GitBucket
Newer
zeynalig
<?xml version="1.0" ?> <tei> <teiHeader> <fileDesc xml:id="0"/> </teiHeader> <text xml:lang="en"> <p>T he clinical entity of cervical flexion myelopa-<lb/>thy proposed in the 1960s by Reid <ref type="biblio">[16,17]</ref> and<lb/> Breig et al <ref type="biblio">[2,3]</ref> has been neglected for a long time.<lb/> Recently, attention has been given to its possible<lb/> involvement in the etiology of juvenile muscular<lb/> atrophy of a unilateral upper extremity (JMAU),<lb/> which was first described by Hirayama et al <ref type="biblio">[4 – 6]</ref>.<lb/> Cervical flexion myelopathy is defined as myelopa-<lb/>thy caused by compression of the lower cervical<lb/> spinal cord by the vertebral bodies, intervertebral<lb/> discs, or forwardly migrated posterior wall of the<lb/> dura mater when the neck is flexed; it may also<lb/> result from " overstretching " of the cervical spinal<lb/> cord. Recently, Mii et al <ref type="biblio">[11]</ref> described a similar<lb/> condition not only in the cervical but also in the<lb/> thoracic region. Cervical flexion myelopathy, in-<lb/>cluding JMAU, is quite a rare disease, found much<lb/> more frequently in Japan than other countries; it<lb/> affects mainly young males. JMAU is considered to<lb/> be a lower cervical anterior horn syndrome with<lb/> chronic muscular atrophy confined to the hand and<lb/> forearm, predominantly of a unilateral upper ex-<lb/>tremity without sensory disturbance; whereas cer-<lb/>vical flexion myelopathy may also include acute<lb/> progression, sensory disturbance, and/or long tract<lb/> signs.<lb/></p> <p>A myelogram in the flexed neck position of pa-<lb/>tients with JMAU shows the characteristic radio-<lb/>logic feature of forward migration of the posterior<lb/> wall of the lower cervical dura mater; therefore,<lb/> JMAU is classified as a cervical flexion myelopathy.<lb/> The cause of the forward migration of the posterior<lb/> wall of the lower cervical dura mater is suspected<lb/> to be disproportion of the cervical spine and spinal<lb/> cord <ref type="biblio">[19,20]</ref>, but this has yet to be confirmed.<lb/></p> <p>A marked male preponderance is recognized<lb/> <ref type="biblio">[5,18]</ref>, but the reason for this is still unknown.<lb/></p> <p>Therefore, we performed cervical magnetic reso-<lb/>nance imaging (MRI) studies on our 6 patients and<lb/> on 34 young, normal controls (22 males and 12<lb/> females) to determine the features and differences<lb/> of the cervical spine between our patients and<lb/> groups of males and females controls.<lb/></p> <head>Subjects and Methods<lb/></head> <head>SUBJECTS<lb/></head> <p>The patient group comprised all six patients with<lb/> juvenile cervical flexion myelopathy admitted to<lb/> our hospital <ref type="table">(Table 1)</ref>. In all of them, cervical flexion<lb/> myelopathy was diagnosed on the basis of findings<lb/> of muscular atrophy of the upper extremities <ref type="figure">(Fig-<lb/>ure 1)</ref> and spinal cord compression by the vertebral<lb/> bodies or intervertebral discs with forward migra-<lb/>tion of the posterior wall of the lower cervical dura<lb/> mater in neck flexion shown by myelography <ref type="figure">(Fig-<lb/>ure 2)</ref>. Electromyogram in all patients showed high-<lb/>amplitude action potentials with long duration in<lb/> association with reduction of the total motor-unit<lb/> voltage during the maximum voluntary contraction<lb/> of the atrophied muscles indicating a neurogenic<lb/> electromyographic pattern. Nerve conduction stud-<lb/>ies showed normal motor and sensory conduction<lb/> velocities in both the median and ulnar nerves. All<lb/> patients were male, aged 19 –25 years (mean: 20.3<lb/> years) and their heights were 168 –178 cm (mean:<lb/> 172.2 cm). Dynamic cervical MRI in flexed, neutral,<lb/> and extended neck positions was carried out on all<lb/> the patients. In all patients, cervical plain X-ray<lb/> films showed malalignment of the posterior sur-<lb/>faces of the cervical vertebrae in the flexed and<lb/> neutral neck positions and resolution of this mal-<lb/>alignment in the extended neck position <ref type="figure">(Figure 3)</ref>.