Pediatric Cervical Spine Fractures: Predominately Subtle Presentation ByAnn
M. Dietrich,
Margaret
E. Ginn-Pease,
Columbus, 0 Previous descriptions dren have emphasized cervical vertebra. spectrum
Our experience
viewed.
in chil-
injuries to the upper
suggests
a much wider
of injury. The medical records of all children with
cervical spine fractures tween
of cervical spine fractures
high mortality
January
admitted
to Children’s
1, 1985 and December
The average
Hospital
be-
31, 1989 were
re-
age of the 50 patients
was 11 years
(range, 2.7 to 18.8 years) and 82% were boys. Motor vehiclerelated accidents (54%). sports injuries (18%). and falls (12%) accounted for the majority of the fractures. Twenty-nine patients (58%) had an associated head injury. Fifty percent of
Henry M. Bartkowski,
and Denis R. King
Ohio
communicate their complaints. A significant potential exists for the symptoms of vertebral injury to be misinterpreted or disregarded. This study was undertaken to correlate the anatomic patterns of injury with the clinical presentation, diagnostic evaluation, and results of management in children with fractures of the cervical spine. MATERIALS
AND METHODS
occurred at all levels of the cervical spine. The anatomic site of the injury did not correlate with age. Sixteen patients (32%) died. Of the 34 who survived, only 8 had a persistent
The medical records of all children with cervical spine fractures who were admitted to the Children’s Hospital, Columbus, OH, between January 1, 198.5 and December 31, 1989 were reviewed. Patients with a diagnosis of Trisomy 21, congenital bone abnormalities, or spinal cord injury without radiologic abnormality (SCIWORA) were excluded. Information from prehospital providers, emergency medical service (EMS) personnel at the accident scene, and referral hospitals was reviewed. The clinical presentation, diagnostic imaging studies, and hospital course were evaluated and analyzed. Indices of injury severity were defined by calculating the Glasgow Coma Scale (GCS),’ Trauma Score (TS),5 Pediatric Trauma Score (PTS),” and Abbreviated Injury Severity Score (ISS).’ Student’s t tests and xZanalysis were used to determine the statistical significance between groups at a 9.5% confidence level.
neurological deficit. Children with cervical spine fractures have two distinct patterns of presentation: lethal or intact.
RESULTS
the patients were transported from the accident scene and 44% were interhospital transfers. All patients receiving medical care prior to referral had appropriate cervical spine stabilization. On admission 30% of the patients were unresponsive. Thirty-one children were alert and verbal at the time of evaluation and 30 complained of neck pain and tenderness (97%). Twenty-five of the 31 patients (83%) had no demonstrable neurological deficit on initial physical examination. Lateral cervical spine radiographs were diagnostic in 49 children (98%). A relatively even distribution of fractures
The majority of children with cervical spine fractures presented with no demonstrable neurological deficit. All children with complaints of neck pain and/or tenderness need a complete radiographic evaluation of their cervical spine. Copyright o 1991 by W.B. Saunders Company INDEX WORDS:
Cervical spine fractures,
childhood;
pediat-
ric trauma; neck injuries.
F
RACTURES OF THE cervical spine are relatively uncommon in childhood, with a reported incidence of 1% to 2% in pediatric trauma victims who require hospital admission.‘,’ Previous studies have reviewed the morbidity and mortality of cervical spine fractures in pediatric patients and have emphasized the lethal lesions involving the upper cervical vertebra.3 However, our experience suggests a wider spectrum of injury, and indicates the need for vigilance in the evaluation and management of children with head and neck trauma. When obvious neurological deficits are documented on physical examination, a thorough evaluation of the vertebral column is mandated and spinal injuries are rarely overlooked. Less dramatic complaints of neck pain and tenderness may be more difficult to assess in pediatric patients. Adults perceive that children localize pain poorly and they may lack the ability to accurately Journal of Pediatric Surgery, Vol26, No 8 (August), 1991: pp 995-1000
Of the 50 children included in this study, 62% were male and 38% were female with an average age of 11.1 years (range, 2.7 to 18.7 years). Ninety-six percent of the fractures were caused by blunt trauma; penetrating wounds were observed in two children. The majority of the fractures (54%) occurred in motor vehicle-related incidents (26% passengers, 28% pedestrians). Four victims (8%) had diving accidents, 12% sustained injuries in falls, and sports activities accounted for 18%: football (5), volleyball (l), wrestling (1) baseball (1), and trampoline (1). Forty-five patients received prehospital care. Twenty-three children were transported directly from the accident scene, and 22 were transferred from From the Divisions of Emergency Medicine, Pediatric Surgery, and Neurosurgery, Departments of Surgery and Pediauics. The Ohio State University College of Medicine, and The Children’s Hospital, Columbus, OH. Presented at the Jens G. Rosenkrantz Resident Competition at the 42nd Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Boston, Massachusetts, October 6- 7, 1990. Address reprint requests to Dents R. King, MD, Division of Pediatric Surgery, Children’s Hospital, 700 Children’s Dr, Columbus, OH 43205. Copyright o 1991 by W.B. Saunders Companv 0022-3468/91/2608-0021$03.00/O 995
DIETRICH ET AL
other facilities. All 45 patients who received prehospital care had appropriate cervical spine immobilization. At the time of admission, 15 patients were unresponsive, 4 had varying levels of consciousness, and 31 were alert and verbal. All 15 unresponsive children were injured in motor vehicle-related accidents, 8 as pedestrians, 4 as passengers, and 3 while riding bicycles. The average age of these severely injured patients was 6.6 years, their mean GCS score was 3, and all required ventilatory assistance prior to admission. Resuscitative medications had been administered by EMS personnel to 6 of the 15 children. On initial evaluation all of these children demonstrated evidence of a devastating neurological injury with fixed and dilated pupils, flaccid extremities, and an absence of rectal sphincter tone. The four children who presented with varying levels of consciousness (mean GCS, 9.5) also sustained injuries in motor vehicle accidents (3 passengers and 1 pedestrian). The mean age of these patients was 8.9 years. Supplemental oxygen was provided during transport but no airway intervention was required. On arrival to the emergency department none of these four children were observed to have any focal neurological deficits. Thirty-one patients (mean age, 13.3 yrs) were alert and verbal at the time of their initial evaluation (mean GCS, 15). Thirteen children had been injured in sports and diving activities, 9 were involved in motor vehicle accidents, and 6 sustained their fractures in falls. Subjective complaints of neck pain and tenderness were elicited from 30 of these children (97%). Only three (10%) had a neurological deficit on clinical examination (Fig 1). Lateral cervical spine radiographs were obtained in all 50 patients. On final review, the screening films obtained in the emergency department were diagnos-
occiput
-
Cl
-
C2-
Table 1. Level of Injury, Incidence, Age, GCS at Initial Evaluation, and Mortality Meall Level of Injury
NO.
