Review Article Pediatric Spinal Trauma Thierry A.G.M. Huisman, MD, Matthias W. Wagner, MD, Thangamadhan Bosemani, MD, Aylin Tekes, MD, Andrea Poretti, MD From The Russell H. Morgan Department of Radiology and Radiological Science, Division of Pediatric Radiology, Section of Pediatric Neuroradiology, The Johns Hopkins University School of Medicine, Baltimore, MD.
ABSTRACT Pediatric spinal trauma is unique. The developing pediatric spinal column and spinal cord deal with direct impact and indirect acceleration/deceleration or shear forces very different compared to adult patients. In addition children are exposed to different kind of traumas. Moreover, each age group has its unique patterns of injury. Familiarity with the normal developing spinal anatomy and kind of traumas is essential to correctly diagnose injury. Various imaging modalities can be used. Ultrasound is limited to the neonatal time period; plain radiography and computer tomography are typically used in the acute workup and give highly detailed information about the osseous lesions. Magnetic resonance imaging is more sensitive for disco-ligamentous and spinal cord injuries. Depending on the clinical presentation and timing of trauma the various imaging modalities will be employed. In the current review article, a summary of the epidemiology and distribution of posttraumatic lesions is discussed in the context of the normal anatomical variations due to progressing development of the child.
Keywords: Trauma, spine, children, imaging. Acceptance: Received May 14, 2014, and in revised form July 23, 2014. Accepted for publication August 16, 2014. Correspondence: Address correspondence to Thierry A.G.M. Huisman, MD, The Russell H. Morgan Department of Radiology and Radiological Science, Director of Pediatric Radiology and Pediatric Neuroradiology, The Johns Hopkins School of Medicine, Charlotte R. Bloomberg Children’s Center, Sheikh Zayed Tower, Room 4174, 1800 Orleans Street, Baltimore, MD 21287-0842. E-mail:
[email protected]. J Neuroimaging 2015;25:337-353. DOI: 10.1111/jon.12201
Introduction Pediatric spinal trauma is unique because of various reasons.1,2 First of all, the biomechanical properties of the initially predominantly cartilaginous pediatric spine are very different compared to the adult spine.3,4 In the early years of life, the stability of the pediatric spine relies predominantly on the cartilaginous spine and the relatively lax ligaments. The pediatric spine is consequently more mobile and deformable compared to adults. Traumatic forces will be absorbed differently in children and varies for the various age groups. Vertebral fractures are less frequent in young children compared to adults. Dislocations, ligamentous injuries, epiphyseal detachments, and lesions of the ossification centers are more frequent. With progressing age and physical activity of the child the paraspinal musculature will develop and contribute to the dynamic stability of the spinal column. In addition the proportions of the pediatric body changes dramatically in the first years of life. Young children have a relatively large and heavy head compared with the torso. Later in life, the head-to-torso ratio progressively decreases. This is of particular importance for the craniocervical junction. Next to the large head and the “weak” neck musculature, the “young” pediatric spine is also more mobile due to the shallow occipital condyles, the horizontal orientation of the facet joints (30° vs.
Copyright
60°-70° in adults), small uncinate processes, immature uncovertebral joints, increased elasticity of the posterior joint capsules, and a cartilaginous junction between the vertebral bodies and their end plates.5 The transverse (extension/flexion) and rotational mobility of the cervical spine is increased compared to adults. This makes the craniocervical junction and upper cervical spine very vulnerable for sudden acceleration and deceleration forces and trauma-related injuries. Most pediatric spinal traumas occur consequently in the cervical spine (80%).3,4,6,7 In children younger than 8 years primarily the first three cervical segments are involved.8 With progressive age the fulcrum of flexion gradually shifts caudally from C2/3 to C5/6. In older children and young adults, the lower cervical spine is consequently more frequently affected.3,9 At about 10 years of age the more typical adult distribution of injury is noted, affecting predominantly the cervicothoracic junction. Similar to adult patients, also in children the thoracic spine is less frequently affected due to the stabilizing effects of the adjacent rib cage. The lumbar spine is again more mobile. Another feature that increases the risks of spinal trauma in children is that young children typically have less welldeveloped protective reflexes if exposed to an approaching force.
◦ 2014 by the American Society of Neuroimaging C
337
Fig 1. Axial FLAIR of the brain and axial T2, sagittal T1 and T2 weighted MRI of a neonate who suffered from nonaccidental injury. A small FLAIR hyperintense subdural hematoma is noted along the right parietal hemisphere, partially extending into the interhemispheric fissure. In addition a co-existing moderate sized intradural, subarachnoid T1 hyperintense, T2-hypointense hematoma is noted within the dorsal, dependent lower part of the dural sac.
Fig 2. Sagittal CT (soft tissue algorithm) and sagittal T2-weighted MR images of two children who have suffered a craniocervical junction injury. On CT a moderately dense retroclival hematoma (arrow) is noted which displaces the adjacent lower brainstem. The apical ligaments between the dens/anterior arch of C1 and the clivus are intact and outlined by hypodense fat. No osseous lesion was noted. In the second child, the exquisite soft tissue resolution of MRI shows a disrupted (arrow head) and displaced T2-hypointense tectorial membrane which in combination with a retroclival hematoma displaces the lower brain stem. In addition, the apical ligament at the tip of the dens is also ruptured (reprinted with permission from Meoded A et al, AJNR Am J Neuroradiol 2011;32:1806-1811).
