Surgical Decompression for Cervical Spondylotic Myelopathy ...

■ Feature Article S P OT L I G H T O N

spine Surgical Decompression for Cervical Spondylotic Myelopathy: Correlation Between Operative Outcomes and MRI of the Spinal Cord CHARALAMBOS A. PAPADOPOULOS, MD*; PAVLOS KATONIS, MD*; PANAYIOTIS J. PAPAGELOPOULOS MD, DSC†; SPYRIDON KARAMPEKIOS, MD‡; ALEXANDER G.HADJIPAVLOU, MD*

abstract Forty-two patients who underwent decompressive surgery for cervical spondylotic myelopathy were studied. The pre– and postoperative magnetic resonance imaging (MRI) findings and the Japanese Orthopedic Association score were analyzed to evaluate whether the different appearances of intramedullary high-signal intensity on T2-weighted MRI are related to the surgical prognosis. Magnetic resonance imaging signal intensities were classified as type 0 if no intramedullary high-signal intensity on T2-weighted images was noted, type 1 if high-signal intensity involved only one segment, and type 2 if high-signal intensity extended over two segments. Statistical analyses of the recovery ratio showed that type 0 and type 1 intramedullary high-signal intensity indicates better prognosis than type 2.

ervical spondylotic myelopathy, which may result from spinal cord compression, disturbed blood supply, or both, develops due to degenerative changes in the cervical spine after the middle years of life.1,2 In the past, the diagnosis of cervical spondylotic myelopathy was dependent on radiography, myelography, computed tomography (CT), and CT myelography. With the advent of magnetic resonance imaging (MRI), diagnostic imaging of spinal cord compression has developed markedly. Magnetic resonance imaging depicts anatomically how the spinal cord is compressed and reflects the pathological changes within the spinal cord by

C

OCTOBER 2004 | Volume 27 • Number 10

showing a change in the signal intensity of the spinal cord.3-7 A number of authors have reported that intramedullary high-signal intensity is a predictor of poor recovery after surgical decompression.8-12 However, some authors have reported no correlation between the surgical outcome and intramedullary highsignal intensity.3,4,7,13 The relation between high-signal intensity changes and clinical symptoms is not clear due to the various types of disease, crisis pattern, and sampling bias, and postoperative clinical symptoms may vary depending on the type of surgery, anterior or posterior, performed. Therefore, we tried to determine whether

the different preoperative appearances of intramedullary high-signal intensity on T2weighted MRI are related to differences in surgical prognosis.

MATERIALS AND METHODS Between January 1999 and January 2003, 42 (27 men and 15 women) consecutive patients with multilevel cervical compression myelopathy were treated by decompression surgery. Average age of patients was 57.5 years (range: 36-82 years). Eighteen patients underwent posterior decompression by laminectomy and lateral mass plate fixation; 7 underwent anterior decompression by corpectomy, excision of posterior longitudinal ligament, bone graft or titanium cage, and plate fixation; and 17 underwent both anterior and posterior procedures. All patients underwent MRI examination preoperatively and 6 months postoperatively. Neurologic status was evaluated From the *Departments of Orthopedics and Traumatology and ‡Radiology, University of Crete, Heracleion, Crete; and the †First Department of Orthopedics, Athens University Medical School, Athens, Greece. Reprint requests: Charalambos Papadopoulos, MD, Iras 39, Gerakas, Athens, Greece.

1087

■ Feature Article

effacement according to axial and sagittal T1- and T2-weighted MRIs. Decompression was considered incomplete if either anterior or posterior cord impingement or effacement was identified. After evaluation of all sagittal MRI sections through the cervical cord, the findings were classified into two major patterns. One pattern showed a high-signal intensity in one segment and in the other pattern, the high-signal intensity was extended over ⭓2 segments. Patients were classified as type 0 if no intramedullary signal

1 Figure 1: Cervical MRI of a 67-year-old man with 35% cord compression ratio and preoperative JOA score of 14 points. Preoperative sagittal T2-weighted image shows type 1 intramedullary high-signal intensity (arrow) (A). One year postoperative sagittal T2-weighted image shows residual intramedullary high-signal intensity. Recovery ratio was 72% (B).

With the advent of magnetic resonance imaging, diagnostic imaging of spinal cord compression has developed markedly.

on T2-weighted MRI was noted, type 1 if the high-signal intensity was limited to one segment, and type 2 if the high-signal intensity was extended over ⭓2 segments. Student t test was used for statistical analysis.

