UT Neuroanesthesia / Neuro 2 · Day 4
Unit Awake craniotomy Read 14 min Reviewed Sep 1, 2026 · JP Next review Aug 2027

Awake Enough to Map

Only one of the three phases requires the patient to be genuinely awake. The rest of the anaesthetic is a choice between techniques that a meta-analysis of forty-seven studies could not separate — and a selection decision that matters more than either.

Physiology · established Clinical claim · uncertain 1 meta-analysis · 47 studies · no adequately powered RCT exists
01

The pearl

An awake craniotomy has three phases and only the resection phase requires the patient to be truly awake and interactive. Everything else — asleep-awake -asleep, monitored anaesthesia care, awake-awake-awake — is a choice about the other two, and a meta-analysis of 47 studies found no relevant outcome difference between the two main techniques except the rate of conversion to general anaesthesia. What is not interchangeable is patient selection, and the airway you cannot reach because the head is in pins.

02

Why it is true

mechanism

Awake craniotomy sounds like a different specialty and it is mostly a sequencing problem. Surgery is performed while the patient is aware and able to cooperate with functional testing of the cortex, because the tissue to be resected sits close to eloquent cortex — movement, vision or language.

The reason to do it is measurable. Awake craniotomy has been shown to reduce the size of the resection, surgery time, postoperative neurologic deficits, early postoperative nausea and vomiting, and length of hospital stay. Vasopressor use and hypertension during head pinning are lower. Stays as short as a day have been reported in patients with good functional status and uncomplicated tumours, and the technique has even been studied as an outpatient procedure without reported adverse outcomes. Patient acceptance is excellent, and patients have better postoperative pain scores and use less opioid — despite the commonest complaints being pain from the head holder, inadequate local anaesthesia and an uncomfortable position.

Three phases, one of which is the point

The procedure divides into preresection, resection and postresection, and the sentence to carry is this: only the resection phase requires the patient to be truly awake and interactive.

That reframes the whole anaesthetic. The two named approaches — asleep-awake-asleep and monitored anaesthesia care — differ only in what happens either side of that window. General anaesthesia in the outer phases can use a laryngeal mask, an endotracheal tube or airway adjuncts such as a nasal airway; all have been described and found safe, and in current practice endotracheal intubation is rarely used. Sedation approaches without a fully protected airway have proven safe and effective, with lower operative time and a similar safety profile, typically using sedative doses of propofol and dexmedetomidine with remifentanil analgesia.

There is a third approach, awake-awake-awake, performed without sedation at all. It depends on therapeutic communication and psychological support from the anaesthetist, with analgesia from the scalp block and remifentanil only as needed. The evidence for it is limited, which is a statement about the literature rather than about the technique.

What the numbers actually are

Across 47 studies, the pooled rates are worth knowing before you consent anyone:

  • Awake craniotomy failure: 2% (95% CI 1–3)
  • Conversion to general anaesthesia: 2% (95% CI 2–3)
  • Intraoperative seizures: 8% (95% CI 6–11)
  • New neurological dysfunction: 17% (95% CI 12–23)

And on the argument everyone has: meta-regression of the two main techniques did not reveal any relevant differences in outcomes explained by technique, except for conversion into general anaesthesia. Both were feasible and safe. The authors conclude that large randomized trials are required to prove superiority of one regime — which has not happened.

So the technique question is genuinely open, and the seizure rate of roughly one in twelve is not. Brief the room for the second rather than arguing about the first.

Selection, which is where the case is actually won

Appropriate patient selection and preparation are described as the most important factors in success. The assessment covers age and maturity, anxiety, claustrophobia, psychiatric disorders, the presence or likelihood of a difficult airway, and a history of reflux or nausea and vomiting.

Three factors are named as risks for sedation failure: hypertension, alcohol abuse, and lack of maturity. Children younger than about 14 are generally not recommended, though developmental status is assessed individually.

And the disqualifying one, which is today’s pitfall: a patient with a potentially difficult airway, or who is likely to obstruct while sedated, is a poor candidate — because the head will be in pins and sudden conversion to intubation may be difficult.

Preparation is not optional either. The anaesthetist should set out what to expect: the varying states of sedation and awareness, the positioning, the possible discomfort, and the testing process. A strong rapport established beforehand is part of the technique rather than a courtesy.

Positioning and the block

Position for the patient’s comfort and your access. Lateral or semilateral is common. Pad generously and check pressure points. Have the patient confirm they are comfortable before sedation, because they will be in pins on emergence and must stay still when repositioning is not feasible. Tenting the drapes upward on the anaesthesia side gives access and reduces the sense of claustrophobia; a deflatable beanbag or fixed backrest stabilizes the position, with wide straps and tape.

