Paediatric head injury

Introduction

• Traumatic brain injury is the leading cause of morbidity and mortality in paediatric trauma
• Significant TBI seen in 75% of blunt trauma causes and 70% of these will die
• To prevent secondary brain injury as a result of hypoxia, hypotension, fever, seizures, poor cerebral perfusion, cerebral bleeding, hyperglycaemia (note hypoglycaemia not identified as cause of secondary injury although no doubt it is) and missed diagnoses

CDRs

Multiple different clinical decision rules have been developed in order to guide imaging and discharge decisions. PECARN, CHALICE and CATCH are the most well known (see below).

Pecarn vs. chalice vs. catch

• PECARN is for all children with GCS >13
• CHALICE considers all children after a head injury
• CATCH was designed for a higher-risk cohort
• Both PECARN and CHALICE have a negative predictive value of 99.9% for significant intracranial injury
• Easter at al. in 2014 published in Annals of Emergency Medicine a comparison study in new population in USA ED. 1009 patients <18yo within 24 hours of injury
• Patients with GCS <13 or coagulopathy were excluded
• Results

Comparison

• PECARN most sensitive but will entail performance of more CT scans than comparator rules
• In this study, all patients with high risk OR moderate risk criteria on PECARN were classed as positive and underwent CT
  • This achieved 100% sensitivity but is not what the original paper described as it offers observation vs. CT option
• Overall 100% sensitivity of PECARN is hard to go past and while it may entail more scanning than the use of CHALICE or CATCH, their 84% and 91% sensitivity profiles in this prospective study are not high enough for safety in an area where there is zero-tolerance for missed injuries

APHIRST study

• Australia and New Zealand observational study
• PECARN had the highest sensitivity and all three rules had NPV >99%
• Strict application of the rules would have resulted in a much higher imaging rate than is the current standard of care across the 10 tertiary and large mixed hospitals in Australia and New Zealand (10% vs.
  • PECARN rate of up to 46% if all intermediate-risk cases were scanned rather than observed)
  • CHALICE 22%
  • CATCH 30%
• Clinical gestalt outperformed any rule in children who presented GCS 13-15 (no missed injuries)

PECARN

• Original study 42412 children <18 presenting to US ED with 33785 in derivation cohort (8502 <2yo) and 8627 in validation cohort (2216 <2yo)
• Only children GCS 14 or greater (<14 all get CT)
• Clinically significant traumatic brain injury (see below) occurred in 0.9% of patients and 0.1% underwent neurosurgery. Zero patients died.
• In patients <2yo, 100% sensitive
• In patients >2yo, 96.8% sensitive. Validation studies have shown 100% sensitivity
• Externally validated in 2 trials
• Number of vomiting episodes and timing of episodes was non-contributory
• Holmes et al. Lancet. 2009
• Excluded patients
  • Trivial injury: Ground-level falls, walking or running into stationary objects, no signs or symptoms of head trauma other than abrasion/laceration
  • Penetrating trauma
  • GCS<14
  • Brain tumors
  • Pre-existing neurological disorders
  • VP shunts
  • Coagulopathy
  • >24 hours after injury

• Children <2yo
  • GCS 14, other signs of altered mental status* or palpable skull fracture = CT (4.4% risk of ciTBI)
  • Non-frontal scalp haematoma, hx of LOC >5s or severe mechanism of injury * or not acting normally as per parent = Observation 4-6 hours vs. CT on basis of physician experience, multiple vs. isolated findings, any deterioration, age <3mo or parental preference (0.9% risk of ciTBI per predictor)
  • Otherwise CT not recommended (<0.02% risk of ciTBI)
• Children >2yo
  • GCS 14 , other signs of altered mental status* or signs of basilar skull fracture = CT (4.3% risk of ciTBI)
  • History of LOC, vomiting, severe MOI* or severe headache = Observation 4-6 hours vs. CT (0.8% risk of ciTBI per predictor)
  • Otherwise CT not recommended (<0.05% risk of ciTBI)

• ciTBI defined as:
  • Death from TBI
  • Neurosurgical intervention
  • Intubation >24 hours
  • Hospital admission 2 or more nights with CT evidence of TBI
• Severe MOI defined as:
  • MVA with patient ejection/death of other passenger/rollover
  • Fall >0.9m (<2yo) or 1.5m (>2yo) or 
  • Head struck by high impact object
• Altered mental status defined as:
  • Agitation, somnolence, repetitive questioning, or slow verbal response to verbal communication

Chalice

• Children 16 and under with history or signs of head injury
• Validated in Australian setting but not as widely utilised or extensively validated
• CT required if any of the following:
  • Witnessed LOC >5min
  • Amnesia (anterograde or retrograde) >5 min
  • Abnormal drowsiness
  • 3 or more vomits after head injury
  • Suspicion of NAI
  • Seizure after head injury in patient with no hx of epilepsy
  • GCS <14 or <15 (in <1yo)
  • Suspicion of penetrating/depressed skull fracture or tense fontanelle
  • Signs of basal skull fracture
  • Focal neurology
  • Bruise/swelling/laceration >5cm if <1yo
  • High-speed RTA as pedestrian, cyclist or passenger (>40mph)
  • Fall >3m
  • High-speed projectile

