- Email: [email protected]
Spinal injuries in children
Spinal Injuries in Children By Bayram Cirak, Suzan Ziegfeld, Vinita Misra Knight, David Chang, Anthony M. Avellino, and Charles N. Paidas Baltimore, Maryland
Background/Purpose: Traumatic spinal injury (TSI) is an uncommon source of morbidity and mortality in children. The aim of this study was to describe childhood TSI in a single level 1 urban pediatric trauma center. Methods: The authors retrospectively analyzed all children younger than 14 years with TSI, treated at a level I pediatric trauma center between 1991 and 2002 (n ⫽ 406, 4% total registry). All children were stratified according to demographics, mechanisms, type and level of injury, radiologic evaluations, associated injuries, and mortality. Results: The mean age was 9.48 ⫾ 3.81 years. The most common overall mechanism of injury was motor vehicle crash (MVC; 29%) and ranked highest for infants. Falls ranked highest for ages 2 to 9 years. Sports ranked highest in the 10 to 14 year age group. Paravertebral soft tissue injuries were
I
NJURIES to the spine, although relatively rare, contribute to significant morbidity and mortality in children. The reported incidence of pediatric traumatic spinal injury (TSI) varies, and most of the previous studies are limited because of the sparse attention to both the thoracic and lumbosacral region.1,2 Limitations of the definitive prevalence studies are that birth injuries and child abuse cases are underreported.3-6 Compared with adults, spinal trauma is relatively rare in pediatric patients. But the mortality rate is higher in children as a result of associated injuries.7,8 There are certain anatomic and biomechanical differences between the immature spine of pediatric patients and adults. These include a greater mobility of the spine owing to ligamentous laxity, shallow angulations of facet joints, immature development of neck musculature, and incomplete ossification of the vertebrae.9 As a result of these differences, 60 to 80% of all pediatric vertebral injuries are in the cervical region. This is in contrast to adults in whom cervical injuries constitute only 30 to 40% of all vertebral injuries.10 Consequently, our overall understanding of characteristics of pediatric TSI is incomplete. The aim of this retrospective review was to identify demographic characteristics of pediatric spinal injury with regard to mechanism, type, level, associated injuries, and mortality in an urban trauma center. MATERIALS AND METHODS After institutional IRB approval at the Johns Hopkins Hospital, a level I urban trauma center, data were obtained from the pediatric Journal of Pediatric Surgery, Vol 39, No 4 (April), 2004: pp 607-612
68%. The most common injury level was the high cervical spine (O-C4). The incidence of spinal cord injury without radiologic abnormality (SCIWORA) was 6%. Traumatic brain injury (37%) was the most common associated injury. Overall mortality rate was 4% in this urban catchment. Conclusions: TSI in children requires a different preventive and therapeutic logarithm compared with that of adults. The potential devastating nature of TSI warrants that the health care team always maintains a high index of suspicion for injury. Future prospective studies are needed to further elucidate injury patterns. J Pediatr Surg 39:607-612. © 2004 Elsevier Inc. All rights reserved. INDEX WORDS: Pediatric trauma, spinal injury, spinal cord injury without radiologic abnormality, abuse.
trauma registry, a comprehensive database of approximately 12,000 consecutively injured children from 1991 to 2002. All children age 14 and younger with spinal trauma diagnoses, defined by International Classification of Diseases 9th Revision, Clinical Modification (ICD-9-CM) codes ranging from 805.0 to 806.9 and also 839.0 to 839.9 and 847.0 to 847.9 were included in this review. In addition, ICD-9 codes 952.0 to 952.9 were utilized to identify cases of spinal cord injury without radiologic abnormality (SCIWORA). Mechanisms of injury were classified by ICD-9 external injury codes. The spinal column was divided into 5 different regions according to anatomic and physiologic differences in each level to facilitate analysis. The regions include occipital to cervical vertebrae (O-C4), lower cervical (C5-C7), upper thoracic (T1-T10), thoraco-lumbar (T11-L1), and lumbo-sacral (L2-sacrum). Statistical analysis was performed using STATA software (College Station, TX). Comparisons between groups were tested using analysis of variance (ANOVA) for continuous outcome variables, with 2 tests for categorical outcome variables and with Kruskal-Wallis tests for differences between group medians. Odds of associated injuries were calculated using forward stepwise regressions, beginning with empty models and adding age, gender, mechanism of injury, level of any spinal column injuries, and level of any paravertebral injuries. The rationale for the selection of these variables was to include only
From the Pediatric Division, Department of Neurosurgery; the Department of Pediatric Surgery; and the Department of Health Policy and Management, Bloomberg School of Public Health, Johns Hopkins Medical Institutions, Baltimore, MD. Address reprint requests to Charles N. Paidas MD, FACS, FAAP, Department of Pediatric Surgery, 600 North Wolfe St, CMSC 7-116, Baltimore, MD, 21287. © 2004 Elsevier Inc. All rights reserved. 0022-3468/04/3904-0021$30.00/0 doi:10.1016/j.jpedsurg.2003.12.011 607
608
CIRAK ET AL
Table 1. Demographics Patients (yr)
Male
Female
Total (%)
Median Final ISS
Number With Final ISS Greater Than 15 (%)
Infants (0-1) Toddlers (2-5) School (6-9) Adolescent (10-14) Total
8 37 50 164 259
9 17 40 81 147
17 (4) 54 (13) 90 (22) 242 (60) 406
4.5 3 2 1 2
4 (23.5) 11 (20.4) 14 (15.6) 25 (10.2) 54 (13.3)
variables that would be available during the initial workup of a patient. Therefore, other variables that may have statistically significant relationship to risks of associated injuries but would not be available on initial patient workup (for example, Injury Severity Score) were not included. P values for inclusion to and exclusion from the models were set at .05, and the likelihood-ratio test was used to test for significance.
RESULTS
Five characteristics were used to evaluate the children with spinal injuries. The definitions of patient groups, etiologies, injury levels, SCIWORA, and associated injuries were standard well accepted literature parameters used to evaluate injury data.7,8,11-14 Demographics A total of 406 patients met our inclusion criteria (Table 1). The mean age was 9.48 ⫾ 3.81 years (2 months to 14 years), and approximately two thirds of the population were boys. The children in the youngest age groups had higher mean and median ISS than older children. Moreover, they have the lowest proportion of severe injuries (ISS ⬎ 15), reflecting very different distribution of patient injury severity levels between the different age groups. Considering the nonlinear and noncontinuous nature of ISS, the group medians (instead of the means) were selected for statistical analysis. Using Kruskal-Wallis test, it was found that the differences in injury severity were not significantly different between age groups (P ⫽ .06). In contrast, when the group means were used and tested with ANOVA, the differences in injury severity between age groups were highly significant (P ⬍ .01). This discrepancy is a reflection of the very different distributions of childhood injury severity levels between the different age groups.
Injury Mechanisms Across all age groups, motor vehicle collisions (MVC) accounted for the majority of injuries (29%; Table 2). However, the distributions of mechanisms of injury are significantly different between the different age groups (P ⬍ .01). Demonstrating the common exposure of this age group, MVCs were the leading cause of injury in infants (71%). Falls were the leading cause of spinal trauma in toddler and school age, 48% and 34%, respectively. The leading mechanisms in the adolescent group were sports-related injuries followed by MVCs. Adolescents accounted for all gunshot wounds, demonstrating the exposure of this age group to this mechanism. Injury Level To facilitate evaluation, we divided spinal injuries of the 4 age cohorts into 2 categories (Table 3A & 3B). The first category includes children with paravertebral soft tissue injury (defined as muscular sprains), and the second group consists of children with only injury to vertebral column or spinal cord. Included in this cohort (Table 3a) is the SCIWORA group. The definition of SCIWORA has evolved with advances in imaging; currently, there are 2 working definitions. One definition is children with positive neurologic findings and negative plain x-ray and computed tomography (CT) scan but pathologic spinal cord MR imaging. According to this definition, 24 children (6%) had SCIWORA. The second definition is defined as abnormal neurologic examination with normal imaging (including MR); our SCIWORA incidence according to this definition decreases to 5 children (1%).
