INTRODUCTION
Acute epidural hematomas (EDHs) in children and infants represent a potentially life-threatening condition after head injury. EDHs in children are a relatively unusual occurrence. The reported incidence of EDH ranges between 1% and 6% after head trauma in hospitalized children5,10). Early identification and adequate management depending on the patient's clinical condition and the size of EDH are prerequisites for good outcome10). However, besides the presence of midline structure shift on head computed tomographic (CT) scan and clinical signs of uncal herniation, other factors, such as associated brain contusion, diffuse axonal injury (DAI), or extracranial injury, may determine the outcome of the children. Also, the causes and outcomes of acute EDHs vary depending on period and region. The aims of this analysis were to review the cause and outcome of surgically treated pediatric EDHs in our hospital and to discuss outcome-related variables in a large consecutive series of surgically treated EDH in children for 10 years.
MATERIALS AND METHODS
Patient population
A total of 175 children were admitted with head trauma between Jan 2000 and Feb 2010. There were 57 patients with EDHs. Among 57 patients, 29 patients with EDH were surgically treated. All patients were documented in a database at the time of treatment. Perioperative course, radiological findings, and outcome were evaluated retrospectively. Items included in the database were demographic features, cause of trauma, preoperative clinical status according to the Glasgow Coma Scale (GCS) score, signs of transtentorial herniation (unilateral or bilateral dilated pupil), isolated head injury or associated substantial extracranial injury, time to treatment (including the reason for delayed treatment, if performed), and the outcome according to the modified Rankin Scale (mRS) and Glasgow Outcome Scale (GOS). The outcome was classified as excellent (mRS score, 0; GOS score, 5), good (mRS score, 1.2; GOS score, 4), or adverse (mRS score ≥3; GOS score ≤3).
Surgical procedures and imaging studies
All patients were treated according to a standard advanced trauma life support protocol if presented immediately to the emergency department of our institution. Trauma workup was initiated according to the clinical status and the mechanism of the trauma and usually included a head and body CT scan and abdominal ultrasound. After documentation of an EDH treatment, indication for surgery was a thickness of the hematoma of at least 10 mm (maximal thickness), dislocated skull fractures associated with EDH, or a documented increase of the hematoma size with clinical deterioration. Operative treatment of EDH was performed as a standardized procedure. After osteoplastic craniotomy, hematoma evacuation was performed. Any craniotomy performed more than 10 hours after injury was classified as "delayed". All patients underwent either postoperative cranial CT or magnetic resonance imaging (MRI) to document adequate removal of the hematoma, trauma-associated brain contusion, or any complication related to the trauma or operative treatment. MRI was the preferred method of control imaging whenever DAI was suspected from the severity and mechanism of trauma.
Statistical analysis
Comparisons of demographic variables including age at trauma, isolated head injury, unilateral or bilateral dilated pupil, severe head injury (GCS <8) before treatment, hematoma size, presence of brain contusion, and time to operative evacuation between the groups were performed using unpaired t-test, χ2 test. Values of p<0.05 were considered significant.
RESULTS
General characteristics
Twenty-nine patients with EDH were surgically treated in study period. Age ranged from 1 day to 15 years (mean age : 109 months). Two patients (6.8%) were younger than 1, two (6.8%) younger than 4, 10 (34.5%) between 4 and 10, and 15 (51.9%) older than 10 years. Nineteen patients (65.5%) were male. The most common cause was traffic accident (14 cases, 48.2%). In cases of traffic accident, pedestrian hit was most common cause (9 cases, 31%) followed by fall down (6 cases, 20.6%), slip down (6 cases, 20.6%), and birth injury (1 case, 3.4%) (Fig. 1). One case was due to unknown cause. He had no history of trauma. The GCS score before hematoma evacuation ranged from 3 to 15 (Mean score, 12.9). Only 1 patient (3.4%) had severe head injury (GCS score <8), 6 patients (20.7%) had a GCS score between 9 and 12, and in 22 patients (75.9%), the GCS score was 13 to 15. The mean time to surgery was 23 hours; however, 11 patients (37.9%) underwent surgical evacuation of the hematoma within 10 hours. Delayed surgery (more than 10 hours after trauma) was performed in 8 cases (27.5%). Abnormal pupil finding was only 2 cases (6.8%) (Table 1).
