Hypothermia and Hyperthermia: Treating Temperature Disruptions

Authors: David Ledrick, MD, MEd, Assistant Residency Director, Student Clerkship Director, St. Vincent Mercy Medical Center, Toledo, OH; and Michael J. Caudell, MD, Clinical Assistant Professor, Department of Emergency Medicine, Medical College of Georgia, Augusta, GA.

Peer Reviewer: Mary Jo Bowman, MD, FAAP, FCP, Associate Professor of Clinical Pediatrics, Pediatric Emergency Medicine Fellowship Director, Nationwide Children's Hospital, Columbus, OH.


Victims of environmental hypo- or hyperthermia may present seasonally and the number of patients any given practitioner will treat in a year may vary geographically, but no region of the country is immune to either presentation. Physicians in the emergency department will be responsible for directing the treatment of these physiologic disruptions through the patient's entire acute phase and often to resolution. Whether practicing in a predominately warm or cool climate, it is important for the emergency department physician to have an updated understanding of both entities.

For the purposes of this discussion, we will distinguish between individuals who become hypo- or hyperthermic as the result of a disease process and those who were subjected to extreme physiologic stress. In the former case, the diagnosis and treatment of the underlying cause must be initiated and/or continued, along with the treatment of the temperature disruption. This article will be confined primarily to work concerned with environmental exposure.

What are the effects of hypothermia on cognitive and physical performance?

Source: O'Brien C, Tharion WJ, Sils IV, et al. Cognitive, psychomotor, and physical performance in cold air after cooling by exercise in cold weather. Aviat Space Environ Med 2007;78:568-573.

To answer the question of how mild hypothermia affects human performance both from a cognitive and motor performance perspective, researchers enrolled healthy volunteers to lower their core body temperatures by exercising in cold water. Specifically, the volunteers walked on a treadmill at variable speeds and were immersed to their waists or chests in 10°C or 15°C water until a core temp of 35.5°C was reached. They then performed a series of cognitive and physical tests. Their performance on the tests was compared to their performance without hypothermia. To determine the effects of local tissue chilling, both upper and lower body physical testing was done.

The 10 soldiers were physically fit men between the ages of 18 and 22. They did not experience any statistically significant changes on the cognitive performance tests, which included assessment of reaction time and logical reasoning; nor did they show a difference in performance on the psychomotor testing, which included marksmanship and weapon assembly. On physical performance testing they experienced a degradation of lower body performance but upper body testing was unaffected.


The chosen conditions and testing were intended for the military but could easily be applied to EMS providers performing in cold environments. The cognitive testing would be directly applicable to the mental process of assessing a patient or situation, and the psychomotor testing would be similar to performing many field procedures. The most significant departure from EMS providers may come from the physical characteristics of the test subjects, most noticeably, their age. One would hardly want to look at this article and decide that EMS providers can be safely chilled, but it does offer some insight into the ability to train and operate in cold weather conditions.

How effective are commercially available fluid warmers in hospital, prehospital environments?

Source: Dubick MA, Brooks DE, Macaitis JM, et al. Evaluation of commercially available fluid-warming devices for use in forward surgical and combat areas. Mil Med 2005;170:76-82.

This study was intended to evaluate the effectiveness of four commercially available fluid warmers. It tested the ability of the devices to warm fluids not only from room temperature but also from 4-7°C, as these temperatures are likely to be encountered in the field. The Level 1, FMS 2000, Thermal Angel and Ranger were tested at high- and low-flow rates to determine their ability to reach a standard of 32-35°C. All devices were tested with cold and room temperature 6% hetastarch, lactated Ringer's solution (LR), and cold packed red blood cells. Additional attention was paid to set-up times, the instrument's steady state temperature, the time to reach that temperature, and the ability of the device to warm fluids above 35°C.

The Thermal Angel and Ranger are lightweight, portable devices that depend on pressure bags to achieve flow. The Thermal Angel was the least expensive and unable to warm cold LR but performed adequately on the other fluids, heating to a temperature of 32°C. The Ranger warmed all fluids at both rates. The Level 1 and FMS 2000 are designed to be attached to poles and require an external power source. The FMS 2000 is the most expensive and also has the most safety features. It was the fastest device and maintained the highest temperatures (37-39°C). The less expensive Level 1 demonstrated comparable performance to the FMS 2000 but only had one flow rate and could not warm cold LR to > 35°C.