Table 2. Injury Mechanisms Etiology
MVC (%)
Fall (%)
Sports (%)
Pedestrian (%)
Other Collisions*
Bike (%)
Diving (%)
Gunshot (%)
Total
Infant Toddlers School Adolescent Total (%)
12 (71) 20 (37) 26 (29) 59 (24) 117 (29)
2 (12) 26 (48) 31 (34) 45 (18) 104 (26)
0 0 9 (10) 72 (29) 81 (20)
0 6 (11) 15 (17) 21 (9) 42 (10)
3† (18) 2 (4) 4 (4) 22 (9) 31 (8)
0 0 4 (4) 13 (5) 17 (4)
0 0 1 (1) 7 (3) 8 (2)
0 0 0 6 (2) 6 (1)
17 54 90 245 406
NOTE. MVC: E810-825 (.1) (Excludes 821.1-all terrain vehicle), sports: 917.0, 917.5, 886.0 (excludes bike and diving), fall: E880.0-E888 (excludes 886.0), pedestrian: E810-825 (0.7 only), other collisions: E916-E928 (excluding 917.0 and 917.5), bike: 826.-E829, diving: approximately 883.0885.0, gunshot wounds: E922.0 ⫹ .9 and E965.0-965.4, and 985.0-985.4. *“Other collisions” includes struck accidentally by falling object, caught between objects. †One child (2-month-old boy) was a victim of child abuse.
SPINAL INJURIES IN CHILDREN
609
Table 3. Injury Level a. Paravertebral Soft Tissue Injury Injury Level
Cervical
Thoracic
Lumbar
Total
Infant Toddler School Adolescent Total (%)
5 (83.3%) 26 (90.0%) 48 (77.4%) 149 (82.8%) 228 (82.3%)
1 (16.7%) 3 (10.3%) 8 (12.9%) 22 (12.2%) 34 (12.3%)
0 0 6 (9.7%) 9 (5.0%) 15 (5.4%)
6 29 62 180 277(100%)
b. Injury to Vertebral Column or Spinal Cord Injury Level
O-C4
C5-7
T1-10
T11-L1
L2-S
Total
Infant Toddler School Adolescent Total (%)
7 (63.6%) 15 (48.4%) 18 (60.0%) 27 (37.0%) 67 (46.2%)
2 (18.2%) 5 (16.1%) 5 (16.7%) 10 (13.7%) 22 (15.2%)
1 (9.1%) 3 (9.7%) 1 (3.3%) 9 (12.3%) 14 (9.7%)
1 (9.1%) 2 (6.5%) 2 (6.7%) 4 (5.5%) 9 (6.2%)
0 6 (19.4%) 4 (13.3%) 23 (31.5%) 33 (22.8%)
11 (7.6%) 31 (21.4%) 30 (20.7%) 73 (50.3%) 145 (100%)
Abbreviations: O, occipital; C, cervical; T, Thoracic; L, Lumbar; S, Sacral.
A striking difference in the analysis of the type of spinal injuries was the observation that 68% (277 of 406) of the entire cohort had muscular sprains without any associated vertebral column or cord injury. The cervical region was the most commonly affected area of all sprains (82%). Likewise, isolated vertebral column or injury to the cord shows a pattern similar to the paravetebral soft tissue injuries, in which the upper cervical region is the most commonly injured level in all age groups (46% of all spinal cord or column injuries). The distributions of levels of paravertebral soft tissue injuries and distributions of levels of isolated vertebral column injuries are not significantly different between the age groups (P ⫽ .60 and P ⫽ .46, respectively). Associated Injuries and Mortality Overall, the most commonly associated injury was craniocerebral injury (37%). However, nearly 58% had no other associated injury (Table 4). Differences exist when separated into paravertebral soft tissue injury and injury to vertebral column or spinal cord. The cohort of spinal injury showed that 42% of the children had one or more system injuries associated with the spinal injury. Seventy-four of those children had solely muscular sprains. Paravertebral injuries generally were not associated with any change in risks of associated injuries, with the exception of cervical sprains, which were noted to be associated with lower risks of most associated injuries. Table 4. Associated Injury Distribution in Spinal Trauma Patients Associated Injury
Patients (%)
Maxillo-facial trauma Craniocerebral injury Genitourinary system Orthopedic injury Gastrointestinal system Respiratory system No other associated injury
40 (9.9) 152 (37.4) 1 (0.2) 64 (15.8) 35 (8.6) 37 (9.1) 235 (57.9)
Overall, 18 children in our subset died, representing 4% of the total population. Spinal injury was the primary cause of death in (6 children), 33% owing to fractures (2 children) and dislocations (4 children) from O-C4. In the 67% (12 children) who died as a result of nonspinal injuries, the majority died as a result of their traumatic brain injuries. Odds Ratios for Associated Injuries: The Effect of Age, Gender and Mechanism of Injury Controlling for mechanism of injury and level of spinal cord, we noted that increasing age was associated with a higher rate of orthopedic injuries, and boys have an increased risk of gastrointestinal system injury (Table 5). Genitourinary system injuries were not evaluated because of the small number of available cases. Pedestrian trauma had a high risk of multisystem injuries. Otherwise, mechanism of injury information in this dataset generally did not have significant association with occurrence of associated injuries. Odds of Associated Injuries: The Effect of Vertebral Column Injury Level Injuries associated with the various levels of vertebral column injuries were evaluated (Table 6). The upper cervical injury had an increased risk of associated orthopedic injury, whereas lower cervical injuries had a high tendency to have associated maxillofacial injury. Thoracolumbar and lumbosacral spine injuries had a higher risk of gastrointestinal system injury, as evidenced by an odds ratio of 35. DISCUSSION
Traumatic spinal injuries in children are uncommon, representing only 1% to 10% of all reported spinal injuries. Each year, approximately 1,000 new spinal injuries are reported in children.15 The true incidence of
610
CIRAK ET AL
Table 5. Odds of Associated Injuries: The Effect of Age, Gender, and Mechanism of Injury Variables Associated Injuries
Age
Boys
MVC
Fall
Sports
Pedestrian
Other
Bike
Diving
GSW
Maxillofacial Craniocerebral Orthopedic injuries Gastrointestinal injuries Respiratory system
— 0.91 1.12
— — —
— — —
0.01 — —
0.01 — —
— 2.71 4.75
— — —
— — —
— (Ref.) —
(Ref.) — (Ref.)
— —
3.08 —
— 4.76
0.05 —
— —
— 11.76
— —
— —
— —
(Ref.) (Ref.)
NOTE. The reference group for comparison was the group of children with gunshot wounds in most cases, which was the smallest group in our dataset. The reference group was changed to children with diving injuries when evaluating the risks of craniofacial injuries, which was the second smallest group, because of zero in one of the cells in the gunshot group when stratified by craniofacial injuries. Only statistically significant odds ratio are reported (P ⬍ .05).
