INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nitric oxide (NO) is an endogenous vasodilator synthesized from L-arginine through catalysis by nitric oxide synthase (NOS) enzymes (EC 1.14.13.39). Mammals have three different NOS enzymes, each coded by a different gene: endothelial (ecNOS) and neuronal (bcNOS), both of which are constitutive enzymes, and inducible NOS (iNOS). Under physiologic conditions, NO synthesis by ecNOS in the vascular endothelium is generally believed to have a role in vascular regulation, including the regulation of vascular permeability (1). However, under pathologic conditions, iNOS can be expressed in endothelial cells, vascular smooth muscle cells, and macrophages in response to various stimuli such as endotoxin or inflammatory cytokines (2). This stimulation results in the overproduction of NO, which can have harmful effects on vascular regulation (3).

Recent studies related to burn injury have held NO to be a putative cause of vascular hyperpermeability (4, 5). Sozumi and associates demonstrated that the NO content of second-degree burn tissue was significantly higher than that of uninjured skin, and that inhibition of NOS suppressed vascular permeability at the site of injury in a mouse ear burn model (5).

In patients with extensive cutaneous burns in which the burned area exceeds 25% to 30% of the total body surface area (TBSA), capillary hyperpermeability occurs not only at injured sites but also in regions distant from the injury (6, 7). This vascular hyperpermeability leads to fluid loss from the circulation. The large fluid losses in such cases result in hypovolemic shock unless rapid and adequate fluid resuscitation is administered (8, 9). However, this resuscitation therapy generates an excess fluid accumulation in the systemic interstitial space and in the lung. The lung edema is a major source of morbidity and mortality (10, 11). Edema is even more severe when thermal damage is combined with inhalation injury (12- 14). Thus, therapeutic approaches to reduce the vascular permeability of patients with combined burn/inhalation injury are desirable.

We hypothesized that induction of iNOS and subsequent increased production of NO are responsible for edema formation after this kind of injury. To test this hypothesis, we gave conscious sheep, subjected to a combined 40% TBSA third-degree burn and smoke inhalation injury, mercaptoethylguanidine (MEG), a selective inhibitor of iNOS [EC₅₀] (with effective concentrations reducing 50% of iNOS and ecNOS activity in tissue homogenates of 11.5 µM and 110 µM, respectively) and scavenger of peroxynitrite that is an oxidative metabolite of NO, and measured these animals' iNOS protein and message concentrations and plasma and lymph concentrations of NOx (15).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The study was approved by the Animal Care and Use Committee of the University of Texas Medical Branch and was conducted in compliance with the guidelines of the National Institute of Health and the American Physiological Society for the care and use of laboratory animals.

Surgical Preparation

Twelve female sheep were surgically prepared for chronic study under halothane anesthesia. The right femoral artery and vein were cannulated with Silastic catheters (Intracath, 16-gauge 24 in.; Becton Dickinson Vascular Access, Sandy, UT). A thermodilution catheter (Swan- Ganz, Model 131F7; Baxter, Edwards Critical-Care Division, Irvine, CA) was introduced through the right external jugular vein into the pulmonary artery. The chest was opened at the fifth intercostal space, and an efferent lymphatic vessel from the caudal mediastinal lymph node was cannulated with a silicone catheter (Alliedsil silicone tubing 1264/T056, I.D. = 0.025 in. O.D. = 0.047 in.; Allied Biomedical, Paso Robles, CA), using a modified method based on the technique of Staub and colleagues (18). Ligation of the tail of the caudal mediastinal lymph node and cauterization of the systemic diaphragmatic lymph vessels removed the systemic lymph contribution. Through the fifth intercostal space, a catheter (Durastic Silicone Tubing DT08, I.D. = 0.062 in., O.D. = 0.125 in.; Allied Biomedical) was positioned in the left atrium. An efferent lymphatic vessel from the prefemoral lymph node that drains the capillary bed of skin and soft tissue on the flank was cannulated with a silicone catheter (Alliedsil silicone tubing 1264/T056, I.D. = 0.025 in., O.D. = 0.047 in.; Allied Biomedical) using a method based on the technique of Demling and coworkers (19). The procedure was performed through an incision anterior to the tensor fascia lata. The animals were given 5 to 7 d to recover from the surgical procedure with free access to food and water During the first 24 h after the procedure the animals were given 0.3 mg of buprenorphine hydrochloride intravenously (Buprenex; Reckett & Colman Products Ltd., Hull, UK) immediately and at 12 h intervals to reduce pain. The narcotic analgesic was given after the first 24 h only if the animal appeared to be in distress (e.g., grinding its teeth, getting up and down).