This article addresses one aspect of the care of hypothermic patients. It was performed independently of any manufacturer and did an excellent job of reviewing the capabilities of each of these devices. Warming intravenous fluids (IVF) alone is hardly treatment for hypothermia, but hypothermic patients with significant dehydration may require large volumes of fluid and their condition can be worsened or treatment compromised if this detail is overlooked.

Where should pulse oximeter sensors be placed on hypo- thermic patients?

Source: MacLeod DB, Cortinez LI Keifer JC, et al. The desaturation response time of finger pulse oximeters during mild hypothermia. Anaesthesia 2005;60:65-71.

Ten healthy male volunteers underwent a series of hypoxic challenges, each lasting 3 minutes, during which pulse oximeter readings were taken on their foreheads, earlobes, and fingers. Pulse oximetry values were compared to the saturations obtained via arterial blood gas measurements. The subjects had baseline temperatures established using aural canal probes. Baseline oxygen saturations also were established. Surface cooling of the subjects was then performed using a cooling mattress set at 14°C. Temperatures were maintained above 36°C, and after 30 minutes the first hypoxic challenge was performed. Then the second cooling phase was initiated using 2 liters of cold (4°C) lactated Ringer's solution infused over 30 minutes to establish a temperature between 35 and 36°C. There was an average drop of 0.8°C from baseline values. Another hypoxic challenge was performed. Response times of the oximeters to desaturation were measured and it was determined that mild hypothermia significantly prolongs the response times of finger oximeters, while the response times of the forehead and ear oximeters were not affected. The authors report that forehead and ear oximeters may provide better monitoring sites for patients with mild hypothermia.


Although there are only a small number of subjects, this study attempts to validate what many practitioners have been attempting to do when they are unable to obtain a pulse oximeter reading from their patient's finger. It seems intuitive that the vasoconstriction caused by hypothermia will affect the reading of the pulse oximeter sensor; of course, studies have demonstrated this effect. It also has been shown that there is less vasoconstriction in the vessels of the forehead and earlobe; this allows for better readings at each of these sites. So should clinicians consider initial pulse oximeter placement on the forehead instead of the finger in hypothermic patients? The answer is yes. However, it is important to note that the sensors used in this study are manufactured specifically for use on the forehead and are not simply digital sensors placed elsewhere on the patients. Caution should be used when interpreting values obtained when using the digital sensors on the forehead because of prior studies reporting unreliability in detecting hypoxemia when doing so. This study supports the placement of a pulse oximeter sensor on the forehead or earlobe in those patients with hypothermia, but only with sensors specifically manufactured for use on those sites.

How are hypothermic patients with ventricular fibrillation best treated? What is the significance of an Osborn wave?

Sources: Clift J, Munro-Davies L. Best evidence topic report. Is defibrillation effective in accidental severe hypothermia in adults? Emerg Med J 2007;24:50-51; and Aslan S, Erdem AF, Uzkeser M, et al. The Osborn wave in accidental hypothermia. J Emerg Med 2007;32:271-273.

The article published by clift and Munro-Davies is a "best evidence" summary. It found 385 papers on the ventricular defibrillation of hypothermic patients, of which 17 were felt to be relevant. All studies were case series or case reports and included a total of 26 patients. These patients all were successfully defibrillated with temperatures as low as 24°C and as high as 35°C, but a number of them required rewarming before the defibrillation was successful. At temperatures below 30°C, there was a 55% success rate in this series. The authors point out that there is no higher level evidence, but suggest that defibrillation attempts should be made even at low temperatures.

In regard to the Osborne wave, our search of the literature found three articles published in the last 24 months, all of which are reviews or case studies. The article by Aslan and colleagues was a Turkish case study of a hypothermic individual who also presented with a subarachnoid hemorrhage (SAH). In their discussion of the case, the authors point out that the differential diagnosis of Osborn waves includes SAH, ischemic heart disease, Chagas disease, and Brugada syndrome. The discussion goes on to point out the need to look for concomitant disease, as the authors initially assumed their patient was hypothermic purely as a result of an environmental stress.


