The Basics

• ECG (risk of arrhythmias)
• Invasive arterial pressure (MAP, CPP)
• Invasive central venous pressure
• ETCO2 (correlate to PaCO2)
• Oxygen saturation

Clinical

• GCS
• Pupils
• Limb strength

Currently Accepted Indications for ICP Monitoring REF

Severe TBI (GCS 3-8 post resuscitation) with abnormal CTB

Or severe TBI with normal CTB if 2 or more of:

• Age > 40
• Uni or bilateral motor posturing
• SBP < 90 mmHg

Optional Multi-Modal Monitoring

• Brain tissue oxygen tension (PbtO2) monitoring
• Cerebral microdialysis
• Transcranial doppler
• cEEG

Fentanyl 20-50 mcg/hour + Propofol < 4 mg/kg/h

Analgesia

Pain and anxiety must be treated, for obvious reasons as well as to avoid EICP, hypertension and rebleeding.

For non-intubated patients analgesics like paracetamol, tapentadol and opiates usually suffice; NSAIDs may be useful but the slight effect on platelet function risking bleeding should be considered.

Intubated patients usually receive opiate (fentanyl) infusions titrated to pain e.g. 20-50 mcg/hour IV fentanyl

Very high doses of opiates do not reduce ICP beyond the effects of reducing pain induced EICP so are not recommended.

Ultra short acting opiates (remifentanil) offer the attraction of quick neuro assessability but rebound hyperalgesia may occur on abrupt cessation.

Ventilator dysynchrony may improve with titrated opiates.

Sedation

Propofol is usually used for sedation due to its short duration of action, permitting faster neuro assessability.

Propofol reduces ICP and is an initial treatment for EICP. Keep propofol < 4 mg/kg/h (less in older patients) to reduce risk of PRIS

In haemodynamically unstable patients midazolam may be used as it causes less hypotension; the main disadvantage is accumulation affecting neurological assessment.

If patients are agitated or posturing without EICP, consider use of clonidine, midazolam, olanzapine REF

Sedation in intubated patients usually require propofol infusions to tolerate the endotracheal tube.

Nausea should be aggressively treated as vomiting raises ICP

EICP

Manage as per algorithm

See the separate section on EICP

Target euvolaemia with isotonic IV fluids

> Normal saline is the default fluid

Avoid hypotonic fluids (e.g. 5% dextrose, Hartmann’s solution), which may worsen cerebral oedema

Avoid albumin REF

Avoid Plasma-Lyte 148 REF

SpO2 ≥ 92%, PaO2 > 60 mmHg

PaCO2 35–40 mmHg

As we know that hypoxia (sats < 90%, PaO2 < 60 mmHg) is definitely associated with poor outcome, we tend to aim slightly higher than this to reduce chances of hypoxia.

Temporary hyperventilation used in management of EICP but don’t allow PaCO2 < 30 mmHg except for very brief periods when managing EICP. This causes cerebral vasoconstriction and ischaemia / hypoxia which is obviously bad.

Avoid high PEEP (potentially affects CPP)

Avoid tight ETT ties that obstruct venous drainage (potentially affects CPP)

Correct any coagulopathy

Coagulopathy occurs in approximately 1/3 of patients with TBI

It is a risk factor for a worse outcome.

In acute TBI, check:

• Platelet count (FBC)
• INR
• PT
• APTT
• Fibrinogen

Obtain drug history, specifically for medications that affect coagulation.

Platelet transfusion for patients previously on anti-platelet agents is controversial and probably not indicated REF

Strongly consider viscoelastic tests (ROTEM or TEG) to guide treatment REF

One Approach (if viscoelastic tests not available)

> Platelet ≥ 100 x 10^9/L

> INR ≤ 1.4

> Fibrinogen > 2.0 g/L

Nurse patients head up (>30 degrees)

For

• There is some suggestion that having the head at 20-30 degrees improves ICP, so may have some role in patients with intracranial hypertension
• As described in the EICP section, sitting a patient up reduces venous congestion in the head, so promoting improved cerebral perfusion
• Weak evidence suggests this is associated with reduced risk of ventilator associated pneumonia (VAP)

Against

• Despite several small exploratory studies, no head position has been shown superior over another, in terms of maximising cerebral blood flow or more importantly improving patient outcomes
• Similarly, there is no strong evidence demonstrating superiority of head up positions in other aspects of neurocritical care

Although there is no convincing evidence for this practice, in the absence of high quality research, a head up position of 20-30 degrees is likely not to be harmful and has some potential to minimise VAP and may reduce ICP and improve CPP.

