ADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Initial Management of:: Closed Head Injury in Adults
HE
AD
IN
JU
RY
GU
IDE
LIN
E
TraumaCvr_FullReport.qxd 3/5/07 6:35 AM Page 1
Suggested citation:
Dr Duncan Reed 2007, Adult Trauma Clinical Practice Guidelines, Initial Management
of Closed Head Injury in Adults, NSW Institute of Trauma and Injury Management.
Author
Dr Duncan Reed (FACEM), Director of Trauma, Central Coast Health
Editorial team
NSW ITIM Clinical Practice Guidelines Committee
Mr Glenn Sisson (RN), Trauma Clinical Education Manager, NSW ITIM
Ms Joan Lynch (RN), Project Manager, Trauma Service, Liverpool Hospital
Assoc. Prof. Michael Sugrue, Trauma Director, Trauma Service, Liverpool Hospital
Ms Nichole Woodward, Trauma Clinical Nurse Consultant, Central Coast Health
Dr Declan Stewart, Emergency Physician, Central Coast Health
Acknowledgments
Ms Gail Long, Secretary, Emergency Department, Gosford Hospital
Assoc. Prof. Michael Fearnside, Neurosurgeon, Director Surgical Services,
Westmead Hospital
Art and Design Unit, Gosford Hospital
This work is copyright. It may be reproduced in whole or in part for study
training purposes subject to the inclusion of an acknowledgement of the source.
It may not be reproduced for commercial usage or sale. Reproduction for purposes
other than those indicated above requires written permission from the NSW Insititute
of Trauma and Injury Management.
© NSW Institute of Trauma and Injury Management
SHPN (TI) 070020
ISBN 978-1-74187-078-7
For further copies contact:
NSW Institute of Trauma and Injury Management
PO Box 6314, North Ryde, NSW 2113
Ph: (02) 9887 5726
or can be downloaded from the NSW ITIM website
http://www.itim.nsw.gov.au
or the
NSW Health website http://www.health.nsw.gov.au
January 2007
TraumaCvr_FullReport.qxd 3/5/07 6:35 AM Page 2
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE i
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Important notice!'The Initial Management of Closed Head Injury in Adults' clinical practice guidelines areaimed at assisting clinicians in informed medical decision-making. They are not intended toreplace decision-making. The authors appreciate the heterogeneity of the patient populationand the signs and symptoms they may present with and the need to often modifymanagement in light of a patient's co-morbidities.
The guidelines are intended to provide a general guide to the management of specifiedinjuries. The guidelines are not a definitive statement on the correct procedures, rather theyconstitute a general guide to be followed subject to the clinicians judgement in each case.
The information provided is based on the best available information at the time of writing, which is December 2004. These guidelines will therefore be updated every five years and consider new evidence as it becomes available.
These guidelines are intended for use in adults only.
All guidelines regarding pre-hospital care should be read and considered in conjunction with NSW Ambulance Service protocols.
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page i
::
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page ii
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE iii
HE
AD
IN
JU
RY
GU
IDE
LIN
E
ContentsAlgorithm 1 :: Initial Management of ‘Adult Closed Head Injuries’ ............................1
Algorithm 2 :: Initial Management of ‘Adult Mild Head Injury’....................................2
Summary of Guidelines....................................3
1 Introduction..............................................15
2 Methods...................................................17
3 What is the definition of Mild Head Injury? ................................19
4 What are the clinically important complications of Mild Head Injury?.........20
5 How should patients with Mild Head Injury be assessed? ...............22
6 Which patients with Mild HeadInjury require a CT scan?.........................28
7 What to do with high risk Mild HeadInjury patients when CT scan isunavailable? .............................................36
8 When can patients with Mild Head Injury be safely discharged home? ...................37
9 What are the proven treatments forpatients with ‘Moderate’ to ‘Severe’ Head Injuries?..........................................39
10 When should patients with closed head injury be transferred to hospitals with neurosurgical facilities? ...........................41
Evidence tables
Evidence Table 1. Definitions of Mild Head Injury .........44
Evidence Table 2. Initial GCS versus
Abnormal CT/Neurosurgery ..............45
Evidence Table 3. Risk factors for intracranial
injury in Mild Head Injury .................46
Evidence Table 4. Safe discharge of Mild Head Injury ...51
Evidence Table 5. Post concussive symptoms
and Mild Head Injury........................54
Evidence Table 6. Management of Severe Head Injury..56
Appendices
APPENDIX A ::
Westmead PTA Scale ...................................................60
APPENDIX B ::
Glasgow Coma Scale ...................................................62
APPENDIX C ::
Mild Head Injury Discharge Advice...............................63
APPENDIX D ::
NSW Brain Injury Rehabilitation Program (BIRP)............65
APPENDIX E ::
List of Search Terms to identify studies.........................66
References .....................................................67
List of tables
Table 1.1 Summary of closed head
injury classification and outcome ................16
Table 2.1 Levels of evidence.......................................18
Table 2.2 Codes for the overall assessment
quality of study checklists ...........................18
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page iii
::
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page iv
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 1
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Algorithm 1 :: Initial management of ‘Adult Closed Head Injuries’
GCS 3-8 GCS 9-13 GCS 14-15
Severe Head Injury (10%)1 Early intubation.1 Supportive care ABCDE's.1 Prevent secondary injury.1 Early CT scan.1 Early neurosurgical consult.1 Early retrieval consult if required.1 Consider ICP monitoring.1 Anticonvulsants optional.1 ICU admission.1 Routine brain injury rehabilitation consult.
NB. Minimum supportive care aims to prevent secondary brain injury:
1 PaO2 >60 1 SaO2 >90 1 PaCO2 35-401 Systolic BP >90 1 Head up 30º
Moderate Head Injury (10%)1 Supportive care ABCDE's.1 Prevent secondary injury.1 Early CT Scan.1 Period of clinical observation.1 Intubation if required to ensure adequate
oxygenation and ventilation or to safely perform CT scan.
1 Early neurosurgical consult if not clinically improving or abnormal CT scan.
1 Early retrieval consult if required.1 Admit to hospital unless early rapid clinical
improvement, normal CT scan and absence of other risk factors as listed below.
1 Consider routine post traumatic amnesia testing and referral to a brain injury rehabilitation service (or neurologist) due to high risk of cognitive behavioural social sequelae.
Mild Head Injury (80%)1 Initial period of clinical observation.1 Assess for risk factors of significant
intracranial injury as listed below.1 CT scan if GCS <15 at 2 hours post
injury or if other risk factors present.1 Admit to hospital if clinically not improving
at 4 hours post injury or abnormal CT scan.1 Consider hospital admission if elderly
(>65 years), known coagulopathy (warfarin, alcoholic), intoxicated or socially isolated.
1 Discharge home with head injury advice at 4 hours post injury if clinically improving with either no risk factors or normal CT scan.
1 Refer to local doctor for routine follow up and monitoring of any post concussion symptoms.
1 Consider referral to a brain injury rehabilitation service (or neurologist) if still has post traumatic amnesia after 24 hours or if still has post concussion symptoms after 4 weeks.
NB. See also separate Mild Head Injury Algorithm.
Initial Management of Adult Closed Head Injuries
Risk factors indicating potentially significant Mild Head Injury1 Persistent GCS <15 at 2 hours post injury.1 Deterioration in GCS.1 Focal neurological deficit.1 Clinical suspicion of skull fracture.1 Prolonged loss of consciousness (>5 min).1 Prolonged anterograde / retrograde amnesia.
What to do if a Closed Head Injury patient acutely deteriorates?
When should patients with Closed Head Injury be transferred to hospitals with neurosurgical facilities?
1 Post traumatic seizure.1 Persistent abnormal alertness / behaviour / cognition.1 Persistent vomiting (2 or more occasions).1 Persistent severe headache.1 Known coagulopathy (eg warfarin / alcoholic).1 Age >65 years.
1 Multi-system trauma.1 Dangerous mechanism.1 Clinically obvious drug or alcohol intoxication.1 Known neurosurgery / neurological impairment.1 Delayed presentation or representation.
Clinical approach1 Re-assess ABCDE's to exclude
non head injury cause.1 Supportive care of ABCDE's
(NB. oxygenation and perfusion).1 Consider early intubation and
short term hyperventilation to PaCO2 25-30.
1 Immediate CT scan if available.1 Consult neurosurgical service.1 Consult retrieval service early.1 Consider use of mannitol
boluses (1g/kg), in consultation with neurosurgical service, to reduce ICP for transfer.
1 Consider local burr holes in consultation with neurosurgical service if transfer likely to take more than 2 hours.
Indications of deterioration1 GCS falls by two or more points.1 Develops dilated pupil(s).1 Develops focal neurological
deficit.1 Delayed or focal seizure.1 Cushings response – bradycardia and hypertension.
Clinical approach1 When in doubt consult you
regional neurosurgical service.1 Patients with closed head
injuries should be observed in facilities that can manage any complications that are likely to arise. Clinical judgment regarding risk of deterioration is required and neurosurgical consultation may be appropriate.
1 Patients with closed head injuries should be transferred to the nearest appropriate hospital with neurosurgical facilities if there is significant risk of intracranial injury. Transfer of patients to hospitals with CT scan facilities but without neurosurgical services should be avoided.
Potential indications for transfer1 Patient with severe head injury
(GCS 3-8).1 Patient with moderate head injury
(GCS 9-13) if: – clinical deterioration – abnormal CT scan – normal CT scan but
not clinically improving – CT scan unavailable.1 Patient with mild head injury
(GCS 14-15) if:– clinical deterioration– abnormal CT scan– normal CT scan but not clinically improving– high risk mild head injury with CT scan unavailable, particularly if:
- persistent abnormal GCS- focal neurological deficit- clinical suspicion of skull fracture- persistent abnormal mental status, vomiting or severe headache at 4 hours post injury.
AMRS (adult) 1800 650 004
NETS (children) 1300 362 500Regional Neurosurgical tel.
'formerly the MRU'
NS
W In
stitu
te o
f Tra
uma
and
Inju
ry M
anag
emen
t ©
Jan
uary
200
7 S
HP
N: (
TI) 0
7003
1
Initial Assessment and Stabilisation of ABCDE'sTrauma Team activation if initial GCS 3-13 or otherwise indicated
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 1
PAGE 2 Inital Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EH
EA
D I
NJ
UR
Y G
UID
EL
INE
Algorithm 2 :: Initial management of ‘Adult Mild Head Injury’
Clinical observation until at least four hours post time of injury.
Initial GCS 14-15 on arrival following blunt trauma. Stabilise ABCDE’s then assess risk factors.
Low risk Mild Head Injury
Normal during observation period.
Normal CT scan. Clinical symptoms improving at 4 hours post time of injury.
Normal CT scan. Clinical symptoms NOT improving at 4 hours post time of injury.
Socially safe for discharge for home observation if:1 Responsible person available to take home and observe.1 Able to return if deteriorates.1 Discharge advice is understood.
Clinically safe for discharge for home observation if:1 Normal alertness / behaviour / cognition.1 Clinically improving after observation.1 Normal CT scan or no indication for CT scan.1 Clinical judgment required if elderly and /
or known coagulopathy due to increased risk of delayed subdural haematoma.
Consider urgent transfer for CT scanning if:1 Persistent GCS <15 at 2 hours post injury.1 Deterioration in GCS.1 Focal neurological deficit.1 Clinical suspicion of skull fracture.1 Persistent abnormal mental status, vomiting or
severe headache at 4 hours post time of injury.
Admit for prolonged hospital observation and consult regional neurosurgical service regardingfurther management and disposition.
CT scan unavailable.
Initial Management of Adult Mild Head Injury
High risk Mild Head InjuryAny of ...1 Persistent GCS <15 at 2 hours post injury.1 Deterioration in GCS.1 Focal neurological deficit.1 Clinical suspicion of skull fracture. ::11 Prolonged loss of consciousness (>5 mins).1 Prolonged anterograde / retrograde amnesia (>30 mins).1 Post traumatic seizure. ::21 Persistent abnormal alertness / behaviour / cognition.1 Persistent vomiting (2 or more occasions).1 Persistent severe headache.1 Known coagulopathy (eg warfarin / alcoholic).1 Age >65 years. ::31 Multi-system trauma. ::41 Dangerous mechanism. ::51 Clinically obvious drug or alcohol intoxication. ::61 Known neurosurgery / neurological impairment. ::71 Delayed presentation or representation. ::8
Discharge for home observation and provide written discharge advice.
Abnormal alertness / behaviour / cognition.Clinically deteriorates or not improving.
Abnormal CT scan.
Note – Clinical judgement required::1 Clinical suspicion of skull fracture includes – history of focal blunt assault or injury; large scalp lacerations or haematomas; signs of base of skull fracture
– haemotympanum / CSF leak / raccoon eyes / Battles sign.
::2 Post traumatic seizures – prolonged, focal or delayed seizures are significant risk factors for intracranial injury but brief generalised seizures immediately following head injury are less concerning.
::3 Age >65yrs – elderly patients have increased risk of significant intracranial injury; routine CT scanning unless totally asymptomatic patient with no other risk factors.
::4 Multi-system trauma – beware patients with unstable vital signs or distracting injuries, as significant head injuries are easily missed.
::5 Dangerous mechanisms include MVA ejection / rollover; pedestrians / cyclists hit by vehicle; falls >own height or five stairs; falls from horses / cycles etc; focal blunt trauma, eg bat / ball / club.
::6 Clinically obvious drug or alcohol intoxication with altered mental status is an indication for CT scanning but drug or alcohol ingestion with normal mental status is not.
::7 Known neurosurgery/neurological impairment – conditions such as hydrocephalus with shunt or AVM or tumour or cognitive impairment from any cause make clinical assessment less reliable and may increase risk of intracranial injury.
::8 Delayed presentation or representation – consider both intracranial injury and post concussion symptoms and have a low threshold for CT scanning.
Indication for early CT scan and prolonged clinical observation.
NS
W In
stitu
te o
f Tra
uma
and
Inju
ry M
anag
emen
t ©
Jan
uary
200
7 S
HP
N: (
TI) 0
7003
2
All of ...1 GCS 15 at 2 hours post injury.1 No focal neurological deficit.1 No clinical suspicion of skull fracture. ::11 Brief loss of consciousness (<5 mins).1 Brief anterograde / retrograde amnesia (<30 mins).1 No post traumatic seizure.1 Mild nausea or single episode of vomiting.1 Mild headache.1 No known coagulopathy.1 Age <65 years.1 Isolated head injury without dangerous mechanism.1 No drug or alcohol ingestion.1 No known neurosurgery / neurological impairment.
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 2
What is the definition ofMild Head Injury?
A patient with an initial GCS score of 14-15 on arrival at hospital III-2following acute blunt head trauma (with or without a definite history of loss of consciousness or amnesia).
Typical characteristics
:: Direct blow to the head or acceleration / deceleration injury.
:: Transient loss of consciousness or amnesia.
:: Transient abnormal alertness, behaviour or cognition.
:: Neurosurgical intervention rare.
:: Post concussion symptoms common.
:: Long term functional outcome good.
Risk stratification III-2
Mild Head Injury may be further sub-classified into 'High' and 'Low' risk groups, based on the risk of having an intracranialinjury requiring neurosurgical intervention. Stratification of
'high' and 'low' risk of intracranial injury is based on:1,6-9,14,16-19,23-46,62
:: initial GCS on admission and at two hours post injury
:: the duration of loss of consciousness or amnesia
:: the presence or absence of other specified risk factors.
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 3
HE
AD
IN
JU
RY
GU
IDE
LIN
EDEFINITION LEVEL OF EVIDENCE
:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 3
PAGE 4 Initial Management of Closed Head Injury in Adults :: NSW ITIM
Clinicians and patients should be aware of both the risk of neurosurgical III-2intervention and the risk of cognitive-behavioural-social sequelae following Mild Head Injury.
Clinicians and patients should also be aware that the absence of a structurallesion on CT scan following Mild Head Injury does not exclude the possibility of significant cognitive-behavioural-social sequelae.
Acute life-threatening complications requiring neurosurgical IVintervention are rare in Mild Head Injury patients:16,17,19,24-32,47,62
:: 'Low risk' Mild Head Injury range 0-3%.
:: 'High risk' Mild Head Injury range 0.5-6.5%.
Post concussion symptoms are common in Mild Head Injury patients and may IVhave significant cognitive-behavioural-social impacts on both patients and their families:3,8,9,36,48-52
:: Typical post concussion symptoms include:– headaches– dizziness – fatigue– memory impairment– poor concentration– mood swings– behavioural changes – social dysfunction.
:: Up to 50% of patients with Mild Head Injury may have significant post concussion symptoms which can last several weeks.
:: About 10% of patients with Mild Head Injury will have persistent disabling post concussion symptoms.
What are the clinically importantcomplications of Mild Head Injury?
GUIDELINE LEVEL OF EVIDENCE
HE
AD
IN
JU
RY
GU
IDE
LIN
E:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 4
Mild Head Injury patients should have a minimum of hourly III-2observations for four hours post injury.
:: These observations include:– GCS– alertness / behaviour / cognition– pupillary reactions– vital signs.
Serial neurological observations should be continued on any Mild Head Injury III-2patients who fail to clinically improve at four hours post injury or who are found to have structural lesions requiring hospital admission.7,53
Assessment for post traumatic amnesia (PTA) should be performed III-2on any Mild Head Injury patients who fail to clinically improve at four hours post injury or who are found to have structural lesions requiring hospital admission.3,55
Structured clinical assessment using clinical decision rules can identify those III-2patients at increased risk of intracranial injury requiring further investigation.16,17,19,62
Skull x-rays are not sufficiently sensitive to be used as a routine Iscreening investigation to identify significant intracranial lesions.37
CT scanning is the most appropriate investigation for the exclusion III-2of neurosurgically significant lesions in mild head injured patients.
CT scanning is indicated for those Mild Head Injury patients III-2identified by structured clinical assessment as being at increasedrisk of intracranial injury.16,17,19,25,38,39,62
If structured clinical assessment indicates the risk of intracranial injury III-2is low, the routine use of CT scanning is neither clinically beneficial nor cost effective.16,17,19,25,38,39
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 5
HE
AD
IN
JU
RY
GU
IDE
LIN
E
How should patients withMild Head Injury be assessed?
GUIDELINE LEVEL OF EVIDENCE
:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 5
PAGE 6 Initial Management of Closed Head Injury in Adults :: NSW ITIM
Patients with ‘High Risk’ Mild Head Injury requiring CT scan
The following risk factors identify patients with Mild Head Injury III-2(initial GCS 14-15) at increased risk of clinically significant lesions requiring acute neurosurgical intervention or prolonged observationin hospital. These patients should have early CT scanningif available, if they have any of the following features:
Any of...
Initial assessment
:: Persistent GCS <15 at two hours post injury.**
:: Focal neurological deficit.
:: Clinical suspicion of skull fracture.
:: Prolonged loss of consciousness (>5min).
