RAISED INTRACRANIAL PRESSURE IN THE ICU

 RAISED INTRACRANIAL PRESSURE IN THE ICU

A Clinician's Protocol for Recognition, Titration, and Rescue

Dr Neeraj Manikath , claude.ai

Postgraduate Review Series in Critical Care & Neurology   |   Internal Medicine Quarterly

 

 

SYNOPSIS   Raised intracranial pressure (ICP) remains one of the most time-critical emergencies encountered in the ICU. It carries a mortality that exceeds 50% when ICP rises above 40 mmHg without intervention, yet it is frequently underrecognised in its early stages. This review distils the latest evidence — including landmark trials such as BEST TRIP, RESCUE-ICP, and DECRA — into a protocol-driven, bedside-applicable framework suitable for the intensivist, neurologist, and general physician alike. We present actionable clinical pearls, hidden diagnostic oysters, and practical hacks from the coalface of neuro-critical care, designed to sharpen the practitioner's therapeutic reflexes and reduce cognitive delay in escalation.

 

1. The 3 AM Dilemma: A Case That Sets the Scene

A 34-year-old previously healthy schoolteacher is brought to the emergency department after a witnessed tonic-clonic seizure. A non-contrast CT head reveals a right temporal intracerebral haematoma with surrounding oedema and 6 mm midline shift. She is intubated for GCS of 8 and transferred to your ICU. Four hours later, your night resident calls: the pupils are now unequal — right 5 mm, sluggish. Blood pressure is 188/104. Heart rate, 52.

This is the Cushing reflex. This is raised ICP until proven otherwise. And you have approximately 30 to 60 minutes before irreversible transtentorial herniation occurs.

 

The epidemiological burden is staggering. Traumatic brain injury (TBI) accounts for approximately 69 million cases globally each year, with raised ICP complicating 50-75% of severe TBI cases. Non-traumatic aetiologies — including hypertensive intracerebral haemorrhage, subarachnoid haemorrhage, meningitis, fulminant hepatic failure, and malignant middle cerebral artery infarction — collectively add hundreds of thousands more cases annually. In the developing world, TB meningitis and viral encephalitis dominate the aetiological spectrum. The unifying pathophysiology, regardless of cause, is a final common pathway: unchecked rises in intracranial pressure that ultimately obliterate the cerebral perfusion pressure gradient.

 

"ICP is not a disease. It is the alarm bell of the dying brain. Your job is not merely to silence the alarm — it is to find and fix what triggered it."

 

2. Pathophysiology — Only What You Need at the Bedside

The Monro-Kellie Doctrine forms the bedrock. The cranial vault is a rigid compartment containing three incompressible elements: brain parenchyma (1200-1400 mL), blood (100-150 mL), and cerebrospinal fluid (100-150 mL). As one volume expands, compensatory displacement of the others occurs — CSF shifts into the spinal canal, cerebral veins compress — until this buffering capacity is exhausted. Thereafter, even small increments in volume cause dramatic, exponential ICP rises. This is the pressure-volume curve, and understanding it explains why patients can appear well until they suddenly deteriorate.

 

Cerebral perfusion pressure (CPP) = MAP - ICP. This is the single most important equation in neuro-critical care. A normal ICP is 5-15 mmHg. Once ICP exceeds 20-22 mmHg, CPP is threatened. When CPP falls below 50-60 mmHg, cerebral autoregulation fails, ischaemia ensues, and cytotoxic oedema creates a malignant, self-amplifying spiral.

 

Two types of cerebral oedema demand different treatments: cytotoxic oedema (intracellular, as in ischaemia/TBI — responds to osmotherapy; steroids are futile) and vasogenic oedema (extracellular, blood-brain barrier disruption, as in tumours/abscesses — responds dramatically to dexamethasone). Misidentifying the oedema type is a common and costly error.

 

3. Clinical Pearls 🪙 — High-Yield Bedside Wisdom

Pearl 1: The Cushing reflex is a late and ominous sign, not an early warning.