Start enteral feeding as soon as feasible as per usual unit protocol with dietician consultation.

Bowel sounds or passing flatus not required to start enteral feeding.

There’s not much good quality evidence to guide TBI specific nutrition.

Evidence Based Guidelines summarised here

Fever (temperature > 38.0 degrees Celsius) should be avoided

> Give regular paracetamol 1 g QID

> Consider NSAIDS e.g. ibuprofen 400 mg QID if evidence active bleeding has stopped (e.g. CT at 48 hours)

> Actively cool with fans and ice packs

> Use surface automated cooling device with temperature feedback

Prophylactic hypothermia and hypothermia for EICP are now NOT recommended

In Severe TBI use Levetiracetam 500 mg BD for 7 days then cease

Not required if on thiopentone infusion for elevated ICP

See Initial Management

Levetiracetam appears to be the best ASM option but is not without risks

Venous thromboembolism is common after TBI

Anticoagulation is not used when there is clinically significant intracranial haemorrhage

If there is imaging evidence of stabilisation or resolution of bleeding then anticoagulation may be considered e.g. heparin 5000 U BD.

This is a discussion between ICU and neurosurgical teams.

In the meantime use mechanical VTE prophylaxis with calf compressors and elastic stockings, even though there’s not much evidence that these make a difference

Screening lower limb dopplers are performed regularly (weekly if patient immobile).

If lower limb thrombosis is identified, the extent and ability to anticoagulate must be taken into account, with consideration of a inferior vena caval filter insertion if risk of PE is deemed high.

One Approach

> Mechanical VTE prophylaxis only initially

> Start heparin 5000 U SC BD once CT shows haematoma stable / resolving often around 48 hours

> If further CT’s remain stable, change to enoxaparin 40 U daily

> With hold anticoagulation when EVD’s are inserted or removed

> Weekly screening lower limb dopplers

> If below knee DVT found, start some heparin (amount depends on TBI factors) if not contraindicated (e.g. deemed high risk of more intracranial bleeding); Repeat doppler USS in 2-3 days and monitor DVT.

> If above knee DVT found, start some heparin (amount depends on TBI factors) if not contraindicated (e.g. deemed high risk of more intracranial bleeding), with low threshold for insertion of IVC filter

Pantoprazole 40 mg IV daily

Stress ulcer prophylaxis (SUP) with a PPI or H2RB is given to most ICU patients

SUP with pantoprazole reduces clinically important GI bleeding in ICU patients (see SUP-ICU trial)

H2RB may have benefits over PPI for SUP (see PEPTIC trial) but follow local guidelines for which agent to choose.

Starting enteral nutrition as soon as feasible probably reduces risk of stress ulcers

Aim BGL 6-10 mmol/L

No strong evidence exists for BSL control in TBI patients specifically.

Very tight BGL control increases risk of hypoglycaemia, a cause of secondary brain injury.

NICE-SUGAR trial.

Transfuse if Hb < 80 g/dL

The optimal haemoglobin target and threshold for transfusion in TBI is not known.

Pros

Improved brain tissue oxygenation REF

Potentially more cardiovascular stability

Cons

Doesn’t necessarily improve cerebral metabolism REF

More adverse events if threshold 10, including more thrmoboembolism REF

Worse long term outcomes if TBI pts receive blood transfusion REF

Blood transfusion may predispose to progression or development of intracerebral hematomas REF

An Approach

In severe TBI without multi-modal monitoring, transfuse if Hb < 80 g/dL

In severe TBI with multi-modal monitoring, transfuse guided by PbtO2

No Corticosteroids

Corticosteroids are not indicated unless there is evidence of adrenocortical insufficiency

No Prophylactic Hyperventilation

Used to be commonly practised. Reduces ICP but causes cerebral ischaemia. Now avoided unless for brief periods treating EICP.

No Prophylactic Hypothermia or Therapeutic Hypothermia for EICP

No evidence of benefit in TBI but does increase chance of ventilator associated pneumonia REF

Always check whether your patient is appropriate for enrolling in trials.

Clinical trial medicine is good medicine

Regular Meetings

It is essential to have regular meetings with families, to ensure they understand the current situation.

Arrange interpreters if required.

Intensive Care and Neurosurgical teams, nursing staff and social work should be available and present.

Establishing patient centred goals of treatment, and establishing the patient’s views on their predicted outcome are important in guiding treatment.

In some circumstances, if the expected neurological outcome is deemed to be unacceptable for the patient and there is medical consensus about prognosis, the focus of care may shift to palliative goals.

