Staged cold-injury pathophysiology and field triage for a polar expedition casualty
Before any ice forms in living tissue, the body mounts a defensive response to cold that itself sets the stage for injury. Peripheral vasoconstriction sacrifices the extremities to protect the core — a trade-off that, if exposure continues, tips from adaptive to destructive. Frostnip is the reversible warning shot before true frostbite begins.
When skin temperature falls, cutaneous thermoreceptors trigger sympathetic release of norepinephrine, constricting arterioles and closing arteriovenous shunts in the fingers, toes, ears, and nose. Blood flow to a fingertip can fall from several hundred milliliters per minute to single digits, preserving core temperature at the direct expense of the extremity.
Paradoxically, the body periodically overrides this shutdown. Every 5–10 minutes, the "hunting reaction" (Lewis wave) briefly dilates peripheral vessels, flushing warm blood through the digit before vasoconstriction resumes. This cyclical rewarming delays frostbite but also increases total heat loss — a trade the body accepts to prevent immediate tissue death.
As exposure continues and core temperature is threatened, the hunting reaction weakens and eventually stops altogether, leaving the extremity persistently vasoconstricted and undefended against freezing.
Frostnip is superficial cooling of skin without ice crystal formation: pale, numb, tingling skin that returns to normal color and sensation within minutes of rewarming, with no blistering and no lasting damage. It is the most common cold injury and, critically, the last fully reversible stage before real tissue freezing begins.
Risk factors that accelerate the transition from frostnip to frostbite include wet clothing or gloves (water conducts heat roughly 25 times faster than air), high wind speed, altitude (lower barometric oxygen impairs peripheral perfusion), nicotine and other vasoconstricting substances, dehydration, and pre-existing peripheral vascular disease or diabetes.
Field teams often underestimate frostnip because early numbness is not painful — the same anesthesia that blunts symptoms also removes the warning signal that would otherwise prompt someone to rewarm before ice actually forms.
Wind dramatically accelerates heat loss by stripping away the thin layer of warmed air next to the skin. Wind-chill equivalent temperature charts (used by meteorological services worldwide) translate air temperature and wind speed into an estimated time until exposed skin develops frostbite.
As a rough field guide: at a wind chill near -20°C, exposed skin can develop frostbite in roughly 30 minutes; near -35°C, that window shrinks to about 10 minutes; below -60°C, frostbite can occur in under 2 minutes. These are population averages — individual susceptibility varies with clothing, activity level, hydration, and prior cold injury history.
Expedition protocols typically mandate covering all exposed skin, buddy-checking faces for early pallor, and setting hard exposure-time limits keyed to the day's wind-chill forecast rather than air temperature alone.
At wind chills below roughly -60°C, exposed skin can develop frostbite in under two minutes — faster than it takes to stop, assess a teammate, and re-glove.
Once skin temperature drops below its freezing threshold, ice begins to form — first outside the cells, then progressively deeper. Superficial frostbite involves the epidermis and upper dermis: the skin turns waxy and white, may blister within a day, and — if rewarmed correctly and promptly — still carries a good prognosis for full tissue salvage.
Ice almost always forms first in the extracellular space, because extracellular fluid is more dilute than the cytoplasm and freezes at a slightly higher temperature. As extracellular ice grows, it draws water osmotically out of surrounding cells, concentrating intracellular electrolytes to toxic levels and shrinking cells — a process called solution effect injury.
If cooling continues or occurs rapidly, ice can also form directly inside cells, and growing crystals mechanically shear membranes, organelles, and the cytoskeleton. Both mechanisms damage tissue even before any vascular effects are considered.
Critically, a single freeze causes measurably less damage than a freeze-thaw-refreeze sequence: once tissue thaws and cells re-absorb water, a second freeze produces larger, more destructive ice crystals in already-injured cells. This is the physiological basis for the cardinal rule never to rewarm frostbitten tissue unless refreezing can be reliably avoided.
