Journal Logo


Cold Exposure Injuries to the Extremities

Golant, Alexander MD; Nord, Russell M. MD; Paksima, Nader DO, MPH; Posner, Martin A. MD

Author Information
Journal of the American Academy of Orthopaedic Surgeons: December 2008 - Volume 16 - Issue 12 - p 704-715
  • Free


The early historical accounts of frostbite are associated with military conflicts.1-3 Describing the military campaigns of the ancient Greeks, Xenophon wrote about the loss of nearly half of Sparta's soldiers to cold injuries in their retreat across the Carduchian Mountains.1,2 In the early 1700s, the Swedish army in an invasion of Russia sustained similar cold injuries that were so numerous that they were a major factor in the Swedes' eventual defeat.3 During the American Revolution, approximately 10 % of the Continental Army had cold exposure.1 A similar fate awaited Napoleon's army in the Franco-Russian War in its retreat from Moscow in 1812. That episode was significant because Baron Dominique Larrey, Napoleon's surgeon, provided the first description of the pathophysiology of frostbite injuries as well as his recommended treatment. He noted that the frozen feet of soldiers would become swollen and blistered when they were warmed at campfires. Larrey was therefore opposed to rapid application of heat and instead recommended gradual rewarming by rubbing the limb in wet snow, a treatment that was generally accepted until the mid 1900s.1,2 Frostbite injuries were also documented in other 19th-century military conflicts: the Crimean War (1853) and the American Civil War (1861-1865), in which more than 15,000 cold exposure injuries were recorded.4

In the early 1900s, a new type of cold injury was noted and referred to as “trench foot” because it occurred in soldiers whose feet were submerged in rain-filled trenches for days and sometimes weeks during the infamous trench warfare battles of World War I. Unlike frostbite, these injuries resulted from prolonged exposure to temperatures above freezing. Similar injuries occurred during World War II and were also noted in sailors whose feet were immersed in sea water as they sat in life rafts after their ships had been torpedoed and sunk; the name of the condition was changed to “immersion foot.”

World War II marked a turning point in cold exposure injuries because, until then, these injuries had been confined to military personnel on the ground. In 1943, high-altitude frostbite was described in bomber aircraft crews. These injuries were so common that they exceeded the number of injuries sustained from hostile fire.2 Cold exposure injuries, particularly immersion foot, were subsequently encountered during the Korean War and during the Vietnam conflict; in warmer regions, these injuries were referred to as “tropical immersion foot” because they occurred in water that was relatively warm.1 Cold exposure injuries arose as a military problem as recently as 1982, in both British and Argentinean troops during the Malvinas (Falklands) Islands conflict.2


Although cold exposure injuries historically were reported to occur primarily in military personnel,5,6 in the past 20 to 30 years, they have been more commonly encountered in the civilian population.7-10 Homeless persons living outside in winter months are at risk for hypothermia and/or frostbite, as are those whose work requires them to spend prolonged periods outside in freezing temperatures.2,7-12 These workers are generally men, and for that reason, frostbite occurs far more commonly in males than females (10:1); the most commonly affected age group is 30 to 50 years. The elderly and the young do not appear to be at greater risk, provided they are healthy. The incidence of frostbite is especially high among those who live in northern latitudes. In a study of approximately 6,000 Finnish men aged 17 to 30 years entering military service, 44% had sustained some degree of frostbite at least once.6

Frostbite is also seen with increasing frequency in individuals participating in outdoor winter sports, such as skiing and highaltitude climbing and hiking, because of such factors as inadequate protective clothing or equipment and unexpected weather changes.2,13,14 Most minor lesions are superficial and occur in areas of the head and face, while deep frostbite injuries that usually necessitate hospitalization affect the hands and feet in more than 90% of patients.2 In one series, 19% of injuries involved the upper extremity, 47% the lower extremity, 31 % both upper and lower extremities, and only 3 % the head and face.8

Host and Environmental Factors

It has been written that the “human capacity for physiologic adaptation to cold is minimal; we survive by insulating ourselves with protective clothing.”15 In the absence of sufficient protection in cold weather, cold exposure injuries are likely. Their severity is mediated by both host and environmental factors (Table 1).

