Structured MRI reporting in spinal dysraphism

Abstract

Magnetic resonance evaluation of spinal dysraphism can be confusing for inexperienced radiologists and a detailed, step-by-step evaluation of the normal and abnormal imaging findings can help garner the diagnosis. The purpose of this article is to review the existing literature and to provide a comprehensive, structured, template checklist-style format for reporting spinal dysraphism that can help inexperienced radiologists to systematically analyze and report all the significant and ancillary findings in cases of spinal dysraphism and efficiently communicate the findings to the treating physician/surgeon.

Keywords

Spinal dysraphism spina bifida magnetic resonance imaging structured reporting

Introduction

Spinal dysraphism is a blanket term applied to a spectrum of congenital aberrations in the formation of the neural arch (1). Magnetic resonance imaging (MRI) has become the investigation of choice for evaluation of these disorders. These disorders can be complicated and variable in imaging appearance and can be confusing for the inexperienced radiologists. A detailed and systematic analysis of each component of the spine is crucial for obtaining a correct diagnosis. The superior diagnostic efficacy of the modern high field MR scanners mandates a systematic, step-by-step evaluation of the normal and abnormal imaging findings so that no critical, surgically relevant finding is missed (2). Although various studies have assessed these disorders on MRI in the last two decades utilizing various classification approaches, little stress has been placed on structured reporting of the findings. We aimed to suggest a standard checklist-style reporting template to describe pediatric spinal dysraphism.

Vertebral anomalies

Reporting of the vertebral anomalies starts with the counting of all vertebrae and designating the abnormality with an addendum to this count. A normal spine should be reported as C-7/D-12/L-5/S-5, whereas a block vertebra at C2 to C4 and an additional hemivertebra at C6 level should be reported as C-8(block 2-4; 6h)/D12/L-5/S5 (3).

Congenital vertebral defects include the anomalies of vertebral formation, segmentation, or combination of both (Fig. 1). Lesions can be classified into simple (1–2 vertebra) or complex (>2 vertebra). Segmentation anomalies include block vertebra (fusion of vertebral body and/or posterior elements), unilateral unsegmented bars, lumbosacral transitional vertebra, and additional vertebra (4). The fusion of unilateral lamina with the contralateral lamina of adjacent vertebra is known as an intersegmental laminar fusion (3). Unilateral unsegmented bars are a cause of congenital scoliosis with unilateral fusion of lamina, pedicles, and posterolateral aspect of vertebral body at the apex of the concave portion of the curve.

Fig. 1.

Vertebral anomalies. (a) Formation defects, i.e. butterfly vertebra (b), and hemivertebra (h); (b) segmentation abnormality consisting of block vertebra (bl).

Defects of vertebral formation include butterfly vertebra, hemivertebra (unilateral wedge vertebra), dorsal wedged vertebra, spina bifida occulta, pedicular anomaly, hypoplasia/agenesis, coronal cleft, asomia (absence of a vertebra despite the presence of posterior elements), and notochord remnant (4). The unilateral wedge vertebra is called hemivertebra and is invariably associated with scoliosis at birth. In hemimetameric segmental displacement, there is the presence of two halves of hemivertebra, with one of the halves fused with the vertebral body above or below the affected segment. A coronal cleft is a vertical gap between the anterior and posterior halves of the vertebral body. Butterfly vertebra (sagittal cleft) is referred to as an unfused medial and lateral half of vertebral bodies (4,5).

An attempt should be made to identify various clinical syndromes that feature vertebral segmentation and formation anomalies (Fig. 2). The presence of multiple unsegmented cervical vertebrae points towards Klippel–Feil anomaly which is also characterized by omovertebral bone, anteroposterior vertebral bodies narrowing, hemivertebra, and congenital elevation of scapula (Sprengel deformity) (6). Multiple mid-thoracic vertebral anomalies with radiating short ribs points towards Jarcho Levin syndrome (6,7). Associated vertebral malalignment should be characterized into scoliosis, kyphosis, and lordoscoliosis (4). Hemivertebra were characterized by scoliosis at birth (5). Dorsal hemivertebra is associated with kyphosis.

Fig. 2.