The scalp block carries all three phases and must be complete. Six nerves are blocked individually: auriculotemporal, zygomaticotemporal, supraorbital, supratrochlear, lesser occipital and greater occipital. Ropivacaine can be used safely up to 4.5 mg/kg and levobupivacaine up to 2.5 mg/kg; mepivacaine may be added where faster onset is wanted. Peak plasma concentrations occur about 15 minutes after injection, so it is worth placing the block with that in mind. Severe bradycardia after scalp block has been reported.

The first test of the block is free: the response to head pin placement. An obvious physical response in a sedated patient, or a rise in heart rate and blood pressure under general anaesthesia, indicates failure — at a point where you can still supplement with local anaesthetic at the pin sites, or rescue with a bolus of propofol, rather than discovering it during mapping.

And if you have chosen the asleep start

One detail that catches people. If you take the patient through the first phase under general anaesthesia and then wake them for testing, they will be emerging in head pins. Bucking must be avoided assiduously — which is Neuro 1 day 16’s problem arriving in a case where the head is fixed to the table.

The awake window Expected complications New deficit
ONLY ONE PHASE ACTUALLY NEEDS THE PATIENT AWAKE Preresection Sedation or general anaesthesia, with or without an airway. Resection AWAKE AND INTERACTIVE Speech, memory, motor testing. the only phase that requires it Postresection Sedation or general anaesthesia again, as chosen. THE SCALP BLOCK CARRIES ALL THREE PHASES · SIX NERVES auriculotemporal zygomaticotemporal supraorbital supratrochlear lesser occipital greater occipital peak plasma concentration ≈ 15 min after injection POOLED ACROSS 47 STUDIES · 95% CI 2% procedure failed 1–3 2% converted to GA 2–3 8% intraoperative seizure 6–11 17% new neurological deficit 12–23 POOLED PROPORTIONS FROM MOSTLY OBSERVATIONAL STUDIES — NOT A RANDOMISED COMPARISON.
Figure 1. The phase structure is the thing to hold, because it converts a frightening-sounding case into one interactive window with ordinary anaesthesia on either side. The rates below are pooled proportions from mostly observational studies, not a randomised comparison — they tell you what to expect and how often to expect it, and they do not establish that one technique causes those numbers rather than another.
03

The evidence

1 meta-analysis · 47 studies · no adequately powered RCT exists

Anaesthesia management for awake craniotomy

Stevanovic A, Rossaint R, Veldeman M, Bilotta F, Coburn M. Anaesthesia management for awake craniotomy: systematic review and meta-analysis. PLoS One. 2016;11(5):e0156448. No association
Design
Systematic review and meta-analysis of 47 studies
Population
Patients undergoing awake craniotomy
Question
Does asleep-awake-asleep or monitored anaesthesia care produce better outcomes, and what are the expected complication rates?
Result
Awake craniotomy failure 2% (95% CI 1–3) · conversion to general anaesthesia 2% (95% CI 2–3) · intraoperative seizures 8% (95% CI 6–11) · new neurological dysfunction 17% (95% CI 12–23). Meta-regression of the two techniques "did not reveal any relevant differences between outcomes explained by the technique, except for conversion into GA."

What 47 studies cannot do These are pooled proportions from a literature that is largely observational, and single-arm rates from case series are vulnerable to selection and reporting effects in a way randomized comparisons are not — the 17% new neurological dysfunction in particular will depend heavily on how hard each series looked and when. The authors' own conclusion is the honest one: both techniques were feasible and safe, data for the awake-awake-awake technique are limited, and large RCTs are required to prove superiority of one anaesthetic regime. That trial has not been done.

Synthesis. This is a day where the evidence answers a different question from the one people argue about.

The argument is technique: asleep-awake-asleep against monitored anaesthesia care. Across 47 studies, meta-regression found no relevant difference in outcomes explained by technique, except for conversion to general anaesthesia. Both are feasible and safe. The awake-awake-awake approach has too little data to place. And the authors call for the randomized trials that would settle it, which is a fair description of a field running on accumulated practice.

What the pooled data do give you is a set of expectations, and those are worth more at the bedside than a technique preference. Roughly 1 in 50 of these procedures fails; roughly 1 in 50 converts to general anaesthesia; roughly 1 in 12 has an intraoperative seizure; and roughly 1 in 6 has new neurological dysfunction. Knowing those numbers changes how you consent the patient, how you brief the room, and what you have prepared.

The decision that is not technique-sensitive is selection. A patient who will obstruct their airway when sedated, or who cannot tolerate lying still and being questioned for an hour, is a poor candidate whichever letters you put after the anaesthetic — because the head will be in pins and sudden conversion to intubation may be difficult.

