INTRODUCTION AND EPIDEMIOLOGY
Cervical spine injuries occur in 1% to 2% of all pediatric trauma patients. While the incidence of cervical spine injuries in children with trauma is lower than adults, children have higher rates of mortality (~18%) compared to adults (~10%).1 In children <8 years old, almost three quarters of all spinal injuries occur in the cervical spine,2 and nearly two thirds of these children have associated neurologic deficits and head or other major organ injury.3 In addition, spinal cord injury without radiographic abnormality (SCIWORA) may occur in children and typically involves the cervical spine. The incidence of SCIWORA among pediatric trauma patients ranges from 0.15% to 0.2%, compromising 4.5% to 35% of pediatric spine injuries.4,5,6 Motor vehicle crashes are the most common mechanism of cervical spine injuries, followed by falls, and in teenagers, diving and sports injuries. Boys are affected more often than girls. Child abuse can result in cervical spine injuries in younger patients via a shaking mechanism, although this is a rare manifestation of nonaccidental trauma.7
A number of anatomic differences between the pediatric and adult cervical spine predispose children to different patterns of injury (Table 139-1). In particular, the relatively larger head-to-body ratio in young children creates a fulcrum at C2-C3 (compared to C5-C6 in adults) that accounts for higher rates of cervical spine injury above C3 in children. Weaker muscles and ligaments combined with anterior wedging and shallow facets connecting cervical vertebrae and immature growth centers together allow for easier anterior-posterior slipping of the vertebrae than in adults.
TABLE 139-1Anatomic Considerations in the Pediatric Cervical Spine ||Download (.pdf) TABLE 139-1 Anatomic Considerations in the Pediatric Cervical Spine
Absent cervical lordosis
Weaker neck muscles
Anterior wedging of vertebrae
Shallow and horizontal vertebral facets
Ossification centers and synchondroses
Patients younger than 8 years of age incur high ligamentous injuries more often than older children and adults. Fractures tend to occur at the weak points in the bones—synchondroses and ossification centers. Dens fractures occur most commonly along the synchondrosis, especially in children younger than age 7 years. The mechanism of injury is usually a forward facing child in a high-speed motor vehicle crash with rapid forward flexion. Atlanto-occipital and atlantoaxial dislocation injuries are devastating vertical distraction injuries that occur in the very young child, most commonly from a motor vehicle crash, and usually result in rapid death (Figure 139-1).
Atlantoaxial dislocation in a 6-year-old boy involved in a motor vehicle crash. A. Lateral plain radiograph reveals atlantoaxial dislocation (blue arrow). B. MRI of the same patient reveals a near-complete transection of the brain stem at the level of the distal medulla, extensive ligamentous injury with resulting atlantoaxial dissociation, extensive intrathecal hematoma and hemorrhage, C1-C2 interspinous ligament tear, and prevertebral soft tissue swelling and edema around the nuchal ligament (blue arrows).
IMMOBILIZATION AND NECK STABILIZATION IN INFANTS AND CHILDREN
Proper immobilization of the cervical spine is of primary concern in the ED. Immobilization may be difficult in young children who are frightened or agitated, and placement of a cervical collar in children <7 years old while on a flat surface (e.g., spine board) may cause unwanted flexion of the cervical spine with further harm. A number of untoward effects have been associated with immobilization on a long backboard and cervical spine immobilization, including development of decubitus ulcers, flexion of the neck causing respiratory compromise, worsening of atlanto-occipital distraction injury, increased intracranial pressure, and musculoskeletal pain that may mimic injury and lead to increased radiologic investigation.8-20
For those at significant risk for cervical spine injury, neutral positioning of the neck is important; consider elevating the torso 2.5 cm (or more for children <4 years of age) from the spine board in order to alleviate neck flexion caused by the large occiput, and place padding beneath the child. Neutral position is achieved by aligning the external auditory meatus with the shoulders. Proper sizing of pediatric cervical collars is equally important and varies depending on the device used. If the proper-size collar is not available, use towel rolls or foam blocks placed on both sides of the child's head and secured to the backboard with tape across the forehead.
