Pathophysiology of ovine smoke inhalation injury treated with extracorporeal membrane oxygenation.

An ovine model was used to study the pathophysiology of smoke inhalation injury treated with extracorporeal membrane oxygenation (ECMO). Smoke inhalation is characterized by leukocyte-oxygen free-radical mediated acute lung injury. Treatment with ECMO was by extracorporeal venoarterial or venovenous perfusion using a venous drainage reservoir, roller pump, heat exchanger, and membrane lung oxygenator capable of oxygen delivery to and carbon dioxide removal from a patient. Blood-foreign surface interactions are known to occur during ECMO. We examined the effects of ECMO on circulating leukocytes, oxygen free-radical activity, thromboxane release, and gas exchange after smoke inhalation injury. Animals treated with smoke and ECMO had significantly increased circulating thromboxane B2 levels and oxygen free-radical activity compared with sham-treated animals and animals treated with smoke and mechanical ventilation (MV). Likewise, there was a significant increase in lung wet-to-dry weight ratios in animals treated with smoke and ECMO compared with those treated with smoke and MV. These data may account for the initial deterioration in native lung function after the initiation of ECMO and imply that ECMO may potentiate the pathophysiology of smoke inhalation injury.

Bacteremia and postmortem microbiology in burned children.

During a three-year period, Staphylococcus aureus and Pseudomonas aeruginosa were the organisms most commonly isolated from blood cultures of burned children. Microorganisms were considered to contribute to the cause of death in 17 of 20 patients who died from various complications of thermal injuries. Pseudomonas aeruginosa was involved in eight deaths, whereas other Gram-negative bacilli or fungi, or both, were involved in the deaths of the remaining nine patients. The microbiologic examination of cardiac blood and pulmonary tissue correlated reasonably well with clinical and anatomic judgments of cause of death, as well as with the defining of some cases of "terminal sepsis".

Tracheal venous blood and lymph collection: a model to study airway injury in sheep.

Airway injury is a frequent result of the inhalation or aspiration of toxic material. Although upper airway damage can be identified endoscopically, pathophysiological changes are difficult to evaluate. This paper describes an animal model in which changes in tracheal blood and lymph flow rates, wet-to-dry weight ratios, and lymph-to-plasma protein ratios can be evaluated after injury. In this model, 12 cm of the cervical trachea were isolated using a double-cuffed endotracheal tube and injured with cotton smoke at near room temperature. Injury to the trachea was evaluated in twenty-five anesthetized sheep 4 (n = 3), 8 (n = 3), 24 (n = 3), 48 (n = 3), 96 (n = 3), and 192 (n = 2) h after smoke exposure and compared with sham control animals (n = 8). A significant increase in tracheal venous blood flow from 1.3 +/- 0.4 (SD) ml.min-1.cm-1 for the noninjured trachea to 2.8 +/- 1.2 was noted 24 h after injury (P less than 0.01). Lymph flow significantly increased from 1.3 +/- 0.4 microliters.min-1.cm-1 for the noninjured trachea to 9.8 +/- 3.3 24 h after injury while wet-to-dry weight ratios were elevated from 3.0 +/- 0.2 for noninjured trachea to 4.6 +/- 0.9 from 4 to 24 h after injury (P less than 0.01) and decreased to 3.7 +/- 0.5 by 96 h. Cast material consisting of airway exudate, cellular debris, and intact ciliated epithelial cells was both expectorated and found in the trachea when the animals were killed.(ABSTRACT TRUNCATED AT 250 WORDS)

Mechanical alteration of blood flow in smoked and unsmoked lung areas after inhalation injury.

The degree of pulmonary perfusion may have an important role in the pathogenesis of inhalation injury. We studied this in sheep that had only one lung exposed to smoke. The right lung and upper airway of 12 chronically instrumented sheep were insufflated with cotton smoke. In six animals, the left pulmonary artery was occluded between 4 and 10 h after smoke insufflation. All animals were studied for 24 h and then killed, and lung tissue was harvested. The smoked as well as the air-insufflated lung of all animals showed an increase in wet-to-dry weight ratio and tissue conjugated dienes (products of lipid peroxidation). Neither the intermittent blood flow increase to the smoked lung nor the simultaneous blood flow reduction with a concomitant polymorphonuclear neutrophil entrapment in the air-insufflated lung significantly affected the histopathological outcome of the respective lung. We conclude that tissue damage after inhalation injury cannot be diminished by increasing the flow to smoked areas. Ischemia-reperfusion injury does not have a major role in the lung damage seen with inhalation injury.