04

What you actually do

in the room
  1. Hold the phase structure in your head, not the acronym. Preresection, resection, postresection — and only the resection phase requires the patient truly awake and interactive. That framing turns the case into one interactive window with ordinary anaesthesia on either side.
  2. Select the patient before you select the technique. Age and maturity, anxiety, claustrophobia, psychiatric history, likelihood of a difficult airway, reflux and nausea history. Hypertension, alcohol abuse and lack of maturity are described as risk factors for sedation failure, and children under about 14 are generally not recommended.
  3. Treat a patient who will obstruct when sedated as a poor candidate. The head will be in pins and positioned by the surgeon, so sudden conversion to intubation may be difficult. Obstructive sleep apnoea, difficult ventilation or difficult intubation raise the risk, and success has been reported in such patients using high-flow oxygen — which is a description of a rescue, not a plan.
  4. Agree the airway rescue plan with the surgeon before you start. The source is explicit that a plan for airway management must be agreed between the anaesthesia and surgical teams in the event airway patency is lost. Two per cent of these cases convert to general anaesthesia; that is not rare enough to improvise.
  5. Have the patient confirm they are comfortable before you sedate them. They will be in pins on emergence and must remain still when repositioning is not feasible. Lateral or semilateral positioning, generous padding, and tenting the drapes upward on your side gives access and reduces the patient's sense of claustrophobia.
  6. Block all six nerves and respect the dose ceilings. Auriculotemporal, zygomaticotemporal, supraorbital, supratrochlear, lesser occipital and greater occipital. Ropivacaine to 4.5 mg/kg and levobupivacaine to 2.5 mg/kg; peak plasma concentration is about 15 minutes after injection, so time it. Severe bradycardia after scalp block has been reported.
  7. Judge the block by the response to the pins. An obvious physical response in a sedated patient, or a rise in heart rate and blood pressure under general anaesthesia, indicates block failure — before the operation has really started, when you can still supplement.
  8. Expect a seizure roughly once in twelve cases, and have the plan ready. Pooled intraoperative seizure rate is 8% (95% CI 6–11). That is common enough that it should be part of the room brief rather than a surprise.
05

The pitfall

The airway you cannot reach

Every other complication in this case is manageable in the room. This one is a geometry problem, and it is set before the incision.

The patient's head is in pins and positioned by the surgeon. If airway patency is lost during the awake phase — obstruction in a sedated patient with obstructive sleep apnoea, oversedation to control anxiety, a seizure — then sudden conversion to endotracheal intubation may be difficult, and the difficulty is mechanical rather than pharmacological. You cannot get the head where you would want it.

Two decisions prevent this and both happen before the case. Selection: a patient with a potentially difficult airway, or who is likely to obstruct when sedated, is a poor candidate for the procedure. Success has been reported in patients with obstructive sleep apnoea using high-flow oxygen, and that is a report of a rescue rather than a reason to proceed. And an agreed plan: the source states that a plan for airway management must be agreed between the anaesthesia and surgical teams in the event patency is lost. Agreed means the surgeon knows what you will ask for and how fast the pins come off.

Two per cent of awake craniotomies convert to general anaesthesia. That is one in fifty — often enough that every list will produce one eventually, and rare enough that nobody in the room will have done it recently.

The related trap is quieter and it is about sedation depth. The temptation when a patient is anxious is to deepen, and the specific hazard of deepening here is that spontaneous ventilation must be maintained and airway reflexes preserved. The alternative is the one the awake-awake-awake technique is built on: rapport, explanation and psychological support, established before the patient ever reaches the operating room.

06

Sources

primary first

Full text · Preston goes through LibKey to whatever Preston holds — the PDF where there is one — and works on a phone. The doi link always reaches the publisher's record, whether or not we hold a subscription. If you are reading on a computer, LibKey Nomad adds the same one-click PDF button to PubMed and publisher pages everywhere else; it is a desktop browser extension, so it does nothing on a phone or tablet.

07

Check yourself

3 items · one attempt each
Item 1 Analysis Neuro 2 · Day 4

A colleague argues that monitored anaesthesia care is superior to asleep-awake-asleep for awake craniotomy, citing the meta-analysis of 47 studies. What is the accurate reading?

Item 2 Application Neuro 2 · Day 4

Which of these patients is the poorest candidate for awake craniotomy?

Item 3 Recall Neuro 2 · Day 4

Which statement about the scalp block for awake craniotomy is correct?

↻

Returning items

from earlier this rotation
Day 01 On T1 and T2, how do CSF and most pathology appear — and which sequence is contrast given with?

That is the whole orientation rule. Fat is bright on T1 and intermediate on T2, cortical bone gives no signal, and flowing blood produces a signal void. The practical summary is that anatomy is better seen on T1 and pathology on T2 — and post-contrast sequences are T1 weighted for exactly the reason given.

Day 01 A patient with a clipped aneurysm needs an MRI. What determines whether they can have one?

Ferromagnetic aneurysm clips are a contraindication; non-ferromagnetic ones are not. The risk from a ferromagnetic object is magnetically induced movement, with dislodgement and injury, and heating — image artefact is the secondary concern. The information you need is in the old operation note, not on any current imaging.