<lb/> Cervical MR imaging in the neutral neck position of<lb/> five of the six patients showed a straight cervical<lb/></p> <figure type="table">1 Summary of Our Six Patients with Juvenile Cervical Flexion Myelopathy<lb/> PATIENT<lb/> AGE/SEX<lb/> AGE AT<lb/> ONSET<lb/> HEIGHT<lb/> (CM)<lb/> MUSCLE<lb/> ATROPHY<lb/> SENSORY<lb/> DISTURBANCE<lb/> 1<lb/> 19/M<lb/> 16<lb/> 174<lb/> Lt. forearm, hand<lb/> 2<lb/> 19/M<lb/> 18<lb/> 168<lb/> Bil. arm, hand<lb/> 3<lb/> 21/M<lb/> 17<lb/> 178<lb/> Lt. arm, hand<lb/> 4<lb/> 25/M<lb/> 15<lb/> 174<lb/> Bil. forearm, hand<lb/> 5<lb/> 20/M<lb/> 17<lb/> 170<lb/> Rt. arm, hand<lb/> Bil. hand<lb/> 6<lb/> 18/M<lb/> 16<lb/> 169<lb/> Rt. arm, hand<lb/> Bil. hand<lb/> Rt., right; Lt., left; Bil., bilateral.<lb/></figure> <figure>1<lb/> Preoperative photograph of Patient 5. Note the mus-<lb/>cular atrophy in the right thenar, hypothenar, tri-<lb/>ceps, forearm, and major pectoral muscles (arrowheads).<lb/></figure> <p>spinal cord running along the shortest route<lb/> through the cervical and upper thoracic spinal<lb/> canal <ref type="figure">(Fig. 4, center)</ref>. Sagittal section MR images<lb/> failed to show atrophy of the lower cervical spinal<lb/> cord in two of our patients; however, postmyelo-<lb/>graphic computed tomographic (CT) scan dem-<lb/>onstrated lower cervical cord atrophy in all of<lb/> them.<lb/></p> <p>The age distribution of the 34 normal controls<lb/> matched that of the patient group; they were 14 –29<lb/> years old (mean, 24.1 years). The normal control<lb/> group comprised 12 female and 12 male patients<lb/> without cervical disease admitted to our hospital<lb/> whose cervical MR images were normal, and who<lb/> were designated as random-selected females and<lb/> males, respectively, and 10 healthy male volun-<lb/>teers. As well as matching the age distribution and<lb/> sex, the height distribution of the patients with<lb/> juvenile cervical flexion myelopathy was compared<lb/> with a height-matched male group (n 17) consist-<lb/>ing of 10 healthy male volunteers and 7 from the<lb/> random-selected male group. The height-matched<lb/> males were 16 –29 years old (mean, 25.4 years) and<lb/> their heights were 168 –180 cm (mean, 174.3 cm);<lb/> these values did not differ significantly from the<lb/> patient group values. All 10 healthy male volun-<lb/>teers (dynamic-MRI male group) in the group of<lb/> 17 height-matched males underwent dynamic cer-<lb/>vical MRI in flexed, neutral, and extended neck<lb/> positions. To investigate the effects of height, the<lb/> 22 normal control males were divided into tall<lb/></p> <figure>2<lb/> Cervical myelograms showing forward migration of<lb/> the posterior wall of the dura mater of the lower<lb/> cervical spine in neck flexion (arrows) (left, flexion; right,<lb/> extension).<lb/></figure> <figure>3<lb/> Dynamic cervical plain X-ray films showing malalignment of the posterior surfaces of the vertebral bodies in the<lb/> flexed and neutral neck positions (arrows) and resolution of this malalignment in the extended neck position.<lb/></figure> <p>(height 170 cm, n 15) and short (height 170<lb/> cm, n 7) groups.