Intended for healthcare professionals


Treating head injuries

BMJ 2002; 325 doi: (Published 31 August 2002) Cite this as: BMJ 2002;325:454

Step by step, we are improving the care of the head injured patient

  1. Jonathan Wasserberg, senior lecturer in neurosurgery (jwasserberg{at}
  1. Department of Neurosurgery, Queen Elizabeth Hospital, Birmingham B15 2TH

    Up to 1 million people a year in the United Kingdom attend an accident and emergency department because of a head injury. Of these 90% are classified as minor (with a Glasgow coma score of 15) or mild (13 or 14), 5% as moderate (9-12), and 5% as severe (3-8).1 Road traffic accidents cause most of the severe head injuries and are likely to become the third most common cause of death and disability worldwide over the next 20 years.2 Even patients with “mild” injury (13 or 14) can suffer long term disability, with up to 47% being classed as moderately or severely disabled one year after injury.3 Magnetic resonance imaging of the brain of patients who have an initial Glasgow coma score of 13 or 14 shows a high incidence of parenchymal lesions, suggesting that even “mild” cases may sustain a significant brain injury.4 For such a major health problem, guidelines for management based on the best available evidence are essential. Although the literature on head injury might be criticised for lacking large randomised trials,5 notable improvements have been made in our management of this complex and heterogeneous condition over the past decade.

    Initial assessment and triage of patients is now based on clear guidelines—in the United Kingdom on those published by the Society of British Neurosurgeons and the Scottish Intercollegiate Guidelines Network. 6 7 Patients are differentiated primarily by Glasgow coma score at admission into patients who require immediate resuscitation, computed tomography scanning and neurosurgical referral, admission, and continuing observation, and patients who can be discharged with appropriate advice. The thorny issue of when an x ray film of the skull is indicated is now clearer as patients with a reduced conscious level and signs of external injury to the head now proceed directly to computed tomographic scanning of the head. Patients with a normal conscious level, no signs of external injury, and a history of a trivial blow to the head can be discharged. Patients who have lost consciousness, fallen more than 60 cm in height, have a full thickness scalp laceration, scalp haematoma or other features of a significant blow to the head, loss of memory, or an inadequate history should have an x ray of the skull. Patients found to have a skull fracture should have a computed tomography head scan. New guidelines covering many aspects of the management of head injury are expected from the United Kingdom's National Institute for Clinical Excellence (NICE) in 2003.

    The need for neurosurgeons to take a leadership role in developing local guidelines and protocols for head injury was highlighted in a recent report from the Royal College of Surgeons of England.8 In most instances patients with severe head injury have to be transferred to a neurosurgical unit. Since the time from injury to evacuation of an intracranial haematoma is critical for a good outcome (four hours is considered the maximum permissible delay), all components of the healthcare system must operate smoothly. Hospitals accepting patients with head injuries should have 24 hour facilities for computerised tomography scanning, with an image link facility to the regional neurosurgical unit. Although rapid transfer is important, it should not compromise basic resuscitation and restoration of physiological stability. Standards for transfer of such patients have been laid out in guidelines that stress the importance of experienced anaesthetic staff travelling with the patient and avoiding hypotension and hypoxia during transfer.9

    Patients with a severe head injury (Glasgow coma score less than 8) are currently treated in either general or specialist intensive care units. In specialist units, up to 75% of patients with severe injury have their intracranial pressure monitored.10 Raised intracranial pressure is treated according to a protocol that introduces successive treatments as the pressure becomes increasingly difficult to control. These treatments include hyperventilation, cerebrospinal fluid drainage, infusion of mannitol, hypothermia, barbiturates, and decompressive craniotomy.10 Determining the effect of an individual treatment on overall outcome is difficult from the studies available to date.11 The use of corticosteroids in all grades of severity of head injury is currently being studied in a large randomised international trial.12 To date over 3000 patients have been recruited, and recruitment of 20 000 patients by 2005 is planned. A recent Cochrane review of therapeutic hypothermia for brain injury concluded that there is no evidence that hypothermia is beneficial.13 Despite these continuing uncertainties over the benefits of individual treatments, there is evidence of an overall improvement in head injury outcome from treatment in a specialist unit that uses protocol driven treatment.10

    Rehabilitation of patients with head injuries is considered to be essential for a good outcome. Analysis of efficacy is hampered by methodological problems such as widely varying outcome measures and a lack of randomised trials. Despite this, reviews of rehabilitation of patients who have brain injury due to a variety of causes such as stroke, subarachnoid haemorrhage, and trauma have shown that rehabilitation is effective in reducing long term functional disability and improving overall outcome.14

    Developing novel neuroprotective drugs has proved challenging and simultaneously frustrating over the past decade. At the cellular level, injury leads to disruption of the neuronal cytoskeleton. This leads ultimately to irreversible division of the axon over a 12 hour period. Microdialysis studies in humans have detected very high concentrations of extracellular glutamate after brain injury. This injures neighbouring cells and leads to a cascade of cell death and progressive release of excitotoxic molecules. Both of these mechanisms have raised the possibility of a therapeutic window in which therapeutic agents might have maximal effect. Clinical trials of potential neuroprotective agents such as glutamate and calcium antagonists have not so far shown efficacy.15

    Current management of head injury should follow national guidelines tailored to local circumstances. Improved resuscitation and triage, safe and rapid transfer of patients, and the expansion of specialist neurocritical care units all contribute to an improved outcome. The use of large trials of current treatments and the continuing effort to develop new therapeutic agents holds the promise of further improving the outcome for the patient with head injury.


    • Competing interests JW is a principal investigator of the Medical Research Council funded CRASH (corticosteroid randomisation after significant head injury) trial


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    View Abstract