Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2311030004
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, C.-J.
Right arrow Articles by See, L.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, C.-J.
Right arrow Articles by See, L.-C.
(Radiology 2004;231:39-44.)
© RSNA, 2004

Neuroradiology

Hirayama Flexion Myelopathy: Neutral-Position MR Imaging Findings—Importance of Loss of Attachment1

Chi-Jen Chen, MD, Hui-Ling Hsu, MD, Ying-Chi Tseng, MD, Rong-Kuo Lyu, MD, Chiung-Mei Chen, MD, PhD, Ying-Chih Huang, MD, Li-Jen Wang, MD, Yon-Cheong Wong, MD and Lai-Chu See, PhD

1 From the Second Departments of Diagnostic Radiology (C.J.C., H.L.H., Y.C.T., L.J.W., Y.C.W.), and Neurology (R.K.L., C.M.C., Y.C.H.), Chang Gung Memorial Hospital and University, 199 Tung-Hwa North Rd, Taipei, Taiwan, ROC; and Public Health and Biostatistics Center, Chang Gung University, Taoyuan, Taiwan, ROC (L.C.S.). Received January 1, 2003; revision requested March 6; final revision received August 3; accepted August 22. C.J.C. supported by National Science Council, Taiwan, grant NSC 89–2314-B-182A-086 and Chang Gung Memorial Hospital, Taiwan, grant CMRP 1112. Address correspondence to C.J.C. (e-mail: radcjc@adm.cgmh.org.tw).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To investigate the sensitivity and specificity of various neutral-position magnetic resonance (MR) imaging findings in the diagnosis of Hirayama flexion myelopathy.

MATERIALS AND METHODS: The neutral-position cervical MR images of 46 patients and 51 control subjects were evaluated for the following findings: localized lower cervical cord atrophy, asymmetric cord flattening, abnormal cervical curvature, loss of attachment (LOA) between the posterior dural sac and subjacent lamina, and noncompressed intramedullary high signal intensity on T2-weighted MR images. The difference in frequency of these findings between the control and patient groups was examined by means of the {chi}2 test. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of these MR imaging findings in the diagnosis Hirayama disease were calculated. Multivariate analysis of these findings was also performed.

RESULTS: There was a significant difference in the frequency of these MR imaging findings between the control and patient groups (all comparisons, P <= .002). Among the MR imaging findings, localized lower cervical cord atrophy, asymmetric cord flattening, and LOA had accuracy of more than 80% in identification of the disease. After multivariate analysis, LOA was the only significantly important predictor of the disease, with odds ratio of 716.7 (95% CI: 71.9, 7,145.2). Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of LOA were 93.5%, 98.0%, 97.7%, 94.3%, and 95.9%, respectively.

CONCLUSION: LOA from posterior dural sac and subjacent lamina is the most valuable finding in the diagnosis of Hirayama disease at neutral-position MR imaging.

© RSNA, 2004

Index terms: Muscles, diseases, 42.838, 43.838, 44.838 • Spinal cord, diseases, 341.41, 341.49 • Spinal cord, MR, 341.121411


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Hirayama disease, also termed juvenile muscular atrophy of the unilateral upper extremity, is a kind of cervical myelopathy related to flexional movements of the neck (1–6). Although the underlying causative mechanism remains unclear, findings in recent studies reveal that the myelopathy is attributed to a forward displacement of the posterior cervical dural sac when the neck flexes, which causes compression of the cervical cord (1,6–13). Clinically, these symptoms are usually seen in 15–25-year-old male patients, with insidious onset and predominantly unilateral muscular atrophy in the hand and forearm. The motor deficit and muscular atrophy may progress for 1–3 years before stabilization (1–3,5). Therefore, when the duration of the symptoms is short, several conditions that also cause localized amyotrophy of the distal arm—including syringomyelia, amyotrophic lateral sclerosis, cervical spondylotic myelopathy, and spinal cord tumor—should be differentiated from Hirayama disease.