Table 1
Table 1:
Predisposing Factors for Cold Exposure Injuries

The primary host factor is alcohol abuse.8-10 Other factors include psychiatric illness, smoking, and peripheral vascular disease. In a study of admission records of US Department of Veterans Affairs hospitals over a 2-year period, psychiatric illnesses that included alcohol abuse accounted for 37% of all admissions, while the incidence in patients admitted for frostbite was 63%.9 The malnourished or chronically ill are also more susceptible to coldinduced injuries because of their decreased capacity for normal thermoregulation. Smoking is an adverse factor because nicotine elevates plasma catecholamine levels and because other chemicals in smoke reduce the synthesis of nitric oxide, a vasodilator.6 Smoking also potentiates the development of thromboses by increasing fibrinogen levels and platelet activity. Peripheral vascular disease as seen in patients with atherosclerosis and diabetes mellitus is also a risk factor and, when present, intensifies the deleterious effects of smoking. Vasospastic disorders, such as Raynaud disease, can also potentiate cooling of skin that is in contact with cold objects.16 In addition to these medical problems, occupational activities, particularly those that require the use of vibrating tools, pose risks.6 Even one prior cold exposure injury increases the risk for reinjury in previously affected areas.17

The notion that individuals who normally reside in exceptionally cold climates “feel the cold less” because of better acclimatization is controversial. Indigenous people living in the Arctic, such as Inuits, certainly benefit from their day-to-day experience of surviving in a cold climate, and those engaged in fishing during the winter months are able to work with their bare hands in freezing water for long periods without any obvious incapacity. Although their hands appear to have a greater vasodilatation response to cold than do those of other indigenous ethnic groups, other anatomic areas, such as their feet, do not fare as well, and they frequently sustain cold exposure injuries.2 However, studies have shown that race does have a role with respect to individuals of African descent.5 These individuals are more likely to sustain frostbite injuries than are Caucasians, who appear to have better capability for coldinduced vasodilatation.1

Environmental factors primarily involve temperature, duration of exposure, altitude (> 17,000 feet), and wind speed. Wind speed is a wellknown risk factor; the greater the wind velocity, the faster the rate of convection heat transfer from exposed areas of the body. It was not until 1945 that this effect was quantified by Siple and Passel,18 who developed a Wind Chill Index based on the time necessary for water in a cylinder to freeze when exposed to different combinations of temperature and air speed. Tissues exposed to an ambient temperature of 0°F (-18°C) and a wind speed of 10 mph will freeze in 1 hour compared with freezing in just 10 minutes when exposed to the same temperature at a wind velocity of 40 mph.1 Generally, the risk of frostbite is low when air temperatures are above 14°F (-10°C), regardless of wind velocity, but is high at temperatures below -13°F (-25°C) even when there is little or no wind.19 The risk of injury is increased when a limb is in direct contact with a conductive material such as a liquid (eg, water, ice) or metal because of a more rapid loss of body heat.20


Nonfreezing and freezing cold exposure injuries to the extremities range in severity from mild, with few or no clinical sequelae, to severe, with serious long-term problems.1,2,19 Frostnip is the mildest cold exposure injury. It occurs at freezing temperatures and affects only the superficial layers of the skin, resulting in skin blanching and numbness. There is no damage to the dermis or deeper tissues, and the condition is completely reversible provided further exposure is prevented.

Chilblain, also referred to as pernio, is a more serious problem that occurs in response to repeated exposure to cold, nonfreezing temperatures in dry conditions. Typically, patients report a burning sensation in the affected area with pruritis, swelling, and erythema. Blistering may develop and sometimes progresses to ulceration. Although chilblain lesions generally resolve within 2 weeks, the condition has been associated with chronic problems, such as a vasculitis that most commonly occurs in young and middle-aged women.21

Trench foot, more commonly referred to today as immersion foot, is next in the spectrum of cold exposure injuries. This condition is most commonly seen in military personnel exposed to prolonged wet conditions at temperatures that are nonfreezing but <50°F (<10°C).

Frostbite is at the severe end of the spectrum of cold exposure injuries; like frostnip, it occurs in freezing temperatures. However, frostbite results in tissue necrosis that can be localized or extensive, sometimes requiring amputation. When the central (ie, “core”) body temperature is affected, the condition is referred to as hypothermia, which if untreated can be fatal.