Frostbite has traditionally been graded like burns, in four degrees: first (hyperemia and edema without blistering), second (superficial blistering with clear or milky fluid), third (hemorrhagic blistering extending into the reticular dermis), and fourth (full-thickness injury into subcutaneous tissue, muscle, tendon, or bone, often ending in mummification).
The problem with degree-based grading is that it can usually only be assigned accurately after the tissue has declared itself over days to weeks — not useful for immediate field or emergency-department triage. Modern practice increasingly uses a simplified two-tier system (superficial versus deep) at initial presentation, supplemented later by imaging-based grading (such as Cauchy's system, which incorporates the level of the visible lesion and early triple-phase bone scan findings) to predict eventual outcome much earlier.
Both systems agree on one point: the true depth of injury is often invisible at first contact and becomes clear only with time — which is why serial reassessment matters as much as the initial exam.
Several bedside signs help field providers estimate prognosis without imaging. Blister character is one of the most useful: clear or milky blisters, forming within the first 24 hours, indicate the underlying dermal microvasculature is largely intact and predict good tissue salvage. Hemorrhagic (blood-filled) blisters indicate injury has reached the deeper, more vascular dermis and predict a much higher chance of eventual tissue loss.
Other useful signs include residual sensation (light touch and pinprick — anesthesia is a poor prognostic sign), skin texture after thaw (soft and pliable is favorable; persistently hard and "woody" is not), and capillary refill once rewarmed (brisk pink refill is reassuring; mottled, cyanotic, or absent refill signals deep injury).
None of these field signs are perfectly reliable in isolation, which is why any frostbite beyond simple frostnip warrants reassessment over the following days rather than a single-exam verdict.
Clear or milky blisters carry a favorable prognosis with over 90% tissue salvage, while hemorrhagic blisters signal deep dermal vascular injury and a high risk of eventual necrosis — the single most useful bedside sign in the first 24 hours.
When freezing pushes past the dermis into subcutaneous fat, muscle, tendon, and bone, the injury becomes deep frostbite. Much of the eventual tissue loss in this stage is not from ice itself but from a "second insult": progressive microvascular thrombosis and ischemia that continue for days after the tissue has already thawed.
With prolonged exposure or very low tissue temperatures, the freezing front advances inward past the dermis into subcutaneous fat, then muscle, tendon, and — in the most severe cases — bone and periosteum. Different tissues tolerate this differently: highly vascular muscle and nerve suffer disproportionate damage, while relatively avascular tendon and bone are more structurally resistant even though the vessels supplying them are just as vulnerable to thrombosis.
As depth of freezing increases, the clinical picture shifts from painful, blistering skin to a cold, hard, insensate, and eventually immobile digit or limb segment. Fourth-degree injury frequently produces little pain at all in the affected part precisely because the nerves supplying it have also frozen and lost function.
The visible skin injury at initial presentation systematically underestimates true depth — a digit that looks only moderately affected on the surface may already have non-viable muscle and bone beneath it.
Rewarming a deeply frozen extremity does not simply reverse the injury — it initiates a second, largely separate injury process. Ice-damaged endothelium becomes intensely pro-thrombotic: platelets aggregate, thromboxane A2 and prostaglandin F2-alpha are released, and progressive microvascular thrombosis develops over the following hours to days, propagating ischemia into tissue that survived the freeze itself.
Arteriovenous shunting compounds the problem, diverting blood away from the capillary beds that need it most. Reperfusion also generates reactive oxygen species and inflammatory mediators that injure cells which had otherwise survived freezing intact.
This is why the ultimate boundary between viable and non-viable tissue is not fixed at the moment of thaw — it continues to evolve for one to three weeks (occasionally months) afterward, and why aggressive early debridement based on the initial post-thaw appearance risks removing tissue that would otherwise have survived.
Because visible injury lags true tissue viability, hospitals increasingly use triple-phase bone scintigraphy or MR angiography within 24–72 hours of rewarming to map perfusion and predict which tissue will survive — allowing earlier, more accurate decisions than waiting for full clinical demarcation.
Where deep frostbite is confirmed early, thrombolytic therapy (intra-arterial or intravenous tissue plasminogen activator, tPA) or vasodilator therapy (intravenous iloprost) given within roughly 24 hours of rewarming can dissolve or limit the microvascular thrombosis driving the "second insult." Published case series applying tPA within this window have reported digit amputation rates falling from a historical ~40% to roughly 10–15%.
These interventions require hospital-level monitoring for bleeding risk and are not available in the field — which is precisely why recognizing deep frostbite early and prioritizing evacuation, rather than prolonged field observation, materially changes outcomes.
In case series where tissue plasminogen activator was given within about 24 hours of rewarming, digit amputation rates fell from a historical ~40% to roughly 10% — the true "golden window" for deep frostbite salvage is measured in hours, and it is only reachable through evacuation, not field care.
Once refreezing can be reliably prevented, frostbitten tissue should be rewarmed as quickly and evenly as possible — the single intervention with the strongest evidence behind it. The method matters as much as the decision to rewarm: circulating warm water, a controlled temperature range, and strict avoidance of mechanical or dry-heat trauma to insensate tissue.
Rewarming speed matters because slow thawing prolongs the time tissue spends in the temperature range where ice crystals continue to grow and recrystallize, worsening cellular damage. Passing quickly through this zone — achieved with circulating water rather than still air or body contact — minimizes additional crystal growth.
Water is used because it conducts heat far more efficiently than air: a limb warmed in a 38°C water bath reaches target temperature many times faster than one warmed by campfire, blankets, or a companion's body heat, all of which are too slow and too uneven.
The 37–39°C range is a deliberate compromise: hot enough to rewarm quickly, but not so hot that it burns tissue that has lost protective sensation. Because frostbitten skin is anesthetic, a patient cannot reliably feel — and therefore cannot warn against — a water bath that is actually too hot, which is why the temperature must be measured with a thermometer, not judged by feel.
Rubbing or massaging frostbitten skin — including the old folk remedy of rubbing snow on it — is actively harmful. Ice crystals already present in the tissue act like microscopic glass shards; mechanical friction shears cell membranes and capillaries that might otherwise have survived, converting a superficial injury into a deeper one.
Dry radiant heat sources such as campfires, vehicle exhaust pipes, or stoves are equally dangerous for a different reason: anesthetic, insensate tissue cannot signal pain from excessive heat, so uneven radiant warming frequently produces a superimposed thermal burn on top of the existing cold injury, without the patient ever feeling it happen.
The single most catastrophic error, however, is allowing thawed tissue to refreeze before reaching definitive shelter or care. A freeze-thaw-refreeze cycle produces larger, more destructive ice crystals in tissue that is already injured and edematous, roughly doubling the ultimate extent of tissue loss compared with a single freeze — which is why field teams are trained to sometimes deliberately delay rewarming.
If a stable, warm environment and a reliable heat source (stove, thermometer, clean water) are available and the party will not need to keep moving through freezing conditions afterward, rewarming should begin as soon as possible — delay only prolongs ice-crystal exposure with no offsetting benefit.
If, however, the casualty must continue traveling on foot through cold terrain to reach safety and refreezing cannot be prevented, most wilderness medicine guidelines favor keeping the limb frozen, padded, and immobilized rather than thawing it in the field only to have it refreeze en route — a still-frozen limb tolerates onward transport far better than a thawed one that refreezes.
Whichever path is chosen, supportive care matters: aggressive analgesia before and during rewarming (rewarming pain is often severe), padding between digits, loose non-adherent dressings after thaw, limb elevation to limit edema, and avoiding any weight-bearing on a thawed lower extremity — walking on a rewarmed frostbitten foot is itself a recognized cause of additional tissue damage.
