Keywords
paediatric - extradural - haematoma - neuropsychology - follow-up
Introduction
Extradural haematomas (EDHs) are responsible for 2 to 3% of all acute traumatic paediatric
head injuries.[1]
[2] Most literature states that children who are unwell or who have a large EDH should
receive rapid surgical intervention to ensure optimal surgical outcomes.[3] If operated in a timely manner, most children appear to recover very well postoperatively,
returning to full-time education and sporting activities.[1]
[4]
Paediatric EDHs are often discharged from neurosurgical follow-up relatively quickly.[4] However, in other forms of head injury, such as brain contusions and subdural haematomas,
it is known that children may develop neuropsychological symptoms that require specialist
treatment, such as emotional lability, memory issues, mood disturbance and sleep issues.[5] These complex domains are rarely evaluated as a standard of care in most neurosurgical
clinics, and thus, children with these types of head injuries are often also routinely
assessed by specialist paediatric neuropsychologists[5]
[6] and close, multidisciplinary follow-up is often advocated.[7] However, most literature on paediatric EDHs focusses on surgical outcomes such as
mortality, recurrence and postoperative motor deficits and does not assess the follow-up
or neuropsychological outcomes in children following surgery for these extra-axial
haematomas.[4]
[8] There is also no set guidance to inform the follow-up of paediatric EDH patients,
which prompted the current study to evaluate the practice for follow-up and neuropsychological
assessments of children who have undergone acute surgery for an EDH in a single tertiary
neurosurgical unit in the United Kingdom.
Materials and Methods
A retrospective review of a prospectively maintained and link-anonymised database
of all acute paediatric EDH admissions was conducted to screen for all cases of paediatric
EDH treated surgically at a single United Kingdom's tertiary paediatric neurosurgery
centre. The inclusion criteria were (1) diagnosis of acute traumatic EDH via CT scan
in a paediatric (<16 years old) patient and (2) urgent neurosurgical evacuation of
the EDH. The study period spanned between March 2007 and May 2021. This retrospective
study was performed in-line with the principles stated in the Declaration of Helsinki.
Anonymised clinical data were collated from electronic patient documentation, including
patient gender, date of birth, mechanism of injury, admission Glasgow Coma Scale (GCS),
neurological and pupillary deficits, date and nature of surgical intervention, postsurgical
course, Glasgow Outcome Score (GOS), number and nature of follow-up appointments and
evaluation of any clinical or neuropsychological concerns at follow-up. CT scans were
reviewed by a consultant neurosurgeon. The location of the EDH, the presence or absence
of brain contusions and the presence or absence of skull fractures were noted.
Follow-up appointments were screened using electronic records, noting the clinic in
which the reviews were conducted, the median follow-up times, the number of follow-up
appointments and the presence of any ongoing physical or psychological concerns at
these appointments.
During the study period, children treated surgically for an EDH were referred for
review with a paediatric neuropsychologist postoperatively. However, due to appointment
constraints during the COVID pandemic and the subsequent departure of the unit's paediatric
neuropsychologist, these reviews were paused from 2020 onwards. Therefore, a subcohort
of children in this study did not have access to neuropsychological support postoperatively.
Of children who did receive neuropsychological follow-up, each appointment consisted
of the child and parents/guardians completing questionnaires about the child's behaviour,
mood and cognitive function. Parents/guardians completed the Behaviour Rating Inventory
of Executive Function (2nd edition), assessing executive functioning skills and ‘metacognitive
skills’, and a Conners 3 assessment, providing ratings of the child's perceived attention
and ability to concentrate.[9]
[10] Children were asked to self-report via the Revised Children's Anxiety and Depression
Scale (RCADS), assessing states of anxiety and low mood in children aged 8 to 18 years.[9]
[11]
[12] The neuropsychological review also assessed cognitive skills relating to intellect,
memory, processing speed, attention skills, executive functioning skills and academic
skills. Recommendations for further treatment considered the results of parental and
patient reporting, behavioural observations during assessment, and the assessment
scores themselves.