TSI may be underestimated because of scene mortality or death in transport. It is estimated that 5% of the catchment of trauma centers caring for children less than 16 years of age will consist of vertebral column or spinal cord injuries.16-19 In our study, spinal injury including muscle sprains were 4% of all admissions to the pediatric trauma service. Similarly, consistent with the literature, we found that adolescent boys were most affected, yet their ISS was much less then other age groups. There is progressive change in anatomic and physiologic structure of the childhood vertebral column and spinal cord, which necessitates unique and specific evaluation of each age group. The upper cervical region (O-C4) is the most mobile portion of the cervical spine. Children have many physiologic characteristics that predispose them to spinal injury. Among these are a disproportionately large head, underdeveloped neck musculature and relatively vertical facet joints, ligamentous laxity, shallow angulations of the facet joints, and incomplete ossification of the vertebral bodies.9,10,12,20,21 This was corroborated in our findings. Specifically, 46% of vertebral column or spinal cord injuries in this study were localized to the upper cervical region. The level of injury and incidence of neurologic compromise varied with the age of the child. Infants, toddlers, school age children, and adolescents have distinctive anatomic and biomechanical spinal features and behavioral patterns that distinguish them not only from Table 6. Odds of Associated Injuries: The Effect of Vertebral Column Injury Level, Controlling for Age, Gender, and Presence of Paravetebral Injuries Associated Injuries
O-C4
C5-C7
T1-T10
T11-L1
L2-Sacral
Maxillofacial Craniocerebral Orthopedic injuries Gastrointestinal injuries Respiratory system
— — 2.31
3.21 — —
— — —
— — —
— — 4.94
—
—
—
35.21
6.73
—
—
—
—
—
NOTE. Odds ratios are relative to patients without spinal column injury in that level. Only statistically significant odds ratio are reported (P ⬍ .05).
adults, but also from each other.22 In infants, spinal trauma was most commonly the result of MVCs (71%). It is suggested that infants are otherwise unlikely to sustain significant TSIs except as the result of traumatic birth, although the true incidence of birth related TSI is unknown.24 Beyond the newborn period, the patterns of TSI are determined by the specific biomechanics of the spine for that age group. These biomechanical events lead to a specific type of injury, for example, fractures were very rare. Disproportionate size of the head leads to a tendency for upper cervical spine and craniovertebral junction injuries to be the most severe. Atlanto-occipital dislocations occur primarily and represent a significant cause of death.11,26 Mechanism of injury can be stratified further by age. Falls were the most common cause of TSI in toddlers and the school age group. Toddlers and preschool-aged children have proportionally larger heads, a higher center of gravity, and decreased coordination than older children, which predisposes them to falls. In contrast to the toddler and school-aged children, sports-related injuries were the most common injuries in adolescents (20% of all injuries). Specifically, football injuries were the leading cause of sports-related injuries (38%). Adolescents have physical characteristics that are similar to adults.23,28-32 But their different behavioral characteristics were an explanation for the fact that the leading mechanism of spinal injury was sports-related injuries followed by MVCs. All gunshot wounds were in the adolescent age group, a reflection of the unfortunate fact of the prevalence of youth violence in our urban catchment. Spinal injury may also occur in the absence of a bony abnormality. This entity has been defined as SCIWORA.33 The estimated incidence of SCIWORA varies widely from 8% to 50%.34-36 This variation depends on the ever-changing definition of SCIWORA. When Pang and Wilberger33 originally defined SCIWORA, magnetic resonance imaging (MRI) did not exist; therefore, SCIWORA was defined as spinal cord injury without evidence on plain x-ray. With advances in CT and MRI, the
SPINAL INJURIES IN CHILDREN
definition evolved to include these imaging techniques.37,38 Some investigators have noted that the term SCIWORA has become a misnomer because most patients actually have a radiographic abnormality detectable on MRI.39-43 As imaging technology continues to improve, patients with a cord injury that is not detectable by any radiographic study may become exceedingly rare. Therefore, the term SCIWORA should be defined each time a study is conducted and include the specific imaging. In this study, SCIWORA has been defined in 2 different ways and in either case our incidence is less than 10%. Spinal trauma frequently occurs in association with multisystem trauma.13,44 We found that pedestrian-related injuries are associated with multisystem injuries. Concomitant injuries may mask the spinal injury. For example, craniocerebral injury may prevent or delay diagnosis and treatment of spinal injury and frequently is complicated by orthopedic injuries as well. Therefore, it is mandated that all pediatric trauma patients must be treated with a high index of suspicion for spinal injury until definitively proven otherwise. In our study, 42% of children had concomitant-associated injury. All spinal injury deaths in this review were associated with 1 or more system injury. Traumatic brain injury (37%) was the most common associated injury. Older children were noted to be less likely to suffer from craniofacial injuries but more likely to suffer from orthopedic injury. Boys were more likely to suffer from gastrointestinal injuries. Controlling for age, gender, and levels of spinal column and paraverterbral injuries, motor vehicle crashes were associated with increased risks of respiratory system injuries, and pedestrian injuries were associated with increased risks of craniofacial, orthopedic, and respiratory system injuries. Falls were associated with decreased risk of maxillofacial injuries and gastrointestinal injuries. Sports-related injuries were associated with a decreased risk of maxillofacial injuries. High cervical (O-C4) level of spinal column injuries were associated with an increased risk of orthopedic injury, whereas lower cervical (C5-C7) spinal column injuries were associated with an increased risk of maxillofacial injury. Thoracic levels of spinal column injuries were generally not noted to be associated with any change in risks of associated injuries. However, injuries in the lower thoracic region to upper lumbar region (T11-L1) were noted to be associated with significant increase in risk of gastrointestinal injury. Injuries in lumbar and sacral regions (L2-sacral) were noted to be associated with risks of orthopedic injuries and gastrointestinal injuries, although the magnitude of the latter is much lower than T11-L1 injuries. Injuries in the lumbosacral region were associated with significant risks of associated gastrointestinal injuries, and a thorough workup should be con-
611
sidered if radiologic findings suggest injuries in this region. The seemingly protective effect of paravetebral column injuries is probably owing to bias in the recording of injuries in less injured patients. In severely injured patients, the presence of minor sprains were more than likely not recorded, leaving the recording of such injuries only in patients with minor injuries—and thus lower risks of associated injuries. Given these findings, it is imperative that trauma patients be fully evaluated for multisystem injuries. The decreased risk of associated craniocerebral injuries may be related to the reduction in the surface area of the head relative to the body as a child grows in age. The increase in risks of associated orthopedic injuries may be related to the increase in activity and mobility of the child. The independent effect of male gender on risks of associated gastrointestinal injuries is provocative yet cannot be readily explained. It is consistent with reports in the trauma literature that boys have worse outcomes than girls. Perhaps the gastrointestinal system is the body system that drives the gender differences in trauma outcome. This may be an area of interest for future research. In this study, mechanism of injury information was generally not as useful in predicting associated injuries. For example, falls and sports were not associated with changes in risks of orthopedic injuries, even though age was associated with orthopedic injuries. We speculate that the crude categorization of injury mechanism itself does not include such important information as the magnitude of the energy transfers. The importance of the energy information can be indirectly surmised by the association of age with orthopedic injuries, because older children are more likely to be involved in more severe sports injuries. Thus, the lack of association of mechanism of injury with orthopedic injuries— but the positive association of age with orthopedic injuries—suggest the importance of energy level in risks of associated injuries. Therefore, the crude categorization of injury mechanisms was generally not helpful in identifying associated injuries, unless one can further identify the energy level of the injuries. These points underscore the significance of careful and accurate documentation regarding specifics of the injury event from prehospital through emergency department management. The mortality rate in our study was 4%. As is reported in the literature, associated injuries contribute to the majority of the deaths.9-11,45 Almost all of the nonsurvivors (94%) were children with combined spinal and craniocerebral injury. The high mortality rate (27%) in chest and respiratory system association was thought to be increased by respiratory failure caused by high cervical spinal cord injuries. The majority of deaths (67%) occurred in the upper cervical injury group.
