The protocol
 
Patients with head injury requiring ventilation and intra-cranial pressure monitoring will be considered for the trial. Entry will be determined using the above inclusion and exclusion criteria and following consent for the next of kin. Patients will be managed on intensive care units using a standard protocol (figure 3). The major objective of this protocol is to maintain ICP<25 mmHg by applying treatment measures in a number of stages.
 
Stage 1 - initial treatment measures:
Patients will be sedated, analgesed and ventilated. Patients may or may not be paralysed but this must be noted. They will be nursed head up with no venous obstruction. Invasive monitoring (central venous pressure and arterial lines as a minimum will be applied). Targets for physiological parameters will be:
   Cerebral perfusion pressure > 60 mmHg (central venous pressure 6-10),
   Oxygen saturation >97%,
   Arterial CO2 = 4.5-5.0 kPa,
   Temperature <37ºC,
   Blood sugar 4-7 mmol/l.
 
The ICP will be assessed at this stage. If the ICP<20mmHg, the above medical treatment will continue. If the ICP>20mmHg, a repeat scan will be considered to investigate the presence of an evolving mass lesion and stage 2 will be applied.
 
Stage 2 - advanced treatment measures:
In stage 2 the following measures can be considered, all of which are optional:

• An external ventricular drain - depending on the size of the lateral ventricles
• Mannitol
• Inotropes to increase the mean arterial pressure to maintain a cerebral perfusion pressure of >60 mmHg.
• Arterial carbon dioxide 4.0 to 4.5kPa (can be monitored with jugular venous oxygen saturation sensors maintaining SjvO2 >55%)
• Hypertonic saline
• Moderate cooling (35-36°C) but not severe hypothermia <34°C
• Loop diuretics

Barbiturates are not implemented as part of stage 2, but are reserved as part of continued medical treatment following randomisation. This clause enables a direct comparison between the efficacy of decompressive craniectomy and extended medical treatment including the introduction of barbiturate coma.
 
If despite stage 1 and 2 measures, the ICP remains above 25mmHg for 1 to 12 hours then patients will be randomised to either continued medical treatment stage 2 but also including barbiturates (e.g. thiopentone boluses + infusion 4-8 mg/kg/hr) or to surgical treatment (decompressive craniectomy). Treatment following randomisation should be implemented within 4-6 hours.
 
The surgical treatment will comprise:
 
(a) for unilateral hemisphere swelling / a large unilateral fronto-temporo-parietal craniectomy
or
(b) for bilateral diffuse hemisphere swelling a large bilateral fronto-temporo-parietal craniectomy from the frontal sinus anteriorly to the coronal suture posteriorly and pterion laterally with a wide dural opening (pedicles based on the  superior sagittal sinus medially and division of the falx anteriorly).
 
If continued medical treatment is drawn no decompressive surgery will be performed at the time of randomisation, but decompressive surgery may be performed later at the clinician's discretion if the patient subsequently deteriorates (for example prolonged and unacceptably high ICP >40mm Hg with compromised CPP). This clause is required if a situation arises whereby the treating physician feels that withholding surgery is acting against the best interest of the individual - "the interests of the patient always prevails over those of science and society". The same principle applies to barbiturates in the decompressive craniectomy group.
 
Imaging:
 
CT scans will be read centrally and include the following:
   (a) Presentation CT scan
   (b) Pre-randomisation CT scan after stage 1 / 2 to check no evolving mass         lesion
   (c) CT scan 72 hours post-randomisation
   
 
 
 
Power
The total number of patients will be 400 (200 in each arm of the study) for a 15% difference in outcome (increase in favourable outcome from 45% to 60%) (power=80%, p=0.05).
 
 
Outcome measures
 
1. The primary endpoint will be assessment of outcome at discharge (Glasgow Outcome Score) and 6 months (Extended Glasgow Outcome Score).22

2.  Secondary endpoints will be:
           a.        Assessment of outcome using the SF-3623 and SF-10 ( for children
                      below 16 years of age) questionnaires
           b.        Assessment of ICP control
           c.        Time in intensive care
           d.        Time to discharge from the neurosurgical unit.
           e.         Detailed health-economic analysis

3. Additional assessment of outcome using the above measures will be performed at 1 year and 2 years after the injury.

For patients undergoing decompressive craniectomy, it is recommended that the bone flap is replaced within six months of the initial injury.
 
 
 
Acknowledgements
 
This protocol has evolved as a result of discussions between representatives of the University of Cambridge and the European Brain Injury Consortium including the participants of the intra-dural lesion study.
 
 
 
 
 
Table 1. Results of East Anglian Regional Audit of Head Injury.
 

Head   injury	Periods	Favourable	Severe disability/ PVS	Dead
		% *<0.05
All	1991 - 1993	56	21	23
	1994 - 1997	66	13	20
Severe	1991 - 1993	40*	32	28
	1994 - 1997	60*	21	22

 
 
 

Randomised Evaluation of Surgery with Craniectomy for
Uncontrollable Elevation of Intra-Cranial Pressure