Statistical evaluations were conducted using Fisher's exact test for categorical data.
p-values <0.05 were considered statistically significant.
Results
Participants
Thirty-five paediatric acute traumatic EDH cases were identified between March 2007
and May 2021 and included in the study. The majority of patients were male (85.7%;
n = 30), and 14.3% (n = 5) were female. Age distribution on the date of surgical intervention ranged from
5 months to 15 years, 11 months. The median age on the date of surgery was 9 years.
Twenty percent (n = 7) of patients were <5 years of age on the date of surgical intervention. Forty-six
percent (n = 16) were 5 to 10 years of age, and 12 patients (34.3%) were 10 to 16 years of age
on the date of surgery. None of these three age categories was statistically associated
with a higher risk of neuropsychological deficits postoperatively.
Mechanisms of Injury
The leading cause of injury was a fall (49%, n = 17). This included falls from heights, standing height and falls downstairs. Non-motorised
and motorised vehicle incidents made up other significant subcohorts of cases in this
study ([Table 1]). There was one child for whom we did not have access to the reported mechanism
of injury and was excluded from this section of the analysis. One hundred percent
(n = 5) of female patients in this series suffered a fall as the reported mechanism
of injury.
Table 1
Breakdown of mechanisms of injury leading to the development of an extradural haematoma
|
Mechanism of injury
|
Impact (low, moderate or high)
|
n (%)
|
|
Fall from standing height
|
Low
|
10 (29.4%)
|
|
Fall from arms of an adult
|
Moderate
|
1 (2.9%)
|
|
Fall downstairs
|
Moderate
|
2 (5.9%)
|
|
Ran into a table
|
Moderate
|
2 (5.9%)
|
|
Fall from a height (10 ft)
|
High
|
1 (2.9%)
|
|
Involving a non-motorised vehicle (bikes and scooters)
|
High
|
9 (26.5%)
|
|
Involving a motorised vehicle (cars and motorbikes)
|
High
|
5 (14.7%)
|
|
Sporting accident resulting in direct head trauma
|
High
|
2 (5.9%)
|
|
No data
|
–
|
1 (2.8%)
|
Preoperative Glasgow Coma Scale and Neurology
The lowest GCS of children prior to surgical evacuation was recorded ([Table 2]). Just over half (58.1%, n = 18) of patients had a worst presenting GCS of 14 to 15 (mild); almost 10% (n = 3) had a moderate GCS (9–13); and almost a third (n = 10) of children had a severe traumatic brain injury (TBI; presenting GCS 3–8).
A severe TBI (preoperative GCS 3–8) was statistically associated with a higher risk
of postoperative neuropsychological deficits (p = 0.0454).
Table 2
Perioperative clinical findings and their associations with postoperative neuropsychological
issues
|
Clinical findings
|
Number of children (%)
|
Association with postop neuropsychological deficits (p-value)
|
|
Pupil dysfunction
|
8 (22.8%)
|
0.083
|
|
Motor deficits
|
5 (14.3%)
|
1.000
|
|
Preoperative worst GCS
|
|
14–15 (Mild)
|
18 (51.4%)
|
0.139
|
|
9–13 (Moderate)
|
3 (8.5%)
|
1.000
|
|
8–3 (Severe)
|
10 (29%)
|
0.0454
|
|
Posttraumatic seizures
|
2 (5.7%)
|
0.152
|
Abbreviation: GCS, Glasgow Coma Scale.
p-values were calculated via Fisher's exact test. p < 0.05 was considered significant.
Fourteen percent (n = 5) of the children had neurological deficits on admission. One child had a transient
left-hand paraesthesia. One child was bilaterally areflexic in their lower limbs.
Another child had severe bilateral upper and lower limb weakness, in addition to Cushing's
response. The fourth child had right-sided hemiplegia, and the final child presented
with Wernicke's aphasia and right-hand paraesthesia. All of these childrens' neurological
deficits resolved immediately after surgery, except for the child exhibiting the Cushing's
response. This child had persistent unilateral hemiparesis postsurgery, although this
did resolve without further treatment prior to the patient's discharge from the hospital,
albeit with residual choreiform movements of those limbs persisting long-term. The
presence of preoperative neurological deficits was not associated with postoperative
neuropsychological deficits in this study cohort (p = 1.000).
