Thermal injury remains a major medical problem throughout the world.1 Although many advances have been made in our understanding and care of burn injuries,2 there are numerous unresolved questions concerning the nature and treatment of this injury. A prerequisite to better understanding of the underlying pathophysiology of burn injuries and the development of novel burn therapies is the existence of in vivo animal models. A variety of animal models have been reported for scald and contact burns.3–6 However, many of these models are associated with risk to the investigator. Furthermore, the burn depth and extent may not always be controllable and consistent. To address these gaps, we recently developed a new burn infliction device that was used to establish two different in vivo burn models in swine and rats. This burn device was designed to create a consistent injury without having the heating source coming into direct contact with the animal’s skin. It was also designed to maximize investigator safety. This article describes the use of the novel burn inflicting device in rats and swine.
Objectives: The objective was to develop a novel animal model of burns in rats and pigs.
Methods: The model uses heat that is delivered via a radiant heater with an opening of 5 cm by 5 cm, set at 400°C, for 20 seconds. An advantage of this model is that the heating source does not come into direct contact with the animal, and the heat dispersion surrounding its center is very constant. The device was evaluated in 40 rats and seven pigs. With rats, three to four burns were created on each rat, resulting in a burn covering a total body surface area of 30% to 50%. In pigs, 16 burns were created on each animal.
Results: In rats, infliction of burns resulted in mortality rates of 0%–50% depending on the size of the burns and the rats. In pigs, the burns reepithelialized within approximately 3 weeks and resulted in hourglass contracted scars in two of three burns within 1 month.
Conclusions: The authors describe a novel animal burn model that utilizes radiant heat to create consistent burns that maximizes safety to the investigators and animals.
ACADEMIC EMERGENCY MEDICINE 2010; 17:514–520 © 2010 by the Society for Academic Emergency Medicine
We conducted a prospective animal experiment to evaluate the utility of a new burn infliction device in swine and rats. Animal handling was in accordance with national guidelines. The study was approved by the Ben Gurion University Institutional Animal Care and Use Committee.
The study was conducted in an accredited laboratory animal research facility affiliated with an academic medical center and university. Forty male Sprague-Dawley rats weighing between 200 and 300 g were used in the study, as well as seven large white juvenile domestic swine weighing between 25 and 45 kg. Animals were procured from the Animal Research Institute, Kibbutz Lahav, Israel.
Burn Infliction Device. To overcome the problems of cutaneous surface factors (topography, irregularities, moisture, etc.), the burn is inflicted remotely by a combination of radiation and hot air convection. A heating coil emits infrared/visible light radiation that heats the skin surface as well as the air within the chamber. The thermal energy that is transferred onto the target skin is the combination of the coil’s radiant heat and heat convection from the heated air. Adjustment of the heat regulator or the time of exposure helps control the depth of injury.
The burn infliction device consists of a heating element (a 350-W electrical coil), which is suspended in a pyramid-shaped stainless-steel box connected to a power supply (Figure 1). The dimensions of the steel box are shown in Figure 2. A thermocouple (Eurotherm 2132, Eurotherm, Leesburg, VA) is used to control the temperature. One side of the box is left open, to allow the heat to escape. The aperture is connected by aluminum prongs to an aluminum plate having a matching window of the heating box. The aluminum plate is hinged to the side of an insulating 7-mm-thick Teflon plate, with a window the size and shape of the desired burn area situated in front of the apertures of the aluminum plate and the heating box. The purpose of the Teflon plate is to protect and insulate the skin surrounding the burned area from direct contact with the hot stainless-steel box or any other heat. Because it is an insulator, the Teflon plate remains unheated. The Teflon plate window can have different sizes and shapes, which will determine the size and shape of the burn. The burn infliction device is placed in direct contact with the animal’s skin and a burn is created by radiation and transfer of heat from the heated air to the underlying skin through the opening in the Teflon plate. While the unheated Teflon comes into direct contact with the skin, the heating element that radiates heat does not, which is unique to this model. As in clinical practice, most burns have a deep area surrounded by more superficial, tapering off burns. This has been classically described by Jackson,7 who refers to the burn’s central zone of necrosis surrounded by a secondary zone of stasis that may or may not progress to full-thickness necrosis over time. To simulate a burn that has a deeper injury at its center, the shape of the device was specifically designed to transfer more heat at the center of the burn with lesser amounts of heat transferred at the periphery. This was done by having the heating element smaller than the aperture. Thus, the hot air is responsible for the more superficial burn, and toward the center, more and more heat is added as radiant energy from the heating element. This pattern was verified by measuring the temperature at the aperture of the Teflon opening. The following setting has been found to be the most useful: the thermo controller was set at a temperature of 400°C, which produced 350°C at the center while it tapered off as the distance from the center increased (Figure 3). The temperature was constant and similar for all measurements because the device contains a thermocouple that controls the temperature to within ±0.5°C. The total cost of all the necessary parts to assemble the burn infliction device is roughly U.S. $200.
