Ossification of the posterior longitudinal ligament

Transcription

1 J Neurosurg Spine 13: , 2010 Using the T2-weighted magnetic resonance imaging signal intensity ratio and clinical manifestations to assess the prognosis of patients with cervical ossification of the posterior longitudinal ligament Clinical article Lin-Fe n g Wa n g, M.D., Yi n g-ze Zh a n g, M.D., Yo n g Sh e n, M.D., Ya n-li n g Su, M.D., Jia-Xin Xu, M.D., We n-yu a n Di n g, M.D., a n d Yi n g-hu a Zh a n g, M.D. Department of Spinal Surgery and MR Imaging, Third Hospital of HeBei Medical University, Shijiazhuang, China Object. The aim of this study was to investigate the clinical significance of both the signal intensity ratio obtained from MR imaging and clinical manifestations on the prognosis of patients with cervical ossification of the posterior longitudinal ligament. Methods. The authors retrospectively reviewed the records of 58 patients with cervical ossification of the posterior longitudinal ligament who underwent cervical laminoplasty from February 1999 to July Magnetic resonance imaging (1.5-T) was performed in all patients before surgery. Sagittal T2-weighted images of the cervical spinal cord compressed by the ossified posterior longitudinal ligament showed increased intramedullary signal intensity, whereas the sagittal images obtained at the C7 T1 disc levels were of normal intensity. The signal intensity ratio between regions of intramedullary increased signal intensity and the normal C7 T1 disc level was calculated based on the signal intensity values generated from the MR imaging workstation. Patients were divided into 3 groups according to their signal intensity ratio (high, intermediate, and low signal intensity groups). Results. There were significant differences between the 3 groups regarding recovery rate (p < 0.001), age (p = 0.022), duration of disease (p = 0.001), Babinski sign (p < 0.001), ankle clonus (p < 0.001), and both pre- and postoperative Japanese Orthopaedic Association score (p < 0.001). There was no significant difference in sex among the 3 groups (p = 0.391). Conclusions. Patients with low signal intensity ratios that changed on T2-weighted imaging experienced a good surgical outcome. Low increased signal intensity might reflect mild neuropathological alteration in the spinal cord and greater recuperative potential. An increased signal intensity ratio with positive pyramidal signs indicates less recuperative potential of the spinal cord and a poor surgical outcome. (DOI: / SPINE09887) Ke y Wo r d s clinical manifestation magnetic resonance imaging ossification of the posterior longitudinal ligament prognosis signal intensity ratio Ossification of the posterior longitudinal ligament is a common cause of cervical spinal cord dysfunction. Magnetic resonance imaging is an invaluable examination tool for patients with compressive cervical myelopathy. Magnetic resonance imaging can show not only the degree of spinal canal stenosis but also the intramedullary state of the spinal cord in detail. 14 Furthermore, increased signal intensity is often noted in Abbreviations used in this paper: IQR = interquartile range; JOA = Japanese Orthopaedic Association; OPLL = ossification of the posterior longitudinal ligament; ROI = region of interest. patients with cervical OPLL. However, the significance of increased signal intensity for prognosis remains controversial. 1,4,8,11,16,17,20 In this study, we attempt to quantify the T2 signal intensity on MR imaging by using both a signal intensity ratio and clinical manifestations to assess the prognosis of the disease. Methods Study Population One hundred and two patients with cervical OPLL who underwent expansive open-door laminoplasty for J Neurosurg: Spine / Volume 13 / September

2 L. F. Wang et al. cervical compressive myelopathy at our medical center from February 1999 to July 2007 were studied retrospectively. Only 58 patients were able to be followed up for more than 12 months (mean 14.6 months, range months). Of the 58 patients, 41 were men and 17 were women, with a mean age of years (range years). Forty-one patients underwent 5 levels of decompression (C3 7), and 17 patients underwent 4 levels of decompression (C3 6, 12 patients; C4 7, 5 patients). Patients with traumatic cervical myelopathy, neoplasia, vitamin B12 deficiency, amyotrophic lateral sclerosis, arteriovenous malformations, congenital anomalies, demyelinating disease, cervical radiation therapy, collagen vascular disease, and vasculitis were excluded from the study. Information regarding age, sex, duration of disease, Hoffmann sign, Babinski sign, ankle clonus, and knee jerk was recorded before surgery. We used a modified JOA scoring system to determine pre- and postoperative functional status. The recovery rate after treatment was calculated using the Hirabayashi method: (postoperative JOA score preoperative score)/(17 preoperative score) 100%. 