Hypothermia

Meet the patient

A 78-year-old widow is found on her kitchen floor in January, confused, cold to the touch, breathing eight times a minute. The heating broke two days ago. Her core temperature on a low-reading rectal probe reads 29.4 degrees C, her ECG shows an Osborn wave at the J point, and she is in slow atrial fibrillation.^[1]

Three decisions now sit on your shoulders. How cold is she (stage with a low-reading core thermometer and the Swiss HT system). Why is she cold (broken boiler, or myxoedema, sepsis, hypoglycaemia, a femoral neck fracture). Can she be touched safely (a cold myocardium may fibrillate with rough handling). Hold those three and every section below slots into place.^[1][5]

What hypothermia is — and the single concept that drives the whole page

Accidental hypothermia is an unintentional drop of core temperature below 35 C, measured at a reliable site with a low-reading thermometer.^[5] ANZCOR: tympanic (thermocouple) if spontaneously breathing; oesophageal (distal) if a tracheal tube or supraglottic device is in place. Infrared tympanic devices that do not seal the ear are not designed for low core readings. Rectal and bladder probes are practical emergency-department defaults.^[3]

It happens when heat loss outruns heat production: a cold, wet, windy environment overwhelms an intact thermoregulator (primary), or thermogenesis itself fails and even modest cold tips the patient under (secondary). Injured and intoxicated patients cool quickly even in subtropical regions.^[5]

The single most tested concept: a hypothermic patient who looks dead may still be salvageable if cooling preceded anoxia. That is why arrest care is modified ALS plus ECLS, and why you do not stop on neurology or a 32 C clock alone — prognosticate with HOPE plus unsurvivable injury.^[5][8]

Stage the patient — two systems, no HT V

ANZCOR temperature bands: mild 32 to 35 C, moderate 28 to 32 C, severe below 28 C.^[3]

Swiss HT I to IV (ERC/Paal; no HT V in the current ERC table):

^{[5] [6] [8]}

The Revised Swiss System stages by AVPU and removes shivering as a stage-defining sign; it uses risk of cardiac arrest rather than a field temperature guess when a core reading is unavailable.^[6]

By aetiology, separate the patient who was overwhelmed from the patient who cannot generate heat:^[5]

• Primary (environmental) — previously healthy thermoregulation overwhelmed by cold: immersion, avalanche, mountain exposure, indoor cold in the vulnerable.
• Secondary (impaired thermogenesis) — illness, intoxication, or trauma: myxoedema coma, sepsis, hypoglycaemia, malnutrition, stroke or spinal injury, and drugs (alcohol, opioids, sedatives). Treat the cause as well as rewarm.^[5]

Enlarge

Who gets hypothermic, and why alcohol is the textbook co-factor

The exam stem almost always names excess heat loss (wet, wind, immersion, indoor cold), impaired heat production (extremes of age, malnutrition, hypothyroidism, sepsis, hypoglycaemia), or impaired judgement (alcohol, dementia, overdose). Ethanol is the classic co-factor because it combines vasodilation, poor behavioural defence, and hypoglycaemia — state that cluster, not an unsourced five-mechanism list as if it were a guideline table.^[1][5]

Neonates and the elderly are the two ends of life at highest risk. Paal: hypothermia and cardiac-arrest risk thresholds are lower in the elderly and comorbid (below 32 C) than in the young and healthy (below 30 C).^[5]

Pathophysiology — why the cold heart kills, and why the clotting screen lies

Normal core temperature is held near 37 C by hypothalamic and peripheral thermoregulation (vasoconstriction, shivering, behaviour).^[5] Heat is lost by radiation, convection, conduction, and evaporation; relative percentages are not sourced this pass.