More about end of life care.

Elevated ICP is associated with bad outcomes

ICP monitoring allows intracranial compliance and CPP to be targeted.

It is typical for intracranial hypertension to rise over the first several days, plateau, and then slowly resolve.

See Fundamentals

Currently Accepted Indications for ICP Monitoring REF

Severe TBI (GCS 3-8 post resuscitation) with abnormal CTB

Or severe TBI with normal CTB if 2 or more of:

• Age > 40
• Uni or bilateral motor posturing
• SBP < 90 mmHg

Does ICP Monitoring Improve Patient Outcomes?

Not necessarily.

Like with many things we measure in ICU, knowing a number on it’s own means nothing, it’s what you do with it!

In a Dutch study mortality rates were similar when ICP guided therapy was compared with empirical management of TBI REF

Several other retrospective studies have come to the same conclusion REF1 REF2

The BEST-TRIP study compared outcomes in 2 groups of patients with severe TBI in Bolivia and Ecuador – one group managed with an intraparenchymal ICP monitor and the other with only imaging and clinical exam. There was no difference in functional outcome or mortality.

Whilst ICP monitoring continues to be used and recommended as best practice around the world, how we use the number and how it is integrated into other forms of multi-modal monitoring has become the more critical question…

Brain oxygenation can be assessed by the measurement of tissue oxygen tension (PbtO2) either electrochemically (Licox, Integra Lifesciences) or by the oxygen-dependent fluorescence quenching (Neurovent-PTO, Raumedic).

PbtO2 is a spatial average of oxygen partial pressure and is determined by the balance between oxygen supply and consumption, but also by diffusion, which may be impaired after TBI.

Why use PbtO2?

Brain tissue hypoxia is common after TBI

The hypoxia is prevalent several days after injury

Brain tissue hypoxia is associated with poor outcome

ICU interventions are effective in improving brain tissue oxygenation; these are:

• Increasing FiO2
• Increasing PEEP
• Increasing PaCO2 (causing cerebral vasodilatation)
• Blood transfusion to increase Hb and oxygen carrying capacity of the blood

The BOOST-2 Phase II study used PbtO2 monitoring with a treatment algorithm and showed it was safe and achievable.

The BOOST-3 and BONANZA phase III trials look to see whether this translates to improved clinical outcomes.

Risks of PbtO2 Monitoring

> Invasive, occasionally causing devastating bleeds from catheter insertion

> Currently no high level evidence that it improves outcomes

> The site of where the monitor is inserted is controversial – the values are different in undamaged brain compared to injured.

> The interventions to increase PbtO2 are controversial e.g. blood transfusion, high FiO2 for hyperoxia, hypercapnoea which are all usually avoided

> Must be used in conjunction with an ICP monitor so 2 invasive devices required

The BOOST-3 Algorithm

Used in non-penetrating severe TBI

ICP and PbtO2 monitors inserted at same time, early (6-12 h post admission)

Not done if pts have fixed dilated pupils, coagulopathic, septic, hypoxic, poor pre-morbid baseline.

PbtO2 monitor inserted into right frontal lobe (unless that’s injured, then into left)

REF

Note,  Type C Interventions include:

Tier 1 interventions

Tier 1 therapies must be started within 15 minutes of the start of the episode, as detected by the continuous PbtO2recordings. These are listed in no particular order and other options may be found in the MOP.

• Adjust head of the bed to improve brain oxygen level.
• Ensure Temperature < 38 degrees C.
• Increase CPP up to a maximum of 70 mm Hg with fluid boluses as clinically appropriate.
• Optimize hemodynamics. Obtain arterial blood gas to confirm that oxygenation is in desired range before treating with PaO2 adjustment. Increase PaO2 by increasing FiO2 to 60%..
• Increase PaO2 by adjusting positive end expiratory pressure (PEEP).
• Consider EEG monitoring.
• Consider anti-seizure medications (AEDs)

Tier 2 interventions

Providers may move to Tier 2 interventions at any point if PbtO2 is < 20 mm Hg and at least one intervention from Tier 1 has been used.  Providers must move on to Tier 2 interventions if PbtO2 < 20 mm Hg for > 60 minutes despite Tier 1 therapies. These are listed in no particular order and other options may be found in the MOP.