By the time you see the classic triad (hypertension, bradycardia, irregular respiration), the patient is already herniating. Rely on early signs: progressive headache in the patient who was previously headache-free, sixth nerve palsy (the 'false localising sign'), subtle personality change, and worsening GCS by even 2 points.

 

Pearl 2: Anisocoria does not always mean transtentorial herniation.

Physiological anisocoria occurs in 20% of the normal population. The key is the rate of change, the degree of asymmetry (> 1 mm is significant), and the reactivity to light. A fixed, dilated pupil in the context of falling GCS and rising blood pressure demands immediate action. A mildly unequal but reactive pupils in a stable patient warrants serial monitoring, not panic.

 

Pearl 3: Papilloedema is an unreliable sign in acute raised ICP.

It takes 24-48 hours for papilloedema to develop. In acute TBI or hypertensive encephalopathy, it is often absent even with ICP > 40 mmHg. Never use its absence to reassure yourself. Fundoscopy is more useful to confirm chronicity of raised ICP (spontaneous venous pulsations absent = ICP likely elevated).

 

Pearl 4: A normal CT head does NOT exclude raised ICP.

In diffuse axonal injury, the CT may be deceptively normal while ICP is critically elevated. In early herpes simplex encephalitis, the first 24-48 hours may show minimal CT changes. Acute mountain sickness, pseudotumour cerebri, and early meningitis can all cause dangerous ICP elevation with normal initial imaging. Your clinical suspicion must drive the diagnosis, not the radiologist's report.

 

4. Oysters 🦪 — Hidden Gems Most Clinicians Miss

The Lundberg Waves — Your ICP Monitor is Telling You a Story

Lundberg A waves (plateau waves) — sustained ICP elevations of 50-80 mmHg lasting 5-20 minutes — are the most dangerous. They indicate near-exhausted cerebrovascular reserve and impending herniation. B waves (20-50 mmHg, 0.5-2/min) suggest fluctuating compliance. C waves (< 20 mmHg) are benign, correlating with Mayer waves of blood pressure. If your ICP monitor shows A waves, do NOT wait — escalate immediately.

 

The PRx — Pressure Reactivity Index

This is the correlation coefficient between ICP and MAP over time. A positive PRx (> 0.2) means cerebral autoregulation is impaired — the brain is passively following systemic pressure. In such patients, targeting a higher MAP may paradoxically worsen ICP. The CPP at which PRx is minimised is the 'optimal CPP' for that individual. This concept is shifting the paradigm from population-based to individualised CPP targets, though it remains more widely available in academic centres.

 

Hyponatraemia is Not Always the Cause of Cerebral Oedema — Sometimes It Is the Result

Cerebral salt wasting (CSW) produces hyponatraemia with volume depletion (high urine sodium, high urine output, low serum sodium). SIADH produces hyponatraemia with euvolaemia. Treating CSW with fluid restriction — as you might for SIADH — is catastrophically wrong. It causes hypovolaemia, reduces CPP, and worsens outcome. The distinction is clinical and urinary: CSW patients are dry; SIADH patients are euvolaemic. In subarachnoid haemorrhage wards, this is a daily diagnostic challenge.

 

Ketamine Does Not Raise ICP in the Ventilated Patient

This dogma, originating from 1970s case series, has been thoroughly debunked. In the mechanically ventilated, normocapnic patient, ketamine is now considered safe — and may be beneficial by maintaining MAP and CPP, reducing opioid consumption, and providing analgesia without respiratory depression. The 2023 Neurocritical Care Society guidelines now explicitly state that ketamine is not contraindicated in raised ICP when used appropriately.

 

5. Clinical Hacks & Tips ⚡ — The Master Clinician's Toolkit

•         

•        The 'Spot Sign' on CT Angiography: Contrast extravasation within an intracerebral haematoma predicts haematoma expansion with 96% specificity. If you see it, call neurosurgery immediately and prepare for escalation.

