From Bird
Late-stage frostbite injury in a peacock (Pavo cristatus). The tissue is mummified from the level of the distal tarsometatarsus.
Frostbite in a hatch-year great blue heron (Ardea herodias) that has just been found recumbent in a stream during winter in Ontario, Canada. This heron had missed migration.
Same bird 24 days later
Mummified toes on a red-tailed hawk (Buteo jamaicensis). The arrowheads indicate the line of demarcation

Frostbite is a localized tissue injury sustained as a result of cooling and thawing of tissues. Although much of the work done on frostbite has been with mammals, birds are susceptible, and it is not an uncommon injury in birds in colder climates.

Risk factors in birds include unseasonable weather[1], use of anesthetics[2], use of wire caging[3], metal leg bands[4], missed migrations, and previous injury or overaggressive bandaging resulting in impaired blood supply. Nonnative species placed in colder habitats are especially at risk, and peafowl (Pavo cristatus) and European starlings (Sturnus vulgaris) are two of the more common species seen with frostbite in upper midwestern North America. Although the majority of frostbite cases involve the feet, a syndrome of distal wing necrosis in falcons may be due to cold injury[5].


Numerous methods of adapting to colder environments, including arteriovenous countercurrent heat exchange and shunting via arteriovenous anastomoses, have evolved in birds[6]. As the environment cools beyond a bird’s ability to maintain homeothermy, heat is conserved in vital organs. Initially, vasoconstriction of vessels of the extremities is seen, with regular, intermittent vasodilation to preserve tissue viability. This vasodilation is known as the hunting reflex and has been demonstrated in the feet of pigeons, ducks, chickens, and fulmars[7]. As cooling continues, vasodilation ceases. Tissue damage occurs as a result of both direct freezing injury and ischemia resulting from impaired vascular supply.

Initially, freezing of tissue leads to extracellular ice crystal formation. As extracellular water freezes, the osmotic gradient is altered, resulting in intracellular dehydration. As crystals become larger, direct mechanical damage to cells also occurs. Initial freezing is typically less significant than resultant ischemic injury. Skin transplants from frostbitten areas to undamaged areas survive, whereas healthy skin transplanted to frostbitten areas does not[8]. Tissue ischemia occurs because of several reasons. Initially, vasoconstriction and sludging of blood results in inadequate blood supply. In a mouse model, a return of blood flow to apparent normal prefreeze rates, followed by the onset of a gradual sludging of blood 15 to 20 minutes postthaw, was seen[9]. Freezing, osmotic, and hypoxic damage to cells results in release of inflammatory mediators, especially by the vascular endothelium. Frostbite blisters in humans and experimentally frostbitten rabbit tissue have been shown to have markedly elevated levels of prostaglandin F2a and thromboxane B2 (a metabolite of thromboxane A2)[10]. Increased numbers of mast cells and polymorphonuclear leukocytes (PMN) are seen in frostbitten rabbit ears, and degranulation of these cells may also be a source of inflammatory mediators. Frostbitten rabbits treated at the time of rewarming, with antibodies blocking PMN adhesion, had significantly less tissue loss[11].

Inflammatory mediators trigger further vasoconstriction, platelet aggregation, and thrombosis, leading to a cycle of further microvascular damage, hypoxia, tissue damage, and inflammatory mediator release. Vascular inflammation and thrombosis may not be limited to the damaged extremity. Cardiac lesions have been associated with frostbite in birds[12]. In a study of 26 birds that died because of cold injury, six (23%) had aseptic vegetative valvular endocarditis. In another study, seven of 10 birds (70%) with frostbite had myocardial, valvular, or vascular lesions.14 Sterile vegetative valvular lesions have also been documented in mammals subjected to cold stress[13].

There are significant differences between avian and mammalian coagulation. Activated partial thromboplastin times in birds are much longer than those in mammals, indicating that the intrinsic coagulation pathway in birds is much less significant than in mammals[14]. The extrinsic coagulation pathway, which depends on tissue factor, is of primary importance in birds.18 Tissue factor expressed by damaged vascular endothelium may result in more significant coagulation abnormalities in birds. Alternatively, it is possible that bacterial infection secondary to frostbite leads to endocarditis. However, in one of the studies, only one of seven birds with vegetative valvular endocarditis had septic lesions.

Clinical signs

The most important part of frostbite diagnosis is the history. It is essential that treatment be initiated as soon as possible, before necrosis is evident. An early sign may be proprioceptive deficit of the affected extremity resulting from nerve injury. Pain may also be present, resulting in self-mutilation. Although coldstressed birds have been shown to have increased levels of serum lactate dehydrogenase, uric acid, and triglycerides19 and frostbitten birds typically have elevated levels of aspartate transferase and creatine kinase, biochemistries are less useful for diagnosing frostbite. In mammals, blister formation occurs during the first 24 hours. This is not typically evident in bird legs, possibly because of the anatomy of the scaled skin on the avian leg. As the injury progresses, edema may be seen within the first 24 hours. Demarcation of viable and nonviable tissue typically occurs slowly. Visual assessment of the line of demarcation in birds may take 3 to 6 weeks. As the line of demarcation forms, mummified tissue is evident. Long-term effects in surviving tissue in mammals may include increased susceptibility to cold reinjury, sensory loss, and osteoarthritis[15].