Ask parents, witnesses, or prehospital personnel about the mechanism of injury: children with cervical spine injury will usually have a history of high-force acceleration/deceleration (as seen in motor vehicle crashes) or axial loading trauma (falls, diving injuries). Trivial mechanisms such as a ground-level fall usually do not lead to serious spine injury unless the patient has a condition associated with cervical spine instability (see "Special Considerations" later in the chapter). In the cooperative, verbal child, ask about symptoms such as neck pain, sensory deficits, or weakness. One large multicenter case-control study looking at 521 children with blunt trauma and cervical spine injury identified eight factors associated with cervical spine injury after blunt trauma (Table 139-2).21 Absence of these eight factors had a 98% sensitivity and 26% specificity for cervical spine injury. Prospective studies for validation of these risk factors are forthcoming.
TABLE 139-2Risk Factors for Pediatric Cervical Spine Injury ||Download (.pdf) TABLE 139-2 Risk Factors for Pediatric Cervical Spine Injury
Altered mental status
Focal neurologic examination
High-risk motor vehicle crash
Substantial torso injury
Predisposing condition associated with cervical spine injury
Examine the child with a primary survey of airway, breathing, and circulation, followed by a complete head-to-toe examination to identify major and potentially distracting injuries that may complicate the specific evaluation of the cervical spine.
Pay careful attention to breathing, because damage to C3-C5 can injure the phrenic nerve, impairing innervation to the diaphragm, compromising respiratory mechanics, and leading to apnea. Injury to the high cervical spine can affect hemodynamic stability from spinal shock. Intubated or obtunded trauma patients should have a cervical collar remain in place, and imaging should be performed.
In children who are cooperative and alert, assess for midline neck tenderness, the presence of torticollis, and neurologic deficits while maintaining inline stabilization of the neck. Sensory symptoms such as numbness or tingling are the most common neurologic deficits among pediatric cervical spine injury patients, and persistent sensory deficits may help localize the level of the injury.
Test for motor function in the cooperative child: shoulder shrug is controlled by C5, elbow flexion and wrist extension by C6, elbow extension and wrist flexion by C7, and finger flexion by C8. Also check deep tendon reflexes: the biceps reflex tests C5, the brachioradialis reflex tests C6, and the triceps reflex tests C7.
Distinguishing between bony cervical spine injury, SCIWORA, and peripheral nerve injury (brachial plexus) from sports-related accidents can be challenging. Transient burning sensation of the hands and fingers has been described with hyperextension injuries among young football players and may indicate central cord contusion. Muscular torticollis causes neck pain in children but typically lacks a history of trauma. Infectious causes of neck pain in children include cervical adenitis and deep space infections such as retropharyngeal and peritonsillar abscess. These conditions are associated with fever and toxic appearance and do not include a history of injury.
Routine laboratory testing is not helpful in the general evaluation of cervical spine injury in children but may be useful in the context of multisystem trauma.
Because the history and physical examination can be difficult, particularly in young children or those with multiple injuries, imaging is often considered, and clinical decision rules have been developed to help identify patients who require radiologic investigation.
Adult decision rules such as NEXUS and the Canadian Cervical Spine Rule (CCR) (see Tables 258-4 and 258-5) for cervical spine clearance can safely be applied to children over 8 years old, the same age that patient anatomy starts to resemble that of an adult. NEXUS should be used with caution in children under 9 years old, because the cohort examined in the study of NEXUS included no patients with cervical spine injury under the age of 2 years and only four patients under the age of 9 years.22 A prospective study applying NEXUS in children showed that NEXUS did in fact miss two fractures, both in children under 2 years of age, highlighting the difficulty of ruling out cervical spine injury in the very young patient.23 Additional clinical algorithms have been published by the Trauma Association of Canada's Pediatric Subcommittee (see Figures 139-4 and 139-5 later).
