CERVICAL SPONDYLOTIC myelopathy is one of the

Transcription

1 ORIGINAL ARTICLE Gait Analysis for Evaluating the Relationship Between Increased Signal Intensity on T2-Weighted Magnetic Resonance Imaging and Gait Function in Cervical Spondylotic Myelopathy Chung Reen Kim, MD, Jong Yoon Yoo, MD, PhD, Sang Hoon Lee, MD, PhD, Dong Ho Lee, MD, PhD, Seung Chul Rhim, MD, PhD ABSTRACT. Kim CR, Yoo JY, Lee SH, Lee DH, Rhim SC. Gait analysis for evaluating the relationship between increased signal intensity on T2-weighted magnetic resonance imaging and gait function in cervical spondylotic myelopathy. Arch Phys Med Rehabil 2010;91: Objective: To determine relationships between increased signal intensity (ISI) on T2-weighted cervical spine magnetic resonance imaging (MRI) and parameters of gait analysis in patients with cervical spondylotic myelopathy (CSM). Design: Retrospective comparative study. Setting: Gait analysis laboratory. Participants: Patients (N 36) who undertook cervical laminectomy or laminoplasty because of CSM. Interventions: Not applicable. Main Outcome Measures: Subjects were evaluated by using the modified Japanese Orthopaedic Association (JOA) scale, the Nurick scale, cervical spine MRI, and gait analysis. Two radiologists classified patients into 3 groups: intense, faint, and no ISI. Results: Relative to patients without ISI, those with ISI showed significantly slower gait speed, longer step time, decreased single-limb support time, increased double-limb support time, and limited range of motion of knee and ankle (P.05). Increased intensity tended to correlate with poor gait function including slower gait speed, longer step time, decreased single-limb support time, and increased double-limb support time. The modified JOA and Nurick scale did not correlate with ISI. Conclusions: In patients with CSM who received surgical treatment, more intense ISI on T2-weighted MRI correlated preoperatively with increased difficulties in gait function. Gait analysis may be a useful tool for evaluating gait functions in cervical myelopathy. Key Words: Gait; Magnetic resonance imaging; Rehabilitation by the American Congress of Rehabilitation Medicine 1587 CERVICAL SPONDYLOTIC myelopathy is one of the common symptomatic dysfunctions of the cervical spinal cord and results from degenerative changes in the cervical spine with subsequent narrowing of the spinal canal. To confirm these degenerative changes, MRI is a useful imaging modality for diagnosing CSM. 1,2 ISI on T2-weighted MRI is commonly observed in CSM, and several studies have been performed by using several clinical scales to determine whether MRI signal changes suggest possible clinical outcomes as well as the postoperative prognosis. However, a consensus has not been reached. According to some studies, the ISI seen on T2-weighted MRI has been reported to indicate edema and irreversible gliosis, associating ISI with a poor prognosis. 3-5 However, in other studies, the ISI seen on T2-weighted MRI has been reported to indicate reversible parenchymal damage, suggesting that there is no correlation between ISI and clinical deficits. 6-8 The Nurick and JOA scales are the most commonly used clinical scales to classify the severity of CSM and to assess the postoperative outcomes. Those scales focus more on the gait function of patients with CSM because gait dysfunction is the most important issue regarding the surgical outcome and clinical deficits. The Nurick and JOA scales are quite simple and easy to assess at the primary clinic but are somewhat subjective and qualitative to describe precisely. However, computerized 3-dimensional gait analysis, which has been used for gait evaluation in various neurologic diseases, 3,9,10 is a well-established quantitative and objective measure of gait function. However, few previous studies performed gait analysis to evaluate the gait function of patients with CSM or reported the usefulness of gait analysis. 