Normal thermoregulation protects core body temperature (CBT) at the expense of peripheral temperatures in the skin and extremities. In hot weather, peripheral circulation is normally increased as a cooling mechanism. This is manifested primarily by perspiration that dissipates heat through conduction, convection, and evaporation. In cold weather, the opposite occurs. To maintain CBT, circulation is shunted centrally from the periphery. With hypothermia, defined as a CBT ≤94°F (≤35°C), the normal thermoregulatory mechanism is overwhelmed by the environmental conditions.7 As core temperature decreases, so does the basal metabolic rate as well as heart rate and cardiac output. The decrease in cardiac function is accompanied by myocardial irritability and conduction problems, as reflected in abnormal electrocardiograms.7

Central nervous system function is also affected; the hypothermic person becomes disoriented and can rapidly progress to a comatose state. Shivering, as the body attempts to generate sufficient heat to warm itself, is common. Shivering is essentially an anaerobic process that depletes muscle glycogen while generating lactic acid. When core temperature drops below 86° to 89.6°F (30° to 32°C), shivering is replaced by muscle rigidity that can resemble rigor mortis. The clinical condition can be confused with death because a person so affected may be rigid and comatose and have slow to absent respirations and pulse, dilated pupils, and a flat electrocardiogram reading. Thus, the standard parameters of death do not apply to hypothermic patients. Instead, they must be rewarmed before the determination is made; “no one is dead until warm and dead.”22,23

Although hypothermia is lifethreatening, frostbite is generally a problem of morbidity. The physiologic responses to the effects of freezing temperatures on limbs have been categorized into four phases: cooling and freezing (phase I), rewarming (phase II), progressive tissue injury (phase III), and resolution (phase IV).1

Phases of Injury

Phase I: Cooling and Freezing

Initially, in exposure to cold temperature there is vasoconstriction and vasospasm. This is soon followed by a physiologic reaction of transient arteriovenous shunting, referred to as the “hunting response,” consisting of cycles of vasodilatation and vasoconstriction that alternate approximately every 10 minutes. The response is not present in all individuals, and those without the response are more prone to cold exposure injuries. The hunting response continues as long as the individual is sufficiently protected to prevent a severe drop in core temperature.

However, when cold exposure persists and core temperature drops to a level that threatens circulation to vital organs, the response cycles cease; the body's physiologic priority is always “life over limb.” With continued cold exposure, the temperature in the affected limb decreases and plateaus at the freezing point of tissues (28°F [-2°C]). Extracellular ice crystals form in the plasma volume, resulting in sludging and stasis, as well as in interstitial cellular tissues, producing an osmotic gradient leading to intracellular dehydration.

As the limb temperature continues to decrease, intracellular ice crystals form and expand, resulting in mechanical destruction of cell membranes. Freezing itself especially causes direct cell membrane damage to endothelial cells in small vessels. The interstitial crystallization process produces an exothermic reaction, and the latent heat that is released maintains limb temperature at its freezing level. However, when the process is complete, limb temperature falls rapidly to the environmental (ambient) temperature.

Phase II: Rewarming

Rewarming reverses the freezing process. Heat absorbed by the limb begins an endothermic reaction as extracellular and intracellular crystals melt. Limb temperature rises and plateaus at its freezing point (28°F [-2°C]) until the melting process is complete. The temperature then continues to rise. Intracellular swelling occurs, and the endothelial cells of small capillaries that are most vulnerable to the effects of freezing become highly permeable, resulting in extravasation of fluid, causing edema and blisters. Rewarming is crucial to recovery, and it is critically important that freezing not recur. Even a single repeated freeze-thaw cycle has severely deleterious effects on the tissues.

Phase III: Progressive Tissue Injury

Inflammation, vascular stasis, and thromboses lead to ischemia and progressive tissue damage. The tissue necrosis, accompanied by blisters, is similar to the changes that occur with thermal burns. Prostaglandins and thromboxanes, both metabolites of arachidonic acid and both inflammatory mediators, have been found in similar levels in the blisters of patients with burn and in those with frostbite.24 Clinical studies have shown that the level of prostaglandin E2, a vasodilator and an antiaggregating platelet substance, is only slightly decreased, while levels of prostaglandin F and thromboxane B2, both vasoconstrictors and platelet-aggregating substances, are markedly increased.25

Phase IV: Resolution

Following the initial cellular and microvascular damage, and after the effects of prostaglandins and thromboxanes have dissipated, recovery proceeds along three possible pathways. There may be complete healing with little or no clinical symptoms, healing associated with later sequelae, or early tissue necrosis leading to gangrene. All three outcomes may be present to varying degrees.

Clinical Presentation


Hypothermia and frostbite can occur separately or together. Hypothermia is potentially the more serious problem because it is lifethreatening. It is classified as mild, moderate, or severe, depending on CBT. Although there is no universally accepted agreement on the temperature levels for each stage, the generally accepted parameters for both temperature scales are mild hypothermia, 90° to 94°F (32° to 35°C); moderate hypothermia, 82° to 89°F (28° to 32°C); and severe hypothermia, <82°F (<28°C).23,25 The lower the core temperature, the more profound the effects on the cardiovascular, respiratory, central nervous, and coagulation systems. Initial tachycardia is followed by bradycardia and decreased cardiac output, both of which are associated with arrhythmias, including atrial and ventricular fibrillation. Respiratory rate is also decreased, resulting in carbon dioxide retention that leads to hypoxia and respiratory acidosis. Central nervous system function is depressed, causing the affected person initially to become disoriented and later comatose if the condition is not reversed. Coagulation defects occur because of a decrease in platelet count and platelet function.