Flexion magnetic resonance (MR) imaging can help differentiate Hirayama disease by depicting forward displacement of the posterior dural sac (8,10,14–17). However, we have encountered some problems in our daily practice. First, the finding of dural shifting may be absent or subtle in patients with disease duration of longer than 10 years (15). Second, most of these patients may undergo only nonflexion cervical MR examinationbecause most clinicians are not familiar with this disorder and do not request a flexion cervical MR examination. Therefore, exploration of neutral-position MR imaging findings to aid the diagnosis seems important. In the literature, a few investigators have suggested some specific, but not yet validated, neutral-position MR imaging or computed tomographic (CT) findings of this disease (14,16,18). They include localized lower cervical cord atrophy, asymmetric cord flattening, noncompressed intramedullary high signal intensity on T2-weighted MR images, and abnormal cervical curvature (straight or kyphotic). Besides these findings, in our practice, we have noted a loss of attachment (LOA) between the posterior dural sac and subjacent lamina. Thus, the purpose of our study was to investigate the sensitivity and specificity of various neutral-position MR imaging findings in the diagnosis of Hirayama flexion myelopathy.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Patients
We studied the neutral-position cervical MR images of all 46 patients (38 men and eight women; age range, 15–51 years; mean age, 23.67 years ± 7.59) with the diagnosis of Hirayama disease who were seen in our institution between January 1996 and July 2002. Retrospective review of the MR images and medical records of patients with Hirayama disease was approved by our institutional review board, and informed consent of patients was waived.

The diagnostic criteria of Hirayama disease include the following: (a) chronic weakness and atrophy of the distal upper limb(s), (b) insidious onset in the 2nd or 3rd decade of life, and (c) absence of substantial sensory deficits, reflex abnormalities, and cranial nerve, pyramidal tract, lower limb, sphincter, or cerebellum involvement (4); no history of poliomyelitis, toxin exposure, or periodic paralysis; and no evidence of an underlying disease causing peripheral neuropathy of axonal motor type shown at biochemical and neurophysiologic work-up (5).

Control Subjects
The neutral-position cervical MR images of 51 control subjects (34 men and 17 women) were studied. These subjects were enrolled consecutively between August 2000 and February 2001, because their age was between 15 and 30 years (an age range similar to that of the patients with Hirayama disease) and they had no evidence of cervical surgery or congenital cervical abnormalities noted at cervical MR imaging. In addition to neutral-position MR imaging, flexion cervical MR imaging was performed in these control subjects. The latter study was approved by our institutional review board, and all of the control subjects gave written informed consent. The control subjects underwent cervical MR imaging to provide a diagnosis for posterior neck pain, weakness or numbness of hands, and gait disturbance; to screen for cerebrospinal seeding; or to rule out multiple sclerosis. Charts of these control subjects were reviewed by one neurologist (R.K.L.) and flexion cervical MR images were reviewed by one radiologist (C.J.C.) to ensure that none of the charts contained the clinical findings or flexion MR imaging evidence of Hirayama disease.

Imaging Protocol
The MR imagers were superconducting 1.5-T systems (Signa, GE Medical Systems, Milwaukee, Wis; Vision, Siemens, Erlangen, Germany). The neutral-position MR protocol included transverse and sagittal T1-weighted (spin echo, repetition time msec/echo time msec of 500–600/15–20) and T2*-weighted (gradient-echo, 400–500/15–20, flip angle of 20°–30°) MR imaging and sagittal T2-weighted MR imaging (fast spin echo, 300–4,000/85–99 [effective]). The flexion MR imaging protocol consisted of sagittal T1-weighted (500–600/15–20) and T2-weighted (4,000/85–99, matrix size of 256 x 256) MR imaging and transverse T2*-weighted MR imaging (400–500/15–20, flip angle of 20°–30°, matrix size of 256 x 192) with 40° neck flexion by using a custom-built positioning sponge. Two signals were acquired at T1- and T2-weighted MR imaging, and four signals were acquired at T2*-weighted MR imaging. Section thickness was 4 mm with 1-mm gap for both sagittal and transverse MR imaging with all sequences.

Image Evaluation
Neutral-position cervical MR imaging findings—including localized lower cervical cord atrophy, asymmetric cord flattening, abnormal cervical curvature, LOA from posterior dural sac and subjacent lamina, and noncompressed intramedullary high signal intensity—were evaluated by two radiologists (L.J.W., Y.C.W.) independently as described later. Both radiologists were blinded to the patient’s history, clinical, and flexion MR imaging findings. Disagreements were resolved with subsequent consensus.