11,12 Based on previous studies, we believed that gait analysis might be useful to assess the unsettled issue of the relationship between signal changes on MRI and a patient s functional status, especially gait function. Therefore, the purpose of the present study was to determine the relationship between ISI on T2-weighted MRI and the severity of gait difficulty when using 3-dimensional gait analyses. From the Departments of Physical Medicine and Rehabilitation (Kim, Yoo), Radiology (S-H Lee), Orthopaedic Surgery (D-H Lee), and Neurosurgery (Rhim), Asan Medical Center, Ulsan University College of Medicine, Seoul, Korea. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Jong Yoon Yoo, MD, PhD, Department of Rehabilitation Medicine, Asan Medical Center, Ulsan University College of Medicine, Pungnap-2 dong, Songpa-gu, Seoul , Korea, jyyoo@amc.seoul.kr /10/ $36.00/0 doi: /j.apmr CSM ISI JOA MRI TE TR List of Abbreviations cervical spondylotic myelopathy increased signal intensity Japanese Orthopaedic Association magnetic resonance imaging echo time repetition time

2 1588 GAIT ANALYSIS IN CERVICAL MYELOPATHY, Kim METHODS Participants The study involved patients treated by cervical laminoplasty or laminectomy for CSM between January 2003 and August 2007 at the Department of Neurosurgery and Orthopaedic Surgery. Patients were excluded if they were unable to walk; had other diseases that could affect walking, such as stroke, traumatic brain injury, or myelitis; or had previous surgery of the cervical and/or lumbar spine. A total of 36 patients, 26 men and 10 women, were included; their clinical parameters and radiologic findings were retrospectively assessed. The protocol was approved by the ethics committee. Outcome Measurement Before surgery, MRI examinations were performed by using a 1.5 Tesla system. a T1-weighted and T2-weighted images of axial and sagittal views of the cervical cord were obtained. A surface coil was used, and the pulse sequence was TR 4000ms and TE 20 to 40ms for short sequences (T1-weighted images) and TR 1600 to 2500ms and TE 80 to 100ms for long sequences (T2-weighted images). Signal intensity on T2- weighted MRI was classified by 2 radiologists blinded to the patients clinical data. Patients were classified into those with no ISI (group 0), faint ISI (group 1), and intense ISI (group 2) (fig 1). 13 The concordance between the 2 radiologists in evaluating signal changes was.86 (.80; P.001). The 2 radiologists established the final classification by consensus. All subjects also underwent gait analysis before surgery. A physiatrist (C.R.K) evaluated the clinical status of subjects and reviewed their medical records. Clinical evaluations included neurologic symptoms related to CSM, motor and sensory function tests, deep tendon reflexes, ankle clonus, the Babinski sign, Rating Table 1: Nurick Scale Description 0 Sign or symptoms of root involvement but without evidence of spinal cord disease 1 Sign of spinal cord disease but no difficulty in walking 2 Slight difficulty in walking that did not prevent full-time employment 3 Difficulty in walking that prevented full-time employment or the ability to do all housework but that was not so severe as to require someone else s help to walk 4 Able to walk only with someone else s help or with the aid of a frame 5 Chair bound or bedridden bowel and bladder function, Nurick score (table 1), and modified JOA score (table 2). The Nurick grading scale, 14 which is widely used to grade CSM, 15 focuses mainly on gait function. The scale ranges from 0 (no evidence of spinal cord disease and no difficulty in walking) to 5 (chair-bound or bedridden state). The JOA scoring system, 16 as modified, 17 is used to assess the motor function of the upper and lower extremities as well as sensation and bladder function. The modified JOA scale ranges from 0 to 18 (no dysfunction of upper and lower extremities, no sensory loss, normative micturition). Gait analysis was performed by using a computerized 3- dimensional gait analysis system b and a 20-m track. Subjects walked continuously and freely along the track 3 or 4 times, and their linear (cadence, gait speed, step time, double-limb support time, single-limb support time) and kinematic (pelvis, hip, knee, ankle angles in the sagittal plane) parameters were computerized and averaged. Fig 1. T2-weighted cervical MRIs were classified by signal intensity. No ISI is shown in group 0 (A). Faintly and intensely ISI is shown in group 1 (B) and group 2 (C), respectively.