Mean GCS
NO
8
7.2
C-l
a
9.7
8
c-2
8
11.1
10
c-3
1
14.9
15
c-4
2
9.4
12
c-5
9
14.5
15
C-6
7
12.5
13
C-7
7
10.7
14
Age (vr)
Mortality
(no.)
3
tic in 49 cases (98%). In five children the initial interpretation of the cervical spine films by emergency department physicians or neurosurgical residents was inaccurate and delays in diagnosis resulted. Computed axial tomography was required in only one child to achieve a definitive diagnosis. The anatomic distribution of the fractures is presented in Table 1. The injuries had a uniform distribution throughout all levels of the cervical spine. Although the level of injury was not correlated statistically with age, there was a predilection for the high cervical fractures to occur in young children (Fig 2). All of the patients who sustained atlantooccipital separation were less than 11 years of age and all were involved in motor vehicle-related accidents. The C, fractures occurred predominantly in older children (mean age, 14.5 years) and two thirds of these resulted from sports injuries. Three patterns of injury were recognized: (1) fractures of the vertebral bodies or posterior elements (54%); (2) fractures with subluxation (18%); and (3) distraction injuries (28%) (Table 2). The average age of the children with simple fractures and fracture/subluxations was 12.6 years. The children with distraction of the vertebral bodies were substantially younger (mean age, 6.3 years). The various indexes of injury severity for those children with
occiput -
I
Cl -
=3-
C2-
C4‘ C5-
C4-
c6C7T1 4 2
I
C3-
C5-
,
, 4
.
, 6
c6I
I
,
6
Age
,
10
,
,
12
,
I
14
,
,
16
,
,
,
16
(Years)
Fig 1. Level and pattern of cervical spine injury and age of the patient. Cl, Fracture; 8, fracture/subluxation; n, distraction; *, noncontiguous.
=7T12
4
6
6
10
12
14
16
16
Age (Years) Fig 2. Age of the child as compared with the level of injury and outcome. 0, Survivors; W, nonsurvivors; *, noncontiguous.
PEDIATRIC CERVICAL SPINE FRACTURES
997
Table 2. Age, Scores of Injury Severity, Incidence of Associated Injuries and Outcome for the Different Patterns of Injury Fracture/Subluxation
Fractures
No.
27
Mean age (yr)
12.5
9
15 2 2.0
14 f 4.5
ISS
14 t 10.2
20 f 18.8
Mortality (%) Abbreviations:
6.3*
14 ?I 4.0
TS Associated injuries (%)
14
12.7
14 f 2.7
GCS
Distraction
3 2 o* 3 + 1.5’ 76 + o*
18
22
85”
4
11
100*
GCS Glasgow Coma Scale; TS, Trauma Score; ISS.
Abbreviated Injury Severity Score. *P < .05 for distractions compared with fractures and distractions compared with fracture/subluxation.
fractures, fracture/subluxation, and distraction injuries are presented in Table 2. In the children with distraction of the vertebral bodies, all of the indexes of injury severity were significantly different when compared with the patients with simple fractures or fracture/subluxations. The incidence of associated injuries (85%) and the mortality rate (100%) were also higher (P < .05). Thirty-one children had associated injuries (62%). Twenty-nine patients sustained head injuries of varying severity. Twenty children were victims of multiple trauma: 9 thoracic injuries that required surgical intervention (18%), 8 femur fractures (16%), and 2 clavicular fractures. Thirty-four patients survived and 16 died. Table 3 contrasts the severity of injury scores (GCS, TS, PTS, ISS) for the survivors and nonsurvivors. The children who died were younger and an analysis of the indexes of injury severity between the two groups demonstrated statistical differences in all parameters. At the time of discharge only 6 of the surviving patients (18%) had persistent neurological disabilities. In two of the six children the residual deficits were consistent with their coexisting head injuries. The other four children had a disability related to their spinal fractures (two triceps, one deltoid weakness), and one left upper extremity paresis). Table 3. Age and injury Severity Scores for Nonsurvivors and Survivors Nonsurvivors
No.
16
(vr)
Survivors
34
6.5 ? 2.2
13.0 2 3.5*
GCS
3 k 1.8
14 f 2.3*
TS
42
15 2 2.1’
Age
1.7
PTS
-2 ? 1.6
ISS
71 2 12.6
9 + 2.8* 14 2 9.4’
Abbreviations: GCS, Glasgow Coma Scale, TS, Trauma Score; PTS, Pediatric Trauma Score; ISS, Abbreviated Injury Severity Score. l/J < .05.