Localized lower cervical cord atrophy was evaluated on sagittal MR images and confirmed on transverse MR images. Lower cervical cord was defined as the cord between C4 and C7. Localized cord atrophy was defined as a decrease in cord size in comparison with the normal cord above and that below the affected level. Atrophy suggested on sagittal MR images was verified on appropriate contiguous transverse MR images, since it is possible to make an erroneous interpretation of atrophy on sagittal MR images if the spinal cord is not truly in the midline. On sagittal MR images, comparison of the suspected atrophy level with the cord below is more reliable than that with the cord above. Physiologically, the cervical cord starts to enlarge at C3, reaches its maximum at C5, and tapers thereafter to T2; thus, for a lower cervical cord lesion, a larger cord below the suspected atrophy confirms atrophy at the suspected level.

Asymmetric cord flattening was evaluated on transverse T2*-weighted MR images. To avoid confusion with cord compression due to adjacent spurs or herniated disks, cord flattening was defined as cord flattening without a narrowed or obliterated adjacent subarachnoid space. An elliptic spinal cord was considered normal, a pear-shaped spinal cord was considered asymmetric cord flattening, and a triangular spinal cord was considered symmetric cord flattening.

Cervical curvature was classified according to the principles suggested by Guigui et al (19) and Batzdorf and Batzdorff (20). Cervical curvature was measured according to the relationship of the dorsal aspect of the vertebral bodies C3 through C6 to a line drawn from the dorsocaudal aspect of the vertebral body C2 to the dorsocaudal aspect of the vertebral body C7 (Fig 1). By definition, normal lordotic cervical curvature is curvature in which no part of the dorsal aspect of the vertebral bodies C3 through C6 crosses the line from C2 through C7 (Fig 1, A). An abnormal (straight or kyphotic) curvature is curvature in which part or all of the dorsal aspects of the vertebral bodies C3 through C6 meet or cross through the line from C2 through C7 (Fig 1, B).



View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. A, Sagittal fast-spin-echo T2-weighted MR image (4,133/85) in a 17-year-old male control subject. B, Sagittal fast-spin-echo T2-weighted MR image (3,250/91.3) in a 17-year-old male patient with slowly progressive weakness and atrophy of the left hand and forearm. Cervical curvature is measured according to the relationship of the dorsal aspect of the vertebral bodies C3 through C6 to a line drawn from the dorsocaudal aspect of the vertebral body C2 to the dorsocaudal aspect of the vertebral body C7 (black line in A, white line in B). By definition, normal lordotic cervical curvature is curvature in which no part of the dorsal aspects of the vertebral bodies C3 through C6 cross the line from C2 through C7 (A). An abnormal (straight or kyphotic) curvature is curvature in which part or all of the dorsal aspects of the vertebral bodies C3 through C6 meet or cross that line (B).

 
LOA between the posterior dural sac and subjacent lamina was evaluated on each side of the lamina at the pedicular levels of C4 through C6 on transverse T2*-weighted MR images. The degree of separation between the posterior dural sac and its subjacent lamina was evaluated within a range defined medially by the point of junction with the lamina and laterally by a tangential line along the medial aspect of the pedicle. A separation of less than 33.3% at all of the observed segments was considered normal, while a separation of more than 33.3% at any of the observed segments was considered positive for LOA.

Noncompressed intramedullary high signal intensity was considered if a patent subarachnoid space and intramedullary high signal intensity were noted. Intramedullary high signal intensity associated with cord compression by adjacent spurs or herniated disks was assumed to be due to these local changes.

