UT Neuroanesthesia / Neuro 1 · Day 11
Unit Cerebral protection Read 12 min Reviewed Aug 28, 2026 · JP Next review Aug 2027

The Metabolic Floor

Why anesthetics cannot suppress cerebral metabolism past a fixed point, why hypothermia can, and why that advantage has never once produced a better patient.

Physiology · established Clinical claim · refuted 3 RCTs · n = 1,899
01

The pearl

Anesthetics can abolish only the electrophysiologic component of cerebral metabolism, roughly 60% of it. Hypothermia lowers that and the basal component underneath. It is the only intervention that reaches the floor, and in three randomized trials that mechanistic advantage produced no benefit and, in one, clear harm.

02

Why it is true

mechanism

The adult brain consumes roughly 3.5 mL O₂ per 100 g per minute at 37 °C. That total is conventionally divided into two accounts. About 60% pays for electrophysiologic work: synaptic transmission and the ion pumping needed to restore gradients after each depolarization. The remaining 40% pays for basal cellular integrity, the housekeeping that keeps a neuron alive whether or not it is signaling.

Anesthetics act on the first account only. Titrate propofol or a volatile agent deeper and CMRO₂ falls until the EEG goes isoelectric, at which point the electrophysiologic account is empty and further drug buys nothing. You get hypotension without additional metabolic suppression. That is the floor, and it sits at roughly 40% of baseline.

Temperature acts on chemistry rather than on electrical activity, so it depresses both accounts. The Q₁₀ for cerebral metabolism is approximately 2 to 2.5, which works out to a 6% to 7% reduction in CMRO₂ per degree Celsius. Cooling to 33 °C therefore lowers total demand by about a quarter, and cooling a brain that is already isoelectric lowers it further still. This is the sense in which hypothermia is uniquely capable, and it is entirely correct.

The corollary matters more in practice than the headline does. The relationship runs in both directions, so a febrile patient with a brain that cannot increase supply is spending 6% to 7% more oxygen per degree above normal. Fever is the temperature abnormality you can act on.

Electrophysiologic (~60%) Basal cellular integrity (~40%)
CMRO₂ mL O₂ · 100 g⁻¹ · min⁻¹ 0 1 2 3 ANESTHETIC FLOOR · 1.40 3.50 1.40 2.63 1.05 Awake Isoelectric EEG Active EEG Isoelectric EEG 37 °C 37 °C 33 °C 33 °C −60% vs awake −25% vs awake −70% vs awake ONLY COOLING CROSSES THE DASHED LINE
Figure 1. The two accounts of cerebral metabolism under four conditions. Deep anesthesia empties the electrophysiologic account and then stops; the basal account is untouched at any dose, which is why the dashed line is a floor rather than a way point. Cooling scales both accounts, so 33 °C combined with an isoelectric EEG is the only column that finishes below it. Values are illustrative, derived from a nominal CMRO₂ of 3.5, a conventional 60:40 split, and 6.3% per degree; the 60:40 partition is a teaching approximation, not a measured constant.
03

The evidence

3 RCTs · n = 1,899

IHAST

Todd MM, Hindman BJ, Clarke WR, Torner JC. Mild intraoperative hypothermia during surgery for intracranial aneurysm. N Engl J Med. 2005;352(2):135-145. No benefit
Design
Multicenter RCT, 30 centers
Population
1,001 adults, WFNS I–III aneurysmal SAH within 14 days, undergoing surgical clipping
Intervention
Cooled to 32.5–33.5 °C by first clip application, rewarmed after final clip
Control
Maintained at 36–37 °C
Primary
Glasgow Outcome Scale score of 1 at 90 days
Result
66% hypothermia vs 63% normothermia · OR 1.14 (95% CI 0.88–1.48) · P = 0.32
Harm signal
Postoperative bacteremia 5% vs 3% · P = 0.05

What it does not settle Good-grade patients only, and the ischemic exposure was brief. It does not speak to poor-grade SAH, and it was not powered for the subgroup most people actually worry about: patients who receive a temporary clip.

IHAST temporary-clipping analysis

Hindman BJ, Bayman EO, Pfisterer WK, Torner JC, Todd MM. No association between intraoperative hypothermia or supplemental protective drug and neurologic outcomes in patients undergoing temporary clipping during cerebral aneurysm surgery: findings from IHAST. Anesthesiology. 2010;112(1):86-101. No association
Design
Prespecified analysis of the IHAST cohort undergoing temporary clipping
Question
Does hypothermia, or any supplemental protective drug, improve outcome during deliberate focal ischemia?
Result
No association between hypothermia or supplemental protective drug and neurologic outcome

Why it matters here This is the closest thing in the literature to a direct test of the temporary-clip rationale, which is the exact situation in which residents are most often told to cool. It is observational within a randomized cohort, so it cannot exclude a small effect, but it removes the last comfortable argument.