Rat Burns. The rats were acclimated to the laboratory for 2 weeks prior to beginning the study and had free access to water and food at all times. The rats were anesthetized with inhaled 0.5%–5.0% isoflurane. The hair of the torso was clipped with an electrical clipper, and two, three, or four 5-cm by 5-cm square burns were inflicted with the burn device set at a temperature of 400°C for 20 seconds. Large burns are created with the device by applying it multiple times to adjacent areas without minor gaps or overlap by careful planning and marking of the skin prior to burning. The device was placed directly on the skin surface with the skin coming into contact with the insulating layer of Teflon. The rats were subjected to burns covering 30%–50% of their total body surface area. The animals were then resuscitated by administration of 5 mL of intraperitoneal normal saline immediately after burn injury.
Porcine Burns. The pigs were fasted for a minimum of 5 hours prior to induction of anesthesia. Approximately 30 minutes prior to induction of anesthesia, they were given acepromazine (1.1 mg/kg intramuscular [IM]) and atropine (0.02 mg/kg IM). The animals were then anesthetized with ketamine (20 mg/kg IM) and xylazine (2 mg/kg IM), intubated, maintained under anesthesia with 1%–5% isoflurane inhalant anesthetic as needed, and delivered through a volume regulated respirator (Vetaflex 5 veterinary anesthesia machine, Ohio Medical, Madison, WI) with a tidal volume of approximately 15 mL/kg. Once anesthetized, the animals were placed prone with a heating lamp over them to maintain their body temperature. A rectal thermometer was placed to assure that each animal’s core temperature was maintained around 37–38°C. An intravenous catheter was placed in an ear vein, and lactated Ringer’s solution was administered intravenously at a rate of 22 mL/kg/hr.
The hair over the pig’s dorsum was clipped with electric clippers, and the entire back and lateral flanks were prepared with three sequential scrubs of 70% alcohol. Sixteen burns were then created with the burn infliction device set at a temperature of 400°C for 20 seconds, evenly divided between the two sides of the pig. All burns were treated by fast and selective enzymatic debridement using Debrase gel dressing (MediWound, Yavne, Israel),8 which removes the coagulated tissue during a 4-hour application, allowing direct visualization and diagnosis of the clean bed and the residual viable structures.8 All burns were dressed with nonadherent gauze and all pigs were followed for a period of 28 days. Animals were observed frequently for signs of pain or discomfort and treated with 0.01 mg/kg IM buprenorphine as needed.
Measures and Outcomes
For the rat burns, the primary outcome was mortality within 24 hours after injury. With the porcine burns, the main outcomes were the depth of the burn and the time to complete reepithelialization treated by different dressings. Other secondary outcomes included wound infection and scarring leading to wound contraction. To determine the depth of the burn, besides visual diagnosis of the predebrided and postselective debrided bed, full-thickness biopsies were taken from the center of the burns 1 hour after injury. This was done to ensure that the center was a full-thickness burn. The thickness of the dermal residues that correspond to the burn depth can be easily diagnosed by the different bleeding patterns. The deep dermis is characterized by larger, widely spaced areas of bleeding, with more, smaller bleeding vessels at the mid dermis and typically, numerous, very small capillaries at the papillary dermis.
