Authors: Tim Harris, Jonathan M Jones / Editors: Adrian Boyle, Gavin Lloyd / Reviewer: Rebecca Ford / Codes: CMP2, HAP11, HMP2, ACCS LO 2, ELP6, EnvC3, RP3, SLO3, SLO5 / Published: 03/12/2018
The temperate climate of the UK means that severe hypothermia leading to cardiovascular instability and cardiac arrest is rare. Despite this it is important to understand the patho-physiology and treatment of this emergency as the prognosis, particularly in the young and previously healthy can be remarkably good. Achieving a good outcome requires aggressive and prolonged input from a multi-disciplinary team.
The human body has evolved to work within a narrow, carefully maintained, core temperature range. As hypothermia develops, many organ systems stop working properly. The physiological effects of hypothermia mean that the management of cardiac arrest requires an altered approach. Remember, too, that hypothermia can be protective greatly prolonged resuscitation efforts may be justified.
Hypothermia is defined as a core temperature below 35 C, with mild hypothermia classified as 32-35 C, moderate 30-32 C and severe disease below 30 C . It is frequently associated with submersion injury and drowning (see forthcoming session).
Two groups of patients seen are winter sports and wilderness enthusiasts and urban poor. Factors contributing to the latter group include alcohol, drugs, malnutrition and poor socioeconomic conditions . Deaths in the UK are rare and occur predominantly in the elderly, more often in women than men . However, in 2017 there were 34 deaths in Scotland that involved hypothermia
This CEMPaedia article concentrates on patients with actual or impending cardiopulmonary arrest. The treatment of mild-moderate hypothermia and localised cold injuries are dealt with in the article on Hypothermia & Frostbite.
Body temperature is normally tightly regulated but is less well-controlled in the very young, and impaired by age, disease, injury and alcohol/drugs.
As the body cools, the metabolic rate falls and neural transmission is inhibited . Multiple derangements occur including:
- Depressed myocardial contractility
- Leftwards shift of the oxygen dissociation curve (leading to reduced availability of oxygen at the tissue level)
- Ventilation-perfusion mismatch
- Increased blood viscosity
All of these reduce tissue oxygenation .
Initially, sympathetic drive increases heart rate and respiratory rate, and heat is produced via shivering. At around 30 C, this process ceases and ventilation, heart rate, blood pressure and cardiac output fall.
Intravascular volume falls due to cold diuresis and fluid shifts into the extravascular space.
Sinus bradycardia develops followed by atrial fibrillation (AF) followed by ventricular arrhythmias (usually below 32 degrees C) including ventricular fibrillation (VF). Finally, asystole  results
|Correlation of Core Temperature and Physiological Changes|
|Mild||35||Maximum shivering, impaired judgment, confusion|
|34||Tachycardia, tachypnoea, increased SVR, cold diuresis|
|33||Bradycardia, respiratory depression, hyperglycaemia, dysarthria, ataxia|
|Moderate||32||Stupor, lethargy, arrest of shivering thermogenesis|
|31||Atrial arrhythmia, Osborn J-waves on ECG, worsening bradycardia|
|30||Insulin ineffective, decreased oxygen consumption|
|29||Progressive decreased level of consciousness, pulse, respiratory rate|
|Severe||28||Increased susceptibility for ventricular fibrillation, pulse rate and oxygen consumption decreased by 50%|
|27||Loss of reflexes and voluntary movement, hypoglycaemia|
|26||Acid-base disturbances, no response to pain|
|25||Cerebral blood flow decreased by 2/3, pulmonary oedema, apnoea|
|23||Corneal and oculocephalic reflexes absent|
|22||Maximum risk for ventricular fibrillation, oxygen consumption 25% of
|Profound||20||Lowest resumption of cardiac activity, pulse 20% of normal|
Note that malignant arrhythmias are unlikely to be hypothermia-induced at temperatures above 32 C consider alternative causes such as acute coronary syndrome (ACS).
Cardiac arrest may occur as a direct result of hypothermia, or hypothermia may occur secondary to cardiac arrest. Primary hypothermic cardiac arrest usually requires severe hypothermia. Because hypothermia protects the vital organs, including the brain, the prognosis of primary hypothermic arrest is better than in non-hypothermic arrest, especially if the initial insult did not involve asphyxia. There are many documented cases of survival after long periods of circulatory arrest [7-9].