3 GAIT ANALYSIS IN CERVICAL MYELOPATHY, Kim 1589 Statistical Analysis SPSS 14.0 statistical software c was used for statistical analysis. The Kruskal-Wallis test was used to analyze differences in the modified JOA and Nurick scores among the 3 groups. Classification and regression tree analysis was performed to determine relationships between the presence of clinical signs and ISI on T2-weighted MRI. For the tree construction, the Gini index was used. The Mann-Whitney test was used to compare linear and kinematic parameters of gait analysis between the non-isi (group 0) and ISI (group 1 and 2) groups. The Kruskal-Wallis test was used to analyze differences in linear parameters and kinematic parameters among the 3 groups. Spearman rank correlation coefficient was used to determine whether ISI was correlated with linear and kinematic parameters. Any association having a P value less than.05 was considered statistically significant. Correction for alpha inflation associated with multiple tests of statistical hypotheses was not undertaken because of the limited statistical power afforded by the small sample. RESULTS Table 3 shows the general characteristics of the subjects. The modified JOA scores and Nurick scores were similar among the 3 groups, which indicated that they were not significantly related to signal intensity. In the regression tree, the first selected clinical sign was paresthesia, and ankle clonus, sensory change, and the Babinski sign followed (fig 2). Groups 1 and 2 tended to have more positive clinical signs than group 0. We observed significant relationships between ISI on T2- weighted MRI and linear parameters on gait analysis, including cadence, gait speed, step time, single-limb support time, and double-limb support time (table 4). Patients with ISI (groups 1 Table 2: Modified JOA Cervical Spine Myelopathy Functional Assessment Scale I. Motor dysfunction score of the upper extremities 0. Inability to move hands 1. Inability to eat with a spoon but able to move hands 2. Inability to button shirt but able to eat with a spoon 3. Able to button shirt with great difficulty 4. Able to button shirt with slight difficulty 5. No dysfunction II. Motor dysfunction score of the lower extremities 0. Complete loss of motor and sensory function 1. Sensory preservation without ability to move legs 2. Able to move legs but unable to walk 3. Able to walk on flat floor with a walking aid 4. Able to walk up and/or down stairs with hand rail 5. Moderate to significant lack of stability, but able to walk up and/or down stairs without hand rail 6. Mild lack of stability but walks with smooth reciprocation unaided 7. No dysfunction III. Sensory dysfunction score of the upper extremities 0. Complete loss of hand sensation 1. Severe sensory loss or pain 2. Mild sensory loss 3. No sensory loss IV. Sphincter dysfunction score 0. Inability to micturate voluntarily 1. Marked difficulty with micturition 2. Mild to moderate difficulty with micturition 3. Normal micturition Table 3: Demographic Characteristics of Patients in Each Group of ISI Characteristics Group 0 Group 1 Group 2 No. of subjects Age (y), mean SD Gender Female Male Modified JOA score, mean SD Nurick score, mean SD Neck pain (%) 8 (62) 7 (50) 3 (33) Increased tendon reflex (%) 3 (23) 7 (50) 6 (67) Ankle clonus (%) 2 (15) 6 (36) 6 (67) Babinski sign (%) 0 (0) 3 (21) 1 (11) Paresthesia (%) 10 (77) 14 (100) 9 (100) Sensory change (%) 3 (23) 6 (36) 4 (44) Bowel symptoms (%) 0 (0) 1 (7) 1 (11) Bladder symptoms (%) 1 (8) 3 (21) 2 (22) and 2) had slower gait speed, longer step time, shorter singlelimb support time, and longer double-limb support time than patients without ISI (group 0). In the kinematic parameters, the angle of maximal knee flexion of the swing phase was significantly different in the patients with and without ISI. In the patients with ISI, the angle of maximal knee flexion of the swing phase decreased, whereas the angle of maximal ankle dorsiflexion of terminal stance phase significantly increased. In fact, the knee flexion increased for toe clearance during the swing phase, and, during the terminal stance phase, ankle plantar flexion increased to push off before the swing phase. Therefore, the results showed that the range of motion of both knees and ankles was limited to the patients with ISI. Among the 3 groups, there were statistically significant relationships between linear parameters and signal intensity (table 5). As ISI increased, cadence, gait