612
CIRAK ET AL
Spinal injuries in children have unique characteristics differentiated by age, mechanism, paravetebral verses bony injury, and the presence of associated injuries. For example, thoraco-lumbar or lumbo-sacral spine injured patient should be evaluated for an associated gastrointestinal injury. Moreover, the high incidence of associated injuries, especially craniocerebral injury, inability to
articulate and behavioral patterns complicate the diagnosis of TSI in children. All these factors combine to warrant scene spine immobilization and judicious care in the initial posttrauma period to potentially decrease morbidity and mortality. In addition, registry-based surveillance systems should continue to be utilized to better understand injury patterns.
REFERENCES 1. Garcia RA, Gaeblar-Spira D, Sisung C, et al: Functional improvement after pediatric spinal cord injury. Am J Physic Med Rehabil 81:458-463, 2002 2. Bliss D, Silen M: Pediatric thoracic trauma. Crit Care Med 30:409-415, 2002 3. Ruge JR, Sinson GP, McLone DG, et al: Pediatric spinal injury: The very young. J Neurosurg 68:25-30, 1988 4. Cramer KE: Orthopedic aspects of child abuse. Pediatr Clin North Am 43:1035-1051, 1996 5. Rooks VJ, Sisler C, Burton B: Cervical spine injury in child abuse: Report of two cases. Pediatr Radiol 28:193-195, 1998 6. Ghatan S, Ellenbogen RG: Pediatric spine and spinal cord injury after inflicted trauma. Neurosurg Clin North Am 13:227-233, 2002 7. Nitecki S, Moir CR: Predictive factors of the outcome of traumatic cervical spine fracture in children. J Pediatr Surg 29:1409-1411, 1994 8. Dowd MD, Keenan HT, Bartton SL: Epidemiology and prevention of childhood injuries. Crit Care Med 30:385-392, 2002 9. Roche C, Carty H: Spinal trauma in children. Pediatr Radiol 31:677-700, 2001 10. Akbarnia BA: Disorders of the pediatric and adolescent spine. Orthop Clin North Am 30:521-536, 1999 11. Reakate HL, Theodore N, Sonntag VKH, et al: Pediatric spine and spinal cord cord trauma. State of the art for the third millennium. Child’s Nerv Syst 15:743-750, 1999 12. Kokoska ER, Keller MS, Rallo MC, et al: Characteristics of pediatric cervical spine injuries. J Pediatr Surg 36:100-105, 2001 13. Vicellio P, Simon H, Pressman BD, et al: A prospective multicenter study of crvical spine injury in children. Pediatrics 108:E20, 2001 14. Turgut M, Akpinar G, Akalan N, et al: Spinal injuries in the pediatric age group: A review of 82 cases of spinal cord and vertebral column injuries. Eur Spine 5:148-152, 1996 15. Hu R, Mustard CA, Burns C: Epidemiology of incident spinal fracture in a complete population. Spine 21:492-499, 1996 16. Krug EG, Sharma GK, Lozano R: The global burden of injuries. Am J Public Health 90:253, 2000 17. Hamilton MG, Myles ST: Pediatric spinal injury: Review of 174 hospital admissions. J Neurosurg 77:700-704, 1992 18. Haffner DL, Hoffer MM, Wiedbusch R: Etiology of children’s spinal injuries at RanchoLosAmigos. Spine 18:679-684, 1993 19. Anderson JM, Shutt AH: Spinal injury in children: A review of 156 cases seen from 1950 through 1978. Mayo Clin Proc 55:499-504, 1980 20. Patel JC, Tepas JJ III, Molitt DL, et al: Pediatric cervical spine injuries: Defining the disease. J Pediatr Surg 36:373-376, 2001 21. Brown LB, Brunn AM, Garcia VF: Cervical spine injuries in children: A review of 103 patients treated consecutively at a level 1 pediatric trauma center. J Pediatr Surg 36:1107-1114, 2001 22. Sledge JB, Allred D, Hyman J: Use of Magnetic resonance imaging in evaluating injuries to the pediatric thoracolumbar spine. J Pediatr Orthop 21:288-293, 2001 23. Massagli TL, Jaffe KM: Pediatric spinal cord injury: Treatment and outcome. Pediatrician 17:244-254, 1990 24. Franken EA Jr: Spinal cord injury in the newborn infant. Pediatr Radiol 3:101-104, 1975
25. Leventhal H: Birth injuries of the spinal cord. J Pediatr 56:447, 1969 26. Subach BR, McLaughlin MR, Albright AL, et al: Current management of pediatric atlantoaial rotatory subluxation. Spine 23: 2174-2179, 1998 27. Chang BA, Ebel BE, Rivara FP: Child passenger safety: Potential impact of the Washington State booster seat law on childcare centers. Inj Prev 8:284-288, 2002 28. Osenbach RK, Menezes AH: Pediatric spinal cord and vertebral column injury. Neurosurgery 30:385-390, 1992 29. Reynolds R: Pediatric spinal injury. Cur Opin Pediatr 12:67-71, 2000 30. Wang YM, Kim A, Griffith PM, et al: Injuries from falls in the pediatric population: An analysis of 729 Cases. J Pediatr Surg 36:15281534, 2001 31. Givens TG, Polley KA, Smith GF, et al: Pediatric cervical spine injury. A three-year experience. J Trauma: Inj Inf Crit Care 41:310314, 1996 32. Baker C, Kadish H, Schunk JE: Evaluation of pediatric cervical spine injuries. Am J Emerg Med 17:230-234, 1999 33. Pang D, Wilberger JE: Spinal cord injury without radiographic abnormalities in children. J Neurosurg 57:114-129, 1982 34. Kriss VM, Kriss TC: SCIWORA (Spinal Cord Injury Without Radiographic Abnormality) in infants and children. Clin Pediatr 35: 119-124, 1996 35. Hendey MN: Spinal cord injury without radiographic abnormality: Results of the National Emergency X-Radiography Utilization Study in Blunt Cervical Trauma. J Trauma 53:1-4, 2002 36. Hadley MN: Spinal cord injury without radiographic abnormality. Neurosurgery 50(suppl):100-104, 2002 37. Grabb PA, Pang D: Magnetic resonance imaging in the evaluation of spinal cord injury without radiographic abnormality in children. Neurosurgery 35:406-414, 1994 38. Marinier M, Rodts MF, Connolly M: Spinal cord injury without radiographic abnormality (SCIWORA). Orthop Nurs 16:57-63; quiz 64-65, 1997 39. Gupta SK, Rajeev K, Khosla VK, et al: Spinal cord injury without radiographic abnormality in adults. Spinal Cord 37:726-729, 1999 40. Demetriades D, Charalambides K, Chahwan S, et al: Nonskeletal cervical spine injuries: Epidemiology and diagnostic pitfalls. J Trauma 48:724-727, 2000 41. Hadley MN: Management of pediatric cervical spine and spinal cord injuries. Neurosurgery 50(suppl):85-99, 2002 42. Elekraky MA, Theodore N, Adams M, et al: Pediatric cervical spine injuries: Report of 102 cases and review of the literature. J Neurosurg 92(suppl):12-17, 2000 43. Bonadio WA: Cervical spine trauma in children: I. General concepts, normal anatomy, radiographic evaluation. Am J Emerg Med 11:158-165, 1993 44. Apple DF Jr, Anson CA, Hunter JD, et al: Spinal cord injury in youth. Clin Pediatr 34:90-95, 1995 45. Curran C, Dietrich AM, Bowman MJ, et al: Pediatric cervicalspine immobilization. Achieving neutral position? J Trauma: Inj, Inf, Crit Care 39:729-732, 1995
Background/Purpose: Traumatic spinal injury (TSI) is an uncommon source of morbidity and mortality in children. The aim of this study was to describe childhood TSI in a single level 1 urban pediatric trauma center. Methods: The authors retrospectively analyzed all children younger than 14 years with TSI, treated at a level I pediatric trauma center between 1991 and 2002 (n ⫽ 406, 4% total registry). All children were stratified according to demographics, mechanisms, type and level of injury, radiologic evaluations, associated injuries, and mortality. Results: The mean age was 9.48 ⫾ 3.81 years. The most common overall mechanism of injury was motor vehicle crash (MVC; 29%) and ranked highest for infants. Falls ranked highest for ages 2 to 9 years. Sports ranked highest in the 10 to 14 year age group. Paravertebral soft tissue injuries were
I
NJURIES to the spine, although relatively rare, contribute to significant morbidity and mortality in children. The reported incidence of pediatric traumatic spinal injury (TSI) varies, and most of the previous studies are limited because of the sparse attention to both the thoracic and lumbosacral region.1,2 Limitations of the definitive prevalence studies are that birth injuries and child abuse cases are underreported.3-6 Compared with adults, spinal trauma is relatively rare in pediatric patients. But the mortality rate is higher in children as a result of associated injuries.7,8 There are certain anatomic and biomechanical differences between the immature spine of pediatric patients and adults. These include a greater mobility of the spine owing to ligamentous laxity, shallow angulations of facet joints, immature development of neck musculature, and incomplete ossification of the vertebrae.9 As a result of these differences, 60 to 80% of all pediatric vertebral injuries are in the cervical region. This is in contrast to adults in whom cervical injuries constitute only 30 to 40% of all vertebral injuries.10 Consequently, our overall understanding of characteristics of pediatric TSI is incomplete. The aim of this retrospective review was to identify demographic characteristics of pediatric spinal injury with regard to mechanism, type, level, associated injuries, and mortality in an urban trauma center. MATERIALS AND METHODS After institutional IRB approval at the Johns Hopkins Hospital, a level I urban trauma center, data were obtained from the pediatric Journal of Pediatric Surgery, Vol 39, No 4 (April), 2004: pp 607-612
68%. The most common injury level was the high cervical spine (O-C4). The incidence of spinal cord injury without radiologic abnormality (SCIWORA) was 6%. Traumatic brain injury (37%) was the most common associated injury. Overall mortality rate was 4% in this urban catchment. Conclusions: TSI in children requires a different preventive and therapeutic logarithm compared with that of adults. The potential devastating nature of TSI warrants that the health care team always maintains a high index of suspicion for injury. Future prospective studies are needed to further elucidate injury patterns. J Pediatr Surg 39:607-612. © 2004 Elsevier Inc. All rights reserved. INDEX WORDS: Pediatric trauma, spinal injury, spinal cord injury without radiologic abnormality, abuse.
trauma registry, a comprehensive database of approximately 12,000 consecutively injured children from 1991 to 2002. All children age 14 and younger with spinal trauma diagnoses, defined by International Classification of Diseases 9th Revision, Clinical Modification (ICD-9-CM) codes ranging from 805.0 to 806.9 and also 839.0 to 839.9 and 847.0 to 847.9 were included in this review. In addition, ICD-9 codes 952.0 to 952.9 were utilized to identify cases of spinal cord injury without radiologic abnormality (SCIWORA). Mechanisms of injury were classified by ICD-9 external injury codes. The spinal column was divided into 5 different regions according to anatomic and physiologic differences in each level to facilitate analysis. The regions include occipital to cervical vertebrae (O-C4), lower cervical (C5-C7), upper thoracic (T1-T10), thoraco-lumbar (T11-L1), and lumbo-sacral (L2-sacrum). Statistical analysis was performed using STATA software (College Station, TX). Comparisons between groups were tested using analysis of variance (ANOVA) for continuous outcome variables, with 2 tests for categorical outcome variables and with Kruskal-Wallis tests for differences between group medians. Odds of associated injuries were calculated using forward stepwise regressions, beginning with empty models and adding age, gender, mechanism of injury, level of any spinal column injuries, and level of any paravertebral injuries. The rationale for the selection of these variables was to include only
From the Pediatric Division, Department of Neurosurgery; the Department of Pediatric Surgery; and the Department of Health Policy and Management, Bloomberg School of Public Health, Johns Hopkins Medical Institutions, Baltimore, MD. Address reprint requests to Charles N. Paidas MD, FACS, FAAP, Department of Pediatric Surgery, 600 North Wolfe St, CMSC 7-116, Baltimore, MD, 21287. © 2004 Elsevier Inc. All rights reserved. 0022-3468/04/3904-0021$30.00/0 doi:10.1016/j.jpedsurg.2003.12.011 607
608
CIRAK ET AL
Table 1. Demographics Patients (yr)
Male
Female
Total (%)
Median Final ISS
Number With Final ISS Greater Than 15 (%)
Infants (0-1) Toddlers (2-5) School (6-9) Adolescent (10-14) Total
8 37 50 164 259
9 17 40 81 147
17 (4) 54 (13) 90 (22) 242 (60) 406
4.5 3 2 1 2
4 (23.5) 11 (20.4) 14 (15.6) 25 (10.2) 54 (13.3)
variables that would be available during the initial workup of a patient. Therefore, other variables that may have statistically significant relationship to risks of associated injuries but would not be available on initial patient workup (for example, Injury Severity Score) were not included. P values for inclusion to and exclusion from the models were set at .05, and the likelihood-ratio test was used to test for significance.
RESULTS
Five characteristics were used to evaluate the children with spinal injuries. The definitions of patient groups, etiologies, injury levels, SCIWORA, and associated injuries were standard well accepted literature parameters used to evaluate injury data.7,8,11-14 Demographics A total of 406 patients met our inclusion criteria (Table 1). The mean age was 9.48 ⫾ 3.81 years (2 months to 14 years), and approximately two thirds of the population were boys. The children in the youngest age groups had higher mean and median ISS than older children. Moreover, they have the lowest proportion of severe injuries (ISS ⬎ 15), reflecting very different distribution of patient injury severity levels between the different age groups. Considering the nonlinear and noncontinuous nature of ISS, the group medians (instead of the means) were selected for statistical analysis. Using Kruskal-Wallis test, it was found that the differences in injury severity were not significantly different between age groups (P ⫽ .06). In contrast, when the group means were used and tested with ANOVA, the differences in injury severity between age groups were highly significant (P ⬍ .01). This discrepancy is a reflection of the very different distributions of childhood injury severity levels between the different age groups.
Injury Mechanisms Across all age groups, motor vehicle collisions (MVC) accounted for the majority of injuries (29%; Table 2). However, the distributions of mechanisms of injury are significantly different between the different age groups (P ⬍ .01). Demonstrating the common exposure of this age group, MVCs were the leading cause of injury in infants (71%). Falls were the leading cause of spinal trauma in toddler and school age, 48% and 34%, respectively. The leading mechanisms in the adolescent group were sports-related injuries followed by MVCs. Adolescents accounted for all gunshot wounds, demonstrating the exposure of this age group to this mechanism. Injury Level To facilitate evaluation, we divided spinal injuries of the 4 age cohorts into 2 categories (Table 3A & 3B). The first category includes children with paravertebral soft tissue injury (defined as muscular sprains), and the second group consists of children with only injury to vertebral column or spinal cord. Included in this cohort (Table 3a) is the SCIWORA group. The definition of SCIWORA has evolved with advances in imaging; currently, there are 2 working definitions. One definition is children with positive neurologic findings and negative plain x-ray and computed tomography (CT) scan but pathologic spinal cord MR imaging. According to this definition, 24 children (6%) had SCIWORA. The second definition is defined as abnormal neurologic examination with normal imaging (including MR); our SCIWORA incidence according to this definition decreases to 5 children (1%).