<lb/></p> <head>METHODS<lb/></head> <p>Cervical T 1 -weighted images were fed into a per-<lb/>sonal computer using an image scanner (EPSON<lb/> GT-6500) and printed out on A3-sized paper with full<lb/> magnification to reduce measurement errors. The<lb/> required parameters on the printed T 1 -weighted im-<lb/>ages were measured using a digitizer (distance/area<lb/> calculating program ACA 9802-20065, WACOM). The<lb/> measured parameters are presented in <ref type="figure">Figure 5</ref>.<lb/> The distances C and J are the spinal cord lengths<lb/> along the center line on the spinal cord from the<lb/> lower surface of C7 to the lower surface of the C2<lb/> vertebral body, C, and the pontomedullary junc-<lb/>tion, J. V is the distance along the posterior surface<lb/> of the vertebral bodies from the lower surface of C7<lb/> to the lower surface of C2, and S is the length of the<lb/> line connecting the posteroinferior point of the C7<lb/> vertebral body with the same point of the C2 ver-<lb/>tebral body. 5P, 6P, and 7P are the distances from<lb/> the center of the spinal cord to the posterior sur-<lb/>face of the spinal canal at the level of the lower<lb/> surfaces of C5, C6, and C7, respectively. The angles<lb/> 4A, 5A, and 6A are composed of the line connecting<lb/> each center point of the spinal cord at the level of<lb/> the lower surface of C7 and C4, C5, C6, respectively,<lb/> and the line connecting each center point of the<lb/> spinal cord at the level of the lower surface of C2<lb/> and C4, C5, C6, respectively. The total post-cord<lb/> distance (TPD) of the lower cervical spine<lb/> (5P6P7P) (A) was considered to indicate the<lb/> location of the lower cervical spinal cord in the<lb/> spinal canal; it was smaller when the spinal cord<lb/> was more stretched and ran along the shorter<lb/> route. The total angle (TA) of the lower cervical<lb/> spinal cord (4A5A6A) (B) was considered to<lb/> reflect the shape of the lower spinal cord and was<lb/></p> <figure>4<lb/> T1-weighted cervical MR images in Patient 2 (left, flexion; center, neutral position; right, extension). MR image in<lb/> the neutral neck position (center) shows kyphotic cervical vertebrae and a straight spinal cord, that appeared to<lb/> be " stretched, " running along the shortest route through the cervical and upper thoracic spinal canal. MR image in the<lb/> flexed neck position (left) showed high signal intensity in the posterior epidural space (arrows) and a compressed<lb/> spinal cord at C5/C6 level by the C5, C6, and C5/C6 intervertebral spaces.<lb/></figure> <p>larger when the spinal cord was straighter. Further-<lb/>more, the cervical spine shape was evaluated ap-<lb/>proximately by scoring (1, kyphotic; 2, slightly ky-<lb/>photic; 3, straight neck; 4, slightly lordotic; 5,<lb/> lordotic). The lengths of the cervical spinal cord (J,<lb/> C) and cervical spine (V, S) of each individual were<lb/> divided by his/her height when any two groups<lb/> were compared to exclude any influence of height.<lb/></p> <p>Each parameter showed a normal distribution;<lb/> Student's t-test for unpaired samples was used to<lb/> compare the mean values of two groups. Differ-<lb/>ences of p 0.05 were considered to be significant.<lb/> The data for each measured parameter were used<lb/> to perform the following studies. 1) Comparison of<lb/> the tall and short male groups to determine<lb/> whether height influenced the cervical spine and<lb/> spinal cord shapes of the normal controls; 2) com-<lb/>parisons of the random-selected female group with<lb/> the random-selected male and short male groups<lb/> (the mean height of the latter did not differ signifi-<lb/>cantly from that of the random-selected females) to<lb/> determine whether the cervical MR images were<lb/> subject to sex differences; 3) comparisons of the<lb/> patients with juvenile cervical flexion myelopathy<lb/> with the height-matched male and random-selected<lb/> female groups to determine whether there were any<lb/> differences between the cervical MR images of the<lb/> patients and those of normal control males and<lb/> females; and 4) comparison of the patient and dy-<lb/>namic MR imaging male groups to determine<lb/> whether their dynamic cervical MR images differed.