Syndromic associations of spinal dysraphism. (a, b) A case with multiple cervical block vertebrae (black arrow), vertebral collapse (star), and left sided omovertebral bar (white arrow) along with elevated scapula, suggestive of Klippel-Feil anomaly. (c) A second case showing multiple mid-thoracic vertebral anomalies with radiating short ribs on right side (small white arrows) pointing towards Jarcho–Levin syndrome.

Spinal cord and spinal canal

Tethered cord syndrome

Tethered cord syndrome (TCS) denotes a clinical entity and not a malformation and results due to traction on conus or other part of spinal cord (8,9). It consists of atypical scoliosis, spastic gait, low backache (worsened by activity), leg pain (predominantly posteriorly), bladder dysfunction, lower extremity sensory abnormality, and abnormal lower extremity reflexes (6,10,11). Most commonly, the cord is tethered to an associated dysraphic masses (such as myelomeningocele/lipomyelomeningocele) lesions producing tethering of cord include tight filum terminale, fibrosis, post-surgical adhesions, or filar lipoma. Filum terminale is a fibrous band extending from the tip of conus traversing from intradural space traversing through all the layers of meninges to the dorsal surface of the first coccygeal vertebra (4,12). Thick and short filum leads to the transmission of pressure over the conus medullaris (Fig. 3), which is most often but not always low-lying depending upon the impact on an ascent of conus (13,14). Fibrolipomas of the filum terminale may also lead to cord tethering and may present with symptoms of tethered cord at any age (4,15,16) . Fibroneural tracts can also lead to spinal cord tethering and consist of dorsal dermal sinus (DDS), limited dorsal myeloschichis, or meningocele manqué. On MRI, the level of conus should be first evaluated on axial T2-weighted (T2W) images. Next, the measurement of filar thickness should be performed at the L5-S1 level, and a thickness >2 mm is considered abnormal (4,13). Proximal to this level, the filum is stretched and may appear falsely as of normal thickness. Next, a search should be done to look for fibro lipoma (lipid component bright on T1-weighted [T1W] image) till the bottom of thecal sac (4).

Fig. 3.

(a, b) Tight filum terminale is seen as filar thickening >2 mm (white arrow). Also seen is a short segment syrinx (s) in a visualized spinal cord.

Split cord syndrome

Complete duplication of the spinal cord is known as dimyelia. The term diplomyelia refers to an isolated accessory spinal cord, ventral or dorsal to normal cord, and diastematomyelia (DSM) is lateral duplication of the spinal cord into two hemicords (Fig. 4) (17). The two hemicords can be symmetric or asymmetric with each containing a dorsal and a ventral nerve root (6). Type 1 DSM contains the two hemicords in separate dural sacs with the presence of fibrous, cartilaginous, or osseous spur between them. Identification of spur is important and should be sought for using a combination of T1W, T2W, and if possible T2*-weighted images. A complete resection of the spur is essential for complete untethering of the cord (4,18–20). The presence of two hemicords in a single dural sac is referred to as type 2 DSM. An attempt should be made to identify abnormal bands in type 2 DSM, which adheres to the dura producing cord tethering, also referred to as meningocele manqué (6). Having identified the type of DSM, scrutiny of other findings should be done that can lead to neurological symptoms. One or both hemicords may be associated with syrinx. One of the hemicords may fail to neurulate and can present as a hemimyelocele or with hemimyelomeningocele with associated expansion of fluid-filled meninges. Teratoma can also be associated with one of the hemicords (8,21).

Fig. 4.

Diastematomyelia (DSM). (a) Type I DSM with a bony spur (white arrow) separating the two hemicords. (b) Type II DSM with two hemicords contained in a single dural sac. Also noted is meningomyelocele (MMC). (c, d) Type II DSM associated with hemimyelocele in which left-sided hemicord along with its non-expanded dural sac herniated through the left vertebral defect all the way to skin (black arrow). One of the cases with type I DSM was associated with a teratoma (*) in the left hemicord (a, e, f).

