|Home | About | Journals | Submit | Contact Us | Français|
Smoke inhalation injury frequently increases the risk of pneumonia and mortality in burn patients. The pathophysiology of acute lung injury secondary to burn and smoke inhalation is well studied, but long-term pulmonary function, especially the process of lung tissue healing following burn and smoke inhalation, has not been fully investigated. By contrast, early burn excision has become the standard of care in the management of major burn injury. While many clinical studies and small-animal experiments support the concept of early burn wound excision, and show improved survival and infectious outcomes, we have developed a new chronic ovine model of burn and smoke inhalation injury with early excision and skin grafting that can be used to investigate lung pathophysiology over a period of 3 weeks.
Eighteen female sheep were surgically prepared for this study under isoflurane anesthesia. The animals were divided into three groups: an Early Excision group (20% TBSA, third-degree cutaneous burn and 36 breaths of cotton smoke followed by early excision and skin autografting at 24 h after injury, n = 6), a Control group (20% TBSA, third-degree cutaneous burn and 36 breaths of cotton smoke without early excision, n = 6) and a Sham group (no injury, no early excision, n = 6). After induced injury, all sheep were placed on a ventilator and fluid-resuscitated with Lactated Ringers solution (4 mL/% TBS/kg). At 24 h post-injury, early excision was carried out to fascia, and skin grafting with meshed autografts (20/1000 in., 1:4 ratio) was performed under isoflurane anesthesia. At 48 h post-injury, weaning from ventilator was begun if PaO2/FiO2 was above 250 and sheep were monitored for 3 weeks.
At 96 h post-injury, all animals were weaned from ventilator. There are no significant differences in PaO2/FiO2 between Early Excision and Control groups at any points. All animals were survived for 3 weeks without infectious complication in Early Excision and Sham groups, whereas two out of six animals in the Control group had abscess in lung. The percentage of the wound healed surviving area (mean ± SD) was 74.7 ± 7.8% on 17 days post-surgery in the Early Excision group. Lung wet-to-dry weight ratio (mean ± SD) was significantly increased in the Early Excision group vs. Sham group (p < 0.05). The calculated net fluid balance significantly increased in the early excision compared to those seen in the Sham and Control groups. Plasma protein, oncotic pressure, hematocrit of % baseline, hemoglobin of % baseline, white blood cell and neutrophil were significantly decreased in the Early Excision group vs. Control group.
The early excision model closely resembles practice in a clinical setting and allows long-term observations of pulmonary function following burn and smoke inhalation injury. Further studies are warranted to assess lung tissue scarring and measuring collagen deposition, lung compliance and diffusion capacity.
The short-term pathophysiology of acute lung injury secondary to burn and smoke inhalation has been studied extensively [1-4] but there are few studies of long-term pulmonary pathophysiology following burn and smoke inhalation . The purpose of present study was to develop an animal model of smoke inhalation injury and cutaneous burn to document the long-term effect on the pulmonary parenchyma. In order to study the long-term effect of injury, the animals need to survive over 2 weeks with appropriate treatment. Early burn excision and skin grafting have become the standard of care in the management of major burns . Many clinical studies and small-animal experiments support the concept of early burn wound excision and show decreased operative blood loss, length of hospitalization, and incidence of infection compared with late excision [7-12]. The effects of early excision and skin grafting have not, to our knowledge, been studied well in a large-animal model. We hypothesized that long-term model of smoke inhalation injury and cutaneous burn could be produced if early burn excision and autografting were utilized.
In the present study, we have developed a new ovine model of smoke inhalation injury and cutaneous burn with early excision and skin autografting to investigate chronically in lung tissue. We also demonstrated the effects of the early excision and skin autografting in our model with inhalation injury.
This study was approved by the Animal Care and Use Committee of the University of Texas Medical Branch (Galveston, TX, USA) and conducted in compliance with the guidelines of the National Institutes of Health and the American Physiological Society for the care and use of laboratory animals.
Eighteen female sheep were surgically prepared for this study under isoflurane anesthesia. The mean animal weight (mean ± SD) was 34 ± 4.6 kg. The right femoral artery was canulated with Silastic catheter (Intracath; 16 gauge, 24 in.; Becton Dickinson Vascular Access, Sandy, UT, USA). A thermodilution catheter (Swan–Ganz model 131F7, Baxter, Edwards Critical-Care Division, Irvine, CA, USA) was introduced through the right external jugular vein into the pulmonary artery. Through the left fifth intercostal space, a catheter (Durastic silicone tubing DT08, 0.062-in. ID, 0.125-in. OD; Allied Biomedical, Paso Robles, CA, USA) was positioned in the left atrium. The animals were given 5–7 days to recover from the surgical procedure, with free access to food and water.