DISCUSSION
Fractures of the cervical spine are particularly disasterous lesions that have a significant potential to produce a lifetime of morbidity. Fortunately, these injuries are uncommon in pediatric trauma patients with an estimated incidence of 1% to 2%.‘*’ The 50 patients included in this study represent approximately 1% of the 5,000 trauma admissions to The Children’s Hospital, Columbus, OH during the review period. Boys sustained cervical spine fractures more frequently than girls (M-F ratio 2:1).2,8 Motor vehicle accidents were the predominant cause of spinal trauma in these patients (54%) and they accounted for 88% of the deaths. Fourteen of the 16 children who died were injured by automobiles and 11 of the 14 were either pedestrians or cyclists. Similarily, all 19 deaths reported by Bohn et al were the result of motor vehicle-related accidents and 47% of the fatally injured children were pedestrians.’ In comparison to previous reports from this institution and others, a lower incidence of diving (8%) and football (10%) injuries was observed.1.3 The automobile continues to be the major cause of cervical spine fractures and resultant morbidity and mortality.3.9 The neurological status at presentation clearly defined the outcome in 49 of the 50 children. The nonsurvivors were found to be unresponsive and apneic at the accident scene but with intubation and ventilatory assistance vital signs were maintained during transport. All of these children had evidence of a devastating neurological injury at the time of admission to the emergency department. The level of injury tended to be in the upper cervical spine (88% above C,) and distraction of the vertebral bodies was commonly observed (88%). Most of the children had evidence of significant associated trauma (85%). Bohn et al recently described a similar group of children with high cervical injuries and absent vital signs who had a uniformly fatal outcome.’ Of the 15 patients in this series who were unresponsive on admission, 8 died in the emergency department and 7 were maintained on life support systems until brain death was confirmed in the intensive care unit. The children who survived had subtle clinical presentations. The four patients with altered sensorium had coexisting closed head injuries. These children had immobilization of the cervical spine until their vertebral injuries were identified. Delay in diagnosis did not occur in this subset of patients and none of them developed a neurological disability that was spinal cord related. Sneed and Stover described four cases of spinal cord damage that resulted from
DIETRICH ET AL
998
delays in diagnosis in patients with coexisting head injuries.” The presence of traumatic brain injury in a child demands comprehensive evaluation of the cervical spine. The 31 children who retained their ability to communicate were able to localize the area of injury with accuracy. Ninety-seven percent of the patients who were alert and verbal on admission expressed complaints of neck pain and tenderness. The one exception was a child with a high cervical cord injury who developed progressive deterioration of his respiratory status during transport and required immediate intubation on arrival in the emergency department. Reports detailing the clinical presentation of large cohorts of children with cervical fractures are not available. Rachesky et al reviewed 25 cases and concluded that any child with complaints of neck pain or with head trauma as a result of involvement in a motor vehicle accident had a significant potential for cervical spine injury.’ They indicated that pain alone should mandate a high index of suspicion because the existence of neurological deficits would not clearly identify the patients at risk, and our experience substantiates this point. Of the 31 children in this study who were awake and alert on admission, only 3 had any evidence of a neurological deficit that might have indicated the presence of vertebral injury. Seven of these 31 patients (22%) had an inappropriate delay in the management of their cervical spine fracture. One child was treated by his family physician for 8 months for suspected cervical strain before being referred to a physiatrist who obtained x-rays of the spine. In two children symptoms of neck pain and tenderness were attributed to coexisting injuries, which in both cases were clavicular fractures. Persistent neck pain prompted referral for further evaluation 4 days and 3 months postinjury. Relatively minor trauma produced significant injuries in four children. The lax ligaments, angulated facets and immaturity of the pediatric skeleton predisposes to this problem.3,8.” The majority (86%) of the children with delayed diagnosis sustained apparently trivial injuries in falls on the playground or while playing sports. In the abscence of a history of significant trauma, a thorough radiographic evaluation of the diagnosis of cervical
spine fracture may not be considered and may be delayed.‘2x13 The final source of diagnostic error was an inaccurate assessment of the cervical spine radiographs. Cervical spine x-rays may be more difficult to evaluate in pediatric patients, because many physicians are not familiar with the appearance of the immature skeleton.13,14These x-rays were initially misinterpreted on 5 occasions, 3 at this institution. Of the seven children who had delays in diagnosis, only one had a minor residual neurological deficit at the time of discharge, but the potential for devastating consequences should be emphasized. Although there was a relatively even distribution of fractures throughout all levels of the cervical spine in this series, all of the lethal injuries were observed in the upper cervical region (above C,) (Figs 1 and 2). The simple fractures had a propensity to occur in older children and in the lower cervical vertebrae. None of the children with lower cervical spine injuries (below C,) as a result of blunt impact trauma died and only two had residual neurological deficits. Hadley et al reported a 28% incidence of disability following fractures of the spine, but their report included all levels of the vertebral column (cervical, thoracic, lumbar).8 The outcome of children who sustained cervical spine fractures was accurately predicted by the scores used to assess injury severity (GCS, TS, ISS). Patients with multiple injuries and scores indicative of severe multisystem trauma did not survive. Children with isolated neck injuries who were neurologically intact survived with minimal morbidity. Although limited by retrospective design, this study demonstrates the need for thorough clinical and radiographic evaluation of any child with neck pain. Physicians should not be deterred by histories of insignificant trauma or lack of neurological findings. An adequate examination may require flexion/ extension views and/or computed axial tomography to evaluate the cervical spine. Medical personnel need to be meticulous in their management of any child with neck pain. Morbidity can be minimized with early recognition and appropriate management of cervical spine fractures.