The depth of the burn was determined histologically using hematoxylin and eosin staining by the presence of denatured collagen and necrotic cells such as the epithelial cells surrounding the hair follicles, sebaceous glands, and endothelial cells. The wound was considered reepithelialized if its surface was opaque and dry, as opposed to the shiny wet appearance of a nonreepithelialized wound. The presence of wound infection was based on the clinical appearance of the burns (i.e., the presence of erythema, warmth, purulent discharge, and systemic fever). A scar was considered to have healed with contraction if it had an hourglass shape (i.e., its center was shorter than its periphery). A contracted scar is considered a poor outcome, generally due to a very deep or full-thickness burn.
Descriptive statistics were used to describe outcomes. Continuous variables are presented as means with standard deviations (SDs). Binomial data are presented as the percent frequency of occurrence with 95% confidence intervals (CIs). Data analysis was performed with SPSS 16.0 for Windows (SPSS Inc., Chicago, IL). No a priori sample size determination was made. The sample size was mostly based on feasibility.
Full-thickness burns were created on 40 rats in four equal groups of ten rats each (Figure 4). The number of burns created on each group of rats was three to four, as noted in Table 1. We also varied the weight of the rats (Table 1). There was a direct relationship between the size or number of the burns and the mortality within 24 hours of injury, ranging from 0% to 50% (Table 1). All burns created were of full thickness (Figure 5).
|Number of Rats (Weight)||Number of Burns||Total Size of Burns, %||Mortality, % (95% CI)|
|10 (200–250 g)||2||15–20||0 (0–30)|
|10 (200–250 g)||3||30||10 (2–40)|
|10 (200–250 g)||4||50||50 (24–76)|
|10 (>300 g)||3||30||10 (2–40)|
Sixteen burns were created on each of seven pigs, for a total of 112 burns (Figure 6). The mean (±SD) burn depth was 1.2 (±0.2) mm (Figure 7). The mean (±SD) time to complete reepithelialization was 18.7 (±6.2) days (Figure 8, Table 2). All burns were completely reepithelialized within the study period; there were no wound infections. Of all burns, approximately 60% healed with an hourglass scar (Figure 9). The remaining burns healed with a square-shaped scar. The rate of contracted scars did not differ by location (data not shown).
|Number of pigs||7|
|Number of burns/pig||16|
|Mean (SD) depth of burn, mm||1.2 (±0.2)|
|No. (%) infections||0 (0)|
|Mean (SD) time to complete reepithelialization, days||18.7 (±6.2)|
|No. (%) hourglass, contracted scars||74 (66.1)|
|Mortality, no. (%)||0 (0)|
Further advances in our understanding and treatment of burns will require reliable and reproducible in vitro and in vivo experimental models. The most accepted animal model for studying the healing of burns is the pig, due to the resemblance of its skin to that of humans.9 Both pig and man have a thick epidermis and dermis with a dermal–epidermal ratio of at least 10:1. The presence of well-formed dermal rete pegs and subcutaneous fat is also common to both man and swine, although the structure of the fat layer is different. Importantly, both man and swine heal by reepithelialization and not wound contraction, due to the lack of a panniculus carnosus. Porcine dermal collagen is also similar in its thickness and architecture to that of man. In spite of their different skin structure, smaller animal models, such as rats, are also used for specific endpoints due to their relative ease of use and lower cost.
Most animal burn models have used direct contact with heated metal objects or exposure to scalding water to inflict burns. The factors that determine the burn’s depth and size depend on the area’s size, time of exposure, initial available heat source (energy capacity), the quality of heat transfer from the source to the skin that depends on contact variables such as matching between the two surfaces; consistency (hardness) of both surfaces; and presence of moisture, grease, and pressure. The characteristic of the heat source (flame, water, hot surface) will determine the characteristics of the burn due to the differences in the mechanism and characteristics of the heat transfer.
