelopathy [7, 9], it was also found that the preoperative canal grade(AP canal diameter of stenosed level/ normal canal diameter 9 100) was correlated with the preoperative JOA score [9], but no further analyses were conducted to provide the critical value, thus unable to guide the clinical diagnosis precisely. Measurement on axial T2weighted MRI is not accurate sometimes. For example, when the subarachnoid space is occluded, the signal of the spinal cord is very close to the signal of ossification (or the posterior margin of the vertebrae) on axial T2-weighted MRI (Fig. 3), thus it is not very easy to measure the CSA or sagittal diameter precisely. In comparison with MRI, CT imaging clearly shows bone structures, allowing for quantitative measurement of the bony canal. Moreover, CT can perform axial scan layer by layer and the layer with severest ossification will not be missed. Liu et al. [19] measured the residual spinal canal area based on CT scans and calculated the critical value of OLF-induced thoracic myelopathy. It was concluded that when the residual area was \80 %, the sensitivity of diagnosis of OLF-induced thoracic myelopathy was 93 %, while the specificity and the diagnostic coincidence rate were 95.5 and 93.8 %, respectively. The residual spinal canal area seems to be an accurate method of diagnosis, although it has several disadvantages. First, measurement of the CSA cannot determine the ossification position; for example, bilateral and unilateral ossification would have significantly different degrees of compression despite having the same residual spinal canal area. Second, the CSA can only be measured with the aid of PACS or some kind of software electronically. It is hardly applicable in some regions. In contrast, measurement of the canal diameter can be performed directly on CT images. According to the growth pattern of OLF, ossification usually begins in the capsular portion where the stress is most concentrated, then spreads to the laminar portion, and lastly enlarges anteriorly toward the spinal cord [1]. Generally, ossification of the lateral articular capsule is the most obvious. As the early period of OLF often occurs in the lateral portion of the spinal canal, it may not compress the spinal cord. Meanwhile, the MRI of the myelopathy group showed that in 34.1 % of the cases, there is no spinal

Eur Spine J

Fig. 3 Signal of the spinal cord is very close to the signal of ossification

cord compression in the lateral boundary of the spinal canal, where ossification is the most obvious. Thus, taking the measurements there cannot directly reflect the degree of compression. Ossification of the middle line of the spinal canal is not always obvious; we found only 56.8 % of cases with obvious ossification in that area. So it also cannot be used to reflect the degree of compression. Therefore, the measurement should be performed at a position between the middle line and lateral boundary of the spinal canal. The paramedian point, therefore, is the most convenient site to take the measurement. The canal grade (paramedian) was therefore determined to be an appropriate indicator. In this study, 95.5 % cases showed significant ossification at this position; therefore, this value can directly reflect the degree of compression. Results from this study prove that this indicator has the highest correlation and diagnostic accordance rate, and can be easily measured. In this study, measurement of the canal grade (paramedian) is intuitively simple and accurate, thus reducing measurement subjectivity. Sensitivity and specificity of diagnosis of symptomatic OLF were 95.5 and 95.5 %, respectively. OLF that only compresses the dural sac can also cause myelopathy possibly because the static pressure of OLF can be conducted through the dural sac and cerebrospinal fluid to the spinal cord. As flowing fluid can also exert pressure on the spinal cord at narrow segments, and due to spinal flexion and extension and a variety of small external forces, a large portion of the OLF surface would intermittently contact the spinal cord [20]. Cases with a

canal grade (paramedian) \60 % may have reached the threshold of myelopathy by this biomechanical mechanism despite having no obvious spinal cord compression. In our study, there were two cases that were missed diagnosed; these CT scans showed that the ossification originated from the middle of the lamina and thickened to form a central tuberous mass. The compression here was mainly in the midline of the spinal canal and the canal grade (sagittal) of the two cases were \50 %. Therefore, for this type of OLF, the canal grade (sagittal) may be a better indicator. In order to determine a simple method and to find the responsible segment in OLF, thoracic ventral spinal cord compression, other spinal disorders and multi-segmental OLF were excluded to eliminate interfering factors that might affect JOA scores. A more accurate and easier critical value was proposed for OLF diagnosis, and we established diagnosis procedures for clinical identification of OLF-related segments: The responsible and non-responsible segment should be determined based on a detailed medical history and comprehensive physical examination, in combination with a full-thoracic MRI. Grade IV compression can be considered responsible and surgical treatment should be considered. Axial CT scans will be helpful in measuring the canal grade (paramedian) for most of the Grade III cases. For cases with tuberous ossification near the midline of the spine canal, the canal diameter (sagittal) should also be measured. However, our study has the following limitations: first, the incidence of OLF is low, and single-segment symptomatic OLF is even lower, resulting in small sample size. Second, most images from the control group do not present the entire spine because they are not specific for OLF; thus, certain segments that are compressed heavily may not be included. Additionally, with whole spine imaging available in the myelopathy group while only selective imaging available in the control group, the observers could not be blinded to the patient group. More cases of single-segment OLF and more detailed imaging data of the control group are needed to confirm the results of this study.