Statistical Analysis
Data were analyzed with software (SPSS, version 10.0; SPSS, Chicago, Ill). The power for the current study was calculated retrospectively by means of the formula of Fleiss (21), with confidence level of 95%, 2% exposure rate for the control group, odds ratio of exposure in patients relative to that in control subjects of 100, ratio of the control group to patient group of 1:1, and sample size of 40 for both groups. The resultant power was 100%. Interobserver agreement for the neutral-position MR imaging findings was evaluated and expressed with the {kappa} statistic. Agreement was excellent with {kappa} > 0.80; good, {kappa} = 0.61–0.80; moderate, {kappa} = 0.41–0.60; fair, {kappa} = 0.21–0.40; and poor, {kappa} < 0.20. The independent t test was used to examine the difference between patients and control subjects with regard to age. The {chi}2 test was used to examine the difference between patients and control subjects with regard to sex, abnormal cervical curvature, LOA between the posterior dural sac and subjacent lamina, localized lower cervical cord atrophy, asymmetric cord flattening, and noncompressed intramedullary high signal intensity. Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of these neutral-position MR imaging findings in the diagnosis of Hirayama disease were also calculated. Multiple logistic regressions with forward selection were made to include important predictors of the disease. Differences with a P value of less than .05 were considered to be statistically significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
The age of the 46 patients at symptom onset ranged from 12 to 28 years (mean, 17.2 years ± 3.30 [SD]) and that at cervical MR imaging ranged from 15 to 51 years (mean, 23.67 years ± 7.59). The age of the 51 control subjects at cervical MR imaging ranged from 15 to 28 years (mean, 21.96 years ± 3.06). No signifi-cant difference in age and sex was found between the control and patient groups (P > .05) (Table 1).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Age, Sex, and Neutral-Position MR Imaging Findings in Patients with Hirayama Disease and Control Subjects

 
The {kappa} values of interobserver agreement for localized lower cervical cord atrophy, asymmetric cord flattening, LOA between the posterior dural sac and subjacent lamina, cervical curvature, and noncompressed intramedullary high signal intensity were 0.897, 0.881, 0.958, 0.932, and 1.000, respectively.

Seventy percent (32 of 46) of the patient group and 0% (zero of 51) of the control group had evidence of asymmetric cord flattening (Figs 2–4). The difference between the groups was significant (P < .001). Fifty-nine percent (27 of 46) of the patient group and 0% (zero of 51) of the control group had localized lower cervical cord atrophy on neutral-position cervical MR images (Figs 3, 4). The difference between groups was significant (P < .001). Ninety-three percent (43 of 46) of the patient group and 2% (one of 51) of the control group had LOA between the posterior dural sac and subjacent lamina (Figs 2–4). The difference between groups was significant (P < .001). Eighty-three percent (38 of 46) of the patient group and 53% (27 of 51) of the control group had abnormal (straight or kyphotic) cervical curvature (Fig 1). The difference between groups was significant (P = .002). Twenty-eight percent (13 of 46) of the patient group and 4% (two of 51) of the control group had noncompressed intramedullary high signal intensity (Figs 3, 4). The difference between groups was significant (P = .001) (Table 1).



View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Transverse neutral-position T2*-weighted MR images (433/17 with 20° flip angle) at the pedicular level. A, Image in 20-year-old male control subject. B, Image in an 18-year-old patient with Hirayama disease. The lamina, defined medially by point of junction of lamina and laterally by tangential line along medial aspect of pedicle (longest white line in A and B), is equally divided into three parts. Images show less than 33.3% LOA between the posterior dural sac and subjacent lamina in A and more than 33.3% LOA in B. Asymmetrically bilateral cord flattening (more severe on the right side) is also depicted in B.

 


View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Neutral-position MR images in 18-year-old male patient with progressive right hand and forearm weakness and atrophy for 21/2 years. A, Sagittal fast-spin-echo T2-weighted image (4,000/85) shows normal cervical curvature, localized cord atrophy (arrowheads), and noncompressed intramedullary high signal intensity (arrow). B, Transverse T2*-weighted image (433/17 with 20° flip angle) shows asymmetric cord flattening (on right side) and nearly 100% LOA between the posterior dural sac and subjacent lamina. Intramedullary high signal intensity (arrowhead) is depicted in right-sided gray mater. C, Sagittal flexion fast-spin-echo T2-weighted image (4,000/85) shows anterior displacement of posterior dural sac (black arrows) and cord compression (white arrow).

 


View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Neutral-position MR images in a 24-year-old male patient with progressive bilateral weakness and atrophy, which is greater on the right side than on the left, for 3 years. A, Sagittal fast-spin-echo T2-weighted image (4,000/85) shows kyphotic cervical curvature, localized cord atrophy at C5 and C6 (arrowheads), and faint noncompressed intramedullary high signal intensity at C5-6 disk (arrow). B, Transverse T2*-weighted MR image (433/17, with 20° flip angle) at pedicular level of C5 shows symmetric cord flattening and 100% LOA between the posterior dural sac and subjacent lamina (arrows).