Translational forces at the craniocervical junction may result in devastating, frequently life-long lasting injuries to the lower brainstem and upper cervical spinal cord. Young children with traumatic brain injury (TBI) should be evaluated for concomitant injuries to the craniocervical junction. Because of the particular biomechanical properties of the pediatric head,
338
Journal of Neuroimaging Vol 25 No 3 May 2015
neck, and spine, cervical spine injury occurs particularly in TBI caused by abrupt acceleration and deceleration as in motor vehicle accidents.10–12 Next to the immediate injury to the spinal cord, injuries to the developing cartilaginous vertebral column and discoligamentary apparatus may also interfere with the ongoing development of the spinal column. Epiphyseal ring
Fig 3. (A) Lateral conventional radiography and sagittal CT of a healthy young child reveal the typical subtle ossifications along the apophyseal ring of the cervical vertebral bodies. These ossifications are physiological and should not be mistaken for posttraumatic lesions. The adjacent prevertebral soft tissues are also within normal limits with intact demarcation of the fat lines. (B) AP conventional radiography of the thoracic spine of an infant, and coronal 3D CT and axial CT image of 2 healthy neonates show the physiological synchondroses of the vertebral bodies which should not be mistaken for posttraumatic lesions. Familiarity with the normal progressing development of the vertebral bodies is essential. (C) Lateral conventional radiography of a healthy infant shows the typical wedge shaped configuration of the developing vertebral bodies as well as the physiological subluxation between C2 and C3 which should be less than 3 mm. In addition the atlantodens interval is wider compared to adults (up to 4.5 mm in children, compared to 3 mm in adults). The spinolaminar line or Swischuk’s line (dotted line) may be helpful to confirm the normal alignment of the vertebral bodies.
detachments, injuries to the ossification centers, or dislocations may result in spinal maldevelopment and long lasting spinal instability including scoliosis. Knowledge about the kind of trauma and risks of associated injuries outside of the central nervous system are a sine qua non for correct diagnosis. Frequently, trauma to the spinal column and cord occurs as part of a more extensive accident including TBI. In addition, simultaneous trauma to the chest and abdomen with possible cardiopulmonary or parenchymatous complications may aggravate the injury of the spinal cord. Hypoxia-ischemia, spinal cord swelling, epidural hematomas, release of excitatory neurotransmitters, and various additional complex inflammatory processes to mention a few, may result in significant secondary injuries.
Epidemiology Most pediatric spinal injuries occur in conjunction with motor vehicle accidents (52%), followed by sports-related injuries (27%), falls (15%), child abuse (3%), and various other less frequent accidents.13,14 In the vast majority of motor vehicle accidents, spinal injury occurs because the child was unrestrained or incorrectly restrained in the car. The cause of the trauma, however, changes depending on the age and physical activity of infants and children. In neonates, the spine and spinal cord may be injured secondary to a traumatic, forceful delivery.15 In young infants shaken baby injuries and nonaccidental injury may result in trauma of the craniocervical junction.16–19 In older children direct blows to the spine may occur after falls or aggressive sport activities.20–22 Later in life, children
Huisman et al: Pediatric Spinal Trauma
339
In addition, subdural and epidural blood may extend from the cranial vault into the spinal canal (Fig 1). Children with cervical spine injury are more likely to have additional neurological injury. The overall mortality rate of pediatric patients with cervical spine injuries ranges between 16% and 18%.13 The morbidity and mortality rate is especially high in children with an upper cervical spine injury or atlantooccipital dislocation.13
Imaging
Fig 4. Lateral conventional radiography of a child with a T10 segmentation/formation anomaly which may mimic an old trauma.
predominantly suffer spinal injuries from high speed motor vehicle accidents.20,21 Airbag-related injuries compromise a relatively new group of spinal injuries that depend on the size and position of the infant or child in relation to the rapidly inflating airbags in the case of a motor vehicle collision. The unique biomechanics of the pediatric developing and growing spine are responsible for different sports-related injuries. The rate and also often the severity of injuries increase with the child’s age and ability level. Injuries may occur while a child is skiing, cycling, diving, wrestling, performing gymnastics, or playing team sports like soccer, football, or ice hockey.23 Unfortunately spinal injury may also occur as consequence of child abuse either by shaking a young baby with resultant injuries to predominantly the craniocervical junction or due to forces exerted directly to the spine while the child is being beaten.
