Frostbite in Children: A Case Series and Development of a New Iloprost-Driven Protocol

Protocol Development

The Yukon Protocol was the starting point for the development of our pediatric protocol, which we named the Frostbite Management in Children Protocol. The protocol was developed after reviewing all published literature on frostbite and uses a multimodality treatment approach to combat the poor outcomes commonly seen in severe cases (grades 3-4). It includes rewarming, grading of injury, involvement of a multidisciplinary team, specific medical management, and applicability of hyperbaric oxygen (HBO) therapy (Figure 2). There are no published data on the pharmacokinetics of iloprost in children. Due to the lack of pediatric-specific data, iloprost dosages were extrapolated from adult dosing guidelines in our protocol. Iloprost is used for grades 2 to 4 frostbite. Our protocol differs from the adult protocols by suggesting that tPA (Activase rT-PA) with low-molecular-weight heparin be considered for grade 2 injuries. ¹³ For grades 3 to 4 frostbite, tPA is recommended and may be given intraarterially. Exception would be the unavailability of interventional radiology or specifics of the case. Furthermore, HBO therapy is added for grades 3 to 4 injuries or any other case where the response to iloprost and tPA is thought to be inadequate.

OPEN IN VIEWER

Figure 2.

Frostbite management in children protocol.

This protocol is more aggressive than the adult protocols and overcorrects for the potential of underestimating the frostbite severity. We know that after rewarming, frostbite injury characteristics change through a 24- to 48-h window making it difficult to identify a grade 2 versus 3 or 4 injury. This window is when iloprost and alteplase need to be administered to be effective. Patients commonly present late to the hospital; therefore, rapid aggressive intervention is necessary. A review of 3 subsequent cases treated using the protocol was performed. Outcomes recorded included digital amputation rate, motor/sensory recovery, and adverse effects of treatment.

Case 1

A 9-year-old female, 33.7 kg, with Fitzpatrick skin type V, ¹⁴ was exposed to −29 °C temperature for approximately 10 min on her walk home from the bus stop without any gloves. Her fingers were rewarmed at home under hot water. The next morning, discoloration and blister formation appeared on her left middle finger. She presented to our center where her left middle finger was cyanotic to the level of the proximal phalanx and determined to have grade 3 frostbite.

Given that the patient presented within 72 h from rewarming but more than 24 h, the protocol called for the administration of iloprost but not tPA. An iloprost infusion was started at a dose of 0.05 µg/kg/h and titrated up to the maximum dose of 0.15 µg/kg/h over 2 h. It was well tolerated, and the patient was transferred to the ward from the PICU. On day 2 of iloprost treatment, there was still poor perfusion to the tip of the left middle finger. The treatment plan was reconsidered. Despite the patient being outside the 24-h postrewarming window, a one-time dose of tPA was administered as an infusion over 6 h at 0.05 mg/kg/h. The patient was started on enoxaparin 72 h after tPA at 1 mg/kg every 12 h. Iloprost was also administered for 6 h each day for the 5 consecutive days. Hyperbaric Oxygen chamber therapy was considered but never initiated, as the closest available HBO chamber was fully booked. By day 5 of iloprost treatment, there was an improvement in warmth and color of the fingertip with no new blisters. At 14 days postinjury, the patient had full recovery with good perfusion, full range of motion and normal sensation (Figure 3).

OPEN IN VIEWER

Figure 3.

(A) Patient 1: Hand injury Frostbite of middle finger, (B) Patient 1 after treatment and recovery.

Case 2

A 15-year-old male, 79.1 kg, with Fitzpatrick skin type II was drinking alcohol, using marijuana and mushrooms with friends outside in January only wearing light gloves and one boot and was exposed to −35 °C for 1.5 h. He rewarmed at home with many blankets. He presented to the emergency department 36 h after the injury, with cyanosis on the tips of the fingers, blistering, and swelling of all digits on the right hand, as well as the index and middle fingers of the left hand. The right foot was also swollen with large blisters over the plantar surface. The frostbite was determined to be grade 2 (Figure 4).

OPEN IN VIEWER

Figure 4.

(A) Patient 2: Hand and Foot Frostbite within 24 h of injury showing signs of swelling, blistering, and discoloration. (B) Patient 2, 28 days postinjury, showing full recovery.

Iloprost was administered at 2 µg/h titrated up to 10 µg/h and administered for 6 h daily for 5 days. After the first dose of iloprost, his cyanosis improved. He received bedside blister debridement and wound dressings for his feet and hands. The patient was discharged on day 6 after the full course of iloprost treatment. He was seen again at 31 days postinjury, and all areas affected by the frostbite had completely recovered (Figure 4).