:: Prolonged anterograde or retrograde amnesia (>30min).
:: Post traumatic seizure.
:: Repeated vomiting (>2 occasions).
:: Persistent severe headache.
:: Known coagulopathy.
:: Age >65 years (clinical judgment appropriate if no other risk factors present).
After a period of observation (four hours post injury)
:: Any deterioration in GCS.
:: Persistent abnormal mental status (abnormal alertness, behaviour or cognition).**
:: Any patient who fails to clinically improve.
Clinical judgment required if:
:: age >65 years
:: drug or alcohol ingestion
:: dangerous mechanism
:: multi-system trauma
:: known neurosurgery / neurological impairment
:: delayed presentation or representation.
** NOTE: Includes patients with abnormal GCS due to drug or alcohol ingestion.
Which patients withMild Head Injury require a CT scan?
GUIDELINE LEVEL OF EVIDENCE
HE
AD
IN
JU
RY
GU
IDE
LIN
E:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 6
Patients with ‘Low Risk’ Mild Head Injury not requiring CT scan
The following features indicate patients with Mild Head Injury III-2(initial GCS 14-15) at low risk of clinically significant lesions requiring acute neurosurgical intervention or prolonged observation in hospital.These patients should not routinely have CT scanning if they have all of the following features:
All of...
Initial assessment
:: GCS 15 at two hours post injury.
:: No focal neurological deficit.
:: No clinical suspicion of skull fracture.
:: Brief loss of consciousness (<5min).
:: Brief anterograde or retrograde amnesia (<30min).
:: No post traumatic seizure.
:: Mild nausea or single episode of vomiting.
:: No persistent severe headache.
:: No known coagulopathy.
:: Age <65 years.
After a period of observation (four hours post injury)
:: Normal mental status including alertness and behaviour and cognition.
:: No deterioration during observation.
:: Clinically improving.
Clinical judgment required if:
:: age >65 years
:: drug or alcohol ingestion
:: dangerous mechanism
:: multi-system trauma
:: known neurosurgery / neurological impairment
:: delayed presentation or representation.
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 7
HE
AD
IN
JU
RY
GU
IDE
LIN
EGUIDELINE LEVEL OF EVIDENCE
:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 7
PAGE 8 Initial Management of Closed Head Injury in Adults :: NSW ITIM
‘High Risk’ Mild Head Injury patients should be admitted for prolonged III-2hospital observation and considered for transfer for CT scanning. Patients at highest risk of intracranial injury who should be transferred for CT scanning include those with:16-19,27-34,37-39,47,70,71,76
:: persistent GCS <15 at two hours post injury.
:: deterioration in GCS.
:: focal neurological deficit.
:: clinical suspicion of skull fracture.
:: persistent abnormal mental status, vomiting or headache at four hours post injury.
Patients at ‘Highest Risk’ of intracranial injury should be discussed with Consensusthe regional neurosurgical service and a management plan established.
If patients are transferred for CT scanning they should ideally be transferred Consensusto a hospital with neurosurgical facilities to avoid secondary transfer.
A skull x-ray may be useful to confirm the presence of a skull fracture Ithat mandates an early CT scan due to the increased risk of deterioration.
All ‘High Risk’ patients who cannot have CT scanning should at a minimum Consensushave prolonged observation in hospital for at least 24 hours and until clinically improving.
What to do with ‘High Risk’Mild Head Injury patients when CT scan is unavailable?
GUIDELINE LEVEL OF EVIDENCE
HE
AD
IN
JU
RY
GU
IDE
LIN
E:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 8
Mild Head Injury patients can be discharged for home observation after III-2initial period of in-hospital observation if they meet the following clinical, social and discharge advice criteria.16,17,19,24,26,27,29,32,66,68,69,77-79
Clinical criteria III-2
:: Normal mental status and behaviour with clinically improving minor post concussion symptoms after observation until at least four hours post injury.
:: No clinical risk factors indicating the need for CT scanning or normal CT scan if performed due to risk factors being present.
:: No clinical indicators for prolonged hospital observation (irrespective of CT scan result) such as:– clinical deterioration– persistent abnormal GCS or focal neurological deficit– persistent abnormal mental status or behaviour– persistent severe post concussion symptoms– persistent drug or alcohol intoxication– presence of known coagulopathy (relative)– presence of multi-system injuries (relative)– presence of intercurrent medical problems (relative)– age >65 (relative).
Social criteria Consensus
:: Responsible person available to take patient home.
:: Responsible person available for home observation.
:: Patient able to return easily in case of deterioration.
:: Written and verbal discharge advice able to be understood.
Discharge advice criteria Consensus
:: Discharge summary for local doctor.
:: Written and verbal head injury advice given to patient and nominated responsible person covering:– symptoms and signs of acute deterioration– reasons for seeking urgent medical attention– typical post concussion symptoms– reasons for seeking routine follow up.
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 9
HE
AD
IN
JU
RY
GU
IDE
LIN
E
When can patients with Mild HeadInjury be safely discharged home?
GUIDELINE LEVEL OF EVIDENCE
:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 9
PAGE 10 Initial Management of Closed Head Injury in Adults :: NSW ITIM
Initial management of ‘Severe Head Injuries’ (GCS 3-8) III-2
Standard care15,22,80-87
:: Initial systematic resuscitation of ABCDE's.
:: Early CT scanning to identify neurosurgically correctable focal intracranial haematomas.
:: Prevention of secondary brain injury by avoiding hypoxaemia (oxygen saturation <90%) and hypotension (systolic BP <90).
:: Supportive care of ABCDE's with appropriate attention, posturing (30o head up), basic nursing care and avoidance of hyperventilation.
:: Early neurosurgical consult.
:: Use of ICP monitoring to guide management of cerebral perfusion pressure in patients with severe brain injury.
:: Optional use of anticonvulsants to prevent early post traumatic seizures.
:: Routine brain injury rehabilitation consult.
:: ICU admission.
Acute neurological deterioration
:: Resuscitation and stabilisation of ABCDE's.
:: Short term hyperventilation to PaCO2 25-30.
:: Mannitol 1g/kg IV Bolus.
:: Early CT scan with neurosurgical intervention as required.
Poor prognostic indicators
:: Low GCS (especially motor component).
:: Age >60 years (prognosis deteriorates with increasing age).
:: Absent pupillary reflexes (after systemic resuscitation).
:: Hypotension (systolic BP <90).
:: Hypoxaemia (oxygen saturation <90%).
What are the proven treatments for patients with ‘Moderate’ to ‘Severe’ Head Injuries?
GUIDELINE LEVEL OF EVIDENCE
HE
AD
IN
JU
RY
GU
IDE
LIN
E:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 10
Initial management of ‘Moderate Head Injuries’ (GCS 9-13) III-2
Standard care15,22,80-87
:: Initial assessment and resuscitation of ABCDE's.
:: Early CT scanning to identify neurosurgically correctable focal intracranial haematomas.
:: Period of ED observation.
:: Prevention of secondary brain injury by avoiding hypoxaemia (O2 saturation <90%) and hypotension (systolic BP <90).
:: Supportive care of ABCDE's.
:: Admit for prolonged hospital observation (24-48 hours) unless rapid clinical improvement, normal CT scan and absence of other risk factors.
:: Early neurosurgical consult if not clinically improving and/or abnormal CT scan.
:: Routine post traumatic amnesia (PTA) testing.
Outcome
:: Approximately 80% of moderate head injuries improve while 20% deteriorate and require management as per severe head injuries.
:: The majority of patients who suffer moderate head injuries will have some degree of cognitive behavioural social sequelae and should be considered for routine follow up with a brain injury rehabilitation service or a neurologist (see Appendix D of the Adult Trauma Clinical Practice Guideline, Initial Management of Closed Head Injury in Adults).
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 11
HE
AD
IN
JU
RY
GU
IDE
LIN
EGUIDELINE LEVEL OF EVIDENCE
:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 11
PAGE 12 Initial Management of Closed Head Injury in Adults :: NSW ITIM
Patients with severe head injuries (GCS 3-8). Consensus
Patients with moderate head injuries (GCS 9-13) if: Consensus
i clinical deterioration
ii abnormal CT scan
iii normal CT scan but not clinically improving
iv CT scan unavailable.
Patients with mild head injuries (GCS 14-15) if: Consensus
i clinical deterioration
ii abnormal CT scan
iii normal CT scan but not clinically improving
iv high risk Mild Head Injury with CT scan unavailable.
NOTE: Consult neurosurgical / retrieval service early.
When should patients with Closed Head Injury be transferred to hospitals with neurosurgical facilities?
GUIDELINE LEVEL OF EVIDENCE
HE
AD
IN
JU
RY
GU
IDE
LIN
E:: SUMMARY OF GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 12
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 13
::
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 14
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 15
HE
AD
IN
JU
RY
GU
IDE
LIN
E
1 IntroductionTrauma is the leading cause of death and disability
in children and young adults in New South Wales and
closed head injuries cause a significant proportion of this
burden.1,2 Closed head injuries often result in lifelong
physical, cognitive, behavioural and social dysfunction for
patients who in turn may place major social and financial
burdens on their families and society.3 Recent Australian
figures indicate there are approximately 150 patients per
100,000 population admitted to hospital each year with
closed head injuries.3-5 Worldwide figures suggest an
incidence range of 200-350 per 100,000 population for
patients with closed head injury with mild head injury
accounting for 80%.6 Despite the fact that closed head
injuries are common, the classification and management
of head injures remains surprisingly controversial and
subject to variation in clinical practice [6-10]. Due to the
large numbers of patients involved it has been estimated
that even small improvements in closed head injury
management could have significant impact.11
Furthermore, it has been suggested that the greatest
improvements can be made in the better management
of those patients with mild to moderate head injury
rather than those with severe head injury.12
Much of the controversy that exists about closed head
injury management stems from the combination of a
lack of uniformity in definitions with a paucity of large
well designed studies in the area.11,13,14 ‘Head injury’ is
typically used to describe the initial clinical presentation
whilst ‘traumatic brain injury’ is used to describe the
subsequent functional outcome. It is difficult to find two
studies that define mild head injury in exactly the same
way so comparison of data becomes difficult.6-9,14
Similarly, comparison of data in moderate to severe head
injury studies is made difficult because controversy exists
about how and when best to apply Glasgow Coma
Score (GCS) to sedated or intubated patients.15 Perhaps
most significantly there have been very few large
prospective randomised controlled trials of sufficient
power and quality to guide management.11,13,14
However, in the past few years there has been some
progress in working toward uniform definitions and
some better quality trials and meta-analyses have been
published.6-9,16-19,62
The variety of clinical practice observed worldwide
cannot be explained solely by the lack of uniformity
of definitions and good quality studies. Much of the
variation in management strategies between the USA,
Canada, Europe and Australasia is driven by local issues
such as the availability of resources and the medico-legal
environment.6, 20 Thus the USA has higher rates of CT
scanning for mild head injuries compared to Canada,
Europe and the UK. Even within countries and within
institutions, considerable variation in practice has been
shown to exist.10,12,20,21 Whilst some variation in clinical
practice is to be expected, the introduction of clinical
practice guidelines can potentially improve care and
ensure adequate access to resources for more isolated
areas.6,19 Furthermore, clinical guidelines can potentially
reduce unnecessary tests and hospital admissions for
mild head injury patients, by identifying those patients
at low risk of neurosurgically significant
lesions.6,14,18,19,62
These guidelines are intended to assist clinicians
throughout NSW in delivering optimal care to patients
with closed head injuries. They are intended to provide a
clinically practical approach to the initial management of
closed head injuries based on the current best available
evidence. However, as with all guidelines, it should be
remembered that they are a clinical tool and should not
replace clinical judgement.
1.1 Defining closed head injuryThis guideline uses the terms ‘closed head injury’ and
‘mild, moderate or severe head injury’ to identify and
classify patients on arrival to hospital. The outcome
following ‘closed head injury’ will vary from rapid
complete recovery to a mixture of structural lesions and
functional deficits ranging from trivial to life threatening.
Important functional deficits following ‘closed head
injury’ range from persistent post concussion symptoms
and post traumatic amnesia to a variety of disabling
physical-cognitive-behavioural-social sequelae.
As this guideline concentrates on the initial management
of the target population it was felt that the term
‘head injury’ was more relevant to the initial clinical
presentation than the term ‘traumatic brain injury’ that
essentially refers to the subsequent functional outcome.
It was also felt that the clinicians at whom this guideline
is aimed would be far more familiar and comfortable
with using the term ‘head injury.’
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 15
1.2 Classification of closed head injuryThis guideline has used the ATLS/EMST classification of head injuries based on initial GCS on admission to hospital after
initial resuscitation and stabilisation.22 This classification is widely used clinically and the ATLS/EMST course provides the
common language of trauma care throughout the world. This system classifies patients with initial GCS score of 13 in the
moderate head injury group due to the patients having similarly patterns of intracranial injury and cognitive behavioural
sequelae. The following table is a summary based on the Brain Trauma Foundation and ATLS/EMST data gives a rough
guide to classification and outcome.15,22
PAGE 16 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
The definition of ‘closed head injury’ in this document is as follows:
‘A patient presenting to hospital following acute blunt head trauma with or without a definite history loss of consciousness or amnesia.’
Typical characteristics:
:: Direct blow to the head or acceleration / deceleration injury.
:: Variable duration of initial loss of consciousness or amnesia.
:: Variable duration and severity of diffuse and / or focal neurological deficits.
Specific exclusions:
:: Clinically obvious penetrating head injury.
:: Non-traumatic brain injury.
CLOSED HEAD INJURY DEFINITION
Initial GCS 14-15 9-13 3-8
% of total 80 10 10
Abnormal CT scan (%) 10-15 40-50 90
Neurosurgical intervention (%) (excluding ICP) 1-3 10-15 40-50
Mortality (%) 1 <1 10-15 20-803
Good functional outcome (%) 1 >902 50 10-503
Mild Moderate Severe
1 Generally the lower the GCS the worse the prognosis.
2 Approximately 10% will have persistent post concussive symptoms at three to six months.
3 Outcome deteriorates with increasing age – ‘children do better and elderly do worse’.
Table 1.1. Summary of closed head injury classification and outcome
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 16
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 17
2 Methods2.1 Scope of the guidelinesThe guidelines are intended for use by clinicians
managing patients with closed head injuries in major
trauma services, regional, urban and rural hospitals.
The Initial Management of Closed Head Injury Guidelines
are concerned with the initial care of the mild, moderate
and severely head injured patient. The guideline
will make evidence based recommendations on
the diagnosis, resuscitation, and disposal of
patients with closed head injuries.
These guidelines however, are not prescriptive nor
are they rigid procedural paths. It is recognised that
the guidelines may not suit all patients in all clinical
situations. The guidelines rely on individual clinicians
to decipher the needs of individuals. They aim to
provide information on what decisions can be made,
rather than dictate what decisions should be made.
2.2 Aims and objectivesThe broad objectives of the guidelines are to reduce
morbidity and mortality in adult patients with closed
head injuries by providing clinicians with practical
evidenced based guidelines to assist them in managing
such patients. It is also hoped that the guidelines may
prevent unnecessary diagnostic tests and hospital
admissions especially in the mild head injury group.
At the start of the process of creating the guidelines
the clinicians on the Clinical Guidelines Committee
posed a series of questions about the initial management
of closed head injuries. The nature of the questions
reflected the feeling among the clinicians that the
greatest needs for guidelines related to the management
of mild head injuries and the timing of transfer of
patients with closed head injuries from centres with
limited resources. The initial management of patients
with moderate to severe head injuries was felt to
be less controversial.
The questions are listed below.
1 What is the definition of a Mild Head Injury?
2 What are the clinically important complications
of Mild Head Injury?
3 How should patients with Mild Head Injury
be assessed?
4 Which patients with Mild Head Injury require
a CT scan?
5 What to do with high risk Mild Head Injury
patients when CT scan is unavailable?
6 When can patients with mild head injury be
safely discharged?
7 What are the proven treatments for patients
with moderate to severe head injuries?
8 When should patients with closed head injuries be
transferred to hospitals with neurosurgical facilities?
2.3 Literature ReviewThe Guideline Review Group used the methods outlined
in A Guide to the Development, Implementation and
Evaluation of Clinical Practice Guidelines (NHMRC 1999)
and the companying handbooks (NHMRC 2000abc,
2001, 2002) to review the literature and assess the
evidence used to support these guidelines.
Evidence from published studies was obtained from
a thorough search of MEDLINE, EMBASE and the
Cochrane Database of Systematic Reviews and the
Cochrane Library. Relevant articles were hand-searched.
A list of search terms used in these databases are listed
in Appendix E, p.66.
These databases were searched from 1980 to 30th
October 2004 for English language articles that met
the following general inclusion and exclusion criteria.
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 17
2.5 Assessment of the evidence for strength, size and relevance
2.5.1 Level of evidenceThe articles were classified according to their general purpose and study type. From this each article was
allocated a level of evidence.
PAGE 18 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
2.4 Inclusion and exclusion criteria
Table 2.1 Levels of evidence
2.5.2 QualityThe included and excluded articles were appraised according to the NHMRC. The MERGE assessment tool
(Liddle, Williamson, and Irwig, 1996) was used. Full text of this document and the checklists (and their criteria)
used are available from http://www.health.nsw.gov.au/pubs/a-z/m.html.
The articles were rated for quality on a 4-point scale as follows:
Table 2.2 Codes for the overall assessment quality of study checklists
Inclusion criteria
Meta-analysis, controlled clinical trials
Closed head injury
Aged >16 years
� ✕
Low risk of bias A All or most evaluation criteria from the checklist are fulfilled,
the conclusions of the study or review are unlikely to alter.
Low-moderate risk of bias B1 Some evaluation criteria from the checklist are fulfilled. Where evaluation
criteria are not fulfilled or are not adequately described, the conclusions
of the study or review are thought unlikely to alter.
Moderate - High risk of bias B2 Some evaluation criteria from the checklist are fulfilled. Where evaluation criteria
are not fulfilled or are not adequately described, the conclusions of the study or
review are thought to alter.
High risk of bias C Few or no evaluation criteria fulfilled. Where evaluation criteria are not fulfilled
or adequately described, the conclusion of the study or review are thought
very likely to alter.
Level I Evidence obtained from a systematic review of all relevant randomised control trials.
Level II Evidence obtained from at least one properly-designed randomised control trial.
Level III-1 Evidence obtained from well-designed pseudo-randomised controlled trials (alternate allocation
or some other method).
Level III-2 Evidence obtained from comparative studies (including systematic reviews of such studies) with concurrent controls
and allocation not randomised, cohort studies, case-control studies, or interrupted time series with a control group.
Level III-3 Evidence obtained from comparative studies with historical control, two or more single arm studies or interrupted
time series without a parallel control group.
Level IV Evidence obtained from a case-series, either post-test or pre-test/post-test.