Conclusion Spinal canal compromise is relevant to spinal cord deficits in patients with thoracic OLF, and a canal grade (paramedian) measured on axial CT scan can be used to quantify spinal cord deficits. Additionally, a canal grade (paramedian) of \60 % can serve as a critical value for diagnosing OLF-induced myelopathy, especially for Grade III compression on T2-weighted MRI. Conflict of interest

None.

123

Eur Spine J

References 1. Aizawa T, Sato T, Sasaki H et al (2006) Thoracic myelopathy caused by ossification of the ligamentum flavum: clinical features and surgical results in the Japanese population. J Neurosurg Spine 5(6):514–519 2. Shiokawa K, Hanakita J, Suwa H et al (2001) Clinical analysis and prognostic study of ossified ligamentum flavum of the thoracic spine. J Neurosurg Spine 94(2):221–226 3. Gao R, Yuan W, Yang L et al (2013) Clinical features and surgical outcomes of patients with thoracic myelopathy caused by multilevel ossification of the ligamentum flavum. Spine J 13(9):1032–1038 4. Li M, Wang Z, Du J et al (2013) Thoracic myelopathy caused by ossification of the ligamentum flavum: a retrospective study in Chinese patients. J Spinal Disord Tech 26(1):E35–E40 5. Kang KC, Lee CS, Shin SK et al (2011) Ossification of the ligamentum flavum of the thoracic spine in the Korean population: clinical article. J Neurosurg Spine 14(4):513–519 6. Park BC, Min WK, Oh CW et al (2007) Surgical outcome of thoracic myelopathy secondary to ossification of ligamentum flavum. Joint Bone Spine 74(6):600–605 7. Yoon SH, Kim WH, Chung SB et al (2011) Clinical analysis of thoracic ossified ligamentum flavum without ventral compressive lesion. Eur Spine J 20(2):216–223 8. Kuh SU, Kim YS, Cho YE et al (2006) Contributing factors affecting the prognosis surgical outcome for thoracic OLF. Eur Spine J 15(4):485–491 9. Sanghvi AV, Chhabra HS, Mascarenhas AA et al (2011) Thoracic myelopathy due to ossification of ligamentum flavum: a retrospective analysis of predictors of surgical outcome and factors affecting preoperative neurological status. Eur Spine J 20(2):205–215 10. Pascal-Moussellard H, Cabre P, Smadja D et al (2005) Symptomatic ossification of the ligamentum flavum: a clinical series from the French Antilles. Spine (Phila Pa 1986) 30((14):E400– E405

123

11. Zhao Y, Xue Y, Shi N et al (2014) The CT and intraoperative observation of pedicel-ossification tunnel in 151 cases of thoracic spinal stenosis from ossification of ligamentum flavum. Eur Spine J 23(6):1325–1331 12. Lang N, Yuan HS, Wang HL et al (2013) Epidemiological survey of ossification of the ligamentum flavum in thoracic spine: CT imaging observation of 993 cases. Eur Spine J 22(4):857–862 13. Guo Q, Ni B, Yang J et al (2011) Simultaneous ossification of the posterior longitudinal ligament and ossification of the ligamentum flavum causing upper thoracic myelopathy in DISH: case report and literature review. Eur Spine J 20(Suppl 2):S195–S201 14. Wang W, Kong L, Zhao H et al (2007) Thoracic ossification of ligamentum flavum caused by skeletal fluorosis. Eur Spine J 16(8):1119–1128 15. Inamasu J, Guiot BH (2006) A review of factors predictive of surgical outcome for ossification of the ligamentum flavum of the thoracic spine. J Neurosurg Spine 5(2):133–139 16. Park JY, Chin DK, Kim KS et al (2008) Thoracic ligament ossification in patients with cervical ossification of the posterior longitudinal ligaments: tandem ossification in the cervical and thoracic spine. Spine (Phila Pa 1986) 33(13):E407–E410 17. Sun CG, Chen ZQ (2007) Clinical diagnosis of myelopathy resulted from thoracic ossification of the ligamentum flavum associated with cervical spondylosis. J Spine Surg 5:18–21 (in Chinese) 18. Baba H, Tomita K, Kawahara N et al (1992) Spinal cord evoked potentials in thoracic myelopathy with multisegmental vertebral involvement. Spine (Phila Pa 1986) 17(11):1291–1295 19. Liu N, Chen ZQ, Qi Q (2007) Correlation between spinal canal encroachment and neurologic deficits in thoracic ossification of ligamentum flavum. Chin J Orthop 27:481–484 (in Chinese) 20. Olmarker K, Rydevik B, Holm S et al (1989) Effects of experimental graded compression on blood flow in spinal nerve roots. A vital microscopic study on the porcine cauda equina. J Orthop Res 7(6):817–823

A diagnostic study of thoracic myelopathy due to ossification of ligamentum flavum.

We set out to establish a magnetic resonance imaging (MRI) and computed tomography (CT)-based diagnostic method for determining the responsible segmen...
668KB Sizes 2 Downloads 6 Views