As a care of the elderly registrar currently working in a district general hospital, I am very familiar with the following story, and it is likely to be recognised by others who take part in the on-call medical take, as well as those who work in accident and emergency, neurosurgery, and palliative medicine.

It is midnight and the stroke thrombolysis bleep goes off. You get there as quickly as possible and meet an 83 year old man, previously independent, who has been brought to the emergency department with a Glasgow coma scale (GCS) score of 9/15 and dense right sided weakness. He has had a CT scan that shows a large intracerebral haemorrhage: there is extensive bleeding into the left cerebral hemisphere, with intraventricular extension.

The A&E senior house officer has already phoned the neurosurgical registrar in the tertiary referral centre “just to check that nothing needs to be done.” Junior doctors often know before they make the phone call what the outcome will be. It is a catastrophic bleed, but they feel they need to consult with a neurosurgical team to confirm the decision for conservative management. The assessing doctor diligently writes in the notes the neurosurgical message: “For conservative management but continue neuro observations and repeat CT scan if deteriorates further.”

The patient deteriorates over the ensuing hours: his GCS drops to 4/15 and plans are made to return him to the CT scanner. His airway is becoming unsafe and the anaesthetic registrar is summoned. She decides the patient should not be intubated for the scan because if the bleed has worsened, the plan is for conservative management in any case. In the corridor on the way to the CT scanner, the patient’s breathing gets worse and he is taken back to the emergency department where he dies in a cubicle.

Our collective inability to acknowledge when patients have a very poor prognosis means that those who are frankly dying are subjected to frequent neurological observations and repeat CT scans, rather than doctors having tactful discussions with families and providing appropriate end of life care. It appears often easier to carry on with investigations and “treatment” rather than recognise that a patient is dying.

The number of phone calls the on-call neurosurgical registrar receives from referring hospitals is onerous. We wrote to the team at our own tertiary referral hospital and they agreed that inappropriate referrals have a substantial impact on the capacity of the neurosurgical unit to provide a safe and effective emergency service.

Given the greater availability of timely emergency radiology reporting and increasing consultant supervision (hence junior doctors working with more support and so treatable neurovascular pathologies are less likely to be missed), now is the time to draw up some guidelines for neurosurgical referral. This would limit inappropriate phone calls, allowing neurosurgeons to respond promptly to patients needing emergency intervention, while encouraging local teams to provide appropriate end of life care to those who will not benefit from neurosurgical expertise.

When I next review a similar patient, I need to think: are repeated neurological observations and repeat CT head scans really the best course of action, or should the patient be receiving appropriate and supportive end of life care?

Note: The patient mentioned in this blog is illustrative of a typical patient and not based on a specific case.

Helen Wood is a geriatric and general medicine registrar currently working in the Severn Deanery. 

Competing interests: I have read and understood BMJ policy on declaration of interests and declare the following interests: None.