Frostbite is essentially a thermal injury to local tissues that is similar to burns and can be classified as a first-, second-, third-, and fourthdegree injury, depending on the depth of tissue damage (Figure 1). However, unlike burns, most frostbite injuries appear similar at the initial evaluation; they can be classified only after rewarming. Firstdegree injuries are characterized by a central whitish area surrounded by erythema; second-degree injuries display clear or cloudy fluid-filled blisters that appear within the first 24 hours; third-degree injuries also have blisters but ones that are hemorrhagic and progress to hard black eschars; and fourth-degree injuries are associated with tissue necrosis of varying severity.

Figure 1
Figure 1:
Illustration demonstrating different levels of tissue damage depending on the severity of cold exposure injury.

In recent years, frostbite injuries have been divided into just two categories: superficial, comprising firstand second-degree injuries, and deep, comprising third- and fourth-degree injuries. The newer classification system appears to be more useful in predicting clinical outcomes. For example, an injury lesion that has clear blisters and skin that deforms easily under digital pressure is likely a superficial injury and has a better prognosis than does the injury characterized by hemorrhagic blisters and blue-black, nondeforming skin.


Treatment of a cold exposure injury begins as soon as it is identified, even if the severity has not yet been determined. The initial objectives are to protect the individual from further exposure to freezing temperatures and to protect the injured extremity from mechanical trauma (eg, injury to a lower extremity caused by walking). Rewarming should not begin until there is assurance that the individual can be maintained in a constant warm environment. Repeat cycles of thawing and refreezing must be avoided because they result in significantly greater tissue damage. The injured skin should not be rubbed because rubbing is likely to cause mechanical trauma, and the frostbitten area should not be placed near any heat. Instead, the extremity is simply padded and splinted; this can be accomplished by wrapping the extremity in a blanket before the individual is transported to a medical facility. Depressants, such as sedatives and alcohol, should not be given because they can blunt the shivering reflex and cause further lowering of body temperature in a person who may already be experiencing hypothermia.

In the medical facility, the first priority is to assess the patient for hypothermia. An accurate assessment of CBT is critically important and can be obtained with an esophageal or rectal probe. Rewarming techniques for hypothermia are either external-surface rewarming or internal-core rewarming (Table 2). External rewarming can be done by the passive method, in which the patient is simply redressed in dry clothes and placed in a warm room, or it can be active, with the room containing convection heaters such as heat lamps and radiant heaters. External rewarming may also involve the application of a heating blanket or immersion of the patient in warm water. External rewarming is used for the patient with mild hypothermia, and generally the passive method is preferred. A potential problem with the active method is that it may cause a too-rapid peripheral vasodilatation, resulting in a bolus of cool, stagnant blood, containing metabolic waste products, that is shunted to the core tissues. The temperature of those tissues then paradoxically drops even lower, a condition referred to as “afterdrop.” 7 The main risk is that myocardial irritability already present will worsen and a severe arrhythmia (eg, ventricular fibrillation) will ensue. Cardiac monitoring is therefore important during rewarming, even with mild cases.

Table 2
Table 2:
Hypothermia Treatment 20,23,25

For moderate and severe hypothermia, internal core rewarming is usually required. Warmed oxygen inhalation and warm intravenous fluids have been used, but neither treatment is particularly effective. Body cavity lavage, either thoracic or abdominal, has also been used, but this highly invasive procedure has limited benefits. Cardiac bypass is more effective, but it requires systemic heparinization, which is contraindicated in trauma patients. In 1991, Gentilello and Rifley26 reported on a method, which they referred to as continuous arteriovenous rewarming, that does not require systemic anticoagulation. The method essentially involves a percutaneously placed femoral arterial catheter connected to the inflow side of a fluid heat exchanger; the rewarmed blood is returned to the body through a subclavian venous catheter. Continuous arteriovenous rewarming has been used successfully in the treatment of severe hypothermia and achieves warming at the rate of 1 °C approximately every 15 minutes, far faster than body cavity lavage methods.26

