Any guideline that gives priority to the prompt identification of patients at risk of brain injury deserves to be supported and timely imaging (CT brain scan or skull radiograph, or both) is an essential part of this process. Recently published guidelines recognise the increased availability of emergency CT in the UK.^1–3 The Scottish Intercollegiate Guidelines Network Publication number 46 (August 2000) recommends the increased use of CT brain scans for selected patients with mild head injury including those with the radiological demonstration of a skull fracture.²

The National Institute for Clinical Excellence: Clinical Guideline (2003)³ includes an adapted form of The Canadian CT head rules,⁴ which lowers the threshold for scanning patients with mild head injury much further and relies very little on the use of skull radiographs.

A concern is whether such guidelines can be safely and effectively implemented if hospitals in the UK lack the necessary resources for easy access to out of hours CT.

The Canadian CT head rules were based on the study of patients who had a history of loss of consciousness or post-traumatic amnesia after blunt head injury. In Steill’s paper, the population studied from the 10 Canadian hospitals appears to be less violent than that of some major inner city A&E departments in the UK. Most of the injuries were attributable to falls or road traffic accidents. Of the 11% attributable to assault most were by use of hands or feet rather than blunt objects and only 8% of assaults suffered significant “brain injury” as evident on CT scans.

In the UK attempts to establish the probable impact of the NICE guideline are underway and not surprisingly indicate that there would be a significant increase in the CT scan rate if the guideline was adhered to. It remains to be seen how well individual hospitals can cope with the increased demand for CT scans.

NICE recommend immediate request for CT of patients who still have depressed GCS (14/15) two hours after injury. This should have support from those responsible for the observation of such patients on wards if it detects those patients needing referral to a neurosurgical unit while they are still in the A&E department. Clinical variables such as coagulopathy and vomiting (twice) are rightly included as warranting a low threshold for CT but the emergency scanning of all patients over 64 years of age with any amnesia may prove difficult to implement.

Most head injured patients are GCS 15 on arrival in A&E, with little or no post-traumatic amnesia or primary brain damage. If they have a skull fracture their risk of needing an operation for intracranial haematoma is increased several hundred times.^5,6 In most Scottish A&E departments skull radiographs are still used as a screening method to detect skull fractures in mild head injury (GCS13–15). Nevertheless, in the past few years since the development of the SIGN guideline there has been a significant increase in the utilisation of CT brain scans, some instead of and some after skull radiography.

In accordance with the SIGN guideline, we believe there is still a role for skull radiological examination, for example, in patients with a significant mechanism of injury who may have sustained a depressed skull fracture attributable to blunt or penetrating trauma. We are concerned that, if NICE guidelines are followed, A&E doctors may not request CT scans until such patients deteriorate from intracranial haematoma or brain injury.

Ultimately, the most contentious aspect of the NICE guidelines is their advising an immediate CT scan in any patient who has been GCS 12 or less at any point since injury. This would include many patients who in retrospect have simply been intoxicated or have other reversible causes of depressed consciousness and may result in a large proportion of unnecessary scans. Furthermore, from a logistical viewpoint, if it were agreed that all such patients did require a CT scan, most radiology services would be unable to cope with the workload. It is ironic that The London Royal College of Radiologists recommends the Canadian CT head rules and yet their members are often not in a position to comply with their implementation!⁷

Some lowering of the threshold for early CT is welcome but not to the degree or in the form required by NICE. While the content of the SIGN guideline is not perfect it is increasingly accepted by radiologists and clinicians and we would recommend it as a more realistic alternative to NICE. If further improvement in the head injury service is to be achieved, the major challenge for A&E is implementation of realistic guidelines. This requires education of medical and nursing staff with an emphasis on early selection for imaging, frequent charting of observations, and improved documentation.