Controlling temperature to protect the brain
"The brain that survives the cardiac arrest can still be lost in the hours that follow. Targeted temperature management is not about making the patient cold — it is about not letting them get hot, and controlling the one variable that reliably worsens reperfusion injury: fever."
Summarised from Marino's The ICU Book; with the Washington Manual of Critical Care and Irwin & Rippe's Intensive Care Medicine.After return of spontaneous circulation (ROSC), reperfusion triggers a cascade of injury — excitotoxicity, calcium influx, free-radical formation, mitochondrial failure and apoptosis. Targeted temperature management (TTM) actively controls core temperature to a defined target to blunt this post-cardiac-arrest brain injury. Every 1 °C rise in temperature increases cerebral metabolic demand by ~6–8%, so fever is actively harmful.
Who gets TTM
- Adults who remain comatose (not obeying commands) after ROSC from cardiac arrest — the core indication, regardless of the initial rhythm (shockable or non-shockable) and of in- vs out-of-hospital arrest.
- Applied as part of a complete post-cardiac-arrest bundle: airway/ventilation, haemodynamic support, urgent coronary angiography where indicated, and treating the cause.
Who does not benefit: a patient who is awake and following commands after ROSC does not need temperature lowering — but should still have fever actively prevented.
Why "33 °C for everyone" is no longer the rule
2002 — HACA & Bernard
Cooling to 32–34 °C after out-of-hospital VF arrest improved neurological outcome vs no temperature control. Hypothermia entered guidelines.
2013 — TTM trial
33 °C was no better than 36 °C. The benefit seemed to be avoiding fever, not deep cooling. Targets shifted to 32–36 °C.
2021 — TTM2
Hypothermia at 33 °C was no better than targeted normothermia (≤37.5 °C, treating fever) for survival or function — and caused more arrhythmias.
2021 → now — ERC/ESICM & ILCOR
Actively prevent fever (keep ≤37.5 °C) for ≥72 h in comatose post-arrest patients. Cooling to 32–36 °C remains an acceptable option; deep hypothermia is no longer mandated.
The modern intervention is fever avoidance. Whether you hold the patient at strict normothermia or actively cool, the essential thing is that the temperature is controlled and fever is not allowed.
Induction, maintenance, rewarming
| Phase | Goal | Key points |
|---|---|---|
| Induction | Reach target quickly | Cold IV fluids and surface/endovascular cooling; start early; sedate + control shivering |
| Maintenance | Hold a constant target (≤37.5 °C, or 32–36 °C if cooling) | Feedback-controlled device; avoid overshoot and swings; continuous core temperature (bladder/oesophageal) |
| Rewarming | Warm slowly: ~0.25–0.5 °C/hour | The dangerous phase — rebound hyperkalaemia, hypotension (vasodilation), hypoglycaemia; then prevent fever for ≥72 h total |
Do not simply stop TTM at 24 h and walk away — rebound hyperthermia after rewarming is common and harmful. Continue fever prevention to 72 h.
Methods and their trade-offs
| Method | Examples | Notes |
|---|---|---|
| Surface | Cooling blankets/pads, ice packs, gel pads with feedback | Non-invasive; feedback pads give tight control; skin care needed |
| Endovascular | Central catheter with circulating coolant | Precise, fast control; invasive; line-related risks |
| Adjunct | Cold IV crystalloid (4 °C), antipyretics | Cold fluid speeds induction but does not maintain; avoid large volumes in cardiogenic shock |
Monitor core, not skin: use a bladder or oesophageal probe (or blood temperature if on an endovascular device). Surface/axillary readings lag and mislead.
What hypothermia does to every organ
| System | Effect during cooling |
|---|---|
| Electrolytes | Hypokalaemia, hypomagnesaemia, hypophosphataemia (intracellular shift) — with rebound hyperkalaemia on rewarming |
| Cardiac | Bradycardia (expected, often benign), prolonged QT, arrhythmia risk if <30 °C |
| Metabolic | Insulin resistance → hyperglycaemia; reduced drug metabolism (sedatives accumulate) |
| Haematology | Impaired platelet function/coagulation; masked by warmed lab samples |
| Immune / renal | Increased infection risk; cold diuresis → hypovolaemia |
| Neuromuscular | Shivering (raises metabolic demand and opposes cooling) |
Hypothermia slows clearance of sedatives and neuromuscular blockers — drug effects (and a falsely "poor" neuro exam) persist longer, which is why prognostication is delayed.