Treatment of frostbite is begun when CBT is ≥95°F (35°C) (Table 3). The affected limb is rapidly rewarmed, a concept developed toward the end of World War II by Fuhrman and Crismon29 in animal experiments they conducted at the request of the US government because of the large number of frostbite injuries among military personnel. Their paper was the first in the English-language medical literature to report on rapid rewarming for the treatment of frostbite, a radical departure from previous treatment methods. In 1960, Mills and Whaley30 reported on the first clinical experience using this treatment, one that they combined with deep ultrasound in their early cases. Rapid rewarming is performed in a water bath containing a mild antibacterial agent at 104° to 107.6°F (40° to 42°C).1,2,7,20,23 This narrow temperature range is critically important; rewarming at a lower temperature reduces the likelihood of tissue survival, while rewarming at a higher temperature may cause thermal burns and worsen the injury.

Table 3
Table 3:
Frostbite Treatment 2,15,27,28

Rewarming continues for 15 to 30 minutes, the time usually required for complete thawing and cessation of vasoconstriction. When rewarming is successful, the skin becomes pliable and has a red-purple appearance, both favorable signs for recovery. However, skin that remains nonpliable, cold, darkly mottled, and anesthetic is likely to progress to tissue necrosis (Figures 2 to 5). Tetanus prophylaxis and intravenous antibiotics should be administered. The injury is observed for signs of a compartment syndrome, which is rare but may occur as a result of tissue reperfusion in the early postthaw period. In such cases, an immediate fasciotomy or escharotomy is required.

Figure 2
Figure 2:
Hands (A and B) and feet (C and D) of a 31-year-old male mugging victim whose shoes were taken by his assailant and who was left unconscious in a ditch in freezing weather for approximately 12 hours. On hospital admission, his core body temperature was 90°, and the digits of both hands and both feet from the ankles distal were cool and bluish. Within 24 hours, the patient's core body temperature returned to normal, but at 1 week, the areas of discoloration in the hands (E and F) and feet (G and H) had worsened. (Clinical case courtesy of Donald Wehmeyer, MD.)
Figure 3
Figure 3:
Same patient as in Figure 2. A and B, At 5 weeks, areas of gangrene in the hands had demarcated to the right ring and little fingers and the left little finger. C, D, and E, In the feet, gangrene involved all the toes beginning at the metatarsal levels as well as more proximal areas of necrosis.
Figure 4
Figure 4:
Same patient as in Figures 2 and 3. At 8 weeks, the gangrenous fingers in both hands (A and B) were amputated, and the areas of necrosis in both the left (C) and right (D) feet, proximal to the gangrenous toes, were débrided. Necrosis in the feet was extensive and involved the extensor tendons.
Figure 5
Figure 5:
Same patient as in Figures 2 through 4. A, At 9 weeks, the gangrenous toes were amputated in an attempt to save the remaining portions of the feet. This was not possible, and Syme amputations (B) were necessary.

Clear blisters should be débrided to reduce the high levels of inflammatory mediators (ie, prostaglandin F and thromboxane B2) that are in the blister fluid. Aloe vera is applied topically to the affected areas every 6 hours. Hemorrhagic blisters are drained but are generally left intact because they represent deeper injuries; débriding them could lead to desiccation of the underlying dermis. With open wounds, a topical antimicrobial agent such as silver sulfadiazine ointment is used. Oral nonsteroidal anti-inflammatory drugs are usually prescribed for several days. Adequate analgesia is also provided to keep the patient comfortable during and immediately after the rewarming process, which is usually painful. Keeping the patient comfortable facilitates daily rehabilitation, which involves whirlpool hydrotherapy and physical and occupational therapy to preserve joint motions.27

A variety of adjunctive treatments for frostbite injuries has been recommended, including low-molecularweight dextran to reduce blood viscosity, anticoagulants, vasodilators, sympathetic-blocking drugs, and thrombolytics such as tissue plasminogen activator (tPA).2,12,27,31 Recent studies have shown that intravenous or intra-arterial tPA administered within 24 hours of cold exposure significantly (P < 0.05)12 reduces the need for digital amputations but may be ineffective for patients with warm ischemia times >6 hours or those with evidence of multiple freeze-thaw cycles.12,31 Twomey et al31 reported that tPA is ineffective for cold exposure >24 hours. Intravenous administration of tPA is probably safer than intra-arterial administration in regard to the risk of bleeding complications.31 Although these treatments have been beneficial in experimental studies, there is no conclusive evidence that they appreciably change the course of recovery. In addition, anticoagulants and/or thrombolytics are contraindicated in alcoholic patients, who are the most prone to develop frostbite injuries because of the risk that they may have sustained concomitant head injuries such medication could cause internal bleeding. Surgical sympathectomies have also been recommended as an adjunctive treatment, and although they shorten the time to resolution of pain and to demarcation of tissue necrosis, they do not reduce the extent of tissue loss.15 Hyperbaric oxygen has been used to increase local oxygen tension and therefore promote better healing, but there is only anecdotal evidence of its effectiveness.32