Heated blocks with hot surfaces are easy to use; if their mass is big enough their heat capacity suffices but must be maintained constant. The contact with the skin is difficult to standardize due to differences and specificities of anatomical areas. The flat surface exerts different pressures in different points with higher pressures in convex and hard skin areas and at the heated block’s sharp corners. Minute quantities of substances such as water or grease may vaporize at the point of contact with the hot surface, changing the heat transfer process. In these models the pressure on the skin should also be standardized.
Use of boiling water controls some of these variables (contact, adherence, temperature, and heat capacity) but area and time of exposure are more difficult to control. One of the disadvantages of the water model is the risk of accidental scald burn to the investigator. Cuttle et al.,10 describe a heated water bottle with an elastic membrane bottom that, when applied directly to the skin surface, may fit and adjust better to skin surface irregularities and eliminate the scalding risk. This system partially overcomes the surface problem of the heating block’s solid surface, but the elastic membrane still cannot adhere intimately to the smallest cutaneous topographic irregularities. Open flame is rarely used, probably due to the investigator’s and animal welfare committee’s psychological reluctance and difficult heat capacity control.
In all these models, the goal is to have a uniform burn of controlled depth. None of these models offer the option of reconstituting the real clinical situation present in most human burns, which are not of uniform depth. Our goal was to develop a simple, safe, burn infliction device that would result in adjustable and controlled temperatures and injuries, even those that are not uniform. The burn infliction device and animal models reported here meet the previously described criteria of an animal burn model. They are easy to use, inexpensive, and safe for both animals and investigators. Furthermore, our results demonstrate that the burn infliction device results in adjustable and consistent temperatures and heat capacity of the heat source, with heat transfer and distribution (not based on contact) that is not impaired by surface irregularities, resulting in controlled depths of injury.
Unlike other burn infliction devices, in the current device the heating element itself also does not need to come into direct contact with the animal’s skin, because it uses radiant heat, thus adding to its safety. The application of radiant heat also circumvents the problem of achieving consistent and even contact with the skin’s surface required with hot metal contact burn models.
Noninvasive measurement of burn depth is very difficult and not validated. Selective, fast enzymatic debridement with Debrase8 allows direct visualization, assessment, and photographic documentation of the entire surface of the wound’s viable bed a few hours after burn infliction, whereas biopsies, besides being invasive, give us only points of reference and not the entire surface.
The healing of the burn by contraction depends on the potential movement of the wound’s edges, which depends on the wound’s shape (square or round) and laxity of the surrounding skin.11 Although the burns were square in shape, only 60% of them healed with a contracted, hourglass-shaped scar, reflecting varying degrees of skin thickness and lateral tensions on different areas of the body.
When applied to average-sized rats in a mortality rate study, the number (relative size) and depth of the burns can be selected to result in a desired and predictable mortality rate to be used as a baseline when comparing various potential treatments. When applied to swine, application of the device set to a temperature of 400°C for 20 seconds results in deep dermal burns with a central area of full thickness. It occupies approximately one third of the entire burned surface and requires approximately 19 days to reepithelialize. The burns that healed with an hourglass-shaped scar, indicating scar contraction, were in areas of relative higher skin mobility and higher tension caused by breathing (chest) or eating (abdomen; Figure 9). Thus, the new burn infliction device can be used both in small and in large animals, resulting in reproducible injury.
Our study was limited to a single site and few investigators. Thus, it will need to be validated by other investigators and other sites. Our study is limited to short-term follow-up; we cannot comment on longer-term outcomes. We did not measure interobserver agreement for the primary outcome. However, wound reepithelialization is a well-accepted and common measure of wound outcomes, even by gross inspection.
This study describes a novel burn infliction device based on radiant heat that can be used to create burns of varying depth and size in both small (rats) and large (pigs) animals.