340
Journal of Neuroimaging Vol 25 No 3 May 2015
Goal of any diagnostic imaging should be to collect as much relevant and specific information about the location, degree, and quality of primary injury to start immediate treatment and to limit or prevent secondary injury. Imaging should be fast, readily available, not interfere with the emergent treatment, and should of course be highly sensitive and specific. The tool box of available imaging modalities is diverse and includes static and dynamic conventional radiography (CXR), computer tomography (CT), and magnetic resonance imaging (MRI). Depending on the patients acute physical status and the available equipment different diagnostic approaches may be chosen. CXR of the spine is still being used as a first line of imaging.7,24 However in many institutions, children who suffered from serious traumas typically first receive a CT study of the region of suspected injury. In particular, children who suffered from an unwitnessed traumatic brain and/or spine injury, unconscious children, rapidly deteriorating children, and children with focal neurological deficit receive primarily a head and spine CT.7 Multiplanar reconstructions of the spine show fractures and dislocations with high sensitivity and specificity.25 Next to bone algorithm reconstructions, soft tissue images are necessary to depict focal spinal cord injuries, epidural hematomas and adjacent soft tissue, and ligamentous injuries (Fig 2).9 In addition, occasionally vascular injuries may be noted. Finally, if additional injuries to the chest and abdomen are present or suspected, whole body CT imaging may be performed. MRI is typically performed after a CT study if a spinal cord lesion is suspected (focal neurological deficits), or if the CXR or CT imaging findings do not explain the clinical/neurological symptoms adequately.26,27 In addition, MRI is frequently used as follow-up imaging tool in order to reduce the radiation dose. Even in children who have stabilizing hardware in place, MRI may be considered. The hardware that is nowadays being used is usually nonferromagnetic and will typically only induce a focal signal loss in the immediate region of the hardware. The segments above and below the surgical site are usually free of significant artifacts. In addition, new developments and sequences have significantly reduced the degree of distortion in the MR images due to the orthopedic hardware (so called susceptibility artifacts). MRI typically includes multiplanar T1 and T2-weighted imaging as well as either short TI inversion recovery or fat saturated T2-weighted sequences.28 Intravenous contrast injection is rarely necessary. High resolution heavily T2-weighted sequences like the 3-dimensional (3D) constructive interference in steady state CISS sequence can be helpful to study the anatomy of the spinal cord and nerve roots as well as the spinous ligaments in exquisite anatomical detail. This sequence is however not sensitive for subtle intramedullary
Fig 5. Sagittal T2-weighted MR images of an infant who sustained a craniocervical acceleration-deceleration injury. The traction forces injured predominantly the cartilaginous and ligamentous components of the craniocervical junction with separation of the T2-hypointense ossification center of the tip of the dens from its base. T2-hyperintense fluid is noted between the separated fragments as well as between the widened joints between C1 and C2. In addition, the dorsal longitudinal ligament is partially torn. Finally, a focal T2-hyperintense focal contusion is noted within the upper cervical spinal cord. Conventional radiography and likely also CT are at risk to underestimate the full extent and degree of injury.
Fig 6. Sagittal T1, T2, T2*, and follow up T2-weighted MR images of a child who was the victim of a high speed motor vehicle accident. An ill-defined T1- and T2-hyperintense injury occurred to the lower brain stem and upper cervical spinal cord. Minimal T1-hyperintense blood clots are noted anterior to the spinal cord at level of C4/5. In addition a large T1-hypointense, T2-hyperintense fluid collection is noted in the prevertebral space which appears to communicate with the intradural cerebrospinal fluid (CSF) space surrounding the spinal cord. A T2-hypointense flow-related signal void is noted connecting both compartments (arrow). Finally the inferior endplate of C6 appears disrupted from C6 and is lining the intact intervertebral disk C6/7 which is attached to C7. This injury implies that the anterior and posterior longitudinal ligaments are torn and that a kind of Slater Harris type I fracture occurred. Follow-up imaging reveals a high grade atrophy of the cervical spinal cord, secondary deformity of the cervical spinal alignment, and a complete resolution of the prevertebral fluid collection.
signal alterations like, (eg, edema due to contusion or infarction). Ultrasound has only a very limited role in the diagnostic work-up of traumatic spine injuries. In the neonates, the spinal canal can be evaluated in between of the ossification centers of the vertebral bodies; however, spinal US is predominantly
used for the evaluation of spinal malformations rather than for posttraumatic lesions. Bone scintigraphy may be considered if a complicating osteomyelitis is suspected or in the case where hardware prevents a diagnostic MRI study of the region of interest.
Huisman et al: Pediatric Spinal Trauma
341
The Normal Developing Pediatric Spinal Column May Mimic Pediatric Spinal Injuries
Fig 7. Sagittal T2-weighted MR of the craniocervical junction/spine and axial FLAIR MR images of the brain in an adolescent who was injured in a high speed motor vehicle accident. Injury to the craniocervical junction (separation of the tectorial membrane from the adjacent clivus with retroclival hematoma) is often combined with diffuse injury to the brain. FLAIR imaging shows the characteristic pattern of diffuse axonal injury to the brain by shear forces. Whenever possible, in children the brain and craniocervical junction should be examined together.
CXR remains important in children suspected of nonaccidental injury (skeletal survey). Depending on the age of the child, various “typical” posttraumatic lesions can be seen.
The developing pediatric spinal skeleton may be challenging. The cartilaginous nature of the young skeleton, the presence of multiple ossification centers, and complex synchondroses (Fig 3A and B) as well as a high variability in the normal development may result in misdiagnosis.29 Synchondroses, not yet ossified parts of skeleton or residual cartilaginous components may be misinterpreted as fractures.13,23,30 The different shape of the young pediatric vertebral bodies with a physiological anterior wedging may be interpreted as a compression fracture (Fig 3C). In addition, the hypermobility of the spine with physiological subluxation of especially C2 relative to C3 may mimic traumatic injury/dislocation (Fig 3C).30 Finally, various metabolic disorders (eg, mucopolysaccharidoses) and skeletal dysplasias (eg, achondroplasia) or segmentation/formation anomalies of the vertebral column may result in a deformity of the osseous elements suggesting traumatic injuries (Fig 4).31–33 Also various connective tissue disorders as well as chromosomal anomalies (eg, trisomy 21) may result in an increased mobility of the spinal column.34 Frequently, widening of the prevertebral space, especially along the cervical spine is used as an indirect sign for an adjacent spinal injury.30 However, depending on the degree of inspiration the width of the prevertebral space may show a significant physiological variability in young children. Moreover, the adenoids and cervical lymphatics (eg, pharyngeal lymphoid ring of Waldeyer) are more prominent in children compared to adults. Finally, it should not be forgotten that traumatic injuries and fractures may be missed on CXR and CT because the injured components may not yet be ossified and consequently remain undetected on “bone imaging.” Each radiologist and physician who is interpreting pediatric spine studies has to be familiar with the normal, developing pediatric skeleton to prevent misdiagnosis.