Syrinx

Fluid-filled cavities extending craniocaudally within the spinal cord are known as syringohydromyelia. The term hydromyelia is applied to the dilatation of the central canal, whereas syrinx refers to the cavities in cord parenchyma usually present in the watershed region of anterior and posterior spinal arteries (dorsolateral to central canal). Syrinx extending to the medulla is called as syringobulbia (Fig. 5) and that extending up to the fourth ventricle is referred to as communicating syringomyelia (4). Sometimes, there can be septations within the syrinx cavity. The cord parenchyma adjacent to the syrinx may exhibit a high T2 signal due to edema or gliotic changes, also known as presyrinx (22). A short segment dilated terminal central canal in the conus is known as a persistent fifth ventricle (23). This finding should be differentiated from a filar cyst, which is present within the filum (24).

Fig. 5.

Syringobulbia. Short segment cervical syrinx seen extending to the medulla.

Linear tethering tracts

Dorsal dermal sinus

DDS refers to an epithelial lined sinus track extending from the skin surface passing through the subcutaneous tissue to the spinal canal (Fig. 6). It may terminate into the dura, causing its posterior tenting. DDS may extend into subarachnoid space, filum terminale, nerve roots, and fibrotic nodule on the dorsal surface of the spinal cord, conus medullaris or rarely into the central spinal canal (4,25–27) . The DDS may pass through a defect in posterior elements of the spine (spina bifida) or through a defect in interspinous ligaments (26,28). It is most commonly located in lumbosacral regions, but can also be seen in cervical, thoracic, or occipital regions (29). DDS can be identified as a dark/gray band in the background of bright subcutaneous fat extending from dura to the skin using axial or sagittal T1W images. DDS passing through the dural or vertebral defects can be assessed on T2W axial images, although intrathecal portions are difficult to see. DDS can be associated with meningitis or subdural, epidural, subarachnoid, or subcutaneous abscess (4). When associated with dermoid or epidermoid, chemical meningitis can also occur (4,31). Complete evaluation of DDS should be achieved with the contrast study to rule out meningitis and diffusion-weighted images to look for an associated epidermoid cyst. Ruptured or infected dermoid or epidermoid leads to heterogenous subarachnoid space and imaging picture similar to arachnoiditis. Approximately 10% of the DDS are associated with dermoid, epidermoid, split cord, tethered cord, intraspinal lipomatous lesions (lipoma/lipomyelocele/lipomyelomeningocele [LMMC]) (6,8,32). Most of the associated lesions exist separately from the sinus tract. However, uncommonly the sinus tract can be seen terminating into LMMC or teratoma.

Fig. 6.

Dorsal dermal sinus (DDS). (a) Sinus track (small black arrow) extending from the skin to dura via the vertebral defect. (b, c) DDS was associated with teratoma entirely in subcutaneous plane which consisted of a solid cystic mass with fatty elements (white arrow) and only the fibrous track (long black arrow) traversing the vertebral defect. (d–f) Another case showing posterior herniation of neural placode (curved short white arrow) through the vertebral defect and a heterogeneous mass attached to it containing fat, solid, and cystic components protruding beyond the skin. The fatty component of the lesion is not contiguous with the subcutaneous tissue suggestive of teratoma arising within the DDS. (g, h) DDS is seen (red arrow) extending from skin to the LMC up to the junction of the lipoma and everted deformed posterior vertebral element.

Limited dorsal myeloschichis

An important imaging differential diagnosis of DDS is limited dorsal myeloschichis (LDM). Also known as dermal sinus-like stalk or pseudodermal sinus, LDM consists of a fibroneural tract connecting the skin to the tethered spinal cord. Unlike DDS, it is a solid tract without a lumen; therefore, there is minimal risk of associated infection. On MRI, LDM can appear as a linear subcutaneous tract identical to DDS. In contrast to DDS, the intrathecal portion of the tract and its attachment site on the spinal cord is distinctly visualized in LDM and can be associated with a characteristic cigarette-burn mark on overlying skin (33).

Meningocele manqué

A linear tethering band extending from the dorsal surface of the spinal cord to the dura and not proceeding to the posterior subcutaneous tissue is known as meningocele manqué (MM). They may consist of anomalous nerve roots, fibrous bands, or a combination of both. It is extremely difficult to be visualized on preoperative imaging and requires high degree of suspicion. Clinical features of tethered cord without any other obvious tethering lesion should prompt the radiologist to search for a thin fibrous band in the posterior aspect of the spinal cord, causing focal cord deformity. Approximately one-fifth of type II distematomyelia may harbor MM. An overlying cigarette-burn mark on the skin may be present in half of the cases (34).