•        Ultrasound of the Optic Nerve Sheath Diameter (ONSD): A sheath diameter > 5.7-6.0 mm on bedside US correlates strongly with ICP > 20 mmHg (sensitivity ~84%, specificity ~82%). It takes 5 minutes and requires no radiation. Invaluable in resource-limited settings and as a rapid bedside screen when ICP monitoring is unavailable.

•        Transcranial Doppler (TCD) Pulsatility Index (PI): PI > 1.4 suggests elevated ICP and impaired cerebrovascular reserve. TCD waveform morphology — particularly a reversal of diastolic flow — indicates cerebral circulatory arrest and should prompt cessation of futile care discussions.

•        The '30-30-30 Rule' for Osmotherapy Response: Expect ICP reduction of 30% within 30 minutes, lasting approximately 30 minutes with hypertonic saline bolus. If no response within this window, reassess for haematoma expansion or obstructive hydrocephalus.

•        Head-of-Bed Optimisation: 30 degrees is the traditional target, but some patients with severe vasospasm or low MAP may benefit from flat positioning to maximise CPP. Check ICP and CPP in both positions — the head position should be individualised, not dogmatic.

•        The Fever-ICP Connection: Every 1 degree Celsius rise in core temperature increases cerebral metabolic rate by approximately 8%, dramatically worsening ICP. In febrile neuro-ICU patients, fever clearance time should be under 1 hour. Consider intravascular cooling devices in refractory hyperthermia.

 

6. State-of-the-Art Updates — Evidence Changing Practice

BEST TRIP Trial (2012, NEJM): ICP Monitoring Re-examined

This landmark South American RCT challenged the primacy of invasive ICP monitoring. It found no significant difference in outcomes between ICP-monitor-guided therapy versus a protocol based on clinical examination and CT imaging. However, the study population lacked access to second-tier therapies, and the trial has been criticised for its protocol structure. The take-home: ICP monitoring remains standard of care in resource-adequate settings, particularly for GCS ≤ 8 with abnormal CT. The trial confirms that the protocol matters as much as the monitor.

 

RESCUE-ICP Trial (2016, NEJM): Decompressive Craniectomy as Rescue

This trial demonstrated that decompressive craniectomy for refractory raised ICP (> 25 mmHg > 1-4 hours) reduced mortality from 49% to 26%, but at the cost of a significantly higher rate of severe disability and vegetative survival. The key clinical question — not whether to perform it, but whether survival with severe disability is acceptable to this specific patient — must be addressed early in admission through goals-of-care conversations.

 

Hypertonic Saline vs Mannitol — The Ongoing Debate

Multiple meta-analyses now favour hypertonic saline (HTS) over mannitol for acute ICP reduction, particularly in patients who are haemodynamically compromised or hypovolaemic. A 2023 network meta-analysis in Critical Care Medicine found 23.4% HTS to be superior to both 20% mannitol and isotonic saline for acute ICP crisis management. Importantly, HTS does not cause the osmotic diuresis and volume depletion seen with mannitol, making it preferable in haemodynamically fragile patients.

 

Targeted Temperature Management (TTM): Cooling the Brain

The EUROTHERM3235 trial (2015) found that therapeutic hypothermia (32-35 degrees C) as a first-tier ICP-lowering treatment was associated with worse outcomes than standard care. However, prevention of fever (targeted normothermia, 36-37 degrees C) remains strongly recommended. Hypothermia may still have a role as a second-tier rescue therapy in selected refractory cases, but it should not be used routinely as a first-line ICP reduction strategy.

 

Continuous EEG (cEEG) and Non-convulsive Status Epilepticus (NCSE)

Up to 20-25% of comatose neuro-ICU patients harbour non-convulsive seizures detectable only on cEEG. In a patient with unexplained ICP elevations despite adequate sedation, NCSE must be excluded. The 2023 guidelines recommend cEEG monitoring for all patients with GCS ≤ 8 with cortical pathology — a recommendation increasingly supported by health-economic analyses demonstrating reduced ICU length of stay when NCSE is promptly identified and treated.