Fig 8. Coronal 3D-MIP (maximum intensity projection) and matching coronal and two axial CISS images in a child with traumatic right brachial plexus injury. The nerve roots are well seen along the anterior spinal cord as T2-hypointense band outlined by T2-hyperintense CSF. At the level of the anterior nerve root injury, posttraumatic psuedomemningoceles are noted with an absent/avulsed anterior nerve root.
342
Journal of Neuroimaging Vol 25 No 3 May 2015
Fig 9. Sagittal T2-weighted and STIR MR images of a young adolescent who suffered from a nonwitnessed, apparently minor head and neck trauma who complained about tingling sensations in his upper extremities. Conventional radiography and CT imaging was unremarkable. MRI shows a focal T2-hyperintesne lesion within the cervical spinal cord (arrows) representing a small area of contusion/injury. This presentation is compatible with the diagnosis of SCIWORA. Follow-up imaging showed a discrete area of gliosis with near complete recovery of the symptoms.
Pediatric Spinal Injuries Craniocervical Junction and Cervical Spine A spectrum of posttraumatic lesions may be encountered at the craniocervical junction. Depending on the trauma mechanism translational, flexion/extension, distraction, and compression injuries or fractures may be noted. Most of the lesions are noted at the immediate craniocervical junction. In its mildest forms the ligaments are stretched or torn with or without dislocation.35–37 In more severe cases epiphyseal detachment or fractures may be seen (Figs 5 and 6). Most importantly, compression, contusion, or transsection of the adjacent brainstem and spinal cord may occur. Neurological deficit is typically acute and has a poor prognosis. In addition, co-existent intracranial lesions may be present and should be ruled out if there is a suggestive neurological presentation or trauma mechanism. In high speed motor vehicle accidents with acute acceleration deceleration forces diffuse axonal injury may be co-existent with ligamentous injuries at the craniocervical junction (Fig 7). In mild forms, the alignment is preserved but the ligaments are stretched or torn. CXR and CT may be unremarkable with exception of a straightening of the physiological cervical lordosis (guarding) and a mild paravertebral edema or an epidural retroclival or intraspinal hematoma. T2-weighted MRI may show an increased T2-signal of the injured ligaments, occasionally with T2-hyperintense edema of the injured ligaments. The high resolution 3D-T2-weighted CISS sequence may directly show the interruption/disruption of the injured ligaments (particularly the alar and apical ligaments as well as the tectorial membrane) and the individual nerve roots adjacent to the spinal
cord if injured (Fig 8). Retroclival hematomas may displace the tectorial membrane and extend into the epidural space of the upper spinal canal (Figs 2 and 7).12 T2-hyperintense edema may be seen along the epiphyseal plates as well as along the insertion of the ligaments within the subchondral bone. Adjacent soft tissue edema or hematomas may be seen. Vascular injuries like dissections should be ruled out. Due to the high mobility and flexibility of the pediatric spine, an intermittent subluxation/luxation may have occurred during the time of trauma which has spontaneously recovered on follow up.35,38 The CXR and CT findings are consequently underestimating the degree of soft tissue and ligamentous injury. MRI is however highly sensitive to show the resulting injuries that may affect the ligaments and possibly also spinal cord and brainstem. This category of injury is also overlapping with an entity known as SCIWORA (spinal cord injury without radiographic abnormality; Fig 9). Pang and Wilberger defined this entity as marked by objective signs of spinal cord injury, without evidence of ligamentous injury or fractures on plain films or tomographic studies.39 This entity has typically been described to affect children and is believed to affect the cervical spine most frequently. Flexion and extension but also lateral bending, distraction, rotation, axial loading, or combinations of these forces are the most common mechanism of injury. Again, MRI usually shows the trauma-related injuries that remained undetected on CXR or CT in better detail.40 In the more severe forms an axial dislocation of C1 in relation to the occiput (atlanto-occipital dislocation) may be encountered or between C1 and C2 (atlanto-axial dislocation; Fig 10A-C).41 The spinal canal is typically narrowed with
Huisman et al: Pediatric Spinal Trauma
343
Fig 10. (A) Sagittal T1 and T2-weighted MRI and matching sagittal CT images of a young child who had a combined head and neck trauma. MRI shows an increased distance between the tip of the dens and the adjacent clivus with blood and edema in between. The tectorial membrane is separated from the clivus. On CT the partially ossified dens and adjacent anterior arch of C1 are lowered compared to the skull base compatible with a posttraumatic C0/C1 dislocation. Moderate amount of fluid and blood is noted in the prevertebral and retropharyngeal space. Significant narrowing of the foramen of magnum is seen. (B) Sagittal, coronal, and 3D CT images of a young child with an asymmetrical dislocation of C1 in relation to C0/skull base. The right-sided ligamentous injury can only be identified indirectly on CT due to the dislocation. The 3D reconstruction shows the physiological nonfusion of the posterior arch with smooth osseous borders while the anterior arch fracture is irregular. (C) Sagittal T2- and T1-weighted initial and follow-up MR imaegs of