speed, and single-limb support time decreased, whereas step time and double-limb support time increased. We also found that the increased ISI was significantly correlated with the decreased angle of maximal knee flexion of the swing phase and increased angle of maximal ankle dorsiflexion of the terminal stance. DISCUSSION We have shown here that ISI on T2-weighted MRI was related to poor gait function in preoperative patients with CSM. Change in signal on MRI may reflect compression-induced histopathologic changes in the cervical spinal cord, 18,19 with a more intense signal indicating a more significant injury. These histopathologic changes may affect neurologic symptoms and signs, including gait function. To date, however, the influence of ISI on clinical status and surgical prognosis has been unclear. Before surgery, patients with ISI on T2-weighted MRI were found to have more severe injuries to the spinal cord than patients with normative signal intensity, 17 with the former having a lower rate of recovery from clinical symptoms after surgery. 20 In contrast, others 6-8 have reported no correlation between signal intensity changes and clinical symptoms; increased signal changes on T2- weighted MRI were found to be indicative of both pathologically reversible and irreversible changes in the spinal cord. 6-8 To assess whether ISI correlated with clinical status, previous studies used several systems for patients with CSM, including the Nurick scale, the JOA or modified JOA scale, the

4 1590 GAIT ANALYSIS IN CERVICAL MYELOPATHY, Kim Fig 2. Diagram shows classification and regression tree generated with clinical signs of CSM. Cooper myelopathy scale, the Prolo scale, and the European Myelopathy Score, all of which have different benefits and problems. 21 Although both the Nurick and JOA scales are widely used, both are considered qualitative and subjective measurements, suggesting that their results may depend on reports of patients and physicians. One crucial limitation of the JOA scale is that patients who show the same recovery rates have been found to differ in preoperative JOA scores, suggesting that preoperative JOA scores do not correlate with actual surgical outcomes. 9 In our study, the modified JOA score and the Nurick score did not significantly differ among the preoperative patients with CSM in our 3 groups. In fact, all the Table 4: Comparison of Linear Parameters and Kinematic Parameters With ISI and Without ISI Gait parameters Group 0 Group 1 Group 2 Linear parameters Cadence (steps/min) ( ) ( ).001* Gait speed (m/s) 1.05 ( ) 0.80 ( ).001* Step time (s) 0.54 ( ) 0.61 ( ).001* Single-limb support time (%) ( ) ( ).001* Double-limb support time (%) ( ) ( ).001* Kinematic parameters Maximal hip flexion ( ) of swing phase ( ) ( ).504 Maximal knee flexion ( ) of swing phase ( ) ( ).001* Maximal ankle dorsiflexion ( ) of terminal stance phase ( ) ( ).001* Abbreviation: CI, confidence interval. *P.05 based on the Mann-Whitney test. P

5 GAIT ANALYSIS IN CERVICAL MYELOPATHY, Kim 1591 Table 5: Comparison and Correlation Coefficient of Linear Parameters and Kinematic Parameters Among 3 Groups Gait parameters Group 0 Group 1 Group 2 P r Linear parameters Cadence (steps/min) ( ) ( ) ( ).004*.294 Gait speed (m/s) 1.05 ( ) 0.82 ( ) 0.79 ( ).001*.422 Step time (s) 0.54 ( ) 0.62 ( ) 0.60 ( ).003*.320 Single-limb support time (%) ( ) ( ) ( ).002*.376 Double-limb support time (%) ( ) ( ) ( ).000*.460 Kinematic parameters Maximal hip flexion ( ) of swing phase ( ) ( ) ( ) Maximal knee flexion ( ) of swing phase ( ) ( ) ( ).001*.296 Maximal ankle dorsiflexion ( ) of terminal stance phase ( ) ( ) ( ).003*.372 Abbreviation: CI, confidence interval. *P.05 based on the Kruskal-Wallis test. P.05 based on the Spearman rank correlation coefficient. subjects recruited in our study had only mild to moderate gait dysfunction, so they could walk independently without any device. Apparently it was difficult to evaluate the difference in gait function among the subjects using clinical assessment scales, such as the modified JOA score and Nurick score. Similar findings were also observed in another study, in which 114 preoperative patients with CSM were classified into 3 groups based on ISI, with preoperative JOA scores showing no significant difference among the 3 groups. 