Case 3

A 16-year-old male, 75.1 kg, with Fitzpatrick skin type V, was shovelling snow in March. Though wearing gloves, he was exposed to −21 °C temperatures for approximately 1 h before rewarming his hands. He presented to the emergency department 3.5 h after rewarming where he was noted to have discoloration, swelling, and blistering on the fingers of both hands. The frostbite was determined to be grade 2.

An Iloprost infusion was started at 2 µg/h and titrated up to 10 µg/h. Because perfusion was improving with the iloprost, alteplase was not used. This patient experienced pain in all digits, especially the left index finger. Blisters were debrided at the bedside. He was discharged after 8 days. By 46 days postinjury, both hands had healed, though he had lost the nail plates on all digits in both hands except the left thumb. He had good range of motion but still complained of tingling in his fingertips. The altered sensation was noted to have disappeared, and the nail plates to have regrown at 3 months post injury (Figure 5).

OPEN IN VIEWER

Figure 5.

(A) Patient 3: Bilateral hand Frostbite within 24 h of injury showing signs of swelling, blistering, and discoloration on both palmar and dorsum aspect of digits. (B) Patient 3: 46 days postinjury showing signs of hyperpigmentation on all fingers and development of new nails.

The cases are summarized in Table 2. Overall, the protocol was well tolerated by patients with no significant complications. There were no concerns from hospital staff regarding implementation of the new protocol.

OPEN IN VIEWER

Table 2.

Summary of Frostbite Treatment and Outcome in Paediatric Frostbite.

Case #	Age & Sex	Frostbite Grade / Body Part	Iloprost	LMW Heparin	TPA (Alteplase)	Analgesics	Wound Care	Outcome
1	9 years—female	Grade 3 hands	Infusion started at 1.68 µg/h and increased to a max of 5.04 µg/h (5 days)	Enoxaparin 100 mg/mL injection 34 mg 1 mg/kg/dose, subcutaneous (3 days)	50 mg in 50 mL 0.05 mg/kg/h, IV, continuous (1 day)	Acetaminophen liquid 480 mg PO q6 h PRN	Polysporin and Bandages Blisters drained	Full ROM Perfusion and sensation normal No tissue loss, no amputation Length of stay: 5 days
2	15 years—male	Grade 2 hands & right foot	Infusion started at 2 µg/h and increased to a max of 10 µg/h (5 days)	NA	NA	Acetaminophen tablet 650 mg, oral, q6 h PRN Ibuprofen tablet 400 mg oral, q6 h PRN Acetylsalicylic acid-81 mg for 4 weeks	Blisters drained & deroofed Wrapped with Mepilex Ag and gauze	Full ROM All areas healed. Some sensitivity on planter surface of right foot. No loss of digits Length of stay: 6 days
3	16 years—male	Grade 2 hands	Infusion started at 2 µg/h and increased to a max of 10 µg/h (5 days)	NA	NA	Ibuprofen 400 mg PO q6 h PRN Ketorolac-30 mg 15 mg q6 h sc Acetaminophen 1000 mg PO q6 h PRN Hydromorphone twice Pregabalin-75 g bid Acetylsalicylic acid-81 mg PO daily	Blisters drained & deroofed	Full ROM Hyperpigmentation on all fingers Loss of all nail plates except left thumb No loss of digits Length of stay: 8 days

Rewarming

Rewarming should be initiated after the risk of further exposure to cold temperatures has passed. Patients should undergo rapid rewarming in normal saline or water at 37 °C to 39 °C. ¹⁷ This typically takes 30 to 60 min. If in a hospital setting 2% chlorhexidine gluconate may be added to the rewarming solution (30 mL per litre of fluid). ¹⁷

Grading

Grading of the frostbite injury should take place after rewarming is completed. The Cauchy grading scale is based on the proximal extent of cyanosis in the worst affected digit. ⁶ Facial frostbite (ie, ears, cheeks, and nose) should also be assessed and considered for infusion therapy, although experience with infusion therapy for such cases is limited. There may be difficulty with the Cauchy grading system in children and in patients with darker skin color. The degree of cyanosis may be worse than is first estimated. We propose that the extent of the blistering, whether clear or hemorrhagic should be taken into consideration. For example, even if cyanosis does not seem to progress to the MCP joints, but blistering does, it might be best to categorize these injuries as grade 3 and be more aggressive with treatment including both iloprost and tPA. We learned about this risk of blistering from the case presented in the introduction. That patient only had clear blisters that extended to the MCP joints, and he went on to develop gangrene and loss of portions of his digits.