Enlarge

Organ by organ — keep what is sourced, drop invented percentages:^[5]

• Brain — cooling reduces resting tissue oxygen consumption and can protect the brain if hypothermia precedes hypoxia; that is the rationale for prolonged CPR and ECLS, not a 32 C death clock. Fixed dilated pupils do not confirm death.^[5]
• Heart — bradycardia and conduction delay; Osborn (J) wave (positive deflection at the J point) is the exam ECG hallmark. Fetched ERC/Paal texts this pass do not give a 32 C J-wave cut-off — do not treat that number as a guideline threshold. Below about 28 C the risk of VF rises; rough movement can precipitate arrest (rescue collapse). Afterdrop is further cooling even after rewarming has started (reperfusion of cold regions).^[5]
• Kidney — cold diuresis contributes to volume depletion and rewarming hypotension.
• Blood — clotting enzymes are temperature-dependent. Laboratories routinely run clotting tests at 37 C, so PT/APTT can look near-normal while the patient bleeds in vivo. Treatment is rewarming, not clotting-factor replacement for the temperature effect alone.^[10] In trauma this is the trauma triad — hypothermia, coagulopathy, and acidosis.

Why drugs fail in the cold heart — the hypothermic heart may be unresponsive to cardioactive drugs, pacing, and defibrillation. ANZCOR: withhold adrenaline if core below 30 C; lengthen the interval to 6 to 10 minutes at 30 to 34 C; if VF persists after three shocks, delay further shocks until core above 30 C.^[3]

Clinical presentation — track consciousness, not a 28-to-24 band

• HT I — alert, typically shivering, the "umbles" (stumbles, mumbles, fumbles, grumbles) as bedside colour.
• HT II — impaired consciousness; shivering may have ceased or not.
• HT III — comatose with vital signs; arrhythmias; pupils may be fixed.
• HT IV — apparent death: no palpable pulse, no respiration. Yet fully recoverable with ECLS in selected patients (Wanscher: arrest victims at 15.5 to 20.2 C rewarmed on ECMO with good functional outcome in six of seven).^[12]

ECG: sinus bradycardia; Osborn (J) waves; prolonged PR/QRS/QT; atrial fibrillation; VF; asystole; shivering artefact. AF in a haemodynamically stable cold patient is usually observed while you rewarm — do not cardiovert as a reflex.^[1]

The elderly trap. Indoor, insidious, often no shivering history. Screen for sepsis, hypothyroidism, and hypoglycaemia.^[5]

Differential — every hypothermic patient gets glucose at the bedside

A low temperature with altered consciousness is not always simple environmental hypothermia.^[5]

• Faulty reading — infrared tympanic devices that do not seal the ear are not designed for low cores. Confirm with a low-reading core site.
• Sepsis — the septic patient may be hypothermic; treat infection in parallel.
• Myxoedema coma — hypothermic, bradycardic, hyponatraemic, comatose. IV levothyroxine plus hydrocortisone, and passive rewarming first (the myxoedematous circulation may not tolerate aggressive active rewarming).
• Hypoglycaemia — check capillary glucose immediately.
• Overdose — opioids, alcohol, sedatives; naloxone trial if opioids are possible.
• CNS event or trauma — collapsed then cooled.
• Adrenal crisis — hyponatraemia, hyperkalaemia, hypotension; parenteral hydrocortisone.^[1][5]

Bedside assessment — measure the core, then handle like glass

Reliable core sites: low-reading tympanic (thermocouple, insulated ear) if breathing; oesophageal if intubated or SGA in place; rectal/bladder as practical defaults. Unreliable as sole staging: oral, axillary, forehead scanners, unsealed infrared tympanic.^[3]

Handle gently. Remove wet clothing while minimising movement. Check signs of life for up to one minute (pulse, respiration, ECG, capnography, ultrasound) before declaring arrest. Do not start chest compressions on a slow but perfusing pulse.^[3]

^{[5] [8]}

Investigations — stage the patient and find the driver

• 12-lead ECG and continuous monitoring
• Capillary glucose immediately
• Arterial blood gas — machines report at 37 C (alpha-stat); pH-stat adjusts to body temperature and reclassifies many PaO2/PaCO2 values. Act on a consistent convention; do not invent a "falsely worse" rule beyond what Eastwood measured.^[11]
• Urea, electrolytes, potassium — a HOPE covariate, not a standalone cease rule except the ANZCOR avalanche potassium above 12 mmol per litre pair
• FBC, coagulation, group and save — remember the 37 C lab trap^[10]
• CK if prolonged immobility
• Cultures if sepsis
• TSH/cortisol if myxoedema or adrenal crisis possible
• Pregnancy test in women of childbearing age

HOPE uses age, sex, mechanism of cooling, core temperature, serum potassium, and CPR duration. External validation: 51/122 survived (42%); AUC 0.825; NPV 97% at a HOPE cut-off below 0.10. The original model was superior to potassium-only triage (AUC 0.895 versus 0.774).^[8][9]

Management — three universal rules, then rewarm by stage

1. Remove from the cold and insulate — cut wet clothes, dry, insulate (including the head), minimise movement.^[3]
2. Handle gently.
3. Monitor and find the cause — continuous core temperature and ECG, glucose, only warmed IV crystalloid (38 to 42 C).^[5]

Cardiac arrest — modified ALS, not a 32 C cease rule:^[3][5]

• Apparent death may be salvageable; continue CPR and rewarming unless HOPE plus the clinical picture support stopping, or there is unsurvivable trauma.
• Start CPR only if careful assessment confirms no signs of life.
• Defibrillation: up to three shocks; if VF persists, delay further shocks until core above 30 C.
• Drugs: withhold adrenaline below 30 C; 6 to 10 minute intervals at 30 to 34 C.^[3]
• Rewarm with ECLS; ECMO preferred over CPB. Transfer high-risk and arrested patients to an ECLS centre with continuous CPR.^[5]
• Avalanche (ANZCOR): burial over 35 minutes and cardiac arrest with an obstructed airway on extrication, or initial potassium above 12 mmol per litre, identifies victims not likely to survive — not a universal indoor-hypothermia potassium rule.^[3]

Rewarming ladder — match the method to the depth

Enlarge

1. Passive external (HT I, still able to shiver). Warm room, dry insulation. ANZCOR: passive rewarming is appropriate as patients are still able to shiver. Do not quote an unsourced 0.5 to 2 C per hour as a universal rate.^[3]

2. Active external (HT II). Forced warm air / chemical heat packs, trunk first to limit afterdrop. Do not quote an unsourced afterdrop magnitude of 1 to 2 C.^[5]

3. Active internal adjuncts and ECLS (HT III to IV, unstable or arrest). Warmed IV fluid 38 to 42 C; humidified gases; cavity lavage as a bridge; ECLS/ECMO for arrest. Body-cavity lavage is an adjunct, not a substitute for ECLS in arrest. Do not quote an unsourced 10 C per hour as a universal ECLS rate.^[5]

Arrhythmias: AF usually reverts with rewarming. VF — three shocks then delay below 30 C. Bradycardia is often physiological — atropine and pacing are typically ineffective; the hypothermic heart may be unresponsive to pacing.^[3]

The scenarios examiners set

• Cold-water submersion — combined asphyxia and hypothermia; oxygenation first, then rewarm; selected deep hypothermia remains salvageable (Wanscher).^[12]
• Avalanche — asphyxia versus hypothermia. ANZCOR: burial over 35 minutes plus obstructed airway, or potassium above 12 mmol per litre. Non-asphyxia cooling was associated with survival after unwitnessed ECLS (Podsiadlo).^[7]
• Elderly indoor hypothermia — search for sepsis and myxoedema; Paal high-risk threshold below 32 C in the elderly.^[5]
• Trauma triad — warm the patient as part of haemorrhage control; do not trust a 37 C clotting screen.^[10]
• Myxoedema coma — levothyroxine plus hydrocortisone; passive rewarming first.^[5]
• Neonate — large surface-area-to-mass ratio; prevention and vigorous resuscitation. Do not apply adult HOPE cut-offs to children.^[8]