13 More precise and quantitative assessments were needed to evaluate the patients with the variable gait dysfunction, and few studies evaluated such patients. Singh and Crockard 22 measured walking times and number of steps taken over 30m in preoperative patients with CSM and healthy persons. Both parameters were significantly lower in the former and were correlated with Nurick scores. Kuhtz-Buschbeck et al 11 and Suzuki et al 12 used gait analysis for evaluating gait function of patients with CSM compared with healthy individuals. The results showed poor gait functions of patients with CSM, such as reduced gait speed, increased step length, prolonged doublelimb support time, and reduced ankle joint extension. Those studies compared only gait parameters between CSM and healthy subjects or between preoperative and postoperative results of patients with CSM. However, we used gait analysis to evaluate gait function of CSM according to ISI. We then found that patients with intense ISI showed decreased cadence, slower gait speed, increased step time, decreased single-limb support time, and increased double-limb support time compared with non-isi patients and that these results corresponded with those of previous other studies. 11 We thought that the reason for gait dysfunction in patients with ISI might be associated with increased upper motor neuron signs, which is affected by histopathologic changes of spinal cord lesion. In regression tree analysis, we observed that upper motor neuron signs such as ankle clonus and the Babinski sign were important in classification of ISI groups. Increased spasticity, another upper motor neuron sign, was also observed in kinematic parameters of gait analysis. Patients in the ISI groups showed limitations in range of motion of the knee during swing phases and of the ankle during terminal stance phase. This is indicative of spasticity that could affect gait function. Patients with spinal cord injuries have been reported to have limited knee flexion because of knee joint spasticity, which decreased knee mobility. 12 Therefore, ISI on T2-weighted MRI, indicative of the histopathologic changes in spinal cord, affected the increased upper motor neuron signs, and the increased upper motor neuron signs might affect the gait function. We suggested that ISI on T2-weighted MRI was reflective of poor gait function. Although the correlation analysis showed that the increased intensity was well correlated with gait dysfunction, this study failed to show clear difference of gait parameters between faint and intense ISI groups. This is thought to be a result of a small number of subjects and the qualitative classification of ISI. Study Limitations There are a few limitations to the results of our study. First, because the study has a retrospective design and a small number of patients was included, it was difficult to control all the compounding factors and to postulate our solid conclusions. It is possible that a bias could have been introduced by these factors and would have affected our conclusions, and that caution is therefore required in interpreting our results. However, through this and previous studies, we can confirm the usefulness of gait analysis in assessing the gait function of patients with CSM as well as the tendency for there to be a correlation between ISI and gait dysfunction. Second, because our study subjects included only patients who could walk more than 20m without assistance or devices, it was difficult to adapt our results to patients who could walk only with a device or with the assistance of another person. Because we also included only subjects who undertook surgery, we could not analyze the gait parameters of patients with ISI on T2-weighted MRI with only minimal clinical symptoms. Analyzing the various conditions of ISI would have provided a better understanding of the effect of ISI on gait function. Third, the classification of ISI severity was not performed with any quantitative measurements because there is not yet any generally accepted valid tool or measurement for classifying the severity of the signal intensity on MRI in CSM. Therefore, as in the previous studies, we divided the study subjects into 3 groups to decrease the number of potential errors. However, despite the high agreement between radiologists regarding classification of the signal changes, our results were still not optimal. Nonetheless, rapidly developing imaging techniques will lead to widely accepted classification scales. It is also difficult to maintain a gait laboratory in all primary clinics because the gait analysis system is expensive and complicated and occupies a wide space. Finally, to perform and interpret the test properly, experienced technicians and clinicians are required. Despite these limitations regarding the use of gait analysis in clinics, we believe it will be helpful to be able to quantify, compare, and