Radiologic finding
Hematoma location was temporal in 13 cases (44.8%), frontal in 7 cases (24.1%), parietal and occipital in 4 cases respectively (13.7%) (Fig. 2). The locations of hematoma were almost same in both sides (left side 15 cases). The mean thickness of the hematoma was 18 mm (range, 5-40 mm), and 14 patients (48.2%) showed midline shift in initial imaging. Twenty-four patients (82.5%) had an associated skull fracture. Five patients had cerebral contusion. Three patients underwent MRI after the treatment of the EDH, and 1 patient showed signs of DAI. Heterogeneous hematomas in CT scans were 20 cases (68.9%). Among these cases, 6 cases showed enlargement of hematoma in follow-up CT scans.
Clinical outcome
None of the patients in this series died of EDH (0% mortality). In this study, only one patient died. He was 1-year-old and fell down on the kitchen floor. When he arrived at the hospital, he was in cardiac arrest. His CT scan showed small amount of EDH in right temporal area. Finally, he died due to hypoxic brain damage and hemoperitoneum. Except this case, no patient died of EDH. The overall outcomes were excellent in 26 patients (92.8%) and good in 2 patients (7.2%).
DISCUSSION
Incidence of epidural hematoma in children was scarcer than that in adults due to tight adherence of the dura to the inner table of the skull10). In other studies, of all head-injured admitted children under 15 years of age admitted, 2-3% were found to have extradural hematomas7). But in our study, 32.6% of all admitted head-injured children (57 of 175) had EDHs. A possible explanation for this difference might be found in the difference of cause of trauma. In other studies, the most common cause was fall down. But in our study, it was traffic accident.
Children sustaining cranial traumas associated with EDH represent a heterogeneous group with various clinical outcomes. The approximated mortality is 5%, although mortality in the quoted literature ranges from 0% to 12%, and was even higher in the pre-CT era3,5,11).
In our study, mortality was zero, 26 cases had excellent outcome. Only 2 cases had good outcome. A possible explanation for the overall very good outcome might be the advanced contemporary pediatric trauma care and urgent surgery in patients with clinical signs of uncal herniation.
There was significant difference between children and adults in the clinical course. 20-30% of adult patients were unconscious (GCS <8). However, only 3% of children in our study were unconscious. It was in concordance with other studies1,4,7-13).
Prognostic factors of EDH in children are GCS, pupillary reaction and light response, and the presence of focal neurological deficit, the laboratory findings of hyperglycemia and hypokalemia5). These are similar to those in adults3). But, mortality of EDH in children is lower than that of adult10). Such a good outcome in children is probably related to the significantly lower incidence of associated intradural injury and the lower incidence of unconsciousness of patients requiring immediate surgery10).
The criteria for determining which cases for conservative and which cases for surgical treatment remained controversial. Bejjani et al.5) found that the most important radiographic parameter dictating surgical evacuation was maximum diameter of hematoma of more than 15 mm and midline shift of more than 5 mm. Only patients with EDHs less than 10 mm and no focal deficits were treated conservatively and followed up closely, but if any deterioration occurred and/or an increase in the size of the hematoma was seen, we decided to perform a surgical evacuation.
It was reported that the incidence of skull fracture in children was 60-82% and it was known that there was no difference in that of adult6,7,9,10,13). In our study, the incidence of skull fracture was 82.5% and it corresponded to reported literature.
The mean time to surgical intervention was 23 hours. Surgery was performed within 24 hours in about 80% of adults, but only in 54% of children. It is thought that it was because of insidious clinical course of EDH in children7,10).
The heterogeneous hematoma appearances of some actively bleeding extradural hematomas have been termed the "swirl sign"2). In our study, 70.0% (20 of 29) had the "swirl sign". Six of 20 patients showed increased hematoma volume in follow-up CT scans.
This study was retrospective and included a small number of patients. Therefore, we could not show statistically meaningful factor which affected prognosis. But, our good outcome may be kept in mind in case of children with extensive EDH and clinical signs of uncal herniation are seen in the emergency department.
CONCLUSION
Extradural hemorrhage is one of the most threatening lesions in patients with craniocerebral trauma, and early diagnosis is essential for its adequate treatment. It can be managed with excellent outcome even in children with low GCS scores, abnormal pupillary reaction, and large hematoma size. However, these cases should be treated surgically without delay when necessary.
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