Effects of allopurinol on smoke inhalation in the ovine model.

We hypothesized that the pulmonary damage induced by smoke inhalation is the result of ischemic reperfusion injury. We determined the effect of allopurinol (xanthine oxidase inhibitor) on the pulmonary microvascular fluid flux in an ovine model after inhalation of cotton smoke (n = 13) and compared these data with those from untreated similarly smoke-injured (n = 7), as well as sham- (air, n = 9) smoked, animals and sheep given an equivalent dose of CO (n = 7). Smoke injury resulted in an increased lung lymph flow, lymph-to-plasma protein ratio, lung content of polymorphonuclear cells, and extravascular lung water (gravametric), in addition to histological evidence of tissue (pulmonary) edema and destruction. No significant difference was found in these variables between the sheep that were injured with smoke whether or not they were pretreated with allopurinol. The sham-smoked and CO-insufflated animals showed no significant changes in cardiopulmonary function or morphology. We conclude that there are few data to support a role of ischemic reperfusion injury in the pulmonary damage seen after smoke inhalation.

Pulmonary hemodynamics and tissue damage after one lung infusion of Pseudomonas aeruginosa in sheep.

The relative roles of hematogenous mediators and direct bacterial toxicity due to phagocytosis by pulmonary intravascular macrophages were determined by selective bacterial infusion into the left pulmonary artery and comparison of right and left lungs at 24 h. Chronically instrumented sheep received 15-min pulmonary arterial infusions of live Pseudomonas aeruginosa (0.35-2.9 x 10(9), n = 6) or saline (n = 5). The saline group demonstrated stable cardiopulmonary function over time. Left lung blood flow, measured by Doppler flow probe, decreased 15 min into the bacterial infusion, with a concomitant sevenfold increase in left lung pulmonary vascular resistance index. The right lung pulmonary vascular resistance index doubled at 1 h, in association with increased plasma thromboxane B2 levels. An increase in cardiac index and decrease in systemic vascular resistance occurred at 12 h. The wet-to-dry weight ratio of the Pseudomonas-infused left lung was increased compared with that of the sham-infused lung. The tissue count of neutrophils in the lungs was doubled in both sides, but neutrophils on the left were more degranulated. The left lung tissue damage was caused by direct bacterial toxicity, including activation of phagocytic cells. Hematogenous mediators induced pulmonary and systemic hemodynamic changes and right lung neutrophil sequestration, but they did not damage the noninfused lung.

Increasing duration of smoke exposure induces more severe lung injury in sheep.

Eighteen sheep previously prepared for chronic study were divided into three groups of six animals each. These were given graded inhalation injury utilizing smoke obtained from burning cotton-toweling material. Smoke was insufflated into animals with a modified bee smoker at temperatures less than 40 degrees C. Group H, which received 64 breaths of smoke, showed the most pronounced changes in pulmonary function. The changes consisted mainly of a profound increase in lung lymph flow following a reduced P/F ratio (PO2 in arterial blood/inspired O2 fraction) and an elevation in both thermal and gravimetrically measured extravascular lung water. Similar changes were seen in group M (48 breaths of smoke) and group L (32 breaths of smoke). However, the injury was graded based on the changes in gravimetrically measured lung water and lung lymph flow. These were highest in group H and lowest in group L. These studies confirm our ability to accurately quantitate the injury induced by smoke inhalation. In addition, it demonstrates that lung injury associated with the inhalation of smoke can be graded depending on the duration of exposure.

The effect of venous obstruction in infected pedicle flap.