 
Table 2 lists the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of these neutral-position MR imaging findings in the diagnosis of Hirayama disease. Among these imaging findings, localized lower cervical cord atrophy, asymmetric cord flattening, and LOA between the posterior dural sac and subjacent lamina had accuracy of greater than 80% in identification of the disease. LOA, however, was the best indicator of Hirayama disease. Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of LOA were between 93.5% and 98.0%.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Sensitivity, Specificity, Accuracy, Positive Predictive Value, and Negative Predictive Value of Neutral-Position MR Imaging Findings in Diagnosis of Hirayama Disease

 
Table 3 presents the results of multiple logistic regression for Hirayama disease. LOA between the posterior dural sac and subjacent lamina was the only significant predictor of the disease (P < .001). The odds ratio of LOA was 716.7 (95% CI: 71.9%, 7,145.2%).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Results of Multiple Logistic Regressions with Forward Selection to Include Significantly Important Predictors of Hirayama Disease

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
Hirayama disease is different from the known types of motor neuron diseases because of its nonprogressive behavior and pathologic findings of focal ischemic changes in the anterior horn of the lower cervical cord (2,4). The pathologic findings of focal ischemia prompted neurologic imaging investigations, which have revealed dynamic changes in the cervical dural sac induced by neck flexion (6–9). Among these imaging methods, cervical flexion MR imaging is the most common and convenient. It shows not only the anteriorly displaced posterior dural wall but also a secondary congested posterior internal vertebral venous plexus (8,10,13–14,16). Although diagnosis at flexion MR imaging is straightforward, diagnosis at neutral-position MR imaging has not been systemically studied and reported previously, to our knowledge. In our study, we found that LOA between the posterior dural sac and subjacent lamina was the most effective finding in the diagnosis of Hirayama disease, with sensitivity and specificity greater than 93.5% and odds ratio of 716.7. The LOA sign, which, to our knowledge, has not been reported previously, is possibly related to a tight dural sac, which causes separation between the posterior dural sac and subjacent lamina in the neutral position.

The pathogenic mechanism of Hirayama disease is still unknown. An assumption of imbalanced growth between the patient’s vertebral column and spinal canal contents has been suggested (6,9,11–12). This imbalanced growth will cause disproportional length between the patient’s vertebral column and the spinal canal contents, which will cause a "tight dural sac" or "overstretch of the cord" in the neutral position and an anteriorly displaced posterior dural wall when the neck is flexed. Toma and Shiozawa (12) also suspected that the different growth rates between male and female patients might be related to the male preponderance of Hirayama disease.

The relationship between these MR imaging findings and the disproportional phenomenon may be explained as follows. In a normal spine, the spinal dura mater is a loose sheath that is anchored in the vertebral canal by the nerve roots and by its attachment to the periosteum in two places: one at the foramen magnum and the dorsal surfaces of C2 and C3 and the other at the coccyx (22). The remainder of the dura mater is only suspended and cushioned in the spinal canal by the epidural fat, venous plexuses, and loose connective tissues (22). In the neutral position, the dura mater of the cervical spine is slack (23). With neck flexion, the dura mater becomes tighter because the length of the cervical canal increases as the neck moves from extension to flexion. The difference in length between extension and flexion from T1 to the top of the atlas is 1.5 cm at the anterior wall and 5 cm at the posterior wall (23).

Normally, the slack of the dura mater can compensate for the increased length in flexion; therefore, although it becomes tightened, the dura mater can still be in close contact with the walls of the spinal canal without displacement. In patients with Hirayama disease, however, the dura mater is no longer slack but is tight in a neutral position. The tight dural sac, therefore, can result in separation of the posterior dural sac and its subjacent lamina or various degrees of abnormal cervical curvature (straight or kyphotic) according to the strength of the tension. The occurrence of dural separation is more likely than abnormal cervical curvature because the dura mater is only suspended in the spinal canal below C3 and can be easily separated when the dural sac becomes tighter. On neck flexion, the tight dural sac cannot compensate for the increased length of the posterior wall, which causes anterior shifting of the posterior dural wall and consequent compression of the cord. This compression may cause microcirculatory disturbances in the territory of the anterior spinal artery at the site of the most kyphotic level (usually C4 through C6) (1,4). The chronic circulatory disturbance resulting from repeated or sustained flexion of the neck may produce localized cord atrophy at the lower cervical region. Sometimes, intramedullary high signal intensity may be seen because of more severe ischemic change or gliosis at the vulnerable areas (17), but its incidence is lower than that of localized lower cervical cord atrophy.