Twenty-three percent (n = 8) of patients had abnormal pupillary responses upon admission. These included
issues such as asymmetrical pupils and bilateral sluggish pupils. Pupillary function
normalised postoperatively in all eight patients. Preoperative pupillary dysfunction
was not associated with postoperative neuropsychological deficits (p = 0.083).
Radiology
All 35 patients received a CT scan on admission to the emergency department. Scans
were assessed for haematoma location, presence/absence of brain contusions and presence/absence
of skull fractures ([Table 2]). No patients had bilateral EDHs, and all EDHs were supratentorial. Forty percent
(n = 14) had a left-sided EDH, and 60% (n = 21) had a right-sided EDH. Fourteen percent (n = 5) of patients had brain contusions as well as their EDH; four of these children
had small cortical contusions, and one had bilateral cerebellar contusions. Skull
fractures were present in 74% (n = 26) of patients. None of these fractures were depressed in nature. None of these
imaging factors had any statistical association with postoperative neuropsychological
deficits ([Table 3]).
Table 3
Radiological findings from presurgical CT scans and their associations with postoperative
neuropsychological issues
|
Radiological findings
|
Number of children (%)
|
Association with postop neuropsychological deficits (p-value)
|
|
Skull fracture
|
26 (74.3%)
|
0.686
|
|
Brain contusions
|
5 (14.3%)
|
0.152
|
|
Side of EDH
|
|
Left
|
14 (40%)
|
1.000
|
|
Right
|
21 (60%)
|
1.000
|
Abbreviation: EDH, extradural haematoma.
p-values were calculated via Fisher's exact test. p < 0.05 was considered significant.
Surgical Outcomes
All patients underwent craniotomy for evacuation of their EDH. One patient also underwent
insertion of a ventricular drain and subsequent ventriculoperitoneal shunting for
posttraumatic hydrocephalus. He achieved a GOS of 4 at discharge from Wales. No follow-up
information was available for this child, as he was on holiday at the time of his
injury and returned to England for his ongoing rehab and follow-up.
Ninety-four percent (n = 33) of patients achieved a GOS of 5 on discharge, and 5.7% (n = 2) of patients achieved a GOS of 4. There were no deaths or recurrences in the
study cohort. There were no postoperative infections in this study cohort. Two patients
had posttraumatic seizures after their head injury. In one of these patients, the
seizures resolved 13 months after surgery, whereas the other patient remained epileptic
long-term. Both patients who had seizures also had neuropsychological concerns identified
at follow-up (p = 0.152; [Table 2]).
Follow-up appointments
The median follow-up duration for the study cohort was 4 months (range 0–55 months).
Five children in the study population had no follow-up postdischarge from hospital.
Of these children, two sustained accidents whilst on holiday in Wales, and as a result,
when discharged home to England, no follow-up documentation was available for review
in this study. These five children were excluded from subsequent analyses.
Follow-up varied considerably between patients. Despite all being treated surgically
for an acute traumatic EDH, there was no discernible pattern to follow-up scheduling.
[Tables 4] and [5] list the number of follow-up appointments each child received and the departments
which conducted their follow-up appointments. Most children who were followed up received
a single follow-up appointment (n = 17, 56.7%), which was most commonly with a paediatric neurosurgeon (n = 26, 86.7%).
Table 4
Table summarising number of postoperative appointments for the study population
|
Number of appointments
|
Number of children (%)
|
|
≥5
|
2 (6.66%)
|
|
4
|
1 (3.33%)
|
|
3
|
2 (6.66%)
|
|
2
|
8 (26.7%)
|
|
1
|
17 (56.7%)
|
|
0
|
5 (16.7%)
|
Table 5
Distribution of specialties conducting follow-up appointments
|
Specialty responsible for follow-up appointments
|
Number of children seen (%) [total number of appointments]
|
|
Paediatric neurosurgery
|
26 (86.7%) [29]
|
|
Paediatric neurology
|
9 (30.0%) [17]
|
|
Paediatric rehabilitation
|
1 (3.33%) [2]
|
|
Paediatric neuropsychology
|
4 (13.3%) [5]
|
Postoperative Neuropsychology
No patients in this study had preoperative neuropsychological diagnoses. Neuropsychological
concerns were identified at postoperative follow-up for 12 patients ([Table 6]), but only four of these children had formal neuropsychology reviews. These children
were referred to the neuropsychology department prior to the departure of the paediatric
neuropsychologist from the unit.