You’re not dead until you’re warm and dead.
It is important to note that these survivors are usually young fit individuals.
The lowest recorded neurologically intact survival following hypothermic cardiac arrest (in this case associated with immersion) is 13.7 C .
As with any critically ill patient, assessment and management will occur simultaneously and follow an ABC approach.
In determining whether hypothermia is playing a significant role in your patients presentation consider:
- Where they were found
- The ambient temperature and weather conditions
- The patients clothing
- The patients age
- Co-morbid conditions and state of nutrition
- Alcohol and drug use
It is important (but difficult) to try to decide whether hypothermia is the primary problem or secondary to another pathology. If they are simply cold because they have been lying in a cold environment following a cardiac arrest from another cause it is doubtful aggressive treatment of hypothermia is going to make much difference.
You are dead if you’re cold because you’re dead
Getting as much information as possible about the circumstances is essential.
Furthermore it is important to bear in mind that while hypothermic arrest is survivable, being frozen is not. If the body is frozen (for example the chest is so stiff that cardiopulmonary resuscitation (CPR) cannot be carried out or there is ice in the pharynx) resuscitation attempts are futile .
Hypothermia can cause a low volume low pressure cardiac output with a blood pressure un-recordable by non-invasive techniques . To avoid missing a weak pulse, standard advanced life support (ALS) technique is modified, in that the pulse check should last 1 minute. There is a role for invasive monitoring and non-invasive Doppler or ultrasound techniques in assessing cardiac output but do not let prolonged attempts stop CPR.
If there is any doubt about finding a pulse, start CPR as for a normothermic patient.
Knowledge of core temperature confirms the diagnosis, directs future therapy and assesses the effectiveness of re-warming. Skin surface temperature measurement is of no use. Different sites of measurement may give different temperature readings and imply differing rates of re-warming so it is sensible to select one site and use it consistently. The rectal thermometer may lag behind the core temperature increases due to impaction into frozen faeces. The oesophageal temperature reading may be falsely elevated when concurrent heated, humidified oxygen is being administered. For these reasons, oesophageal or vascular temperature probes are recommended .
Blood should be taken for arterial blood gas, full blood count, electrolytes/urea/creatinine, calcium, magnesium, coagulation studies and amylase .
A variety of abnormalities are possible:
- Hypoglycaemia and obvious electrolyte disturbance should be treated .
- Hypokalaemia is very commonly observed due to intracellular potassium shifts. Treat if severe but use caution as the potassium will increase with re-warming.
- Hyperkalaemia my represent cell necrosis. Very high levels (>10 mmol/L) are predictive of death and contraindicate re-warming.
- Elevated creatinine is also seen and may not be a true reflection of renal function.
- A coagulopathy similar to disseminated intravascular coagulopathy is common.
- Loss of intravascular fluid causes haemocrit to rise around 2% for each 1 C fall in core temperature. Lower than expected haemocrit may suggest blood loss.
- Thrombocytopenia may occur due to marrow suppression or hepato-splenic sequestration .
Blood gas analysis can cause some confusion in hypothermia. Blood gas machines will warm the sample to 37 C and report those values. In vivo, the values will be different partly because of the increased solubility of gases as the temperature of a liquid falls.
It is possible to mathematically correct the values to the patients actual body temperature but interpretation of these values is very difficult partly because we have little idea of what the normal values for hypothermia should be.
To avoid confusion and ensure consistency use the uncorrected values: that is the values the machine reports after warming the blood to the standard 37 C.
Once cardiac arrest is confirmed:
- Apply basic life support (BLS) and ALS as described in the European Council Guidelines (ERC) guidelines .
- Re-warm the patient as fast as possible re-warming is the best anti-arrhythmic and inotrope in this situation.
- Look for, and treat, co-morbidities and consider alternative diagnoses. Why are they hypothermic? Is there a significant pathological process other than hypothermia? Remember the 4Hs and 4Ts.
- Anticipate a prolonged resuscitation that could require significant multi-specialty input.
Key changes to standard ALS management
Defibrillation and pacing
Defibrillation is less effective in hypothermia. For ventricular fibrillation/ventricular tachycardia (VF/VT) defibrillation may be tried up to three times but is then not tried until the temperature reaches 30 C.