Plain Radiographs For patients in whom a cervical spine injury is suspected, plain radiographs should be considered first. The sensitivity of plain radiographs for identifying fractures, dislocations, and subluxations in children varies across the literature, ranging from 74% to 98%, with sensitivity increasing with the number of views taken.24,25,26,27,28,29,30 Combining the cross-table lateral and anterior-posterior views identifies 87% of significant cervical spine injury in children younger than 8 years of age.30 Consider the addition of an odontoid (open mouth) view in cooperative older children, although the added value in children younger than age 8 years has been questioned.30
Interpreting pediatric cervical spine radiographs can be challenging as a result of the anatomic differences across the spectrum of age. The principles of evaluation are similar to adults: assess the anterior and posterior vertebral lines, the spinolaminal line, and the spinous processes for alignment; and carefully examine the soft tissue spaces for swelling that might indicate subtle fracture. Widening of the soft tissue space >7 mm in the retropharyngeal space and >14 mm in the retrotracheal space on lateral x-ray may be an indirect sign of cervical spine injury. In addition, assess the pediatric cervical spine for atlantoaxial instability. Useful measurements include Wackenheim's clivus line (a line along the posterior edge of the clivus should intersect the odontoid) and the rule of thirds (the dens, cord, and empty space should each occupy one third of the spinal space). Craniocervical dislocation is suggested by a C1-C2:C2-C3 ratio >2.5. Important differences between pediatric and adult cervical spine radiographs include:
Normal cervical lordosis may be absent in children.
The posterior arch of C1 fuses by age 3 years and the anterior arch by age 10 years.
Anterior wedging of the vertebrae caused by secondary growth centers may be mistaken for compression fractures in children under 7 years old.
Posterior laminar fusion lines may be mistaken for fractures in children under 7 years old.
Children younger than 8 years old may demonstrate pseudosubluxation (up to 46%) on lateral x-ray, usually at the C2-C3 level (Figure 139-2).31
Pseudosubluxation of C2 on C3. [Reprinted, with permission, from Yamamoto LG: Cervical spine malalignment—true or pseudosubluxation? In: Yamamoto LG, Inaba AS, DiMauro R (eds): Radiology Cases in Pediatric Emergency Medicine, Vol. 1, Case 5. Honolulu, HI: University of Hawaii John A. Burns School of Medicine, Department of Pediatrics, 1994. http://www.hawaii.edu/medicine/pediatrics/pemxray/v1c05.html.]
Swischuk developed a method for distinguishing between true subluxation and pseudosubluxation: draw a line connecting the posterior cortex of the spinous process of C1 to the cortex of the spinous process of C3 (Figure 139-3). If this line is more than 2 mm anterior to the spinous process of C2, suspect cervical pathology, such as a hangman's fracture. There is little role for flexion/extension plain x-rays in the ED for children. Obtain an MRI if ligamentous instability is suspected.
Posterior cervical line of Swischuk. [Reprinted, with permission, from Yamamoto L: Cervical spine malalignment—true or pseudosubluxation? In: Yamamoto LG, Inaba AS, DiMauro R (eds): Radiology Cases in Pediatric Emergency Medicine, Vol. 1, Case 5. Honolulu, HI: University of Hawaii John A. Burns School of Medicine, Department of Pediatrics, 1994. http://www.hawaii.edu/medicine/pediatrics/pemxray/v1c05.html.]
CT Scan Although CT has been advocated as the most cost-effective modality for imaging the cervical spine in adults, it is unclear if this is true for children, in whom radiation exposure is of particular concern. CT scanning exposes a child to 10 to 90 times the radiation of plain films, with an estimated 18-fold increase in thyroid malignancy in a theoretical model if all plain films are replaced by CT scans.32,33 Given the potential risks and unclear advantages of CT in children, this modality should be reserved for children with inadequate visualization of the cervical spine on plain radiographs, those with abnormal or suspicious plain radiographs, and those with significant mechanism of injury who are obtunded or intubated and undergoing CT evaluation for other injuries. Some experts advocate scanning up to C3 in young children and stopping there if no injury is detected, avoiding irradiation to radiosensitive thyroid tissues.
MRI MRI is the modality of choice for the diagnosis and evaluation of SCIWORA and can be helpful in determining prognosis as well. Patients with normal MRI examinations and SCIWORA have very favorable prognoses; those with minor findings such as cord edema or a minor hemorrhage have good prognoses, whereas those with major hemorrhage or cord transections have poor prognoses for recovery.34 MRI should be reserved for children who have an obvious neurologic deficit, those with a persistently unreliable clinical examination, the very young with a dangerous mechanism of injury, or those with persistent symptoms and negative plain films and CT scan.4
The ED management of the child with potential cervical spine injury should follow the principles of advanced trauma life support with primary attention to assessment and management of the airway and breathing (especially with a high cervical spine injury, which may compromise respiratory effort) and circulation (particularly in spinal shock).
In adult cervical spine injury, steroids are no longer favored (see chapter 258, "Spine Trauma"), and no large studies have examined the efficacy of steroids in pediatric spine injury, so steroids are not standard of care for children.
Definitive management of cervical spine injury is primarily surgical, through consultation with pediatric neurosurgery or orthopedics.
DISPOSITION AND FOLLOW-UP
Unstable patients and those with radiographic evidence of cervical spine injury require admission, usually to intensive care; a minority require immediate surgical intervention. Children with normal plain radiographs or CT imaging who have persistent neurologic complaints need MRI evaluation. Children with persistent neck pain but without focal neurologic signs or symptoms and normal x-rays or CT imaging will usually be evaluated with an MRI. If imaging is normal and the mechanism of injury is mild, the child may be discharged with an appropriately sized cervical collar and close outpatient medical follow-up. Children without neck pain or neurologic symptoms or deficits and those who have been cleared radiographically may be discharged with instructions to return for symptoms of numbness, tingling, or weakness in the arms, hands, legs, or feet.
THORACIC, LUMBAR AND SACRAL SPINE INJURIES
Injuries to the spine and spinal cord outside of the cervical spine are discussed in chapter 110, "Pediatric Trauma."
Certain conditions predispose patients to cervical spine injury as a result of associated abnormalities causing cervical spine instability. These are listed in Table 139-3. Children with these conditions should be considered at high risk for cervical spine injury and undergo conservative evaluation.
TABLE 139-3Predispositions to Cervical Spine Injury in Children ||Download (.pdf) TABLE 139-3 Predispositions to Cervical Spine Injury in Children
Down syndrome: 15% atlantoaxial instability
Connective tissue disorders (e.g., Marfan's syndrome and Ehlers-Danlos syndrome): ligamentous laxity
Klippel-Feil syndrome: cervical vertebral defects
Morquio's syndrome (type IV mucopolysaccharidosis): odontoid hypoplasia
Previous cervical spine surgeries or arthritis
In 2011, the Trauma Association of Canada's Pediatric Subcommittee National Pediatric Cervical Spine Evaluation Pathway released a consensus guideline for evaluation of the pediatric cervical spine summarized in Figures 139-4, 139-5, and 139-6.35 They divided their algorithms into two parts: the reliable patient (Figure 139-4) and the unreliable patient (Figure 139-5). These algorithms have not been prospectively studied, but represent a reasonable, expert-driven tool to evaluate the pediatric cervical spine. For the youngest victims of cervical spine trauma (age 0 to 3 years), a clinically challenging population, an algorithm for cervical spine evaluation has been developed and prospectively studied on a small scale (Figure 139-6). In a moderate-sized, two-center study, no injury was missed using this protocol.36
Algorithm for evaluation of the pediatric cervical spine (C-spine) in the patient with a reliable clinical exam. AP = anteroposterior; GCS = Glasgow Coma Scale. [Reproduced with permission from Chung S, Mikrogianakis A, Wales PW, et al: Trauma Association of Canada Pediatric Subcommittee National Pediatric Cervical Spine Evaluation Pathway: consensus guidelines. J Trauma. 2011 Apr;70(4):873-884. Copyright Wolters Kluwer Health.]
Algorithm for evaluation of the pediatric cervical spine (C-spine) in the patient with an unreliable clinical exam. GCS = Glasgow Coma Scale; LOC = level of consciousness; T/L = thoracic/lumbar. [Reproduced with permission from Chung S, Mikrogianakis A, Wales PW, et al: Trauma Association of Canada Pediatric Subcommittee National Pediatric Cervical Spine Evaluation Pathway: consensus guidelines. J Trauma. 2011 Apr;70(4):873-884. Copyright Wolters Kluwer Health.]
Algorithm for evaluation of the pediatric cervical spine in the patient age 0 to 3 years. AP = anteroposterior; ROM = range of motion; STIR = short T1 inversion recovery. [Reproduced with permission from Anderson RC, Kan P, Vanaman M, et al: Utility of a cervical spine clearance protocol after trauma in children between 0 and 3 years of age. J Neurosurg Pediatr. 2010 Mar;5(3):292-296.]