Table 6
Features of all patients who had neuropsychology concerns identified at follow-up
after surgery for extradural haematoma
|
Age at surgery (years, months)
|
Preop GCS
|
GOS on discharge
|
Number of follow-up appointments
|
Concerns identified
|
Number of neuropsychology team reviews
|
|
0, 5
|
8
|
5
|
2
|
Dyslexia, atypical behaviour, unusual seizures (possibly some non-epileptic events)
|
0
|
|
7, 11
|
5
|
5
|
2
|
Hypersexuality, ADHD-like symptoms
|
0
|
|
14, 1
|
5
|
5
|
1
|
Paranoia
|
0
|
|
4, 0
|
14
|
5
|
3
|
Personality change, poor engagement with authority
|
0
|
|
9, 0
|
3
|
5
|
5
|
Mood disturbance, forgetfulness,
|
0
|
|
5, 11
|
14
|
5
|
2
|
Mood disturbance, anxiety
|
1
|
|
6, 8
|
6
|
5
|
2
|
Poor attention, learning difficulties
|
1
|
|
7, 0
|
14
|
5
|
4
|
Poor communication skills, personality change, mood disturbance, deliberate self-harm
|
2
|
|
1, 5
|
9
|
5
|
6
|
Poor memory (transient)
|
0
|
|
15, 11
|
8
|
4
|
3
|
Severe behavioural issues, mood disturbance
|
1
|
|
8, 3
|
7
|
5
|
1
|
Mood disturbance, poor concentration
|
0
|
|
2, 1
|
14
|
5
|
2
|
Sleep disturbance
|
0
|
Abbreviations: GCS, Glasgow Coma Scale; GOS, Glasgow Outcome Score.
The most common neuropsychological symptom identified was mood disturbance (n = 5/12, 41.6%; [Table 6]). Four patients received a formal neuropsychology assessment with a paediatric neuropsychologist.
Eight children had follow-up notes which referred to neuropsychology concerns, but
no documented assessments with a neuropsychologist have taken place to date. Eleven
children with neuropsychological issues were male, and one was female.
Both patients who had posttraumatic seizures had neuropsychological concerns raised
at follow-up ([Table 2]). Children who had a low preoperative GCS (3–8) were more likely to suffer from
neuropsychological repercussions than those with a preoperative GCS >8 (p = 0.0454; [Table 2]). There was no statistically significant association between brain contusions, skull
fractures or EDH laterality with the development of postoperative neuropsychological
concerns ([Table 3]).
The first child who received a formal neuropsychological review had surgery at age
5 years, 11 months. The time between injury and neuropsychological review was 70 months,
essentially serving as a long-term follow-up review at 6 years postinjury. He was
not taking any medications at the time of his neuropsychology review. He exhibited
signs of high levels of anxiety and low mood, noted as a new change since his brain
injury, according to his parents. During self-reporting, using the Impact of Events
Scale Revised, this child showed no indications of ongoing trauma to suggest a posttraumatic
stress spectrum syndrome. This child was fatigued and slow to respond, and the assessments
were terminated early by his parents. The examinations that were completed indicated
deficits in several areas, such as verbal intellect, processing speed and memory recall.
There were also concerns about a lack of self-regulation for his age.
The next neuropsychology assessment was undertaken on a child who had surgery at 6
years, 8 months. The time between injury and neuropsychological review was 29 months
(roughly 2 years). The patient was not taking any regular medications at the time
of the neuropsychological review. Conners 3rd Edition assessment completed by the
child's parents indicated marked levels of inattention and learning difficulties compared
with peers of his age. The Impact of Events Scale Revised showed no indication of
ongoing trauma. This child was cooperative during the assessment.