Dysraphic mass lesions

Evaluation of spinal dysraphism associated with masses on MRI can be confusing and should be achieved by step-by-step evaluation of the contents of these lesions (fat, fluid, soft tissue, and neural components), extent (intraspinal, both intra and extraspinal, subcutaneous), the components passing through the dural and vertebral defect, and interface of the intratumoral fat with the neural placode and with the subcutaneous fat.

Fluid-containing lesions

Dorsal herniation of cerebrospinal fluid (CSF) containing a meningeal sac through vertebral defect is called a meningocele. MRI is performed only uncommonly to assess its shape, to look for the neural tissue within the sac, to assess its relation with conus and filum, and to evaluate for associated malformations. Uncommonly, a CSF-filled sac can herniate laterally into thorax via enlarged neural foramina (lateral thoracic meningocele) (Fig. 7) (35). They then course anteriorly in intercostal space into the extrapleural thoracic gutter. Intradural extramedullary cyst in the presence of spinal dysraphism can be an epidermoid cyst, arachnoid cyst, filar cyst, or neurenteric cyst. Intraspinal enterogenous cysts (neurenteric cysts) are intradural extramedullary cysts usually ventral to the cord; however, they may show an intramedullary component in 15% of cases. Less commonly, they may be dorsal to the cord or between two hemicords in DSM (4,36,37). Other cystic lesions in relation to the conus are an ependymal cyst or persistent fifth ventricle.

Fig. 7.

(a–c) Meningomyelocele containing cerebrospinal fluid (CSF) and neural elements with both CSF and neural elements (black arrow) crossing the duro-vertebral defect (a), only neural elements (white arrow) crossing the defect (b), and only meningeal lining (without visible fluid or neural placode) (white arrow) crossing the defect (c), which expanded into a fluid- and soft-tissue containing sac in the subcutaneous place. (d) Lateral thoracic meningocele in which fluid filled meningeal sac is seen herniating through left neural foramina (f) extending in left lateral thorax in extrapleural space.

Fluid with soft tissue (neural elements)

The neural placode is abnormal neural tissue located at either the terminal end of the open spinal cord or can be segmental, with the spinal cord extending further inferiorly (38). Herniation of the fluid-filled sac with neural elements points towards MMC, which is an open defect and is seldom referred for imaging (Fig. 7). The neural placode in MMC is seen protruding beyond the dysraphic spinal defect due to the subarachnoid cystic component. If there is no cystic component, the placode remains within the canal, and the condition is known as myeloschisis or myelocele (39).

If the protruded sac contains conus and neural placode, which contains the dilated central canal (syringocele), the condition is known as terminal myelocystocele. The identification of the hydromyelic cord through the defect is essential for diagnosis. Fluid may not extend to the subcutaneous tissue. Within the subcutaneous tissue, the herniated cord with syringocele may be surrounded by a CSF-filled meningocele. The syringocele and myelocele should not communicate with each other (40 –42). The fluid in the dilated central canal may be multiloculated. The outer meningocele communicates with the spinal subarachnoid space (40).

Fat-containing masses

Assessment of fat-containing masses traversing through the vertebral defect or connected to the meninges presents a diagnostic puzzle to the radiologists. These lesions can be divided into four primary groups (4,43): (i) intradural lipoma; (ii) lipomyeloceles/lipomyelomeningocele – dorsal/caudal/combined (transitional); (iii) lipomas of caudal cell mass – terminal lipoma, filar fibro lipoma; and (iv) teratoma.

Intradural lipomas are lipid-containing masses completely encapsulated within the intact dural lining. These lesions abut the neural placode (juxtamedullary) and are subpial in location. They may expand the spinal canal. Most intradural lipomas are near the conus or may be present in the cervicothoracic spine (6). Terminal lipoma also occurs at the conus but is invariably tethered and low-lying (Fig. 8). They may extend through the bifid spine. Lipomas also may occur within the filum terminale and might contain some fibrous elements (fibro lipoma) (4,6).

Fig. 8.

(a) Terminal lipoma (white arrow) which is seen as T1 hyperintense lesion attached to a low-lying tethered cord. (b) Lipoma of the filum terminale (black arrow) attached to the meningomyelocele and was seen herniated posteriorly along with it.

In lipomyelocele (LMC), the lipoma is seen extending from the subcutaneous tissues, traversing the dural and vertebral defect entering the spinal canal attaching to and tethering the spinal cord (Fig. 9). The placode lipoma junction is present within the spinal canal or at the edge of the canal (38). The junction may be smooth or irregular with fatty projections extending into the central canal, leading to dilated central canal (9). The lipoma is diffusely or focally continuous with the subcutaneous fat (4). In LMMC, the expanded CSF space is seen herniating dorsally through the bony and dural defect along with the neural placode (Fig. 10). The placode lipoma interface usually is seen outside the spinal canal; however, depending on the degree of herniation, it may lie partially within and partly outside the canal. The intraspinal component of lipoma can also pass upward or enter within the central canal or may form an isolated intramedullary lipoma at higher levels (4). The lipoma may enter the epidural space and form an epidural lipoma. Asymmetrical lipoma can enter the spinal canal on one side and force the neural placode and fluid-filled meningeal sac to herniate on the other side of the duro-vertebral defect. One of the cases of LMMC presenting at our center showed a fat placode interface entirely within the spinal canal with fat and neural elements traversing the canal and the fluid component seen entirely in the subcutaneous plane (Fig. 11). This observation is in contrast with the existing literature and highlights that the distinction between LMMC and LMC should be based upon the presence or absence of fluid component in the mass rather than the assessment of the location of the fat placode interface. Moreover, the distinction between the neural placode and other neural elements itself may become confusing in these complex masses. It should be noted that these lesions may contain a variable degree of mesenchymal components (muscle fiber, neuroglial tissue, meningeal tissue calcification, and osseous components) and may mimic teratoma (4,44).

Fig. 9.

Lipomyelocele (LMC). (a) Subcutaneous fat entering through the canal in all the cases (L), with the interface of placeode (p) and lipoma within the canal. The fat seen in all these cases were contiguous with subcutaneous fat and often with the expanded epidural fat as well (b, c). Some neural elements (nerve roots) seen within the intralesional fat extending into subcutaneous fat as well as epidural fat (white arrows, b, c).

Fig. 10.

Lipomyelomeningocele (LMMC). (a–c). A case of LMMC showing fat (l), neural placode (p), and cerebrospinal fluid (CSF) traversing the defect with lipoma-placode interface extending from inside the canal to outside. (d, e) Another case of LMMC showing neural placode (p) and lipoma (L) traversing the defect with the placode–lipoma interface extending from spinal canal to the subcutaneous fat. CSF component (c) in this case is entirely seen in subcutaneous plane.

Fig. 11.

Atypical LMMC. Case 1 (a–c) and case 2 (d–f). A case of LMMC showing fat (L) and neural elements (nerve roots) crossing the vertebral defect with the placode–lipoma interface entirely within the spinal canal similar to LMC. However, in both these the backward herniating neural elements expanded into CSF-filled cystic component (c) in the subcutaneous plane pointing towards LMMC. Case 2 showed posterior beaking of the meningeal sac towards the vertebral defect (e) . CSF, cerebrospinal fluid; LMC, lipomyelocele; LMMC, lipomyelomeningocele.

Congenital inclusion tumors (CIT) arise from a disorder of entrapment and embryogenesis of germ cell layers and include epidermoid, dermoid, teratoma, and teratoid cysts. These complex disorders with the variable histological spectrum are challenging to classify (45,46). Some authors consider intraspinal cystic teratoma to be related to neurenteric cysts. These lesions can present as intramedullary, intradural- extramedullary, extradural masses, or a combination of these. Various reported associations with spinal dysraphic states include MMC, LMMC, split cord malformation, and dorsal dermal sinus (45). On MRI, teratomas are seen as cystic, solid, unilocular, or multilocular lesions showing the presence of fat (T1 hyperintensity). Calcification, bony, and other mesenchymal elements may be present. The intratumoral fat in these lesions shows no communication with subcutaneous fat, a point that can help differentiation from LMC/LMMC.