Exclusion criteria
Case, Series, Case reports
Penetrating head injury
Aged <16 years
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 18
A patient with an initial GCS score of 14-15 on arrival at hospital III-2following acute blunt head trauma (with or without a definite history of loss of consciousness or amnesia).
Typical characteristics
:: Direct blow to the head or acceleration / deceleration injury.
:: Transient loss of consciousness or amnesia.
:: Transient abnormal alertness, behaviour or cognition.
:: Neurosurgical intervention rare.
:: Post concussion symptoms common.
:: Long term functional outcome good.
Risk stratification III-2
Mild Head Injury may be further sub-classified into 'High' and 'Low' risk groups, based on the risk of having an intracranialinjury requiring neurosurgical intervention. Stratification of
'high' and 'low' risk of intracranial injury is based on:1,6-9,14,16-19,23-46,62
:: initial GCS on admission and at two hours post injury
:: the duration of loss of consciousness or amnesia
:: the presence or absence of other specified risk factors.
DEFINITION LEVEL OF EVIDENCE
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 19
HE
AD
IN
JU
RY
GU
IDE
LIN
E
3 What is the definition of Mild Head Injury?
3.1 Discussion
Recently published studies and guidelines use a variety ofcriteria to define mild head injury, which itself, is variablyreferred to as either mild head injury or mild traumaticbrain injury.6-9,14,16-19,62 The most common variationsconcern the initial classification according to GCS anddifferent requirements for loss of consciousness andamnesia (as summarised in Evidence Table 1. Definitions of Mild Head Injury, p.44). This variation in the literaturemakes comparison between studies difficult.
The main reason for this variability is a uniform desire toidentify those patients at higher risk of intracranial injuryin what is a heterogenous but essentially low risk group.There is ample evidence to suggest that patients with aninitial GCS of 13 should be considered as part of themoderate head injury group due to the frequency ofintracranial lesions (25-38%) and cognitive functionalsequelae (see Evidence Table 2. Initial GCS versus
abnormal CT/neurosurgery, p.45).24-26,29,31,35
Further sub-classification of mild head injury is thendependent on the presence of associated risk factors and different authors have different approaches. The approaches of some of the best quality studies and guidelines are summarised in Evidence Table 1.Definitions of Mild Head Inury, p.44). It is interesting to note that when all the initial GCS criteria, inclusion/exclusion criteria and sub-classification systems are all taken into account, that the findings are very similar. These findings are that mild head injury is a heterogenous group with patients at higher risk ofincreased intracranial injury identified by abnormal initialGCS, persistently abnormal GCS, presence of prolongedLOC or amnesia and certain other risk factors.1, 6-9, 14,
16-19, 23-46, 62 It is important to recognise that these risk factors for intracranial injury do not necessarilypredict the risk of post concussive symptoms which are the more common complication of mild head injury.
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 19
Clinicians and patients should be aware of both the risk of neurosurgical III-2intervention and the risk of cognitive-behavioural-social sequelae following Mild Head Injury.
Clinicians and patients should also be aware that the absence of a structurallesion on CT scan following Mild Head Injury does not exclude the possibility of significant cognitive-behavioural-social sequelae.
Acute life-threatening complications requiring neurosurgical IVintervention are rare in Mild Head Injury patients:16,17,19,24-32,47,62
:: 'Low risk' Mild Head Injury range 0-3%.
:: 'High risk' Mild Head Injury range 0.5-6.5%.
Post concussion symptoms are common in Mild Head Injury patients and may IVhave significant cognitive-behavioural-social impacts on both patients and their families:3,8,9,36,48-52
:: Typical post concussion symptoms include:– headaches– dizziness – fatigue– memory impairment– poor concentration– mood swings– behavioural changes – social dysfunction.
:: Up to 50% of patients with Mild Head Injury may have significant post concussion symptoms which can last several weeks.
:: About 10% of patients with Mild Head Injury will have persistent disabling post concussion symptoms.
GUIDELINE LEVEL OF EVIDENCE
PAGE 20 Initial Management of Closed Head Injury in Adults :: NSW ITIM
4 What are the clinically importantcomplications of Mild Head Injury?
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 20
I The identification of patients requiring early acute neurosurgical intervention.
II The identification of patients requiring admission to hospital due to the increased risk of deterioration from complications.
III The identification of patients who can be safely discharged for home observation.
IV The provision of discharge advice to allow the identification and early return of patients with unexpected deterioration.
V The provision of discharge advice to allow the identification, treatment and follow-up of patients who develop disabling post
concussion symptoms.
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 21
HE
AD
IN
JU
RY
GU
IDE
LIN
E
4.1 DiscussionClinically important complications of mild head injury
include both structural lesions and functional deficits.
The most important structural lesions to identify are
those requiring acute neurosurgical intervention.
However, functional deficits resulting in cognitive-
behavioural-social sequelae are far more common
and may have significant impact on patients and their
families. It is important that doctors, patients and their
families understand that the absence of a structural
lesion on CT scan following a mild head injury does
not exclude the possibility of significant cognitive-
behavioural-social sequelae.
Acute intracranial haematomas requiring acute
neurosurgical intervention are the most dramatic
and life threatening complications of mild head injury.
The identification of structural lesions requiring acute
neurosurgical intervention is the most important function
of CT scanning because the presence or absence of
other structural lesions does not usually significantly
alter outcome. However, multiple studies have shown
that these neurosurgically significant lesions are relatively
uncommon with incidences of 0.1-3.2% for GCS 15
and 0.5-6.5% for GCS 14 (see Evidence Table 2.
Initial GCS versus abnormal CT/neurosurgery, p.45).
Other intracranial injuries and skull fractures are more
frequently noted on CT scans but are usually only
clinically important as indicators of the potential for
clinical complications such as delayed intracranial
haematomas, post traumatic seizures and post
concussion symptoms.26,36 Delayed acute intracranial
haematomas requiring neurosurgical intervention are
uncommon following normal CT scans (range <1%).
Post concussion symptoms are common following
mild head injury and may have significant cognitive-
behavioural-social impact on patients and their
families.3,8,9,36,48-52 Post concussion symptoms have
been shown to occur in up to 50% of patients with
mild head injury.3,8,9,36,48-50 Post concussion symptoms
include headaches, dizziness, fatigue and cognitive,
behavioural and social dysfunction. These symptoms
usually resolve within three months, but in about 10%
of cases they may persist with significant psychological
overlay as post concussion syndrome.3,8,9,36,48-50
The best clinical predictors of post concussion symptoms
and functional outcome are initial GCS on admission,
duration of loss of consciousness and duration of post
traumatic amnesia (PTA).3 If post traumatic amnesia
does not persist beyond 24 hours then significant post
concussion symptoms are less common.3 The most
widely used screening tool for PTA in New South Wales
is the Westmead PTA Scale (see Appendix A, p.61).
Evidence Table 5. Post concussive symptoms and Mild
Head Injury, p.54 details the most relevant studies
relating to recovery from mild head injury and post
concussion symptoms. The cognitive-behavioural-social
dysfunction caused by mild head injury can be quite
disabling, and some researchers have suggested that the
severity of impact on lifestyle makes the term ‘mild’
inappropriate for some patients.3,48,49 Patients with
significant post concussive symptoms should be referred
to a brain injury rehabilitation service or neurologist (see
Appendix D, p.65).
:: 4 WHAT ARE THE COMPLICATIONS OF MILD HEAD INJURY?
The clinically important complications of Mild Head Injury suggest that the management priorities should be:
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 21
Mild Head Injury patients should have a minimum of hourly III-2observations for four hours post injury.
:: These observations include:– GCS– alertness / behaviour / cognition– pupillary reactions– vital signs.
Serial neurological observations should be continued on any Mild Head Injury III-2patients who fail to clinically improve at four hours post injury or who are found to have structural lesions requiring hospital admission.7,53
Assessment for post traumatic amnesia (PTA) should be performed III-2on any Mild Head Injury patients who fail to clinically improve at four hours post injury or who are found to have structural lesions requiring hospital admission.3,55
Structured clinical assessment using clinical decision rules can identify those III-2patients at increased risk of intracranial injury requiring further investigation.16,17,19,62
Skull x-rays are not sufficiently sensitive to be used as a routine Iscreening investigation to identify significant intracranial lesions.37
CT scanning is the most appropriate investigation for the exclusion III-2of neurosurgically significant lesions in mild head injured patients.
CT scanning is indicated for those Mild Head Injury patients III-2identified by structured clinical assessment as being at increasedrisk of intracranial injury.16,17,19,25,38,39,62
If structured clinical assessment indicates the risk of intracranial injury III-2is low, the routine use of CT scanning is neither clinically beneficial nor cost effective.16,17,19,25,38,39
GUIDELINE LEVEL OF EVIDENCE
PAGE 22 Initial Management of Closed Head Injury in Adults :: NSW ITIM
5 How should patients with Mild Head Injury be assessed?
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 22
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 23
HE
AD
IN
JU
RY
GU
IDE
LIN
E
5.1 DiscussionClinical history and examination remain the basis for
the initial assessment of patients with closed head injury.
However, additional clinical tools are available to assist in
assessment and management. These clinical tools include
GCS, serial neurological observation, skull x-rays, CT
scanning and clinical decision rules.
5.1.1 Glasgow Coma ScaleThe Glasgow Coma Scale (see Appendix C, p.64) was
originally developed by Teasdale and Jennett53 for the
neurological observation of patients with prolonged
coma. It was intended to ensure inter-observer reliability
and to identify deterioration of patients over time. Since
its original introduction its use has been extended such
that it is now the standard tool for assessment of level
of consciousness in many clinical settings.
GCS is used both for the initial assessment and
classification of closed head injuries and for serial
assessment of closed head injuries. Initial GCS on
admission to hospital is used to classify head injuries
into the broad prognostic groups of mild (GCS 14-15),
moderate (GCS 9-13) and severe (GCS 3-8). The Brain
Trauma Foundation concluded that there is good quality
evidence to relate initial GCS score to outcome.15
However, it must be noted that these are broad outcome
measures and initial GCS is only about 75% accurate.15
GCS is unreliable if measured before initial resuscitation
and stabilisation of the ABCDE’s has been completed.15
Fearnside et al54 identified that both intubation and
sedation interfered with accurate assessment of initial
GCS in more severely injured patients and there
is lack of uniformity of approach to classifying
GCS in these situations.
The other significant area of controversy relates
to timing of initial GCS. In unstable patients requiring
resuscitation, the optimal time to record initial GCS
remains controversial.15,54 Similarly, in mild head injury
patients the time of presentation related to time
of injury will influence the initial GCS and therefore
potentially influence clinical decision making in relation
to CT scanning.6,9,16,19 Perhaps the most crucial point to
note about initial GCS is that it cannot predict outcome
for patients with similar GCS scores.
This is particularly important for mild head injury
patients because GCS primarily assesses the risk of
structural lesions. Subtle cognitive changes are not
well discriminated within the mild head injury group.
This led to the development of the extended Glasgow
Coma Scale that assesses the duration of post traumatic
amnesia as a means of identifying patients at increased
risk of cognitive problems.55 Thus, a patient with
an initial GCS of 14-15 may have no significant injury,
long-term cognitive-behavioural dysfunction or a life
threatening extradural haematoma. Despite these
limitations initial GCS on admission remains the
standard method for initial classification of
head injuries.
When assessing initial GCS in patients with head injury
it is worthwhile considering time of injury. Clearly initial
GCS for a given patient may vary depending on time of
presentation to hospital. Few studies have related GCS
to the time of injury with the exception of Stiell et al19
who found GCS <15 at two hour post injury was a
significant predictor of intracranial injury for mild head
injury patients.19 Fabbri et al16 showed that most mild
head injuries present around one hour post injury. Initial
GCS is therefore most likely to overestimate the risk of
intracranial injury in mild head injuries who present early.
Prehospital GCS, motor score and return of orientation
are other factors to consider when assessing initial GCS.
Prehospital GCS was felt to be unreliable56 but with
more organised prehospital systems is gaining further
attention.15,57 Motor score is the best predictor
of outcome of the GCS components.15 Orientation
returns most commonly in the sequence of person,
place then time.58
The Glasgow Coma Scale is also used as one of the
parameters in serial observation of head injury patients.
Both the original studies and subsequent studies have
validated its use in this fashion and prior to the advent
of CT scanning alteration in GCS was the most useful
tool in predicting intracranial injury.7,53 Currently, serial
GCS remains a standard tool in the monitoring of head
injuries when CT scanning is unavailable or when clinical
symptoms persist despite normal CT scanning.
:: 5 HOW SHOULD PATIENTS WITH MILD HEAD INJURY BE ASSESSED?
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 23
Summary
:: Initial GCS score is useful to classify closed head
injuries into broad prognostic groups requiring
further assessment.
:: The extended GCS which assesses duration of
PTA may help identify mild head injury patients
at increased risk of post concussion symptoms.
:: Serial GCS scores are useful for identifying patients
with significant risk of intracranial injury in:
– patients with an initial abnormal GCS score
who fail to improve
– patients whose initial GCS score deteriorates.
:: Initial and serial GCS scores are not as valuable
in excluding significant injury in:
– patients with an initial normal GCS score who
remain normal
– patients with an initial abnormal GCS score
who improve.
:: Initial GCS should only be used for prognostic
purposes after initial resuscitation and stabilisation
of ABCDE’s.
5.1.2 Serial clinical neurologicalobservations
In mild head injury patients the primary aims of serial
neurological observation are the early identification of
acute neurological deterioration and the identification
and monitoring of persistent mild neurological deficits.
Serial neurological observations remain the basic
standard of care for the initial management of mild head
injury patients and should be used in conjunction with
clinical decision rules to determine the need for CT
scanning and / or prolonged observation.
Serial neurological observations remain a standard tool
for assessing mild head injury patients despite the advent
of CT scanning. CT scanning is primarily used to identify
structural abnormalities at a given point in time while
serial neurological observations are used to monitor
clinical conditions over a longer period of time. Serial
neurological observations typically consist of hourly
pupillary reactions and GCS assessment in conjunction
with vital signs. The symmetry of motor responses should
be routinely noted as part of the GCS assessment.
Neurological assessment should also include qualitative
assessment of alertness, behaviour and cognition as this
may identify more subtle neurological impairment.
Controversy exists over the appropriate duration of
serial neurological observation for both mild head
injury patients who are improving and those who have
persistent clinical symptoms or abnormalities on CT scan.
Although four hours of initial neurological observations
post injury are fairly standard following mild head injury,
there is little evidence to support this. There is also some
debate as to whether the initial period of observation
should be until four fours post injury or for four hours
following arrival at hospital. Stiell et al19 demonstrated
that GCS<15 at two hours post injury was one of the
most significant risk factors for intracranial injury. Since
this guideline has used this important criteria it was felt
that the initial clinical observation period should also
be based on time post injury. The initial period of
neurological observation should therefore be until
four hours post time of head injury.
If serial neurological observations are not improving at
four hours post injury then clinical decisions need to be
made about the need for CT scanning and/or prolonged
observation. The period of prolonged observation is also
controversial as there is little evidence to support the
general recommendation of twenty four hours. This
period of observation is derived from limited studies
and case reports that show that clinical deterioration
is unusual in mild head injury patients after twelve to
twenty four hours. The best location for prolonged
neurological observation for lower risk patients is also
debated because some studies have shown that
admission to hospital does not guarantee that
regular neurological observation will occur.29
Summary
:: Serial neurological observations are a useful tool
for the early identification of acute neurological
deterioration and the identification and monitoring
of persistent mild neurological deficits.
:: Serial neurological observations should include
a minimum of hourly GCS assessment, pupillary
reactions and vital signs. Neurological observations
should also include qualitative assessment of
alertness, behaviour and cognition to detect
subtle changes in mental status not assessed
by the Glasgow Coma Scale.
:: Mild head injury patients should have initial serial
neurological observations until four hours post injury
at which point decisions about further management
should be made.
PAGE 24 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 24
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 25
HE
AD
IN
JU
RY
GU
IDE
LIN
E
5.1.3 Post traumatic amnesia (PTA) testingThe assessment of the duration of PTA following
mild head injury is the best tool for predicting whether
a patient will have significant post concussion
symptoms.3,55 PTA is most simply defined as that period
of time following head injury during which a patient is
unable to lay down new memories. While it has been
identified that prolonged PTA is a risk factor for
significant post concussion symptoms, the standard tool
for identifying PTA in NSW, the Westmead PTA Scale
(see Appendix A, p.61), is designed to be used over
several days. There has been considerable interest in
developing bedside PTA screening tools to allow early
prediction of which patients are at greatest risk of
developing post concussion symptoms. The Extended
Glasgow Coma Score and the Modified Westmead PTA
Score are examples of tools developed to try to identify
persistent PTA in mild head injury patients as a marker
for increased risk of post concussion symptoms.55,90
Summary
:: The identification of persistent PTA in mild head
injury patients is a potentially useful marker for the
risk of developing post concussive symptoms.
:: PTA testing should be performed on any patient
admitted to hospital following mild head injury.
5.1.4 Clinical decision rulesClinical decision rules are increasingly being used to
assist clinicians in determining the need for particular
investigations or management. By identifying individual
risk factors and combining them to establish clinical
decision rules, which are then prospectively validated,
useful evidence based diagnostic tools can be
established. Well established clinical decision rules
also include the NEXUS criteria for cervical spine
assessment and the Ottawa Ankle Rules.59,60
Although clinical decision rules are potentially very
useful clinicians need to be aware of the specific
inclusion/exclusion criteria used to develop them and
the overall quality of the original studies before applying
them to their patients. Sound clinical decision rules have
been developed through review of large cohorts of
patients and prospectively applied to patients to
determine their sensitivity and specificity.16,17,19
In recent years a number of attempts have been made
to develop clinical decision rules for mild head injuries
for both adults and children. The studies aim to identify
those patients at increase risk of intracranial injury and
develop clinical decision rules about those who require
CT scanning or prolonged observation. The main design
features of these trials and those of NEXUS II, which is
currently in progress, are summarised in Evidence Table
4. Safe discharge of Mild Head Injury, p.51.14,16,17,19
The main findings of other studies examining
the accuracy of certain risk factors are summarised
in Evidence Table 3. Risk factors for intracranial injury
in Mild Head Injury, p.46.
The most consistent findings of these studies are that
abnormal initial GCS or mental status, clinical suspicion
of skull fracture and focal neurological deficit are the
best predictors of risk of intracranial injury. Loss of
consciousness, amnesia, vomiting, headache and
seizure are also relevant predictors of risk. Depending
on their inclusion / exclusion criteria the authors used
combinations of those risk factors to derive clinical
decision rules of varying sensitivity and specificity
(see Evidence Table 3. Risk factors for intracranial
injury in Mild Head Injury, p.46).
At present, the findings of Haydel et al17 have been
adopted as an ACEP policy whilst the findings of Stiell
et al19 have been adopted by the NICE guidelines (UK).