the same child as shown in Fig 9(B). On MRI the full extent of soft tissue injury is seen in much better detail. A focal contusion of the lower brain stem and upper cervical spinal cord is identified as well as a significant ligamentous injury of the craniocervical junction. On follow-up imaging a significant spinal cord atrophy is noted. In addition, a significant deformity is noted of the lower cervical vertebral bodies which did not appear injured on the initial imaging. This deformity of noninjured vertebral elements is typically seen in young children and is believed to be secondary to the fact that an early trauma may result in a secondary, aberrant development of the spinal column. compression and injury to the spinal cord. Anterior dislocations are more frequently seen than posterior dislocations. The facet joints as well as the occipital condyles should be carefully evaluated for additional injuries and dislocations (Fig 11A-C). In the most severe forms a significant axial/vertical atlantooccipital dislocation, better known as dissociation may be
344
Journal of Neuroimaging Vol 25 No 3 May 2015
encountered (Fig 12).41 These injuries typically occur in high speed motor vehicle accidents and are characterized by a complete rupture of the ligaments between the occiput and C1/C2 with subsequent “separation” of the spinal column from the skull. The spinal cord is usually severely injured with poor prognosis. These injuries may also be seen in young children
Fig 11. (A) Sagittal, coronal, and axial CT images of a child with posttraumatic torticollis and neck pain due to a fracture of the left occipital condyle. The ossified tip of the dens is mildly dislocated in relation to the atlas which is best seen on the axial images. No additional fractures were noted. MRI showed a small epidural hemorrhage at the foramen of magnum without evidence of spinal cord injury (not shown). (B) Coronal, sagittal and axial images of a child with an odontoid fracture and C0/C1 dislocation. The tip of the dens is fractured at the level of the previous synchondrosis between tip and base of the dens. The foramen magnum is secondarily narrowed due to the anterior dislocation of the atlas. (C) Coronal, axial 2D, coronal 3D CT, and sagittal, axial T2-weighetd MR images of a child with a left lateral odontoid tip fracture and an associated rotational subluxation of C1 in relation to C2. MRI excluded a spinal cord injury, or spinal canal stenosis. Mild amount of T2-hyperintense fluid is seen between the anterior dens and adjacent anterior arch of the atlas. 3D CT is very helpful to evaluate the degree of subluxation.
Huisman et al: Pediatric Spinal Trauma
345
Fig 12. Lateral CT localizer, soft tissue and bone algorithm sagittal and coronal CT images of the cervical spine of an adolescent boy who sustained severe craniocervical junction dissociation after a high speed motor vehicle accident. He was ejected from the car and resuscitated at the scene. A significant vertical and anterior dissociation of the craniocervical junction is noted. High-grade narrowing of the spinal canal as well as a large prevertebral and moderate sized anterior epidural hematoma is seen. Subarachnoid blood is seen outlining the brainstem. The child subsequently died.
Fig 13. Conventional lateral radiography, sagittal bone and soft tissue algorithm, coronal bone algorithm CT images of an adolescent with a type II odontoid fracture. Mild dorsal dislocation of the dens in relation to the base of C2, the craniocervical alignment is intact.
who were struck by a rapidly deploying airbag while seated in a forward facing car seat. Large pre- and paravertebral hematomas occur. The spectrum of cervical fractures in children are similar to the “adult” fractures, however the incidence, distribution, and location differs (Figs 11B, C and 13).30,35 Atlas fractures like the Jefferson fracture is typically seen as a result of an axial compression after a fall onto the head (diving accidents) while anterior and posterior arch fractures of the atlas typically result from a focal C1/2 hyperextension. Fractures of the axis include the Anderson fracture which typically occurs as complication of a traumatic hyperflexion and the Hangman fracture which in turn is seen after a traumatic hyperextension. Compression fractures (C3-C7) typically result from traumatic hyperflexion and present with an increased anterior wedging
346
Journal of Neuroimaging Vol 25 No 3 May 2015
of the cervical vertebral bodies. CXR and especially multiplanar coronal and sagittal CT reconstructions usually easily identify the kind and degree of injury/fracture. MRI may help to characterize and date the fracture. Occasionally, it is difficult to differentiate between an anterior wedging of the vertebral bodies as a normal developmental variant versus secondary to a pre-existent systemic disorder versus a posttraumatic anterior reduction in height. T2-hyperintense bone marrow edemas as well as paravertebral soft tissue edema will suggest acute injury. With progressing age, the middle and lower cervical spine will be more frequently affected. The region of maximal mobility migrates from the C2/C3 region towards the adult region at the level of C5/C7/T1. Consequently, more frequently posttraumatic lesions resulting from hyperextension/hyperflexion
Fig 14. Conventional lateral radiography and coronal and sagittal CT images of an adolescent with classical, stable L1 anterior compression fracture with wedge shaped vertebral body. The dorsal alignment is intact, no spinal canal stenosis. will be seen in the middle and lower cervical spine at older age.