Caudal cell mass

Missing caudal elements of the spine is known as caudal agenesis or caudal regression syndrome (CRS). It can range from a clinically asymptomatic absence of few terminal sacral segments to a significantly shortened spine (Fig. 12a). Type I CRS presents with high-lying (above the L1 level), abrupt termination of a wedge-shaped conus, significantly shortened vertebral column, and severe associated orthopedic deformity (6). Cauda equine nerve roots show abnormal course and the cord is not tethered. Type 2 CRS has a low-lying tethered cord with partial agenesis of conus medullaris. An associated filar lipoma or LMMC may be present (9). Rarely, segmental spinal dysgenesis (SSD) or agenesis of the cord can occur in thoracolumbar or lumbar level (6).

Fig. 12.

Caudal mass abnormalities. (a) Shortened vertebral column with low-lying conus along with lipomyelocele, suggestive of type II caudal regression syndrome. (b) Sacrococcygeal teratoma (T) attached to coccygeal bone containing solid, cystic, and fat components; however, it did not insinuate into the spinal canal.

Anterior meningocele can also occur in the sacral region protruding in the presacral region via sacral neural foramina (47). Sacrococcygeal teratoma is a mass containing varying proportions of fat, fluid, calcific/bony elements, and soft-tissue elements occurring in relation to sacrococcygeal bone (Fig. 12b) (48). However, no extension into the spinal canal is seen in these cases.

Conclusion

In conclusion, a structured checklist-style reporting format can help the trainee radiologists to systematically analyze and report all the significant and ancillary findings in cases of spinal dysraphism and efficiently communicate the findings to the treating physician/surgeon. The present study provides the first iteration of the utilization of a structured report template to retrospectively analyze the cases of spinal dysraphism (Table 1). We further envisage that the suggested reporting system can evolve after widespread prospective application of this template and gathering surgical feedback. Subsequently, the gathered data can be utilized to facilitate research in artificial intelligence (AI) and machine learning applications.

Table 1.

Suggested reporting template for spinal dysraphism.

Vertebral body lesions
Vertebral count	Cervical/thoracic/lumbar
Formation anomalies	<butterfly> <hemivertebra> <dorsal wedge vertebra> <hypoplasia/agenesis/asomia> <coronal cleft>
Segmentation anomalies	<block vertebra> <unsegmented bar> <additional vertebra> <lumbosacral transitional vertebra>
Mixed anomalies	Describe
Vertebral alignment	<normal> <scoliosis> <kyphosis> <kyphoscoliosis>
Spinal canal
Low-lying conus/tethered cord	<present> <absent>
Filar thickness	<normal> <tight filum >
Split cord	<complete> < partial: level of involvement> <anteroposterior > <lateral split cord>Type: type I/type II DSMHemimyelomenigocele
Syrinx	Extent: <syringomyelia> <hydromyelia> <syringobulbia>Morphology: <septation> <diameter> <presyrinx>
Dural ectasia	Present/absent
Posterior vertebral elements
Spina bifida	Level and description
Linear sinus like lesion
Extent	<intrathecal> <intrathecal and subcutaneous> <only subcutaneous>
Terminating structure	<dura> <cord/conus/filum terminale> <associated epidermoid/dermoid/LMC/LMMC>
Intrathecal portion	<clearly visible > <non-visible>
Attachment on the spinal cord	<visible> <non-visible>
Evidence of meningitis	<present> <absent>
Mass-containing lesions
Location	<intradural> <intradural extension to posterior subcutaneous fat> <intradural with anterior extension>
Contents	<CSF> <neural elements> <fat>
Contents traversing the vertebral defect	<CSF> <neural elements> <fat> <sinus>
Interface of lipoma	<Continuity with subcutaneous fat> <Interface with neural elements: outside spinal canal/ inside spinal canal/ extending from outside to inside spinal canal >
Caudal cell mass
Caudal regression syndrome	Type I/II
Sacrococcygeal teratoma	Type I/II/III/IV
Skin covering
Intact/mass protruding through skin/sinus like opening/other appearance
Associated anomalies
Brain/CV junction	Chiari malformation/cortical dysplasia/others
Pelvis/abdomen	Renal/anorectal/others

CSF, cerebrospinal fluid; CV, craniovertebral; DSM, diastematomyelia; LMC, lipomyelocele; LMMC, lipomyelomeningocele.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Ishan Kumar

References

Venkataramana

NK.

