Physiology

Humans have a high capacity to dissipate heat but a relatively poor capacity to increase heat production. Thus, humans rely heavily on environmental regulation in the form of clothing and warm shelter to maintain normal body temperature. The body can compensate to a great degree for mild hypothermia. The hypothalamus sends signals that produce cutaneous vasoconstriction, increased muscle tone, and increased metabolic rate. When muscle tone reaches a certain level, shivering thermogenesis begins. The clinical manifestations depend on the severity, acuity, and duration of temperature reduction; the patient’s age; premorbid conditions; and superimposed disease states. Children are more susceptible to hypothermia than adults because of a large surface area relative to body mass and less subcutaneous tissue. Neonates have a capacity for nonshivering thermogenesis primarily by metabolism of brown fat; however, this is at the cost of greatly increased oxygen consumption. Therefore, neonates are extremely sensitive to relatively minor deviations from a neutral thermal environment.

Central nervous system

Cerebral oxygen requirement decreases with cooling, and mentation is progressively impaired. Mild hypothermia may be associated with confusion, dysarthria, and impaired judgment. Deep tendon reflexes are depressed at core temperature below 32°C because of slowed peripheral nerve conduction. As body temperature drops, many victims no longer complain of cold. Shivering thermogenesis ceases at about 31°C. Pupillary responses decline and dilated unreactive pupils may be noted at temperatures below 30°C. Victims may experience hallucinations and sometimes paradoxically remove their clothes. The electroencephalogram shows abnormal activity at temperatures less than 32°C; at 20°C, the electroencephalogram may appear consistent with brain death.

Cardiovascular

The initial cardiovascular responses are vasoconstriction, tachycardia, and increased cardiac output. Further hypothermia results in decreased pacemaker and conduction velocity, causing bradycardia, heart block, and prolongation of PR, QRS, and QT intervals. The myocardium becomes irritable, and atrial fibrillation is common when core temperature is below 32°C. The risk of ventricular arrhythmias is substantial below 30°C. ^{, ,} A recent review of witnessed hypothermic cardiac arrest revealed that it occurred at a mean body temperature of 23.9°C and in no cases with a temperature >30°C. The electrocardiogram may show characteristic J or Osborne wave following the QRS complex ( Fig. 114.1 ). The presence of this wave is not pathognomonic for hypothermia and has no prognostic implications. ^, Myocardial contractility, cardiac output, and systemic blood pressure are often decreased dramatically in hypothermic victims. These changes may be persistent during and after rewarming. Hypovolemia due to cold-induced diuresis and capillary leakage may potentiate the problem. ^{, ,}

Characteristic J or Osborne wave of hypothermia closely follows QRS. It may be mistaken for a T wave with narrow QT interval if the true T wave is not appreciated. The slightly rounded peak distinguishes it from R′ of bundle branch block.

Respiratory

Hypothermia affects tissue oxygenation through several complex physiologic mechanisms. Initially, the respiratory rate may be increased. As hypothermia worsens, the respiratory center becomes depressed and hypoventilation causes carbon dioxide retention, although carbon dioxide production decreases with increasing hypothermia. Respiratory arrest is a late occurrence. Suppression of cough and mucociliary reflexes leads to atelectasis and pneumonia. Oxygen delivery to the tissues is further compromised by shifting of the oxyhemoglobin dissociation curve to the left. Blood gas analyzers warm blood to 37°C before analysis. In patients with hypothermia, arterial blood gases show higher oxygen and carbon dioxide levels and a lower pH than a patient’s actual values. The best approach to interpretation is to compare the uncorrected blood gas values with the normal values at 37°C (Alpha-Stat strategy). ^,

Renal

Renal injury may occur either because of hypothermia or during the rewarming process. The mechanisms involved in cold diuresis may include peripheral vasoconstriction and blunted response to antidiuretic hormone. Renal vasoconstriction and ischemia to the kidney may lead to oliguria and acute tubular necrosis in those with severe hypothermia. Progressive hypokalemia develops during hypothermia, probably because of the shifting of potassium from the extracellular to intracellular compartment, and significant hyperkalemia may develop during rewarming. ^, The electrocardiogram manifestation of hyperkalemia may be obscured or attenuated by hypothermia. Renal replacement therapy may be required for renal failure and has also been used for rewarming. ^,

Coagulation

Hypothermia inhibits the intrinsic and extrinsic pathways in the clotting process. The degree of coagulopathy, however, is often underestimated because dynamic coagulation tests are generally performed at 37°C in the laboratory. Bone marrow suppression and splenic sequestration can lead to thrombocytopenia, and platelets become dysfunctional. ^, This leads to increased bleeding tendency; the combination of hypothermia and trauma carries a grave prognosis. ^,

Treatment

The hallmark of rescue in all individuals with hypothermia is prevention of further heat loss, careful transport, and rewarming. ^, Wet clothes should be removed, and the individuals should be insulated and shielded from wind and cold. Paying special attention to the head and neck is important because radiant heat loss from those areas can be profound.

Collapse of deeply hypothermic victims around the time of rescue may be explained by further cooling, circulatory collapse due to hypovolemia, or arrhythmias, sometimes triggered by procedures such as central venous catheterization. ^{, ,} Sudden changes in hydrostatic conditions may contribute during extrication of victims from cold water. This has been attributed to sudden fall in blood pressure and inadequate coronary blood flow, precipitating ventricular fibrillation. ^, Victims should therefore be kept horizontal during rescue, if possible.

Rewarming

The rewarming of victims who are conscious with mild hypothermia and still able to shiver can be achieved with passive techniques (e.g., blankets, wool cap, warm shelter). Active external rewarming with chemical heat packs or forced warm air can be useful in the prehospital setting to prevent further heat loss in victims with moderate hypothermia. Some researchers claim that standing and exercising immediately after rescue may be hazardous, as this may accentuate core temperature afterdrop. The term core temperature afterdrop refers to a continued decrease in core temperature with potential for clinical deterioration of a victim after rescue. There are probably two mechanisms behind this. One is the inevitable equilibration of temperature between the periphery and core. It follows that the magnitude of afterdrop is greater if cooling is rapid because the temperature gradient between the surface and core is greater. The other mechanism is increased blood flow to the cold periphery and return of cold blood to the core (convection), possibly causing cardiovascular instability. Any intervention that causes increased blood flow to the periphery, including exercise, can potentially accentuate the afterdrop, but the significance of this in the clinical management of hypothermia victims is still being debated. ^{, , , ,} However, avoidance of excessive activity and abrupt movements of any hypothermia victim seems prudent. Attempts to rewarm the victim of hypothermia should not delay transport to the hospital. It is difficult to rewarm hypothermic patients during transport, and no specific recommendations can be given. The use of mechanical cardiopulmonary resuscitation (CPR) devices should be considered if body size fits the device, as this increases safety and may improve outcome. ^{, , ,} Fig. 114.2 shows a simple triage and management algorithm, a slight modification of a previously published algorithm.

Algorithm for management and triage of victims with accidental hypothermia. CPB, Cardiopulmonary bypass; CPR, cardiopulmonary resuscitation; ECMO, extracorporeal membrane oxygenation.

Several techniques have been used in-hospital for active rewarming of hemodynamically stable patients, but no controlled clinical trials exist in which rewarming methods are compared. This includes warming with forced air, warm humidified gases, warmed IV fluids (up to 42°C), and lavage of gastric, peritoneal, pleural, or bladder cavities with warmed fluids (40°C). There are many reports of successful use of different techniques in children with severe hypothermia, even with cardiac arrest. Thoracic lavage using two large chest tubes can accomplish a temperature increase of 3°C to 6°C/h, and warming with forced air (Bair Hugger) can also be quite effective (approximately 2°C/h). However, victims with a core temperature less than 28°C, hemodynamic instability, or cardiac arrest should be taken directly to a center capable of extracorporeal rewarming, unless trauma or other coexisting conditions mandate transport to a closer hospital. ^{, , , , ,} Extracorporeal rewarming may be undertaken with traditional cardiopulmonary bypass (CPB) or venoarterial extracorporeal membrane oxygenation (VA-ECMO). The latter may be preferred, as some patients will require prolonged support for hemodynamic instability or pulmonary edema. Compared with CPB, ECMO allows for lower levels of anticoagulation and may be associated with better outcome. ^{, ,} Slower rate of extracorporeal rewarming may be associated with better neurologic outcome, possibly due to lower risk of cerebral temperature overshoot. However, the optimal rate of rewarming remains unknown. Clinical judgment must be exercised, and it should be kept in mind that children with accidental hypothermia have tremendous potential for good outcome despite a catastrophic presentation. A reasonable approach, in most cases, is to resuscitate and rewarm the child aggressively until the core temperature reaches 32°C and continue resuscitation for 30 more minutes. At that point, in the absence of signs of life or response to aggressive life support measures, termination of resuscitation may be indicated.

There are no formal studies of care after return of spontaneous circulation in hypothermic victims; management should follow usual standards. Inotropic support with levosimendan or milrinone should be considered rather than relying on β-adrenergic agents, which may have reduced effects during hypothermia and rewarming. ^{, ,} Therapeutic hypothermia for at least 24 hours may be indicated. Continuous electroencephalogram, early neuroimaging, and neuroprotective treatment strategies may help in estimating brain injury severity, diagnosing seizure activity, and detecting injury requiring neurosurgical intervention.