6 1592 GAIT ANALYSIS IN CERVICAL MYELOPATHY, Kim evaluate the outcomes of the operative and nonoperative treatments of CSM, and mostly useful for clinical studies of CSM. CONCLUSIONS By using gait analysis, we could assess patients gait dysfunction using linear, kinematic, and kinetic parameters. Gait analysis is a quantitative and objective measure to evaluate gait function, making it useful and possibly superior to the modified JOA and Nurick scales in evaluating preoperative patients with CSM who can walk independently. We also found that ISI correlated with gait function of CSM, indicating that signal changes on T2-weighted MRI may reflect poor gait function. Future studies comparing preoperative and postoperative gait analysis in a larger number of patients may determine the usefulness of gait analysis in preoperative patients with CSM. References 1. Emery SE. Cervical spondylotic myelopathy: diagnosis and treatment. J Am Acad Orthop Surg 2001;9: Edwards CC, Riew KD, Anderson PA, Hilibrand AS, Vaccaro AF. Cervical myelopathy: current diagnostic and treatment strategies. Spine J 2003;3: De Quervain IA, Simon SR, Leurgans S, Pease WS, McAllister D. Gait pattern in the early recovery period after stroke. J Bone Joint Surg Am 1996;78: 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: Takahashi M, Yamashita Y, Sakamoto Y, Kojima R. Chronic cervical cord compression: clinical significance of increased signal intensity on MR images. Radiology 1989;173: Hirabayashi S, Tsuzuki N, Abe R, Saiki K, Takahashi K. The value of a new method for assessing the separate functions of the long tracts and involved segments in patients with cervical myelopathy. Int Orthop 2000;24: Matsumoto M, Toyama Y, Ishikawa M, Chiba K, Suzuki N, Fujimura Y. Increased signal intensity of the spinal cord on magnetic resonance images in cervical compressive myelopathy: does it predict the outcome of conservative treatment? Spine 2000;25: Wada E, Ohmura M, Yonenobu K. Intramedullary changes of the spinal cord in cervical spondylotic myelopathy. Spine 1995;20: Suda Y, Saitou M, Shibasaki K, Yamazaki N, Chiba K, Toyama Y. Gait analysis of patients with neurogenic intermittent claudication. Spine 2002;27: Winters TF Jr, Gage JR, Hicks R. Gait patterns in spastic hemiplegia in children and young adults. J Bone Joint Surg Am 1987;69: Kuhtz-Buschbeck JP, Johnk K, Mader S, Stolze H, Mehdorn M. Analysis of gait in cervical myelopathy. Gait Posture 1999;9: Suzuki E, Nakamura H, Konishi S, Yamano Y. Analysis of the spastic gait caused by cervical compression myelopathy. J Spinal Disord Tech 2002;15: Yukawa Y, Kato F, Yoshihara H, Yanase M, Ito K. MR T2 image classification in cervical compression myelopathy: predictor of surgical outcomes. Spine 2007;32: Nurick S. The pathogenesis of the spinal cord disorder associated with cervical spondylosis. Brain 1972;95: Epstein NE. Circumferential cervical surgery for ossification of the posterior longitudinal ligament: a multianalytic outcome study. Spine 2004;29: Yonenobu K, Abumi K, Nagata K, Taketomi E, Ueyama K. Interobserver and intraobserver reliability of the Japanese Orthopaedic Association scoring system for evaluation of cervical compression myelopathy. Spine 2001;26: Benzel EC, Lancon J, Kesterson L, Hadden T. Cervical laminectomy and dentate ligament section for cervical spondylotic myelopathy. J Spinal Disord 1991;4: al-mefty O, Harkey HL, Marawi I, et al. Experimental chronic compressive cervical myelopathy. J Neurosurg 1993;79: Ohshio I, Hatayama A, Kaneda K, Takahara M, Nagashima K. Correlation between histopathologic features and magnetic resonance images of spinal cord lesions. Spine 1993;18: 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 2001; 26: Vitzthum HE, Dalitz K. Analysis of five specific scores for cervical spondylogenic myelopathy. Eur Spine J 2007;16: Singh A, Crockard HA. Quantitative assessment of cervical spondylotic myelopathy by a simple walking test. Lancet 1999;354: Suppliers a. Philips Medical Systems International B.V., Veenpluis 4-6, 5684 PC Best, Netherlands. b. Motion Analysis, 3617 Westwind Blvd, Santa Rosa, CA c. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL