The articles prior to January 2010 are part of the back file collection and are not available with a current paid subscription. To access the article, you may purchase it or purchase the complete back file collection here

Case Reports 

Severe Frostbite of the Knees After Cryotherapy

Charles K. Lee, MD; Jeff Pardun, MD; Rudolf Buntic, MD; Darrell Brooks, MD; Mark Kiehn, MD; Harry J. Buncke, MD

Cryotherapy has been used to treat pain and inflammation since the time of Hippocrates.1 Ice, snow, cold water, and cold compresses have been used to treat a multitude of soft-tissue traumas.2 More recently, cryotherapy has been used increasingly in sports injuries and in the postoperative orthopedic setting.3 However, a number of complications from cryotherapy have been reported–most commonly frostbite and peripheral nerve injury–that point out its benefits but also its dangers.1-5 These previously reported complications have been diverse in location and severity. This article reports a significant complication of cryotherapy as a result of a relatively common regimen of application of ice packs to knees in a postoperative setting.

Case Report

A 53-year-old man was referred to our microsurgery service 4 weeks after bilateral patellar tendon repair. The patient had sustained bilateral patellar tendon ruptures after a fall down his home stairs. He underwent standard patellar tendon repair through midline incisions using braided polyester suture. He had no intraoperative complications.

Postoperatively, the patient was admitted for 3 days, during which he had continuous cryotherapy with ice packs circulating cold water from an ice cooler. On postoperative day 3 he was discharged and used continuous cryotherapy at home for an additional two weeks. The home regimen consisted of an OpSite dressing (Smith & Nephew, Largo, Fla), a paper towel interface, and the cooling pad with circulation connected to the enclosed water and ice cooler. At 2 weeks postoperatively, skin changes were noted that alerted him to see his orthopedic surgeon. At 4 weeks postoperatively he was referred for full thickness skin eschars of the bilateral knees. He was admitted for debridement, dressing changes, and wound coverage (Figure 1).

Figure 1: Necrosis of skin over knee

Figure 2: After debridement, exposed pattaler tendon, and suturing

Figure 1: Initial presentation with full thickness necrosis of skin over knee. Figure 2: After debridement, exposed pattaler tendon, and suturing.

The patient underwent debridement of clearly nonviable tissue followed by subatmospheric pressure dressings. Further debridement and subatmospheric pressure dressings were applied on two more occasions until the wounds were clean and granulating (Figure 2). Bacterial cultures obtained at the first debridement revealed the presence of Enterobacter cloacae and Staphylococcus epidermidis. The patient was treated with intravenous antibiotics at initial debridement. After debridement the wound dimensions were 15×13 cm on the right and 17×13 cm on the left with exposed patellar tendon and patella. The patellar tendons were severely compromised, with evidence of infection and necrosis. After a third debridement, the patient underwent bilateral rectus abdominus microvascular transplants to cover both knees and remaining patellar tendons (Figure 3). His postoperative course was without complication and he was discharged on postoperative day 5. By 3 weeks the flaps had healed well with no evidence of open wounds. The patient began to ambulate with knee immobilizers at 6 weeks postoperatively and without immobilizers at 10 weeks postoperatively.


This case represents a severe frostbite injury after cryotherapy. With proper instruction and use of a cooling device, complications like this are mainly avoidable. In this case, the patient had minimal padding between the cooling wrap and skin. In addition, the patient used the device continuously for two weeks.

Frostbite occurs when ice crystals form in the intracellular and extracellular space. During the cooling process, the extracellular ice crystals form and osmotic pressure increases, drawing water out of the cells. This leads to intracellular dehydration with an increase in intracellular electrolytes, proteins, and enzymes that lead to cell death. Additionally, vascular endothelial damage leads to intravascular thrombosis and reduced blood flow. Arteriovenous shunting occurs at the capillary level and end organ tissue damage is compounded. During the warming process, an influx of fluid back into the cells causes intracellular swelling. The warming process also allows reflow with vasodilation and reactive hyperemia leading to increased inflammatory mediators, causing further cell death.6

Cryotherapy works by three main processes. First is the reduction in the inflammatory process by inducing a hypometabolic state. Decreasing inflammation decreases the amount of cellular damage by inflammatory mediators, ultimately reducing the amount of capillary permeability and thereby decreasing edema. Second is the decrease in hematoma formation, which is produced from vasocapillary constriction and decreased blood flow. Third is the induction of analgesia by cold. This is thought to be due to decreased nerve conduction velocity and decreased muscle spasm. In combination, cryotherapy is an ideal postoperative therapy which decreases pain, inflammation, hematoma, and the amount of postoperative narcotic usage.2

Figure 3: Three months after free tissue transplant

Figure 3: Three months after free tissue transplant.