Thoracic and Lumbar Spine With progressing static and dynamic stability, the predominant location of spinal injuries migrates toward the thoracic and lumbar spine (children >10 years).42,43 Fractures are typically seen at the thoracolumbar junction and in the region of the lumbar spine.44 Fractures occur less frequently in the thoracic spine because of the stabilizing effects of the rib cage. Thoracic and lumbar fractures include lateral shear-translation fractures, compression fractures (typically from falls), burst fractures, Chance fracture (flexion-distraction injury), and focal, direct impact fractures. Compression fractures are characterized by a wedge shaped deformity of the involved vertebral body with interruption/fracture of the anterior vertebral contour (Fig 14). Compression fractures are typically stable if they involve only the anterior column. Burst fractures occur as complication of an axial force resulting in a fracture of both the anterior and posterior contour of the involved vertebral body. Bony fragments may be dislocated into the spinal canal, compressing the adjacent neuronal structures (Fig 15A and B). Additional dislocations may be seen in more severe injuries. These fractures result from various combinations of flexion and axial compression. Lateral forces may also result in compression or burst fractures. In this kind of trauma, however, additional lateral dislocations can be noted (Figs 16 and 17). Diffusion weighted and diffusion tensor imaging (DTI) of the spinal cord may be helpful to evaluate the degree of spinal cord injury and integrity of fiber tracts. However, the available literature on
the application of DTI to study children with traumatic spinal cord injury is scant.45,46 In children with spinal cord injury, fractional anisotropy (FA) values are reduced and diffusivity values are increased compared to controls.45 In addition, DTI metrics correlate with the neurological function and level of injury. These preliminary results suggest DTI metrics (particularly FA) as a potential neuroimaging biomarker of neurological outcome in children after traumatic spinal cord injury. The Chance fractures, named after the British radiologist G.C. Chance who described this group of fractures first in 1948 for the lumbar region, is characterized by a transverse or oblique fracture that involves all three longitudinal vertebral columns (Figs 18 and 19).47,48 This fracture results from a combined flexiondistraction mechanism around a fulcrum, most commonly a seat belt. The Chance fracture is consequently also known as seat belt or lap belt fracture. The anterior vertebral body is typically compressed while the posterior vertebral body height is increased by the distraction component of the injury. The posterior extension of the distracting forces distracts the posterior elements with increased interspinous distance and widened facet joints. Concomitant ligamentous injury, eg, rupture of the posterior longitudinal ligament and interspinous ligament occurs in variable degrees and determines stability. Anterolisthesis at the fracture level may result in significant compression of the spinal cord or cauda equina. Most injuries of the lower back are at the thoracolumbar junction because of its relative high mobility. Children with seat belt injuries frequently have additional internal injuries because of a somewhat less stable rib cage and pelvis compared to adults (Fig 20). Imaging should consequently also include evaluation of the thoracic and abdominal organs and vasculature.
Huisman et al: Pediatric Spinal Trauma
347
Fig 15. (A) Sagittal and coronal CT and sagittal T2-weighted MR images of a child with multilevel vertebral body fractures. Significant compression fracture with retropulsion of the fracture fragments and spinal canal stenosis is seen at the level of L1. An additional superior endplate impression fracture is noted of L4. MRI shows additional subchondral posttraumatic edema along the superior endplate of L3 indicating additional injury. MRI also shows significant compression of the conus medullaris with intramedullary edema. (B) Conventional lateral radiography, sagittal, and coronal CT images show a similar L1 compression/burst fracture with retropulsion of the fracture fragments into the spinal canal as in the child shown in Fig 14(A). An additional superior endplate fracture affecting T11 is noted. This case illustrates that the individual vertebral segments may be skipped after trauma.
Various additional fractures involving the neural arch and posterior or lateral elements of the spine may be encountered as complication of direct impact.
Nonaccidental Injury The spinal column may also be involved in nonaccidental injury or child abuse.17,49 The literature about spinal injuries in nonaccidental trauma in scant; however, the presence of spinal injury represents a severe form of child abuse with a high risk of longterm neurological impairment.17 Depending on the used force and mechanism of trauma nearly any kind of spinal injury may result. Cervical spinal injuries are more common in the younger
348
Journal of Neuroimaging Vol 25 No 3 May 2015
infants (median age 5 months) who present with impaired consciousness and respiratory distress, while thoracolumbar injury are seen in older infants (median age 13.5 months) with visible spinal deformity or focal neurological signs.17 However, there are no pathognomonic injury patterns to confirm the nonaccidental nature of the trauma and a complete diagnostic work-up with skeletal survey, physical examination, and a psychosocial evaluation should be performed.50 Some lesions may suggest nonaccidental injury like isolated fractures of the spinous processes (resulting from direct blows to the back of the child), especially when no adequate or matching trauma history is given by the caregivers. In addition, Choudhary et al showed that cervical spine ligamentous injuries (predominantly involving the
Fig 16. (A) Coronal 2D and 3D CT images of a 17-year-old male who fell 12 feet off a roof. In addition to multilevel compression fractures a significant lateral dislocation is noted with partial overriding of the vertebral bodies. Despite the significant dislocation no serious neurological deficits were noted on clinical examination. (B) Sagittal STIR, T1- and curved reformatted T2-weighted MR images confirm the multilevel compression fractures with significant T2-hyperintense bone marrow edema. The spinal canal is mildly narrowed; the spinal cord appears intact without significant edema.
nuchal, atlanto-occipital, and atlanto-axial ligaments) are significantly more common in children with nonaccidental trauma compared to children with accidental trauma.49 Moreover, in children with nonaccidental trauma the presence of injury of the cervical ligaments correlates with the evidence of hypoxicischemic brain injury.49 Finally, spinal subdural hemorrhages are significantly more common in children with nonaccidental trauma compared to children with accidental trauma.18
Birth-Related Injury Traumatic vaginal delivery may result in injuries of the craniocervical junction and cervicothoracic junction.15 Next to liga-
mentous and cartilago-osseous lesions, the spinal cord may be injured ranging from smaller focal hemorrhages up to complete transsections. In addition, nerve root avulsions and brachial plexus injuries may be seen.
Additional Risk Factors Preexisting spinal pathologies including spinal anomalies with defective or incomplete development of the neural elements, segmentation, or formation anomalies of the osseous spine, spondylolysis as well as various systemic diseases like metabolic disorders, bony dysplasia, or infectious diseases including spondylitis and retropharyngeal abscesses may enhance the risk
Huisman et al: Pediatric Spinal Trauma
349
Fig 17. (A) AP conventional radiography, coronal and sagittal CT images of a child with a lateral/superiorly dislocated L1 fracture similar to the patient in Figure 15. This child also had very minimal neurological deficits despite the significant mal-alignment. (B) Matching sagittal and coronal T2- and diffusion weighted MR images show a similar significant dislocated L1 fracture with lateral and anterior dislocation of the distal spine. The diffusion weighted imaging is unremarkable without focal spinal cord lesions supporting the minimal clinical symptoms.