Complications and pitfalls

VF and arrest; aspiration; AKI and rhabdomyolysis; coagulopathy; afterdrop and rewarming shock; frostbite.^[5]

Classic pitfalls: declaring death on a 32 C clock or on pupils; rough handling; cold IV fluid; infrared tympanic as the sole reading; missing the secondary cause; standard 3-to-5-minute adrenaline below 30 C; potassium-only indoor triage; HT V as current ERC staging; quoting ECLS survival "approaching 100%" as typical (Paal 2016 notes that in some series; Podsiadlo unwitnessed ECLS survival was 27%, with 83% of survivors neurologically intact).^[2][7]

Prognosis and disposition

HOPE is the in-hospital ECLS triage tool (validation survival 42% in that cohort; NPV 97% if HOPE below 0.10).^[8] Unwitnessed hypothermic arrest rewarmed with ECLS: overall survival 27%.^[7] Transfer high-risk HT III and all HT IV to an ECLS centre.^[5]

Disposition: HT I observe and discharge once normothermic with a cause addressed; HT II monitored bed; HT III ICU with ECLS on the horizon; HT IV ECLS centre with CPR in transit.^[5]

Special populations

• Elderly — Paal: arrest risk below 32 C; indoor hypothermia; search for sepsis/myxoedema.^[5]
• Children — large surface area; resuscitate fully; do not apply adult HOPE cut-offs.^[8]
• Pregnancy — rewarm the mother; fetal monitoring. No unsourced labour-risk percentage this pass.^[5]
• Intoxicated — glucose, thiamine if malnourished, airway.^[5]
• Trauma / anticoagulated — rewarm as haemorrhage control.^[10]

Evidence, guidelines, and regional differences

• Paal 2021 / ICAR — definition, high-risk transfer, ECMO for ECLS rewarming, afterdrop.^[5]
• ERC 2021 — special-circumstances ALS (three shocks, adrenaline withholding).^[3]
• WMS 2019 — out-of-hospital evaluation and treatment.^[4]
• ANZCOR 11.10 — temperature bands, one-minute life check, three-shock then delay, adrenaline 6 to 10 minutes at 30 to 34 C, avalanche 35 minutes / K above 12.
• HOPE papers — replace dichotomous potassium as the primary ECLS triage tool.^[8][9]
• India / winter indoor cold — secondary hypothermia in the elderly and neonates is a public-health reality; the physiology and ALS rules are the same. Do not invent unsourced Delhi mortality counts.^[5]

Exam pearls — the mantra and the memory devices

The mantra: measure a low-reading core, handle like glass, rewarm by Swiss stage, three shocks then delay, HOPE not a 32 C clock.^[5]

^{[5] [8] [3] [5]}

Ward-round test — five stems with answers in the Reveal.

A collapsed hill-walker, core 29 C, no shivering, slow atrial fibrillation. What rewarming method, and what must you NOT do?ShowHide

This is HT II to III border (29 C, impaired). Use active external rewarming plus warmed IV fluid 38 to 42 C, and prepare ECLS transfer if high-risk (core below 30 C in a young/healthy patient). Do NOT give atropine for the bradycardia, do NOT cardiovert stable AF, and do NOT handle roughly or give cold fluid. Absence of shivering does not by itself define the stage.^[5]

An avalanche victim in cardiac arrest, core 24 C, potassium 14 mmol/L, buried 90 minutes with an obstructed airway. Resuscitate or cease?ShowHide

ANZCOR: burial over 35 minutes with arrest and an obstructed airway, or potassium above 12 mmol per litre, identifies avalanche victims not likely to survive. This patient meets both. That is an avalanche pair — not a reason to stop every indoor hypothermic arrest at potassium 12. For non-avalanche ECLS triage use HOPE.^[8]

A drunk man found on a park bench, core 31 C, glucose 2.1 mmol/L, GCS 10. Two diagnoses, two treatments?ShowHide