Before the experiment, the vascular catheters were connected to the monitoring devices, and maintenance fluid (Ringer lactate, 2 mL/kg) was started. After baseline measurements and sample collections were completed, the animals were randomized into three groups: Early Excision group (20% TBSA, third-degree cutaneous burn and 36 breaths of cotton smoke followed by early excision and skin autografting at 24 h after injury, n = 6), Control group (20% TBSA, third-degree cutaneous burn and 36 breaths of cotton smoke without early excision, n = 6) and Sham group (no injury, no early excision, n = 6).
Immediately after injury, anesthesia was discontinued. The animals were allowed to awaken but were maintained on mechanical ventilation (Servo Ventilator 900C, Siemens-Elema AB, Sweden) for at least a 48 h experimental period. This was continued until the weaning process was completed. Ventilation was performed with a positive end-expiratory pressure of 5 cm H2O and a tidal volume of 15 mg/kg. During the first 3 h after injury, the inspiratory O2 concentration was maintained at 100% to induce rapid clearance of carboxyhemoglobin after smoke inhalation. The ventilation was then adjusted according to blood gas analysis to maintain arterial O2 saturation >90% and PCO2 between 25 and 30 mmHg. At 48 h post-injury, weaning from ventilator was begun if PaO2/FiO2 was above 250. Animals were then monitored for 3 weeks. Fluid resuscitation was given during the first 48 h experimental period with Ringer’s lactate solution following the Parkland formula (4 mL/% burned surface area/kg body weight for first 24 h and 2 mL/% burned surface area/kg body weight/day for the next 24 h). One-half of the volume for the first day was infused in the initial 8 h, and the remainder was infused in the next 16 h. From 48 h to 432 h, the animals received Ringer’s lactate (2 mL/% burned surface area/kg body weight/day). For 96 h post-injury, animals were allowed free access to food but not to water, to accurately measure fluid intake. Free access to water was permitted after this period. A Foley catheter was inserted to measure urine output until 96 h post-injury. At 24 h post-injury, early excision was carried out to 20% TBSA and skin autografting was performed at the time of excision under isoflurane anesthesia in the Early Excision group. Antibiotics (Cefazolin, 2 g/day, IV; Marsam Pharmaceuticals Inc., Cherry Hill, NJ and Tobramycin, 0.24 g/day, IV; Abraxis Pharmaceutical Products, Schaumburg, IL, USA) were given for 432 h to prevent possible infection. The animals were monitored for 3 weeks and were euthanized to assess lung tissue after an injection of ketamine (Ketaset, Fort Dodge Animal Health, Fort Dodge, IA, USA) followed by saturated KCl.
A tracheostomy was performed on each animal under ketamine anesthesia and a cuffed tracheostomy tube (10-mm diameter; Shiley, Irvine, CA, USA) was inserted. The anesthesia was continued with isoflurane, and the wool on both sides of the flank was shaved with electric clippers (Fig. 1A). The hair on one side of flank was removed using a depilatant (Nair®, Church & Dwight Co., Inc., Princeton, NJ, USA) to harvest skin. A 20% TBSA third-degree burn was inflicted on one side of the flank with a Bunsen burner until the skin was thoroughly contracted (Fig. 1B). Smoke inhalation was induced with a modified bee smoker. The bee smoker was filled with 40 g of burning cotton toweling and then attached to the tracheostomy tube via a modified endotracheal tube containing an indwelling thermistor from a Swan–Ganz catheter . Three sets of 12 breaths of smoke (total 36 breaths) were delivered, and the carboxyhemoglobin level was determined immediately after each set. The temperature of the smoke was not allowed to exceed 40 °C during the smoking procedure.