The third child assessed by the neuropsychology team was 7 years old. The time between
injury and neuropsychological review was 55 months (4.5 years). The patient was not
taking any regular medications at the time of the neuropsychological review. His neuropsychological
assessment revealed several new executive function issues, mood disturbance and poor
verbal intellect. The child also performed acts of deliberate self-harm following
his EDH, suggestive of ongoing posttraumatic stress-related issues. At his most recent
follow-up (27 months postsurgery), these symptoms, including his mood and school performance,
had improved.
The final child who received a neuropsychological review was 15 years old. He was
slow to recover following his craniotomy. Whilst his presurgery left-sided hemiparesis
did improve prior to discharge from the neurosurgical unit, he was left with persistent
choreiform movements of the left side. The time between injury and neuropsychological
review was 1 month. At the time of this neuropsychological review, he was not taking
any regular medications. His neuropsychological review confirmed the presence of severe
behavioural and mood issues and identified him a highly vulnerable for posttraumatic
stress-related concerns going forward. He was started on tetrabenazine for his choreiform
movements after the review by neuropsychology.
Discussion
Most patients in this study cohort achieved excellent surgical outcomes, in-line with
those in the published paediatric EDH literature.[4]
[8] However, using surgical markers of good outcomes such as mortality, recurrence,
infection and scales such as the raw GOS may not capture more subtle deficits that
may affect children following head injuries. The follow-up scheduling for these children
has also received comparatively little attention.[1]
[4]
It is established that paediatric head trauma can lead to neuropsychological issues
as the child develops in the proceeding months or years postinjury.[10]
[11]
[12] Historically, the relationship between head injury and psychological issues had
been thought to correlate with the severity of brain injury, and in particular, the
degree of frontal intraparenchymal injury; therefore EDH is often considered lower
risk for neuropsychological issues versus injuries that involve brain contusions,
subdural haematomas or diffuse axonal injury.[6]
[7] However, more recent work has now established that 30 to 40% of children can be
at risk of neuropsychological disturbances after even a mild brain injury, without
radiological evidence of intraparenchymal injury.[7]
[13] However, neuropsychological outcomes following EDH, specifically, are rarely reported.[4]
[8] The present study was designed to describe existing follow-up regimes for children
after EDH surgery and assess whether neuropsychological deficits are being identified
at follow-up for these children, and if so, whether such symptoms are being appropriately
assessed and treated.
Most patients in this study cohort did receive at least one follow-up review within
a median duration of follow-up in-line with that reported in previous EDH case series.[2]
[4] Most children had a single follow-up appointment within the paediatric neurosurgery
outpatient clinic, after which they were discharged. Children who had multiple follow-up
appointments were more likely to have long-term sequelae, physically and neuropsychologically,
than children discharged after one appointment. However, delayed onset of neuropsychiatric
symptoms following paediatric head injury, beyond the typical EDH follow-up regimes
reported in this study and others, has been described.[10]
[13] Therefore, it remains possible that some EDH patients are at risk of developing
delayed neuropsychological symptoms, which may not be identified if only a single,
acute follow-up appointment is arranged. It may be wise to consider both an acute
and a later, delayed follow-up review for children after any significant head injury,
including EDH, to permit screening for delayed neurocognitive and psychological issues,
posttraumatic stress disorder (PTSD) and impaired school performance, which are associated
with paediatric head injury.[12]
Forty percent (n = 12) of patients who received follow-up in the study cohort had neuropsychological
concerns identified. The presence of these symptoms does not seem to be related to
confounding factors such as concomitant contusions, skull fractures, seizures or medications
([Tables 2] and [3]). Of note, there were no patients in the present study who were taking antiseizure
medications at the time of their neuropsychological reviews, which are known to be
associated with neuropsychological side effects.[5] Only low preoperative GCS, a gross surrogate marker of a more severe overall TBI,
was significantly associated with neuropsychological morbidity in this study, which
has been widely reported by others.[10]
[14] This suggests that in children with a low preoperative GCS, EDH in of itself carries
a risk of neuropsychological morbidity in children, and so routine screening for these
deficits may be important to optimise care for children, even after an isolated EDH.