A recent Australian study looked at applying the clinical
decision rules developed by Haydel et al17 and Stiell
et al19 to Australian practice and concluded that both
had limitations.61 The NEXUS II study which intends
to enrol the largest patient groups yet, is likely to
provide the most definitive clinical decision rule
once completed.14.
In a recent study Ibanez et al62 attempted to
prospectively identify clinical risk factors predicting
intracranial injury and to assess the reliability of
previously published clinical guidelines. They found that
while clinical risk factors could not detect all intracranial
injuries they could be used to detect clinically relevant
lesions with a negative predictive power approaching
99%. They concluded clinicians should be aware
of the limitations of clinical decision rules when
using clinical guidelines.
Summary
:: Clinical decision rules provide useful adjuncts
to the assessment and management of mild
head injury patients.
:: Studies by Haydel et al17 and Stiell et al19
have provided evidence based clinical decision
rules for mild head injury in adults.
:: 5 HOW SHOULD PATIENTS WITH MILD HEAD INJURY BE ASSESSED?
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 25
5.1.5 Skull x-raysThe literature clearly identifies that both the clinical
suspicion of skull fractures and the radiological
evidence of skull fracture are significant risk factors
for the presence of an intracranial lesion requiring
neurosurgical intervention.6,16,17,19,27,37
If CT scanning is available, the current indications
for skull x-rays are few. However, if CT scanning is
unavailable, the role of skull x-rays as a screening test
is less clear. A detailed meta analysis by Hofman et at37
concluded that whilst the presence of skull fracture
greatly increased the risk of intracranial injury, the
absence of a skull fracture did not rule it out
(calculated sensitivity 38% calculated specificity 95%).
In subsequent clinical guidelines, authors have differed
as to whether skull x-rays should be used to detect
patients at higher risk of intracranial injury. Jagoda
et al7 argued that the sensitivity of skull x-ray is not
sufficient to be used as a screening test. Vos et al9 and
Servadei et al6 felt that in the absence of CT scanning,
a positive skull x-ray can be useful to help allocate
patients into higher risk groups for management
purposes. On existing evidence, both approaches seem
reasonable depending on local management guidelines.
Summary
:: Clinical evidence or suspicion of skull fracture is
associated with increased risk of intracranial injury.
:: Skull x-rays are not sufficiently sensitive to be used
as a routine screening test to identify patients at
increased risk of intracranial injury.
:: Where CT scanning is unavailable, skull x-ray may
be used as an adjunct to identify patients at greater
risk of intracranial injury (but not to exclude
intracranial injury).
5.1.6 CT scanningThe widespread availability of CT scanning has greatly
assisted the management of patients with head injuries.
CT scanning has been particularly useful in identifying
focal injuries in patients with altered level of
consciousness. CT scanning is regarded as mandatory
for all head injury patients with an initial or persistent
altered level of consciousness. However, the role of
the CT scanning in a patient with mild head injury with
a normal level of consciousness remains controversial.
Particular concerns about the routine use of CT scanning
for mild head injury include the financial-resource
burden, the potential hazards of radiation and the
potential pitfalls of reliance on technology at the
expense of clinical assessment.20,38,47,63-65
Furthermore CT scans do not identify patients who have cognitive dysfunction which is the most significantcomplication for most patients. The various pros andcons of CT scanning are summarised below:
Pros
:: Early identification of patients with intracranialinjuries requiring acute neurosurgical intervention(see Evidence Table 3. Risk factors for intracranialinjury for Mild Head Injury, p.46 and Evidence Table 4. Safe discharge of Mild Head Injury, p.51).
:: Early identification of patients with other intracranialinjuries requiring admission to hospital due to risk ofdeterioration (see Evidence Table 1. Risk factors forintracranial injury for Mild Head Injury, p.46 andEvidence Table 4. Safe discharge of Mild Head Injury, p.51).
:: Identification of patients at low risk of deteriorationand suitable for discharge (see Evidence Table 4. Safe discharge of Mild Head Injury, p.51).
:: Identification of patients with structural lesionsindicating increased risk of post concussive symptoms.36
:: Potential cost benefit due to early CT scanningallowing discharge home rather than hospitaladmission in some patients.63
Cons
:: Routine use of CT scanning for mild head injury has a huge financial and resource impact given thatmore than 90% of scans are negative and less than1% of scans indicate the need for neurosurgicalintervention.14,19
:: Patients may develop focal neurosurgical lesions(especially subdurals) despite initial normal CTscanning.32,66,67
:: Early CT scans may not demonstrate intra-cerebralcontusions which take time to become apparent onCT scanning.
:: CT scanning will not demonstrate diffuse axonalinjury in most patients.3
:: Patients may suffer significant post concussivesymptoms despite an initial normal CT scan.3,36,48-50
:: Routine use of CT scanning does not guarantee better identification of significant intracranial injuriesif different institutions are compared.12,20
:: The risk of cumulative radiation exposure especiallyamong children is of concern.64,65
:: May delay definitive management of more significantinjuries in multi system trauma patients.
PAGE 26 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 26
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 27
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Timing of CT scanning
There is no direct evidence to confirm what the best
time to perform CT scanning in relation to time of injury
is. The primary role of early CT scanning in mild head
injury is early recognition of extradural haematomas
prior to clinical deterioration.32 Early neurosurgical
intervention prior to clinical deterioration is associated
with improved outcome. However, early CT scan may
potentially miss other intracranial injuries such as
subdurals or contusions which are slower to become
evident.32 Fortunately, most studies have shown that an
initial normal CT scan allows safe discharge and that the
few patients who deteriorate usually have good
outcome.7,29,68,69 Therefore, it is reasonable
to suggest that CT scans should be performed shortly
after a decision is made that one is necessary.
Summary
:: CT scanning is the best investigation for the early
identification of neurosurgically significant focal
intracranial lesions following mild head injury.
:: Early CT scanning may not demonstrate some
subdural haematomas and cerebral contusions.
:: CT scanning should be used as an adjunct to clinical
assessment.
:: CT scanning does not accurately predict the risk
of post concussion symptoms in mild head injury
patients.
:: Where clinical assessment indicates the risk of
significant intracranial lesion is low, the routine
use of CT scanning is unlikely to be of benefit.
:: 5 HOW SHOULD PATIENTS WITH MILD HEAD INJURY BE ASSESSED?
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 27
Patients with ‘High Risk’ Mild Head Injury requiring CT scan
The following risk factors identify patients with Mild Head Injury III-2(initial GCS 14-15) at increased risk of clinically significant lesions requiring acute neurosurgical intervention or prolonged observationin hospital. These patients should have early CT scanningif available, if they have any of the following features:
Any of...
Initial assessment
:: Persistent GCS <15 at two hours post injury.**
:: Focal neurological deficit.
:: Clinical suspicion of skull fracture.
:: Prolonged loss of consciousness (>5min).
:: Prolonged anterograde or retrograde amnesia (>30min).
:: Post traumatic seizure.
:: Repeated vomiting (>2 occasions).
:: Persistent severe headache.
:: Known coagulopathy.
:: Age >65 years (clinical judgment appropriate if no other risk factors present).
After a period of observation (four hours post injury)
:: Any deterioration in GCS.
:: Persistent abnormal mental status (abnormal alertness, behaviour or cognition).**
:: Any patient who fails to clinically improve.
Clinical judgment required if:
:: age >65 years
:: drug or alcohol ingestion
:: dangerous mechanism
:: multi-system trauma
:: known neurosurgery / neurological impairment
:: delayed presentation or representation.
** NOTE: Includes patients with abnormal GCS due to drug or alcohol ingestion.
GUIDELINE LEVEL OF EVIDENCE
PAGE 28 Initial Management of Closed Head Injury in Adults :: NSW ITIM
6 Which patients with Mild Head Injury require a CT scan?
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 28
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 29
HE
AD
IN
JU
RY
GU
IDE
LIN
E
:: 6 WHICH PATIENTS WITH MILD HEAD INJURY REQUIRE A CT SCAN?
Patients with ‘Low Risk’ Mild Head Injury not requiring CT scan
The following features indicate patients with Mild Head Injury III-2(initial GCS 14-15) at low risk of clinically significant lesions requiring acute neurosurgical intervention or prolonged observation in hospital.These patients should not routinely have CT scanning if they have all of the following features:
All of...
Initial assessment
:: GCS 15 at two hours post injury.
:: No focal neurological deficit.
:: No clinical suspicion of skull fracture.
:: Brief loss of consciousness (<5min).
:: Brief anterograde or retrograde amnesia (<30min).
:: No post traumatic seizure.
:: Mild nausea or single episode of vomiting.
:: No persistent severe headache.
:: No known coagulopathy.
:: Age <65 years.
After a period of observation (four hours post injury)
:: Normal mental status including alertness and behaviour and cognition.
:: No deterioration during observation.
:: Clinically improving.
Clinical judgment required if:
:: age >65 years
:: drug or alcohol ingestion
:: dangerous mechanism
:: multi-system trauma
:: known neurosurgery / neurological impairment
:: delayed presentation or representation.
GUIDELINE LEVEL OF EVIDENCE
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 29
6.1 DiscussionMild head injury patients (initial GCS 14-15) require a CT
scan if they are clinically assessed as being at significant
risk of acute deterioration from an underlying intracranial
injury. The clinical risk factors that indicate patients are
at increased risk of intracranial injury are discussed in
the following text, and most of the relevant studies
summarised in the evidentiary tables. The absence
of clinical risk factors on initial assessment combined
with a period of observation during which the patients
clinically improves makes the probability of a significant
intracranial injury extremely unlikely. These low risk
mild head injury patients can be discharged for home
observation without CT scanning. In the high risk group,
which includes initially low risk patients who fail to
clinically improve, both CT scanning and prolonged
clinical observation are indicated.
Early CT scanning allows identification of acuteintracranial haematomas requiring urgent neurosurgicalintervention and other structural lesions that put thepatient at increased risk of deterioration fromcomplications such as delayed intracranial haematomas,raised intracranial pressure, post traumatic seizures,Syndrome of Inappropriate Antidiuretic Hormone (SIADH)and disabling post concussive symptoms. However it isimportant to recognise that the absence of a structurallesion on CT scan does not exclude the possibility ofsignificant post concussion symptoms.
Patients with an abnormal CT scan should be admittedfor prolonged observation in hospital. Conversely, anormal CT scan makes acute clinical deterioration highlyunlikely and allows safe discharge for home observationas long as the patient is clinically improving. It isimportant to stress that CT scanning should be used as aclinical tool in conjunction with clinical assessment andobservation as part of an overall management strategyfor mild head injury patterns.
PAGE 30 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
GUIDELINE
A persistent GCS <15 at two hours post injury is an absolute indication for CT scanning. An initial GCS <15 is a relative indication for CT scanning.16,19,24-29,34,35,62,70
Several studies have noted the heterogeneity of the
original GCS 13-15 mild head injury classification and
these findings are summarised in Evidence Table 2. Initial
GCS versus abnormal CT/neurosurgery, p.45. Patients
with an initial GCS 13 have been shown to have similar
rates of intracranial injury to those with initial GCS 9-
12.29,54 Although patients with initial GCS 14-15 have
lower rates of intracranial injury as a group, the more
recent studies16,17,19 have confirmed a higher risk of
intracranial injury for initial GCS 14. This higher risk
of intracranial injury for initial GCS 14 does not take
into account either associated risk factors nor the
time of injury.
A number of recent, well designed studies have
attempted to identify risk factors which can predict
which patients with initial GCS 14-15 are at highest risk.16,17,19 The evidence would suggest that other risk
factors can be used to stratify risk within the initial GCS
14-15 group.
Perhaps the most relevant finding to clinical practice is
that of Stiell et al [19] who showed that for patients
with an initial GCS 13-15 that GCS <15 at two hours
post injury was a useful predictor of risk of intracranial
injury. By applying this criteria, both time of injury and
GCS are usefully combined in a clinically practical
assessment tool.
Individual factors predicting risk of intracranial injury and therefore the need for CT scanning in patients with mild head injury:
6.1.1 Initial GCS
GUIDELINE
Focal neurological deficit mandates CT scanning [16-19, 28-30, 32-34, 62, 70, 71].
6.1.2 Focal neurological deficits
Focal neurological deficits have been shown to
significantly increase the risk of intracranial
injury.16,18,29,31 Both Haydel et al17 and Stiell et al19
excluded focal neurological deficits in their studies due
to the previously proven nature of risk. Conversely
Vilke et al45 showed that a normal neurological
examination does not rule out underlying brain injury
in mild head injuries.
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 30
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 31
HE
AD
IN
JU
RY
GU
IDE
LIN
EClinical suspicion or evidence of skull fracture has been
shown to be a significant risk factor for intracranial
injury. The meta analyses by Hofman et al [37] showed
that the x-ray presence of skull fracture had a specificity
of 95% for intracranial injury. Clinical suspicion or
evidence of skull fracture has been shown by several
authors including Stiell et al [19], Haydel et al [17] and
Palchek et al [18] to be a major risk factor for the
presence of intracranial injury. Clinical suspicion of open,
depressed or base of skull fractures is based on the
presence of large scalp lacerations or haematomas
(especially in children <2 years), obvious skull
depression, and base of skull signs such as raccoon
eyes, haemotympanum, Battles sign, or CSF leak.
The presence of significant facial fracture may also
indicate the possibility of skull fracture.
:: 6 WHICH PATIENTS WITH MILD HEAD INJURY REQUIRE A CT SCAN?
GUIDELINE
Clinical suspicion or evidence of skull fracture is an absolute indication for CT scanning. 17-19,27,28,30,31,34,37,38,47,62,71
6.1.3 Skull fractures
GUIDELINE
Brief loss of consciousness (<5 minutes) should not be considered an independent indication for CT scan. Prolonged loss of consciousness (>5 minutes) should be considered a strongindication for CT scanning.1,6,16-18,27,33,39-44
6.1.4 Loss of consciousness
Loss of consciousness increases the risk of intracranial
injury and is used by many authorities to define mild
head injury (See Evidence Table 1. Definitions of Mild
Head Injury, p.44).
However, the absolute risk associated with loss
of consciousness is small.6,27 The absence of loss
of consciousness or amnesia has been used to classify
patients as ‘minimal risk’.6-9 The absence of loss
of consciousness has been shown not to rule out
intracranial lesions in children.18,41 Similarly,
adult studies have shown that transient LOC does
not accurately predict the risk of intracranial
injury.33,39,40,42,62
Duration of loss of consciousness is also controversial.
‘Brief’ loss of consciousness in mild head injury patients
is usually associated with good functional outcome while
‘prolonged’ loss of consciousness is not clearly
associated with poorer outcome.9,27,43,44 However, the
exact definitions of what should be considered a low-risk
duration vary greatly from momentary to five minutes, to
20 minutes to 30 minutes.6-9
From a practical viewpoint obtaining a definite history
and duration of loss of consciousness is often difficult.
Head injuries are frequently unwitnessed, observers
unreliable and patients often affected by alcohol.
It is perhaps simplest to consider loss of consciousness
in terms of no loss of consciousness, brief loss of
consciousness less than five minutes, prolonged loss
of consciousness greater than five minutes or unreliable
history. The clinical recommendation of the neurosurgical
committee of the Royal Australasian College of Surgeons
(RACS) is that loss of consciousness should be
considered brief if less than five minutes.1
GUIDELINE
Persistent anterograde or retrograde amnesia of greater than 30 minutes duration mandates CTscanning. Amnesia for the event does not warrant CT scanning.16,17,19,27
6.1.5 Amnesia
The presence of amnesia is associated with higher risk
of intracranial injury and recent studies have established
certain durations as significant. Amnesia for the event
implies transient neurological dysfunction and indicates
mildly increased risk of intracranial injury although the
absolute risk remains small.27 Retrograde amnesia
(defined as the period of loss of memory prior to the
event) has been shown by Stiell et al19 to be
of significance if greater than 30 minutes duration.
Duration of anterograde amnesia or post traumatic
amnesia (the period of inability to lay down new
memories post event) has been shown to be associated
with both risk of intracranial injury and cognitive –
behavioural sequelae.3,17,19
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 31
Haydel et al identified short term memory deficit
(a surrogate for PTA) as a significant risk factor for
intracranial injury.17 Post traumatic amnesia that persists
for more than 24 hours has been shown to
be a significant risk factor for persistent cognitive-
behavioural-social dysfunction.3 Stiell et al19
also identified anterograde amnesia of more than
30 minutes as a risk factor but did not include
it in their clinical decision rule.
From a practical point of view, any patient with
persistent anterograde or retrograde amnesia for more
than 30 minutes duration should be considered at higher
risk. It is useful to assess the patients recall of events
following their injury by asking specific questions such as
what is their first clear memory, who helped them at the
scene and how they got to hospital. This can be used to
estimate the period of anterograde amnesia. A simple
memory assessment technique such as three object
recall can be used as a bedside screening test for
anterograde amnesia, to supplement the ‘history’
of amnesia for events.
PAGE 32 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
GUIDELINE
Post traumatic seizures are an indication for CT scanning.17,18
6.1.4 Post traumatic seizure
Brief generalised post traumatic seizures are common
immediately following mild head injury and are not
usually associated with poor outcome. They are
frequently seen on sporting fields and in young children.
Prolonged, delayed or focal post traumatic seizures are
more likely to be associated with significant intracranial
injury. However, most mild head injury studies do not
differentiate between types of seizures when assessing
risk factors for intracranial injury.17,18 Most published
guidelines have identified any post traumatic seizures as
risk factors warranting routine CT scanning.
GUIDELINE
Persistent or recurrent vomiting is an indication for CT scanning.16-19,32,38,47
6.1.5 Vomiting
Vomiting has been identified as a significant risk factor
for intracranial injury in many studies.17-19,32,38 There has
been some debate whether persistent vomiting is more
relevant than isolated vomiting. Stiell et al19 identified
repeated vomiting (more than one occasion) as being a
significant risk factor.
GUIDELINE
Persistent severe headache is an indication for CT scanning.16-18,32,38,47
6.1.6 Headache
Many studies have identified headache as a significant
risk factor for intracranial injury.17,18,32,38
The general trend of the literature would suggest
that persistent moderate to severe headache
should be considered a significant risk factor.
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 32
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 33
HE
AD
IN
JU
RY
GU
IDE
LIN
E
:: 6 WHICH PATIENTS WITH MILD HEAD INJURY REQUIRE A CT SCAN?
GUIDELINE
Known coagulopathy is both an indication for CT scan and also an indication to consider prolongedhospital observation.6-9,16,17,19,72
6.1.7 Coagulopathy
Most guidelines and studies on mild head injury mention
coagulopathy either as a significant risk factor or an
exclusion criteria in the assessment of risk of intracranial
injury.6-9,16,17,19 However, there is no compelling
evidence to either support or refute this reasonable
assertion in the mild head injury patient group.
Mina et al72 demonstrated that pre-existing
anticoagulation significantly increased the risk of death
from intracranial injury in trauma patients with head
injury. However, this was a heterogenous patient group
with significantly abnormal ISS (mean 17.0 +/- 7.8) and
GCS (mean 11.8 +/- 4.0).