Spinal dysraphism. J Pediatr Neurosci 2011; 6:S31–S40.

European Society of Radiology (ESR). ESR paper on structured reporting in radiology. Insights Imaging 2018; 9:1–7

Barkovich AJ. Congenital Anomalies of the spine. In: Paediatric Neuroimaging. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2011; pp. 857–916.

Trenga

Singla

Feger

, et al. Patterns of congenital bony spinal deformity and associated neural anomalies on X-ray and magnetic resonance imaging. J Child Orthop 2016; 10:343–352.

Kumar

Guinto

Jr Madewell

, et al. The vertebral body: radiographic configurations in various congenital and acquired disorders. Radiographics 1988; 8:455–485.

Moore

Congenital Abnormalities of the Spine In: BD

Coley

, editor. Caffey’s Pediatric Diagnostic Imaging. Philadelphia, PA: Elsevier Saunders, 2013:449–460.

Vázquez-López

López-Conde

Somoza-Rubio

, et al. Anomalies of vertebrae and ribs: Jarcho Levin syndrome. Description of a case and literature review. Joint Bone Spine 2005; 72:275–277.

Rossi

Cama

Piatelli

, et al. Spinal dysraphism: MR imaging rationale. J Neuroradiol 2004; 31:3–24.

Rossi

Biancheri

Cama

, et al. Imaging in spine and spinal cord malformations. Eur J Radiol 2004; 50:177–200.

10.

Tortori-Donati

Rossi

Biancheri

, et al. Magnetic resonance imaging of spinal dysraphism. Top Magn Reson Imaging 2001; 12:375–409.

11.

Warder

DE.

Tethered cord syndrome and occult spinal dysraphism. Neurosurg Focus 2001; 10:e1.

12.

Emery

Lendon

RG.

Lipomas of the cauda equina and other fatty tumors related to neurospinal dysraphism. Dev Med Child Neurol 1969; 11:62–70.

13.

Raghavan

Barkovich

Edwards

, et al. MR imaging in the tethered cord syndrome. Am J Neuroradiol 1989; 10:27–36.

14.

Hall

Albright

Brunberg

JA.

Diagnosis of tethered cords by magnetic resonance imaging. Surg Neurol 1988; 30:60–64.

15.

Pang

Wilberger

Jr.

Tethered cord syndrome in adults. J Neurosurg 1982; 57:32–37.

16.

Pierre-Kahn

Zerah

Renier

, et al. Congenital lumbosacral lipomas. Childs Nerv Syst 1997; 13:298–335.

17.

Hori

Fischer

Dietrich-Schott

, et al. Dimyelia, diplomyelia, and diastematomyelia. Clin Neuropathol 1982; 1:23–30.

18.

Pang

Dias

Ahab-Barmada

Split cord malformation: part I: a unified theory of embryogenesis for double spinal cord malformations. Neurosurgery 1992; 31:451–480.

19.

Guthkelch

AN.

Diastematomyelia with median septum. Brain 1974; 97:729–742.

20.

Naidich

Harwood-Nash

DC.

Diastematomyelia: hemicord and meningeal sheaths; single and double arachnoid and dural tubes. Am J Neuroradiol 1983; 4:633–636.

21.

Duckworth

Sharrard

Lister

, et al. Hemimyelocele. Dev Med Child Neurol 1968; 10:69–75.

22.

Fischbein

Dillon

Cobbs

, et al. The “presyrinx” state: a reversible myelopathic condition that may precede syringomyelia. Am J Neuroradiol 1999; 20:7–20.

23.

Coleman

Zimmerman

Rorke

LB.

Ventriculus terminalis of the conus medullaris: MR findings in children.

Am J Neuroradiol 1995; 16:1421–1426.

24.

Irani

Goud

Lowe

LH.