5 All patients were treated using cervical expansive laminoplasty. Imaging Procedures All patients underwent high-resolution MR imaging using a 1.5-T system (Siemens Magnetom Symphony) before surgery. The sagittal T1- and T2-weighted images of the cervical spinal cord were obtained using a spin echo sequence for T1-weighted images and a fast spin echo sequence for T2-weighted images. A surface coil was used. Slice width was 4 mm and the acquisition matrix was The sequence parameters were TR 612 msec/te 13 msec for T1-weighted images, and TR 2400 msec/te 114 msec for T2-weighted images. Sagittal T2- weighted images of the compressed spinal cord showing increased intramedullary signal intensity were obtained, from which a 0.05-cm 2 ROI was delineated. Sagittal T2- weighted images were obtained of the normal cord at the C7 T1 disc level were obtained, and a 0.3 cm 2 ROI was delineated. If high intramedullary signal intensity was not present, then we defined the ROI from the severely compressed cord. Signal intensity values of both compressed and normal cord can be obtained from the MR imaging workstation and the signal intensity ratio can be calculated. The determination of the ROI is the result of a balance of a number of factors; for instance, too large an area would not include all patients in the group, whereas too small an area would jeopardize the accuracy of the signal intensity value. Statistical Analysis The 58 patients were divided into 3 groups using hierarchical clustering analysis of the signal intensity ratios. There were 23 patients in Group 1, with a low signal intensity ratio (< 1.396). Twenty patients were included in Group 2, with an intermediate signal intensity ratio ( and < 1.689). Fifteen patients were included in Group 3, and had a high signal intensity ratio ( 1.689). An ANOVA was used to compare continuous variables in Group 1, the Pearson chi-square test was used to compare categorical variables in Group 2, and the Kruskal-Wallis H-test was applied in Group 3. A probability value < 0.05 was considered statistically significant. In addition, a multiple comparisons test was performed in the 3 groups. Multiple linear regression equations were used to determine the relationship between each independent variable and the recovery rate (R entry = 0.10, R removal = 0.15). Statistical analyses were performed using SPSS version 13.0 (SPSS, Inc.). Results In this study, the median and IQR of the JOA scores before surgery were 8.00 and 5.25, respectively; and 5.25, respectively, at 6 months after surgery; and and 5.30, respectively, at the final follow-up evaluation. The median and IQR of the recovery rate were 53% and 47%, respectively, at the final follow-up. The range of the signal intensity ratio was for all patients. Figure 1 gives examples of MR images and CT scans for each signal intensity ratio group. There was a statistically significant difference in age (p = 0.022) among the 3 groups. The Student-Newman- Keuls test showed a significant difference in mean age between Groups 1 and 2 (p < 0.05) and Groups 1 and 3 (p < 0.05), but no significant difference between Groups 2 and 3 (p > 0.05; Table 1). The statistical analysis for the Hoffmann sign and knee jerk show no significant difference among the 3 groups, so these 2 variables were removed from the study. Using the Pearson chi-square test, a statistically significant difference was found in the presence of a Babinski sign and ankle clonus among the different groups (p < 0.001), but no significant difference in sex (p = 0.391; Table 2). After further adjusting the test level for Babinski sign and ankle clonus, a multiple comparisons test showed no significant difference in the presence of a Babinski sign between Groups 1 and 2 (p = 0.381), but a significant difference between Groups 1 and 3 (p < 0.001) and Groups 2 and 3 (p = 0.001). Although no significant difference in ankle clonus was found between Groups 1 and 2 (p = 0.095), a significant difference was found between Groups 1 and 3 (p < 0.001) and Groups 2 and 3 (p = 0.006; Table 3). The Kruskal-Wallis H-test showed significant differences among the 3 groups for duration of disease (p = 0.001), pre- and postoperative JOA score (p < 0.001), and recovery rate (p < 0.001; Table 4). A subsequent Bonferroni test for duration of disease showed no significant difference between Groups 1 and 2 (p = 0.448), but showed a significant difference between Groups 1 and 3 (p = 0.001) and Groups 2 and 3 (p = 0.002; Table 5). The Bonferroni test for preoperative JOA score showed a significant difference between Groups 1 and 2 (p = 0.006), Groups 1 and 3 (p < 0.001), and Groups 2 and 3 (p = 0.001). This same test for postoperative JOA score showed a significant difference between Groups 1 and 2 (p = 0.001), Groups 1 and 3 (p < 0.001), and Groups 2 and 3 (p < 0.001). The comparison of recovery rates showed a significant difference between Groups 1 and 2 (p = 0.002), Groups 1 and 3 (p < 0.001), and Groups 2 and 3 (p < 0.001). 320 J Neurosurg: Spine / Volume 13 / September 2010