However, cold also has its dangers. The temperature at which the beneficial effects end is 15° C, where the permeability of the lymphatics in conjunction with edema increases. Additionally, slow cooling is more destructive than fast cooling because it allows intra- and extracellular ion and protein shifts to occur over a longer period of time. Current recommendations for cryotherapy include 20-30 minutes of cryotherapy with a maximum time of 40 minutes, always with a protective covering (usually a towel) between the cryotherapy wrap and the skin. The cycle can be repeated every 2 hours while the patient is awake.

The most common complications of cryotherapy are frostbite and peripheral nerve palsy. Frostbite has been documented in the literature and anecdotally in the head and neck and extemities. In the head and neck, frostbite of the upper and lower eyelids requiring full thickness skin graft reconstruction has been reported.7 In the extremities a case of partial thickness frostbite has been reported over the calf where the patient applied an ice pack directly over the skin.4 Another patient sustained full thickness necrosis on the dorsum of her foot requiring skin grafting. The patient reportedly fell asleep with frozen ice chips wrapped in a towel for at least 40 minutes.3

Peripheral nerve palsies have also been documented, particularly around the peroneal and ulnar nerve. These injuries are mainly transient and with appropriate splinting and range of motion exercising there is little permanent sequelae. One paper documented 5 cases of transient peroneal nerve palsy. Only one case had EMG documented axonotmesis of the peroneal nerve with complete paralysis of the great toe and ankle dorsi flexion. The patient recovered after 19 months.1

In all of these cases, the patients used cryotherapy inappropriately. Cryotherapy was continued beyond the 20-30 minute recommended time, or there was no protective layer between the ice and skin. Absolute contraindications for cryotherapy include: Raynaud’s phenomenon, cold allergy, cryoglobulinemia, and paroxysmal cold hemoglobinuria. Relative contraindications are arthritic conditions, pheochromocytoma, anesthetic skin, and severe cardiovascular disease. Although the incidence of complications from cryotherapy appear rare, estimated at 0.00225%,5 it is likely that this number is an underestimation as there are many unreported cases in conjunction with the increased use of continuous cryotherapy in the postoperative setting. We see this as an opportunity to emphasize the importance of education between the patient and doctor about this device and point out its potentially devastating risks.


  1. Drez D, Faust DC, Evans JP. Cryotherapy and nerve palsy. Am J Sports Med. 1981; 9:256-257.
  2. Swenson C, Sward L, Karlsson J. Cryotherapy in sports medicine. Scand J Med Sci Sports. 1996; 6:193-200.
  3. Graham CA, Stevenson J. Frozen chips: an unusual cause of severe frostbite injury. Br J Sports Med. 2000; 34:382-383.
  4. O’Toole G, Rayatt S. Frostbite at the gym: a case report of an ice pack burn. Br J Sports Med. 1999; 33:278-279.
  5. Wilke B, Weiner RD. Postoperative cryotherapy: risks versus benefits of continuous-flow cryotherapy units. Clin Podiatr Med Surg. 2003; 20:307-322.
  6. Su C, Lohman R, Gottlieb LJ. Frosbite of the upper extremity. Hand Clinics. 2000; 16:235-247.
  7. Quist LH, Peltier G, Lundquist KJ. Frostbite of the eyelids following inappropriate application of ice compresses. Arch Ophthalmol. 1996; 114:226.


Drs Lee, Pardun, Buntic, Kiehn, Brooks, and Buncke are from The Buncke Clinic, San Francisco, Calif.

Correspondence should be addressed to: Charles K. Lee, MD, The Buncke Clinic, 45 Castro St, 140N, San Francisco, CA 94114.



Sign up to receive

Journal E-contents