<lb/></p> <head>Results<lb/></head> <head>INFLUENCE OF HEIGHT<lb/></head> <p>Comparison of the tall (n 15) and short (n 7)<lb/> male groups <ref type="table">(Table 2)</ref> showed that the TA of the<lb/> lower cervical spinal cord was significantly larger,<lb/> as well as height (H), cervical spinal cord length (J,<lb/> C) and cervical spine length (V, S) (p 0.05, respec-<lb/>tively), and the cervical spine shape score (CSSS)<lb/> was smaller in the tall than in the short male group.<lb/> In summary, the cervical spines and spinal cords<lb/> of the tall males were much straighter than those of<lb/> the short males.<lb/></p> <head>INFLUENCE OF SEX<lb/></head> <p>Comparison of the random-selected female (n 12)<lb/> and random-selected male (n 12) groups <ref type="table">(Table<lb/> 2)</ref> showed that TA was significantly larger (p<lb/> 0.005) and CSSS was significantly smaller (p<lb/> 0.005) in the former despite being significantly<lb/> shorter in height (p 0.05) than the latter. The<lb/> cervical spinal cord length/height (J/H, C/H) and<lb/> cervical spine length/height (V/H, S/H) ratios of the<lb/> two groups were almost identical.<lb/></p> <p>Comparison of the random-selected female (n<lb/> 12) and short male (n 7) groups, the mean heights<lb/> of which did not differ significantly, also revealed<lb/> that TA was significantly larger (p 0.005) and CSSS<lb/> was significantly smaller (p 0.0005) in the females<lb/> than in the short males.<lb/></p> <p>In summary, the cervical spines and cervical spi-<lb/>nal cords of the random-selected females were sig-<lb/>nificantly straighter than those of the random-<lb/>selected males despite being significantly shorter<lb/> and also straighter than those of the short males,<lb/> whose mean height was matched to that of the<lb/> random-selected females.<lb/></p> <head>COMPARISON OF THE PATIENT (N<lb/> 6) AND HEIGHT-MATCHED MALE (N<lb/> 17) GROUPS<lb/></head> <p>Comparison of cervical MR images in the neutral<lb/> neck position <ref type="table" >(Table 2)</ref> showed that the cervical<lb/> spinal cord length/height (C/H) and cervical spine<lb/> length/height (V/H, S/H) ratios of the patients were<lb/> significantly higher (p 0.05) than those of the<lb/> height-matched males. TA and J/H were larger and<lb/> CSSS and TPD appeared to be smaller in the pa-<lb/>tients than in the height-matched males, but the<lb/> differences failed to reach significance.<lb/> In summary, the cervical spinal cord length/<lb/> height and cervical spine length/height ratios of the<lb/> patients were significantly higher than those of the<lb/> height-matched males. The cervical spine and spi-<lb/>nal cord shapes were straighter and the spinal cord<lb/></p> <figure>5<lb/> Schematic illustrations of measured parameters.<lb/></figure> <figure type="table">2 Results of the Comparisons Among the Groups in This Study<lb/> MEASUREMENT<lb/> PARAMETERS<lb/> TALL<lb/> MALE GROUP<lb/> (N 15)<lb/> SHORT<lb/> MALE GROUP<lb/> (N 7)<lb/> RANDOM-SELECTED<lb/> MALE GROUP<lb/> (N 12)<lb/> RANDOM-SELECTED<lb/> FEMALE GROUP<lb/> (N 12)<lb/> PATIENT GROUP<lb/> (N 6)<lb/> HEIGHT-MATCHED<lb/> MALE GROUP<lb/> (N 17)<lb/> NEUTRAL<lb/> POSITION<lb/> EXTENSION<lb/> R<lb/> R<lb/> R<lb/> H (cm)<lb/> 175.1 3.4<lb/> 163.1 5.4<lb/> 167.6 7.0<lb/> 159.9 6.8<lb/> 172.2 3.8<lb/> 174.3 3.8<lb/> R<lb/> R<lb/> R<lb/> J (mm)<lb/> 150.2 9.1<lb/> 141.6 5.1<lb/> 143.9 9.6<lb/> 136.4 6.7<lb/> 152.3 2.7<lb/> 145.7 5.9<lb/> 148.9 9.3<lb/> R<lb/> R<lb/> R<lb/> C (mm)<lb/> 89.5 5.4<lb/> 84.4 