Burn and Smoke Inhalation Injury

Before the induction of burn/inhalation injury, a tracheostomy was performed under ketamine anesthesia (Ketaset; Fort Dodge Animal Health, Fort Dodge, IA) and a cuffed tracheostomy tube (10-mm diameter; Shiley, Irvine, CA) was inserted. Anesthesia was then continued with halothane. Following this, all animals were given a combined injury consisting of a 40% TBSA third-degree burn and 48 breaths of cotton-smoke inhalation. After the animal's wool had been shaved, a 20% TBSA third-degree flame burn was made on one side of flank. The burn was produced with a Bunsen burner until the skin was thoroughly contracted. We have previously determined this degree of injury to be a full-thickness burn including both epidermis and dermis, in which the nerve endings are destroyed by heat (20). Thereafter, inhalation injury was induced with a modified bee smoker. The bee smoker was filled with 40 g of burning cotton toweling and was attached, via a modified endotracheal tube containing an indwelling thermistor from a Swan-Ganz catheter, to the tracheostomy tube that had been inserted in the animal. Four sets of 12 breaths of smoke (a total of 48 breaths) were delivered, and the animal's carboxyhemoglobin level was determined immediately after smoke inhalation. The temperature of the smoke was not allowed to exceed 40° C during the smoke-inhalation procedure. After smoke insufflation, another 20% TBSA third-degree burn was made on the remainder of the animal's flank.

Measured Variables

Vascular pressures, mean arterial pressure (MAP), pulmonary arterial pressure (Ppa), left atrial pressure (Pla), and central venous pressure (CVP) 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). Pressures were measured with the animal in the standing position. Zero calibrations were taken on the right front leg of the animals, at the level of the olecranon process. Cardiac output (CO) was measured with the thermodilution technique, using a CO computer (COM-1; Baxter, Edwards Critical-Care Division). A 5% dextrose solution was used as the indicator. Blood gases were measured with a blood gas analyzer (model IL 1600; Instrumentation Laboratory, Lexington, MA). The blood gas results were corrected for the body temperature of the sheep. Oxyhemoglobin saturation and carboxyhemoglobin concentration were analyzed with a Co-oximeter (Model IL 482; Instrumentation Laboratory). Hematocrit (hct) was measured in heparinized microhematocrit capillary tubes (Fisherbrand; Fisher, Inc., Pittsburgh, PA). Systemic prefemoral lymph flow, which represents burned tissue lymph flow (bQ_L) and lung lymph flow (lQ_L) were measured with a graduated test tube and stopwatch. Lymph and blood samples were collected in tubes containing ethylendiamine tetraacetic acid (EDTA). The colloid oncotic pressures in plasma (_P), in burned-tissue lymph (b_i), and in lung lymph (l_i) were determined through a semipermeable membrane in a colloid osmometer (Model 4100; Wescor, Logan, UT). The burned-tissue permeability index (bPI) and lung permeability index (lPI) were calculated according to the following equations: bPI = bQ_L × (b_i /_P) and lPI = lQ_L × (l_i /_P). Levels of NOx, which was intermediate and end products of NO oxidation, were measured in plasma and lymph with a chemiluminescence assay as subsequently described.