RESULTS 2 Figure 2: Cervical MRI of a 52-year-old woman with 46% cord compression ratio, and preoperative JOA score of 12 points. Preoperative sagittal T2-weighted image shows type 1 intramedullary high-signal intensity (arrow) (A). Eight-month postoperative sagittal T2-weighted image shows withdrawal of the high-signal intensity. Recovery ratio was 78% (B).

using the modified Japanese Orthopedic Association (JOA) score for clinical myelopathy and recovery was determined using Hirabayashi’s formula.14,15 One radiologist (S.K.) who was blinded to clinical status and one of the authors (C.A.P.) evaluated MRIs for intramedullary highsignal intensity, spinal cord deformity, and

1088

adequate decompression. Cord deformity was expressed by the ratio of the sagittal diameter divided by the transverse diameter of the spinal cord observed on T1weighted transverse images at the site of maximal cord compression.6 Decompression was considered complete if the cord was free of any impingement or

In 16 patients, no high-signal intensity was detected on T2-weighted MRIs (type 0), 13 patients with high-signal intensity were limited to one segment (type 1), and 13 patients demonstrated high-signal intensity extended over ⭓2 or more segments (type 2). None of the patients demonstrated intramedullary low-signal intensity on T1weighted sequences. The preoperative JOA type 0, 1, and 2 scores were 13.9⫾1.6, 12.69⫾0.99, and 10.2⫾0.9, respectively. The mean preoperative cord compression ratios of types 0, 1, and 2 were 49.02%⫾11.03, 41.13%⫾12.62, and 31.17%⫾9.11, respectively. Statistical

ORTHOPEDICS |

www.orthobluejournal.com

CERVICAL SPONDYLOTIC MYELOPATHY | PAPADOPOULOS ET AL

analyses showed no significant difference in age, sex, duration of symptoms, and the surgical procedure between the three groups. However, the mean cord compression ratio and the preoperative JOA type 2 score were more severe than type 0 and type 1. Postoperatively, the mean cord compression ratio of type 0, 1, and 2 was 85.16%⫾10.13, 81.27%⫾10.26, and 66.32%⫾9.19, respectively. Six months to 2 years postoperatively, the recovery ratio of all 42 patients was 57.3%⫾14.17. The recovery ratios of types 0, 1, and 2 were 66.14%⫾17.60, 71.5%⫾12.44, and 34.26%⫾12.48, respectively. Comparing the recovery ratio of the different types of high-signal intensity on T2-weighted images, a significant difference was found between types 2 and 1 (P⬍.001) and types 2 and 0 intramedullary high-signal intensity (P⬍.001); however, no significant difference was noted between types 0 and 1 (P⫽.232). Significant differences of the final JOA score were noted between type 2 and type 0 or type 1 intramedullary highsignal intensity (P⬍.001) (Table). In 4 (15%) of 26 patients with intramedullary high-signal intensity, T2weighted MRI demonstrated improved signal intensity after complete decompression. Of the 4 patients, 1 underwent anterior decompression, 1 underwent posterior decompression, and 2 underwent both anterior and posterior decompression. The mean recovery ratio was 77.53%⫾18.32. All 4 had type 1 intramedullary high-signal intensity on preoperative T2-weighted MRIs.

3 Figure 3: Cervical MRI of 72-year-old woman with abnormal cervical curvature, 26% cord compression ratio, and preoperative JOA score of 11 points. Preoperative sagittal T2-weighted image shows type 2 intramedullary high-signal intensity (arrows) (A). Fourteen months postoperative sagittal T2-weighted image shows stationary findings of the intramedullary high-signal intensity (arrows). Recovery ratio was 41% (B).

been the treatment of choice for patients with cervical spondylotic myelopathy with neurological deficits. An improvement rate of 48%-87% after posterior or anterior decompression or a combination of both has been reported.14-16 In our series the improvement rate was 57.3%, which is compatible with past studies. It has been speculated that a change in

TABLE

Mean JOA Scores, Cord Compression Ratio, and Recovery Rates in Intramedullary High-Signal Intensity

DISCUSSION Cervical spondylosis is a degenerative disease in which its compressive effects result in cervical spondylotic myelopathy, which includes herniation of intervertebral disks, posterior osteophytes, and infolding of hypertrophied ligamentum flavum. These lesions cause impingement on the cervical spinal cord. Surgical decompression of the narrowed spinal canal or spondylotic protrusion has


the signal intensity of the spinal cord observed on MRI reflects pathologic changes in the spinal cord according to autopsy studies17,18 of the pre-MRI period in patients with compression myelopathy. Takahashi et al19 reported that a change in high-signal intensity of the spinal cord reflected myelomalacia or cord gliosis secondary to long-standing compression of