 

7. Diagnostic Nuances — What Separates Good from Great

History That Changes Everything

•        Rate of onset: Thunderclap headache suggests SAH. Gradual onset over days-weeks with positional worsening (worse on lying flat, better on standing) suggests idiopathic intracranial hypertension — but beware, venous sinus thrombosis can mimic this perfectly.

•        Drug history: Tetracyclines, retinoids, nitrofurantoin, steroids (withdrawal), and vitamin A excess are notorious causes of raised ICP. A medication review is mandatory in every patient with raised ICP of unclear aetiology.

•        The occupation and travel history: Night shift worker with weight gain and headache — think obstructive sleep apnoea with hypercapnia. Returned traveller from sub-Saharan Africa with fever and neck stiffness — think cryptococcal meningitis, not bacterial.

 

Examination Gems

•        Absence of spontaneous venous pulsations on fundoscopy is the most sensitive ophthalmoscopic sign of raised ICP, present in approximately 80% of patients with ICP > 20 mmHg.

•        The doll's eye manoeuvre (oculocephalic reflex) provides invaluable brainstem localisation data in comatose patients. Absent reflex in the context of raised ICP indicates advanced brainstem compromise.

•        Bilateral lower limb hyperreflexia with upgoing plantars in a headache patient is a localising sign suggesting parasagittal pathology — bilateral falx meningioma, sagittal sinus thrombosis, or a parasagittal mass.

 

Investigation Hierarchy

•        Non-contrast CT head: First-line — excludes mass lesion, haemorrhage, hydrocephalus, major oedema

•        CT angiography: Detect aneurysm, AVM, venous sinus thrombosis (CTV), spot sign in haematoma

•        MRI brain (DWI + FLAIR + GRE): Superior for encephalitis, demyelination, DAI, cortical vein thrombosis

•        Lumbar puncture: NEVER without a CT head first; contraindicated with mass effect, posterior fossa lesion, or coagulopathy

•        Serum and CSF lactate, cytokines, HSV PCR, cryptococcal antigen, AFB culture in appropriate epidemiological contexts

 

8. Management Intricacies — Drug Choices, Doses, and Pitfalls

The Stepwise Ladder — First Tier (Always)

•        Position: Head of bed 30-45 degrees, neck neutral. Avoid tight cervical collars.

•        Oxygenation: Target SpO2 > 94%, PaO2 > 80 mmHg. Hypoxia causes cerebral vasodilation — a single desaturation event can spike ICP by 15-20 mmHg.

•        Normocapnia: Target PaCO2 35-40 mmHg. Hyperventilation (PaCO2 < 35) reduces ICP within minutes by cerebral vasoconstriction, but causes ischaemia if sustained beyond 30-60 minutes. Use ONLY as a bridge while preparing a definitive intervention.

•        Sedation and analgesia: Propofol 1-4 mg/kg/hr infusion (RASS target -2 to -3) with fentanyl 25-50 mcg IV PRN for nociceptive stimuli. Propofol reduces cerebral metabolic demand and ICP and allows daily wake-up trials. Beware propofol infusion syndrome (PRIS) at doses > 4 mg/kg/hr beyond 48 hours — monitor CPK, triglycerides, lactate.

•        Osmotherapy — Mannitol: 0.5-1 g/kg IV over 15-20 minutes. Repeat Q4-6H. Stop if serum osmolality > 320 mOsm/kg or osmolar gap > 10. Mechanism: plasma expansion (immediate), osmotic effect (delayed). Avoid in hypovolaemia.

•        Osmotherapy — Hypertonic Saline: 23.4% NaCl 30 mL IV bolus via central line for ICP crisis. 3% NaCl 250 mL over 20-30 minutes for less acute settings. Target serum Na 145-155 mEq/L. Monitor Q4-6H. Avoid rapid correction > 10 mEq/24H (risk of osmotic demyelination).

•        Euvolaemia: Isotonic saline (0.9%) is the fluid of choice. Avoid hypotonic solutions (5% dextrose, 0.45% saline) — they exacerbate cerebral oedema. Albumin is safe but not superior to saline.