Measurement of NOx in Plasma and Lung Lymph

The concentration of NOx (total amount of NO metabolites) in plasma and lymph was measured with an NO chemiluminescence detector (Model 7020; Antek Instruments, Inc., Houston, TX). The plasma and lymph samples to be analyzed were converted to NO gas by treatment with vanadium (III) and hydrochloric acid at 90° C in a nitrate/ nitrite reduction assembly (Model 745; Antek Instruments, Inc.) before being introduced into the detector. Thereafter, the NO reacted with ozone in the reaction chamber of the detector, and the emitted light signal was recorded by dedicated software as the NO content (µM/L) as compared with standards.

Experimental Protocol

Twenty-four hours before the experiment, vascular catheters were connected to the monitoring devices and maintenance fluid administration (Ringer's lactate, 2 ml/kg) was begun via the right femoral vein. After baseline measurements and sample collections were completed, all animals were given combined burn and smoke inhalation injuries, as previously described. A silicone Foley catheter (Dover, size 14 French, 5 ml; Sherwood Medical, St. Louis, MO) was placed in the urinary bladder to determine urine output. Immediately after injury, anesthesia was discontinued and the animals were allowed to awaken but were maintained on mechanical ventilation (Servo Ventilator 900C; Siemens-Elema, Sweden) throughout the 48-h experimental period. Ventilation was administered with a positive end-expiratory pressure (PEEP) of 5 cm H₂O and a tidal volume of 15 mg/kg. During the first 3 h after injury, the fraction of inspired O₂ was maintained at 100% and the respiratory rate was kept at 30 breaths/min to induce rapid clearance of carboxyhemoglobin after smoke inhalation. Ventilation was then adjusted according to the results of blood gas analysis to maintain the Sa_O2 above 90% and the arterial oxygen tension between 25 and 30 mm Hg. Fluid resuscitation during the experiment was done with Ringer's lactate solution according to the Parkland formula (4 ml/ 1% burned surface area/kg body weight for the first 24 h and 2 ml/1% burned surface area/kg body weight 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. Urine was collected and urine output was recorded every 3 h. Fluid balance was determined from the total fluid volume infused minus the urine output every 3 h. Net fluid balance accumulation was calculated and represented as ml/kg/h. During this experiment, the animals were allowed free access to food but not to water, for the purpose of accurate measurement of fluid intake.

The animals were randomized into two groups: An MEG group (given 30 mg/kg of MEG at 1 h after injury and then every 8 h for 41 h; n = 6) and a control group (given 0.9% NaCl in same manner as the MEG; n = 6). MEG was prepared by the Inotek Corporation (Beverly, MA) as previously described (16). We dissolved the MEG in normal saline, and adjusted its pH to 4.0 at 15 min before administration. The infusion rate was 150 ml/h. The dose and the timing of administration of MEG were based on previous study that showed the beneficial effect of MEG on hemorrhagic shock without evidence for inhibition of constitutive NOS (36). In addition, the MEG was administrated to healthy sheep without any injury in the same manner as to sheep in hemorrhagic shock, and it was confirmed prior to our experiments that the compound itself had no effect on cardiopulmonary function in this dose. The lymph and blood samples for determination of total protein concentration, colloid osmotic pressure, and NOx were collected from both groups of animals in our study at 3, 6, 12, 18, 24, 36, and 48 h after injury. Hemodynamic variables and blood gases were measured at 3, 6, 12, 18, 24, 30, 36, 42, and 48 h after injury in both groups. At 48 h after injury, animals were killed with saturated KCl after injection of ketamine. Immediately after killing, a piece of right lung tissue was harvested for measurement of iNOS messenger RNA (mRNA) expression in the lung tissue, as subsequently described, and the rest of the right lung was used for measurement of blood-free wet weight-to-dry weight (wet/dry) ratios, as described by Pearce and colleagues (27).