Pacing is generally ineffective. Do not try it unless bradycardia persists when normothermia is reached. Sinus bradycardia may be a physiological response and is not treated specifically.
Normocapnia will be achieved at lower minute volumes than normal and hyperventilation risks cerebral hypoxia through reduction of cerebral blood flow [1-2,4,5]. Aim for a normal CO2 on ABG (not corrected for the patients temperature).
In a patient with a perfusing rhythm, intubation (or other rough handling of the patient) may precipitate VF, although the evidence for this is mainly animal- based and it is rare.
Drugs are often ineffective and will undergo reduced metabolism; so these are withheld below 30 C then given with twice the time interval between doses until either normothermia is approached (35 C) or circulation restored. So, adrenaline would be given about every 8-10 minutes once the core temperature is above 30 C.
Hypothermia causes muscular stiffness: chest compressions may be harder work than normal. Make sure that the individual performing chest compressions is swapped frequently. Trials are under way to clarify the role of mechanical compression devices.
Re-warming can be passive or active (internal or external). In a patient with a perfusing rhythm, passive measures may be sufficient.
Passive measures will only achieve rewarming in a patient with effective circulation.
- Remove wet clothing and dry the patient.
- Cover the patients head (theatres will usually stock insulated elasticated hats).
- Ensure a warm environment.
- Cover the patient with a blanket a forced-air warming blanket
In patients in cardiac arrest, active internal re-warming is mandatory. The therapeutic manoeuvre of choice is cardiopulmonary bypass (CPB) [4,5]. This may be instituted using femoro-femoral or aortic-right atrial bypass. The former is more suited to the emergency department (ED) environment as it is easier and faster to establish and prevents further heat loss, which will result from opening the chest . If femoro-femoral bypass is established, chest compressions should generally be continued to ensure acceptable flow through the cardio-pulmonary circulation.
Temperature increases of 1-2 C every 5 minutes may be achieved. Bypass should be instituted for patients in cardiac arrest, patients with haemodynamic instability and a core temperature below 32 C, frozen extremities and rhabdomyolysis with hyperkalaemia .
Note the size of the cardiopulmonary bypass equipment in Fig 1. It may be easier to take the patient to theatre than the machine to Resus.
Fig 1. Cardiopulmonary bypass equipment
Active warming without cardiopulmonary bypass
CPB simply will not be available to a lot of departments. If it is not an option (or while you are waiting for CPB) you should:
- Start forced-air warming (e.g. Bair Hugger )
- Warm IV fluids. Warm fluids will not do too much to increase core temperature but they will certainly prevent any further drop that might result from using room temperature infusions. Large volumes of fluid may be required.
- Warm inspired air to 40-46 C This requires a heater humidifier, which if not available in the ED, should be obtained from the ICU. If unavailable, ensure that a heat and moisture exchange filter is in place.
- Try gastric, bladder, peritoneal and/or pleural lavage Use warmed Hartmanns solution for gastric lavage via an NG tube or bladder lavage with a three-way catheter. Use warmed peritoneal dialysate (4L) via a peritoneal catheter. Leave the fluid in place for 10-15 minutes to allow heat exchange before draining out and replacing with fresh warm fluid. Pleural lavage requires an apical and basal chest drain on each side. 2L of warmed Hartmanns is infused into each hemithorax via the apical drain and then removed through the basal drain about 10-15 minutes later.
- Use a high volume renal haemofilter A high volume renal haemofilter may be a readily available tool in smaller hospitals as most intensive care units have the facility for renal support. It may be technically difficult in a patient with no blood pressure but if flow can be established the machine will generally allow heating of the blood as it flows through the circuit. The kit is also fairly easily moved to the patient. Intravascular temperature management devices exist too and some case reports have highlighted there use in re-warming hypothermic patients. The rate at which they achieve re-warming in severe hypothermia is unclear.