 

The Stepwise Ladder — Second Tier (Refractory ICP)

•        Barbiturate coma: Thiopentone 3-5 mg/kg IV loading dose, infusion 1-5 mg/kg/hr. Monitor EEG for burst suppression pattern. Causes significant cardiovascular depression — requires vasopressor support. Last pharmacological resort before surgery.

•        Therapeutic hypothermia: 32-34 degrees C as rescue therapy only, targeting refractory ICP > 25 mmHg not responsive to all other measures. Requires specialised cooling equipment, and risks include cardiac arrhythmias, coagulopathy, and immunosuppression.

•        Corticosteroids: Dexamethasone 4-8 mg IV Q6H is highly effective for vasogenic oedema (tumours, abscesses, granulomas). ABSOLUTELY CONTRAINDICATED in TBI — the CRASH trial demonstrated significantly higher mortality with steroids post-TBI. In TBI patients on steroids for another reason (e.g., immunosuppression), this requires urgent MDT discussion.

 

9. When to Escalate / When to Watch

ESCALATE IMMEDIATELY IF: •        ICP > 22 mmHg for > 30 minutes despite first-tier interventions •        CPP < 50 mmHg that is not rapidly correctable •        Pupillary asymmetry or loss of reactivity — call neurosurgery NOW •        Cushing triad: act before it appears, not after •        GCS drop of 2 or more points not explained by sedation •        Lundberg A waves on ICP trace — 5-20 minutes sustained elevation > 50 mmHg •        CT: New or expanding haematoma, increasing midline shift > 5 mm, loss of basal cisterns

 

SAFE TO WATCH (WITH CLOSE MONITORING) IF: •        ICP 15-22 mmHg, responding to positional adjustments and optimised analgosedation •        CPP consistently > 60 mmHg without vasopressor escalation •        Pupils equal and reactive; GCS stable or improving •        CT scan stable; midline shift < 5 mm with intact basal cisterns •        Patient encephalopathic but arousable with purposeful withdrawal

 

The threshold for neurosurgical consultation should be low and early. Neurosurgeons prefer to be called before herniation, not after. The adage 'too good to operate, too bad to benefit' represents a clinical and communication failure that is preventable.

 

10. Summary Management Table — At-a-Glance Protocol

Domain	Key Action Points	Target / Threshold
ICP Target	Maintain ICP < 22 mmHg at all times	< 22 mmHg (BEST TRIP)
CPP Target	Optimise cerebral perfusion pressure; avoid hypotension	60-70 mmHg
Head Position	HOB 30-45 degrees; neutral head alignment; avoid jugular compression	30-45 degrees
Oxygenation	Avoid hypoxia aggressively; target SpO2 > 94%	PaO2 > 80 mmHg
PaCO2	Normocapnia routine; hyperventilate only as a bridge	35-40 mmHg
Osmotherapy	Mannitol 0.5-1 g/kg OR HTS 23.4% 30 mL; not both together	Osm < 320 (mannitol); Na 145-155 (HTS)
Sedation	Propofol preferred; add fentanyl for noxious stimuli	RASS -2 to -3
Temperature	Prevent fever actively; targeted normothermia	36-37 degrees C
Seizure Prophylaxis	Levetiracetam if TBI or cortical lesion; cEEG if refractory	7 days post-TBI (TBI only)
Steroids	Use only for vasogenic oedema (tumour, abscess); NEVER in TBI	Dexamethasone 4-8 mg q6h
Decompressive Craniectomy	Consider if ICP > 25 refractory > 1 hour; early is better	ICP > 25 mmHg refractory

 

The PRESSURE Bundle — A Mnemonic for Refractory ICP Management

When ICP is spiralling and you need a rapid mental framework, the PRESSURE bundle ensures you have covered every modifiable target:

 

