* Magnetic resonance imaging of the region of suspected neurological injury is recommended in a patient with spinal cord injury without radiographic abnormality (SCIWORA).
* Radiographic screening of the entire spinal column is recommended.
* Assessment of spinal stability in a SCIWORA patient is recommended with flexion-extension radiographs in the acute setting and at late follow-up, even in the presence of a magnetic resonance imaging negative for extraneural injury.
* Neither spinal angiography nor myelography is recommended in the evaluation of patients with SCIWORA.
* External immobilization of the spinal segment of injury is recommended for up to 12 weeks.
* Early discontinuation of external immobilization is recommended for patients who become asymptomatic and in whom spinal stability is confirmed with flexion and extension radiographs.
* Avoidance of “high-risk” activities for up to 6 months following SCIWORA is recommended.
Pang and Wilberger1 defined the term spinal cord injury without radiographic abnormality (SCIWORA) in 1982 as “objective signs of myelopathy as a result of trauma” with no evidence of fracture or ligamentous instability on plain spine radiographs and tomography. The definition specifically excluded all magnetic resonance imaging (MRI) findings and any injuries from penetrating trauma, electric shock, and obstetrical complications and those associated with congenital spinal anomalies.5 Although many practitioners may consider this diagnostic terminology anachronistic in light of the current near-universal availability of MRI, pediatric neurosurgeons continue to refer to this predominantly pediatric phenomenon as SCIWORA.
In their original article, Pang and Wilberger1 cautioned, “If the early warning signs of transient symptoms could be recognized and promptly acted upon before the onset of neurological signs, the tragic fate of some of these children might be duly averted.” Hamilton and Myles,2 Osenbach and Menezes,3 and Pang and Wilberger1-3 have documented the delayed onset of SCIWORA in children as late as 4 days following injury. Therefore, a concern is whether a child with a normal neurological examination but with a history of transient neurological symptoms or persisting subjective neurological symptoms referable to traumatic myelopathy should be assigned the diagnosis of SCIWORA and managed accordingly, despite the absence of “objective signs of myelopathy.”
Pang and Pollack4 have recommended obtaining a computed tomography scan focused at the neurological level of injury to exclude an occult fracture in a child with a neurological deficit referable to the spinal cord without abnormalities on plain radiographs of the spine. In addition, dynamic flexion and extension radiographs or fluoroscopy has been advocated to exclude pathological intersegmental motion consistent with ligamentous injury without fracture. If paraspinous muscle spasm, pain, or uncooperation prevents dynamic studies, they recommended external immobilization until the child can cooperatively flex and extend the spine for dynamic x-ray assessment. The finding of fracture, subluxation, or abnormal intersegmental motion at the level of neurological injury excludes SCIWORA as a diagnosis. In the initial report by Pang and Wilberger,1 1 of 24 children showed pathological motion on initial dynamic radiographs. By their own definition of SCIWORA, this 1 child would not be diagnosed with SCIWORA because the initial flexion and extension radiographs were abnormal. Although concern exists for the development of pathological intersegmental motion in children with SCIWORA following normal flexion and extension studies, there has been no documentation of such instability ever developing.
MRI findings in children with SCIWORA have spanned the spectrum from normal to complete cord disruption, along with evidence of ligamentous and disk injury in some.5-7 Possible roles for MRI of children with SCIWORA include identifying of signal change or intramedullary injury, excluding compressive lesions of the cord or roots or spinal ligamentous disruption that might warrant surgical intervention, guiding treatment regarding length of external immobilization, and/or determining when to allow patients to return to full activity.
Pang7 has also recommended somatosensory evoked potential (SSEP) screening of children with presumed SCIWORA. Possible roles for SSEPs in children with presumed SCIWORA include detecting subtle posterior column dysfunction when clinical findings are inconclusive, evaluating head-injured, comatose, or pharmacologically paralyzed children, distinguishing between intracranial, spinal, or peripheral nerve injuries, and/or providing a baseline MRI examination for comparison with subsequent evaluations.
Because subluxation and/or malalignment are, by definition, absent in SCIWORA, the mainstay of treatment has been immobilization and avoidance of activity that may either lead to exacerbation of the present injury or increase the potential for recurrent injury. Medical management issues such as blood pressure support and pharmacological therapy are of concern to this population as well and have been addressed in other guidelines. (Of note, the often-cited prospective studies of pharmacological therapy in the treatment of acute spinal cord injuries did not include children < 13 years of age.8)
Pang and Pollack4 originally recommended 12 weeks of external immobilization to allow adequate time for the healing of the presumed ligamentous strain/injury and to prevent exacerbation of the myelopathy. Conversely, Bosch et al9 found no evidence that bracing prevents recurrent SCIWORA and therefore recommended bracing only on a case-by-case basis. It is unclear, however, what role immobilization plays in this population once dynamic radiographs have confirmed the absence of instability. The duration of and even the need for immobilization remain debatable given the current literature. If the incidence of delayed pathological intersegmental motion in children with SCIWORA who have been proven to have normal dynamic radiographs approaches zero, then the role of spinal immobilization for SCIWORA patients needs to be considered in light of the available literature. If (normal) physiological motion of the spinal column can potentiate spinal cord injury (SCIWORA) in these patients when there is no malalignment, subluxation, or lesion causing cord compression, then immobilization may be warranted in these patients.
SCIWORA has been shown to be associated with a high incidence of complete neurological injuries, particularly in children < 9 years of age. Hadley et al10 reported 4 complete injuries in 6 children < 10 years of age with SCIWORA. The regions of complete injury tend to be cervical and upper thoracic. Pang and Wilberger1 and Pang7 found the presenting neurological examination to relate strongly to outcome. Some data suggest that MRI abnormalities (or lack of abnormalities) of the cord may be more predictive of outcome than presenting neurological status.11-13 Because no child has been documented to develop spinal instability following the diagnosis of SCIWORA and has, by definition, normal flexion and extension radiographs, there has been little impetus to define predictors of instability. On the other hand, children have been documented to suffer recurrent SCIWORA,7,14 and predictors of a “high-risk” subgroup of children with SCIWORA for recurrent injury may exist. The guidelines author group of the Joint Section on Disorders of the Spine and Peripheral Nerves of the American Association of Neurological Surgeons and the Congress of Neurological Surgeons have previously produced a medical evidence-based guideline on this topic.15 The purpose of this updated review is to provide a contemporary analysis of the literature on the diagnosis and treatment of SCIWORA since that original publication.
A National Library of Medicine (PubMed) computerized literature search from 1966 to 2011 was undertaken using Medical Subject Headings in combination with “spinal cord injury”: “pediatric,” “spinal cord concussion,” “cervical cord neurapraxia,” and “SCIWORA.” Approximately 188 citations were acquired. Non-English language citations were deleted. Articles written in English were reviewed for those that identified children who incurred an SCIWORA. Those articles that described the clinical aspects and management of children with SCIWORA were used to generate these guidelines. Case reports were excluded from review. Of the 19 articles meeting selection criteria, none provided Class I or Class II medical evidence. All were case series representing Class III medical evidence. Summaries of these 19 articles are provided in Evidentiary Table format (Table 1).
One concern is whether the child with a normal neurological examination and either a history of transient neurological deficit (ie, paraparesis or quadriparesis) or persisting subjective symptoms (ie, numbness or dysesthesias) would be a candidate for the diagnosis of SCIWORA. Pang and Wilberger1 described 13 of their 24 children to have a “latent” period from 30 minutes to 4 days (mean, 1.2 days) before the onset of objective sensorimotor deficits. All 13 of these children had transient subjective complaints at the time of their initial trauma that cleared within 1 hour before their subsequent neurological decline. Those who developed mild neurological deficits often improved to normal, and those who developed severe neurological deficits were often left with permanent neurological dysfunction. Hamilton and Myles,2 Osenbach and Menezes,3 and Pang and Pollack4 also reported a 22%, 23%, and 27% incidence, respectively, of delayed onset of myelopathy within their series of children with SCIWORA. Dickman et al,6 Eleraky and associates,16 and Hadley et al10 described no child having a latent period of neurological normalcy following injury. The observations of delayed deterioration by different investigators, however, raises the concern that any child presenting with a history of transient neurological deficit or symptoms following an appropriate mechanism of injury may be considered for the diagnosis of SCIWORA despite the absence of objective evidence of myelopathy on the initial neurological examination.
Pang and Wilberger1 reported 1 child of 24 whom they managed who demonstrated pathological intersegmental spinal motion on flexion and extension radiographs 1 week after injury, following resolution of that child’s neck pain and paraspinous muscle spasm. By definition, this child would not be considered to have had SCIWORA because the initial flexion and extension radiographs were abnormal. That child was treated successfully with external immobilization alone for 8 weeks. No child with SCIWORA has been documented in the literature to have had normal dynamic radiographs and then subsequently develop intersegmental instability.
In 1994, a series of 7 children with SCIWORA were demonstrated to have ligamentous, disk, and intramedullary abnormalities identified on MRI.13 Soft-tissue findings consisted of anterior longitudinal ligament disruption in association with a hyperextension injury, posterior longitudinal ligament disruption, and a noncompressive C2-3 disk herniation associated with lateral flexion and 1 case of C6-7 disk abnormality associated with hyperflexion. Intramedullary findings reported included cord transection and rostral stump hemorrhage, severe hematomyelia, a minor intramedullary hemorrhage, and edema without hemorrhage. Davis et al12 described 7 children with SCIWORA who were imaged with MRI. They described no abnormalities of muscles, ligaments, or disks but correlated the presence of intramedullary hemorrhage or cord “infarction” with permanent neurological deficit. The lack of intramedullary findings correlated with a normal neurological outcome. Dickman et al6 commented on 7 children with SCIWORA who were imaged with MRI. Five of the 7 studies revealed no abnormality, and 2 studies documented intramedullary signal changes. Osenbach and Menezes3 commented in their series of childhood SCIWORA that MRI and computed tomography--myelography performed on their patients did not demonstrate a single compressive lesion. In addition, they performed spinal arteriograms in 4 of 5 children with thoracic SCIWORA and found no angiographic abnormalities. Rossitch and Oakes17 performed myelograms on neonates and found no abnormalities that changed their management. Hadley et al15 obtained MRIs before 1988 on 5 children with SCIWORA and identified no abnormalities. Dare et al11 reviewed early MRI studies on 20 consecutive pediatric SCIWORA patients from 1992 to 1999, reporting abnormalities only in those with complete neurologic deficits. These reports and their results need to be viewed in the context of the technology and image quality available at the time of investigation.
SSEP evaluations of pediatric SCIWORA have been reported by Pang.7 In his series of 50 pediatric SCIWORA patients evaluated with both MRI and SSEPs, the SSEPs demonstrated abnormalities more frequently than MRI for both permanent and transient deficits.
Pang7 is the only investigator to report any alteration in the clinical management of pediatric SCIWORA based on MRI/SSEP findings. No child with MRI-documented ligamentous injury and SCIWORA has developed spinal instability, early or delayed. There has been no correlation between the ligamentous findings on MRI in SCIWORA patients and subsequent spinal instability to date. The appearance of the spinal cord on MRI does provide prognostic information regarding ultimate neurological outcome.
Hadley et al10 noted a 16% incidence of multiple noncontiguous injuries of the spine or spinal cord in children with any type of spinal column or spinal cord injury. Ruge et al had a similar incidence (17%) of multiple levels of spinal injury in children. Although neither of these 2 studies dealt with an isolated population of children with SCIWORA, they provide consistent observations that 1 in 6 children with spinal trauma will have multiple levels of injury. Pang and Wilberger1 reported 1 of 24 children with a second-level injury (L2 Chance fracture) who had a T6 neural injury (SCIWORA), but they did not obtain complete spine radiographs on every child. Pollina and Li19 reported 1 case of tandem SCIWORA with distinct cervical and thoracolumbar junction lesions identified on MRI. Because of these observations, one should consider radiographs of the entire spinal column when any traumatic spinal injury is identified in a child, SCIWORA or otherwise.
In the initial series of children with SCIWORA reported by Pang and Wilberger,1 treatment routinely consisted of 4 weeks of external immobilization with a “cervical collar” for cervical injuries. In cases of thoracic injury, if repeat plain radiographs showed no abnormality following 1 week of bed rest, the child was mobilized without a brace. In a later report in 1989, Pang and Pollack4 recommended 12 weeks of external immobilization for SCIWORA patients to allow healing of the presumed ligamentous strain/injury and to prevent exacerbation of the myelopathy. They also advocated external immobilization for this time frame to prevent recurrent injury during the healing phase. They reported 7 children who sustained recurrent SCIWORA of greater severity with lesser degrees of force when external immobilization was removed before 12 weeks or they were allowed to participate in activities against physician advice within 6 months of the initial injury. For these reasons, they recommend 12 weeks of external immobilization and 12 additional weeks of activity restriction following SCIWORA. In his updated series, Pang7 recommends immobilization with a “hard collar” in cases of transient deficits (< 24 hours) or “symptoms only,” provided that both MRI and SSEPs are normal. Those patients are then reevaluated after 1 to 2 weeks.
Dickman et al6 Eleraky et al,16 and Hadley and colleagues10 reported no neurological deterioration in any patient with SCIWORA following admission or discharge. None of these 3 reports described the length of time children with SCIWORA were immobilized. Bosch et al9 reviewed 189 cases of pediatric SCIWORA over 35 years, reporting no instances of neurologic deterioration and no benefit to bracing because it did not appear to reduce the frequency of recurrent SCIWORA. It has not been routine among treating physicians to prescribe 12 weeks of immobilization for children with SCIWORA.5 Although the single report by Pollack et al14 describes recurrent SCIWORA within 12 weeks of the original injury, this has not been validated by other observations.9 Because MRI evaluation was not available for those with recurrent injury, it is not known whether certain MRI characteristics (eg, ligamentous disruption) could predict an “at-risk” group for recurrent SCIWORA.
Additional case series of mild, transient SCIWORA have been reported using different terminology, namely “spinal cord concussion” or “cervical cord neurapraxia.”19-22 The diagnostic criteria for these entities fall within the definition of SCIWORA and further require that the symptoms/deficits resolve within 48 to 72 hours. Most of the patients in these reports are adolescents and young adults injured during athletic activities, especially American tackle football. Many of these series report an association between radiographically documented spinal stenosis and recurrent episodes20-22; however, Pang asserts that “congenital cervical stenosis and resultant [spinal cord injury], such as in young athletes, most definitely should be excluded from the SCIWORA umbrella” (D. Pang, MD, personal communication, November 2010).7 Return-to-play recommendations proposed in these reports are arbitrary, controversial, and beyond the scope of these guidelines.
While Pang and Wilberger1 reported that in their series neurological outcome correlated with the presenting neurological status, the MRI appearance of the spinal cord has been shown to be predictive of neurological outcome in children with SCIWORA.7,11-13,23 Absence of signal change within the cord is associated with an excellent outcome. Signal change consistent with edema or microhemorrhages, but not frank hematomyelia, is associated with significant improvement of neurological function over time. The presence of frank hematomyelia or cord disruption is associated with a severe, permanent neurological injury.11-13 The correlation of neurological outcome with spinal cord MRI findings in SCIWORA remains consistent with the findings in much larger numbers of patients with spinal cord injury (non-SCIWORA) who have been studied with MRI.4,23 Recognized prognostic factors related to SCIWORA are summarized in Table 2.
SCIWORA is a widely recognized form of spinal cord injury, occurring almost exclusively in children, and is characterized by the absence of any radiographically evident fracture, dislocation, or malalignment. Children presenting with a history of transient neurological signs or symptoms referable to the spinal cord after a traumatic event, despite the absence of objective neurological deficits with normal radiographs, may develop SCIWORA in a delayed fashion.
No child with SCIWORA has developed pathological intersegmental motion with instability when early flexion and extension radiographs have been normal.
MRI has not identified any abnormal findings in a child with SCIWORA when the management scheme would be changed by the results of the MRI. Similarly, no child with SCIWORA in whom a subsequent MRI has documented ligamentous injury has developed evidence of spinal instability.
Treatment consisting of cervicothoracic bracing for patients with cervical-level SCIWORA for 12 weeks and avoidance of activities that encourage flexion and extension of the neck for an additional 12 weeks has not been associated with recurrent injury. Patients with normal MRI and SSEP findings following transient deficits or “symptoms only” may be managed with a cervical collar for 1 to 2 weeks.
The spinal cord findings on MRI provide prognostic information regarding long-term neurological outcome in patients with SCIWORA. Myelography and angiography have no defined role in the evaluation of children with SCIWORA.
KEY ISSUES FOR FUTURE INVESTIGATION
The treatment end points of spinal immobilization and activity restriction for patients with SCIWORA have been arbitrarily chosen. MRI may be helpful to guide the duration of immobilization and activity restriction for that child. For example, the absence of ligamentous injury by MRI may indicate that there is no need for external immobilization or activity restriction. It has been observed that SCIWORA can recur despite the absence of demonstrable spinal instability and may not be prevented by bracing. A study that obtained MRI on all children with SCIWORA and followed up their clinical status longitudinally may highlight the utility of MRI in the management of children who go on to develop recurrent SCIWORA.
The literature provides little guidance as to the likelihood for subsequent catastrophic injury in children presenting with SCIWORA of any severity who are found to have a preexisting spinal or neurological abnormality such as congenital cervical stenosis or a Chiari malformation.19,21,24 Longitudinal clinical follow-up of SCIWORA patients of this type may provide information to appropriately counsel these children.
There are no data to elucidate the role of age in the success or failure of various treatments for this condition. This could be undertaken in a longitudinal study of a patient population of reasonable size. Serious attempts to address the topics above cannot be forthcoming from a single institution or investigator because of the relatively low numbers of children who sustain SCIWORA annually.25,26 A multiple-institution protocol-directed study of SCIWORA patients may provide answers to some of the questions that accompany this unique spinal cord injury subtype.
The authors have no personal financial or institutional interest in any of the drugs, materials, or devices described in this article.
1. Pang D, Wilberger JE Jr. Spinal cord injury without radiographic abnormalities in children. J Neurosurg. 1982;57(1):114–129.
2. Hamilton MG, Myles ST. Pediatric spinal injury: review of 174 hospital admissions. J Neurosurg. 1992;77(5):700–704.
3. Osenbach RK, Menezes AH. Spinal cord injury without radiographic abnormality in children. Pediatr Neurosci. 1989;15(4):168–174; discussion 175.
4. Pang D, Pollack IF. Spinal cord injury without radiographic abnormality in children: the SCIWORA syndrome. J Trauma. 1989;29(5):654–664.
5. Bruce DA. Efficacy of barbiturates in the treatment of resistant intracranial hypertension in severely head-injured children. Pediatr Neurosci. 1989;15(4):216.
6. Dickman CA, Zabramski JM, Hadley MN, Rekate HL, Sonntag VK. Pediatric spinal cord injury without radiographic abnormalities: report of 26 cases and review of the literature. J Spinal Disord. 1991;4(3):296–305.
7. Pang D. Spinal cord injury without radiographic abnormality in children, 2 decades later. Neurosurgery. 2004;55(6):1325–1342.
8. Bracken MB, Shepard MJ, Collins WF, Holford TR. A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury: results of the Second National Acute Spinal Cord Injury Study. N Engl J Med. 1990;322(20):1405–1411.
9. Bosch PP, Vogt MT, Ward WT. Pediatric spinal cord injury without radiographic abnormality (SCIWORA): the absence of occult instability and lack of indication for bracing. Spine (Phila Pa 1976). 2002;27(24):2788–2800.
10. Hadley MN, Zabramski JM, Browner CM, Rekate H, Sonntag VK. Pediatric spinal trauma: review of 122 cases of spinal cord and vertebral column injuries. J Neurosurg. 1988;68(1):18–24.
11. Dare AO, Dias MS, Li V. Magnetic resonance imaging correlation in pediatric spinal cord injury without radiographic abnormality. J Neurosurg. 2002;97(1 suppl):33–39.
12. Davis PC, Reisner A, Hudgins PA, Davis WE, O'Brien MS. Spinal injuries in children: role of MR. AJNR Am J Neuroradiol. 1993;14(3):607–617.
13. Grabb PA, Pang D. Magnetic resonance imaging in the evaluation of spinal cord injury without radiographic abnormality in children. Neurosurgery. 1994;35(3):406–414.
14. Pollack IF, Pang D, Sclabassi R. Recurrent spinal cord injury without radiographic abnormalities in children. J Neurosurg. 1988;69(2):177–182.
15. Spinal cord injury without radiographic abnormality (SCIWORA). In: Guidelines for the management of acute cervical spine and spinal cord injuries. Neurosurgery. 2002;50(3 suppl):S100–S104.
16. Eleraky MA, Theodore N, Adams M, Rekate HL, Sonntag VK. Pediatric cervical spine injuries: report of 102 cases and review of the literature. J Neurosurg. 2000;92(suppl 1):12–17.
17. Rossitch E Jr, Oakes WJ. Perinatal spinal cord injury: clinical, radiographic and pathologic features. Pediatr Neurosurg. 1992;18(3):149–152.
18. Ruge JR, Sinson GP, McLone DG, Cerullo LJ. Pediatric spinal injury: the very young. J Neurosurg. 1988;68(1):25–30.
19. Pollina J, Li V. Tandem spinal cord injuries without radiographic abnormalities in a young child. Pediatr Neurosurg. 1999;30(5):263–266.
20. Del Bigio MR, Johnson GE. Clinical presentation of spinal cord concussion. Spine (Phila Pa 1976). 1989;14(1):37–40.
21. Torg JS, Corcoran TA, Thibault LE, et al.. Cervical cord neurapraxia: classification, pathomechanics, morbidity, and management guidelines. J Neurosurg. 1997;87(6):843–850.
22. Torg JS, Naranja RJ Jr, Pavlov H, Galinat BJ, Warren R, Stine RA. The relationship of developmental narrowing of the cervical spinal canal to reversible and irreversible injury of the cervical spinal cord in football players. J Bone Joint Surg Am. 1996;78(9):1308–1314.
23. Liao CC, Lui TN, Chen LR, Chuang CC, Huang YC. Spinal cord injury without radiological abnormality in preschool-aged children: correlation of magnetic resonance imaging findings with neurological outcomes. J Neurosurg. 2005;103(suppl 1):17–23.
24. Ramón S, Domínguez R, Ramírez L, et al.. Clinical and magnetic resonance imaging correlation in acute spinal cord injury. Spinal Cord. 1997;35(10):664–673.
25. Osenbach RK, Menezes AH. Pediatric spinal cord and vertebral column injury. Neurosurgery. 1992;30(3):385–390.
26. Turgut M, Akpinar G, Akalan N, Ozcan OE. Spinal injuries in the pediatric age group: a review of 82 cases of spinal cord and vertebral column injuries. Eur Spine J. 1996;5(3):148–152.
27. Rathbone D, Johnson G, Letts M. Spinal cord concussion in pediatric athletes. J Pediatr Orthop. 1992;12(5):616–620.