At 24 h post-injury, early excision was carried out to muscular fascia in the burn area (Fig. 1C) under isoflurane anesthesia in the Early Excision group. Skin autografting was performed at the time of excision (Fig. 1D). Split-thickness skin sections (20/1000 in., 0.5 mm) were harvested from the flank of the other side using an electric dermatome (Padgett Electro-Dermatome, Padgett Instruments Inc., KC, USA) and meshed in 4:1 ratio using a mesh dermatome (Padgett Mesh-Dermatome, Padgett Instruments Inc., KC, USA). The graft area was covered using non-adhering dressing (ADAPTIC®, Johnson & Johnson, Skipton, UK). The tie-over dressing was performed using rubber band and removed 4–6 days after placement. After removing the dressing, the wound was treated using vaseline without dressing. In order for the animal to maintain body temperature the operating time was limited to a maximum of 2 h. The ambient temperature in the operating room was maintained at 30 °C to prevent hypothermia. The Control and Sham groups were exposed to anesthesia with isoflurane for 100 min in the same operation room. Seventeen days after surgery, the wound was evaluated and photographed by the same person (for consistency) each time the site was evaluated. Each photograph was transferred in digital format and measured for the size of raw surface (RS; in cm2) and total graft area (TGA; in cm2) using computer software (ImageJ 1.40g, National Institutes of Health, USA) [14,15]. The percentage of the wound healed area (WHA) was calculated using the following equation:
Mean arterial (MAP; in mmHg), mean pulmonary arterial (MPAP; in mmHg), left atrium (LAP; in mmHg), and central venous (CVP; in mmHg) pressure were measured with pressure transducers (model PX-1800, Baxter, Edwards Critical-Care Division) that were adapted with a continuous flushing device. The transducers were connected to a hemodynamic monitor (model 78304A, Hewlett-Packard, Santa Clara, CA, USA). The pressures were measured with the animal in the standing position. Zero calibrations were taken at the level of the olecranon joints on the front leg of the animal. Cardiac output was measured with the thermodilution technique with a cardiac output computer (COM-1, Baxter, Edwards Critical-Care Division). A 5% dextrose solution was used as the indicator. For evaluation of cardiac function, cardiac index (CI; in L min−1 m−2) was calculated with standard equations. Blood gases were measured with a blood gas analyzer (IL GEM Premier 3000 Blood Gas Analyzer; GMI, MN, USA). The blood gas results were corrected for the body temperature of the animal. Oxyhemoglobin saturation and hemoglobin concentration were analyzed with a CO-Oximeter (model IL 482, Instrumentation Laboratory, USA). White blood cells (WBC; in cells/μL); neutrophil (in cells/μL) were counted (HEMAVET® HV950FS; Drew Scientific, Inc., TX, USA). Blood samples for determination of total protein concentration and oncotics pressure were collected in all groups. Hematocrit (Hct; in %) was measured in heparinized microhematocrit capillary tubes (Fisherbrand, Pittsburgh, PA, USA). Infused fluid volume and urine output were recorded every 6 h and net fluid balance was calculated by subtracting urine output from fluid intake. After sheep were euthanized, the entire right lung was harvested for measurement of wet-to-dry weight ratios (an index of pulmonary edema) as described by Pearce et al.  and aliquots of lung tissue were taken for various assays.
Significance was determined using a two-factor analysis of variance with repeated measures. The two factors were treatment and time. The differences in the wet-to-dry weight were evaluated by means of Student’s unpaired t-test. p-Value < 0.05 was considered to be significant.
The arterial carboxyhemoglobin levels (mean ± SD), as measured immediately after smoke exposure, amounted to 58.6 ± 10.1% in the Early Excision group and 61.5 ± 9.6% in the Control group. There were no significant differences between two groups. The Sham group had a significantly lower mean carboxyhemoglobin level of 6.5 ± 0.4%. All animals were survived for 3 weeks without infectious complication in Early Excision and Sham groups, whereas two out of six animals had abscess in lung in the Control group.
Early excision and skin autografting were performed in safety in the Early Excision group. The operative blood loss (mean ± SD) was 82.8 ± 44.2 g, operation time (mean ± SD) was 86.7 ± 18.6 min and the weight of excised eschar (mean ± SD) was 1127 ± 139 g. There was no bleeding (Fig. 2A) or wound infection (Fig. 2B) after operation. In the donor site, the wound was closed 2 weeks after surgery. The percentage of the wound healed area (mean ± SD) was 74.7 ± 7.8% on 17 days post-surgery (Fig. 2C).
No significant differences were noted in CI, MAP, LAP and CVP between the groups at any time point. MPAP and pulmonary vascular resistance index (PVRI) in the Early Excision group were lower compared to the Control group (Fig. 3C and D). However, there were no significant differences between Early Excision and Control groups.
