Refractory intracranial hypertension after traumatic brain injury
Mr R is a 34-year-old man admitted after a motorcycle collision. CT shows a right frontal contusion and a thin acute subdural haematoma, with midline shift 4 mm. He is intubated and sedated; GCS is 3T. An intraparenchymal ICP monitor is in situ. Current values: ICP 25 mmHg, MAP 78 mmHg, CPP 53 mmHg, PaCO2 33 mmHg, PaO2 95 mmHg, sodium 139 mmol/L, haemoglobin 92 g/L, platelets 88 × 10^9/L, temperature 38.1°C. He is receiving propofol 4 mg/kg/h, fentanyl, and noradrenaline 0.15 µg/kg/min. He has received 1 g/kg of 20% mannitol 2 hours ago. There is no active bleeding on repeat CT 6 hours ago. The neurosurgeon is reluctant to operate because the contusion is not dominant.
1. Calculate CPP and outline a tiered approach to his raised ICP, prioritising the next three interventions and justifying them.
Read the worked answer
CPP = MAP − ICP = 78 − 25 = 53 mmHg, below the 60–70 mmHg target. This is the priority.
Tiered approach:
- Check and fix the basics: ensure the ICP monitor is zeroed and the waveform is real; confirm head-up 30°, neutral neck, no cuff ties obstructing venous drainage; check PaCO2 (33 is acceptable; do not hyperventilate below 30 except as a bridge), temperature, seizures, and bladder or bowel distension.
- Osmotherapy: he had mannitol 2 h ago. Check sodium and osmolality; if sodium is 139, hypertonic saline (e.g. 3% bolus) may be preferable to repeated mannitol because it is less diuretic and can be titrated to a sodium target of 145–155. Do not give both simultaneously without monitoring.
- Deepen sedation/analgesia and consider a propofol bolus; if ICP remains >22 mmHg for >5 min, escalate to a short trial of hyperventilation (PaCO2 30–35 mmHg) as a bridge to definitive treatment, and consider neuromuscular blockade.
- If refractory, consider decompressive craniectomy or barbiturate coma; the neurosurgeon's reluctance is reasonable for a non-dominant contusion, but a rising ICP with clinical deterioration changes the balance.
The order matters: correct reversible factors before escalating to rescue therapies.
2. Comment on the propofol dose, temperature, haemoglobin and platelets, and how each modifies your management.
Read the worked answer
Propofol 4 mg/kg/h is high; prolonged infusion at this rate risks propofol-related infusion syndrome (PRIS) - metabolic acidosis, rhabdomyolysis, arrhythmia, renal failure. Check creatine kinase, lactate, triglycerides and acid–base status, and consider switching to an alternative sedative or reducing the dose. Sedation should be the minimum needed to control ICP.
Temperature 38.1°C: treat fever aggressively with paracetamol and cooling; fever raises cerebral metabolic demand and ICP. Look for VAP, line infection and drug fever.
Haemoglobin 92 g/L: in TBI, a restrictive threshold (7–9 g/L) is generally safe, but at 92 g/L there is no strong indication for transfusion; consider it if there is evidence of cerebral hypoxia or ongoing bleeding.
Platelets 88 × 10^9/L: acceptable for monitoring, but check trends and coagulation before any neurosurgery; correct if <50 × 10^9/L before an invasive procedure or if bleeding.
Each of these is a secondary insult that worsens outcome; the principle is to prevent episodes of hypotension, hypoxia, fever, anaemia and hyperglycaemia.
3. The family asks about prognosis and decompressive craniectomy. How do you approach this conversation and what evidence informs the decision?
Read the worked answer
Prognosis in severe TBI is uncertain, especially early. The conversation should be structured:
- Prepare: review the CT, ICP trends, sedation and confounders; involve the family's cultural and religious context and identify the main contact.
- Explain what is known and what is not: the injury is severe, but early prognostic scores are unreliable and recovery can continue for months. Avoid definitive statements in the first 72 h.
- Discuss decompressive craniectomy honestly: in the DECRA and RESCUEicp trials, surgery reduces ICP and may reduce mortality but increases the proportion of survivors with severe disability. It is a reasonable option for refractory ICP in a young patient, but it is not a guarantee of good outcome.
- Use shared decision-making: elicit the patient's values from the family, explain the burdens of prolonged ICU care, and agree on a time-limited trial with clear reassessment points.
- Document the discussion and the agreed plan.
The judgement call is to avoid both premature nihilism and unrealistic promises, and to keep the family involved without transferring the entire decision to them.