GUIDELINE
Patient age >65 years is a relative indication for CT scanning. Clinical judgment may be used in anasymptomatic patient age >65 years if no other risk factors are present.6,16,17,19,24,30,31
6.1.8 Age
Extremes of age have been shown to be associated
with increased risk of intracranial injury. The paediatric
literature identifies young children (age <5 years
and particularly <2 years) as being at increased
risk of intracranial injury due to a combination of
communication difficulties, non-accidental injury,
anatomical considerations and the lack of localising
clinical signs and symptoms.18,73
Several published studies have shown that there is an
increased risk of intracranial injury for patients aged
over 60-65 years with mild head injury.17,19,24,30,31
In particular, both Haydel et al17 and Stiell et al19
included age >60-65 years as indicators of the need
for CT scan. Mack et al74 recommended routine head
CT for elderly patients suffering mild head injury as they
could not identify any useful predictors of intracranial
injury in the elderly. The Brain Trauma Foundation
concluded that ‘increasing age is a strong independent
factor in prognosis, with significant increase in poor
outcome above 60 years of age’ for patients with severe
head injuries.15 Similarly Williams et al36 demonstrated
that elderly patients were more likely to sustain
complications of mild head injury. Although there
is no doubt advanced age is a risk factor for mild head
injury, there are practical difficulties in applying this
to the general population. As Servadei et al6 have
pointed out it is unlikely there is a specific age at which
risk of intracranial injury dramatically increases. Fabbri et
al16 found that using age >60 years alone to predict the
need for CT scanning in patients with mild head injury
was impractical from a cost-resource consideration
during a study to validate a set of guidelines.
Interestingly, of the 705 patients meeting guidelines
criteria for CT scanning based on age >60 years alone
(who did no have CT scans) Fabbri et al found that
only one patient deteriorated within 48 hours. From
a practical viewpoint extremes of age need to be
recognised as risk factors for intracranial injury. Whether
all patients age >60 years with mild head injury should
have CT scans based on age alone remains controversial.
It is worth noting that the NSW Institute of Trauma and
Injury Management Trauma Death Review Committee
has identified that in 2003 / 2004 elderly NSW patients
with head injuries represented a significant number
of potentially preventable deaths.75
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 33
PAGE 34 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
GUIDELINE
Persistent abnormal mental status manifested by drowsiness, abnormal behaviour or cognitiveimpairment is an indication for CT scanning.17-19,28,32,39,70
6.1.9 Abnormal mental status
Although initial GCS partially assesses mental status
it primarily addresses level of consciousness. Clinical
observation of alertness, behaviour and cognition will
detect more subtle changes in mental status than GCS
and should be part of routine neurological observation.
Abnormal alertness, abnormal behaviour and cognitive
impairment have also been shown to be associated with
intracranial injury [17-19]. Abnormal alertness or
behaviour and cognitive impairment requires careful
observation by staff and relatives and is particularly
important in children or adults with pre-existing
neurological impairment. In patients with pre-existing
neurological conditions a lower threshold for CT
scanning is appropriate [6, 62].
GUIDELINE
Drug or alcohol ingestion with a normal mental state is not an indication for CT scanning. Drug or alcohol intoxication resulting in an abnormal mental state is an indication for CT scanning.16,17,19,23
6.1.10 Drug or alcohol intoxication
Drug or alcohol intoxication is frequently present in
patients with head injuries and makes patients difficult
to assess and manage. Cook et al23 in a study
of alcohol intoxicated patients found that clinical
examination could not predict which alcohol intoxicated
patients had abnormal CT scans. However, they
observed that the rate of abnormal CT scan and
neurosurgical intervention was similar to that of the
non-intoxicated mild head injury population. Several
studies have shown that drug or alcohol intoxication
is a risk factor for intracranial injury but the exact
definition of intoxication remains vague.16,17
Stiell et al19 took a different approach and used GCS
<15 at two hours post injury as their variable to predict
abnormal mental status for whatever reason. They
argued that drug or alcohol intoxication was not an
independent predictor of intracranial injury if the patient
had a normal mental status at two hours post injury.
From a practical viewpoint, clinically obvious drug or
alcohol intoxication should be treated as a risk factor
for intracranial injury because it manifests as abnormal
mental status which impairs clinical assessment and
must be assumed to be due to intracranial injury.
GUIDELINE
In the absence of other risk factors, dangerous mechanism is only a relative indication for CT scan.Clinical judgment is required.19,30,31,46
6.1.11 Dangerous mechanisms of injury
Epidemiological studies have generally identified motor
vehicle accidents, falls and assaults as the commonest
causes of head injuries. In studies on patients with mild
head injuries, specific high risk factors for intracranial
injury that have been identified include assault,30,46
pedestrians or cyclists struck by motor vehicles19,30,31
and falls >1m.19
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 34
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 35
HE
AD
IN
JU
RY
GU
IDE
LIN
E
:: 6 WHICH PATIENTS WITH MILD HEAD INJURY REQUIRE A CT SCAN?
GUIDELINE
In mild injury patients with multi-system trauma clinical judgement is required regarding the needfor CT scanning.
6.1.11 Multi-system trauma
Most guidelines and studies on mild head injury have
specifically excluded patients with multi-system trauma
or unstable vital signs. It is therefore difficult to make
evidence based recommendations and clinical judgment
is required. A relatively low threshold for performing
head CT scan in multi-system trauma patients should
be used as subtle signs of neurological deficit are easily
missed in the presence of distracting injuries.
GUIDELINE
Pre-existing neurosurgery or neurological impairment is a relative indication for CT scanning [6, 62].
6.1.12 Pre-existing neurosurgery / neurological impairment
Pre-existing neurosurgery has been suggested as an
indication for CT scanning particularly in the presence
of hydrocephalus and shunt placement.6 The World
Federation of Neurosurgical Societies guideline6
recommended routine CT scanning for patients with
either previous neurosurgery or epilepsy. From a practical
viewpoint any pre-existing medical condition resulting
in neurological impairment (eg stroke, dementia,
and developmental delay) may make clinical
assessment difficult.
6.1.13 Presentation
Although mild head injuries are very common it is
thought that the majority do not present to hospital.13,14
Therefore, those that do present to hospital are already a
group at slightly increased risk. Of particular concern are
those who have a delayed presentation due to
persistence of symptoms or those who represent
because of ongoing or new symptoms.
However, as Fabbri et al16 showed the overall risk
of intracranial injury in patients who represent after
mild head injuries is low if their initial risk was low.
6.1.14 Unwitnessed event / unreliable history
A good history of injury may help predict risk of
intracranial injury by identifying dangerous mechanism
of injury or significant features such as prolong loss
of consciousness or seizures. This is of particular
importance in children. However, in the absence
of other significant risk factors there are few studies
identifying unwitnessed event or unreliable history
as a significant independent risk factor.71
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 35
PAGE 36 Initial Management of Closed Head Injury in Adults :: NSW ITIM
7 What to do with ‘High Risk’Mild Head Injury patients when CT scan is unavailable?
‘High Risk’ Mild Head Injury patients should be admitted for prolonged III-2hospital observation and considered for transfer for CT scanning. Patients at highest risk of intracranial injury who should be transferred for CT scanning include those with:16-19,27-34,37-39,47,70,71,76
:: persistent GCS <15 at two hours post injury.
:: deterioration in GCS.
:: focal neurological deficit.
:: clinical suspicion of skull fracture.
:: persistent abnormal mental status, vomiting or headache at four hours post injury.
Patients at ‘Highest Risk’ of intracranial injury should be discussed with Consensusthe regional neurosurgical service and a management plan established.
If patients are transferred for CT scanning they should ideally be transferred Consensusto a hospital with neurosurgical facilities to avoid secondary transfer.
A skull x-ray may be useful to confirm the presence of a skull fracture Ithat mandates an early CT scan due to the increased risk of deterioration.
All ‘High Risk’ patients who cannot have CT scanning should at a minimum Consensushave prolonged observation in hospital for at least 24 hours and until clinically improving.
GUIDELINE LEVEL OF EVIDENCE
7.1 DiscussionIn patients with high risk mild head injury, a normal CT
scan combined with clinical assessment will allow the
patient to be safely discharged for home observation.
If CT scan is unavailable then the patient will require
either admission for prolonged observation or early
transfer for CT scanning depending on clinical
assessment of risk. Prolonged clinical observation
for at least 24 hours, associated with clinical
improvement, has been shown to make a significant
injury unlikely in the majority of mild head injury
patients. However, those patients at highest risk for
an intracranial injury identified by persistently abnormal
GCS or mental status, deterioration in GCS, focal
neurological deficit, or clinical suspicion of skull fracture
should be transferred for CT scan to allow the early
identification of potentially neurosurgically significant
injury.
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 36
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 37
8 When can patients with Mild HeadInjury be safely discharged home?
Mild Head Injury patients can be discharged for home observation after III-2initial period of in-hospital observation if they meet the following clinical, social and discharge advice criteria.16,17,19,24,26,27,29,32,66,68,69,77-79
Clinical criteria III-2
:: Normal mental status and behaviour with clinically improving minor post concussion symptoms after observation until at least four hours post injury.
:: No clinical risk factors indicating the need for CT scanning or normal CT scan if performed due to risk factors being present.
:: No clinical indicators for prolonged hospital observation (irrespective of CT scan result) such as:– clinical deterioration– persistent abnormal GCS or focal neurological deficit– persistent abnormal mental status or behaviour– persistent severe post concussion symptoms– persistent drug or alcohol intoxication– presence of known coagulopathy (relative)– presence of multi-system injuries (relative)– presence of intercurrent medical problems (relative)– age >65 (relative).
Social criteria Consensus
:: Responsible person available to take patient home.
:: Responsible person available for home observation.
:: Patient able to return easily in case of deterioration.
:: Written and verbal discharge advice able to be understood.
Discharge advice criteria Consensus
:: Discharge summary for local doctor.
:: Written and verbal head injury advice given to patient and nominated responsible person covering:– symptoms and signs of acute deterioration– reasons for seeking urgent medical attention– typical post concussion symptoms– reasons for seeking routine follow up.
GUIDELINE LEVEL OF EVIDENCE
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 37
8.1 DiscussionMild head injury patients can be safely discharged from
hospital for home observation when the risk of acute
deterioration from an underlying intracranial injury is
assessed as being low. Safe discharge also requires that
the patient has adequate social supports and appropriate
advice on when to return to hospital.
The duration of in-hospital observation required will
be determined by clinical assessment combined with
selective use of imaging. Deterioration following
mild head injury may occur due to missed or delayed
intracranial haematomas or other complications such as
SIADH, post traumatic seizures or severe post concussive
symptoms. Although clinical assessment and observation
combined with appropriate imaging will identify most
at risk patients, the risk of deterioration is never zero.
Although uncommon, deterioration may occur even
after prolonged periods of observation and / or following
normal CT scanning. The challenge of managing mild
head injuries is to identify what is reasonable risk and
to ensure that the patient is aware of the potential
for delayed deterioration.
The mild head injury management algorithm summaries
the key points in management relating to safe discharge
and Evidence Table 4. Safe discharge of Mild Head Injury,
p.51 present some of the significant studies relating to
safe discharge. Mild head injury patients should
essentially be divided into low and high risk groups
based on clinical assessment. Low risk mild head injury
patients can be discharged for home observation after
a short period of observation in hospital if clinically
improving. High risk patients require CT scanning and/or
prolonged observation. High risk patients with clinically
important abnormalities on CT scan require admission
for prolonged observation. High risk patients with
normal CT scanning should also be admitted for
prolonged observation unless rapid clinical improvement
occurs. In both high and low risk mild head injury
patients’, potential clinical indications for admission such
as intercurrent medical problems and injuries need to be
considered. Whatever the period of observation selected,
the provision of safe discharge advice and assessment
of the patients social situation is mandatory because
occasional cases of deterioration following discharge
are unavoidable. An example of a suitable head injury
discharge advice sheet is attached in Appendix C,p.63.
PAGE 38 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 38
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 39
HE
AD
IN
JU
RY
GU
IDE
LIN
E
9 What are the proven treatmentsfor patients with ‘Moderate’
to ‘Severe’ Head Injuries?
Initial management of ‘Severe Head Injuries’ (GCS 3-8) III-2
Standard care15,22,80-87
:: Initial systematic resuscitation of ABCDE's.
:: Early CT scanning to identify neurosurgically correctable focal intracranial haematomas.
:: Prevention of secondary brain injury by avoiding hypoxaemia (oxygen saturation <90%) and hypotension (systolic BP <90).
:: Supportive care of ABCDE's with appropriate attention, posturing (30o head up), basic nursing care and avoidance of hyperventilation.
:: Early neurosurgical consult.
:: Use of ICP monitoring to guide management of cerebral perfusion pressure in patients with severe brain injury.
:: Optional use of anticonvulsants to prevent early post traumatic seizures.
:: Routine brain injury rehabilitation consult.
:: ICU admission.
Acute neurological deterioration
:: Resuscitation and stabilisation of ABCDE's.
:: Short term hyperventilation to PaCO2 25-30.
:: Mannitol 1g/kg IV Bolus.
:: Early CT scan with neurosurgical intervention as required.
Poor prognostic indicators
:: Low GCS (especially motor component).
:: Age >60 years (prognosis deteriorates with increasing age).
:: Absent pupillary reflexes (after systemic resuscitation).
:: Hypotension (systolic BP <90).
:: Hypoxaemia (oxygen saturation <90%).
GUIDELINE LEVEL OF EVIDENCE
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 39
PAGE 40 Inital Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
9.1 Discussion
Recent exhaustive reviews by the Brain Trauma
Foundation and The Cochrane Review Group have
looked at the management of severe head injuries.15,
22,80-87,91 The findings of those reviews are summarised
in Evidence Table 6. Management of severe head injury,
p.56. For the purposes of this guidelines it is assumed
that moderate head injury patients (GCS 9-13) either
improve (80%) and should be managed in a similar
fashion to complicated mild head injuries or deteriorate
(20%) and should be managed as severe head injuries.22
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Initial management of ‘Moderate Head Injuries’ (GCS 9-13) III-2
Standard care15,22,80-87
:: Initial assessment and resuscitation of ABCDE's.
:: Early CT scanning to identify neurosurgically correctable focal intracranial haematomas.
:: Period of ED observation.
:: Prevention of secondary brain injury by avoiding hypoxaemia (O2 saturation <90%) and hypotension (systolic BP <90).
:: Supportive care of ABCDE's.
:: Admit for prolonged hospital observation (24-48 hours) unless rapid clinical improvement, normal CT scan and absence of other risk factors.
:: Early neurosurgical consult if not clinically improving and/or abnormal CT scan.
:: Routine post traumatic amnesia (PTA) testing.
Outcome
:: Approximately 80% of moderate head injuries improve while 20% deteriorate and require management as per severe head injuries.
:: The majority of patients who suffer moderate head injuries will have some degree of cognitive behavioural social sequelae and should be considered for routine follow up with a brain injury rehabilitation service or a neurologist (see Appendix D of the Adult Trauma Clinical Practice Guideline, Initial Management of Closed Head Injury in Adults).
GUIDELINES LEVEL OF EVIDENCE
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 40
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 41
10 When should patients with Closed Head Injury be transferred to
hospitals with neurosurgical facilities?
10.1 Discussion:: Patients with closed head injuries should be observed
in facilities that can manage any complications that
are likely to arise. Clinical judgment regarding risk
of deterioration is required and early neurosurgical
consultation is advisable.
:: Patients with closed head injuries should be
transferred to the nearest appropriate hospital if
there is significant risk of intracranial injury. Transfer
of patients to a hospital with CT scanning facilities
but without neurosurgical services should be avoided.
:: The evidence for which patients are at significant risk
is outlined in the preceding text and evidence tables.
Patients with severe head injuries (GCS 3-8). Consensus
Patients with moderate head injuries (GCS 9-13) if: Consensus
i clinical deterioration
ii abnormal CT scan
iii normal CT scan but not clinically improving
iv CT scan unavailable.
Patients with mild head injuries (GCS 14-15) if: Consensus
i clinical deterioration
ii abnormal CT scan
iii normal CT scan but not clinically improving
iv high risk Mild Head Injury with CT scan unavailable.
NOTE: Consult neurosurgical / retrieval service early.
DEFINITION LEVEL OF EVIDENCE
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 41
::
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 42
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 43
PAGE 44 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Haydel et al Prospective Minor HI 15 LOC / Minimal head injury. Excluded minimal head injury.
200017 study amnesia. Penetrating. Excluded GCS 13-14.
Neurodeficit. Excluded neurodeficit.
Identified risk factors.
Stiell et al Prospective Minor HI 13-15 LOC / Minimal head injury. Excluded minimal head injury.
20019 study amnesia. Penetrating. Excluded neurodeficit.
Neurodeficit. Excluded some risk factors.
Seizure. Identified risk factors.
Coagulopathy.
Representation.
Unstable vitals.
Fabbri et al Prospective Minor HI 14-15 All. Penetrating. Sub-classification into mild,
20046 study medium and high risk based on:
i initial GCS
ii LOC / amnesia
iii risk factors.
Servadei et al Guideline Mild HI 14-15 All. Penetrating. Sub-classification into mild,
20016 (WFNS) medium and high risk based on:
i initial GCS
ii LOC/amnesia
iii risk factors.
Vos et al Guideline Mild TBI 13-15 LOC <30min. Penetrating. Sub-classified categories
20029 (EFNS) PTA <60min. 0-3 based on:
i initial GCS
ii LOC/amnesia
iii risk factors.
Jagoda et al Guideline Mild TBI 15 LOC/ Minimal head injury. Excluded GCS 13-14.
20027 (ACEP) amnesia. Penetrating. Excluded minimal head injury.
Neurodeficit. Identified risk factors.
Coagulopathy.
Multisystem trauma.
Cushman et al Guideline Mild TBI 13-15 LOC/ Penetrating. Excluded minimal head injury.
20018 (EAST) amnesia. Neurodeficit. Used CT scanning to define
LOC <20min. Seizure. mild head injury.
Normal CT scan.
Mower et al Nexus II Study Minor HI 15 All. Penetrating. Excluded GCS 13-14.
200214 methodology Aim to identify risk factors.