If tissue is still frozen at presentation, rapid rewarming in a warm (body temperature) water bath is indicated. Although past recommendations for birds have suggested more gradual rewarming[16], rapid rewarming has been shown to be more efficacious.

Pharmaceutical therapy should be directed toward maintaining optimal blood supply to damaged tissue, avoiding secondary infection, and providing analgesia. This therapy must be initiated before microthrombi form and irreversible tissue ischemia occurs.

Administration of NSAIDs serves to block inflammatory products of cyclooxygenases, such as prostaglandin F2a and thromboxane A2, preventing platelet aggregation and thrombosis. Intramuscular flunixin or ketoprofen at 5 mg/kg has been shown to significantly decrease thromboxane B2 in ducks, although flunixin resulted in significant muscle necrosis[17]. Pharmacokinetic data suggested that administration every 12 hours might be appropriate. Negative fecal occult blood tests suggested that administration of this single dose did not produce significant gastrointestinal bleeding. Ketoprofen has also been shown to have analgesic effects in ducks at a dose of 5 mg/kg IM[18].

Because frostbite is an extremely painful condition, analgesia is essential in patients. In addition to NSAID therapy, opioids may be used. Butorphanol at 1 mg/kg IM has been shown to be efficacious for analgesia in African grey parrots[19]. Antibiotic therapy should be directed against common skin flora and clostridial infection; pharmacokinetic data have shown that oral clavulanic acid–amoxicillin at a dosage of 125 mg/kg q8h is appropriate in Amazon parrots[20].

In passerine birds, clavulanic acid–amoxicillin at a dosage of 200 mg/kg q8h has been used without the detection of adverse effects. Early surgical debridement is contraindicated unless uncontrolled infection is present. The line of demarcation of viable tissue may take weeks to develop. An old adage from human medicine is, “frostbite in January, amputate in July,” and current recommendations in humans are to debride mummified tissue at 4 to 6 weeks or longer.


  1. Wallach JD, Flieg GM (1969) Frostbite and its sequelae in captive exotic birds. JAVMA 155:1035–1038
  2. Cooper JE (2002) Birds of Prey: Health and Disease, ed 3. Malden, MA, Blackwell Science, pp:80–81
  3. Coles BH (1997) Avian Medicine and Surgery, ed 2. Malden, MA, Blackwell Science, p:43
  4. Quesenberry KE, Hilyer EV (1994) Supportive care and emergency therapy, in Ritchie BW, Harrison GJ, Harrison LR (eds): Avian Medicine: Principles and Applications. Lake Worth, FL, Wingers Publications, pp:382–416
  5. Forbes NA, Harcourt-Brown NH (1991) Wing tip oedema and dry gangrene of raptors. Vet Rec 128:575–576
  6. Midtgard U (1989) Circulatory adaptations to cold in birds, in Bech C, Reinertsen RE (eds): Physiology of Cold Adaptation in Birds. New York, Plenum Press, pp:211–222
  7. Ostnes JE, Bech C (1998) Thermal control of metabolic cold defence in pigeons Columbia livia. J Exp Biol 201:793–803
  8. Weatherly-White RCA, Sjostrom B, Paton B (1964) Experimental studies in cold injury: The pathogenesis of frostbite. J Surg Res 4:17–22
  9. Bourne MH, Piepkorn MW, Clayton F, Leonard LG (1986) Analysis of microvascular changes in frostbite injury. J Surg Res 40(1):26–35
  10. Robson MC, Heggers JP (1981) Evaluation of hand frostbite fluid as a clue to pathogenesis. J Hand Surg 6:43–47
  11. Mileski WJ, Raymond JF, Winn RK, et al (1993) Inhibition of leukocyte adherence and aggregation for treatment of severe cold injury in rabbits. J Appl Physiol 74(3):1432–1436
  12. Calle PP, Montali RJ, Janssen DL (1982) Distal extremity necrosis in captive birds. J Wildl Dis 18(4):473–479
  13. Angrist AA, Oka M, Nakao K, Marquiss J (1956) Studies in experimental endocarditis. Am J Pathol 36:181–190
  14. Lewis JH (1996) Comparative Hemostasis in Vertebrates. New York, Plenum Press, pp:97–114
  15. Britt LD, Dascombe WH, Rodriguez A (1991) New horizons in management of hypothermia and frostbite injury. Surg Clin North Am 71(2):345–370
  16. Heidenreich M (1995) Birds of Prey: Medicine and Management. Malden, MA, Blackwell Science, p:211
  17. Machin KL, Tellier LA, Lair S, Livingston A (2001) Pharmacodynamics of flunixin and ketoprofen in mallard ducks (Anas platyrhynchos). J Zoo Wildl Med 32(2):222–229
  18. Machin KL, Livingston A (2002) Assessment of the analgesic effects of ketoprofen in ducks anesthetized with isoflurane. Am J Vet Res 63(6):821–826
  19. Paul-Murphy JR, Brunson DB, Miletic V (1999) Analgesic effects of butorphanol and buprenorphine in conscious African grey parrots (Psittacus erithacus erithacus and Psittacus erithacus timneh). Am J Vet Res 600(10):1218–1221
  20. Orosz SE, Jones MP, Cox SK, et al (2000) Pharmacokinetics of amoxicillin plus clavulanic acid in blue-fronted Amazon parrots (Amazona aestiva aestiva). J Avian Med Surg 14(2):107–112