Shivering defeats cooling — treat it stepwise
Assess
Use the Bedside Shivering Assessment Scale. Shivering raises metabolic rate and cerebral oxygen demand — the opposite of the goal.
Non-sedating first
Skin counter-warming (warm hands/face), paracetamol, magnesium.
Sedation / analgesia
Opioids (e.g. pethidine lowers the shivering threshold), dexmedetomidine, propofol titrated to effect.
Neuromuscular blockade — last resort
Only if shivering is refractory and target cannot be held; requires deep sedation and ideally continuous EEG (paralysis masks seizures).
Never on a single test, never too early
Prognostication in the comatose post-arrest patient must be delayed to ≥72 h after ROSC (longer if sedation/hypothermia may still be confounding) and must be multimodal.
- Clinical: bilaterally absent pupillary AND corneal reflexes at ≥72 h; myoclonus status is worrying but not alone decisive.
- Electrophysiology: bilaterally absent N20 on somatosensory evoked potentials; malignant EEG patterns (suppression, burst-suppression).
- Biomarker: markedly elevated/rising neuron-specific enolase (NSE).
- Imaging: diffuse anoxic injury on CT/MRI (loss of grey–white differentiation).
Use ≥2 concordant predictors in a patient off sedation and normothermic. A poor exam under residual sedation/hypothermia is not grounds to withdraw care.
Where teams go wrong
- Chasing deep hypothermia (33 °C) as if it were mandatory — fever avoidance is the evidence-based goal.
- Stopping TTM at 24 h and allowing rebound fever.
- Rewarming too fast → rebound hyperkalaemia, hypotension, hypoglycaemia.
- Aggressively replacing potassium during cooling, then hyperkalaemia on rewarming.
- Prognosticating too early or on one test while sedation/hypothermia still confound the exam.
- Monitoring skin/axillary rather than core temperature.
High-yield one-liners
Q: What did TTM2 show?
Hypothermia (33 °C) was no better than targeted normothermia (≤37.5 °C) — fever prevention is the key intervention.
Q: Temperature target & duration now?
Keep ≤37.5 °C (or 32–36 °C if cooling) and prevent fever for ≥72 h in comatose post-arrest patients.
Q: Electrolyte shift during cooling vs rewarming?
Hypokalaemia during cooling; rebound hyperkalaemia on rewarming.
Q: When and how to prognosticate?
≥72 h off sedation, multimodal (exam + SSEP + EEG + NSE + imaging); never a single test.
Q: Rewarming rate?
Slow — about 0.25–0.5 °C per hour.
Side-by-side comparisons
The distinctions the examiners love: the two temperature strategies, and cooling vs rewarming physiology.
Hypothermia (32–36 °C) vs Targeted Normothermia (≤37.5 °C)
| Feature | Hypothermia | Targeted normothermia |
|---|---|---|
| Target | 32–36 °C | ≤37.5 °C (treat fever) |
| Outcome evidence | Not superior (TTM, TTM2) | Equivalent survival/function |
| Arrhythmia/complications | More (bradycardia, arrhythmia) | Fewer |
| Current status | Acceptable option | Reasonable default |
Cooling vs Rewarming — the physiology flips
| Variable | Cooling | Rewarming |
|---|---|---|
| Potassium | Falls (intracellular shift) | Rises (rebound hyperkalaemia) |
| Vascular tone | Vasoconstriction | Vasodilation → hypotension |
| Glucose | Rises (insulin resistance) | Can fall → hypoglycaemia |
| Main risk | Shivering, arrhythmia | Doing it too fast |
References
- Marino PL. Marino's The ICU Book. 5th ed. Wolters Kluwer; 2025.
- Kollef MH, Isakow W, Burks AC, Despotovic VN (eds). The Washington Manual of Critical Care. 4th ed. Wolters Kluwer; 2024.
- Irwin RS, Lilly CM, Mayo PH, Rippe JM (eds). Irwin & Rippe's Intensive Care Medicine. 9th ed. Wolters Kluwer; 2023.
- Dankiewicz J, Cronberg T, Lilja G, et al. (TTM2). Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest. N Engl J Med. 2021;384:2283–2294.
- Nolan JP, Sandroni C, Böttiger BW, et al. ERC-ESICM Guidelines 2021: Post-resuscitation care. Intensive Care Med. 2021;47:369–421.