Termin- Initial Sub-classification of risk
Author Source ology GCS Inclusion Exclusion of intracranial injury
Evidence Table 1. Definitions of Mild Head Injury
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 44
Dacey III-2 B2 18 N/A 33 59 N/A 6.5 533 N/A 1.5
198635
Teasdale III-2 B1 – – – – – – 7,838 0.03
199027
Shackford III-2 B2 221 33 10.8 646 17.5 3.8 1,899 14.8 3.20
199229
Stein IV B2 120 37.5 – 301 24.2 – 1,117 13.2 –
199278
Jeret III-2 B2 – – – – – – 712 9.4 0.30
199330
Borczuk III-2 C 40 27.5 7.5 197 18.2 3.6 1,211 5.9 0.10
199531
Dunham III-2 B1 1,160 25 3.5 435 12.4 0.5 1,481 3.0 0.13
199624 13 30 – 57 28 – 150 10 –
Culotta III-2 B1 173 28 4.5 755 16 1.6 2,179 4 0.40
199625
Hsiang III-2 B2 45 57.8 20 138 35.5 5.1 1,177 18.5 2.20
199726
Miller III-2 B1 – – – – – – 2,143 6.4 0.20
199747
Nagy III-2 B1 – – – – – – 1,170 3.3 0.34
199968
Haydel III-2 B1 – – – – – – 1,429 6.5 0.40
200017
Stiell III-2 A 110 41 – 522 17 – 2,489 4.8 –
200119
GCS 13 GCS 14 GCS 15
Author Neuro Neuro Neuro
and Level of CT scan SX CT scan SX CT scan SX
Year evidence Quality Patients abnormal required Patients abnormal required Patients abnormal required
No. % % No. % % No. % %
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 45
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Age >14-60 Age>14-60 Age 14-60
Age >60 Age >60 Age >60
:: EVIDENCE TABLES
Evidence Table 2. Initial GCS versus abnormal CT / neurosurgeryStudies showing the relationship between Initial GCS and frequency of abnormal CT scans or neurosurgical intervention
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 45
PAGE 46 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Evidence Table 3. Risk factors for intracranial injury in Mild Head InjuryEvidence table for risk factors predicting intracranial injury based on abnormal CT scan and / or neurosurgical intervention
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Dacey
198635
III-2 – Prospective 610
patients.
– LOC / amnesia.
– GCS 13-15.
Neurosurgical
intervention.
– Initial GCS <15.
– Skull fractures.
Primarily looking at skull x-rays. B2
Masters
198771
III-2 Major risk factors
include:
– persistent abnormal
– mental state focal
– neurological deficit
skull fracture.
Skull x-ray may be useful in
identifying higher risk patients if
CT scan unavailable although
does not exclude intracranial
injury.
B1
Rockswold
198776
IV 215 patients with
severe head injury.
Talk and
deteriorate.
Patients who talk and
deteriorate are high risk.
Confirms seriousness
of ‘deterioration’.
C
Feuerman
198870
IV – Retrospective
373 patients.
– GCS 13-15.
Assess skull x-ray,
CT scan and
observation.
– Initial GCS <15.
Abnormal mental states.
– Neurological deficit.
Loss of consciousness
not useful.
B2
Teasdale
199027
III-2 Prospective 8406
head injuries adults
and children.
Neurosurgical
intervention.
– Initial GCS <15.
– Skull fracture.
– LOC / amnesia
with GCS 15.
History of LOC / amnesia with
initial GCS 15 was only a minor
absolute risk (1 in 6,663 vs 1 in
31,370 if no LOC / amnesia).
B1
Shackford
199229
III-2 – Retrospective
2,166 patients.
– GCS 13-15
isolated head
injuries.
Positive CT or
deterioration.
– GCS 13.
– Neurological deficit.
Also found that ‘admission
to hospital does not guarantee
skilled neurological observation’.
B2
Jeret
199330
III-2 – Prospective
712 patients.
– LOC / amnesia
– GCS 15.
Clinical predictors
of abnormality on
CT scan.
– Age.
– Skull fracture.
– Dangerous
mechanism.
Concluded that no clinical
prediction rule could be
developed to exclude
intracranial injury.
B2
Borczuk
199531
III-2 – Retrospective
1,228 patients.
– GCS 13-15.
Clinical predictors
of abnormal CT
scan.
– Age >60.
– Skull fracture.
– Neurological deficit.
– Cranial soft tissue
injury.
Combined variables
91.6% sensitive for CT
abnormality (46.2% specific).
Absence of LOC / amnesia did
not exclude significant injury.
C
Madden
199528
III-2 – Phase 1,540.
– Phase 2,273.
– Inclusion criteria
not defined.
Clinical predictors
for CT selection.
– Initial GCS <15.
– Abnormal mental status.
– Deterioration.
– LOC.
– Skull fracture (any).
– Focal neurological
signs (pupils).
– Facial injury.
Criteria detected all patients
requiring neurosurgical
intervention in Phase II but
missed two patients with
abnormal scans.
B1
Lee
199532
III-2 Prospective 1,812
patients GCS 15
with one of blow
to head / LOC /
amnesia.
Clinical predictors
of deterioration.
Risk factors fordeterioration:– age >60 years– abnormal mental
status (drowsiness)– focal neurological
deficit– headache– vomiting.
Majority of patients who
deteriorated (28 or 1.5%) did so
in first 24 hours (57%). Of those
who deteriorated 23 patients
required neurosurgery.
B2
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 46
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 47
HE
AD
IN
JU
RY
FU
LL
GU
IDE
LIN
E
:: EVIDENCE TABLES
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Lucchi
199533
IV Retrospective 300
patients ‘Mild Head
Injuries’.
Role of CT in Mild
Head Injuries.
Focal neurological
deficit.
Found there was no
relationship between LOC
and positive CT scans.
B2
Gomez
199634
III-2 – Retrospective
2,484 patients.
– GCS 13-15.
Abnormal CT
findings.
Initial GCS <15.
Skull fracture.
Focal neurological deficit.
Relatively few CT scans
were performed.
B2
Dunham
199624
III-2 – Prospective
2,587 patients.
– GCS 15 with
LOC / amnesia.
– GCS 14 or GCS 13.
To identify risk
factors for
abnormal CT.
Risk factors were:
– GCS <15
– age >60
– cranial soft
tissue injury.
Time since injury average
one hour. Noted that initial
GCS may suffer interobserver
and inter-centre variability.
B1
Culotta
199625
III-2 – Consecutive
3,370 patients.
– GCS 13-15.
LOC present.
Relationship of
GCS to abnormal
scans
neurosurgery.
Initial GCS <15 a
significant risk for
both abnormal CT
scan and neurosurgical
intervention.
Significant difference in injury
patterns between GCS 13-14
and GCS 15.
B1
Miller
199638
III-2 Prospective 1,382
patients with GCS
15 and LOC or
amnesia.
Clinical value of
routine CT scans.
Neurosurgery required
in only 0.2%. Higher
risk if nausea / vomiting
/ headache or skull
fracture.
Routine CT scan not
recommended for patients
with GCS 15 and LOC / amnesia
without other clinical signs and
symptoms of head injury or
skull fracture.
B2
Miller
199747
III-2 – Prospective
2,143 patients.
– GCS 15 LOC.
Predictive value of
severe headache,
nausea, vomiting
and skull fracture
for abnormal
CT scans.
– Severe headache.
– Nausea.
– Vomiting.
– Skull fracture.
Clinical criteria selected shown
to be predictive of need for
neurosurgery (PPV 100%)
but not sensitive for abnormal
CT scans (65% sensitivity).
B1
Hsiang
199726
III-2 – Prospective
1,360 patients.
– GCS 13-15.
Define Mild
Head Injury.
Initial GCS <15
associated with higher
risk abnormal CT,
neurosurgery or
poor outcome.
Vomiting not found
to be a significant risk.
B2
Stiell
199720
III-3 – Retrospective
1,699 patients.
– GCS 13-15.
LOC/amnesia.
– Seven Canadian
major teaching
hospitals.
Assess need for
and use of CT
scanning for
Mild Head Injury.
6.2% with abnormal
CT scan and 0.5% with
extradural. Significant
variability in use of
CT scanning. Missed
haematomas occurred
in those institutions
with higher rates of
CT scanning.
CT scanning not necessarily
useful if use not standardised.
Need for clinical decision.
B2
Nagy
1999 [68]
III-2 – Prospective
1,170 patients
GCS 15.
– LOC/amnesia.
– All CT.
– All 24 hour
observation.
– Using CT to
allow safe
discharge.
– Positive CT or
deterioration.
Neurosurgical
incidence of 4 (0.34%).
No patients deteriorated
following normal
CT scan.
Recommended discharge if
initial CT normal. Study had
969 patients with unknown
LOC emphasising clinical
difficulties in assessment.
B1
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 47
PAGE 48 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Vilke
200045
IV – Prospective
58 patients.
– GCS 15.
– LOC / amnesia.
To determine if
intracranial injury
could be excluded
in initial GCS 15
and normal
complete Neuro
examination.
Need for CT scanning
could not be ruled out
by GCS 15 and normal
neurological examination
on presentation.
B2
Hofman
200037
I Meta analysis of
risk of intracranial
injury after MHI
related to skull
fractures.
Assess use of
skull fracture
in predicting
intracranial injury.
– Prevalence of ICH
0.083 (8 in 100).
– Skull x-ray sensitivity
0.39 specificity 0.95.
– Skull fracture is a
significant risk for ICH.
Skull fracture is associated with
increased risk of intracranial
injury. Skull x-rays are not
useful screening tools.
B1
Haydel
200017
III-2 – Prospective.
– Phase I 520.
– Phase II 909.
– GCS 15.
– LOC / amnesia.
– Age >10.
– Exclusion criteria:
– neurodeficit
– coagulopathy
(only one
patient).
To identify clinical
predictors of
patients who
do not require
CT scan.
All patients with
positive CT scan had
one of seven criteria
(likelihood ratio):
– deficit in STM (15.0)
(anterograde amnesia)
– trauma above clavicle
(11.0) (mostly skull
fracture)
– drug or alcohol
intoxication (11.0)
– seizure (3.0)
– age >60 (3.0)
– headache (2.0)
– vomiting (0.2).
Clinical decision rule of seven
criteria was 100% sensitive
(95% CT 95-100) and 25%
specific with a negative
predictive value of 100%.
Application of the criteria
would have reduced the
number of CT scans by 22%.
B1
Horowitz
200142
IV Chart review
665 patients.
Evaluation of
pre-hospital triage
using brief LOC
or head injuries.
Transient LOC did
not predict need
for neurosurgical
intervention.
Transient LOC not predictive
of neurosurgical intervention.
B2
Sharma
200140
III-2 100 patients Assess importance
of history of
unconsciousness.
– GCS useful.
– Presence of LOC
not useful.
– Duration of LOC
more useful.
A history of brief LOC
did not accurately predict
risk of intracranial injury
in Mild Head Injury.
B1
Stiell
200119
III-2 Prospective multi-
centre 3,121
patients GCS
13-15 one of
LOC / amnesia/
disorientation
Exclusion:
– neurodeficit
– seizure
– coagulopathy
– reassess
– unstable
multi-system.
To develop clinical
decision rule for
use of CT scan
in mild head injury
to identify clinically
important brain
injury.
Developed seven risk
factors.
High risk (OR):
– GCS <15 at 2 hr
post injury (7.3)
– skull fracture (base)
(5.2)
– skull fracture (non-
base) (3.6)
– age > 65 (4.1)
– vomiting (3.8)
Medium risk:
– dangerous mechanism
(2.8)
– retrograde amnesia
>30 min(1.4).
High risk factors alone
were 100% sensitive and
69% specific for predicting
neurosurgical intervention
and require only 32% of
patients to undergo CT scan.
High risk and medium risk
factors together were 98.4%
sensitive and 49.6% specific for
clinically important brain injury
and would require 54% of
patients to have CT scan.
A
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 48
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 49
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Mina
200272
III-2 – Retrospective
inception cohort.
380 patients.
– Any injury with
pre-injury
anticoagulation.
Effect of
anticoagulation
on incidence of
intracranial lesion.
Preinjury anticoagulation
and intracranial injury
has a four- to-five-fold
higher risk of death than
the non-anticoagulated
patient.
Not typical mild head injury
patients, average ISS 17
and GCS 11.
B1
Falmirski
200339
III-2 – Prospective
331 patients.
– GCS 14-15.
– LOC.
To identify
significance of
constitutional signs
and symptoms
in predicting need
for CT scan.
Constitutional signs
and symptoms
predicted presence
of abnormalities on CT
scan requiring a change
in management.
LOC alone is not predictive of
significant head injury and the
need for CT scan. Other clinical
features need to present.
B1
Mack
200374
III-2 – 133 patients.
– GCS 13-15.
Identify risk factors. – Acute Intracranial
injuries 14.3%.
– No risk factors
useful for identifying
intracranial injury
in elderly.
Suggest routine CT scan
if age>65 years.
C
Palchak
200318
III-2 – Prospective
2,043 children.
– Blunt head
trauma.
– All GCS.
– Exclude trivial
injury.
Clinical decision
rule for low risk
traumatic brain
injury.
Significant risk factors
identified for abnormal
CT or neurosurgical
intervention (relative
risk of neurosurgical
intervention brackets).
– Abnormal mental
status (21.7).
– Skull fracture (11.3).
– Focal neurological
deficit (10.6).
– Loss of
consciousness (7.6).
– Seizure (5.3).
– Amnesia (4.7).
– Headache (4.5).
– Vomiting (3.5).
Clinical decision rule identified
using four factors:
1 Abnormal mental status.
2 Evidence of skull fracture
(including scalp haematoma
<2 yr old).
3 Headache.
4 Vomiting.
Abnormal CT prediction:
– 99% sensitive (95-100).
– 26% specific (23-28).
– 99% NPV (98-100).
Neurosurgical intervention
prediction:
– 100% sensitive (97-100).
– 43% specific (40-45).
– 100% NPV (99-100).
B2
:: EVIDENCE TABLES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 49
PAGE 50 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Fabbri
200416
III-2 – Prospective
5,578.
– GCS 14-15.
– Blunt trauma.
– Excluded
penetrating multi-
system injury.
– Validation of
mild head injury
guideline.
– Outcomes of
post traumatic
lesion on CT,
neurosurgical
intervention,
outcome at 6/12.
High risk
GCS 14 or
GCS 15 with
any of:
– neurodeficit
– skull fracture
– risk factors
- coagulopathy
- age >60
- previous
neuro-
symptoms
- epilepsy
- alcohol / drug
intoxication.
Medium risk
GCS 15 with
clinical findings:
– LOC
– amnesia
– vomiting
– diffuse headache.
Low risk
GCS 15 with
none of above.
– Best predictors of post
traumatic injury and
neurosurgical injury
were abnormal GCS,
skull fracture and to a
lesser extent clinical
findings.
– Unfavourable outcome
was most related to
abnormal GCS skull
fracture and not
surprisingly risk factors
or neurological deficits.
– Sensitivity and
specificity of high vs.
medium / low risk
were:
- post traumatic 92%
- sensitive 54%
- specificity
- neurosurgical 93%
- sensitive 53%
specificity
- unfavourable outcome
100% sensitive
- 52% specificity.
– Risks of post concussive
symptoms were similar in
all groups, while rates of
intracranial injury were much
higher in the high risk group.
– Neurosurgical intervention
occurred in 71 patients. None
from low risk, five (0.4%) from
medium risk and 66 (2.4%)
from the high risk group.
– Only 67% of the high risk
group had CT scans due to
cost and local resource issues.
– Age alone was not used as a
risk factor due to cost and
resource issues. Only one
patient (1 of 705) with age
>60 as the sole high risk
factor had an undiagnosed
ICH. Suggests age alone need
not be used as a risk factor.
B1
Ibanez
200462
III-2 – Prospective
1,101 patients.
– GCS 14-15.
– Identify risk
factors.
– Evaluate
guidelines
Acute intracranial
lesions 7.5%.
Risk factors identified
– GCS 14
– LOC
– Neurological deficit
– Skull fracture
– Vomiting
– Severe headaches
– coagulopathies
– Aged>65years
– significant extracranial
lesions
– hydrocephalus
with shunt.
Guidelines
– miss some abnormal
CT scans
– identify clinically
important lesions.
– Findings consistent with
previous studies identifying
risk factors. Support use of
clinical decision rules for
identifying clinically important
lesions. Excellent discussion
on the pros and cons of
clinical guidelines.
– Identify absence of LOC
as not being useful in ruling
out intracranial injury.
A
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 50
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 51
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Stein
199079
III-2 – Retrospective
658 patients.
– GCS 13-15.
Detect abnormal
CT scan and
deterioration in
mild head injury
patients.
None of 542 patients
with initial normal CT
scan deteriorated.
Safe to discharge Mild Head
Injury if normal CT and
neurological examination.
B2
Stein
199278
IV – Retrospective
1,538.
– GCS 13-15.
– LOC / amnesia
‘Normal’
neurology.
– All CT scanned
and admitted.
Detect abnormal
CT scan or
deterioration in
mild head injury
patients.
None of 1,339 patients
with initial normal CT
scan deteriorated.
Safe to discharge mild
head injury if normal CT
scan and normal neurological
examination.
B2
Shackford
199229
III-2 – Retrospective
multi-centre 2.
– 66 patients.
– GCS 13-15.
– Isolated head
injuries.
– To prove that
a patient with
normal
neurological
examination and
normal CT scan
would have
negligible risk of
neurosurgical
lesion.
– Distinguished
abnormal, positive
and relevant
positive scans.
– Normal
neurological
examination
included
abnormal mental
states or focal
neurological
deficit.
– None of 1,170
patients with normal
CT scan deteriorated.
– None of 933
patients with
normal CT scan and
normal neurological
deteriorated.
– Identified significant
risk factors as
GCS 13 or focal
neurological deficit.
– ‘Reliable patients with a
mild head injury with normal
neurological examination and
negative CT scan can be
safely discharged.’
– The utility of abnormal CT
scan or abnormal neurological
examination in detecting
neurosurgical lesions was:
– sensitivity 100%
– specifically 51%
– NPV 100%.
B2
Taheri
199377
IV – Retrospective
310 patients.
– GCS 15.
– Assess safe
discharge for
Mild Head Injury.
Patients who required
neurosurgical
intervention with initial
GCS 15 had either:
i skull fracture
ii neurological deficit.
Safe discharge required:
i initial GCS 15
ii no skull fracture clinically
or radiologically
iii no neurological deficit.
B2
Teasdale
199027
III-2 – Prospective
8,406 patients.
– All head injuries.
Assess risk of
intracranial
haematoma.
Best predictors of injury:
– abnormal LOC
– focal neurology
– skull fractures.
Safe discharge if:
– initial GCS 15
– no focal neurological deficit
– no skull fracture.
B1
Evidence Table 4. Safe discharge of Mild Head Injury
Lee
199532
III-2 – Prospective
1,812 patients.
– GCS 15.
Clinical predictors
of deterioration.
Risk of deterioration
associated with:
– age >60
– abnormal
mental states
– focal deficit
– headache
– vomiting.
Deterioration most common in
first 24 hours due to extradural.
Delayed deterioration due to
subdurals occurred up to a
week later. Initial CT scan may
not rule out risk of deterioration
due to subdurals SIADH
or seizure.
B2
:: EVIDENCE TABLES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 51
PAGE 52 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Dunham
199624
III-2 – Prospective
2,587 patients.
– GCS 13-15.