Table 2. Injury Mechanisms Etiology
MVC (%)
Fall (%)
Sports (%)
Pedestrian (%)
Other Collisions*
Bike (%)
Diving (%)
Gunshot (%)
Total
Infant Toddlers School Adolescent Total (%)
12 (71) 20 (37) 26 (29) 59 (24) 117 (29)
2 (12) 26 (48) 31 (34) 45 (18) 104 (26)
0 0 9 (10) 72 (29) 81 (20)
0 6 (11) 15 (17) 21 (9) 42 (10)
3† (18) 2 (4) 4 (4) 22 (9) 31 (8)
0 0 4 (4) 13 (5) 17 (4)
0 0 1 (1) 7 (3) 8 (2)
0 0 0 6 (2) 6 (1)
17 54 90 245 406
NOTE. MVC: E810-825 (.1) (Excludes 821.1-all terrain vehicle), sports: 917.0, 917.5, 886.0 (excludes bike and diving), fall: E880.0-E888 (excludes 886.0), pedestrian: E810-825 (0.7 only), other collisions: E916-E928 (excluding 917.0 and 917.5), bike: 826.-E829, diving: approximately 883.0885.0, gunshot wounds: E922.0 ⫹ .9 and E965.0-965.4, and 985.0-985.4. *“Other collisions” includes struck accidentally by falling object, caught between objects. †One child (2-month-old boy) was a victim of child abuse.
SPINAL INJURIES IN CHILDREN
609
Table 3. Injury Level a. Paravertebral Soft Tissue Injury Injury Level
Cervical
Thoracic
Lumbar
Total
Infant Toddler School Adolescent Total (%)
5 (83.3%) 26 (90.0%) 48 (77.4%) 149 (82.8%) 228 (82.3%)
1 (16.7%) 3 (10.3%) 8 (12.9%) 22 (12.2%) 34 (12.3%)
0 0 6 (9.7%) 9 (5.0%) 15 (5.4%)
6 29 62 180 277(100%)
b. Injury to Vertebral Column or Spinal Cord Injury Level
O-C4
C5-7
T1-10
T11-L1
L2-S
Total
Infant Toddler School Adolescent Total (%)
7 (63.6%) 15 (48.4%) 18 (60.0%) 27 (37.0%) 67 (46.2%)
2 (18.2%) 5 (16.1%) 5 (16.7%) 10 (13.7%) 22 (15.2%)
1 (9.1%) 3 (9.7%) 1 (3.3%) 9 (12.3%) 14 (9.7%)
1 (9.1%) 2 (6.5%) 2 (6.7%) 4 (5.5%) 9 (6.2%)
0 6 (19.4%) 4 (13.3%) 23 (31.5%) 33 (22.8%)
11 (7.6%) 31 (21.4%) 30 (20.7%) 73 (50.3%) 145 (100%)
Abbreviations: O, occipital; C, cervical; T, Thoracic; L, Lumbar; S, Sacral.
A striking difference in the analysis of the type of spinal injuries was the observation that 68% (277 of 406) of the entire cohort had muscular sprains without any associated vertebral column or cord injury. The cervical region was the most commonly affected area of all sprains (82%). Likewise, isolated vertebral column or injury to the cord shows a pattern similar to the paravetebral soft tissue injuries, in which the upper cervical region is the most commonly injured level in all age groups (46% of all spinal cord or column injuries). The distributions of levels of paravertebral soft tissue injuries and distributions of levels of isolated vertebral column injuries are not significantly different between the age groups (P ⫽ .60 and P ⫽ .46, respectively). Associated Injuries and Mortality Overall, the most commonly associated injury was craniocerebral injury (37%). However, nearly 58% had no other associated injury (Table 4). Differences exist when separated into paravertebral soft tissue injury and injury to vertebral column or spinal cord. The cohort of spinal injury showed that 42% of the children had one or more system injuries associated with the spinal injury. Seventy-four of those children had solely muscular sprains. Paravertebral injuries generally were not associated with any change in risks of associated injuries, with the exception of cervical sprains, which were noted to be associated with lower risks of most associated injuries. Table 4. Associated Injury Distribution in Spinal Trauma Patients Associated Injury
Patients (%)
Maxillo-facial trauma Craniocerebral injury Genitourinary system Orthopedic injury Gastrointestinal system Respiratory system No other associated injury
40 (9.9) 152 (37.4) 1 (0.2) 64 (15.8) 35 (8.6) 37 (9.1) 235 (57.9)
Overall, 18 children in our subset died, representing 4% of the total population. Spinal injury was the primary cause of death in (6 children), 33% owing to fractures (2 children) and dislocations (4 children) from O-C4. In the 67% (12 children) who died as a result of nonspinal injuries, the majority died as a result of their traumatic brain injuries. Odds Ratios for Associated Injuries: The Effect of Age, Gender and Mechanism of Injury Controlling for mechanism of injury and level of spinal cord, we noted that increasing age was associated with a higher rate of orthopedic injuries, and boys have an increased risk of gastrointestinal system injury (Table 5). Genitourinary system injuries were not evaluated because of the small number of available cases. Pedestrian trauma had a high risk of multisystem injuries. Otherwise, mechanism of injury information in this dataset generally did not have significant association with occurrence of associated injuries. Odds of Associated Injuries: The Effect of Vertebral Column Injury Level Injuries associated with the various levels of vertebral column injuries were evaluated (Table 6). The upper cervical injury had an increased risk of associated orthopedic injury, whereas lower cervical injuries had a high tendency to have associated maxillofacial injury. Thoracolumbar and lumbosacral spine injuries had a higher risk of gastrointestinal system injury, as evidenced by an odds ratio of 35. DISCUSSION
Traumatic spinal injuries in children are uncommon, representing only 1% to 10% of all reported spinal injuries. Each year, approximately 1,000 new spinal injuries are reported in children.15 The true incidence of
610
CIRAK ET AL
Table 5. Odds of Associated Injuries: The Effect of Age, Gender, and Mechanism of Injury Variables Associated Injuries
Age
Boys
MVC
Fall
Sports
Pedestrian
Other
Bike
Diving
GSW
Maxillofacial Craniocerebral Orthopedic injuries Gastrointestinal injuries Respiratory system
— 0.91 1.12
— — —
— — —
0.01 — —
0.01 — —
— 2.71 4.75
— — —
— — —
— (Ref.) —
(Ref.) — (Ref.)
— —
3.08 —
— 4.76
0.05 —
— —
— 11.76
— —
— —
— —
(Ref.) (Ref.)
NOTE. The reference group for comparison was the group of children with gunshot wounds in most cases, which was the smallest group in our dataset. The reference group was changed to children with diving injuries when evaluating the risks of craniofacial injuries, which was the second smallest group, because of zero in one of the cells in the gunshot group when stratified by craniofacial injuries. Only statistically significant odds ratio are reported (P ⬍ .05).