3 Imaging signal intensity ratio and prognosis in cervical OPLL TABLE 1: Comparison of mean age among the 3 groups Group No. Mean Age (yrs) F, p Value* ± , ± ± 8.87 * Among the 3 groups. Statistically significant difference between Groups 1 and 2 (Student- Newman-Keuls test). Statistically significant difference between Groups 1 and 3 ((Student- Newman-Keuls test). Fig. 1. Sagittal MR images (A C) and corresponding CT scans (a c) of representative patients from each group. A: The signal intensity ratio of this patient at the C4 5 disc level is 1.09 (Group 1). B: The signal intensity ratio of this patient at C-5 is 1.68 (Group 2). C: The signal intensity ratio of this patient at C-4 is 1.92 (Group 3). a c: Corresponding CT scans show severe canal narrowing resulting from OPLL. By using multiple linear regression we get the following equation: recovery rate = age duration of disease signal intensity ratio preoperative JOA score Babinski sign ankle clonus. The R value at entry was 0.10, the R value at removal was 0.15; r = 0.887, r 2 = Discussion Increased signal intensity of the spinal cord on T2- J Neurosurg: Spine / Volume 13 / September 2010 weighted MR imaging is often observed in patients with OPLL. With recent developments and advances in MR imaging techniques and software, we can quantify various signal intensities of the spinal cord in these patients. It has been found that low intensity signal change on T1-weighted MR imaging is a prognostic factor for poor outcome after surgical treatment of cervical compressive myelopathy. 1,8,10,13 Many researchers have investigated the association between increased signal intensity and surgical outcomes. Some research showed that patients with increased signal intensity have a poor prognosis after surgery. 9,15,18,19 However, others found no correlation between surgical outcome and intramedullary high signal intensity on T2-weighted imaging. 1,8,11,13 To the best of our knowledge, Takahashi et al. 18,19 first reported the MR findings of intramedullary high signal intensity in patients with cervical spondylotic myelopathy. These investigators considered intramedullary high signal intensity as an indicator of poor prognosis. Some authors concluded that high signal change on T2-weighted imaging was nonspecific and indicated edema, inflammation, vascular ischemia, gliosis, or myelomalacia. 15,20 Al-Mefty et al. 2 reported that high signal intensity on T2-weighted imaging reflected myelomalacia, and that low signal change on T1-weighted imaging indicated cystic necrosis or secondary syrinx. In this study, we attempted to use quantitative data to investigate the relationship between signal intensities, clinical manifestations, and prognosis. We have noticed that the T2-weighted imaging signal intensity value of the compressed spinal cord is different for each patient. However, the signal intensity value of the normal cord is different for each patient as well. As such, we do not believe that increased signal intensity value alone can objectively reflect the increase in the signal intensity. Therefore, we use the signal intensity ratio to balance the baseline difference in values, given that every patient has a different T2-weighted imaging signal intensity value of the normal spinal cord at baseline. Houten and Noce 6 reported that, in patients with cervical myelopathy treated surgically, the Hoffmann sign was more prevalent and more likely to be observed in patients with less severe neurological deficits than the Babinski sign, which was also found in our study. The statistical analysis for the Hoffmann sign and knee jerk show no significant difference among the 3 groups, so the two variables were removed from the study. In this patient series, presence of a Babinski sign and ankle clo- 321