Isolated filar cyst on lumbar spine sonography in infants: a case-control study. Pediatr Radiol 2006; 36:1283–1288.

25.

Schwartz

HG.

Congential tumors of the spinal cord in infants. Ann Surg 1952; 136:183–192.

26.

Matson

DD.

Neurosurgery of infancy and childhood. 2nd ed. Springfield: Charles C. Thomas, 1969.

27.

Friede

RL.

Developmental neuropathology. 2nd ed. Berlin: Springer-Verlag, 1989.

28.

Harwood-Nash

Fitz

CR.

Normal spine, abnormal spine, myelography, mass lesions of the spinal canal. In: DC

Harwood-Nash

Fitz

, editors. Neuroradiology in infants and children. St. Louis, MO: Mosby, 1976:1054–1227.

29.

Ackerman

Menezes

Spinal congenital dermal sinuses: A 30-year experience. Pediatrics 2003; 112:641.

30.

Mount

LA.

Congenital dermal sinuses as a cause of meningitis, intraspinal abscess and intracranial abscess. JAMA 1949; 139:1263–1268.

31.

List

CF.

Intraspinal epidermoids, dermoids, and dermal sinuses. Surg Gynecol Obstet 1941; 73:525–538.

32.

Barkovich

Edwards

Cogen

PH.

MR evaluation of spinal dermal sinus tracts in children.

Am J Neuroradiol 1991; 12:123–129.

33.

Lee

Cheon

J-E

Choi

, et al. Limited dorsal myeloschisis and congenital dermal sinus: comparison of clinical and MR imaging features. AJNR Am J Neuroradiol 2017; 38:176–182.

34.

Schmidt

Bryant

Iwanaga

, et al. Menigocele manque: a comprehensive review of this enigmatic finding in occult spinal dysraphism. Childs Nerv Syst 2017; 33:1065–1071.

35.

Oner

Uzun

Tokgöz

, et al. Isolated true anterior thoracic meningocele. AJNR Am J Neuroradiol 2004; 25:1828–1830.

36.

D’Almeida

Steward

Jr. Neurenteric cyst: case report and literature review. Neurosurgery 1981; 8:596–599.

37.

Rizk

Lahoud

Maarrawi

, et al. Acute paraplegia revealing an intraspinal neurenteric cyst in a child. Childs Nerv Syst 2001; 17:754–757.

38.

Badve

Khanna

Phillips

, et al. MRI of closed spinal dysraphisms. Pediatr Radiol 2011; 41:1308–1320.

39.

Mirsky

Schwartz

Zarnow

DM.

Diagnostic features of myelomeningocele: the role of ultrafast fetal MRI. Fetal Diagn Ther 2015; 37:219–225.

40.

Kumar

Afsal

Garg

Imaging spectrum of spinal dysraphism on magnetic resonance: A pictorial review. World J Radiol 2017; 9:178–190.

41.

Byrd

Harvey

Darling

CF.

MR of terminal myelocystoceles. Eur J Radiol 1995; 20:215–220.

42.

Peacock

Murovic

JA.

Magnetic resonance imaging in myelocystoceles. Report of two cases. J Neurosurg 1989; 70:804–807.

43.

Arai

Sato

Okuda

, et al. Surgical experience of 120 patients with lumbosacral lipomas. Acta Neurochir (Wien) 2001; 143:857–864.

44.

Friede

RL.

Developmental neuropathology. 2nd ed. Berlin: Springer-Verlag, 1989.

45.

Koen

McLendon

George

TM.

Intradural spinal teratoma: evidence for a dysembryogenic origin. Report of four cases. J Neurosurg 1998; 89:844–851.

46.

Kubie

Fulton

JF.

A clinical and pathological study of two teratomatous cysts of the spinal cord, containing mucus and ciliated cells. Surg Gynecol Obstet 1928; 47:297–311.

47.

Manson

Comalli-Dillon

Moriaux

Anterior sacral meningocele: management in gynecological practice. Ult Obstet Gynecol 2007; 6:893–896.

48.

Keslar

Buck

Suarez

ES.

Germ cell tumors of the sacrococcygeal region: radiologic-pathologic correlation.

Radiographics 1994; 14:607–620.