TSI may be underestimated because of scene mortality or death in transport. It is estimated that 5% of the catchment of trauma centers caring for children less than 16 years of age will consist of vertebral column or spinal cord injuries.16-19 In our study, spinal injury including muscle sprains were 4% of all admissions to the pediatric trauma service. Similarly, consistent with the literature, we found that adolescent boys were most affected, yet their ISS was much less then other age groups. There is progressive change in anatomic and physiologic structure of the childhood vertebral column and spinal cord, which necessitates unique and specific evaluation of each age group. The upper cervical region (O-C4) is the most mobile portion of the cervical spine. Children have many physiologic characteristics that predispose them to spinal injury. Among these are a disproportionately large head, underdeveloped neck musculature and relatively vertical facet joints, ligamentous laxity, shallow angulations of the facet joints, and incomplete ossification of the vertebral bodies.9,10,12,20,21 This was corroborated in our findings. Specifically, 46% of vertebral column or spinal cord injuries in this study were localized to the upper cervical region. The level of injury and incidence of neurologic compromise varied with the age of the child. Infants, toddlers, school age children, and adolescents have distinctive anatomic and biomechanical spinal features and behavioral patterns that distinguish them not only from Table 6. Odds of Associated Injuries: The Effect of Vertebral Column Injury Level, Controlling for Age, Gender, and Presence of Paravetebral Injuries Associated Injuries
O-C4
C5-C7
T1-T10
T11-L1
L2-Sacral
Maxillofacial Craniocerebral Orthopedic injuries Gastrointestinal injuries Respiratory system
— — 2.31
3.21 — —
— — —
— — —
— — 4.94
—
—
—
35.21
6.73
—
—
—
—
—
NOTE. Odds ratios are relative to patients without spinal column injury in that level. Only statistically significant odds ratio are reported (P ⬍ .05).
adults, but also from each other.22 In infants, spinal trauma was most commonly the result of MVCs (71%). It is suggested that infants are otherwise unlikely to sustain significant TSIs except as the result of traumatic birth, although the true incidence of birth related TSI is unknown.24 Beyond the newborn period, the patterns of TSI are determined by the specific biomechanics of the spine for that age group. These biomechanical events lead to a specific type of injury, for example, fractures were very rare. Disproportionate size of the head leads to a tendency for upper cervical spine and craniovertebral junction injuries to be the most severe. Atlanto-occipital dislocations occur primarily and represent a significant cause of death.11,26 Mechanism of injury can be stratified further by age. Falls were the most common cause of TSI in toddlers and the school age group. Toddlers and preschool-aged children have proportionally larger heads, a higher center of gravity, and decreased coordination than older children, which predisposes them to falls. In contrast to the toddler and school-aged children, sports-related injuries were the most common injuries in adolescents (20% of all injuries). Specifically, football injuries were the leading cause of sports-related injuries (38%). Adolescents have physical characteristics that are similar to adults.23,28-32 But their different behavioral characteristics were an explanation for the fact that the leading mechanism of spinal injury was sports-related injuries followed by MVCs. All gunshot wounds were in the adolescent age group, a reflection of the unfortunate fact of the prevalence of youth violence in our urban catchment. Spinal injury may also occur in the absence of a bony abnormality. This entity has been defined as SCIWORA.33 The estimated incidence of SCIWORA varies widely from 8% to 50%.34-36 This variation depends on the ever-changing definition of SCIWORA. When Pang and Wilberger33 originally defined SCIWORA, magnetic resonance imaging (MRI) did not exist; therefore, SCIWORA was defined as spinal cord injury without evidence on plain x-ray. With advances in CT and MRI, the
SPINAL INJURIES IN CHILDREN
definition evolved to include these imaging techniques.37,38 Some investigators have noted that the term SCIWORA has become a misnomer because most patients actually have a radiographic abnormality detectable on MRI.39-43 As imaging technology continues to improve, patients with a cord injury that is not detectable by any radiographic study may become exceedingly rare. Therefore, the term SCIWORA should be defined each time a study is conducted and include the specific imaging. In this study, SCIWORA has been defined in 2 different ways and in either case our incidence is less than 10%. Spinal trauma frequently occurs in association with multisystem trauma.13,44 We found that pedestrian-related injuries are associated with multisystem injuries. Concomitant injuries may mask the spinal injury. For example, craniocerebral injury may prevent or delay diagnosis and treatment of spinal injury and frequently is complicated by orthopedic injuries as well. Therefore, it is mandated that all pediatric trauma patients must be treated with a high index of suspicion for spinal injury until definitively proven otherwise. In our study, 42% of children had concomitant-associated injury. All spinal injury deaths in this review were associated with 1 or more system injury. Traumatic brain injury (37%) was the most common associated injury. Older children were noted to be less likely to suffer from craniofacial injuries but more likely to suffer from orthopedic injury. Boys were more likely to suffer from gastrointestinal injuries. Controlling for age, gender, and levels of spinal column and paraverterbral injuries, motor vehicle crashes were associated with increased risks of respiratory system injuries, and pedestrian injuries were associated with increased risks of craniofacial, orthopedic, and respiratory system injuries. Falls were associated with decreased risk of maxillofacial injuries and gastrointestinal injuries. Sports-related injuries were associated with a decreased risk of maxillofacial injuries. High cervical (O-C4) level of spinal column injuries were associated with an increased risk of orthopedic injury, whereas lower cervical (C5-C7) spinal column injuries were associated with an increased risk of maxillofacial injury. Thoracic levels of spinal column injuries were generally not noted to be associated with any change in risks of associated injuries. However, injuries in the lower thoracic region to upper lumbar region (T11-L1) were noted to be associated with significant increase in risk of gastrointestinal injury. Injuries in lumbar and sacral regions (L2-sacral) were noted to be associated with risks of orthopedic injuries and gastrointestinal injuries, although the magnitude of the latter is much lower than T11-L1 injuries. Injuries in the lumbosacral region were associated with significant risks of associated gastrointestinal injuries, and a thorough workup should be con-
611
sidered if radiologic findings suggest injuries in this region. The seemingly protective effect of paravetebral column injuries is probably owing to bias in the recording of injuries in less injured patients. In severely injured patients, the presence of minor sprains were more than likely not recorded, leaving the recording of such injuries only in patients with minor injuries—and thus lower risks of associated injuries. Given these findings, it is imperative that trauma patients be fully evaluated for multisystem injuries. The decreased risk of associated craniocerebral injuries may be related to the reduction in the surface area of the head relative to the body as a child grows in age. The increase in risks of associated orthopedic injuries may be related to the increase in activity and mobility of the child. The independent effect of male gender on risks of associated gastrointestinal injuries is provocative yet cannot be readily explained. It is consistent with reports in the trauma literature that boys have worse outcomes than girls. Perhaps the gastrointestinal system is the body system that drives the gender differences in trauma outcome. This may be an area of interest for future research. In this study, mechanism of injury information was generally not as useful in predicting associated injuries. For example, falls and sports were not associated with changes in risks of orthopedic injuries, even though age was associated with orthopedic injuries. We speculate that the crude categorization of injury mechanism itself does not include such important information as the magnitude of the energy transfers. The importance of the energy information can be indirectly surmised by the association of age with orthopedic injuries, because older children are more likely to be involved in more severe sports injuries. Thus, the lack of association of mechanism of injury with orthopedic injuries— but the positive association of age with orthopedic injuries—suggest the importance of energy level in risks of associated injuries. Therefore, the crude categorization of injury mechanisms was generally not helpful in identifying associated injuries, unless one can further identify the energy level of the injuries. These points underscore the significance of careful and accurate documentation regarding specifics of the injury event from prehospital through emergency department management. The mortality rate in our study was 4%. As is reported in the literature, associated injuries contribute to the majority of the deaths.9-11,45 Almost all of the nonsurvivors (94%) were children with combined spinal and craniocerebral injury. The high mortality rate (27%) in chest and respiratory system association was thought to be increased by respiratory failure caused by high cervical spinal cord injuries. The majority of deaths (67%) occurred in the upper cervical injury group.