Early intervention for EDH in children with a poor GCS is known to improve surgical
outcomes[4]; whether the timing of surgery influences neuropsychological outcomes was not the
focus of the present study, but it is likely.[10]
[12] Of note, half of the children in the present study who had neuropsychological reviews
also had a reasonably high preoperative GCS of 14/15 ([Table 6]). This highlights the potential need for vigilance for neuropsychological issues
after surgery for EDH in children, even if their preoperative GCS was not particularly
low.
Neuropsychological symptoms are significant issues after TBI and come with marked
treatment challenges within the context of paediatrics, but are rarely considered
in children who have had an EDH.[3]
[6]
[9]
[10]
[14] These symptoms can be severely debilitating and progress over time, affecting school
performance and social skills development.[10]
[12] They thus require prompt recognition and subsequent referral for specialist assessment.
Two-thirds of the patients who had neuropsychological issues in the present study
did not receive neuropsychology reviews and, therefore, received no neuropsychological
treatment. This discrepancy may relate to financial constraints, particularly within
the context of the coronavirus pandemic, which spanned 2 years of this study's timeline.
Furthermore, the departure of our unit's resident paediatric neuropsychologist undoubtedly
affected the referral stream to the neuropsychology department. The nature of tertiary
services may also influence patient attendance for follow-up and/or neuropsychology
reviews; this is especially pertinent in the realm of neurosurgical patients, who
are pooled from wide geographical areas. Finally, neuropsychological symptoms can
be subtle and difficult for clinicians, parents and school workers to detect without
a high index of suspicion.[10]
[11] It is therefore vital that clinicians are aware that EDH patients are at risk of
neuropsychological morbidity, even when their surgical outcomes appear excellent,
and that parents and caregivers are appropriately counselled on the neuropsychological
symptoms to be vigilant for in the months following EDH surgery.
The implementation of specialist, multidisciplinary paediatric head injury clinics,
where financially feasible, may allow streamlining of follow-up and provide a standard
framework in which all children receive appropriate access to a neurologist, neurosurgeon
and paediatric neuropsychologist following TBI, including postoperatively following
an EDH. This would reduce the number of visits to a tertiary care centre for patients
who struggle to travel, potentially improving attendance and patient/parent concordance
with treatment plans. It may also prove cost-effective, particularly if run in a hybrid
fashion, combining virtual and face-to-face follow-ups with multiple specialities
as required, on a case-by-case basis. As a result of this study, combined with other
data on the need for neuropsychology services in South Wales, more resources have
recently been agreed by the Specialist Commissioners for paediatric neuropsychology.
Gross TBI outcome scales, such as the GOS, are typically designed for adult patients.
This has led to the development of more detailed neurological and psychological assessments
for children who have suffered head injuries, such as the Paediatric GOS-Extended
and the King's Outcome Scale for Childhood Head Injury.[2]
[15]
[16] These scales include considerations that are more likely to capture neuropsychological
deficits in children post-TBI, such as screening for changes in family/friend relationships,
effects on school performance and mood disturbances. However, these scales are rarely
used in the EDH literature to assess outcomes.[4]
[8] Utilising these succinct, paediatric-specific tools at neurosurgical follow-up appointments
after EDH could aid clinicians in identifying subtle, yet potentially disabling, neuropsychological
issues in these children, who may otherwise appear to be very well. This, in turn,
may lead to more holistic paediatric care and rehabilitation after surgery for EDH.
The children within this series of surgically treated EDHs had excellent surgical
outcomes. However, this study highlights that follow-up of children following surgery
for EDH can be sporadic and neuropsychological reviews may be sparse, despite neuropsychological
symptoms developing in 40% of children after their EDH was evacuated. We, therefore,
advocate standardising follow-up regimes and consideration of routine neuropsychological
review for all children who have surgery for an EDH.