Eurotherm3235

Andrews PJD, Sinclair HL, Rodriguez A, et al. Hypothermia for intracranial hypertension after traumatic brain injury. N Engl J Med. 2015;373(25):2403-2412. Harm
Design
International RCT, stopped early
Population
387 randomized; closed TBI with ICP > 20 mm Hg for ≥ 5 min despite stage 1 measures
Intervention
Hypothermia 32–35 °C as the stage 2 ICP therapy, titrated to the minimum cooling needed
Control
Standard care, escalating to stage 2 and 3 therapies as needed
Primary
GOS-E at 6 months
Result
Favorable outcome 25.7% hypothermia vs 36.5% standard care · adjusted common OR 1.53 (95% CI 1.02–2.30) favoring standard care · P = 0.04 · NNH ≈ 10
Mortality
Higher at 6 months in the hypothermia group

The lesson underneath the result Hypothermia worked on the surrogate. ICP came down. Patients did worse anyway. Keep this trial within reach for any conversation in which someone defends an intervention by the number it moves rather than the outcome it changes.

POLAR

Cooper DJ, Nichol AD, Bailey M, et al. Effect of early sustained prophylactic hypothermia on neurologic outcomes among patients with severe traumatic brain injury: the POLAR randomized clinical trial. JAMA. 2018;320(21):2211-2220. No benefit
Design
Multicenter RCT, prehospital or emergency-department randomization
Population
511 recruited, 466 completed primary outcome; severe blunt TBI, GCS < 9, age 18–60
Intervention
Early prophylactic cooling to 33–35 °C for at least 72 h, extended to 7 days if ICP remained elevated
Control
Normothermia 36.5–37.5 °C
Primary
Favorable GOS-E (5–8) at 6 months
Result
48.8% vs 49.1% · risk difference −0.3% · no significant difference
Harm signal
More pneumonia in the cooled group (per-protocol 70% vs 57%), more ventilator days, bradycardia, higher vasopressor requirement

What it closes POLAR tested the most favorable version of the hypothesis: cooling started early, before secondary injury was established, in young patients. Prophylactic hypothermia for TBI has now failed both prophylactically and therapeutically.

Synthesis. The physiology in section 02 is not in dispute and does not need softening. What is in dispute, and now settled against, is the inference from it. Across three randomized trials totaling 1,899 patients, induced hypothermia for neuroprotection produced no benefit in aneurysm surgery, no benefit prophylactically in TBI, and worse outcomes when used to control intracranial hypertension.

The most defensible reading is that basal metabolic demand was never the binding constraint on perioperative ischemic injury, while cooling reliably delivers infection, coagulopathy, hemodynamic instability, and a confounded neurologic exam. The mechanism was real. The mechanism was not the point.

04

What you actually do

in the room
  1. Target normothermia and treat fever, do not induce hypothermia for protection. Fever is the temperature abnormality with an evidence base worth acting on. Warm the patient, and warm them deliberately.
  2. Protect the pressure. Hypotension remains the best-documented secondary insult in every injured-brain population. Avoiding it is worth more than every pharmacologic protection strategy combined.
  3. Keep glucose in a wide, safe band. Avoid hyperglycemia and avoid the hypoglycemia that tight control produces. This is not a target to be aggressive about.
  4. Hold normocapnia unless the brain is acutely herniating. Hyperventilation is a rescue for minutes, not a management strategy for hours.
  5. Reserve burst suppression for refractory intracranial hypertension. It has an ICP indication, not a protection indication. Below an isoelectric EEG there is no further metabolic suppression to buy, only hypotension.
  6. Do not let the patient drift cold by accident either. Unintended hypothermia gives you coagulopathy in a field where hemostasis decides the case, plus shivering and a CMRO₂ surge during rewarming, and a delayed emergence that hides the exam you need.
  7. Deep hypothermic circulatory arrest is a different argument. When cooling is used for a giant or complex aneurysm, it buys tolerance of a deliberate global circulatory arrest. That is a surgical strategy for a specific operation, not evidence that cooling protects brains generally.
05

The pitfall

Mechanism laundering

The failure mode this day exists to prevent is not forgetting the physiology. It is reasoning from correct physiology straight to a clinical action without checking whether anyone ever tested it. Hypothermia is the cleanest example in neuroanesthesia, and burst suppression for routine protection is the same error wearing different clothes.

Expect to encounter the temporary-clip version of this: the surgeon announces a temporary clip and someone suggests cooling or a bolus of propofol to burst suppression "for protection." The Hindman analysis addressed exactly that population and found nothing. Meanwhile the propofol bolus drops the pressure during the one interval in the case when collateral perfusion to the territory beyond the clip is the only thing keeping that tissue alive.

The correct answer in that moment is unglamorous and is the whole point of the rotation: keep the pressure up, keep the CO₂ normal, keep the patient warm, and be ready to reperfuse.

06

Sources

primary first

Can't get the full text? The doi link always reaches the publisher's record. For one-click access through Preston holdings, install LibKey Nomad once — it follows you between devices instead of depending on which link you clicked.

07

Check yourself

3 items · one attempt each
Item 1 Application Neuro 1 · Day 11

A 58-year-old is undergoing clipping of a WFNS II ruptured anterior communicating artery aneurysm. The surgeon plans a period of temporary clipping. Your attending asks you to cool the patient to 33 °C "for protection" before the clip goes on. What is the best response?

Item 2 Analysis Neuro 1 · Day 11

Hypothermia has a real metabolic advantage over anesthetics, yet three randomized trials found no benefit. Which statement best explains the gap?

Item 3 Recall Neuro 1 · Day 11

In Eurotherm3235, hypothermia was used as a stage 2 therapy for intracranial pressure above 20 mm Hg after traumatic brain injury. What happened?

↻

Returning items

from earlier this rotation

Nothing to return yet — days 1 through 8 of Neuro 1 have not been authored.