2.2<lb/> 86.3 5.0<lb/> 81.7 5.5<lb/> 91.7 3.4<lb/> 87.1 3.7<lb/> 88.9 5.4<lb/> R<lb/> R<lb/> R<lb/> R<lb/> V (mm)<lb/> 91.2 5.3<lb/> 87.2 2.2<lb/> 88.7 5.1<lb/> 82.9 4.8<lb/> 93.5 3.4<lb/> 91.2 2.2<lb/> 90.6 5.2<lb/> R<lb/> R<lb/> R<lb/> R<lb/> S (mm)<lb/> 90.5 5.1<lb/> 86.4 2.2<lb/> 87.8 4.7<lb/> 82.2 4.7<lb/> 92.9 3.3<lb/> 89.7 2.8<lb/> 90.0 5.1<lb/> R<lb/> J/H (%)<lb/> 8.6 0.4<lb/> 8.7 0.4<lb/> 8.6 0.5<lb/> 8.5 0.3<lb/> 8.8 0.2<lb/> 8.5 0.5<lb/> 8.5 0.4<lb/> R<lb/> C/H (%)<lb/> 5.1 0.3<lb/> 5.2 0.2<lb/> 5.2 0.3<lb/> 5.1 0.2<lb/> 5.3 0.2<lb/> 5.1 0.3<lb/> 5.1 0.2<lb/> R<lb/> R<lb/> V/H (%)<lb/> 5.2 0.2<lb/> 5.3 0.2<lb/> 5.3 0.3<lb/> 5.2 0.2<lb/> 5.4 0.2<lb/> 5.3 0.2<lb/> 5.2 0.2<lb/> R<lb/> R<lb/> S/H (%)<lb/> 5.2 0.2<lb/> 5.3 0.2<lb/> 5.2 0.3<lb/> 5.1 0.2<lb/> 5.4 0.2<lb/> 5.2 0.2<lb/> 5.2 0. 2<lb/> TPD (mm)<lb/> 17.9 4.2<lb/> 15.7 1.1<lb/> 17.5 4.6<lb/> 17.0 2.1<lb/> 15.4 2.7<lb/> 14.4 1.8<lb/> 17.6 4.0<lb/> R<lb/> TA (°)<lb/> 519.9 15.7<lb/> 503.9 16.3<lb/> 507.8 17.8<lb/> 531.2 15.7<lb/> 522.8 15.1<lb/> 518.0 16.0<lb/> CSSS<lb/> 2.5 1.5<lb/> 3.7 1.4<lb/> 3.6 1.7<lb/> 1.6 0.8<lb/> 2.2 1.6<lb/> 2.6 1.5<lb/> Values are means SD; R, significant (p 0.05); , significant (p 0.005); H, height; CSSS, Cervical spine shape score.<lb/></figure> <p>ran along a shorter route in the patients than in the<lb/> height-matched males although the differences<lb/> failed to reach significance.<lb/></p> <p>Comparison of the cervical MR images in exten-<lb/>sion of the patients and those in the neutral neck<lb/> position of the height-matched males resolved the<lb/> significant differences demonstrated by compari-<lb/>son of the neutral cervical MR images of these two<lb/> groups.<lb/></p> <head>COMPARISON OF THE CERVICAL MR<lb/> IMAGES IN THE NEUTRAL NECK<lb/> POSITION OF THE PATIENT (N 6)<lb/> AND RANDOM-SELECTED FEMALE<lb/> (N 12) GROUPS<lb/></head> <p>The cervical spinal cord length/height (J/H) and<lb/> cervical spine length/height (V/H, S/H) ratios of the<lb/> patients <ref type="table">(Table 2)</ref> were significantly higher (p<lb/> 0.05), whereas TA and CSSS of these two groups did<lb/> not differ significantly.<lb/></p> <head>COMPARISON OF THE DYNAMIC<lb/> CERVICAL MR IMAGES OF THE<lb/> PATIENT (N 6) AND DYNAMIC-MRI<lb/> MALE (N 10) GROUPS<lb/></head> <p>Differences between the cervical MR images in neck<lb/> flexion and extension: comparison of the flexed and<lb/> extended cervical MR images of the patient group<lb/> and of the dynamic-MRI male group showed that<lb/> TA, TPD, and the lengths of the spinal cord (J, C)<lb/> and cervical spine (V, S) were significantly higher in<lb/> flexion than in extension (p 0.05) in both groups<lb/> <ref type="table">(Table 3)</ref>.<lb/></p> <p>Comparison of the dynamic MR images of the<lb/> patient and dynamic-MR imaging male groups: the<lb/> cervical spinal cord length/height (J/H, C/H) ratios<lb/> and the flexion/extension ratios of the cervical<lb/> spinal cord length (J, C) and the cervical spine<lb/> length (V, S) of these two groups did not differ<lb/> significantly.<lb/></p> <p><ref type="table">Table 4</ref> summarizes the features of the cervical<lb/> spine and spinal cord shapes obtained in our study.<lb/></p> <head>Discussion<lb/></head> <p>The usual neuroradiological findings in patients<lb/> with JMAU are forward migration of the posterior<lb/> wall of the dura mater of the lower cervical spine on<lb/> myelogram in neck flexion and high signal intensity<lb/> changes in the epidural space on the T1-weighted<lb/> MR image in neck flexion <ref type="figure">(Figure 4, left)</ref>. This high<lb/> signal intensity appearing in the flexed neck posi-<lb/>tion was pointed out and confirmed to be an en-<lb/>gorged internal vertebral venous plexus with dy-<lb/>namic CT scan <ref