4 L. F. Wang et al. TABLE 2: Comparison of Babinski sign, ankle clonus, and sex in each group* Babinski Sign Sex Ankle Clonus Variable Negative Positive Female Male Negative Positive Group Group Group total χ p Value < <0.001 * Pearson chi-square test. TABLE 4: Comparison of duration of disease, pre- and postoperative JOA score, and recovery rate in each group* Variable Duration of Disease Preop JOA Score Postop JOA Score Recovery Rate Group (16.00) 9.00 (2.00) (2.00) 0.75 (0.26) Group (14.25) 8.00 (1.75) (4.00) 0.56 (0.42) Group (70.00) 7.00 (1.00) 9.00 (2.00) 0.20 (0.15) c p Value <0.001 <0.001 <0.001 * Kruskal-Wallis H-test. Values for the 3 groups given as median (IQR). nus, duration of disease, and age-related increase of the signal intensity ratio was found. In particular, a statistically significant difference emerged in the recovery rate and pre- and postoperative JOA score between all groups, demonstrating that as the signal intensity ratio increases, the recuperative potential of the spinal cord gradually decreases. In a neuropathological study of patients with cervical spondylotic myelopathy, there appeared to be a common pattern of lesion progression, from mild to severe alteration of the spinal cord. 7 In studies of histopathological examination and MR imaging, increased signal intensity without signal change on T1-weighted images appeared nonspecifically in mildly altered lesions such as loss of nerve cells, gliosis, edema in gray matter, Wallerian degeneration, demyelination, and edema in white matter Increased signal intensity of the spinal cord on T2-weighted images has been considered to include a wide spectrum of compressive myelomalacic pathologies and reflects a wide range of spinal cord recuperative potential. 7 Most previous reviews divided groups mostly according to whether there was any intramedullary signal intensity abnormality on T2-weighted imaging, but did not consider various signal intensities on T2-weighted imaging. Thus, we speculated that the selection bias of patients with different increased signal intensities caused the controversy regarding the relationship between increased signal intensity and prognosis. Based on our findings and previous studies, we speculate that a low signal intensity ratio reflects mild neuropathological alteration in the spinal cord and greater recuperative potential, and a high signal intensity ratio with the presence of pyramidal signs TABLE 3: Comparison of the Babinski sign and ankle clonus in each group* Babinski Sign Ankle Clonus Groups Compared c 2 p Value c 2 p Value 1 & & & < <0.001 * Multiple comparisons test. αʹ = reflects severe alteration and less recuperative potential. In this study, there were 4 patients with multisegmental high-intensity change on T2-weighted MR imaging in Group 3. Among the 4 patients, 3 had low-intensity signal change on T1-weighted imaging, and the mean recovery rate was 14% at the final follow-up. In accordance with the conclusions of Fernandez de Rota et al., multisegmental high-intensity change on T2-weighted MR imaging indicated a poor functional recovery. 4,16 However, in our study, these 4 cases are not sufficient to permit us to draw this conclusion. To our knowledge, this is the first report to demonstrate a significant association between preoperative increased signal intensity, clinical manifestations, and outcome of surgical treatment of cervical compressive myelopathy due to OPLL, based on the method of using the signal intensity ratio. In addition, we used multiple linear regression analysis to study the relationship between recovery rate and other variables. In our study, the recovery rate was correlated with age, Babinski sign, duration of disease, and ankle clonus (r = 78.6%). The results show that the method of quantifying signal intensity ratio and clinical manifestations can reflect surgical outcome and prognosis in patients with cervical OPLL in a short follow-up duration. There are some limitations to our study. We do not have enough cases of multisegmental increased signal intensity on T2-weighted MR imaging. Moreover, the lack of a multicenter study and short follow-up duration limit the statistical power of our analysis. In addition, each high-resolution MR imaging system has different characteristics and working parameters. Thus, these limitations suggest that further study of these questions is needed. Conclusions A low signal intensity ratio reflects mild neuropathological alteration in the spinal cord and greater recuperative potential, and patients might experience a good surgical outcome. In contrast, a high signal intensity ratio with the presence of a pyramidal sign reflects severe alteration in the spinal cord and less recuperative potential, and patients may experience a poor surgical outcome. This method should be a reliable way to quantify the signal intensity ratio with clinical manifestations to reflect sur- 322 J Neurosurg: Spine / Volume 13 / September 2010