Table 1 shows approximate rewarming rates for various techniques (Reproduced from Emerg Med J2007;24:511-512 doi:10.1136/emj.2006.040261 with the permission of the BMJ publishing group)
|Classification||Method||Rewarming rate oC/h (approx rates)|
|External passive||Remove wet clothing, blankets, warm drinks||0.5 4|
|External active||Forced heated air||1-2|
|Internal active||Inhalation of warm air||0.7-1.2|
|Body lavage: peritoneal||1-3|
Return of spontaneous circulation
If you get a return of spontaneous circulation (ROSC), the ERC guidelines suggest following standard post-resuscitation care
The act of re-warming may precipitate cardiovascular collapse (re-warming shock) and rhythm change. The reasons are not completely understood but include electrolyte shifts, temperature fluctuation, intravascular depletion and drug therapy . This will, of course, not be an issue if bypass is used to re-warm. Patients should be adequately volume loaded to maximise cardiac output during the re-warming process. In the absence of cardiac output monitoring aim for a central venous pressure (CVP) of 8-12 cm H2O.
Morbidity and mortality
With CPB, survival rates in arrested patients approach 50%, even with a time to put bypass in place of around an hour [2,11,12]. The best figures are obtained from units dealing with cold mountain regions, as the population in question are younger and fitter than the urban population which typify the UK case mix. However, reasonable figures have also been reported from urban populations .
Common problems in the survivors include adult acute respiratory distress syndrome, acute renal failure and pneumonia . The majority of the survivors can be expected to have good neurological outcome . The majority of deaths are due to failure to wean the patients from bypass.
Patients with a potassium level over 10 mmol/L or severe traumatic injuries will not benefit from bypass. Similarly, patients with pre-existing cardiopulmonary, renal or neurological disorders require careful selection as they have a poorer prognosis [4,5].
In primary hypothermic cardiac arrest, death should not be confirmed until:
- The patient has been re-warmed
- Other unsurvivable injuries have been identified
- Re-warming has failed despite all available measures
Key Learning Points
- A few modifications are necessary to standard ALS and BLS [recommendation B, evidence 2c]
- It can be challenging to identify evidence of cardiac output in hypothermic patients. If in doubt CPR is instituted [recommendation B, evidence 2c]
- The key to treating hypothermic cardiac arrest is an accurate core temperature measurement [recommendation B, evidence 2c]
- The usual cardiac arrest drugs should not be given below 30 C core temperature (then at twice normal intervals until normothermia)
- Defibrillation is less effective in hypothermia: a maximum of 3 shocks should be tried at core temperatures below 30 C [recommendation B, evidence 2c]
- The most effective method of re-warming patients in cardiopulmonary arrest is using cardiopulmonary bypass. [recommendation A, evidence 1c]
- The prognosis in this group is better than other non-traumatic cardiopulmonary arrests. [recommendation B, evidence 2c]
- ERC Guidelines 2015: Section 4. Cardiac arrest in special circumstances
- Ulrich AS, Rathlev NK. Hypothermia and localized cold injuries. Emerg Med Clin N Am 2004;22:281-298.
- Chantler C, Kelly S. Deaths from hypothermia in England and Wales Office for National Statistics., ONS.
- Giesbrecht GG. Emergency treatment of hypothermia. EMA 2001;13:9-16.
- Lazer HL. The treatment of hypothermia. NEJM 1997 337;21:1545-1547.
- Mattu A, Brady W, Peron A. Electrocardiographic manifestations of hypothermia. Am J Emerg Med 2003;10:187-191.
- Farstad M, Andersen K, Koller M, et al. Re-warming from accidental hypothermia by extracorporal circulation. Eur J Cardiothorac surg. 2001;20:58-64.
- Schneider S. Hypothermia from recognition to re-warming. Emerg Med Rep. 1992;13:1-20
- Gilbert M, Busund R, Skagseth P, et al. Resuscitation from accidental hypothermia of 13.7 C with circulatory arrest. Lancet 2000;355:375-37.6
- Danzi D, Pozos R, Auerbach P, et al. Multicenter hypothermia survey. Ann Emerg Med 1987;16:1042-1055.
- Walpoth B, Walpoth-Asian B, Mattle H, et al. Outcome of survivors of accidental deep hypothermia and circulatory arrest treated with extracorporal blood warming. NEJM 1997;337:1500-1505.
- Vretenar DF, Urschel JD, Parrot JC, et al. Cardiopulmonary bypass resuscitation for accidental hypothermia. Ann Thorac surg 1994;58:895-898