REFERENCES 1. Jaffe DM, Binns H, Radkowski MA, et al: Developing a clinical algorithm for early management of cervical spine injury in child victims. Arm Emerg Med 16270-276, 1987 2. Rachesky I, Boyce WT, Duncan B, et al: Clinical prediction of cervical snine iniuries in children. Am J Dis Child 141:199-201. 1987 * ” 3. Hill SA, Miller CA, Kosnik EJ, et al: Pediatric neck injuries: A clinical study. J Neurosurg 60:700-706,1984
4. Teasdale consciousness. 5. Champion 91672~676,198l
G, Jennett A practical H,
Sacco
B: Assessment of coma and impaired scale. Lancet 2:81-84, 1974 W: Trauma
score.
Crit
Care
Med
6. Tepas J, Mollitt D, Bryant M: The pediatric trauma score as a predictor of injury severity in the injured child. J Pediatr Surg 22:14-18,1987
PEDIATRIC
CERVICAL
SPINE
FRACTURES
999
7. Greenspan L, McClellan B, Greig H: Abbreviated injury scale and injury severity score: A scoring chart. J Trauma 2560-64, 1985 8. Hadley MN, Zabramski JM, Browner CM, et al: Pediatric spinal trauma: Review of 122 cases of spinal cord and vertebral column injuries. J Neurosurg 68:18-24, 1988 9. Bohn D, Armstrong D, Becker L, et al: Cervical spine injuries in children. J Trauma 30:463-469, 1990 10. Sneed RC, Stover SL: Undiagnosed spinal cord injuries in brain-injured children. Am J Dis Child 142:965-967,1988
11. Ruge JR, Sinson GP, McLone DG, et al: Pediatric spinal injury: The very young. J Neurosurg 68:25-30,1988 12. Stauffer ES, Mazur JM: Cervical spine injuries in children. Pediatr Ann 11:502-511,1988 13. Ehara E-K: Cervical spine injury in children: Radiologic manifestations. AJR 152:1175-1178, 1988 14. Apple JS, Kirks DR, Merten DF, et al: Cervical spine fractures and dislocations in children. Pediatrics 17:45-49, 1987
Discussion A. Cooper (New York, NY): This study will add significantly to the growing body of literature on this subject. Unlike other studies, the investigators found that cervical spine fractures are not necessarily limited to the upper cervical spine. They also confirmed that while the SCIWORA syndrome (spinal cord injury without radiographic abnormality) has been much written about in recent years, the overwhelming majority of spinal cord injuries are in fact associated with vertebral column deformities. The results also stress the importance of a complete evaluation of the cervical spine at the time of injury, as 30 or 31 patients who were awake at the time of the presentation in fact had symptoms and signs of injury, while 25 of 31 patients were neurologically normal at the time of admission. I do have two questions: first, of the 16 patients who died, 8 had atlantooccipital distractions while this injury was present in none of the patients who survived. Of what specific injuries did the other patients die? Second, the cervical spine x-rays taken at the time of admission presumably occasionally are misinterpreted. Do you take any precautions to ensure that the cervical spine remains immobilized until the official interpretation is available on the morning following admission? D. Lloyd (Liverpool, England): I am concerned that so much emphasis is placed on the x-ray because, as Dr Cooper mentioned, the SCIWORA syndrome does occur. The critical thing with high-risk trauma is to suspect that the spine is unstable until the patient is conscious and can be properly evaluated. Clinical symptoms even if minimal or bizarre, should not be ignored simply because the cervical x-ray is normal; there is a very real risk of this if too much emphasis is placed on the x-rays. B. Beaver (Baltimore, MD): This is the largest collected series in North America depicting this most serious and very puzzling dilemma. Did management of airway priorities, nasotracheal intubation versus oral-tracheal intubation associated with the index of suspicion of cervical spine injury, affect outcome?
M. Olsen (Cleveland, OH): Do you have any incidence of cervical cord injury without fracture or dislocation in your series? Also, do you have any data regarding the use of seat belts and car seats in these children? h4. Fallat (Louisville, KY): I’m sure you’re aware that there’s a debate in the adult literature about whether or not some patients need evaluation of the cervical spine at all. Specifically, these are patients who have been in some type of traumatic incident but do not have any symptoms, neck tenderness or pain, and no focal neurological deficit. We have a very active emergency room residency program and some of the residents have been interested in trying to do a study in our pediatric population. Do you do just a lateral cervical spine as a screening film in your patients, or do you do a complete series? Do you ever not evaluate a trauma patient with at least a lateral cervical spine film? A.M. Dietrich (response): Dr Cooper, in reference to the 16 patients who died, 8 did have documented atlantooccipital distraction. The autopsies on the other patients (and even on some of the patients that had atlantooccipital distractions) were incomplete and the number of autopsies that were performed was low. Most of these children (85%) had major associated injuries determined from the emergency room record and physical examination. The autopsy information was rather poor, so it is difficult for me to say if there were other causes of death. In terms of our protocol, most of these children who have complaints of neck pain and neck tenderness should be immobilized until the point when you have done a lateral cervical spine, trauma series, and x-rays in flexion and extension, CT scans, and whatever is determined appropriate by the neurosurgical staff at that hospital. In this institution we take very seriously any children who have complaints of neck pain and tenderness and evaluate them thoroughly. Dr Fallat had also addressed the question of how our patients were evaluated. All patients who had a significant injury such as
DIETRICH ET AL
motor vehicle accidents, or children struck by cars, or children with complaints of neck pain and tenderness receive appropriate cervical spine immobilization. After a patient gets a lateral cervical spine series we examine the patient. If there is no tenderness we stop at that point. If there is any tenderness then we proceed on until we have effectively done all of the testing. We excluded all patients with SCIWORA. To my knowledge we’ve only had three cases in the last 2 years. There were 19 patients who required some type
of airway management in the field. Three children were sent to another hospital and there were no cases in which nasotracheal intubation was used. All of the patients were intubated orally and sedated. All of the children were intubated with cervical spine traction being held or cervical spine stabilized. Most of the children that did require intubation were apneic, and a large number of them were pulseless at the scene. It’s difficult to say whether that had any effect on our outcome.
M. Dietrich,
Margaret
E. Ginn-Pease,
Columbus, 0 Previous descriptions dren have emphasized cervical vertebra. spectrum
Our experience
viewed.
in chil-
injuries to the upper
suggests
a much wider
of injury. The medical records of all children with
cervical spine fractures tween
of cervical spine fractures
high mortality
January
admitted
to Children’s
1, 1985 and December
The average
Hospital
be-
31, 1989 were
re-
age of the 50 patients
was 11 years
(range, 2.7 to 18.8 years) and 82% were boys. Motor vehiclerelated accidents (54%). sports injuries (18%). and falls (12%) accounted for the majority of the fractures. Twenty-nine patients (58%) had an associated head injury. Fifty percent of
Henry M. Bartkowski,
and Denis R. King
Ohio
communicate their complaints. A significant potential exists for the symptoms of vertebral injury to be misinterpreted or disregarded. This study was undertaken to correlate the anatomic patterns of injury with the clinical presentation, diagnostic evaluation, and results of management in children with fractures of the cervical spine. MATERIALS
AND METHODS
occurred at all levels of the cervical spine. The anatomic site of the injury did not correlate with age. Sixteen patients (32%) died. Of the 34 who survived, only 8 had a persistent
The medical records of all children with cervical spine fractures who were admitted to the Children’s Hospital, Columbus, OH, between January 1, 198.5 and December 31, 1989 were reviewed. Patients with a diagnosis of Trisomy 21, congenital bone abnormalities, or spinal cord injury without radiologic abnormality (SCIWORA) were excluded. Information from prehospital providers, emergency medical service (EMS) personnel at the accident scene, and referral hospitals was reviewed. The clinical presentation, diagnostic imaging studies, and hospital course were evaluated and analyzed. Indices of injury severity were defined by calculating the Glasgow Coma Scale (GCS),’ Trauma Score (TS),5 Pediatric Trauma Score (PTS),” and Abbreviated Injury Severity Score (ISS).’ Student’s t tests and xZanalysis were used to determine the statistical significance between groups at a 9.5% confidence level.
neurological deficit. Children with cervical spine fractures have two distinct patterns of presentation: lethal or intact.
RESULTS
the patients were transported from the accident scene and 44% were interhospital transfers. All patients receiving medical care prior to referral had appropriate cervical spine stabilization. On admission 30% of the patients were unresponsive. Thirty-one children were alert and verbal at the time of evaluation and 30 complained of neck pain and tenderness (97%). Twenty-five of the 31 patients (83%) had no demonstrable neurological deficit on initial physical examination. Lateral cervical spine radiographs were diagnostic in 49 children (98%). A relatively even distribution of fractures
The majority of children with cervical spine fractures presented with no demonstrable neurological deficit. All children with complaints of neck pain and/or tenderness need a complete radiographic evaluation of their cervical spine. Copyright o 1991 by W.B. Saunders Company INDEX WORDS:
Cervical spine fractures,
childhood;
pediat-
ric trauma; neck injuries.
F
RACTURES OF THE cervical spine are relatively uncommon in childhood, with a reported incidence of 1% to 2% in pediatric trauma victims who require hospital admission.‘,’ Previous studies have reviewed the morbidity and mortality of cervical spine fractures in pediatric patients and have emphasized the lethal lesions involving the upper cervical vertebra.3 However, our experience suggests a wider spectrum of injury, and indicates the need for vigilance in the evaluation and management of children with head and neck trauma. When obvious neurological deficits are documented on physical examination, a thorough evaluation of the vertebral column is mandated and spinal injuries are rarely overlooked. Less dramatic complaints of neck pain and tenderness may be more difficult to assess in pediatric patients. Adults perceive that children localize pain poorly and they may lack the ability to accurately Journal of Pediatric Surgery, Vol26, No 8 (August), 1991: pp 995-1000
Of the 50 children included in this study, 62% were male and 38% were female with an average age of 11.1 years (range, 2.7 to 18.7 years). Ninety-six percent of the fractures were caused by blunt trauma; penetrating wounds were observed in two children. The majority of the fractures (54%) occurred in motor vehicle-related incidents (26% passengers, 28% pedestrians). Four victims (8%) had diving accidents, 12% sustained injuries in falls, and sports activities accounted for 18%: football (5), volleyball (l), wrestling (1) baseball (1), and trampoline (1). Forty-five patients received prehospital care. Twenty-three children were transported directly from the accident scene, and 22 were transferred from From the Divisions of Emergency Medicine, Pediatric Surgery, and Neurosurgery, Departments of Surgery and Pediauics. The Ohio State University College of Medicine, and The Children’s Hospital, Columbus, OH. Presented at the Jens G. Rosenkrantz Resident Competition at the 42nd Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Boston, Massachusetts, October 6- 7, 1990. Address reprint requests to Dents R. King, MD, Division of Pediatric Surgery, Children’s Hospital, 700 Children’s Dr, Columbus, OH 43205. Copyright o 1991 by W.B. Saunders Companv 0022-3468/91/2608-0021$03.00/O 995
DIETRICH ET AL
other facilities. All 45 patients who received prehospital care had appropriate cervical spine immobilization. At the time of admission, 15 patients were unresponsive, 4 had varying levels of consciousness, and 31 were alert and verbal. All 15 unresponsive children were injured in motor vehicle-related accidents, 8 as pedestrians, 4 as passengers, and 3 while riding bicycles. The average age of these severely injured patients was 6.6 years, their mean GCS score was 3, and all required ventilatory assistance prior to admission. Resuscitative medications had been administered by EMS personnel to 6 of the 15 children. On initial evaluation all of these children demonstrated evidence of a devastating neurological injury with fixed and dilated pupils, flaccid extremities, and an absence of rectal sphincter tone. The four children who presented with varying levels of consciousness (mean GCS, 9.5) also sustained injuries in motor vehicle accidents (3 passengers and 1 pedestrian). The mean age of these patients was 8.9 years. Supplemental oxygen was provided during transport but no airway intervention was required. On arrival to the emergency department none of these four children were observed to have any focal neurological deficits. Thirty-one patients (mean age, 13.3 yrs) were alert and verbal at the time of their initial evaluation (mean GCS, 15). Thirteen children had been injured in sports and diving activities, 9 were involved in motor vehicle accidents, and 6 sustained their fractures in falls. Subjective complaints of neck pain and tenderness were elicited from 30 of these children (97%). Only three (10%) had a neurological deficit on clinical examination (Fig 1). Lateral cervical spine radiographs were obtained in all 50 patients. On final review, the screening films obtained in the emergency department were diagnos-
occiput
-
Cl
-
C2-
Table 1. Level of Injury, Incidence, Age, GCS at Initial Evaluation, and Mortality Meall Level of Injury
NO.
Mean GCS
NO
8
7.2
C-l
a
9.7
8
c-2
8
11.1
10
c-3
1
14.9
15
c-4
2
9.4
12
c-5
9
14.5
15
C-6
7
12.5
13
C-7
7
10.7
14
Age (vr)
Mortality
(no.)
3
tic in 49 cases (98%). In five children the initial interpretation of the cervical spine films by emergency department physicians or neurosurgical residents was inaccurate and delays in diagnosis resulted. Computed axial tomography was required in only one child to achieve a definitive diagnosis. The anatomic distribution of the fractures is presented in Table 1. The injuries had a uniform distribution throughout all levels of the cervical spine. Although the level of injury was not correlated statistically with age, there was a predilection for the high cervical fractures to occur in young children (Fig 2). All of the patients who sustained atlantooccipital separation were less than 11 years of age and all were involved in motor vehicle-related accidents. The C, fractures occurred predominantly in older children (mean age, 14.5 years) and two thirds of these resulted from sports injuries. Three patterns of injury were recognized: (1) fractures of the vertebral bodies or posterior elements (54%); (2) fractures with subluxation (18%); and (3) distraction injuries (28%) (Table 2). The average age of the children with simple fractures and fracture/subluxations was 12.6 years. The children with distraction of the vertebral bodies were substantially younger (mean age, 6.3 years). The various indexes of injury severity for those children with
occiput -
I
Cl -
=3-
C2-
C4‘ C5-
C4-
c6C7T1 4 2
I
C3-
C5-
,
, 4
.
, 6
c6I
I
,
6
Age
,
10
,
,
12
,
I
14
,
,
16
,
,
,
16
(Years)
Fig 1. Level and pattern of cervical spine injury and age of the patient. Cl, Fracture; 8, fracture/subluxation; n, distraction; *, noncontiguous.
=7T12
4
6
6
10
12
14
16
16
Age (Years) Fig 2. Age of the child as compared with the level of injury and outcome. 0, Survivors; W, nonsurvivors; *, noncontiguous.
PEDIATRIC CERVICAL SPINE FRACTURES
997
Table 2. Age, Scores of Injury Severity, Incidence of Associated Injuries and Outcome for the Different Patterns of Injury Fracture/Subluxation
Fractures
No.
27
Mean age (yr)
12.5
9
15 2 2.0
14 f 4.5
ISS
14 t 10.2
20 f 18.8
Mortality (%) Abbreviations:
6.3*
14 ?I 4.0
TS Associated injuries (%)
14
12.7
14 f 2.7
GCS
Distraction
3 2 o* 3 + 1.5’ 76 + o*
18
22
85”
4
11
100*
GCS Glasgow Coma Scale; TS, Trauma Score; ISS.
Abbreviated Injury Severity Score. *P < .05 for distractions compared with fractures and distractions compared with fracture/subluxation.
fractures, fracture/subluxation, and distraction injuries are presented in Table 2. In the children with distraction of the vertebral bodies, all of the indexes of injury severity were significantly different when compared with the patients with simple fractures or fracture/subluxations. The incidence of associated injuries (85%) and the mortality rate (100%) were also higher (P < .05). Thirty-one children had associated injuries (62%). Twenty-nine patients sustained head injuries of varying severity. Twenty children were victims of multiple trauma: 9 thoracic injuries that required surgical intervention (18%), 8 femur fractures (16%), and 2 clavicular fractures. Thirty-four patients survived and 16 died. Table 3 contrasts the severity of injury scores (GCS, TS, PTS, ISS) for the survivors and nonsurvivors. The children who died were younger and an analysis of the indexes of injury severity between the two groups demonstrated statistical differences in all parameters. At the time of discharge only 6 of the surviving patients (18%) had persistent neurological disabilities. In two of the six children the residual deficits were consistent with their coexisting head injuries. The other four children had a disability related to their spinal fractures (two triceps, one deltoid weakness), and one left upper extremity paresis). Table 3. Age and injury Severity Scores for Nonsurvivors and Survivors Nonsurvivors
No.
16
(vr)
Survivors
34
6.5 ? 2.2
13.0 2 3.5*
GCS
3 k 1.8
14 f 2.3*
TS
42
15 2 2.1’
Age
1.7
PTS
-2 ? 1.6
ISS
71 2 12.6
9 + 2.8* 14 2 9.4’
Abbreviations: GCS, Glasgow Coma Scale, TS, Trauma Score; PTS, Pediatric Trauma Score; ISS, Abbreviated Injury Severity Score. l/J < .05.