Although the disproportional theory may explain most of the neutral-position MR imaging findings of Hirayama disease, the asymmetric cord flattening cannot be explained well with this theory. Seventy percent (32 of 46) of the patients in our study and 68% (39 of 57) of the patients in the study of Hirayama and Tokumaru (15) had asymmetric cord flattening at MR imaging or CT myelography. This finding indicates that there is another predisposing factor. A "posterior epidural ligament factor" has been proposed by Shinomiya et al (24). Anatomically, there are two kinds of posterior epidural ligaments between the posterior dura mater and the ligamentum flavum. One kind consists of fine elastic ligaments, whereas the other consists of large ligaments (approximately 1–3 mm in diameter). These ligaments have a tendency to be abundant at C1 through C2, decreased below C2, and sparse at C6 and C7 (25). It has been assumed that these ligaments may contribute to resistance against the separation of the posterior dura mater from the ligamentum flavum. Abnormally unequal distribution or lack of these ligaments may be the essential cause of asymmetric cord compression.

There were some limitations of this study. First, old and new MR imagers were used in this study, and their imaging qualities were somewhat different. The differences in old and new MR images may have affected the readers and thus introduced bias. Second, even though we tried to ensure that none of the control subjects had the clinical or flexion MR evidence of Hirayama disease at the time of imaging, we cannot completely rule out the presence of Hirayama disease at a very early stage. This condition would only trivially change the results of this study.

In conclusion, we found that LOA between the posterior dural sac and subjacent lamina is the most valuable imaging feature in the identification of Hirayama disease at neutral-position MR imaging. Thus, in patients with adolescent onset of distal upper limb weakness, the LOA sign on neutral-position MR images should raise suspicion for Hirayama disease. An additional flexion MR imaging study or detailed clinical evaluation to confirm this diagnosis can provide further confirmation.