Measurement of iNOS mRNA Expression in Lung Tissue

mRNA isolation. Lung tissue from animals at 48 h after injury was excised at the time of killing and immersed in liquid nitrogen. Total RNA was obtained with a commercially available total RNA purification kit (Purescript; Gentra Systems, Inc., Minneapolis, MN). Briefly, 100 mg of freshly frozen lung tissue was lysed and homogenized with a mortar and pestle with 3 ml of lysis buffer containing EDTA, citric acid, and sodium dodecylsulfate (SDS) according to the kit manufacturer's protocol. Precipitation buffer was then added and the preparation was incubated for 10 min on ice to precipitate protein and DNA, before being centrifuged at 3,000 × g. The supernatant was placed in 3 ml of isopropanol and centrifuged at 3,000 × g for 5 min. The pellet was washed with 3 ml of 70% ethanol, centrifuged again, and air dried for 10 min. The pellet was then resuspended in diethylpyrocarbonate-treated water. Total RNA was quantitated spectrophotometrically at 260 nm. mRNA was isolated from total RNA with the Straight A's mRNA Isolation System (Novagen, Madison, WI) purification procedure, in which mRNA is first hybridized to oligodeoxythymidine-coupled magnetic beads, washed, and then eluted to obtain polyadenylated mRNA according to the manufacturer's protocol.

Preparation of sheep lung complementary DNA. First-strand complementary DNA (cDNA) was synthesized by reverse transcription (RT) of the mRNA samples obtained, using Moloney murine leukemia virus-derived reverse transcriptase (Perkin Elmer, Branchburg, NJ) and random hexamers for priming according to standard procedure (22). The cDNA was then used as a template for quantitative polymerase chain reaction (PCR).

Quantitative PCR of iNOS cDNA. A competitive internal standard based on sequences from the v-erB gene was used to quantitate iNOS cDNA, using a 414-bp fragment of this gene that was constructed to contain 5' and 3' ends that were identical to the corresponding primers utilized for the amplification of iNOS cDNA (23). Primers for sheep iNOS were as follows: the sense primer was 5'-GCTCATCTTCGCCACCAAGC-3' and the antisense primer was 5'-GCCATCTGGCATGTGGTAGC-3', yielding a product of 271 bp. These primers were chosen on the basis of sequences previously determined for bovine and caprine mRNAs for iNOS (24, 25). These highly homologous sequences (~ 97% homology at the nucleotide level) were taken from a portion of the iNOS gene that differed significantly from the genes for other known types of NOS in order to ensure specificity. Initially, 10-fold dilutions of the competitive cDNA template (v-erB gene sequence) were mixed with a constant amount of sample cDNA and subjected to amplification with a final concentration of 2 mM MgCl₂, 0.375 U Taq polymerase (AmpliTaq; Perkin-Elmer), and 0.2 mM deoxynucleotide triphosphates in a reaction volume of 15 µl. The mixture was amplified for 35 cycles at a melting temperature of 94° C for 1 min (6 min on the first cycle) and an annealing temperature of 57° C for 1 min, with extension at 72° C for 1 min (7 min on the final cycle).

The amplification products were electrophoresed on a 1% agarose gel containing ethidium bromide. After an approximate equivalence point was ascertained from the 10-fold dilutions of competitive cDNA template, twofold dilutions were made and the resulting bands were digitized. Curves of the band intensities were constructed to allow interpolation between dilutions for more accurate determination of equivalence points.