It is recognized that both atrial and ventricular dysrhythmias are common as the patient becomes progressively more hypothermic and that these disturbances are best treated with rewarming rather than medications. Many clinicians may feel comfortable watching a slow atrial fibrillation during rewarming, but ventricular fibrillation (VF) produces far more anxiety. The American Heart Association points out in their 2005 ACLS guidelines that a cold heart may not respond to standard therapy but do encourage an initial attempt at defibrillation.1 They go on to state it may be ineffective. The guidelines note that repeated doses of cardioactive medications can cause accumulation to toxic levels in the peripheral circulation while doing nothing for treatment. Ultimately, patient outcomes seem unaffected by the rhythm disturbance alone and are rather tied to the overall treatment of the cold exposure.

In our own literature search we reviewed one of the papers included by Clift and Munro-Davies in which the patient was initially refractory to cardioversion but was successfully defibrillated after a period of warming with cardiopulmonary bypass. In this 2004 paper published in Perfusion, the authors discuss the neuroprotective role of hypothermia that has resulted in good patient outcomes even after prolonged resuscitation.2 The conclusion of this work is in agreement with others that state it is worth the effort to try to treat severely hypothermic patients. To know the true answer to the best treatment of VF in hypothermia, a reasonably large number of patients would need to be enrolled prospectively into separate treatment arms. This study is unlikely to ever occur in humans.

In 1999, Vassallo and researchers published a prospective evaluation of the electrocardiographic (ECG) manifestations of hypothermia in 43 patients who generated 100 ECGs over their course of treatment.3 None were in VF, and most were in a sinus rhythm. Of rhythm abnormalities, atrial fibrillation (AF) was the most common. Only four of the patients died. They had all presented with temperatures below 30°C and their demise occurred many hours to days after their initial presentations. This study also found that Osborne waves were present in all patients with a temperature less than 30.5°C, and that the size of the wave correlated with the degree of hypothermia but did not predict outcome or have an association with laboratory abnormalities.

In answering the questions posed at the start of this review, the simplified answers would be that the primary treatment of VF is rewarming and it is appropriate to attempt cardioversion but not repeated doses of cardiac medications. Patients with VF refractory to initial shocks may still have a good outcome provided aggressive warming and ongoing resuscitation occur. Patient outcomes will be affected more by the hypothermia and its mechanism than the rhythm disturbance, which is more likely to be AF than VF. Osborne waves may be present in mild hypothermia and are nearly universal in moderate to severe hypothermia. They are not ominous in themselves and are a sign of the underlying cause only.


1. ECG Committee. Subcommittees and Task Forces of the American Heart Association. Circulation 2005; 112(24 Suppl):IV1-203.

2. Mulpur AK, et al. Perfusion 2004;19:311-314.

3. Vassallo SU, et al. Acad Emerg Med 1999;6:1121-1126.

How safe, effective is warm water pleural lavage?

Source: Kjaergaard B, Bach P. Warming of patients with accidental hypothermia using warm water pleural lavage. Resuscitation 2006;68:203-207.

Kjaergaard and bach report a case series of five patients suffering from accidental hypothermia. All had bladder temperatures 32°C, were unconscious, intubated and ventilated, and had spontaneous heart rhythms. Each patient had bilateral pleural drains placed and pleural lavage performed. This procedure consisted of unilateral infusion of approximately 500 mL of isotonic saline at 40°C into the pleural space. The tubes were clamped for approximately 2 minutes, the fluid was drained, and the procedure was then performed on the contralateral side. The procedure was continued until bladder temperature reached 35°C. This took 3-5 hours and 32-102 liters of saline. No complications were reported. All patients were discharged to their own homes neurologically intact.


This case series reports great success with the use of pleural lavage in hypothermic patients. The patients' actual measured bladder temperatures ranged from 25.3 to 30.2°C. Three of the five patients involved had complicating factors of near drowning, multiple trauma, and drug overdose. All patients had intact circulation and were receiving ventilatory support. Although the potential for worsening is always a consideration, would these patients have continued to improve without such invasive treatment? That question is impossible to answer, but the use of pleural lavage as reported here is a viable mode of therapy that warrants further investigation and comparison to other modes of therapy, especially in patients without an intact circulation.

A remarkable side note to this study is the authors' unique regional approach to hypothermia. They report the formation of an organization in Denmark that utilizes an emergency response team dispatched to other hospitals specifically for the treatment of patients with accidental hypothermia.