Surgery is generally reserved for the late treatment of frostbite. The grim aphorism “frostbite in January, amputate in July,” although an exaggeration, reflects the considerable time necessary for frostbite lesions to demarcate.33 Demarcation commonly occurs between 1 and 3 months from the time of initial exposure, and it is during this period that débridement is usually performed (Figures 2-5). Earlier débridement is reserved for patients with uncontrolled infections.28 Various imaging studies have been used in an effort to make an early determination of the extent of tissue necrosis and thereby move up the time at which débridement and even amputation can be performed. These studies have included Doppler ultrasound, arteriography, magnetic resonance imaging, magnetic resonance angiography, and bone scans.13,34-36 Doppler ultrasound and angiography have not been reliable because they do not accurately show blood flow at the arteriole and capillary level. The same is true of magnetic resonance angiography.2,15 Magnetic resonance imaging, particularly T2-weighted images, will show enhanced signal intensity of necrotic muscles because of disruption of cell membranes and increase in extracellular fluid.34 However, the specificity of signal changes is often insufficient for making a decision to proceed with an amputation before there is a clinical demarcation of tissue necrosis.

Technetium 99m (99mTc) pertechnetate scintigraphy, either as twoor three-phase injection studies, is commonly used to assess tissue viability.13,35,36 In a large retrospective study of 92 patients treated at Chamonix Hospital in the French Alps, 99mTc bone scans were performed between 2 and 4 days of the injury and again at 7 to 10 days.13 Absence of uptake on the first scan was a poor prognostic finding. The second scan in such cases generally continued to show absence of uptake, and in almost all patients, some type of amputation was necessary. However, a decreased but not absent uptake on the first scan offered a better prognosis. Either the subsequent scan remained unchanged and the patient's clinical condition deteriorated or the subsequent scan showed increased uptake and was accompanied by clinical recovery. The authors reported that 99mTc bone scans within the first few days of the frostbite injury indicated the level of amputation in more than 84% of their cases.

The authors of a later prospective study provided different statistics and conclusions, noting that absent or low uptake beyond 10 days did not correlate with the need for amputation.36 None of the 20 patients in their study who had ischemic lesions on bone scans after 10 days required amputation. The level of arterial perfusion required to sustain tissue viability was lower than the sensitivity of the bone scan, and absence of uptake did not necessarily indicate tissue necrosis that leads to gangrene. Absence of uptake can also occur with ischemia that results in fibrosis (ie, chronic stable ischemia) and with another condition referred to by Bhatnagar et al36 as “hibernating microvasculature.”

These studies show that the specificity of bone scans is not sufficient for use as the sole criterion for deciding on surgical treatment, particularly when an amputation proximal to the wrist or ankle is being contemplated. Bone scanning is a useful diagnostic tool, but the patient's clinical presentation is the key factor for determining the necessity for and timing of surgery.

Late Sequelae

Although frostbite in extremities does not always lead to an amputation, it often results in persistent symptoms and in impaired function that may last for years or the patient's lifetime. Frequently, these problems affect leisure and work activities.6,14 In one study, approximately 50% of frostbite victims reported they had persistent chronic pain in the affected limb or limbs, and in 15%, the pain was characterized as “intolerable.”10

Late sequelae affect vasomotor function, nerve function, and/or the musculoskeletal system (Table 4). Vasomotor changes are the most common and include cold sensitivity, persistent abnormal color changes, and hyperhidrosis.14,28 Treatment measures and agents include sympathetic nerve blocks, vasodilators, and β blockers. Increased susceptibility to future cold injuries is common because of impaired circulation in affected areas. Surgical sympathectomy in the hand has been effective for chronic digital vasospasm. Injuries to superficial and deep nerves may result in neuropathies that are manifested by cold and/or heat hypersensitivity, hypesthesia, and paresthesia. Electrodiagnostic studies of patients with cold exposure injury to the extremities have detected decreased motor and sensory nerve conduction velocities, particularly in lower extremity injuries.37 Surgical decompression of peripheral nerves, such as the median nerve in the carpal tunnel, may be required.