type="biblio">[13,14]</ref>. Biondi et al <ref type="biblio" >[1]</ref> observed<lb/> atrophy of the lower cervical spinal cord without<lb/> signal intensity changes on both T1 and T2-<lb/>weighted images in five of seven patients. Atrophy<lb/></p> <figure type="table">3 Results of the Comparisons Between the Data in Flexed and Extended Neck Position<lb/> PATIENT GROUP<lb/> DYNAMIC-MR IMAGING MALE GROUP<lb/> FLEXION<lb/> EXTENSION<lb/> FLEXION<lb/> EXTENSION<lb/> J (mm)<lb/> 156.5 4.5 145.7 5.8*<lb/> 156.0 4.1 146.4 7.1**<lb/> C (mm)<lb/> 96.2 5.0 87.1 3.1**<lb/> 94.3 3.8 85.1 5.8***<lb/> V (mm)<lb/> 95.4 4.0 91.2 2.2*<lb/> 94.3 3.2 89.3 5.1*<lb/> S (mm)<lb/> 94.0 3.4 89.7 2.8*<lb/> 93.5 3.3 87.6 5.5*<lb/> TPD (mm)<lb/> 22.5 3.9 14.4 1.8**<lb/> 19.0 3.7 15.2 2.0*<lb/> TA (°)<lb/> 560.3 16.8 481.5 31.0***<lb/> 550.6 14.0 477.3 20.0***<lb/> Values are means SD; *p 0.05; **p 0.005; ***p 0.0005.<lb/></figure> <figure type="table">4<lb/> Features of the Cervical Spine and Spinal Cord in the Groups of Males, Females, and Patients with Juvenile Cervical<lb/> Flexion Myelopathy<lb/> GROUP<lb/> HEIGHT<lb/> LENGTH OF<lb/> CERVICAL<lb/> SPINAL CORD<lb/> LENGTH<lb/> OF<lb/> CERVICAL<lb/> SPINE<lb/> SHAPE OF<lb/> CERVICAL<lb/> SPINAL<lb/> CORD<lb/> SHAPE OF<lb/> CERVICAL<lb/> SPINE<lb/> LENGTH OF<lb/> CERVICAL SPINAL<lb/> CORD/HEIGHT<lb/> LENGTH OF<lb/> CERVICAL<lb/> SPINE/HEIGHT<lb/> Males<lb/> Tall<lb/> Long<lb/> Long<lb/> Lordotic<lb/> Lordotic<lb/> Normal<lb/> Normal<lb/> Females<lb/> Short<lb/> Short<lb/> Short<lb/> Straight<lb/> Straight<lb/> Normal<lb/> Normal<lb/> Cervical flexion<lb/> myelopathy<lb/> Tall<lb/> Very long<lb/> Very long<lb/> Straight<lb/> Straight<lb/> Large<lb/> Large<lb/></figure> <p>of the lower cervical spinal cord had already been<lb/> pointed out with postmyelographic CT scan <ref type="biblio">[9]</ref> and<lb/> is widely accepted.<lb/></p> <p>The myelopathy associated with JMAU is thought<lb/> to be derived from circulation disturbance of the<lb/> cervical spinal cord in the light of the autopsy find-<lb/>ings of a patient reported by Hirayama et al <ref type="biblio">[6]</ref>. <lb/></p> <p>Reid <ref type="biblio">[16,17]</ref> and Breig et al <ref type="biblio">[2,3]</ref> hypothesized<lb/> that compression of the cervical spinal cord in<lb/> flexed neck position by anterior structures, such as<lb/> the vertebral bodies or intervertebral discs, was the<lb/> etiological mechanism responsible for cervical flex-<lb/>ion myelopathy. Yada et al <ref type="biblio">[19,20]</ref> and Mii et al <ref type="biblio" >[10]<lb/></ref> considered that overstretching of the cervical spi-<lb/>nal cord itself in the up-down direction (overstretch<lb/> mechanism) was the cause, whereas Iwasaki et al<lb/> <ref type="biblio">[7]</ref> and Kikuchi et al <ref type="biblio">[8]</ref> regarded compression of<lb/> the cervical spinal cord by the migrated posterior<lb/> wall of the dura mater of the lower cervical spine<lb/> (tight dural canal mechanism) to be the etiological<lb/> mechanism. Toma and Shiozawa <ref type="biblio">[18]</ref> proposed a<lb/> new hypothesis that regarded relatively shortened<lb/> cervical dorsal roots during the juvenile growth<lb/> period as the cause of forward migration of the<lb/> spinal cord and the dura mater. However, the etiol-<lb/>ogy or the cause of forward migration has yet to be<lb/> confirmed.<lb/></p> <p>Reid <ref type="biblio">[16,17]</ref> and Breig et al <ref type="biblio">[2,3]</ref>, who studied<lb/> cadavers, mentioned that the degree of cervical<lb/> spine and spinal cord elongatation in various neck<lb/> positions was smallest in extension, moderate in<lb/> neutral position, and greatest in flexion. In our<lb/> study, we confirmed these facts using dynamic cer-<lb/>vical MR imaging.