Practical Intensive Care Medicine
(Problem Solving in the ICU)
You have admitted a 23-year-old man with a head injury to the Intensive Care Unit. He has had a CT scan which shows some frontal contusion and diffuse cerebral oedema. He has a Glasgow Coma Score of 8 and is intubated.

The surgical registrar loudly asserts that evidence does notsupport the routine use of hyperventilation in head injury. A medical student attached to the ICU asks about this, as he has previously heard that both mannitol or hyperventilation are effective.
What will you tell him?

The registrar is right. Routine hyperventilation is contraindicated and it is recommended that a PaCO2 of < 35 mmHg (4.66 kPa) is avoided during the first 24 hours after severe head injury.


Raised intracranial pressure can be rapidly reduced by acute hyperventilation, and this treatment been used routinely over the last 20 years. It is clear that cerebral blood flow following severe head injury is less than half of that in normal individuals, and that this falls further in the face of hyperventilation, with an increased risk of causing cerebral ischaemia.
Recent clinical studies have failed to show benefit, and on the contrary have been associated with adverse outcomes. Brief periods of hyperventilation may be necessary for acute neurological deterioration refractory to other treatment, but close monitoring for cerebral hypoxia is recommended when hyperventilation is employed.
    The use of hyperventilation in the management of severe traumatic brain injury.
    J. Neurotrauma 1996; 13: 699–703.

A 25-year-old man is brought in to hospital after falling from the top of a hay shed. He is conscious but complains that his chest is painful and he is having difficulty breathing. He obviously has a fracture of his left femur and both tibias. An iv is in situ and he has been given 500 ml Haemaccel and 500 ml N/saline on his way to hospital. He has an oxygen mask on at 15 l/min flow.

On examination he has a flail segment right chest with reduced breath sounds on the right. His heart rate is 140/min, BP 150/80, and respiratory rate 42/min. The pulse oximetry is not reading well, but intermittently there is a reading of 80% SpO2 .

 Arrangements have been made for him to be admitted toanother hospital (because your hospital is on by-pass). The ambulance is ready to take the patient and the ambulance officers are keen to do so.
Do you feel he is fit to transfer?

No, he is certainly not fit to transfer. This is an important issue, which is often delegated to inexperienced staff. Perceived urgency often prevents considered thought.
Readiness to transfer needs to be considered in basic terms – airway, breathing, circulation.
His airway is satisfactory.
His breathing is unstable. He is hyperventilating. This may be due to thoracic problems or he may be hyperventilating to compensate formetabolic acidosis. In either case he is not stable.
He is also hypoxic (SpO2 80%). This is probably due to thoracic injury which requires investigation. He may require intubation and ventilation. An intercostal catheter may be indicated.
His circulation is unstable. He is inadequately resuscitated. His pulse rate is too high. Although the blood pressure is normal, this is not the most sensitive indicator of hypovolaemia. Where practical, demonstration of a postural drop is suggestive of hypovolaemia.
Poor peripheral perfusion is a common cause of failure of the pulse oximeter to function reliably.
He has fractures which can predictably be associated with at least 4 unit blood loss.
Thirst, urine output and low JVP may be other useful indicators of hypovolaemia.
His fractures have not been stabilised. Stabilisation is likely to reduce further blood loss, reduce the chance of fat embolism syndrome and minimise pain during transfer.