Immunohistochemical detection of nitrotyrosine. Paraffin sections (3 µm) were deparaffinized in xylene and rehydrated in decreasing concentrations (100%, 95%, and 75%) of ethanol, with an ensuing 10-min incubation in phosphate-buffered saline (PBS) (pH 7.4). Sections were treated with 0.3% hydrogen peroxide for 15 min to block endogenous peroxidase activity, and were then rinsed briefly in PBS. Nonspecific binding was blocked by incubating the slides for 1 h in PBS containing 2% goat serum. To detect nitrotyrosine, routine histochemical procedure was applied, as previously described (2), with minor modifications, as follows. Rabbit polyclonal antinitrotyrosine antibody (Upatate Biotechnology, Lake Placid, NY) was applied in a dilution of 1:300 for 2 h at room temperature. Reaction with the primary antibody was also conducted in the presence of 10 mM nitrotyrosine, as a negative control. After extensive washing of each section (five washes for 5 min each) with PBS, immunoreactivity was detected with a biotinylated goat antirabbit secondary antibody and the avidin-biotin- peroxidase complex (ABC), both supplied in the Vector Elite kit (Vector Laboratories, Burlingame, CA). Color was developed with Ni-diaminobenzidine (DAB) substrate (95 mg DAB, 1.6 g NaCl, and 2 g NiSO₄ in 200 ml 0.1 M acetate buffer). Sections were then counterstained with nuclear fast red, dehydrated, and mounted in Permount. Photomicrographs were taken with an Axiolab microscope (Zeiss, Jena, Germany) equipped with a Fuji HC-300C digital camera (Figure 5).

View larger version (78K): [in this window] [in a new window]   Figure 5.   Immunostaining with anti-iNOS antibody produced labeling of the epithelium of bronchi and bronchioles that tended to be patchy, and staining of bronchial glands. On average, somewhat more intense staining was seen in animals treated with MEG, and the staining tended to be more uniform within individual bronchioles.

Immunostaining for iNOS. Six-micron-thick paraffin sections were rehydrated and treated for 10 min with 0.3% hydrogen peroxide in 100% methanol to quench endogenous peroxidase, and were then treated with protease for 8 min. Immunostaining was done with an automated instrument (Ventana ES, Tucson, AZ). The primary antibody was a polyclonal rabbit antibody to human iNOS (Catalog #PA3-030A; Affinity Bioreagents, Golden, CO). Incubation with primary antibody was done under optimized conditions, using a 1:200 antibody dilution, for 32 min at 42° C. Incubation in secondary antibody and development of reaction product with an ABC complex and DAB were done with Ventana reagents, including the DAB kit. Controls included omission of the primary antibody and substitution of a nonimmune rabbit serum control provided by Ventana, and no reaction product was seen in these controls (Figure 6).

View larger version (56K): [in this window] [in a new window]   Figure 6.   Tracheas from (A) vehicle- and (B) MEG-treated animals were stained for nitrotyrosine. Arrows in A indicate nitrotyrosine positive chondrocytes.

Statistical Methods

Summary statistics of data are expressed as mean ± SEM. Data were analyzed through analysis of variance for a two-factor experiment with repeated measures at specific times. The two factors were experimental groups and time. Fisher's least significant difference procedure was used for multiple comparisons (or post hoc analysis). The differences between groups in iNOS mRNA expression and wet/dry weight ratios of the right lung were evaluated by means of Student's unpaired t test. A value of p < 0.05 was considered to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All animals survived the 48-h experimental period after the combined injury of 40% TBSA third-degree burn and smoke inhalation, with fluid resuscitation strictly following the Parkland formula. The arterial carboxyhemoglobin levels immediately after smoke exposure were 62.2 ± 9.08 (mean ± SEM) % in the control group and 65.5 ± 7.34% in the MEG group. There was no statistical difference between these values.

The quantity of iNOS mRNA in the right lung tissue obtained at the end of experimentation is shown in Table 1. The sham group represents the normal value in our laboratory for iNOS mRNA expression in lung tissue from healthy sheep without any injury (n = 5). There was a significant increase in iNOS mRNA expression after injury in both study groups as compared with noninjured animals. Although there was no significant difference between the study groups, iNOS mRNA in the MEG group tended to be higher than in the control (injury without treatment) group. Immunostaining of the lungs showed the presence of iNOS on the epithelium of the bronchi, bronchioles, and bronchial glands. The staining was more intense in the animals treated with MEG (Figure 6). Nitrotyrosine was present in the same areas as iNOS in the injured group that was not treated. Nitrotyrosine was not present in the MEG-treated animals. The NOx levels in plasma (PNOx), lung lymph (LLNOx), and burned-tissue lymph (BLNOx) increased significantly from baseline values in the control group, whereas in the MEG group no significant increase in NOx was observed during the study (Figure 1). Notably, significant differences between the groups were observed beginning at 24 h after injury.