350
Journal of Neuroimaging Vol 25 No 3 May 2015
Fig 18. (A) Axial and sagittal CT images of a child with a classical Chance fracture with a compression fracture of the anterior L3 vertebral body and distraction of the posterior elements resulting in an increased posterior interspinous distance and widened facet joints. (B) Axial abdominal contrast enhanced CT of the same child shows a bowel wall hematoma and free fluid/blood within the peritoneal cavity/small pelvis related to the seat belt injury.
Fig 19. (A) Coronal soft window and coronal and sagittal bone window algorithm CT images of a young child with a classical Chance fracture affecting T8 (arrow). Moderate amount of blood/free fluid is noted within the abdomen. (B) Axial contrast enhanced CT in soft tissue and lung window settings show a laceration of the anterior stomach with air escaping into the peritoneal cavity (free air) partially outlining the falciform ligament (arrow). The stomach was injured due to the seat belt mechanism of injury.
for and severity of spinal injury. Finally, various chromosomal. abnormalities like Down syndrome increase the risk for injuries due to the increased ligamentous laxity.
Summary The pediatric spine is very different compared to the adult spine; incidence, epidemiology, distribution, and character of
Huisman et al: Pediatric Spinal Trauma
351
Fig 20. Axial contrast enhanced CT of a child with posttraumatic pancreatic fracture. The fractured area is hypo-enhancing (arrow). In addition, contusion/laceration of the adjacent spleen is noted. spinal injuries are unique for the age of the patient. Imaging should evaluate the ligamentous and osteo-cartilaginous elements of the spine as well as the spinal cord including nerve roots and paraspinal plexuses. CXR and CT predominantly study the osseous complications, MRI the soft tissue injuries of spinal trauma.
References 1. Tortori-Donati P, Rossi A, Calderone M, et al. Spinal trauma. In: Tortori-Donati P, ed. Pediatric Neuroradiology Head, Neck and Spine. Heidelberg: Springer, 2005:1683-1704. 2. Hollingshead MC, Castillo M. Trauma to the spinal column. In: Naidich TP, Castillo M, Cha S, Raybaud C, Smirniotopoulos J, Kollias S, Kleinman GM, eds. Imaging of the Spine. Philadelphia: Saunders, Elsevier, 2011:219-236. 3. Hadley MN, Zabramski JM, Browner CM, et al. Pediatric spinal trauma. Review of 122 cases of spinal cord and vertebral column injuries. J Neurosurg 1988;68:18-24. 4. Dickman CA, Rekate HL, Sonntag VK, et al. Pediatric spinal trauma: vertebral column and spinal cord injuries in children. Pediatr Neurosci 1989;15:237-255. 5. Vanderhave KL, Chiravuri S, Caird MS, et al. Cervical spine trauma in children and adults: perioperative considerations. J Am Acad Orthop Surg 2011;19:319-327. 6. McCall T, Fassett D, Brockmeyer D. Cervical spine trauma in children: a review. Neurosurg Focus 2006;20:E5. 7. Viccellio P, Simon H, Pressman BD, et al. A prospective multicenter study of cervical spine injury in children. Pediatrics 2001;108:E20. 8. Ruge JR, Sinson GP, McLone DG, et al. Pediatric spinal injury: the very young. J Neurosurg 1988;68:25-30. 9. Hamilton MG, Myles ST. Pediatric spinal injury: review of 174 hospital admissions. J Neurosurg 1992;77:700-704. 10. Sun PP, Poffenbarger GJ, Durham S, et al. Spectrum of occipitoatlantoaxial injury in young children. J Neurosurg 2000;93:28-39. 11. Kwon TH, Joy H, Park YK, et al. Traumatic retroclival epidural hematoma in a child: case report. Neurol Med Chir (Tokyo) 2008;48:347-350. 12. Meoded A, Singhi S, Poretti A, et al. Tectorial membrane injury: frequently overlooked in pediatric traumatic head injury. AJNR Am J Neuroradiol 2011;32:1806-1811. 13. Jones TM, Anderson PA, Noonan KJ. Pediatric cervical spine trauma. J Am Acad Orthop Surg 2011;19:600-611.