In the Early Excision and Control groups, the average of the PaO2/FiO2 ratio decreased from 12 h post-injury and demonstrated significantly lower level vs. Sham group from 48 h to 96 h post-injury (Fig. 3A). The average of worst level in the PaO2/FiO2 ratio was 246 ± 85 in the Early Excision group, 281 ± 71 in the Control groups and 489 ± 16 in the Sham group throughout the experimental time period. In the PaO2/FiO2 ratio, no statistical difference was found between the Early Excision and the Control group. All animals gradually recovered in the PaO2/FiO2 ratio and were successfully weaned from the ventilator. All animals recovered to baseline level in the PaO2/FiO2 at 2 weeks post-injury. No animals showed a progressive fall in the early postoperative period. The pulmonary shunt fraction (Qs/Qt) increased from 12 h to 96 h post-injury in the Early Excision and the Control groups (Fig. 3B). However, these values could not be shown to be statistically different from baseline. There were no significant differences between Early Excision and Control groups in the PaO2/FiO2 and Qs/Qt at any time point.
The urine output decreased at 30 h post-injury and was maintained in the range of 0.5–1 mL/kg/h in the Early Excision group (Fig. 4B). Despite the same amount of fluid resuscitation (Fig.4A), significantly higher urine output was noted at 54 h post-injury in the Control group compared to the Early Excision group. The calculated net fluid balance significantly increased in the Early Excision compared to those seen in the Sham and Control groups (Fig. 4C). Accumulated positive fluid balance in the Early Excision group significantly increased vs. Sham group during the first 48 h and Control group during the second 48 h (Fig. 4D).
In the Early Excision group, plasma protein and colloid oncotic were significantly decreased during the first 96 h and increased up to 432 h post-injury, whereas in the Control group, these were increased from 48 h post-injury (Fig. 5A and B). Statistical differences were shown in the plasma protein and the oncotic pressure vs. Control and Sham groups.
Hct of % baseline in the Early Excision group decreased after injury up to 432 h post-injury and was significantly lower than the Control group at 408 and 432 h post-injury (Fig. 6A). Hemoglobin also decreased and showed statistical difference from 360 h to 432 h post-injury vs. Control group (Fig. 6B).
The WBC and neutrophil counts were significantly decreased after burn wound excision and autografting (Fig. 7A and B). Statistical differences were found at 48, 60 and 72 h post-injury in the neutrophil vs. Control group and at 60 h post-injury vs. Sham group.
Lung wet-to-dry weight ratio, an indicator of lung water content, was significantly increased in the Early Excision group compared to the Sham group (Fig. 8).
Several authors have described long-term clinical effects of smoke inhalation injury [17,18]. Fogarty et al. showed ventilatory defect and small airway obstruction were present in 11 survivors of the King’s Cross underground station fire after 6 months . Desai et al. reported that 64% of pediatric patients (mean burn size of 44% total body surface) with inhalation injury had abnormal spirometry and lung volumes at rest 2 years post-injury . Park et al. demonstrated the long-term effects of smoke inhalation, by examining airway responsiveness, airway inflammation, and systemic effects, and concluded that inflammatory reaction in the airways and peripheral blood continues for at least 6 months after smoke inhalation . However, there are no studies using the same criteria that grade simultaneously the degree of smoke inhalation and the same methodology to evaluate lung function. Palmieri suggests the first step in determining the effects of smoke on long-term pulmonary function is to evaluate it in an animal model . Many animal models with smoke inhalation and cutaneous burn have been described in the literature involving rodents without early excision [1-5,23,24], but there have been no clinically relevant large-animal models which could monitor pulmonary function and hemodynamics for over and extended period of 2 weeks or more.
Darling et al. demonstrated the high mortality from inhalation injuries is most significant in burns >15% TBSA . Suzuki et al. also reported that the mean full thickness burn size of 1690 patients with inhalation injury was 20.4% TBSA in Tokyo . In our model, the size of cutaneous burn was determined in consideration of the effects on pulmonary function. It is well established that inhalation injury increases the mortality in burn patients, but there are few studies to determine whether early excision at 24 h post-injury would aggravate pulmonary function in inhalation injury .