Are new methods of rewarming available?

Source: Willekes T, Naunheim R, et al. A novel method of intravascular temperature modulation to treat severe hypothermia. Emerg Med J 2006;23:e56.

The icy™ catheter was developed to provide core temperature cooling in patients who have suffered cardiac arrest or undergone neurosurgery. In this case report, the Icy catheter was used to rewarm a hypothermic patient who had an initial rectal temperature of 27.9°C. The catheter was placed in the femoral vein, and rewarming was initiated at a rate of 0.74°C/h. The patient did not experience core afterdrop, and the goal temperature of 37.5°C was reached within 12 hours.


As the title of the article states, this is a "novel" method of rewarming a hypothermic patient. The Icy catheter (Alsius Corporation) was developed for use in the induction of hypothermia and subsequent rewarming in a specific population of patients, mainly unconscious patients with spontaneous circulation after out-of-hospital cardiac arrest when the initial rhythm was ventricular fibrillation. The Icy catheter is a multilumen central venous catheter intended to be placed in the femoral vein. One lumen of the catheter is a standard guide-wire lumen that is used for primary infusions. The other two lumens form a closed-loop heat-exchange unit, with an inflow and an outflow lumen through which heated (or chilled) saline is circulated. This fluid is not infused into the patient, but the heating/cooling membranes of the distal portion of the catheter interface with the patient's circulating blood, thus producing the desired temperature modulating effect. The catheter is used in conjunction with the Alsius CoolGard® 3000, which is an integrated system consisting of a temperature monitor, a temperature controller unit, a heat exchanger unit, and a roller pump. This system was not developed for use in those patients who had succumbed to environmental or accidental hypothermia, but the report of the absence of core afterdrop could make this a promising indication for this device. As in the pleural lavage case series noted above, this patient was intubated and had an intact circulation. It is interesting to compare the rate of rewarming using pleural lavage (3-5 hours) and the rate of rewarming in this case (12 hours), but no specific recommendations can be made as a result of this comparison.

What are best methods for cooling an exertional heatstroke patient?

Source: Smith JE. Cooling methods used in the treatment of exertional heat illness. Br J Sports Med 2005;39:503-507.

This article attempted to resolve the question of the best way to cool an exertional heat illness patient. A thorough review of the literature, focusing on total body immersion, evaporative methods, immersion of hands and forearms, ice packs in the groin and axillae, invasive techniques, and chemical assistance with dantrolene was conducted. A total of 17 papers were included.

The literature supports total body immersion as the fastest cooling technique, achieving rates between 0.1 and 0.2°C per minute and faster with colder water. This technique is not always well tolerated by the patient and presents some logistical difficulties. There are no studies randomizing patients to an evaporative treatment arm and an immersion arm. The published rates of cooling by evaporation are more variable, with rates between 0.05 and 0.3°C per minute. The higher rates of cooling were obtained in healthy volunteers and not achieved in patients suffering from heatstroke. Cooling with more invasive means (cold gastric, peritoneal or pleural lavage) has been studied in animals but there is limited experience in humans. Using ice packs in the axillae and groin or immersion of the extremities may be a convenient field technique but when used alone do not achieve results as fast as evaporative methods; however, the two methods, when used together, are effective. Dantrolene has been used experimentally. One study did show a difference in cooling times but not in patient outcomes. It was flawed by being non-blinded, having small numbers, using different cooling techniques with the subjects, and poor randomization. Another study showed no difference at all between the groups.


This article is useful as a starting point when reading about the treatment of heatstroke. It does not present any new information, but rather gives an overview of the best evidence. When looking at the included articles, it is clearly evident that relatively little research on humans has been done. The 17 studies have enrolled only a combined 272 patients, which is a remarkably small number. In the case of severe disease, presentations can be highly variable and it is difficult for a single center to accumulate data on a large number of individuals. To date, there has not been a large, multi-center co-operative effort. There are some excellent studies on healthy volunteers but there are obvious difficulties in extrapolating the treatment of a 22-year-old marine volunteer to a 65-year-old diabetic man without air conditioning.

How quickly can a hyperthermic patient be cooled?

Source: Proulx CI, Ducharme MB, Kenny GP. Safe cooling limits from exercise-induced hyperthermia. Eur J Appl Physiol 2006;96:434-445.