Outcome

It is difficult to predict outcome of accidental hypothermia victims, and our knowledge about optimal management is limited. It may help to apply the concept of sufficient-cause model to elucidate this problem. Accidental hypothermia is frequently the cause of neurologic injury and death but by itself not a sufficient cause, as demonstrated by the fact that good neurologic recovery is possible after prolonged hypothermic cardiac arrest in adults with initial core temperature as low as 13.7°C and CPR up to almost 7 hours. ^, There are also many reports of successful resuscitation of children and adolescents with severe hypothermia. ^{, , , , , ,} The term accidental hypothermia is too imprecise. One must specify the type of accidental hypothermia (e.g., exposure, immersion, or submersion), severity (e.g., cardiac arrest or hemodynamic instability), duration, age and condition of the victim, whether there is comorbidity, and methods of resuscitation and rewarming. However we define severe accidental hypothermia, it will not cause neurologic injury or death in everyone exposed, within some limits. So, who is less susceptible to the effect of hypothermia and what are the limits? We know that good neurologic outcome is unlikely when asphyxia precedes hypothermia. ^{, , , ,} Management of snow avalanche victims is outside the scope of this chapter, but asphyxia often precedes hypothermia in that setting. Avalanche victims found without an air pocket have poor prognosis, whereas extracorporeal rewarming has been successful in cases in which asphyxia was not the primary factor. ^{, , , , ,} There are many reports of dramatic recovery after prolonged cold-water submersion of pediatric patients. ^{, , , ,} Cases of good outcome in drowning victims are generally associated with water temperatures at or near freezing and hypothermia on arrival to the hospital. Therefore, it is reasonable to assume that good outcome in these cases must be associated with cerebral hypothermia. ^{, , ,} Rapid cooling of the pediatric body, owing to large surface area relative to body mass and little subcutaneous tissue, may offer protection of the brain against anoxia. ^, Furthermore, it is thought that rapid cooling of the brain may largely depend on repeated aspiration of cold water into the lungs. ^{, , , ,} However, reviews from Finland and Canada did not find an association between age or water temperature and outcome, ^{, ,} and a recent meta-analysis including 658 cases of accidental hypothermia requiring extracorporeal support found that age, water temperature, and initial core temperature were not independently associated with survival with good neurologic outcome. Severe coagulopathy, acidosis, and hyperkalemia (>10 mmol/L) are among factors that have long been associated with poor outcome. ^{, , ,} Initial lactate, pH, serum potassium, cardiac rhythm, and asphyxiation were quite significant predictors in a univariate model. However, of those, only potassium and asphyxiation remained significant in a multivariate model. Even so, this does not mean that variables left out of the multivariate model are useless in predicting survival, as two highly correlated variables will effectively cancel each other out when modelled together. It is difficult to provide cut-off values for potassium, exemplified by the fact that a 31-month-old girl survived extreme hypothermia after exposure to cold weather despite having serum potassium of 11.8 mmol/L on admission, which is the highest reported value in a survivor. Another remarkable case is a 7-year-old drowning victim with serum potassium of 11.3 who had an excellent outcome. However, serum potassium level greater than 12 mmol/L is probably an indication to terminate resuscitation. ^, It is important to note that the duration of CPR is not clearly associated with outcome, and there are indeed reports of patients with good recovery after many hours of CPR before successful rewarming. ^{, ,} Other factors that may be associated with better outcome include female gender, possibly explained by higher body fat percentage, and slow rewarming rate on bypass. ^,

Reported mortality and the need for prolonged organ support for victims with severe hypothermia and circulatory arrest treated with CPB or VA-ECMO varies widely. ^{, , , , , , , , , ,} Pooled data from 44 observational studies and 40 case reports (n = 658) revealed a mean survival rate of 46% (n = 303) and 87.5% (n = 265) of survivors had good neurologic outcome. Asphyxiation was presumed in 42% (n = 279) of cases and only 12.2% (n = 34) of those had good neurologic outcome. These data indicate that likelihood for survival with good neurologic outcome is five times higher if victims are not asphyxiated (231 of 379 vs. 34 of 279). However, we still have much to learn about susceptibility to hypothermia and limits of survivability, and we clearly need better tools for prognostication and brain resuscitation. A recent promising study found that in hypothermia victims remaining unconscious the day after admission, the biomarkers neuron-specific enolase and S100 calcium-binding protein were strong predictors of mortality and poor neurologic outcome. More quality research on accidental hypothermia is necessary. It is hoped that data from the International Hypothermia Registry, which receives reports from more than 50 centers worldwide, may increase our insight into this important health problem. In the meantime, available literature suggests that children with severe hypothermia and absent circulation, even after submersion accidents and undergoing extracorporeal rewarming, can have a high survival rate with good neurologic function. In nonresponders, support can usually be withdrawn within a few days. ^,