352
Journal of Neuroimaging Vol 25 No 3 May 2015
14. Basu S. Spinal injuries in children. Front Neurol 2012;3:96. 15. Huisman TA, Phelps T, Bosemani T, et al. Parturitional injury of the head and neck. J Neuroimaging 2014;epub July 5. 16. Brennan LK, Rubin D, Christian CW, et al. Neck injuries in young pediatric homicide victims. J Neurosurg Pediatr 2009;3:232-239. 17. Kemp AM, Joshi AH, Mann M, et al. What are the clinical and radiological characteristics of spinal injuries from physical abuse: a systematic review. Arch Dis Child 2010;95:355-360. 18. Choudhary AK, Bradford RK, Dias MS, et al. Spinal subdural hemorrhage in abusive head trauma: a retrospective study. Radiology 2012;262:216-223. 19. Knox J, Schneider J, Wimberly RL, et al. Characteristics of spinal injuries secondary to nonaccidental trauma. J Pediatr Orthop 2014;34:376-381. 20. Akbarnia BA. Pediatric spine fractures. Orthop Clin North Am 1999;30:521-536. 21. Vialle LR, Vialle E. Pediatric spine injuries. Injury 2005;36(Suppl 2):B104-112. 22. Stracciolini A, Casciano R, Levey Friedman H, et al. Pediatric sports injuries: an age comparison of children versus adolescents. Am J Sports Med 2013;41:1922-1929. 23. Maxfield BA. Sports-related injury of the pediatric spine. Radiol Clin North Am 2010;48:1237-1248. 24. Hoffman JR, Mower WR, Wolfson AB, et al. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. National Emergency X-Radiography Utilization Study Group. N Engl J Med 2000;343:94-99. 25. Gargas J, Yaszay B, Kruk P, et al. An analysis of cervical spine magnetic resonance imaging findings after normal computed tomographic imaging findings in pediatric trauma patients: ten-year experience of a level I pediatric trauma center. J Trauma Acute Care Surg 2013;74:1102-1107. 26. Flynn JM, Closkey RF, Mahboubi S, et al. Role of magnetic resonance imaging in the assessment of pediatric cervical spine injuries. J Pediatr Orthop 2002;22:573-577. 27. Frank JB, Lim CK, Flynn JM, et al. The efficacy of magnetic resonance imaging in pediatric cervical spine clearance. Spine (Phila Pa 1976) 2002;27:1176-1179. 28. Henry M, Scarlata K, Riesenburger RI, et al. Utility of STIR MRI in pediatric cervical spine clearance after trauma. J Neurosurg Pediatr 2013;12:30-36. 29. Kim HJ. Cervical spine anomalies in children and adolescents. Curr Opin Pediatr 2013;25:72-77. 30. Lustrin ES, Karakas SP, Ortiz AO, et al. Pediatric cervical spine: normal anatomy, variants, and trauma. Radiographics 2003;23:539560. 31. Lachman RS. The cervical spine in the skeletal dysplasias and associated disorders. Pediatr Radiol 1997;27:402-408. 32. McMaster MJ, Singh H. Natural history of congenital kyphosis and kyphoscoliosis. A study of one hundred and twelve patients. J Bone Joint Surg Am 1999;81:1367-1383. 33. Tortori-Donati P, Fondelli MP, Rossi A, et al. Segmental spinal dysgenesis: neuroradiologic findings with clinical and embryologic correlation. AJNR Am J Neuroradiol 1999;20:445456. 34. Herman MJ, Pizzutillo PD. Cervical spine disorders in children. Orthop Clin North Am 1999;30:457-466. 35. Roche C, Carty H. Spinal trauma in children. Pediatr Radiol 2001;31:677-700. 36. Caird MS, Hensinger RN, Vander Have KL, et al. Isolated alar ligament disruption in children and adolescents as a cause of persistent torticollis and neck pain after injury. A report of three cases. J Bone Joint Surg Am 2009;91:2713-2718. 37. Kvien TK, Zeidler HK, Hannonen P, et al. Long term efficacy and safety of cyclosporin versus parenteral gold in early rheumatoid arthritis: a three year study of radiographic progression, renal function, and arterial hypertension. Ann Rheum Dis 2002;61:511-516.
38. Muniz AE, Belfer RA. Atlantoaxial rotary subluxation in children. Pediatr Emerg Care 1999;15:25-29. 39. Pang D, Wilberger JE Jr. Spinal cord injury without radiographic abnormalities in children. J Neurosurg 1982;57:114-129. 40. Mahajan P, Jaffe DM, Olsen CS, et al. Spinal cord injury without radiologic abnormality in children imaged with magnetic resonance imaging. J Trauma Acute Care Surg 2013;75:843-847. 41. Martinez-Lage JF, Alarcon F, Alfaro R, et al. Severe spinal cord injury in craniocervical dislocation. Case-based update. Childs Nerv Syst 2013;29:187-194. 42. Daniels AH, Sobel AD, Eberson CP. Pediatric thoracolumbar spine trauma. J Am Acad Orthop Surg 2013;21:707-716. 43. Sayama C, Chen T, Trost G, et al. A review of pediatric lumbar spine trauma. Neurosurg Focus 2014;37:E6. 44. Bollmann C, Fernandez FF, Eberhardt O, et al. Comparison of the diagnostic value of X-ray versus MRI in paediatric spine injuries. Z Orthop Unfall 2011;149:77-82.
45. Mulcahey MJ, Samdani AF, Gaughan JP, et al. Diagnostic accuracy of diffusion tensor imaging for pediatric cervical spinal cord injury. Spinal Cord 2013;51:532-537. 46. Middleton DM, Mohamed FB, Barakat N, et al. An investigation of motion correction algorithms for pediatric spinal cord DTI in healthy subjects and patients with spinal cord injury. Magn Reson Imaging 2014;32:433-439. 47. Chance GQ. Note on a type of flexion fracture of the spine. Br J Radiol 1948;21:452. 48. Davis JM, Beall DP, Lastine C, et al. Chance fracture of the upper thoracic spine. AJR Am J Roentgenol 2004;183:1475-1478. 49. Choudhary AK, Ishak R, Zacharia TT, et al. Imaging of spinal injury in abusive head trauma: a retrospective study. Pediatr Radiol 2014;44:1130-1140. 50. Kadom N, Khademian Z, Vezina G, et al. Usefulness of MRI detection of cervical spine and brain injuries in the evaluation of abusive head trauma. Pediatr Radiol 2014;44:839-848.
Huisman et al: Pediatric Spinal Trauma
353