A single freeze-thaw-refreeze cycle roughly doubles the ultimate tissue loss compared with a single freeze — making the decision of when, not just how, to rewarm one of the most consequential calls in frostbite field management.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Rapid water-bath rewarming | As soon as refreeze can be prevented | Circulating water at 37–39°C for 15–30 min until tissue is flushed and pliable | Minimizes ice recrystallization — the strongest predictor of tissue salvage |
| Analgesia before immersion | Immediately before / during rewarming | NSAIDs blunt thromboxane-mediated vasoconstriction; opioids control severe reperfusion pain | Reduces pain and may limit progressive dermal ischemia |
| Delay rewarming, evacuate frozen | When refreeze en route is unavoidable | Keep the limb frozen, padded, and immobilized until a stable environment is reached | A still-frozen limb tolerates transport far better than one that thaws and refreezes |
| Rubbing or massaging with snow | Never appropriate | Mechanical shear from existing ice crystals lacerates cell membranes and capillaries | Folk remedy with no physiologic basis — actively worsens injury |
| Dry radiant heat (fire, exhaust) | Never appropriate | Uneven, uncontrolled heat applied to anesthetic, insensate tissue | High risk of an undetected superimposed thermal burn |
| Thaw, then re-expose to freezing | Never appropriate | A second freeze forms larger ice crystals in already-injured, edematous tissue | Roughly doubles ultimate tissue loss versus a single freeze |
Frostbite does not finish declaring itself at the moment of rewarming. Over the following days to weeks, a demarcation line gradually separates tissue that will recover from tissue that will not — and the classic teaching to "wait for demarcation" before any amputation remains central to good outcomes, with a narrow set of exceptions.
The boundary between viable and non-viable tissue after deep frostbite is not fixed at the moment of thaw; it evolves as microvascular thrombosis either resolves (restoring flow to marginal tissue) or propagates (extending ischemia). Collateral circulation can open over days to weeks, sometimes rescuing tissue that initially looked doomed.
In tissue that will not survive, the body may eventually perform spontaneous auto-amputation, walling off and separating dead (mummified, dry-gangrenous) tissue from living tissue along a natural demarcation line — without surgical intervention.
Because of this ongoing biology, premature surgical debridement based on the initial post-thaw appearance risks removing tissue that would have survived given more time. This is the origin of the old expedition-medicine proverb: frostbite sustained in January is not amputated until well after full demarcation is complete.
Isolated frostnip or clearly first-degree injury with full return of sensation and color after rewarming can reasonably be managed in the field with protective dressing, pain control, and monitoring, provided refreezing risk is eliminated.
Any suspicion of deep (third- or fourth-degree) frostbite, hemorrhagic blistering, persistent numbness or hard woody tissue after rewarming, involvement of a large body-surface area, coexisting hypothermia, or any sign of infection (fever, spreading redness, foul odor, or wet — rather than dry — gangrene) should trigger urgent evacuation rather than field observation.
The rationale is time-sensitive: the pharmacologic rescue window for thrombolytic or vasodilator therapy described earlier closes within roughly 24 hours of rewarming and is only accessible at a hospital capable of angiography and monitored infusion — a decision to "wait and see" in the field can quietly forfeit that window even when the eventual evacuation itself is not delayed by more than a day or two.
Grading systems such as Cauchy's combine the level of the visible lesion on exam with early triple-phase bone scan findings to stratify amputation risk soon after injury — from grade 1 (distal lesion, normal perfusion, essentially no amputation risk and no systemic involvement) up to grade 4 (lesion extending to or beyond the wrist or ankle, absent perfusion on bone scan, high risk of proximal amputation and of systemic complications such as rhabdomyolysis or compartment syndrome).
This kind of early stratification lets clinicians and expedition physicians set realistic expectations well before the months-long demarcation process is complete, and helps decide who genuinely needs urgent transfer versus who can be safely monitored.
Even frostbite survivors who avoid amputation frequently experience long-term sequelae: permanent cold sensitivity and pain in the affected digits, altered sweating, nail deformity, and — in children whose growth plates were involved — subsequent digit shortening or angular deformity as they continue to grow.
Cauchy's grading system, combining early clinical exam with triple-phase bone scintigraphy, can predict amputation risk with roughly 84–100% accuracy within the first days after injury — turning what was once a months-long waiting game into an early, evidence-based evacuation decision.