Hypothermia plus hypoglycaemia (alcohol as co-factor). Give IV dextrose immediately, thiamine if malnourished, and active external rewarming (HT II range). Protect the airway. Then search for head injury, overdose, and sepsis.^[5]

A bleeding trauma patient arrives cold and coagulopathic, PT and APTT reported 'normal'. Why, and what do you do?ShowHide

Clotting tests are run at 37 C, so they miss in-vivo hypothermic coagulopathy. Rewarm and use damage-control resuscitation. Do not be reassured by the lab.^[10]

A hypothermic arrest in VF does not convert after three shocks at core 28 C. What now?ShowHide

Delay further shocks until the core is above 30 C. Continue CPR, withhold adrenaline below 30 C, rewarm by ECLS/ECMO, and resume shocks and drugs once above 30 C using 6 to 10 minute adrenaline intervals until 34 C. Do not stop because the core is not yet 32 C — use HOPE.^[3][8]

References12ShowHide

1. [1]Brown DJ, Brugger H, Boyd J, Paal P. Accidental hypothermia N Engl J Med, 2012.PMID 23150960
2. [2]Paal P, Gordon L, Strapazzon G, et al. Accidental hypothermia-an update : The content of this review is endorsed by the International Commission for Mountain Emergency Medicine (ICAR MEDCOM) Scand J Trauma Resusc Emerg Med, 2016.PMID 27633781
3. [3]Lott C, Truhlář A, Alfonzo A, et al. European Resuscitation Council Guidelines 2021: Cardiac arrest in special circumstances Resuscitation, 2021.PMID 33773826
4. [4]Dow J, Giesbrecht GG, Danzl DF, et al. Wilderness Medical Society Clinical Practice Guidelines for the Out-of-Hospital Evaluation and Treatment of Accidental Hypothermia: 2019 Update Wilderness Environ Med, 2019.PMID 31740369
5. [5]Paal P, Pasquier M, Darocha T, Lechner R, Kosinski S, Wallner B, Zafren K, Brugger H. Accidental Hypothermia: 2021 Update Int J Environ Res Public Health, 2022.PMID 35010760
6. [6]Musi ME, Sheets A, Zafren K, Brugger H, Paal P, Hölzl N, Pasquier M. Clinical staging of accidental hypothermia: The Revised Swiss System Resuscitation, 2021.PMID 33675869
7. [7]Podsiadło P, Darocha T, Svendsen ØS, Kosiński S, Silfvast T, Blancher M, Sawamoto K, Pasquier M. Outcomes of patients suffering unwitnessed hypothermic cardiac arrest rewarmed with extracorporeal life support: A systematic review Artif Organs, 2021.PMID 32920881
8. [8]Pasquier M, Rousson V, Darocha T, et al. Hypothermia outcome prediction after extracorporeal life support for hypothermic cardiac arrest patients: An external validation of the HOPE score Resuscitation, 2019.PMID 30940473
9. [9]Pasquier M, Hugli O, Paal P, Darocha T, Blancher M, Husby P, Silfvast T, Carron PN. Hypothermia outcome prediction after extracorporeal life support for hypothermic cardiac arrest patients: The HOPE score Resuscitation, 2018.PMID 29481910
10. [10]Reed RL 2nd, Johnson TD, Hudson JD, Fischer RP. The disparity between hypothermic coagulopathy and clotting studies J Trauma, 1992.PMID 1404519
11. [11]Eastwood GM, Suzuki S, Lluch C, Schneider AG, Bellomo R. A pilot assessment of alpha-stat vs pH-stat arterial blood gas analysis after cardiac arrest J Crit Care, 2015.PMID 25449882
12. [12]Wanscher M, Agersnap L, Ravn J, et al. Outcome of accidental hypothermia with or without circulatory arrest: experience from the Danish Præstø Fjord boating accident Resuscitation, 2012.PMID 22634431