DISCUSSION
Fractures of the cervical spine are particularly disasterous lesions that have a significant potential to produce a lifetime of morbidity. Fortunately, these injuries are uncommon in pediatric trauma patients with an estimated incidence of 1% to 2%.‘*’ The 50 patients included in this study represent approximately 1% of the 5,000 trauma admissions to The Children’s Hospital, Columbus, OH during the review period. Boys sustained cervical spine fractures more frequently than girls (M-F ratio 2:1).2,8 Motor vehicle accidents were the predominant cause of spinal trauma in these patients (54%) and they accounted for 88% of the deaths. Fourteen of the 16 children who died were injured by automobiles and 11 of the 14 were either pedestrians or cyclists. Similarily, all 19 deaths reported by Bohn et al were the result of motor vehicle-related accidents and 47% of the fatally injured children were pedestrians.’ In comparison to previous reports from this institution and others, a lower incidence of diving (8%) and football (10%) injuries was observed.1.3 The automobile continues to be the major cause of cervical spine fractures and resultant morbidity and mortality.3.9 The neurological status at presentation clearly defined the outcome in 49 of the 50 children. The nonsurvivors were found to be unresponsive and apneic at the accident scene but with intubation and ventilatory assistance vital signs were maintained during transport. All of these children had evidence of a devastating neurological injury at the time of admission to the emergency department. The level of injury tended to be in the upper cervical spine (88% above C,) and distraction of the vertebral bodies was commonly observed (88%). Most of the children had evidence of significant associated trauma (85%). Bohn et al recently described a similar group of children with high cervical injuries and absent vital signs who had a uniformly fatal outcome.’ Of the 15 patients in this series who were unresponsive on admission, 8 died in the emergency department and 7 were maintained on life support systems until brain death was confirmed in the intensive care unit. The children who survived had subtle clinical presentations. The four patients with altered sensorium had coexisting closed head injuries. These children had immobilization of the cervical spine until their vertebral injuries were identified. Delay in diagnosis did not occur in this subset of patients and none of them developed a neurological disability that was spinal cord related. Sneed and Stover described four cases of spinal cord damage that resulted from
DIETRICH ET AL
998
delays in diagnosis in patients with coexisting head injuries.” The presence of traumatic brain injury in a child demands comprehensive evaluation of the cervical spine. The 31 children who retained their ability to communicate were able to localize the area of injury with accuracy. Ninety-seven percent of the patients who were alert and verbal on admission expressed complaints of neck pain and tenderness. The one exception was a child with a high cervical cord injury who developed progressive deterioration of his respiratory status during transport and required immediate intubation on arrival in the emergency department. Reports detailing the clinical presentation of large cohorts of children with cervical fractures are not available. Rachesky et al reviewed 25 cases and concluded that any child with complaints of neck pain or with head trauma as a result of involvement in a motor vehicle accident had a significant potential for cervical spine injury.’ They indicated that pain alone should mandate a high index of suspicion because the existence of neurological deficits would not clearly identify the patients at risk, and our experience substantiates this point. Of the 31 children in this study who were awake and alert on admission, only 3 had any evidence of a neurological deficit that might have indicated the presence of vertebral injury. Seven of these 31 patients (22%) had an inappropriate delay in the management of their cervical spine fracture. One child was treated by his family physician for 8 months for suspected cervical strain before being referred to a physiatrist who obtained x-rays of the spine. In two children symptoms of neck pain and tenderness were attributed to coexisting injuries, which in both cases were clavicular fractures. Persistent neck pain prompted referral for further evaluation 4 days and 3 months postinjury. Relatively minor trauma produced significant injuries in four children. The lax ligaments, angulated facets and immaturity of the pediatric skeleton predisposes to this problem.3,8.” The majority (86%) of the children with delayed diagnosis sustained apparently trivial injuries in falls on the playground or while playing sports. In the abscence of a history of significant trauma, a thorough radiographic evaluation of the diagnosis of cervical
spine fracture may not be considered and may be delayed.‘2x13 The final source of diagnostic error was an inaccurate assessment of the cervical spine radiographs. Cervical spine x-rays may be more difficult to evaluate in pediatric patients, because many physicians are not familiar with the appearance of the immature skeleton.13,14These x-rays were initially misinterpreted on 5 occasions, 3 at this institution. Of the seven children who had delays in diagnosis, only one had a minor residual neurological deficit at the time of discharge, but the potential for devastating consequences should be emphasized. Although there was a relatively even distribution of fractures throughout all levels of the cervical spine in this series, all of the lethal injuries were observed in the upper cervical region (above C,) (Figs 1 and 2). The simple fractures had a propensity to occur in older children and in the lower cervical vertebrae. None of the children with lower cervical spine injuries (below C,) as a result of blunt impact trauma died and only two had residual neurological deficits. Hadley et al reported a 28% incidence of disability following fractures of the spine, but their report included all levels of the vertebral column (cervical, thoracic, lumbar).8 The outcome of children who sustained cervical spine fractures was accurately predicted by the scores used to assess injury severity (GCS, TS, ISS). Patients with multiple injuries and scores indicative of severe multisystem trauma did not survive. Children with isolated neck injuries who were neurologically intact survived with minimal morbidity. Although limited by retrospective design, this study demonstrates the need for thorough clinical and radiographic evaluation of any child with neck pain. Physicians should not be deterred by histories of insignificant trauma or lack of neurological findings. An adequate examination may require flexion/ extension views and/or computed axial tomography to evaluate the cervical spine. Medical personnel need to be meticulous in their management of any child with neck pain. Morbidity can be minimized with early recognition and appropriate management of cervical spine fractures.
REFERENCES 1. Jaffe DM, Binns H, Radkowski MA, et al: Developing a clinical algorithm for early management of cervical spine injury in child victims. Arm Emerg Med 16270-276, 1987 2. Rachesky I, Boyce WT, Duncan B, et al: Clinical prediction of cervical snine iniuries in children. Am J Dis Child 141:199-201. 1987 * ” 3. Hill SA, Miller CA, Kosnik EJ, et al: Pediatric neck injuries: A clinical study. J Neurosurg 60:700-706,1984
4. Teasdale consciousness. 5. Champion 91672~676,198l
G, Jennett A practical H,
Sacco
B: Assessment of coma and impaired scale. Lancet 2:81-84, 1974 W: Trauma
score.