In this study, subjects had hyperthermia induced through a combination of increased ambient temperature and exercise. The subjects were fit individuals in their early 20s. The four men and three women had their core body temperatures monitored by esophageal, rectal, and aural canal probes. The subjects entered a thermal chamber (ambient temperature, 38.8 ± 0.6°C) when normothermic, and exercised on a treadmill until their rectal temperatures were 39.8 ± 0.2°C. They were then rapidly immersed in a circulated cold water bath at either 2, 8, 14, or 20°C until the heat gained during the exercise was lost. Immersion was discontinued when the rectal temperature reached 37.5°C. Each individual participated in a trial at each water bath temperature. During submersion, the subjects wore neoprene mitts and socks to prevent localized cold injury. The heat that was gained during the exercise was determined to have been eliminated in 5.4 ± 1.5, 7.9 ± 2.9, 10.4 ± 3.8, and 13.1 ± 2.8 minutes after immersion in 2, 8, 14, and 20°C water, respectively. No complications were reported. The subjects reported no great discomfort with immersion in the colder water.


It has been postulated that cold water immersion for the treatment of hyperthermia would likely produce such a significant constriction of cutaneous vasculature that heat loss through convection and conduction would be significantly decreased. This study reports no such impediment to heat loss, and showed rapid rates of temperature reduction. This finding is consistent with other previous studies, particularly those from military sources reporting great success with this cooling technique. The main limitation of this method is its practicality given the usual lack of appropriate equipment to perform cold water immersion. Also, the study subjects were lean, otherwise healthy, and able to cooperate physically with the methods used in the study protocol. For a population of physically fit and cooperative individuals who are exercising in a hot environment, this may be an ideal method of cooling, particularly if health care providers can plan ahead with the necessary materials and equipment. In the typical emergency department setting, these results may be less applicable. Despite its small study size, this study reached valid conclusions but with the caveat that the results may be limited to a specific setting and patient population.

To what temperature should hyperthermic patients be cooled?

Source: Proulx CI, Ducharme MB, Kenny GP. Safe Cooling limits from exercise-induced hyperthermia. Eur J Appl Physiol 2006;96:434-445.

In the study previously discussed, Proulx and colleagues also address the issue of when to stop cooling a hyperthermic patient. After the aforementioned cold water immersion was discontinued (at rectal temperatures of 37.5°C), subjects were observed for a period of 30 minutes in an environment with an ambient temperature of approximately 26°C. Rectal temperatures were continually measured. A decrease in rectal temperature was noted after the subjects were removed from the cold water bath, with a nadir reached prior to the end of the observation period. This occurred at all water temperatures. The lowest of these temperatures was 35.7 ± 0.74°C and occurred approximately 20 minutes after immersion in the 2°C water bath was discontinued. The authors' final recommendations support cooling to a rectal temperature between 37.8°C (if immersed in water > 10°C) and 38.6°C (if immersed in water < 10°C).


There is a significant risk of inducing hypothermia when using cold water immersion. Once removed from the cooling stimulus, the temperature of the patient does continue to decrease. This iatrogenic hypothermia could potentially result in life-threatening cardiac dysrhythmias and/or dysfunction of hypothalamic temperature regulation. There have been other reports with recommendations for temperatures at which clinicians should cease cooling efforts, with temperatures ranging from 37.5 to 38.9°C. This study attempts to address the continued cooling effects of cold water immersion and to set specific temperature readings at which one should remove the patient from the cold water bath based on the rate of heat lost at each water bath temperature. The same limitations previously noted also apply to this aspect of the study, namely the small study size, the population of study subjects involved, and the hyperthermia being induced by exertion. These authors report clear and sensible safety guidelines to prevent the potential complication of hypothermia when using cold water immersion to treat exercise-induced hyperthermia.

Are there other options for cooling hyperthermic patients, especially in prehospital settings?

Source: Giesbrecht GG, Jamieson C, Cahill F. Cooling hyperthermic firefighters by immersing forearms and hands in 10 degrees C and 20 degrees C water. Aviat Space Eviron Med 2007;78:561-567.