612
CIRAK ET AL
Spinal injuries in children have unique characteristics differentiated by age, mechanism, paravetebral verses bony injury, and the presence of associated injuries. For example, thoraco-lumbar or lumbo-sacral spine injured patient should be evaluated for an associated gastrointestinal injury. Moreover, the high incidence of associated injuries, especially craniocerebral injury, inability to
articulate and behavioral patterns complicate the diagnosis of TSI in children. All these factors combine to warrant scene spine immobilization and judicious care in the initial posttrauma period to potentially decrease morbidity and mortality. In addition, registry-based surveillance systems should continue to be utilized to better understand injury patterns.
REFERENCES 1. Garcia RA, Gaeblar-Spira D, Sisung C, et al: Functional improvement after pediatric spinal cord injury. Am J Physic Med Rehabil 81:458-463, 2002 2. Bliss D, Silen M: Pediatric thoracic trauma. Crit Care Med 30:409-415, 2002 3. Ruge JR, Sinson GP, McLone DG, et al: Pediatric spinal injury: The very young. J Neurosurg 68:25-30, 1988 4. Cramer KE: Orthopedic aspects of child abuse. Pediatr Clin North Am 43:1035-1051, 1996 5. Rooks VJ, Sisler C, Burton B: Cervical spine injury in child abuse: Report of two cases. Pediatr Radiol 28:193-195, 1998 6. Ghatan S, Ellenbogen RG: Pediatric spine and spinal cord injury after inflicted trauma. Neurosurg Clin North Am 13:227-233, 2002 7. Nitecki S, Moir CR: Predictive factors of the outcome of traumatic cervical spine fracture in children. J Pediatr Surg 29:1409-1411, 1994 8. Dowd MD, Keenan HT, Bartton SL: Epidemiology and prevention of childhood injuries. Crit Care Med 30:385-392, 2002 9. Roche C, Carty H: Spinal trauma in children. Pediatr Radiol 31:677-700, 2001 10. Akbarnia BA: Disorders of the pediatric and adolescent spine. Orthop Clin North Am 30:521-536, 1999 11. Reakate HL, Theodore N, Sonntag VKH, et al: Pediatric spine and spinal cord cord trauma. State of the art for the third millennium. Child’s Nerv Syst 15:743-750, 1999 12. Kokoska ER, Keller MS, Rallo MC, et al: Characteristics of pediatric cervical spine injuries. J Pediatr Surg 36:100-105, 2001 13. Vicellio P, Simon H, Pressman BD, et al: A prospective multicenter study of crvical spine injury in children. Pediatrics 108:E20, 2001 14. Turgut M, Akpinar G, Akalan N, et al: Spinal injuries in the pediatric age group: A review of 82 cases of spinal cord and vertebral column injuries. Eur Spine 5:148-152, 1996 15. Hu R, Mustard CA, Burns C: Epidemiology of incident spinal fracture in a complete population. Spine 21:492-499, 1996 16. Krug EG, Sharma GK, Lozano R: The global burden of injuries. Am J Public Health 90:253, 2000 17. Hamilton MG, Myles ST: Pediatric spinal injury: Review of 174 hospital admissions. J Neurosurg 77:700-704, 1992 18. Haffner DL, Hoffer MM, Wiedbusch R: Etiology of children’s spinal injuries at RanchoLosAmigos. Spine 18:679-684, 1993 19. Anderson JM, Shutt AH: Spinal injury in children: A review of 156 cases seen from 1950 through 1978. Mayo Clin Proc 55:499-504, 1980 20. Patel JC, Tepas JJ III, Molitt DL, et al: Pediatric cervical spine injuries: Defining the disease. J Pediatr Surg 36:373-376, 2001 21. Brown LB, Brunn AM, Garcia VF: Cervical spine injuries in children: A review of 103 patients treated consecutively at a level 1 pediatric trauma center. J Pediatr Surg 36:1107-1114, 2001 22. Sledge JB, Allred D, Hyman J: Use of Magnetic resonance imaging in evaluating injuries to the pediatric thoracolumbar spine. J Pediatr Orthop 21:288-293, 2001 23. Massagli TL, Jaffe KM: Pediatric spinal cord injury: Treatment and outcome. Pediatrician 17:244-254, 1990 24. Franken EA Jr: Spinal cord injury in the newborn infant. Pediatr Radiol 3:101-104, 1975
25. Leventhal H: Birth injuries of the spinal cord. J Pediatr 56:447, 1969 26. Subach BR, McLaughlin MR, Albright AL, et al: Current management of pediatric atlantoaial rotatory subluxation. Spine 23: 2174-2179, 1998 27. Chang BA, Ebel BE, Rivara FP: Child passenger safety: Potential impact of the Washington State booster seat law on childcare centers. Inj Prev 8:284-288, 2002 28. Osenbach RK, Menezes AH: Pediatric spinal cord and vertebral column injury. Neurosurgery 30:385-390, 1992 29. Reynolds R: Pediatric spinal injury. Cur Opin Pediatr 12:67-71, 2000 30. Wang YM, Kim A, Griffith PM, et al: Injuries from falls in the pediatric population: An analysis of 729 Cases. J Pediatr Surg 36:15281534, 2001 31. Givens TG, Polley KA, Smith GF, et al: Pediatric cervical spine injury. A three-year experience. J Trauma: Inj Inf Crit Care 41:310314, 1996 32. Baker C, Kadish H, Schunk JE: Evaluation of pediatric cervical spine injuries. Am J Emerg Med 17:230-234, 1999 33. Pang D, Wilberger JE: Spinal cord injury without radiographic abnormalities in children. J Neurosurg 57:114-129, 1982 34. Kriss VM, Kriss TC: SCIWORA (Spinal Cord Injury Without Radiographic Abnormality) in infants and children. Clin Pediatr 35: 119-124, 1996 35. Hendey MN: Spinal cord injury without radiographic abnormality: Results of the National Emergency X-Radiography Utilization Study in Blunt Cervical Trauma. J Trauma 53:1-4, 2002 36. Hadley MN: Spinal cord injury without radiographic abnormality. Neurosurgery 50(suppl):100-104, 2002 37. Grabb PA, Pang D: Magnetic resonance imaging in the evaluation of spinal cord injury without radiographic abnormality in children. Neurosurgery 35:406-414, 1994 38. Marinier M, Rodts MF, Connolly M: Spinal cord injury without radiographic abnormality (SCIWORA). Orthop Nurs 16:57-63; quiz 64-65, 1997 39. Gupta SK, Rajeev K, Khosla VK, et al: Spinal cord injury without radiographic abnormality in adults. Spinal Cord 37:726-729, 1999 40. Demetriades D, Charalambides K, Chahwan S, et al: Nonskeletal cervical spine injuries: Epidemiology and diagnostic pitfalls. J Trauma 48:724-727, 2000 41. Hadley MN: Management of pediatric cervical spine and spinal cord injuries. Neurosurgery 50(suppl):85-99, 2002 42. Elekraky MA, Theodore N, Adams M, et al: Pediatric cervical spine injuries: Report of 102 cases and review of the literature. J Neurosurg 92(suppl):12-17, 2000 43. Bonadio WA: Cervical spine trauma in children: I. General concepts, normal anatomy, radiographic evaluation. Am J Emerg Med 11:158-165, 1993 44. Apple DF Jr, Anson CA, Hunter JD, et al: Spinal cord injury in youth. Clin Pediatr 34:90-95, 1995 45. Curran C, Dietrich AM, Bowman MJ, et al: Pediatric cervicalspine immobilization. Achieving neutral position? J Trauma: Inj, Inf, Crit Care 39:729-732, 1995