Admission to ICU and Involvement of Multidisciplinary Team

Patients are assessed by emergency room physicians and the plastic surgery service. Those with grades 2, 3, or 4 injuries are considered for iloprost and tPA. They are admitted to the PICU, where a review by the multidisciplinary team (plastic surgery, intensive care staff, hematology, and interventional radiology [if needed]) takes place.

Iloprost

A 2011 controlled trial indicated the effectiveness of iloprost in minimizing amputation rates in grades 2, 3, and 4 injuries in an adult population. ¹⁸ There are, however, only a few reports of the use of Iloprost in grade 4 injuries in children and these involved adolescent-aged patients. Patient history must be carefully evaluated for any contraindications to the use of Iloprost: hypersensitivity to iloprost, pregnancy, high risk of hemorrhage, hepatic injury, recent cerebrovascular event, and heart disease or arrhythmia. If less than 72 h from the time of rewarming, iloprost is indicated as follows:

Patients Under 40 kg

Patients Over 40 kg

Continue Infusion for 6 h

Deviations from Protocol

In Case 1, given the significant risk of amputation, alteplase treatment was supported despite the injury occurring more than 24 h after rewarming. The Yukon protocol and the Colorado protocol both suggest a 0.15 mg/kg loading dose and others recommend a 0.5 mg/kg bolus.^5,7 These were not deemed to be a suitable option for our pediatric patient, and an alternative approach of low-dose therapy was used—alteplase (0.01-0.06 mg/kg/h). We know that there is a significant, albeit small, risk of bleeding with the use of alteplase in children that may be higher than seen in adults, and as such the optimal pediatric dose is not known. Lower dose regimes may be equivalent in efficacy to higher dose regimens based on case series on nonfrostbite-related arterial thrombosis. ²² If low-dose alteplase is used, anticoagulation via low-molecular-weight heparin should be continued for 72 h; however, some centers continue for 14 days.

Pain Management

Adequate pain relief is important from the rewarming period onward. Some cases required narcotics, and our center used hydromorphone; this has been supported in the literature.^27,28 Acetaminophen can also be taken for managing pain from frostbite injury. Our protocol suggests NSAIDs, for the benefit of the antiplatelet effect, but should be delayed until after alteplase administration is completed. Acetylsalicylic acid can be used for its anti-inflammatory and antiplatelet effects. No published studies compare the effects of these options in frostbite cases. ¹⁷ Gabapentin or pregabalin is helpful for the control of neuropathic pain associated with frostbite. ²⁹

Hyperbaric Oxygen Therapy

The Yukon adult frostbite protocol does not include recommendations for HBO therapy. There is some weak evidence that it may have additional benefits in tissue preservation and minimizing amputation rate when added to iloprost as an early or late treatment option.^23,24 Accelerated healing occurs in many types of wounds due to increased tissue oxygenation from HBO therapy. ²⁵ We have included HBO therapy recommendation in our pediatric frostbite protocol for grades 2 or greater daily for up to 14 days, if the therapeutic response to iloprost and alteplase is deemed inadequate by day 2. The article by Magen et al proposes recommendations for HBO therapy, but these recommendations are not yet supported by large-scale clinical trials. However, we have included it as an adjunctive option based on theoretical benefits. ²⁶ We recognize that there are practical issues around access to HBO in many centers, including our own.

Delayed Treatment

Our protocol suggests that patients must present within the first 72 h to be considered for iloprost, but we may not be rigid on that recommendation. Time limit recommendations of less than 72 h for iloprost and less than 24 h for alteplase should be considered suggestions and not absolute. There are case reports of benefit from late treatment and in our series, case 1 is an example of this for late alteplase treatment. ²² Smaller regional centers do not have Iloprost available. After discussions with our intensive care team, urgent alteplase administration could be recommended before transfer to the hospital for grade 3 or 4 disease. If iloprost and alteplase are not used, HBO should still be considered.

Future Directions

This case series is intended to be the starting point for research in pediatric frostbite injuries treated with iloprost and alteplase. It describes the journey that led to the creation of the Frostbite Management in Children Protocol and the following cases treated under this protocol. For future studies, we hope to further support the validity of the suggestions presented as more cases are treated under this protocol. Furthermore, there is a frostbite website and registry in Canada organized by Dr Alex Poole that is a useful resource for the public and health care professionals (https://frostbitecare.ca/). With collection of data from multiple pediatric centers, this registry has the potential to help modify and update our protocol, and other protocols, to treat severe frostbite in children.