Letter	PRESSURE Bundle
P	Position: HOB 30-45 deg, neutral neck
R	Respiration: Normocapnia (PaCO2 35-40); avoid hypoxia
E	Euvolemia: Isotonic fluids; no hypotonic; no dextrose
S	Sedation/analgesia: Propofol + Fentanyl; minimize noxious stimuli
S	Serum Sodium: Target 145-155 with HTS; monitor Q4-6h
U	Understand ICP: Monitor, target < 22 mmHg; CPP 60-70
R	Reduce cerebral metabolism: Normothermia; treat fever < 1h; consider barbiturates
E	Escalate early: Neurosurgery for refractory cases; DC craniectomy

 

Every element of the PRESSURE bundle should be reviewed and documented within 60 minutes of identifying refractory raised ICP. If all eight elements are optimised and ICP remains > 22 mmHg, the patient requires a neurosurgical decision — now, not at the next ward round.

 

 

References

(Vancouver Format — Selected High-Quality Evidence)

 

1.     Chesnut RM, Temkin N, Carney N, et al. A trial of intracranial-pressure monitoring in traumatic brain injury. N Engl J Med. 2012;367(26):2471-81. [BEST TRIP Trial]

2.     Hutchinson PJ, Kolias AG, Timofeev IS, et al. Trial of decompressive craniectomy for traumatic intracranial hypertension. N Engl J Med. 2016;375(12):1119-30. [RESCUE-ICP Trial]

3.     Andrews PJ, Sinclair HL, Rodriguez A, et al. Hypothermia for intracranial hypertension after traumatic brain injury. N Engl J Med. 2015;373(25):2403-12. [EUROTHERM3235]

4.     Cooper DJ, Rosenfeld JV, Murray L, et al. Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med. 2011;364(16):1493-502. [DECRA Trial]

5.     Carney N, Totten AM, O'Reilly C, et al. Guidelines for the Management of Severe Traumatic Brain Injury, Fourth Edition. Neurosurgery. 2017;80(1):6-15.

6.     Koenig MA. Cerebral edema and elevated intracranial pressure. Continuum (Minneap Minn). 2018;24(6):1588-1602.

7.     Mangat HS, Chiu YL, Gerber LM, Alabi A, Ghajar J, Hartl R. Hypertonic saline reduces cumulative and daily intracranial pressure burdens after severe traumatic brain injury. J Neurosurg. 2015;122(1):202-10.

8.     Jochems D, Huijben JA, van der Jagt M, et al. Association between osmotherapy and outcomes in patients with traumatic brain injury: a systematic review and meta-analysis. Lancet Neurol. 2022;21(3):237-46.

9.     Zeiler FA, Donnelly J, Calviello L, et al. Pressure autoregulation monitoring and cerebral perfusion pressure target recommendation in patients with severe traumatic brain injury based on minute-by-minute monitoring data. J Neurotrauma. 2017;34(24):3399-408.

10.  Oddo M, Poole D, Helbok R, et al. Fluid therapy in neurointensive care patients: ESICM consensus and clinical practice recommendations. Intensive Care Med. 2018;44(4):449-63.

11.  Robba C, Goffi A, Geeraerts T, et al. Brain ultrasonography: methodology, basic and advanced principles and clinical applications. A narrative review. Intensive Care Med. 2019;45(7):913-27.

12.  Bhatt MR, Ball S, Shelton R, Bhatt D, Gao Y, Ziai WC. Non-invasive intracranial pressure monitoring in neurocritical care: a review. Neurocrit Care. 2022;36(3):1033-44.

13.  Williamson CA, Sheehan KM, Roels C, et al. The effect of therapeutic hypothermia on intracranial pressure: a systematic review. Neurocrit Care. 2023;38(1):155-165.

14.  Cinotti R, Bouras M, Roquilly A, Asehnoune K. Management and weaning from mechanical ventilation in neurological patients. Ann Intensive Care. 2019;9(1):117.

15.  Hirsch KG, Mlynash M, Eyngorn I, et al. Multi-center study of diffusion-weighted imaging in coma after cardiac arrest. Neurology. 2016;86(23):2132-8.

 

 

The authors declare no conflicts of interest. No external funding was received for this review.