• Adjust ventilator parameters to increase PaO2 by increasing FiO2 to 100%.
• Increase PaO2 by adjusting PEEP.
• Increase CPP above 70 mm Hg with fluid boluses or vasopressors. Adjust ventilatory rate to increase PaCO2 to 45 – 50 mm Hg.
• Transfuse pRBCs to Hgb > 10 g/dL.
• Decrease ICP to < 15 mm Hg.
• CSF drainage
• Increased sedation

See more with the Trial Protocol if you’re interested!

Cerebral microdialysis is another form of invasive neuromonitoring, less established than PbtO2 monitoring.

A crystalloid perfusate is passed through a coaxial intraparenchymal catheter with a semipermeable membrane so that the effluent reflects brain chemistry.

Typically, hourly measurements are made using an automated bedside colorimetric assay.

The lactate:pyruvate ratio (LPR) and brain tissue glucose have received most application clinically.

LPR is an assay for anaerobic glucose utilization – a marker of ischaemia, shown to have prognostic significance.

Brain tissue glucose is sensitive to ischemia/hyperemia, and the balance between supply and hyper-/hypometabolism and low brain glucose concentrations may be an important driver of secondary injury.

May be used in conjunction with PbtO2 monitoring and ICP monitoring.

cEEG has two main uses following severe TBI:

1. To look for non-convulsive seizures (NCS)
2. To monitor for burst suppression in the context of a therapeutic thiopentone coma

Patients with TBI are at high risk of seizures, including in the early stages postinjury.

cEEG has demonstrated a high prevalence (up to one-third) of NCS, particularly in patients with penetrating injuries, depressed skull fractures, or large haematomas.

Persistent seizures are associated with poor outcomes

Seizure activity is undesirable because seizures may increases metabolic rate and oxygen demands and may increase ICP, therefore contributing to secondary brain injury.

Routine EEG (e.g. for 20-30 minutes) misses a proportion of NCS

cEEG is used in some centres to allow better detection and specific treatment of NCS.

No studies have shown patient-centred outcome benefits by using cEEG compared to spot EEGs

One recent prospective MC RCT that included (but was not only) patients with TBI, cEEG translated into a higher rate of seizures/status epilepticus detection and anti-seizure treatment modifications but did not improve mortality compared with rEEG.

To monitor for burst suppression in the context of thiopentone coma, a limited montage EEG is sufficient.

Refer to seizure module for more!

Pressure Reactivity Index (PRx)

PRx is calculated as the correlation coefficient between a 10-second moving window of MAP and ICP.

Autoregulation often fails after severe TBI, so that increases in MAP cause an increase in cerebral blood volume and therefore ICP, rather than the reflex vasoconstriction expected in health.

Thus, the correlation between MAP and ICP becomes positive and PRx>0 indicates impaired autoregulation’

PRx may be automatically calculated in real time by a computer and has clinical significance.

Positive PRx has been shown to be an independent predictor of mortality and PRx-derived patient-specific thresholds for CPP show a stronger association with outcome than average population values.

Over longer periods of time, PRx often shows a U-shaped relationship with CPP, reflecting the autoregulatory range.

This leads to the concept of the optimal CPP (CPPopt), which is the CPP for which PRx is lowest, and therefore for which autoregulation is best preserved.

In retrospective observational studies, patients managed below their CPPopt had a higher mortality, whereas those above had worse than expected disability

As with other forms of MMM, there is no strong evidence that targeting a PRx or CPPopt improves patient centred outcomes

MAP Challenge

If you don’t have access to a PRx device, one alternative to address the same concept is the MAP challenge

Essentially, in a patient with a severe brain injury and where cerebral autoregulation may not be working, you can try to reduce ICP by increasing the blood pressure. If increasing the MAP doesn’t reduce the ICP or makes it worse, abandon the trial and revert to the previous MAP target as autoregulation would appear to be defective. However if the ICP is reduced (by > 2 mmHg), you can assume autoregulation is preserved and continue this new higher MAP target.

The practical steps are:

1. Patient should be sedated and paralysed, PaCO₂ stable, CPP should be in current target range
2. Record baseline monitor parameters at the beginning of the challenge (ICP, MAP and CPP)
3. Initiate or titrate a vasopressor to increase the MAP by 10 mmHg for up to 20 minutes
4. Observe the interaction between the MAP, ICP, and CPP during the challenge
5. Record monitor parameters of MAP, ICP and CPP at the end of the challenge
6. Evaluate the observed responses and recorded values for evidence of pressure reactivity
7. If ICP decreases by >2 mmHg, assume autoregulation is intact and adopt this new higher MAP target
8. If ICP stays the same or increases, autoregulation is defective so revert to pr­­evious MAP target