To identify
risk factors for
abnormal CT scan.
– No patients required
neurosurgical
intervention if initial CT
scan was normal.
– All patients requiring
craniotomy
deteriorated within
four hours of arrival.
Safe discharge implied after
observation if:
i initial GCS 15:
– age <60
– no evidence skull fracture
– no neurodeficit
– no headache
– no vomiting
ii initial GCS 13-14:
– age <60
– normal CT scan
– no persistent neurodeficit
– no persistent headache
– no persistent vomiting.
B1
Hsiang
199726
III-2 – Prospective
1,360 patients.
– GCS 13-15.
– Good follow
up at 6/12.
Define high risk
Mild Head Injury.
GCS 15 with normal
CT scan or skull x-ray
associated with good
outcome.
Safe discharge if initial
GCS 15 with normal
CT scan or skull x-ray.
B2
Nagy
199968
IV – Prospective
1,170 patients.
– GCS 15.
– LOC / amnesia.
– All CT.
– All 24 hour
admissions.
Assess role of
CT scanning in
identifying lesions
and risk of
deterioration.
No patients with
a normal initial
CT scan deteriorated.
Safe discharge if initial GCS 15
and normal CT scan.
B1
Livingston
200069
III-2 – Prospective.
– GCS 14-15.
– LOC / amnesia.
Discharge after
normal CT scan.
– One of 1,788 patients
with initial negative
CT scan deteriorated
requiring neurosurgery.
– Thirty-three others
required variable
intervention.
Safe to discharge if normal CT
scan and clinical improvement.
B2
Jagoda
20027
III-2 Systematic review – Can a patient
with MTBI be
safely discharged
from ED if a non
contrast head CT
scan shows no
acute injuries?
– Exclusion criteria.
– GCS <15.
– Coagulopathy.
– Focal neurological.
– Multi-system trauma.
Safe discharge if:
– GCS 15
– normal examination
– normal CT scan
– six hours observation
– responsible observer.
B1
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 52
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 53
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Fabbri
200416
III-2 – Prospective,
multi-centre.
– 5,578 patients.
– GCS 14-15.
– Blunt trauma.
– Excluded
penetrating
multi-system
injury.
– Validation of a
set of criteria to
classify mild
head injury
patients into
different risk
categories for
management.
– Outcome
measures:
- post traumatic
lesions
- neurosurgical
intervention
- outcome
at 6/12.
1.676 low risk patients:
– 1 missed intracranial
injury
– 52 post concussive
symptoms
1,200 medium
risk patients:
– 22 intracranial injuries
– 0 missed intracranial
injuries
– 49 post concussive
symptoms.
2,702 high risk patients:
– 301 intracranial injuries
– 15 missed intracranial
injuries
– 76 post concussive
symptoms.
Sensitivity and specificity
of high risk vs. medium
/ low risk for identifying
outcomes were:
– post traumatic
lesions:
92% sensitivity
54% specificity
– neurosurgical
intervenion:
93% sensitivity
53% specificity
– unfavourable
outcomes:
100% sensitivity
52% specificity.
Basically validated safe to
discharge patients if:
1 Initial GCS 15:
– no LOC/amnesia
– no headache/ vomiting
– no neurodeficit
– no skull fracture
– no risk factors
– no imaging
– brief ED observation.
2 Initial GCS 15:
– LOC or amnesia or
vomiting or headache
– no neurodeficit
– no skull fracture
– no risk factors
– normal CT scan and
brief ED observation
or
normal skull x-ray and
24 hr hospital observation
or
24 hr hospital observation.
Note that there was no direct
association between risk
categorisation and frequency
of post concussive symptoms.
Note that even on those patients
who had CT scans there were
eight cases of delayed
intracranial haemorrhage not
detected on initial CT (eight of
1774). Thus while CT may
identify most acute extradurals it
may miss delayed subdurals.
B2
Fabbri
200466
III-2 Prospective
1,480 patients.
Safety discharging
‘high risk’ Mild
Head Injury after
normal CT scan.
No significant difference
in outcome between in
hospital and home
monitoring.
Early home discharge ‘safe’. B2
:: EVIDENCE TABLES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 53
PAGE 54 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Rimel
198148
IV – 538 patients.
– GCS 13-15.
Evaluate disability
at three months.
– Persistent headaches
(79%).
– Persistent memory
problems (59%).
– Persistent
unemployment (34%).
Included potentially complicated
patients with initial GCS 13-14.
B2
Alves
198651
III-2 847 patients. Post traumatic
symptoms
characteristics for
mild head injury.
– Persistent headache
commonest symptom.
– Post concussive
symptoms usually
resolved by three
months.
Headache common,
most improve by 3/12.
B1
Hugenholtz
198850
III-2 22 patients. Time to recovery. Mild cognitive
impairment with
deficits in attention and
information processing
gradual recovery over
three months.
Headache and mild cognitive
impairment common.
B2
Williams
199036
III-2 Compared
neurobehavioural
outcome over
six months for:
– uncomplicated
mild head injury
(GCS 13-15)
– complicated mild
head injury (GCS
13-15) with brain
injury or skull
fracture
– moderate head
injury (GCS 9-12).
Assess relationship
of neurobehaviour
complications to
initial injury.
– Outcome for
uncomplicated mild
head injury was better
than complicated
mild head injury or
moderate head injured
(which were similar).
– Age was associated
with poor outcome.
– An abnormal
intracranial lesion
was more predictive of
poor outcome than an
isolated skull fracture.
– Uncomplicated mild head injury
had a good outcome
in 97% of patients. However
complicated mild head injury
or moderate head injury had a
good outcome in only 70-80%.
– Patients with complications
need follow up.
A
Lee
19932
III-2 – 1,812 patients.
– GCS 15.
Outpatient follow
up for three
months.
– At three months post
injury nearly all had
good outcome.
Delayed deterioration
may occur with mild
head injury.
Outcome is generally good. B2
Evidence Table 5. Post concussive symptoms and Mild Head InjurySelected studies and guidelines looking at post concussive symptoms follow up and outcome for Mild Head Injury
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 54
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 55
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Author& Year
Level ofevidence
Study population
Studyobjectives
Relevant findings Comment Quality
Chambers
199688
III-2 Follow up of
mild head
injury patients.
Assessment of
complications.
– Post concussive
symptoms are
common and gradually
reduce with time.
– Headache or memory
problems most
common.
Post concussive symptoms
common.
B2
Nell
200055
III-2 561 patients. – Evaluate
extended
Glasgow
Coma Scale.
A potentially useful tool
to flag mild head injury
patients at increased risk
of cognitive impairment
using an amnesia scale.
B2
Thornhill
200049
III-2 2,962 patients with
head injury (362
with mild head
injury analysed).
To determine
outcome at
one year.
Among patients with no
pre-existing problems
about 1/3 failed to
achieve good outcome
significant problems
included cognitive
behavioural and
employment dysfunction.
Significant disability may
occur in the undifferentiated
GCS 13-15 ‘mild’ head injury
group.
B2
Cushman
20018
III-2 EAST guidelines. Recommendations
(paraphrased).
– Mild cognitive
impairment is common
and usually resolved
by one month post
injury (Level II).
– Patients with post
concussive symptoms
that persist for more
than 6/52 should
have formal
neuropsychological
testing (level II).
Persistent post concussive
symptoms may identify a
subgroup at risk of prolonged
cognitive deficits.
N/A
Vos
20029
III-2 European (EFNS)
guidelines.
Recommendations
(paraphrased).
– Patients with high risk
head injury admitted to
hospital should have
outpatient follow up.
– Post concussive
symptoms are
common but usually
resolve by 3-6 months.
Post concussive symptoms
persisting after six months may
benefit from neuropsychological
testing.
N/A
Savola
200389
III-2 172 patients with
mild head injury
GCS 13-15.
Early predictors of
post concussive
symptoms.
Best beside predictors
of post concussive
symptoms were:
– skull fracture
– dizziness on admission
– headache on
admission.
Serum protein S-100 B
was also found to be a
good predictor of post
concussive symptoms.
Initial GCS and duration
of PTA were not found
to be good predictors.
B1
:: EVIDENCE TABLES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 55
PAGE 56 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
Evidence Table 6. Management of Severe Head Injury
Author& Year
Level ofevidence
Quality Study objectives Relevant findings
Brain Trauma
Foundation15
I B1 Systematic review of the
management and prognosis
of severe traumatic brain
injury.
Recommendations
Overall management strategies:
– Organised trauma systems.
– Initial systemic ABCDE resuscitation fundamental to successful
neurological outcome.
– Prevention of secondary brain injury from hypoxaemia
or hypotension crucial to outcome.
– Specific therapy directed at raised intracranial pressure should
not interfere with systemic resuscitation.
Hypoxaemia and hypotension:
– Systemic hypoxaemia (SaO2<90) and hypotension
(systolic BP <90) following head injury are both associated
with poor outcome.
– Adequate oxygenation and fluid resuscitation should be the
priority in multiply injured patients.
ICP Monitoring
– ICP monitoring should be used as a guide to optimise cerebral
perfusion pressure.
– Specific indications for ICP monitoring include:
i GCS 3-8 abnormal CT scan.
ii GCS 3-8 normal CT scan if two of:
- age >40
- motor posturing
- systolic BP <90.
Supportive care ABCDE’s
– Supportive care with attention to stabilising ABCDE’s, adequate
nutrition, appropriate posturing (30O head up), basic nursing
care and prevention of complications is more effective than
most other interventions.
Anticonvulsants
– Anticonvulsants such as phenytoin are effective at preventing
early post traumatic seizures but do not prevent late post
traumatic seizures.
Hyperventilation
– Routine hyperventilation (PaCO2<35mmHg) should be avoided
in head injury patients as it is associated with poor outcome.
– Acute hyperventilation (PaCO225-30mmHg) is proven effective
for short term reduction of raised ICP associated with acute
neurological deterioration.
Mannitol
– Mannitol (0.5g-1g/kg) is effective at reducing raised ICP.
– Care should be used to avoid hypovolemia and serum
osmolarity >320.
– It should be largely reserved for patients with acute neurological
deterioration.
Therapies not shown to be effective
– Barbiturates.
– Steroids.
Factors clearly associated with poor prognosis
– GCS (lower GCS = worse outcome with motor component
most predictive).
– Age >60 (gradual trend with age most significant at age >60).
– Absent pupillary reflexes (when systemic ABCDE causes
eliminated).
– Hypotension (systolic BP <90mmHg).
– Hypoxaemia (O2 sat <90%).
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 56
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 57
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Author& Year
Level ofevidence
Quality Study objectives Relevant findings
Forsyth
200380
I A To determine if routine
ICP monitoring in acute
coma is beneficial.
Insufficient evidence to clarify role.
Roberts et al
200381
I A To determine role
of hyperventilation in
head injury.
Insufficient evidence to clarify role.
Roberts et al
200382
I A To determine role of
mannitol for acute
traumatic brain injury.
Mannitol beneficial for pre-operative management of raised ICP.
Schierhout et al
200191
I A To determine role of
anti-epileptics for acute
traumatic brain injury.
– Anti-epileptics reduce early post traumatic seizures.
– Anti-epileptics do not alter outcome.
– Anti-epileptics do not prevent late post traumatic seizures.
Alderson et al
200383
I A To determine role of
corticosteroids for acute TBI.
Insufficient evidence.
Roberts
200384
I A To determine role of
barbiturates for acute TBI.
No evidence that barbiturates improve outcome.
Langham
200385
I A To determine role of
calcium channel blockers
for acute TBI.
Insufficient evidence.
Gadkary
200386
I A To determine role of
therapeutic hypothermia
for acute TBI.
No evidence that hypothermia is beneficial.
Fleminger
200387
I A To determine role
of pharmacological
management for agitation
and aggression following
acquired brain injury.
B blockers (especially propranolol) may be effective in reducing
aggression and agitation in the long term.
:: EVIDENCE TABLES
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 57
::
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 58
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 59
The Westmead Post Traumatic Amnesia (PTA) Scale, developed by NEV Marosszeky, L Ryan, EA Shores, J Batchelor and
JE Marosszeky, was published in 1986. The Westmead PTA Scale consists of seven orientation questions and five memory
items designed to objectively measure the period of PTA. The Westmead PTA Scale is a standardised and prospective
measure of PTA. A person is said to be out of PTA if they can achieve a perfect score on the Westmead PTA Scale for
three consecutive days.
The Westmead PTA Scale form (as seen by the example on the following page) and nine picture cards are required
to perform the test. As the test was designed to measure PTA in a standard fashion to enable comparison of patients
from different hospitals, the supplied picture cards must be used. They are available for purchase with instruction on
their use from the Department of Rehabilitation Medicine, Westmead Hospital, Westmead NSW 2145 for a minimal fee.
More information is available on the Westmead PTA Scale at:
web. http://www.psy.mq.edu.au/pta/index.html or
email. [email protected]
PAGE 60 Initial Management of Closed Head Injury in Adults :: NSW ITIM
Westmead PTA scale
HE
AD
IN
JU
RY
GU
IDE
LIN
EAPPEND IX 1
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 60
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 61
HE
AD
IN
JU
RY
GU
IDE
LIN
E
APPEND IX A :: WESTMEAD PTA SCALE
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 61
PAGE 62 Initial Management of Closed Head Injury in Adults :: NSW ITIM
Glasgow Coma Scale
HE
AD
IN
JU
RY
GU
IDE
LIN
EAPPEND IX B
Eye Opening (E) Verbal Response (V) Motor Response (M)
4 = Spontaneous 5 = Normal conversation 6 = Normal
3 = To voice 4 = Disoriented conversation 5 = Localizes to pain
2 = To pain 3 = Words, but not coherent 4 = Withdraws to pain
1 = None 2 = No words... only sounds 3 = Decorticate posture
1 = None 2 = Decerebrate
1 = None
Total (Glasgow Coma Score) = E+V+M
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 62
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 63
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Mild Head Injury discharge advice
APPEND IX C
Important points about Mild Head Injury:: You have been assessed as having a Mild Head Injury. This means that the doctor has not found any evidence
to suggest a serious underlying brain injury and feels you are suitable for observation at home.
:: Due to the small risk of you developing a serious complication after discharge from hospital, it is important that
a responsible adult observes you for the next 24 hours and returns you to hospital if necessary.
:: Although you do not have evidence of a serious underlying brain injury, it is common to have some mild post concussion symptoms for up to several weeks.
:: If you still have significant post concussion symptoms that are not improving after a few days you
should see your local doctor. Your local doctor may refer you for specialist review if your symptoms persist.
Please read the following fact sheet that details what you should expect while you recover from your head injury.
WARNING SIGNS!
Return to HOSPITAL IMMEDIATELY if you deteriorate and develop any of the following:
:: excessive drowsiness or lethargy
:: confusion or disorientation
:: abnormal behaviour or irritability
:: fitting or seizures
:: blurred vision or slurred speech
:: severe headache
:: persistent vomiting
:: abnormal clumsiness
FIRST 24–48 HOURS
! Warning signs You should be observed and returned to hospital if you develop any of the above warning signs.
Rest / Sleeping Rest and avoid strenuous activity for at least 24 hours. It is alright for you to sleep
tonight but you should be woken every four hours by someone to make sure you
are alright.
Driving Do not drive for at least 24 hours, as you may be unable to concentrate properly.
This is a legal requirement.
Drinking / Drugs Do not drink alcohol, take sedatives or use recreational drugs in the next 48 hours,
as they may make you feel worse and / or mask symptoms of deterioration.
Pain relief Use paracetamol or paracetamol/codeine for headaches. Do not use aspirinor an anti-inflammatory pain reliever such as ibuprofen or naproxen (NSAIDs)
which may increase the risk of bleeding.
zz
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 63
PAGE 64 Initial Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EAPPEND IX C :: MILD HEAD INJURY D ISCHARGE ADV ICE
THE FIRST FOUR WEEKS
Post concussion symptomsThe following symptoms are common after a Mild Head Injury and usually resolve within a few weeks:
:: fatigue :: poor concentration :: mood swings :: dizziness :: mild headache
:: forgefulness :: mild nausea :: mild behavioural change
Rest / Sleeping It is important to get adequate amounts of sleep as you may feel more tired
than normal.
Relationships Sometimes your post concussion symptoms will affect your relationship with family and
friends, as you may suffer irritability and mood swings. See your local doctor if you or
your family are worried.
Driving Do not drive until you feel much better and can concentrate properly.
Drinking / Drugs Do not drink alcohol or use recreational drugs until you are fully recovered
– they will make you feel much worse.
Work / Study Most people return to work or study within a few days following a Mild Head Injury.
Some people find it difficult to concentrate properly when returning to work or study
following a Mild Head Injury. See your doctor and notify your employer or teachers
if you are having problems at work or with study. You may need to take on reduced
work or study duties for a short period.
Sport / Lifestyle It is important to avoid another head injury. You should avoid contact sports and
other activities where you may suffer another head injury for at least four weeks
– 'if in doubt – sit it out'.
Recovery You should start to feel better within a few days and be ‘back to normal’ within
about four weeks. See your local doctor if you are not starting to feel better
within a few days of your head injury.
zz
PERSISTENT POST CONCUSSION SYMPTOMSThe majority of people recover rapidly following a Mild Head Injury. However, a minority of people may
suffer from persistent post concussion symptoms. If you still have post concussion symptoms a few days after
a Mild Head Injury, you should see your local doctor. Your local doctor will monitor these symptoms, which
would normally improve within four weeks. Your local doctor may refer you to a Traumatic Brain Injury Clinic
or a Neurologist for specialist review if your symptoms persist.
SUMMARY:: Return to hospital immediately if you develop any of the warning signs
:: Most people recover rapidly from mild head injuries and have no significant problems
:: Mild post concussion symptoms are common but usually resolve within a few weeks
:: See your local doctor if you still have post concussion symptoms a few days after your head injury
Dr. Duncan Reed, Director of Trauma, Gosford Hospital
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 64
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 65
HE
AD
IN
JU
RY
GU
IDE
LIN
E
NSW Brain Injury Rehabilitation Program (BIRP)
APPEND IX D
SYDNEY
WEST
GREATER SOUTHERN
HUNTER /
NEW ENGLAND NO
RTH
CO
AST
MET
RO
North Coast Head Injury
•
• New EnglandBrain InjuryRehabilitation
•Dubbo Brain InjuryRehabilitation
• Mid Western BrainInjury Rehabilitation
South West BrainInjury Rehabilitation•
• Southern AreaBrain Injury Service
• Hunter BrainInjury Service
Westmead BrainInjury Rehabilitation • • Royal
RehabilitationCentreSydney
•Liverpool Hospital
Brain Injury Rehabilitation Unit
• Illawarra AreaBrain Injury Service
SYD STH
WEST
SOUTH EASTERN
SYDNEY / ILLAWARRA
NORTHERN
SYDNEY /
CENTRAL COAST
GREATER WESTERN
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 65
The following search phrases were used in Medline:
1 exp Head Injuries, Closed/
2 exp *tomography, x-ray/
3 Patient Discharge/
4 Patient Transfer/
5 intubation/ or exp intubation, intratracheal/
6 *Intracranial Pressure/
7 Drainage/
8 (7 and (ventricular or intra?ventricular or extra?ventricular).mp.) or ((ventricular or intra?ventricular or extra?ventricular) adj drain)).mp.