5 Imaging signal intensity ratio and prognosis in cervical OPLL TABLE 5: Comparison of duration of disease, pre- and postoperative JOA score, and recovery rate among groups* Duration of Disease Preop JOA Score Postop JOA Score Recovery Rate Groups Compared z value p Value z value p Value z value p Value z value p Value 1 & & < < & < < <0.001 * Bonferroni test. αʹ = gical outcome and prognosis in patients with cervical OPLL and a short follow-up duration. However, patients with multisegmental high-intensity change on T2-weighted MR imaging would probably show a very poor prognosis after surgery. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Shen. Acquisition of data: Wang, Xu. Analysis and interpretation of data: Wang, Zhang. Drafting the article: Wang. Critically revising the article: Shen, Zhang, Su, Ding. Reviewed final version of the manuscript and approved it for submission: all authors. Statistical analysis: Xu. Administrative/ technical/material support: Zhang. References 1. Alafifi T, Kern R, Fehlings M: Clinical and MRI predictors of outcome after surgical intervention for cervical spondylotic myelopathy. J Neuroimaging 17: , Al-Mefty O, Harkey LH, Middleton TH, Smith RR, Fox JL: Myelopathic cervical spondylotic lesions demonstrated by magnetic resonance imaging. J Neurosurg 68: , Chatley A, Kumar R, Jain VK, Behari S, Sahu RN: Effect of spinal cord signal intensity changes on clinical outcome after surgery for cervical spondylotic myelopathy. J Neurosurg Spine 11: , Fernández de Rota JJ, Meschian S, Fernández de Rota A, Urbano V, Baron M: Cervical spondylotic myelopathy due to chronic compression: the role of signal intensity changes in magnetic resonance images. J Neurosurg Spine 6:17 22, 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 6: , Houten JK, Noce LA: Clinical correlations of cervical myelopathy and the Hoffmann sign. J Neurosurg Spine 9: , Ito T, Oyanagi K, Takahashi H, Takahashi HE, Ikuta F: Cervical spondylotic myelopathy. Clinicopathologic study on the progression pattern and thin myelinated fibers of the lesions of seven patients examined during complete autopsy. Spine 21: , Mastronardi L, Elsawaf A, Roperto R, Bozzao A, Caroli M, Ferrante M, et al: Prognostic relevance of the postoperative evolution of intramedullary spinal cord changes in signal intensity on magnetic resonance imaging after anterior decompression for cervical spondylotic myelopathy. J Neurosurg Spine 7: , 2007 J Neurosurg: Spine / Volume 13 / September Matsuda Y, Miyazaki K, Tada K, Yasuda A, Nakayama T, Murakami H, et al: Increased MR signal intensity due to cervical myelopathy. Analysis of 29 surgical cases. J Neurosurg 74: , Matsuyama Y, Kawakami N, Yanase M, Yoshihara H, Ishiguro N, Kameyama T, et al: Cervical myelopathy due to OPLL: clinical evaluation by MRI and intraoperative spinal sonography. J Spinal Disord Tech 17: , Mehalic TF, Pezzuti RT, Applebaum BI: Magnetic resonance imaging and cervical spondylotic myelopathy. Neurosurgery 26: , Mizuno J, Nakagawa H, Inoue T, Hashizume Y: Clinicopathological study of snake-eye appearance in compressive myelopathy of the cervical spinal cord. J Neurosurg 99 (2 Suppl): , Morio Y, Teshima R, Nagashima H, Nawata K, Yamasaki D, Nanjo Y: Correlation between operative outcomes of cervical compression myelopathy and mri of the spinal cord. Spine 26: , Ohshio I, Hatayama A, Kaneda K, Takahara M, Nagashima K: Correlation between histopathologic features and magnetic resonance images of spinal cord lesions. Spine 18: , Okada Y, Ikata T, Yamada H, Sakamoto R, Katoh S: Magnetic resonance imaging study on the results of surgery for cervical compression myelopathy. Spine 18: , Papadopoulos CA, Katonis P, Papagelopoulos PJ, Karampekios S, Hadjipavlou AG: Surgical decompression for cervical spondylotic myelopathy: correlation between operative outcomes and MRI of the spinal cord. Orthopedics 27: , Suri A, Chabbra RP, Mehta VS, Gaikwad S, Pandey RM: Effect of intramedullary signal changes on the surgical outcome of patients with cervical spondylotic myelopathy. Spine J 3: 33 45, Takahashi M, Sakamoto Y, Miyawaki M, Bussaka H: Increased MR signal intensity secondary to chronic cervical cord compression. Neuroradiology 29: , Takahashi M, Yamashita Y, Sakamoto Y, Kojima R: Chronic cervical cord compression: clinical significance of increased signal intensity on MR images. Radiology 173: , Yukawa Y, Kato F, Yoshihara H, Yanase M, Ito K: MR T2 image classification in cervical compression myelopathy: predictor of surgical outcomes. Spine 32: , 2007 Manuscript submitted November 4, Accepted March 31, Address correspondence to: Yong Shen, M.D., Department of Spinal Surgery, Third Hospital of HeBei Medical University, 139 Ziqiang Road, Shijiazhuang, China wlf7730@163. com. 323