• Accident & Emergency
• Health Economics
• Epidemiology
• Neurosurgery

The spectre of the child who ‘talked, yet died’ (whose head injury was not severe enough to cause loss of consciousness, yet who subsequently succumbed to a complication, such as an expanding extradural haematoma) haunts all who assess the many thousands of children attending emergency departments (ED) annually after apparently minor head injury. Worse yet, perhaps, is the child who survives with devastating disability.

Over the last decade, there has been a major change in strategy in many countries away from admission for a night or two of nursing observation, toward imaging-based triage1; with a parallel move from plain skull x-ray to CT as the imaging modality of choice as availability made this possible. Children with normal CT imaging can be confidently discharged; and the small proportion of children with CT abnormalities can be observed in regional centres with appropriate neurological nursing expertise, with a vigilance appropriate to the knowledge that the child has a CT abnormality that may require urgent neurosurgical intervention.

Any screening strategy involves the optimisation of the benefits of prompt recognition and treatment of a haematoma, and the avoidance of large numbers of unnecessary admissions; versus the disbenefits of performing large numbers of CTs in children, the very large majority of which will be normal. This optimisation is complicated however by an asymmetry in the immediacy and tangibility of the various possible outcomes. The impact of a missed extradural is very immediate; an iatrogenic brain malignancy will occur much …