<lb/></p> <p>Generally, the cervical spine and cervical spinal<lb/> cord are considered to be longer, and the shapes of<lb/> these are considered to be straighter, in taller peo-<lb/>ple. We confirmed that this was indeed so in our<lb/> comparative study of the tall and short male<lb/> groups. However, in the random-selected females,<lb/> who were significantly shorter than the random-<lb/>selected males but whose height did not differ from<lb/> that of the short males, the cervical spines and<lb/> cervical spinal cords were significantly straighter<lb/> than those of both the random-selected and short<lb/> males. These results suggest that there is a " female-<lb/>type " of straight cervical spine and cervical spinal<lb/> cord, at least in young females aged 14 –29 years<lb/> <ref type="table">(Table 4)</ref>.<lb/></p> <p>The cervical spine and cervical spinal cord<lb/> shapes in the patient group did not differ signifi-<lb/>cantly from those in the random-selected female<lb/> group. Despite being males, our patients showed<lb/> nearly female-type spinal shapes. To our knowl-<lb/>edge, this is the first report referring the features of<lb/> the cervical spine and spinal cord shapes using<lb/> cervical MRI in patients with cervical flexion my-<lb/>elopathy and in young normal males and females.<lb/></p> <p>The reason for the marked male preponderance<lb/> in cervical flexion myelopathy is unclear. Why are<lb/> females almost never affected by cervical flexion<lb/> myelopathy? The cervical spine and cervical spinal<lb/> cord shapes of young females were straighter than<lb/> those of young males, but those of the patients were<lb/> actually very similar to the female-type. The patient<lb/> and random-selected female groups were compared<lb/> and the height and the cervical spine length/height<lb/> and spinal cord length/height ratios were found to<lb/> be significantly smaller in the latter. Consequently,<lb/> the absolute shortness of the cervical spine and<lb/> spinal cord in females group may be a reason why<lb/> females rarely suffer from cervical flexion myelopa-<lb/>thy, despite their cervical spine and cervical spinal<lb/> cord shapes being the same as those of the pa-<lb/>tients. A short cervical spine is considered to have<lb/> a more limited range of motion, resulting in less<lb/> disproportion of the cervical spine and spinal cord,<lb/> than that of a long cervical spine. However, the<lb/> male preponderance in cervical flexion myelopathy<lb/> has not been proven to derive from the differences<lb/> in the cervical spine and spinal cord shapes and<lb/> therefore, other causes, including gene-associated<lb/> factors, must be considered. Toma and Shiozawa<lb/> <ref type="biblio">[18]</ref> suspected that the difference of the growth<lb/> velocity between males and females was related to<lb/> this male preponderance.<lb/></p> <p>Comparison of the cervical MR images in the neu-<lb/>tral neck position of the patient and height-matched<lb/> male groups showed that the cervical spine length/<lb/> height and cervical spinal cord length/height ratios<lb/> were significantly larger in the former than in the<lb/> latter.<lb/></p> <p>What do the greater cervical spine length/height<lb/> and cervical spinal cord length/height ratios mean<lb/> when considering the etiology of cervical flexion<lb/> myelopathy? If a patient has disproportion, by na-<lb/>ture or in the process of growth, between the cer-<lb/>vical spine and spinal cord due to a long cervical<lb/> spine and a spinal cord of normal length, the spinal<lb/> cord in the cervical region would inevitably be<lb/> found to be elongated using our measuring meth-<lb/>ods. In this situation, without knowing the actual<lb/> tension in the spinal cord, it is unclear whether the<lb/> cervical spinal cord itself is elongated or the upper<lb/> thoracic spinal cord is contained in the cervical<lb/> region because of upward migration of the entire<lb/> spinal cord. Alternatively, if a patient grows with a<lb/> slightly extended neck as his/her " natural " position,<lb/> the cervical spine and spinal cord may be stretched<lb/> even in the neutral neck position, causing an elon-<lb/>gated cervical spine and spinal cord, which would<lb/> be a " relative flexed neck position " for this patient.