Six firefighters who were fully dressed in "turn-out gear" (including a breathing apparatus) participated in five experimental trials in which they performed three sets of exercises in a 40°C environmental chamber. There was a rest interval of 20 minutes between each 20-minute set, during which subjects were allowed to exit the environmental chamber and remove their turn-out gear. Ambient temperature during rest periods was 21°C. Subjects' temperatures were measured with a probe that was inserted into the auditory canal. Each rest period involved either no active cooling (control) or one of four different randomly assigned cooling conditions: 1) no hand or forearm immersion (control); 2) hand-only immersion in 20°C water; 3) hand and forearm immersion in 20°C water; 4) hand-only immersion in 10°C water; or 5) hand and forearm immersion in 10°C water.

The amount and rate of cooling were measured. There was no significant difference between hand-only immersion in 20°C water and the control group. However, there was a decrease in measured temperatures in the other three groups, with hand and forearm immersion in 10°C water showing the greatest effect.


Hand and forearm immersion is an appealing alternative to total body immersion for treating hyperthermia. It requires less equipment and is more feasible in both the hospital and pre-hospital settings. The necessary degree of patient cooperation would be much less compared to that required for total body immersion. It is important to note that the small number of involved subjects neither had temperatures above 39°C or 2°C above their baseline nor reported signs or symptoms of heat illness. The accuracy with which auditory canal temperature reflects true core temperature is controversial; however, this study measured both absolute aural canal temperature and the change in temperature compared to baseline. Hand and forearm immersion in cold water is a significantly easier means of cooling than total body immersion, but our literature search could not find a direct comparison between the two methods. Hand and forearm immersion in cold water may be very valuable in the prehospital treatment of heat illness given the ease with which it can be performed. This technique also may be useful for the prevention of environmental hyperthermia.

What outcome can be expected for severely hyperthermic patients? What are poor outcome predictors?

Source: LoVecchio F, Pizon AF, Berrett C. Outcomes after environmental hyperthermia. Am J Emerg Med 2007;25:442-444.

This study is a retrospective chart review of patients presenting to Good Samaritan Medical Center and Maricopa Medical Center over a two-year period. Patients with a diagnosis of heat cramps, heat stroke, or heat exhaustion were included. Data were abstracted independently by two trained research assistants who were blinded to the study's purpose.

Ultimately, 52 patients met the inclusion criteria, with a mean presenting body temperature of 40.6°C and a mean daily temperature of 39.8°C. Altered mentation was common, laboratory abnormalities occurred in about one-third to one-half of the patients, and alcohol was present in about one-third of the patients. Hypoglycemia was rare. About one-third of the patients died, and odds ratios (OR) for death were calculated. Hypotension (OR 17, 95% CI, 3.8-75.8) and a Glasgow Coma Score (GCS) of less than 12 (OR 3.39, CI 0.91-12.6) were the best predictors. A presenting body temperature of > 42°C also was associated with death (OR 3.28, CI 0.91-11.8), but the lab parameters were not clinically significant.


Retrospective reviews are an appropriate means of studying patient characteristics and outcomes provided data are available. In this case, the authors were trying to do exactly that; they point out that few studies have described the clinical features of fatal and near fatal heatstroke and none have analyzed patient characteristics. Prospective enrollment of patients may increase the sample size but are unlikely to lead to different conclusions.

While the original question for this article may have been "what outcome can I expect for my hyperthermic patient?", the rejoinder might easily be, "how is knowing the outcome going to change anything I do acutely?" In the end, the principles of initial treatment may not be significantly altered by knowing that a patient has a poor prognosis, but a good physician needs to know what to expect for any disease process.


In the case of a true environmental emergency, the circumstances regarding any particular patient are highly variable; therefore, most conclusions regarding the presentation of heat or cold exposures are drawn from smaller studies. Unlike research into the treatment of myocardial infarction, where we argue over what the data mean, the arguments here begin with whether the data are applicable to a given patient.

Despite these limitations, the studies that have been done allow us to draw some conclusions and suggest areas for further study. What would be the result of a large prospective study comparing immersion to evaporative cooling? Could the Icy catheter be used in the treatment of accidental hypothermia? Is it feasible to institute a system approach to hypothermia much as they do in Denmark?

Individual physicians will need to make decisions about how lessons from the current literature change their practice in regard to a specific patient presentation. Emergency physicians must take the lead in treating individual patients with environmental illnesses and also advocate for research on hypothermia/hyperthermia.