Table 4
Table 4:
Late Sequelae of Cold Exposure Injuries

Although radiographs have little application in the early evaluation of frostbite injuries, they are useful for late sequelae. Frostbite often causes later musculoskeletal problems, such as localized osteopenia, subchondral bone loss resulting in articular damage (ie, frostbite arthropathy), and joint contractures.38,39 In the hands and feet, the distal interphalangeal joints are most commonly affected, followed by the proximal interphalangeal joints. Prosthetic joint arthroplasties or resectional arthroplasties are sometimes required. Children are susceptible to growth disturbances from frostbite injuries, which most commonly affect the fingers. Frostbite can cause stunting of growth that is usually manifest 1 to 2 years after exposure and that is associated with premature epiphyseal plate closures that are evident on radiographs40,41 (Figure 6). Joint laxity and/or angular deformities may subsequently develop and, if significant, may require surgery (eg, epiphyseal arrest, osteotomy, and occasionally, arthrodesis).

Figure 6 A,
Figure 6 A,:
Hands of a 10-year-old boy who sustained cold injuries to both hands while walking to school without mittens in freezing weather. Cold exposure was for approximately 20 minutes. B, Posteroanterior radiograph of the patient's right hand, demonstrating premature closure, caused by the injuries, of the epiphyseal plates in the middle and distal phalanges of all fingers, resulting in growth stunting.


Cold exposure injuries have affected human beings since the beginning of recorded history, but only in the past 50 years has our understanding of the pathophysiology of these injuries provided effective treatment protocols. With respect to frostbite, future studies could advance treatment by facilitating earlier, more accurate identification of necrotic tissues than is currently possible. Potential for advancement lies especially in prophylaxis, primarily with better protective garments, particularly for exposed areas of the face and head, and for the hands and feet. Future garments may contain sensors to monitor peripheral temperatures and CBT in individuals who work in cold climates or who engage in winter sports activities.


Evidence-based Medicine: No level I/II studies are cited. Most of the studies cited are level III/IV case reports and case-control studies (references 4, 5, 7-13, 15, 16, 20, 29, 30, and 33-40) or level V expert opinion (references 1-3, 6, 14, 19, 21, 22, 24-27, 31, and 32).

Citation numbers in bold type indicate references published within the past 5 years.