   FOOTNOTES
 
Abbreviation: LOA = loss of attachment

Author contributions: Guarantor of integrity of entire study, C.J.C.; study concepts, C.J.C., H.L.H., R.K.L.; study design, C.J.C., H.L.H., Y.C.T., R.K.L.; literature research, C.J.C., H.L.H.; clinical studies, R.K.L., Y.C.H., C.M.C.; data acquisition, H.L.H., Y.C.T., L.J.W., Y.C.W.; data analysis/interpretation, C.J.C., H.L.H., R.K.L., C.M.C.; statistical analysis, C.J.C., H.L.H., L.C.S.; manuscript preparation, C.J.C., H.L.H., Y.C.T.; manuscript definition of intellectual content, C.J.C.; manuscript editing, C.J.C., H.L.H.; manuscript revision/review and final version approval, C.J.C., H.L.H., R.K.L.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hirayama K. Non-progressive juvenile spinal muscular atrophy of the distal upper limb (Hirayama’s disease). In: De Jong JM, eds. Handbook of clinical neurology. Vol 15. Amsterdam, the Netherlands: Elsevier, 1991; 107-120.
  2. Hirayama K, Toyokura Y, Tsubaki T. Juvenile muscular atrophy of unilateral upper extremity: a new clinical entity. Psychiatr Neurol Jpn 1959; 61:2190-2197.
  3. Hirayama K. Juvenile non-progressive muscular atrophy localized in hand and forearm: observation in 38 cases. Rinsho Shinkeigaku 1972; 12:313-324.[Medline]
  4. Hirayama K, Tomonaga M, Kitano K, Yamada T, Kojima S, Arai K. Focal cervical poliopathy causing juvenile muscular atrophy of distal upper extremity: a pathological study. J Neurol Neurosurg Psychiatry 1987; 50:285-290.[Abstract/Free Full Text]
  5. Sobue I, Saito N, Iida M, Ando K. Juvenile type of distal and segmental muscular atrophy of upper extremities. Ann Neurol 1978; 3:429-432.[CrossRef][Medline]
  6. Kikuchi S, Tashiro K, Kitagawa K, Iwasaki Y, Abe H. A mechanism of juvenile muscular atrophy localized in the hand and forearm (Hirayama’s disease): flexion myelopathy with tight dural canal in flexion. Rinsho Shinkeigaku 1987; 27:412-419.[Medline]
  7. Tokumaru Y, Hirayama K. Anterior shift of posterior lower cervical dural mater in patients with juvenile muscular atrophy of unilateral upper extremity. Rinsho Shinkeigaku 1989; 29:1237-1243.[Medline]
  8. Mukai E, Matsuo T, Muto T, Takahashi A, Sobue I. Magnetic resonance imaging of juvenile-type distal and segmental muscular atrophy of upper extremities. Rinsho Shinkeigaku 1987; 27:99-107.[Medline]
  9. Mukai E, Sobue I, Muto T, Takahashi A, Goto S. Abnormal radiological findings on juvenile-type distal and segmental muscular atrophy of upper extremities. Rinsho Shinkeigaku 1985; 25:620-626.[Medline]
  10. Chen CJ, Chen CM, Wu CL, Ro LS, Chen ST, Lee TH. Hirayama disease: MR diagnosis. AJNR Am J Neuroradiol 1998; 19:365-368.[Abstract]
  11. Kohno M, Takahashi H, Yagishita A, Tanabe H. "Disproportion theory" of the cervical spine and spinal cord in patients with juvenile cervical flexion myelopathy. Surg Neurol 1998; 50:421-430.[CrossRef][Medline]
  12. Toma S, Shiozawa Z. Amyotrophic cervical myelopathy in adolescence. J Neurol Neurosurg Psychiatry 1995; 58:56-64.[Abstract/Free Full Text]
  13. Gandhi D, Goyal M, Bourque PR, Jain R. Case 68: Hirayama disease. Radiology 2004; 230:692-696.[Free Full Text]
  14. Pradhan S, Gupta RK. Magnetic resonance imaging in juvenile asymmetric segmental spinal muscular atrophy. J Neurol Sci 1997; 146:133-138.[CrossRef][Medline]
  15. Hirayama K, Tokumaru Y. Cervical dural sac and spinal cord in juvenile muscular atrophy of distal upper extremity. Neurology 2000; 54:1922-1926.[Abstract/Free Full Text]
  16. Biondi A, Dormont D, Weitzner I, Jr, Bouche P, Chaine P, Bories J. MR imaging of the cervical cord in juvenile amyotrophy of distal upper extremity. AJNR Am J Neuroradiol 1989; 10:263-268.[Abstract]
  17. Schroder R, Keller E, Flacke S, et al. MRI findings in Hirayama’s disease: flexion-induced cervical myelopathy or intrinsic motor neuron disease? J Neurol 1999; 246:1069-1074.[CrossRef][Medline]
  18. Hirabuki N, Mitomo M, Miura T, Hashimoto T, Kawai R, Kozuka T. Computed tomographic myelopathy characteristics of spinal cord atrophy in juvenile muscular atrophy of the upper extremity. Eur J Radiol 1991; 13:215-219.[CrossRef][Medline]
  19. Guigui P, Benoist M, Deburge A. Spinal deformity and instability after multilevel cervical laminectomy for spondylotic myelopathy. Spine 1998; 23:440-447.[CrossRef][Medline]
  20. Batzdorf U, Batzdorff A. Analysis of cervical spine curvature in patients with cervical spondylosis. Neurosurgery 1988; 22:827-836.[Medline]
  21. Fleiss JL. Statistical methods for rates and proportions 2nd ed. New York, NY: Wiley & Sons, 1981; 38-46.
  22. Williams PL, Warwick R, Dyson M, Bannister LH. Anatomy of the human body 37th ed. London, England: Churchill Livingstone, 1989; 1086-1092.
  23. Bland JH. Basic anatomy. In: Bland JH, eds. Disorders of the cervical spine: diagnosis and medical management. 2nd ed. Philadelphia, Pa: Saunders, 1994; 41-70.
  24. Shinomiya K, Dawson J, Spengler DM, Konrad P, Blumenkopf B. An analysis of the posterior epidural ligament role on the cervical spinal cord. Spine 1996; 21:2081-2088.[CrossRef][Medline]
  25. Shinomiya K, Sato T, Spengler DM, Dawson J. Isolated muscle atrophy of the distal upper extremity in cervical spinal cord compressive disorders. J Spinal Disord 1995; 8:311-316.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, C.-J.
Right arrow Articles by See, L.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, C.-J.
Right arrow Articles by See, L.-C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE