Pediatric Cervical Spine Injuries: Report of 102 Cases and Review of the Literature

January 2000 Volume 92 Number 1

Mohammed A. Eleraky, M.D., Nicholas Theodore, M.D., Mark Adams, M.D., Harold L. Rekate, M.D. and Volker K.H. Sonntag, M.D.
 
Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, Arizona

Object.   To evaluate and review their experience with pediatric cervical injuries and factors affecting outcome, the authors conducted a retrospective clinical study of 102 cases (65% boys, 35% girls) of pediatric cervical spine injuries treated in the last decade. This study is an extension of and comparison with their earlier experience.

Methods.   Patients were divided into two age groups-birth to 9 years (Group 1) and 10 to 16 years of age (Group 2)-;and managed according to status at presentation and type of injury. Thirty patients were managed surgically and 72 nonsurgically (42 wore a halo brace and 30 wore hard collars or custom-molded braces).

Motor vehicle accidents were the most common cause of injury, and 40% were associated with head injury. Patients in the youngerage group (Group 1) sustained more neurological injuries than the older patients in Group 2, and most injuries were in the upper cervical spine. Of the 38 children in Group 1, in 39% a subluxation was present and in 29% a fracture or fracture/subluxation was demonstrated. Of the patients in Group 2, 80% had sustained fractures or fracture/subluxations. Vertebral fractures were the most common radiological findings (32%). At late followup review (mean 5 years), solid fusions were demonstrated in all patients. Neurological deterioration did not occur in any patient. The mortality rate was 16%. Compared with the authors' earlier report, the incidence of cases with pediatric cervical injuries increased, as did the number managed surgically. Various fusion techniques were used, and neurological and fusion outcomes improved as compared with the previous report.

Conclusions.   The prognosis of neurological recovery from pediatric cervical spine injuries is related to the severity of the initial neurological injury. Management must be tailored to the patient's age, neurological status, and type and level of injury. Compared with our earlier experience, fusion and instrumentation procedures were used more frequently. Different types of fusion and instrumentation procedures can be performed safely in children and produce good outcomes.

Key Words.   cervical spine, spinal cord injury, children

Abbreviations.   MR = magnetic resonance, MVA = motor vehicle accident, SCI = spinal column injury

In the pediatric population, cervical spine injuries are rare, constituting between 0.65% and 9.47% of all cervical spine injuries.16, Despite their rarity, the economic, social, and emotional consequences of these injuries are significant in children. 1, 4, This study represents a continuation of an earlier report 11, from this institution, adding to the literature 102 new pediatric cases treated in the decade subsequent to the earlier report. The present series includes treatment with a variety of fusion techniques not used in the previous study. It also represents a longer followup period in which more favorable outcomes were achieved.

Clinical Material and Methods
Between 1986 and 1997 we treated 102 pediatric patients (65% boys, 35% girls) with cervical spine injuries with and without SCI. Patients were divided into two age categories: birth to 9 years of age (38 patients; Group 1) and 10 to 16 years of age (64 patients; Group 2). These groups facilitated a comparison of age with injury pattern, incidence of neurological compromise, and outcome after therapy. The median age of the entire group was 11.3 years (range 1-16 years).

Twenty-two patients were directly referred to us; the other patients were evaluated, diagnosed, and initially treated at our tertiary neurosurgical trauma institution. Patients with missile injuries, brachial plexus or peripheral nerve injuries, and extensive fractures of the upper and lower extremities were excluded from this study.

The medical records and radiographic studies of the patients were reviewed retrospectively. Patients were characterized according to their age, gender, mechanism of injury, diagnosis, neurological deficit, associated injuries, radiographic findings, and treatment outcome. Neurological function was graded according to the Frankel classification system at admission, discharge, and follow up.

Diagnostic Studies
Diagnostic radiographic studies, including plain cervical radiographs (anteroposterior, lateral, swimmer's, and open-mouth odontoid views), were obtained in all patients. Flexion-extension radiographs were also obtained to assess spinal stability when plain radiography and computerized tomography studies demonstrated normal results and in patients with SCI without radiographic evidence of abnormality. Fine-slice computerized tomography scans were obtained if the presence of a fracture was uncertain. Magnetic resonance imaging was performed in 88 patients. Follow-up assessment included clinical evaluation and 3 and 6month radiographic studies (mean follow up 32 months; range 1250 months).

Results
Motor vehicle accidents were the most common cause of injury (48%), followed by sports (35%) and fallrelated injuries (15%). Most injuries in the younger patients (Group 1) were caused by falls (30%) and MVAs (42%). In the older patients (Group 2), sportsrelated accidents (38%) and MVAs (50%) were the most common cause of injury.

Overall, 42 (41%) of the 102 patients were neurologically intact at admission. Fourteen patients (14%) had sustained complete SCIs (Frankel Grade A), and 46 patients (45%) had incomplete injuries: 18 were Frankel Grade D, 22 were Grade C, and six were Grade B. Fiftythree patients (52%) suffered upper cervical spine injuries, and 49 (48%) sustained lower cervical injuries. Altogether, there were 18 patients (18%) with SCI in whom radiographic evaluation showed no abnormality. Neurological injuries were most common in the youngeraged patients (Group 1; 60%), as were upper cervical spine injuries (80% of all cases of upper injury; Table 1). Head trauma was the most common concurrent injury and was associated with 40% of all the cases and with 60% of those with upper cervical spine injuries.

Overall, four patterns of injury were recognized on radiographic studies: vertebral fractures (33%); fractures with subluxation (27%); subluxation without fractures (22%); and SCI without radiographic evidence of fractures or subluxation (18%; Table 1 ). Twelve patients with SCI without radiographic evidence of abnormality underwent MR imaging. The results of these studies were normal in four of these patients, showed segmental areas of spinal cord swelling in five cases, and showed spinal cord contusions in three cases. In the patients in Group 1 a higher incidence of subluxation only (39%) and SCI without radiographic evidence of abnormality (30%) was demonstrated as compared with the older group. The children in Group 2 tended to sustain fracture only or fracture/subluxation injuries (42% and 38% of all cases in this age group, respectively; Table 1). Overall, 11 patients in Group 1 sustained fractures compared with 51 patients in Group 2.

Surgical and Nonsurgical Treatment
Thirty patients (30%) were treated surgically (Table 2). The anterior cervical approach was performed in 18 patients: in four with hyperflexion injuries, in six with vertebral body fractures, in six with unreduced vertebral dislocations, and in two with Type II odontoid fractures. The latter two underwent a fixation procedure in which odontoid screws were placed. Corpectomies were performed in five patients, the youngest of whom was 14 years of age. The other 11 patients, all between the ages of 10 and 16 years (that is, Group 2 only), were treated by anterior cervical discectomy and fusion.

Twelve patients underwent fusion via a posterior approach in which bone grafting and instrumentation were applied. Seven patients underwent occiptocervical fusion: one for a combined atlantoaxial and atlantooccipital dislocation, two for occipitocervical dislocations (Fig. 1), and four for atlantoaxial dislocations. The youngest patient to undergo occipitocervical fusion was 17 months of age and had suffered an atlantoaxial dislocation. The remaining five posterior fusion procedures consisted of three with posterior wiring and one with a lateral mass plate system. Fusion was performed using a posterior transarticular screw and autograft in a 15yearold patient with instability at C12.

In the conservatively managed group, 42 patients wore halo vests, and 30 wore hard collars and custommolded braces for an average of 8 to 12 weeks.

Outcome After Conservative and Surgical Treatment
None of the conservatively or surgically treated patients exhibited neurological deterioration (Table 2). In all patients who underwent fusion procedures solid fusions were demonstrated at 6month followup examination. There was no incidence of surgeryrelated death or complications.

Of the 46 patients with incomplete SCI, 38 (83%) returned to normal, three (7%) improved mildly, and five died. Altogether, 16 (16%) of the 102 patients died. Eleven of the deaths occured in children with complete SCIs and five in those with incomplete SCIs. In patients with lower cervical spine injuries, the mortality rate was 10%, whereas it was 20% in patients with upper cervical spine injuries. Five patients with complete upper SCIs died in the hospital within the first 72 hours of injury, and six died 1 to 3 years after injury of causes unrelated to injury. Two patients with incomplete SCIs died within the first 24 hours after sustaining multiple traumatic injuries. One patient with an incomplete SCI died of a concomitant severe head injury 2 weeks after injury. Two other patients with incomplete SCIs died of unrelated causes 3 years after injury.

Discussion
Cervical spine injuries in children are fundamentally different from those in adults because of anatomical differences in the developing spine. In children the mass of the head is disproportionately large, and the neck muscles are relatively underdeveloped. The vertebral bodies are wedge shaped, the articulating facets are angled horizontally, the end plates are cartilaginous, and the interspinous ligaments are elastic and lax. These features predispose children to upper cervical spine injuries, SCIs without radiographic evidence of abnormality, or severe ligamentous injuries. 2, 6, 7, 25, Such cervical spine injuries can also present special problems regarding the use of external immobilization systems and surgical intervention when the child's potential for growth is still significant.

In children 9 years of age or younger, such as those in our Group 1, the spine retains its immature features. Between the ages of 10 and 16 years, as in our Group 2, however, the vertebrae become more like those of adults. The different types of injuries in the two age groups reflect this agerelated maturation of the spine. There are fewer fractures in children younger than age 10 years as compared with those between the ages of 10 and 16 years because of the greater mobility of the spine and laxity of the ligaments in the younger group. The vertebrae start to ossify and mature when the child reaches approximately 9 years of age. Anteriorly, the vertebral body loses its wedge shape and becomes more rectangular. The orientation of the facets becomes less horizontal and more vertical, and the uncinate process begins to protrude. 8, 11,17, 19,

Some congenital abnormalities and pathological conditions may predispose afflicted children to injury even from mild trauma. Congenital diseases (for example, Down's syndrome, KlippelFeil syndrome, Chiari malformation, and basilar invagination) and pathological conditions (for example, neoplasm, infection, and metabolic disorders) 1, 5, increase the risk of cervical spine injuries because of a combination of ligamentous laxity and bony abnormalities usually present in the upper cervical spine.

Among children, cervical spine injuries constitute 60 to 80% of vertebral injuries. Young children tend to sustain upper cervical injuries, whereas adolescents sustain a greater proportion of lower cervical injuries. This pattern reflects the fulcrum of cervical motion at C56 in adolescents, which is located at C23 in younger children. 15, 22, Osenbach and Menezes 24, have reported that in 66% of 179 cases of pediatric spine injuries, cervical spine injuries were demonstrated (Table 3), as they were in 42% of 174 cases in other series. 12, 13, Minor neck injuries such as cervical strains or whiplash, the most common causes of cervical injuries, have not been accounted for accurately because the victims may not seek medical attention or are rarely admitted to the hospital.

Injury Profile
The 102 cases of cervical injury in this study represent 72% of the cases of pediatric spine injuries treated at our institution in the last 10 years. Compared with our previous experience (59%), 11, this relatively increased rate likely represents both an increase in referral rates and an increase in survivors in the field due to advanced life support. Furthermore, MR imaging has likely contributed to an increase in the diagnosis of cervical injuries in children.

In our study 38% of the patients were 9 years of age or younger (Group 1), and 78% of them sustained upper cervical injuries. In contrast, 62% were between the ages of 10 and 16 years (Group 2), and 70% of these patients sustained lower cervical injuries. Fiftyone patients in Group 2 had sustained a fracture or subluxation in contrast with 11 patients in the younger patient group. In our previous report, 11, 18% of the 72 patients were between the ages of 1 day and 9 years, and in 69% occiputC2 injuries and in 31% lower cervical spine injuries were demonstrated. Of the 59 patients between 10 to 16 years, 27% had upper cervical injuries and 73% had lower cervical injuries. In the report by Osenbach and Menezes, 24, 79% of the patients younger than 9 years of age sustained cervical injuries, and upper cervical injuries were documented in 53%. Of their patients between the ages of 10 and 16 years, cervical injuries were demonstrated in 53% of patients, in whom 26% had upper cervical injuries.

In the current study, the most common causes of injury were MVAs (48%) and sportsrelated accidents (35%). Orenstein, et al., 23, have reported that 65% of the injuries in their series were caused by MVAs and 30% involved sports. In the series reported by Givens, et al., 9, MVAs caused 68% of the injuries and sportsrelated accidents caused 28%.

As many as 66% of the cervical spine injuries in children may be associated with head injury. 3, 14, 20, Orenstein, et al., 23, reported that only 30% of the injuries in their patients were associated with head injuries, whereas Givens, et al., 9, reported that 53% of the injuries in their patients were associated with head trauma. In our study 40% of the cases with cervical spine injury were associated with head trauma, 60% of which were associated with upper cervical injuries.

In our study, there were 18 patients (18%) with SCI in whom there was no radiographic evidence of abnormality; and in none of these patients was delayed onset or recurrence demonstrated. In other studies, 15, 21, 22, 26, the rate of SCI without radiographic evidence of abnormality in cases of cervical spine injuries has ranged from 21% to as low as 1 to 4%, again with no reported cases of delayed onset. Pang and Wilberger, 25, however, reported a delayed onset (up to 4 days) in 52% of their cases. Magnetic resonance images were obtained in only 12 of these cases because MR imaging was unavailable at the start of that study. The findings, however, did not alter their management practice, which was conservative in all cases. Grabb and Pang 10, have reported findings on MR imaging in six cases, but again, the findings did not alter their management strategy. Pang and Wilberger have recommended early immobilization and aggressive supportive treatment. To rule out instability, delayed dynamic radiographic studies must be obtained in all patients with SCI in whom initial radiographic studies revealed no evidence of abnormality.

Management and Outcome
The treatment of cervical spinal injuries in the pediatric population must be individualized, and management depends on the patient's age, the severity and level of injury, the degree of neurological compromise, and the presence of associated injuries. 26, Many authors 7, 28, 29, have found no difference in outcome between patients treated surgically and those treated nonsurgically, recommending surgery for patients with markedly unstable injuries, irreducible dislocations, and incomplete injuries associated with progressive neurological symptoms.

In our study, 30 patients (29%) were managed surgically by different methods of stabilization, and in all cases solid fusions were achieved. In these cases, the indications for surgery were an unstable spine in 19 patients who wore a halo orthosis, deteriorating neurological status in five patients, irreducible fracture/dislocations in three patients, and failed conservative treatment in three patients who had worn a hard collar. Onlay bone graft may be adequate; however, fusion in which bone graft and instrumentation are used is optimal for attaining longterm stability. 26, 27, Fusion should be confined to unstable segments to preserve physiological mobility and the potential for growth. 24, 27, 30, Roy and Gibson 27, have reported that posterior fusion in pediatric patients is suitable to provide solid arthrodesis and excellent range of motion. Lowry, et al., 18, have reported eight patients who underwent posterior fusion of the upper cervical spine, and there was one case in which the fusion failed due to resorption of the graft. They recommended placing patients in halo vests until bone growth is visible.

Compared with the earlier report from this institution,11, the number of surgical interventions has almost doubled. Whereas only 12 (16.6%) of the 72 patients in the earlier study underwent surgery, the rate of surgical treatment in the current study was 29%. This increase likely reflects the increasing comfort level of neurosurgeons in using instrumentation and fusion techniques to treat cervical instability. In the series reported by Osenbach and Menezes, 24 the 25% rate of surgical treatment is similar. In both series the authors reported excellent fusion rates.

In the present series, 72 patients were managed conservatively. Of these, 42 patients were managed successfully with halo vests (the youngest of whom was 2 years of age). Patients with unstable fractures but normal spinal cord alignment should be considered for halo immobilization. Osenbach and Menezes 24, have recommend the use of a halo vest to treat injuries of the upper cervical spine. They found that vest therapy provided superior immobilization, caused minimal morbidity, and could be used in patients as young as 1 year of age. Halo vests are dependable and may be particularly suited for treating uncooperative children. The use of a halo vest, however, requires special consideration and diligent followup evaluation to avoid complications. The decision to use a halo vest must reflect the requirements each patient's clinical and radiological circumstances. We used custommolded braces and hard collars to treat patients with stable fractures. In all the patients ultimately managed conservatively good stability was shown.

All neurologically intact patients remained so. In patients with incomplete injuries, 83% (38 of the 46 cases) recovered completely, and six patients had improved neurological function by one or two Frankel grades. Five deaths were associated with severe head injuries. In several reports the authors suggest that in children with incomplete SCIs, good recoveries are demonstrated and they attribute this propensity to the immature spinal cord's plasticity and capacity for greater functional recovery. In our study the mortality rate was 16% (16 patients). Eleven patients had sustained complete SCIs, and the other five had suffered incomplete SCIs with severe head trauma. Orenstein, et al., 23, have reported a 19% mortality rate, and Givens, et al., 9, reported a 35% mortality rate.

Conclusions
The outcome of cervical injuries in children is more favorable than in adults, especially in patients with incomplete injuries. Upper cervical spine injuries are more common between birth and the age of 9 years than in older children, but fractures and fracture/subluxations are rare in this group. When indicated, surgical intervention in which instrumentation and fusion are used is very effective in the pediatric population. A longterm followup study in which the effects of spinal instrumentation and fusion in the pediatric population are assessed is currently underway. The prognosis for children with complete SCIs, however, is still discouraging.

References
1. Anderson JM, Schutt AH: Spinal injury in children: a review of 156 cases seen from 1950 through 1978. Mayo Clin Proc 55:499504, 1980
2. Apple JS, Kirks DR, Merten DF: Cervical spine fractures and dislocations in children. Pediatr Radiol 17:4549, 1987

3. Birney TJ, Hanley EN Jr: Traumatic cervical spine injuries in childhood and adolescence. Spine 14:1277-1282, 1989

4. Bohn D, Armstrong D, Becker L: Cervical spine injuries in children. J Trauma 30:463-469, 1990

5. Dickman CA, Rekate HL, Sonntag VK: Pediatric spinal trauma: vertebral column and spinal cord injuries in children. Pediatr Neurosci 15:237-255, 1989

6. Dickman CA, Zabramski JM, Hadley MN: Pediatric spinal cord injury without radiographic abnormalities: report of 26 cases and review of the literature. J Spinal Disord 4:296-305, 1991

7. Farley FA, Hensinger RN, Herzenberg JE: Cervical spinal cord injury in children. J Spinal Disord 5:410-416, 1992

8. Fesmire FM, Luten RC: The pediatric cervical spine: developmental anatomy and clinical aspects. J Emerg Med 7:133-142, 1989

9. Givens TG, Polley KA, Smith GF: Pediatric cervical spine injury: a threeyear experience. J Trauma 41:310-314, 1996

10. Grabb PA, Pang D: Magnetic resonance imaging in the evaluation of spinal cord injury without radiographic abnormality in children. Neurosurgery 35:406-414, 1994

11. Hadley MN, Zabramski JM, Browner CM: Pediatric spinal trauma. Review of 122 cases of spinal cord and vertebral column injuries. J Neurosurg 68:18-24, 1988

12. Hamilton MG, Myles ST: Pediatric spinal injury: review of 174 hospital admissions. J Neurosurg 77:700-704, 1992

13. Hamilton MG, Myles ST: Pediatric spinal injury: review of 61 deaths. J Neurosurg 77:705-708, 1992

14. Henrys P, Lyne ED, Lifton C: Clinical review of cervical spine injuries in children. Clin Orthop 129:172-176, 1977

15. Hill SA, Miller CA, Kosnik EJ: Pediatric neck injuries. A clinical study. J Neurosurg 60:700-706, 1984

16. Kalfas I, Wilberger J, Goldberg A: Magnetic resonance imaging in acute spinal cord trauma. Neurosurgery 23:295-299, 1998

17. Kewalramani LS, Kraus JF, Sterling HM: Acute spinalcord lesions in a pediatric population: Epidemiological and clinical features. Paraplegia 18:206-219, 1980

18. Lowry DW, Pollack IF, Clyde B: Upper cervical spine fusion in the pediatric population. J Neurosurg 87:671-676, 1997

19. Manary MJ, Jaffe DM: Cervical spine injuries in children. Pediatr Ann 25:423-428, 1996

20. Michael DB, Guyot DR, Darmody WR: Coincidence of head and cervical spine injury. J Neurotrauma 6:177-189, 1989

21. Mirvis SE, Diaconis JN, Chirico PA: Protocoldriven radiologic evaluation of suspected cervical spine injury: efficacy study. Radiology 170:831-834, 1989

22. Nitecki S, Moir CR: Predictive factors of the outcome of traumatic cervical spine fracture in children. J Pediatr Surg 29:1409-1411, 1994

23. Orenstein JB, Klein BL, Gotschall CS: Age and outcome in pediatric cervical spine injury: 11year experience. Pediatr Emerg Care 10:132-137, 1994

24. Osenbach RK, Menezes AH: Pediatric spinal cord and vertebral column injury. Neurosurgery 30:385-390, 1992

25. Pang D, Wilberger JE Jr: Spinal cord injury without radiographic abnormalities in children. J Neurosurg 57:114-129, 1982

26. Rauzzino MJ, Hadley MN: Pediatric spinal cord injuries, in Menezes AH, Sonntag VKH (eds): Principles of Spinal Surgery., New York: McGrawHill, 1996, pp 817-840

27. Roy L, Gibson DA: Cervical spine fusions in children. Clin Orthop 73:146-151, 1970
28. Ruge JR, Sinson GP, McLone DG: Pediatric spinal injury: the very young. J Neurosurg 68:25-30, 1988

29. Saleh J, Raycroft JF: Hyperflexion injury of cervical spine and central cord syndrome in a child. Spine 17:234-237, 1992

30. Shacked I, Ram Z, Hadani M: The anterior cervical approach for traumatic injuries to the cervical spine in children. Clin Orthop 292:144-150, 1993.