Crit
Care
Med
6. Tepas J, Mollitt D, Bryant M: The pediatric trauma score as a predictor of injury severity in the injured child. J Pediatr Surg 22:14-18,1987
PEDIATRIC
CERVICAL
SPINE
FRACTURES
999
7. Greenspan L, McClellan B, Greig H: Abbreviated injury scale and injury severity score: A scoring chart. J Trauma 2560-64, 1985 8. Hadley MN, Zabramski JM, Browner CM, et al: Pediatric spinal trauma: Review of 122 cases of spinal cord and vertebral column injuries. J Neurosurg 68:18-24, 1988 9. Bohn D, Armstrong D, Becker L, et al: Cervical spine injuries in children. J Trauma 30:463-469, 1990 10. Sneed RC, Stover SL: Undiagnosed spinal cord injuries in brain-injured children. Am J Dis Child 142:965-967,1988
11. Ruge JR, Sinson GP, McLone DG, et al: Pediatric spinal injury: The very young. J Neurosurg 68:25-30,1988 12. Stauffer ES, Mazur JM: Cervical spine injuries in children. Pediatr Ann 11:502-511,1988 13. Ehara E-K: Cervical spine injury in children: Radiologic manifestations. AJR 152:1175-1178, 1988 14. Apple JS, Kirks DR, Merten DF, et al: Cervical spine fractures and dislocations in children. Pediatrics 17:45-49, 1987
Discussion A. Cooper (New York, NY): This study will add significantly to the growing body of literature on this subject. Unlike other studies, the investigators found that cervical spine fractures are not necessarily limited to the upper cervical spine. They also confirmed that while the SCIWORA syndrome (spinal cord injury without radiographic abnormality) has been much written about in recent years, the overwhelming majority of spinal cord injuries are in fact associated with vertebral column deformities. The results also stress the importance of a complete evaluation of the cervical spine at the time of injury, as 30 or 31 patients who were awake at the time of the presentation in fact had symptoms and signs of injury, while 25 of 31 patients were neurologically normal at the time of admission. I do have two questions: first, of the 16 patients who died, 8 had atlantooccipital distractions while this injury was present in none of the patients who survived. Of what specific injuries did the other patients die? Second, the cervical spine x-rays taken at the time of admission presumably occasionally are misinterpreted. Do you take any precautions to ensure that the cervical spine remains immobilized until the official interpretation is available on the morning following admission? D. Lloyd (Liverpool, England): I am concerned that so much emphasis is placed on the x-ray because, as Dr Cooper mentioned, the SCIWORA syndrome does occur. The critical thing with high-risk trauma is to suspect that the spine is unstable until the patient is conscious and can be properly evaluated. Clinical symptoms even if minimal or bizarre, should not be ignored simply because the cervical x-ray is normal; there is a very real risk of this if too much emphasis is placed on the x-rays. B. Beaver (Baltimore, MD): This is the largest collected series in North America depicting this most serious and very puzzling dilemma. Did management of airway priorities, nasotracheal intubation versus oral-tracheal intubation associated with the index of suspicion of cervical spine injury, affect outcome?
M. Olsen (Cleveland, OH): Do you have any incidence of cervical cord injury without fracture or dislocation in your series? Also, do you have any data regarding the use of seat belts and car seats in these children? h4. Fallat (Louisville, KY): I’m sure you’re aware that there’s a debate in the adult literature about whether or not some patients need evaluation of the cervical spine at all. Specifically, these are patients who have been in some type of traumatic incident but do not have any symptoms, neck tenderness or pain, and no focal neurological deficit. We have a very active emergency room residency program and some of the residents have been interested in trying to do a study in our pediatric population. Do you do just a lateral cervical spine as a screening film in your patients, or do you do a complete series? Do you ever not evaluate a trauma patient with at least a lateral cervical spine film? A.M. Dietrich (response): Dr Cooper, in reference to the 16 patients who died, 8 did have documented atlantooccipital distraction. The autopsies on the other patients (and even on some of the patients that had atlantooccipital distractions) were incomplete and the number of autopsies that were performed was low. Most of these children (85%) had major associated injuries determined from the emergency room record and physical examination. The autopsy information was rather poor, so it is difficult for me to say if there were other causes of death. In terms of our protocol, most of these children who have complaints of neck pain and neck tenderness should be immobilized until the point when you have done a lateral cervical spine, trauma series, and x-rays in flexion and extension, CT scans, and whatever is determined appropriate by the neurosurgical staff at that hospital. In this institution we take very seriously any children who have complaints of neck pain and tenderness and evaluate them thoroughly. Dr Fallat had also addressed the question of how our patients were evaluated. All patients who had a significant injury such as
DIETRICH ET AL
motor vehicle accidents, or children struck by cars, or children with complaints of neck pain and tenderness receive appropriate cervical spine immobilization. After a patient gets a lateral cervical spine series we examine the patient. If there is no tenderness we stop at that point. If there is any tenderness then we proceed on until we have effectively done all of the testing. We excluded all patients with SCIWORA. To my knowledge we’ve only had three cases in the last 2 years. There were 19 patients who required some type
of airway management in the field. Three children were sent to another hospital and there were no cases in which nasotracheal intubation was used. All of the patients were intubated orally and sedated. All of the children were intubated with cervical spine traction being held or cervical spine stabilized. Most of the children that did require intubation were apneic, and a large number of them were pulseless at the scene. It’s difficult to say whether that had any effect on our outcome.