9 (icp monitor$ or intracranial pressure monitor$).mp.
10 exp Aggression/
11 exp Mannitol/
12 exp Hyperventilation/
13 Adrenal Cortex Hormones/
14 Craniotomy/
15 Trephining/
16 exp emergency treatment/
17 exp *head injuries, closed/su, th or (exp head injuries, closed/ and management.mp.)
18 (or/2-6) or (or/8-16)
19 (1 and 18) or 17
The following search phrases were used in Embase:(head injury/ and closed$.mp.) or (closed head injury or closed head trauma$).mp. [mp=title, abstract, subject headings, drug trade name, original title, device manufacturer, drug manufacturer name]
2 exp computer assisted tomography/
3 patient transport/ or discharg$.mp.
4 exp RESPIRATORY TRACT INTUBATION/ or INTUBATION/
5 Intracranial Hypertension/
6 cerebrospinal fluid drainage/
7 [(ventricular or intra?ventricular or extra?ventricular) adj drain]).mp.
8 (icp monitor$ or intracranial pressure monitor$).mp.
9 exp aggression/
10 Mannitol/
11 exp Hyperventilation/
12 exp Corticosteroid/
13 craniotomy/
14 (trephin$ or burr hole$).mp.
15 exp emergency treatment/
16 (su or th).fs. or management.mp.
17 1 and (or/2-16)
18 limit 17 to human
PAGE 66 Initial Management of Closed Head Injury in Adults :: NSW ITIM
List of search terms to identify studies
HE
AD
IN
JU
RY
GU
IDE
LIN
EAPPEND IX E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 66
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 67
HE
AD
IN
JU
RY
GU
IDE
LIN
E
References
1 Newcombe R, G Merry 1999, The management of
acute neurotrauma in rural and remote locations:
a set of guidelines for the care of head and spinal
injuries, Journal of Clinical Neuroscience,
6(1): p.85-93.
2 NSW Institute of Trauma & Injury Management
2002, NSW Trauma Minimum Data Set 2002
Annual Report.
3 Khan F, Baguley IJ, Cameron ID 2003, Rehabilitation
after traumatic brain injury, Medical Journal of
Australia, 178(6): p.290-5.
4 Fortune N 1999, The definition incidence and
prevalence of acquired brain injury in Australia,
Australian Institute of Health and Welfare,
Catalogue DIS 15.
5 Tate RL, McDonald S, Lulham JM 1998, Incidence
of hospital-treated traumatic brain injury in an
Australian Community, Australian and New Zealand
Journal of Public Health, 22(4): p.419-23.
6 Servadei F et al 2001, Defining acute mild head injury
in adults: a proposal based on prognostic factors,
diagnosis, and management [Review], Journal of
Neurotrauma, 18(7): p.657-64.
7 Jagoda AS et al 2002, Clinical policy: neuroimaging
and decision making in adult mild traumatic brain
injury in the acute setting, Annals of Emergency
Medicine, 40(2): p.231-49.
8 Cushman JG et al 2001, Practice Management
Guidelines for the Management of Mild Traumatic
Brain Injury: the EAST practice management
guidelines work group, Journal of Trauma-Injury
Infection & Critical Care, 51(5): p.1016-26.
9 Vos PE et al 2002, EFNS guidelines on mild traumatic
brain injury: report of an EFNS task force, European
Journal of Neurology, 9(3): p.207-19.
10 Bellner J, Ingebrigtsen T, Romner B 1999, Survey of
the management of patients with minor head injuries
in hospitals in Sweden, Acta Neurologica
Scandinavica, 100(6): p.355-359.
11 Dickinson K et al 2000, Size and quality of
randomised controlled trials in head injury:
review of published studies, British Medical
Journal, 320(7245): p.1308-11.
12 Klauber MR et al 2002, Determinants of head injury
mortality: importance of the low risk patient,
Neurosurgery, 24(1): p.31-6.
13 Narayan RK et al 2002, Clinical trials in head injury.
Journal of Neurotrauma, 19(5): p.503-57.
14 Mower WR et al 2002, Developing a clinical decision
instrument to rule out intracranial injuries in patients
with minor head trauma: methodology of the NEXUS
II investigation, Annals of Emergency Medicine,
40(5): p.505-14.
15 The Brain Trauma Foundation 2000, The American
Association of Neurological Surgeons, and The Joint
Section of Neurotrauma and Critical Care,
Management and prognosis of severe traumatic brain
injury, Journal of Neurotrauma, 17: p.6-7.
16 Fabbri A et al 2004, Prospective validation
of a proposal for diagnosis and management
of patients attending the emergency department
for mild head injury, Journal of Neurology,
Neurosurgery & Psychiatry 75(3): p.410-6.
17 Haydel MJ et al 2000, Indications for computed
tomography in patients with minor head injury,
New England Journal of Medicine, 343(2): p.100-5.
18 Palchak MJ et al 2003, A decision rule for identifying
children at low risk for brain injuries after blunt head
trauma, Annals of Emergency Medicine,
42(4): p.492-506.
19 Stiell IG et al 2001, The Canadian CT Head rule for
patients with minor head injury, Lancet, 357(9266):
p.1391-6.
20 Stiell IG et al 1997, Variation in ED use of computed
tomography for patients with minor head injury.
Annals of Emergency Medicine, 30(1): p.14-22.
21 Bulger EM et al 2002, Management of severe
head injury: institutional variations in care and
effect on outcome Critical Care Medicine,
30(8): p.1870-6.
22 American College of Surgeons Committee
on Trauma 1997, Advanced Trauma Life Support
for Doctors (6th Edition).
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 67
PAGE 68 Inital Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
HE
AD
IN
JU
RY
GU
IDE
LIN
E
23 Cook LS et al 1994, Identification of ethanol-
intoxicated patients with minor head trauma
requiring computed tomography scans,
Academic Emergency Medicine, 1(3): p.227-34.
24 Dunham CM, S Coates, G Cooper 1996,
Compelling evidence for discretionary brain
computed tomographic imaging in those patients
with mild cognitive impairment after blunt trauma,
Journal of Trauma-Injury Infection & Critical Care,
41(4): p.679-86.
25 Culotta VP et al 1996, Clinicopathological
heterogeneity in the classification of mild
head injury, Neurosurgery, 38(2): p.245-50.
26 Hsiang JN et al 1997, High-risk mild head injury.
[comment], Journal of Neurosurgery, 87(2): p.234-8.
27 Teasdale GM et al 1990, Risks of acute traumatic
intracranial haematoma in children and adults:
implications for managing head injuries, BMJ,
300(6721): p.363-7.
28 Madden C et al 1995, High-yield selection criteria
for cranial computed tomography after acute trauma,
Academic Emergency Medicine, 2(4): p.248-53.
29 Shackford SR et al 1992, The clinical utility of
computed tomographic scanning and neurologic
examination in the management of patients with
minor head injuries, Journal of Trauma-Injury
Infection & Critical Care, 33(3): p.385-94.
30 Jeret JS et al 1993, Clinical predictors of abnormality
disclosed by computed tomography after mild head
trauma, Neurosurgery, 32(1): p.9-15.
31 Borczuk P 1995, Predictors of intracranial injury in
patients with mild head trauma, Annals of
Emergency Medicine, 25: p.731-6.
32 Lee ST et al 1995, Relative risk of deterioration
after mild closed head injury, Acta Neurochirurgica,
135(3-4): p.136-40.
33 Lucchi S et al, The role of computerized tomography
in the management of mild head injury patients.
Rivista di Neuroradiologia, 1995. 8(3): p.415-428.
34 Gomez, P.A., et al., Mild head injury: differences in
prognosis among patients with a Glasgow Coma
Scale score of 13 to 15 and analysis of factors
associated with abnormal CT findings. British Journal
of Neurosurgery, 1996. 10(5): p.453-60.
35 Dacey RGJ et al 1986, Neurosurgical complications
after apparently minor head injury, Assessment
of risk in a series of 610 patients, Journal of
Neurosurgery, 65(2): p.203-10.
36 Williams DH, Levin HS, Eisenberg HM 1990,
Mild head injury classification, Neurosurgery,
27(3): p.422-8.
37 Hofman PA et al 2000, Value of radiological diagnosis
of skull fracture in the management of mild head
injury: meta-analysis, Journal of Neurology,
Neurosurgery & Psychiatry, 68(4): p.416-22.
38 Miller EC, Derlet RW, Kinser D 1996, Minor head
trauma: Is computed tomography always necessary?
Annals of Emergency Medicine, 1996. 27(3): p.290-4.
39 Falimirski ME et al 2003, The need for head
computed tomography in patients sustaining loss
of consciousness after mild head injury, Journal of
Trauma-Injury Infection & Critical Care, 55(1): p.1-6.
40 Sharma P et al 2001, How important is history of
unconsciousness in head injury patients? Journal of
the Indian Medical Association, 99(2): p.81-3.
41 Haydel MJ, AD Shembekar 2003, Prediction of
intracranial injury in children aged five years and
older with loss of consciousness after minor head
injury due to nontrivial mechanisms Annals of
Emergency Medicine, 2003. 42(4): p.507-14.
42 Horowitz BZ, Earle OJ 2001, Should transient loss of
consciousness in blunt head trauma be a pre-hospital
trauma triage criterion? Journal of Emergency
Medicine, 21(4): p.381-6.
43 Ruff RM, Jurica P 1999, In search of a unified
definition for mild traumatic brain injury, Brain Injury
13(12): p.943-52.
44 Hahn YS, DG McLone 1993, Risk factors in the
outcome of children with minor head injury,
Pediatric Neurosurgery, 19(3): p.135-42.
45 Vilke GM, Chan TC, Guss DA 2000, Use of a
complete neurological examination to screen for
significant intracranial abnormalities in minor head
injury, American Journal of Emergency Medicine.
18(2): p.159-63.
46 Harad FT, Kerstein MD 1992, Inadequacy of bedside
clinical indicators in identifying significant intracranial
injury in trauma patients, Journal of Trauma-Injury
Infection & Critical Care, 32(3): p.359-61; discussion
361-3.
47 Miller EC, Holmes JF, Derlet RW 1997, Utilizing
clinical factors to reduce head CT scan ordering for
minor head trauma patients, Journal of Emergency
Medicine,15(4): p.453-7.
48 Rimel RW et al 1981, Disability caused by minor head
injury. Neurosurgery, 9(3): p.221-8.
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 68
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 69
HE
AD
IN
JU
RY
GU
IDE
LIN
E
:: REFERENCES
49 Thornhill S et al 2000, Disability in young people and
adults one year after head injury: prospective cohort
study, BMJ, 320(7250): p.1631-5.
50 Hugenholtz H et al 1988, How long does it take to
recover from a mild concussion? Neurosurgery,
22(5): p.853-8.
51 Alves WM 1986, Understanding posttraumatic
symptoms after minor head injury, Journal of
Head Trauma Rehabilitation, 1:p.1-12.
52 Mittenberg W, Strauman S 2000, Diagnosis of mild
head injury and the postconcussion syndrome
[Review], Journal of Head Trauma Rehabilitation,
15(2): p.783-91.
53 Teasdale G, Jennett B 1974, Assessment of coma
and impaired consciousness: a practical scale, Lancet,
2(7872): p.81-4.
54 Fearnside M, McDougall P 1998, Moderate head
injury: a system of neurotrauma care, Australia and
New Zealand Journal of Surgery, 68(1): p.58-64.
55 Nell V, Yates DW, Kruger J 2000, An extended
Glasgow Coma Scale (GCS-E) with enhanced
sensitivity to mild brain injury, Archives of Physical
Medicine & Rehabilitation, 81: p.614-7.
56 Winkler J, Rosen P, Alfry EJ 1984, Prehospital use
of the Glasgow Coma Scale in severe head injury,
Journal of Emergency Medicine, 2(1): p.1-6.
57 Garner AA, Schoettker P 2002, Efficacy of
pre-hospital interventions for the management
of severe blunt head injury [Review] [32 refs],
Injury. 33(4): p.329-37.
58 High WMJ, Levin HS, Gary HEJ 1990, Recovery
of orientation following closed-head injury,
Journal of Clinical & Experimental Neuropsychology,
12(5): p.703-14.
59 Hoffman JR et al 1998, Selective cervical spine
radiography in blunt trauma: methodology of
the National Emergency X-Radiography Utilization
Study (NEXUS), Annals of Emergency Medicine,
32(4): p.461-9.
60 Stiell IG et al 1992, A study to develop clinical
decision rules for the use of radiography in acute
ankle injuries, Annals of Emergency Medicine,
21(4): p.384-90.
61 Rosengren D et al 2004, The application of North
American CT scan criteria to an Australian population
with minor head injury, Emergency Medicine
Australasia, 16: p.195-200.
62 Ibanez JA, Pedraza F, Sanchez E, Poca M,
Rodriguez A, Rubio D 2004, Reliability of clinical
guidelines in the detection of patients at risk
following mild head injury: results of a prospective
study, Journal of Neurosurgery, 100(5): p.825-34.
63 Geijerstam AF, Britton M, Marke LA 2004, Mild head
injury: observation or computed tomography?
Economic aspects by literature review and decision
analysis, Journal of Emergency Medicine,
21(1): p.54-8.
64 Brenner D et al 2001, Estimated risks of radiation-
induced fatal cancer from pediatric CT, American
Journal of Roentgenology, 176(2): p.289-96.
65 Hall P et al 2004, Effect of low doses of ionising
radiation in infancy on cognitive function in
adulthood: Swedish population based cohort study,
BMJ. British Medical Journal, 2004. 328(7430): p.19.
66 Fabbri A et al 2004, Which type of observation
for patients with high-risk mild head injury and
negative computed tomography? European
Journal of Emergency Medicine, 11(2): p.65-9.
67 Riesgo P et al 1997, Delayed extradural hematoma
after mild head injury: report of three cases, Surgical
Neurology, 48(3): p.226-31.
68 Nagy KK et al 1999, The utility of head
computed tomography after minimal head injury,
Journal of Trauma-Injury Infection & Critical Care,
46(2): p.268-70.
69 Livingston DH et al 2000, Emergency department
discharge of patients with a negative crainial
computed tomography scan after minimal head
injury, Annals of Surgery, 232(1): p.126-32.
70 Feuermann T et al 1988, Value of skull radiography,
head computed tomography scanning, and admission
for observation in cases of minor head injury,
Neurosurgery, 22(3): p.449-53.
71 Masters SJ et al, Skull x-ray examinations after head
trauma. Recommendations by a multidisciplinary
panel and validation study, New England Journal of
Medicine, 316(2): p.84-91.
72 Mina AA et al, Intracranial complications of preinjury
anticoagulation in trauma patients with head injury,
Journal of Trauma-Injury Infection & Critical Care,
2002, 53(4): p.668-72.
73 Haug RH, Pittman T 2002, Computed tomography
of head injury. Atlas of the Oral & Maxillofacial
Surgery Clinics of North America, 10(2): p.149-66.
HE
AD
IN
JU
RY
GU
IDE
LIN
E
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 69
PAGE 70 Inital Management of Closed Head Injury in Adults :: NSW ITIM
HE
AD
IN
JU
RY
GU
IDE
LIN
EADULT TRAUMA CL IN ICAL PRACT ICE GU IDEL INES
HE
AD
IN
JU
RY
GU
IDE
LIN
E
74 Mack LC, Silva SB, Hogan JC 2003, The use of
computed tomography in elderly patients sustaining
minor head trauma, Journal of Emergency Medicine
24(2): p.157-162.
75 NSW Institute of Trauma & Injury Management,
Trauma Death Review Committee 2003/2004.
76 Rockswold GL, Leonard PR, Nagib MG,
Analysis of management in thirty-three closed
head injury patients who ‘talked and deterioriated’,
Neurosurgery, 1987. 21(1): p.51-5.
77 Taheri PA et al 1993, Can patients with minor head
injuries be safely discharged home? Archives of
Surgery, 128(3): p.289-92.
78 Stein SC, Ross SE 1992, Mild head injury: a plea for
routine early CT scanning, Journal of Trauma-Injury
Infection & Critical Care, 33(1): p.11-13.
79 Stein SC, Ross SE, The value of computed
tomographic scans in patients with low-risk head
injuries, Neurosurgery, 1990. 26(4): p.638-40.
80 Forsyth R P, Baxter, Elliott T 2003, Routine intracranial
pressure monitoring in acute coma, The Cochrane
Library (3).
81 Roberts I, Schierhout G 2003, Hyperventilation
therapy for acute traumatic brain injury, The
Cochrane Library (3).
82 Roberts I 2003, Aminosteriods for acute traumatic
brain injury, The Cochrane Library (3).
83 Alderson P, Roberts I 2003, Corticosteriods for acute
traumatic brain injury, The Cochrane Library, (3).
84 Roberts I 2003, Barbiturates for acute traumatic brain
injury, The Cochrane Library (3).
85 Langham J et al 2003, Calcium channel blockers for
acute traumatic brain injury, The Cochrane Library, (3).
86 Gadkary CS, Alderson P, Signorini DF 2002,
Therapeutic hypothermia for head injury,
[update of Chochrane Dateabase Syste Rev.
2000:(2):CD001048:PMID: 10796743]. [Review]
[34 refs], Cochrane Database of Systematic
Reviews, 1(CD001048).
87 Fleminger S, Greenwood RJ, Oliver DL,
Pharmacological management for agitation
and aggression in people with acquired brain
injury, Cochrane Database of Systematic Reviews,
2003. 1(CD003299).
88 Chambers J et al 1996, Mild traumatic brain injuries
in low-risk trauma patients, Journal of Trauma-Injury
Infection & Critical Care, 41(6): p.976-80.
89 Savola O, Hillbom M 2003, Early predictors of
post-concussion symptoms in patients with mild
head injury, European Journal of Neurology,
10(2): p.175-81.
90 Ponsford J et al 2004, Use of the Westmead PTA
Scale to monitor recovery of memory after mild
head injury, Brain Injury, 18(6): p.603-614.
91 Schierhout G, Roberts I 2001, Anti-epileptic drugs
for preventing seizures following acute traumatic
brain injury. Cochrane Database of Systematic
Reviews 2001, Issue 4.
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 70
Initial Management of Closed Head Injury in Adults :: NSW ITIM PAGE 71
HE
AD
IN
JU
RY
GU
IDE
LIN
E
Notes
HeadTrauma_FullReport.qxd 6/8/07 1:02 PM Page 71
::
SHPN (TI) 070020
TraumaCvr_FullReport.qxd 3/5/07 6:35 AM Page 4