The present study suggests that early excision and skin autografting do not aggravate pulmonary function in PaO2/FiO2 ratio, Qs/Qt, PAP and PVRI compared with no excision group. At 3 weeks post-injury, these indices showed recovery from lung injury. Excision therapy and autografting were safely performed in sheep with impaired lung function and long-term model of smoke inhalation injury and cutaneous burn without wound infection. In the present model, a cotton smoke insufflation injury (36 breaths) combined with a 20% TBSA third-degree cutaneous flame burn produces a predictable (PaO2/FiO2 < 300) model of acute lung injury (ALI). One of the long-term effects of smoke inhalation injury was demonstrated in lung wet-to-dry weight ratio, an index of pulmonary edema, in the Early Excision group (Fig. 8). To show the long-term effects clearly, further studies are needed using the measurement of collagen deposits, lung compliance and diffusion capacity tests.
There are several wound models in swine for burn treatment [14,29-31]. Unfortunately, it is difficult to maintain tight dressing within the first 4–6 days and keep wound clean without dressing after first dressing change on an awake pig . In contrast, in the ovine model it is easy not only to measure pulmonary and hemodynamic function, but also to treat and observe the wound as the animals can be maintained in an upright position. In the present model, burn wound of approximately 1900 cm2 could be monitored for 3 weeks without infection. Porcine skin is more similar to human skin than that of sheep , as both porcine and human have sparse body hair and hair follicles play an important role in reepithelialization. However, we speculate that the differences in skin do not have effects on wound healing in our model because wound excision was carried out to muscular fascia and split-thickness skin (20/1000 in. or 0.5 mm) was harvested for grafting. The present burn and smoke inhalation model with early excision is clinically relevant and, to our knowledge, is the first in the world to be used for long-term studies.
Many clinical studies have reported that early excision of the burn wound decreased operative blood loss, reduced the length of hospitalization and incidence of infection [7,9,12]. In the present study, hemoglobin of % baseline and hematocrit of % baseline did not decrease to a statistically significant degree in the early postoperative period, and the average of WHA (mean ± SD) was over 70% at 18 days post-injury. In the Early Excision group, the animals demonstrated no incidence of infection and could have been discharged from hospital had they been patients. In the Control group, two out of six animals (33%) had abscess in lung at 3 weeks post-injury.
At the same time, some differences were exposed between Early Excision and Control groups. In the net fluid balance, early excision and skin grafting statistically increased fluid requirements compared to the Control and the Sham groups (Fig. 4C). In the Control group, the urine volume was statistically increased in the refilling period (Fig. 4B), and less fluid volume was required compared to the Early Excision in the second 48 h post-injury (Fig. 4D). Hypoproteinemia and statistical lower oncotic pressure were found after operation and recovered from 1 week post-injury in the Early Excision group (Fig. 5A and B). These results showed that we have to know the differences between burn/inhalation injury with early excision and burn/inhalation injury alone to determine the fluid resuscitation volume. Progressive anemia, which was measured as Hb and Hct, had appeared in the Early Excision group throughout the experiments though frank bleeding was not observed after surgery (Fig. 6A and B). In a clinical setting, hospitalized burn patients often become anemic because of hemodilution, relative bone marrow suppression, and frequent laboratory draws . Early eschar excision traditionally has been associated with significant operative blood loss . Our current findings suggest that supplement of albumin and late blood transfusion should be considered in extensive burn patients after early excision and grafting. Early wound excision had been shown in small-animal study to increase pulmonary leukosequestration compared with the burn injury alone . In the present model, neutrophil counts statistically significantly decreased compared to Control group and baseline value after burn wound excision. We speculate that neutrophil counts were decreased because of leukosequestration. Xiao-Wu et al. showed that some patients have postoperative pulmonary complications that may counter any benefits from immediate excision such that the 2 effects cancel each other .
Early excision and skin autografting were performed in sheep with 20% burn and moderate smoke inhalation injury without changes of hemodynamics and pulmonary dysfunction. This model closely resembles clinical setting, exposes the effects of early excision and autografting and allows to chronically monitor pulmonary function following burn and smoke inhalation injury. Further studies are warranted to assess lung tissue scaring measuring collagen deposition, lung compliance and diffusion capacity.
National Institute supported this work for General Medical Sciences Grant GM66312-01 and Grants 8450, 8630, 8520 and 8954 from the Shriners of North America.
Conflict of interest
The authors declare that there is no conflict of interest.