CATCH

• Patients up to 16yo with minor head injury, GCS of at least 13, injury within 24 hours, plus at least one of:
  • Blunt trauma with witnessed LOC
  • Amnesia
  • Witnessed disorientation
  • Vomiting 2+ times at least 15 minutes apart
  • Persistent irritability if <2yo
• DO NOT USE if:
  • Penetrating skull injury, depressed skull fracture, focal neuro deficits, developmental delay, suspected NAI, re-evaluation after prior head injury or pregnant
• Not as well validated as PECARN
• CT recommended if high risk or medium risk score
  • High risk predicts neurosurgical intervention while medium risk predicts brain injury on CT scan
• Criteria
  • GCS <15 at 2 hours
  • Suspected open or depressed skull fracture
  • Worsening headache
  • Irritability
  • Any sign of basal skull fracture
  • Large boggy scalp haematoma
  • Dangerous mechanism: MVC, fall >0.9m or 5 stairs, fall from bicycle with no helmet

PREDICT ALGORITHM

5: Other signs of altered mental status: agitation, drowsiness, repetitive questioning, slow response to verbal communication

6: Severe mechanism of injury: MVA with patient ejection or rollover, death of another passenger, pedestrian or cyclist without helmet struck by motor vehicle, fall >=1m (<2yo) or >1.5m(>2yo), head struck by high impact object

12: Observation is to be half-hourly for 2 hours, then q1h up to 4 hours. After 4 hours, continue q2h as long as patient is in hospital

Benefits of this are clarity around delayed presentations (24-72hrs) being managed as for those <24 hours from time of injury AND definition of a very low risk group who DO NOT need observation.

QCH Guideline

Low-risk

• No risk factors
• If remains GCS 15 = Very low risk
• Can be discharged home with minor head injury information
• This is not NO RISK and thus requires education and advice to return in case of any concussive-type symptoms or concerns

Intermediate-risk

• Warrant either further observation or CT
• Scanning all of these children is almost certainly unnecessary and would result in a significantly higher scan rate across Australia and New Zealand with no appreciable increase in identification rates of clinically significant traumatic brain injury
• Factors that may influence the decision to perform a CT are:
  • Multiple risk factors
  • Clinician experience
  • Worsening or unresolving symptoms
  • Age of the child (need for sedation in younger children)
  • Availability of local resources for imaging (+/- sedation)

How long to observe?

• The NICE guidelines from the UK advice minimum 4 hours for intermediate-risk children, while a large retrospective study from Canada found the vast majority of children with clinically significant traumatic brain injuries were symptomatic by 6 hours

Infant cerebral USS

• Not a practical option at this stage but in the future may be applicable to infants with open fontanelles
• Can also be useful in identifying skull fractures in the PECARN children in the moderate risk group, identifying a fracture would necessitate a CT while ruling out of skull fracture may allow the clinician to more safely observe the child only
  • However, if skull fracture is suspected, CT is warranted anyway as per guidelines

Paediatric radiation exposure

• Retrospective cohort study found a single head CT in children/adolescents <22yo tripled the risk of leukaemia and brain tumours
• In the 10 years after a scan of a child <10yo, one extra case of leukaemia and one extra brain tumour for every 10 000 scans

Skull x-ray

• Children with skull fracture on X-ray have higher rates of ICH
• Many children with ICH do not have a skull fracture
• Sensitivity and specificity of skull X-ray are around 21 and 53% respectively
• May have a role in children <2yo with head trauma, in whom keeping still for CT scanning may pose its own risks

Management of severe head injury

• ABCDE
  • Ketamine or fentanyl safest for intubation
  • Ketamine raises ICP but also raises MAP so CPP should be maintained (better than drop in MAP with propofol for instance)
• Normothermia
• Normoglycaemia
• Treat seizures
  • Occur in 5-15% of all TBI patients (up to 40% if persistent GCS <8)
  • 80% occur within first 24 hours
  • Prevention can be considered if GCS <10, depressed skull fracture or penetrating head wound, subdural/extradural or intracerebral haematomas or cortical contusion
  • Discuss with neurosurgery
• Treat raised ICP
  • Mannitol 0.25 – 1g/kg IV repeated as required
  • Hypertonic saline 3% – 3mL/kg bolus (5mL/kg) repeated as required
    • No controlled trial demonstrating improved outcomes
    • Expect to raise serum sodium by 2-3mmol/L

CT head

• Poor prognostic signs
  • Subdural haemorrhage (indicates severe trauma and damage to underlying brain tissue)
  • Ablated basal cisterns and midline shift
  • Lack of reversal of grey-white differentiation

CPP

• Normal ICP is <5mmHg at 2 years and <10mmHg at 5 years
• For adolescents, target CPP 60-70mmHg
• Physiological compensation for raised ICP in children
  • Displacement of CSF into distensible spinal subarachnoid space
  • Compression of intracranial venous system
  • Increased CSF resorption
  • Reduced CSF production
  • Stretching of dura, unfused skull bones and skin (<18mo)

NAI

• Retinal haemorrhages are pathognomonic
  • If multiple, involving >1 layer of retina and extending to periphery, more likely abusive in origin
  • Present in 80% of NAI
• 60% of subdural haemorrhages in children are due to NAI
• Fractures seen in 35% of NAI

Last Updated on December 4, 2024 by Andrew Crofton