A new model of soft-tissue infection is used to investigate the effect of the local wound environment on the septic focus. Island pedicle flaps were raised on the buttock of 24 adult ewes and multiply inoculated with Staphylococcus aureus. Flaps with bacterial inoculation, without compromise of venous outflow, showed distal necrosis (mean +/- SEM percent of surface area, 25.8% +/- 8.6%) and developed septic foci with bacterial counts one log less than the amount injected. Flaps with inoculation and venous outflow obstruction underwent subtotal necrosis (mean percent of surface area, 73.3% +/- 11.2%) and had counts two logs higher than the nonobstructed flaps but without discrete septic foci. Flaps without inoculation, with or without venous obstruction, survived completely. Venous outflow obstruction is shown herein to potentiate tissue necrosis by raising bacterial counts in a septic focus and preventing defensive abscess formation by the host.

Thromboxane receptor blockade with BM 13,177 following toxic airway damage by smoke inhalation in sheep.

Thromboxane may play an important role in the pathogenesis of smoked mediated injury. We studied this possibility in 13 chronically instrumented sheep, which had the left lung exposed to smoke. BM 13,177, a thromboxane receptor antagonist, was given intravenously to six animals prior to smoke inhalation and during the experimental period. Seven animals received the vehicle. All animals were studied for 24 h under ventilatory support, then killed prior to harvesting lung tissue. Airway peak and plateau pressures in the vehicle-treated animals were elevated by 27% and 25% from baseline at 24 h post smoke inhalation. Concomitantly, the left pulmonary vascular resistance index rose continuously throughout the study period (baseline = 822 +/- 58; 24 h = 1819 +/- 84 dyn.s.cm-5.m2).BM 13,177 treatment completely prevented the rise in airway pressure, while the left pulmonary vascular resistance index was significantly attenuated (baseline = 726 +/- 79; 24 h = 1470 +/- 158 dyn.s.cm-5.m2) resulting in a significantly higher percentage of cardiac output being delivered to the smoked lung, compared to vehicle-treated animals. Thromboxane receptor blockade did not prevent smoke induced pulmonary edema formation. There was likewise no effect of BM 13,177 on the systemic hemodynamic changes seen following smoke inhalation. There was a decrease in cardiac index and an increase in systemic vascular resistance index in both groups. We conclude that smoke induced changes in airway and pulmonary vascular resistances may be mediated by thromboxanes. However, thromboxanes appear to play no role in the development of pulmonary edema and elevation of systemic vascular resistance following smoke inhalation injury.

Pulmonary injury in burned patients.

Inhalation injury has emerged as the number one cause of fatality in the burn patient. Fiberoptic bronchoscopy and 133Xe scanning complement traditional clinical signs of inhalation injury and have led to discovery of a higher incidence of these injuries among patients with burns. Patients with inhalation injury typically demonstrate three stages: acute pulmonary insufficiency, pulmonary edema, and bronchopneumonia, all of which carry at least 50 per cent mortality rates. The major early pathophysiologic changes in the lungs of burned patients are related to upper-airway obstruction and lower-airway permeability edema. Treatment consists of intubation for signs of respiratory distress, pulmonary toilet, humidification of inspired air, and antibiotics for documented infection.

Proteoglycan-lymphocyte association in the development of hypertrophic scars.

Parallel histological stains and immunohistological stains were made of hypertrophic scars and normal scars in order to identify the type of cells and associated proteoglycans present in perivascular cuffs in hypertropic scars. Tissue sections were treated with monoclonal antibodies which specifically recognize T-cells and unsulphated, 4-sulphated and 6-sulphated chondroitin proteoglycans. There was a striking association between the perivascular lymphocytic infiltration and chondroitin-4-sulphate, which may be an important contributory factor in the development of hypertrophic scars.

Burn prevention: the need for a comprehensive approach.

Traditionally, burn prevention efforts have been directed at public education and numerous burn prevention campaigns have been undertaken during the past 30 years. How successful have these campaigns been? Contrary to burn mortality, statistical data on burn morbidity are extremely difficult to obtain because little national or international data is available. An analysis of the absolute death rates caused by fire and flames for various countries from 1975 through 1986 seems to indicate that world-wide burn mortality has not decreased. On the contrary, in many countries the rates have increased, whereas in countries like the USA, UK, France and FR Germany, the decrease has been disappointingly low. Furthermore, one wonders if the decrease in mortality rates experienced by these countries is the result of effective prevention programmes or if these could be attributed to advances in technology and improved medical care. Burn prevention campaigns aimed at public education have failed to provide the expected decrease in burn injuries. Education may increase knowledge but does not necessarily lead to behavioural and/or lifestyle change. A restructuring of our burn prevention programmes is needed. Modern techniques of motivational theory must be used to promote public concern and action for individual behaviour change, pertinent legislation and product safety.

Pulmonary embolism in burned children.

There are occasional reports in the literature concerning the incidence of pulmonary embolism in the postburn population, but reports of burned children are especially rare. The clinical diagnosis of pulmonary embolism is particularly difficult in these populations due to the postburn pulmonary complications of pneumonia, bronchopneumonia, respiratory distress syndrome, and changes incurred through inhalation injury. A retrospective review of all patient deaths occurring at this institution during the past 22 years was performed in order to document the incidence of pulmonary embolism in burned children. Of the 6589 patients admitted during this time, 178 patients died (2.7%) and three (1.7%) deaths were attributable to pulmonary embolism. Two other deaths (1.1%) were associated with deep vein thrombosis. The incidence of pulmonary embolism can then be calculated at 46 per 100,000 admissions in this population of burned children. Burned patients always pose an increased risk for the development of pulmonary embolism. These patients are traumatized, require multiple venous and/or arterial cannulations, undergo multiple surgical procedures, are immobile for prolonged periods, prone to infectious processes and fluid and electrolyte imbalances. Despite all these risk factors, the incidence of pulmonary embolism is less than 2 per cent of all deaths in this postburn paediatric population.

Containment of a multiresistant Serratia marcescens outbreak.

A 3-year-old male from Bolivia who sustained a full-thickness 80 per cent TBSA burn complicated by smoke inhalation on the 28 March 1995 was admitted to our burn centre on 6 April 1995. On 11 April the patient's wounds were colonized with a Serratia marcescens sensitive only to ciprofloxacin and imipenem. Sputum cultures revealed the same phenotypic S. marcescens. Two patients who were admitted days later had the same phenotypic S. marcescens. Their TBSA burns ranged from 54 to 80 per cent. Both were injured in early April. Sputum and wound cultures were also positive for S. marcescens. Precautionary measures were instituted immediately. All potential reservoirs were cultured. Cultures were negative for S. marcescens. Patient therapy was maintained via strict isolation. The first patient died on 17 May. The two remaining patients survived and were discharged colonized with S. marcescens. However, the biotype of the initial S. marcescens was different from the latter two. Early recognition of a multiresistant S. marcescens resulted in negating the spread of this agent to other patients.

Measurement of collagen-proteoglycan interaction in hypertrophic scars.

Collagen-proteoglycan interaction in hypertrophic scars was measured by staining tissue collagen with Sirius red, followed by elution of the bound stain, and photometric quantitation before and after papain digestion. Control sections from each of the specimens were treated with buffer solution instead of papain. The difference in dye concentration observed between respective untreated and digested sections was considered to be due to the unmasked basic groups of collagen originally bound to proteoglycans. The results suggest a significantly increased collagen-proteoglycan interaction in human hypertrophic scars as compared with nonhypertrophic scars, normal skin, and granulation tissues.

Proteoglycans and collagenase in hypertrophic scar formation.

The collagen fibers of the nodules and whorl-like figures in hypertrophic scars are "coated" with proteoglycans, mainly chondroitin-4-sulfate. The latter was shown to prevent collagenase from breaking down collagen. This suggests that the presence of great amounts of chondroitin-4-sulfate in hypertrophic scars may contribute to the overabundance of collagen deposition which is characteristic of this abnormal healing process.

The effect of denervation on soft-tissue infection pathophysiology.