1. Vogel JE, Dellon AL: Frostbite injuries of the hand. Clin Plast Surg 1989;16: 565-576.
2. Murphy JV, Banwell PE, Roberts AH, McGrouther DA: Frostbite: Pathogenesis and treatment. J Trauma 2000;48:171-178.
3. Paton BC: A history of frostbite treatment. Int J Circumpolar Health 2000;59:99-107.
4. Cold Weather Effects on Major Military Campaigns. Available at: weather%20effects%20on%20milita ry%20campaigns.PDF. Accessed October 17, 2008.
5. DeGroot DW, Castellani JW, Williams JO, Amoroso PJ: Epidemiology of US Army cold weather injuries, 1980-1999. Aviat Space Environ Med 2003;74:564-570.
6. Ervasti O, Juopperi K, Kettunen P, et al: The occurrence of frostbite and its risk factors in young men. Int J Circumpolar Health 2004;63:71-80.
7. Grace TG: Cold exposure injuries and the winter athlete. Clin Orthop Relat Res 1987;216:55-62.
8. Valnicek SM, Chasmar LR, Clapson JB: Frostbite in the prairies: A 12-year review. Plast Reconstr Surg 1993;92: 633-641.
9. Pinzur MS, Weaver FM: Is urban frostbite a psychiatric disorder? Orthopedics 1997;20:43-45.
10. Koljonen V, Andersson K, Mikkonen K, Vuola J: Frostbite injuries treated in the Helsinki area from 1995 to 2002. J Trauma 2004;57:1315-1320.
11. Tanaka M, Tokudome S: Accidental hypothermia and death from cold in urban areas. Int J Biometeorol 1991; 34:242-246.
12. Bruen KJ, Ballard JR, Morris SE, Cochran A, Edelman LS, Saffle JR: Reduction of the incidence of amputation in frostbite injury with thrombolytic therapy. Arch Surg 2007;142:546-553.
13. Cauchy E, Marsigny B, Allamel G, Verhellen R, Chetaille E: The value of technetium 99 scintigraphy in the prognosis of amputation in severe frostbite injuries of the extremities: A retrospective study of 92 severe frostbite injuries. J Hand Surg [Am] 2000; 25:969-978.
14. Hassi J, Mäkinen TM: Frostbite: Occurrence, risk factors and consequences. Int J Circumpolar Health 2000;59:92-98.
15. Su CW, Lohman R, Gottlieb LJ: Frostbite of the upper extremity. Hand Clin 2000;16:235-247.
16. Jay O, Havenith G: Differences in finger contact cooling response between an arterial occlusion and a vasodilated condition. J Appl Physiol 2006; 100: 1596-1601.
17. Cattermole TJ: The epidemiology of cold injury in Antarctica. Aviat Space Environ Med 1999;70:135-140.
18. Siple PA, Passel CF: Measurement of dry atmospheric cooling in subfreezing temperatures. Proc Am Philos Soc 1945;89:177-199.
19. Danielsson U: Windchill and the risk of tissue freezing. J Appl Physiol 1996;81:2666-2673.
20. Petrone P, Kuncir EJ, Asensio JA: Surgical management and strategies in the treatment of hypothermia and cold injury. Emerg Med Clin North Am 2003;21:1165-1178.
21. Simon TD, Soep JB, Hollister JR: Pernio in pediatrics. Pediatrics 2005; 116:e472-e475.
22. Gregory RT, Patton JF III, Whitby JD, Berkeley JS, Hillman H: Treatment after exposure to cold. Lancet 1972;1: 377-378.
23. Jurkovich GJ: Environmental coldinduced injury. Surg Clin North Am 2007;87:247-267.
24. Robson MC, Heggers JP: Evaluation of hand frostbite blister fluid as a clue to pathogenesis. J Hand Surg [Am] 1981;6:43-47.
25. Biem J, Koehncke N, Classen D, Dosman J: Out of the cold: Management of hypothermia and frostbite. CMAJ 2003;168:305-311.
26. Gentilello LM, Rifley W: Continuous arteriovenous rewarming: Report of a new technique for treating hypothermia. J Trauma 1991;31:1151-1154.
27. Mc Cauley RL, Hing DN, Robson MC, Heggers JP: Frostbite injuries: A rational approach based on the pathophysiology. J Trauma 1983;23:143-147.
28. McAdams TR, Swenson DR, Miller RA: Frostbite: An orthopedic perspective. Am J Orthop 1999;28:21-26.
29. Fuhrman FA, Crismon JM: Studies on gangrene following cold injury: VII. Treatment of cold injury by means of immediate rapid rewarming. J Clin Invest 1947;26:476-485.
30. Mills WJ, Whaley R: Frostbite experience with rapid rewarming and ultrasonic therapy: 1960-1961. Wilderness Environ Med 1998;9:226-247.
31. Twomey JA, Peltier GL, Zera RT: An open-label study to evaluate the safety and efficacy of tissue plasminogen activator in treatment of severe frostbite. J Trauma 2005;59:1350-1355.
32. von Heimburg D, Noah EM, Sieckmann UP, Pallua N: Hyperbaric oxygen treatment in deep frostbite of both hands in a boy. Burns 2001;27: 404-408.
33. Britt LD, Dascombe WH, Rodriguez A: New horizons in management of hypothermia and frostbite injury. Surg Clin North Am 1991;71:345-370.
34. Barker JR, Haws MJ, Brown RE, Kucan JO, Moore WD: Magnetic resonance imaging of severe frostbite injuries. Ann Plast Surg 1997;38:275-279.
35. Greenwald D, Cooper B, Gottlieb L: An algorithm for early aggressive treatment of frostbite with limb salvage directed by triple-phase scanning. Plast Reconstr Surg 1998;102: 1069-1074.
36. Bhatnagar A, Sarker BB, Sawroop K, Chopra MK, Sinha N, Kashyap R: Diagnosis, characterisation and evaluation of treatment response of frostbite using pertechnetate scintigraphy: A prospective study. Eur J Nucl Med Mol Imaging 2002;29:170-175.
37. Arvesen A, Wilson J, Rosén L: Nerve conduction velocity in human limbs with late sequelae after local cold injury. Eur J Clin Invest 1996;26:443-450.
38. Kahn JE, Lidove O, Laredo JD, Blétry O: Frostbite arthritis. Ann Rheum Dis 2005;64:966-967.
39. Chalmers IM, Bock GW: Cold injury in 2 patients with connective tissue disease: Frostbite arthritis plus. J Rheumatol 2000;27:2526-2529.
40. Carrera GF, Kozin F, Flaherty L, Mc-Carty DJ: Radiographic changes in the hands following childhood frostbite injury. Skeletal Radiol 1981;6:33-37.
41. Nakazato T, Ogino T: Epiphyseal destruction of children's hands after frostbite: A report of two cases. J Hand Surg [Am] 1986;11:289-292.
© 2008 by American Academy of Orthopaedic Surgeons