<lb/> Also, the range of motion of a long cervical spine is<lb/> considered to be greater than that of a short one,<lb/> and this may contribute to flexion myelopathy.<lb/></p> <p>We suspect that the spinal cords in the cervical<lb/> regions of the patients were subject to a " stretch-<lb/>ing " force, because their cervical spines and spinal<lb/> cords appeared to be straighter and their cervical<lb/> spinal cords ran along shorter routes (lower cervi-<lb/>cal spinal cords were located more posteriorly in<lb/> their cervical spinal canals) than those in normal<lb/> males. If the cervical spinal cord is stretched even<lb/> in the neutral neck position, further stretching force<lb/> will be added to the spinal cord when the neck is<lb/> flexed, resulting in the overstretched condition de-<lb/>scribed by Yada et al <ref type="biblio">[19]</ref>. Under such " over-<lb/>stretched " conditions, the spinal cord and dural sac<lb/> will migrate forward to run along the shortest route,<lb/> regardless of compression by the vertebral bodies<lb/> and intervertebral discs. Therefore, we consider<lb/> that the cause of the malalignment of the posterior<lb/> surfaces of the cervical vertebrae shown on the<lb/> cervical plain X-rays in the flexed and neutral neck<lb/> positions was force, affecting the cervical spine as a<lb/> result of the compressive power in the opposite<lb/> up-down direction to the force due to tension in the<lb/> stretched spinal cord and the dural sac.<lb/></p> <p>Mukai et al <ref type="biblio">[12]</ref> observed straight neck or kypho-<lb/>sis on the cervical X-rays of patients with JMAU. We<lb/> found malalignment of the posterior surfaces of the<lb/> vertebrae in the flexed and neutral neck positions<lb/> and resolution of this finding in neck extension on<lb/> the cervical X-rays of patients with juvenile cervical<lb/> flexion myelopathy. On the dynamic cervical plain<lb/> X-rays of a case of JMAU <ref type="biblio">[15]</ref>, this phenomenon was<lb/> noticeable but not referred to by the authors.<lb/> Therefore, this phenomenon may be common in<lb/> patients with juvenile cervical flexion myelopathy<lb/> but this has never, to our knowledge, been<lb/> described.<lb/></p> <p>The extended neck position, in which the cervical<lb/> spine and spinal cord are at their shortest, is con-<lb/>sidered to be the most natural and physiological<lb/> position for patients with cervical flexion myelopa-<lb/>thy, because in this position, the malalignment of<lb/> the cervical vertebrae was corrected and the cervi-<lb/>cal spine length/height and spinal cord length/<lb/> height ratios in the patient group did not differ from<lb/> those on the neutral neck MR images in the normal<lb/> controls in this study. In light of these findings, we<lb/> now try to maintain a slightly extended neck posi-<lb/>tion when treating such patients with a cervical<lb/> collar or surgical cervical fusion.<lb/></p> <p>In conclusion, in our study, the spinal cords in<lb/> the cervical regions of the patients with juvenile<lb/> cervical flexion myelopathy were considered to be<lb/> stretched, even in the neutral neck position. We<lb/> believe this reflects disproportion of the cervical<lb/> spine and spinal cord.</p> </text> </tei>