Pressure is the sine qua non in the etiology of pressure sores; however, ischemia, denervation, edema, and infection also have been implicated. The role of denervation in tissue infection was studied in an isolated in vivo ovine flap model. Twenty-six adult ewes, divided into three groups, had 29 island pedicle flaps raised on their buttocks. In group I, the cutaneous nerve remained intact, while group II had its nerve divided acutely. Group III had prolonged denervation, where the nerve was divided 7 days before flap elevation. All flaps received intradermal inoculations of 10(7) Staphylococcus aureus. Ninety-six hours later, quantitative bacteriology showed counts of 10(7), 10(7), and 10(9) colony-forming units (CFU) per gram of tissue in groups I, II, and III, respectively. Septic foci were larger in group III, and there was a significant increase in tissue edema between groups I and III. A 25-fold increase in bacterial counts seen in the prolonged denervation group may help explain why neurologically injured patients are more susceptible to infection and pressure ulcerations.

Comparative effects of dopamine, naloxone, and prostacyclin in the resuscitation of fecal-Escherichia coli peritonitis-induced septic shock in neonatal swine.

To explain the high neonatal mortality from peritonitis-induced septic shock despite current resuscitation practices, the efficacy of dopamine, naloxone, and prostacyclin was evaluated in an experimental neonatal model. Hemodynamics were monitored and survival was measured in anesthetized neonatal swine, which were subjected to fatal fecal-Escherichia coli peritonitis-induced septic shock. All the animals received fluid resuscitation, antibiotics, and bicarbonate to correct acidosis. Pharmacologic resuscitation began when cardiac output dropped below baseline in the experimental groups. Although significant differences were observed between groups in cardiac output, mean arterial and mean pulmonary arterial pressures, left ventricular stroke work, stroke volume, and pulmonary vascular resistance indices (P less than 0.02), and each animal exhibited favorable hemodynamic responses during the first several hours of dopamine and naloxone infusion, these drugs failed to prolong survival. Also, 5 of the 9 naloxone-treated pigs (56%), died with histologically proven intestinal ischemia (P less than 0.02). Thus, dopamine, naloxone, and prostacyclin (at doses commonly recommended for the treatment of septic shock) fail to positively influence the fatal course of this condition, and the use of naloxone in this model is associated with profound intestinal ischemia.

Sequence of morphologic events in experimental smoke inhalation.

The experimentally induced smoke inhalation injury in sheep is in many respects similar to that observed in smoke-injured humans. Our sheep model allows us to describe a sequence of morphologic changes that may be divided into four fairly discrete although overlapping phases. The magnitude of the destructive process and the length and characteristics of each phase are related to the severity of the injury and the superimposed complications. The availability of this animal model may allow the experimental manipulation of each of the four described phases to investigate physiopathologic and therapeutic implications toward the prompt normalization of pulmonary function.

Effects of inhalation injury on airway blood flow and edema formation.

We measured airway blood flow in unanesthetized sheep under control conditions and after lung injury induced by inhalation of cotton smoke. Blood flows in trachea, carina, main stem bronchi, intraparenchymal bronchi, and whole lung were measured by injection of radioactive microspheres. In 10 control sheep mean blood flow (+/- SD) was trachea, 17.2 +/- 10.5; main stem bronchi, 17.5 +/- 7.6; and whole lung (parenchyma inclusive of all small intraparenchymal airways), 20.5 +/- 11.9 ml.min-1/100 gm tissue weight. After injury, measurements were made 8 to 30 hours after smoke inhalation when respiratory distress was evident by arterial oxygen tensions of less than 60 mm Hg. Inhalation injury had little effect on cardiac output or blood flow to peripheral tissue. However, after inhalation injury airway blood flow (n = 6) was increased nine times in trachea, eight times in main stem bronchi, twelve times in intraparenchymal bronchi, and two times in whole lung. The increased airway blood flow resulted from a selective vasodilation of the airway vasculature because arterial driving pressures were unchanged by inhalation injury. Other investigators have shown that the microvascular permeability of the bronchial circulation is remarkably sensitive to inflammation, and the present experiments suggest that a selective vasodilation of the airway vasculature is another aspect of the airway response to inflammation. Increased airway blood flow through a leaky microvasculature may increase capillary filtrate from the bronchial circulation and contribute to the pulmonary edema of inhalation injury.