Preoperatively JOA score Cord compression ratio (%) Postoperatively JOA score Cord compression ratio (%) Recovery rate (%)

Type 0 (n=16)

Type 1 (n=13)

Type 2 (n=13)

13.9⫾2.7 53.12⫾10.12

12.69⫾2.2 48.23⫾9.87

10.2⫾1.9 34.57⫾8.13

15.9⫾1.9 85.16⫾11.13 66.14⫾22.14

14.38⫾2.1 81.27⫾10.26 71.5⫾20.73

12.38⫾1.8 66.32⫾9.19 34.6⫾11.31

Abbreviation: JOA=Japanese Orthopaedic Association.

1089

■ Feature Article

What is already known on this topic ■ Surgical decompression of the narrowed spinal canal or spondylotic protrusion has been the treatment of choice for patients with cervical spondylotic myelopathy and neurological deficits. ■ Various factors such as age at surgery, duration of myelopathy, MRI signal changes, and the transverse area of the spinal cord at the level of maximum compression have been reported to predict the surgical outcome.

What this article adds ■ Preoperative signal intensity of the cervical spinal cord on MRI could be classified into three types—type 0, no intramedullary high-signal intensity on T2-weighted images; type 1, (focal) high-signal intensity involves only one segment; and type 2 (multisegmental), high-signal intensity extends over ⬎2 segments. ■ Not all patients with cervical spondylotic myelopathy and intramedullary high-signal intensity on T2-weighted MRI have poor postoperative outcome. ■ Patients with cervical spondylotic myelopathy and type 1 (focal) high-signal intensity have the same surgical outcome as those without intramedullary high-signal intensity (type 0). ■ Independent of the decompressive procedure, age, sex, and duration of symptoms, patients with multisegmental areas of high-signal intensity (type 2) in the cervical spinal cord will have poor surgical outcome.

the spinal cord, which is an indicator of poor prognosis. Ramanauskas et al20 divided myelomalacia into three stages and speculated that in the early stage, a change in signal intensity of the spinal cord observed on MRI reflected cord edema, in the intermediate stage, a change reflected cystic necrosis of the central gray matter after prolonged cord edema, and cavity or syrinx in the late stage. Ramanauskas et al20 also reported that in the early and intermediate stages, the spinal cord showed a high-signal intensity on T2-weighted sequences, and that in the late stage, a lowsignal intensity on T1-weighted sequences and a high-signal intensity on T2-weighted sequences were noted, which is an indicator of poor prognosis. Al-Mefty et al21 reported that low-signal changes on T1weighted sequences and high-signal changes on T2-weighted sequences indicated cystic necrosis or secondary syrinx. Mehalic et al9 reported that high-signal changes on T2-weighted sequences were nonspecific and indicated edema, inflammation, vascular ischemia, gliosis, or myelomalacia, which might be postopera-

1090

tively reversible. Therefore, the controversy of predicting the value of intramedullary high-signal intensity on T2-weighted MRI might be attributed to the reversible and

Patients with focal areas of high-signal intensity will have the same surgical outcome as those without intramedullary high-signal intensity.

studies. Wada et al4,22 found a significant correlation between multisegmental areas of high-signal intensity on T2-weighted MRI and areas of low-signal intensity on T1-weighted images. This significant correlation supports the cavitation theory, because there is a general consensus that increased signal on T2 with decreased signal on T1 is suggestive of a cavity in the spinal cord. They, therefore, concluded that most multisegmental areas of high-signal intensity on T2-weighted MRIs may represent irreversible changes in the spinal cord, such as cavitations or cystic necrosis. However, focal areas of high-signal intensity on T2-weighted MRI were not of a valuable prognostic factor because they represent minor pathologic changes of the central spinal cord, such as edema. In our study, the poor prognosis of multisegmental highsignal intensity on T2-weighted MRIs supports this subdivision of two types of intramedullary high-signal intensity.

CONCLUSION Not all patients with intramedullary high-signal intensity on T2-weighted MRI have poor postoperative outcome. Patients with focal areas of high-signal intensity will have the same surgical outcome as those without intramedullary high-signal intensity. Therefore, independent of the decompressive procedure, age, sex, and duration of symptoms, patients with multisegmental areas of high-signal intensity in the cervical spinal cord will have poor surgical outcome.

REFERENCES irreversible components of intramedullary high-signal intensity as a whole. In this study, we subdivided intramedullary high-signal intensity into two types. In type 1, the high-signal intensity is located at one level, and in type two, high-signal intensity extended in two or more segments. This subdivision was based on our own observations and two other

1. Crandall PH, Batzdorf U. Cervical spondylotic myelopathy. J Neurosurg. 1966; 25:57-66. 2. Taylor AR. Vascular factors in the myelopathy associated with cervical spondylosis. Neurology. 1964; 14:62-68. 3. Morio Y, Yamamoto K, Kuranobu K, Murata M, Tuda K. Does increased signal intensity of the spinal cord on MR images due to cervical myelopathy predict prognosis? Arch Orthop Trauma Surg. 1994; 113:254-259. 4. Wada E, Ohmura M, Yonenobu K. Intramedullary changes of the spinal cord in

ORTHOPEDICS |

www.orthobluejournal.com

CERVICAL SPONDYLOTIC MYELOPATHY | PAPADOPOULOS ET AL

cervical spondylotic myelopathy. Spine. 1995; 20:2226-2232. 5. Kasai Y, Uchida A. New evaluation method using preoperative magnetic resonance imaging for cervical spondylotic myelopathy. Arch Orthop Trauma Surg. 2001; 121:508-510. 6. Okada Y, Ikata T, Yamada H, Sakamoto R, Katoh S. Magnetic resonance imaging study on the results of surgery for cervical compression myelopathy. Spine. 1993; 18:2024-2029. 7. Yone K, Sakou T, Yanase M, Ijiri K. Preoperative and postoperative magnetic resonance image evaluations of the spinal cord in cervical myelopathy. Spine. 1992; 17(suppl):S388-S392. 8. Matsuda Y, Miyazaki K, Tada K, et al. Increased MR signal intensity due to cervical myelopathy. Analysis of 29 surgical cases. J Neurosurg. 1991; 74:887-892. 9. Mehalic TF, Pezzuti RT, Applebaum BI. Magnetic resonance imaging and cervical spondylotic myelopathy. Neurosurgery. 1990; 26:217-227. 10. Takahashi M, Sakamoto Y, Miyawaki M, Bussaka H. Increased MR signal intensity secondary to cervical cord compression. Neuroradiology. 1987; 29:550-556. 11. Harada A, Mimatsu K. Postoperative changes


in the spinal cord in cervical spondylotic myelopathy demonstrated by magnetic resonance imaging. Spine. 1992; 17:1275-1280. 12. Hukuda S, Xiang LF, Imai S, Katsuura A, Imanaka T. Large vertebral body, in addition to spinal canal are risk factors for cervical myelopathy. J Spinal Disord. 1996; 9:177-186. 13. Okais N, Moussa R, Hage P. Value of increased MRI signal intensity in cervical arthrosis in myelopathies. [French]. Neurochirurgie. 1997; 43:285-291. 14. Hirabayashi 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. 15. Houten JK, Cooper PR. Laminectomy and posterior cervical plating for multilevel cervical spondylopathy and ossification of the posterior longitudinal ligament: effects on cervical alignment, spinal cord compression, and neurological outcome. Neurosurgery. 2003; 52:1081-1088. 16. Swank ML, Sutterlin CE III, Bossons CR, Dials BE. Rigid internal fixation with lateral mass plates in multilevel anterior and posterior reconstruction of the cervical spine. Spine. 1997; 22:274-282.

17. Hashizume Y, Iijima S, Kishimoto H, Yanagi T. Pathology of spinal cord lesions caused by ossification of the posterior longitudinal ligament. Acta Neuropathol (Berl). 1984; 63:123130. 18. Ono K, Ota H, Tada K, et al. Cervical myelopathy secondary to multiple spondylotic protrusions: a clinicopathologic study. Spine. 1977; 2:109-125. 19. Takahashi M, Yamashita Y, Sakamoto Y, Kojima R. Chronic cervical cord compression: clinical significance of increased signal intensity on MR images. Radiology. 1989; 173:219-224. 20. Ramanauskas WL, Wilner HI, Metes JJ, Lazo A, Kelly JK. MR imaging of compressive myelomalacia. J Comput Assist Tomogr. 1989; 13:399-404. 21. Al-Mefty O, Harkey LH, Middleton TH, Smith RR, Fox JL. Myelopathi cervical spondylotic lesions demonstrated by magnetic resonance imaging. J Neurosurg. 1988; 68:217-222. 22. Wada E, Yonenobu K, Suzuki S, Kanazawa A, Ochi T. Can intramedullary signal change on magnetic resonance imaging predict surgical outcome in cervical spondylotic myelopathy? Spine. 1999; 24:455-462.

1091