View larger version (50K): [in this window] [in a new window]   Figure 1.   Values of PNOx, LLNOx and BLNOx. The control group showed a significant increase in both PNOx and LLNOx/BLNOx beyond 24 h after injury. The MEG group did not show any significant increase in NOx throughout the experiment.  p < 0.05 versus baseline value, *p < 0.05 versus control group, p < 0.05.

Cardiopulmonary hemodynamic data are summarized in Table 2. Although a large amount of fluid was administered rapidly, the cardiac index (CI) in the control group decreased significantly immediately after injury, remained depressed for more than 24 h after injury, and then gradually returned toward baseline values (Table 2). The decrease in CI was associated with a decrease in the stroke-volume index. A transient increase in the hct was also observed in the control group (Table 2). The hct in the control group peaked at 24 h after injury (152.9 ± 5.60% of its baseline value). In the MEG group, the CI increased from baseline values after injury with fluid resuscitation, rather than decreasing. The hct did not show any significant change from its baseline value in the treatment group, and there was a statistically significant difference from the control group between 12 h and 30 h after injury. However, the MAP was maintained in the normal range in both groups throughout the experimental period. The Ppa, CVP, and Pla were increased after injury and under mechanical ventilation with PEEP. There was no statistically significant difference between the study groups in the MAP, Ppa and CVP, although CVP trended to be higher in the control group. The Pla was significantly higher in the control group (Table 2).

The urine output was significantly higher in the MEG group than in the control group beyond 12 h after injury (Figure 2a). The net fluid accumulation in the MEG group was significantly less than in the control group (100.3 ± 16.4 ml/kg/ min in the control group and 41.5 ± 15.5 ml/kg/min in the MEG group at 48 h after injury; p < 0.05) (Figure 2b). The control group showed a sustained increase in net fluid accumulation during the 48-h experimental period, whereas in the MEG group fluid accumulation reached a plateau within 12 h and subsequently decreased at 24 h after injury.

View larger version (39K): [in this window] [in a new window]   Figure 2.   (a) Urine output. Urine output was significantly higher in the MEG than in the control group. (b) Net fluid accumulation: The net fluid accumulation was significantly smaller in the MEG than in the control group. *p < 0.05 versus control group.

Lung lymph flow was markedly attenuated in the MEG group as compared with the control group (Figure 3a). There was a significant difference between the two groups from 18 h after injury. IPI was also significantly lower in the MEG group (Figure 3b). In the control group, lPI increased significantly, and reached a peak at 36 h after injury. In the MEG group, IPI was maintained at its baseline level.

View larger version (22K): [in this window] [in a new window]   Figure 3.   (a) Changes in lQL. The increase in lQL was attenuated significantly by iNOS inhibition. (b) Changes in lPI. lPI was significantly lower in the MEG than in the control group.  p < 0.05 versus baseline, *p < 0.05 versus control group.

The blood-free wet/dry weight ratio of the right lung, which represents the degree of pulmonary edema, was significantly lower in the MEG group than in the control group (4.86 ± 0.20 in the control group, versus 3.78 ± 0.15 in the MEG group; p < 0.05) (Figure 4). The sham group represents the normal value for the wet/dry weight ratio of the right lung in our laboratory for healthy sheep without any injury (3.37 ± 0.17, n = 6). The study control group showed a significant increase in the wet/dry weight ratio as compared with the normal value, whereas in the MEG group there was no significant difference from the normal value.

View larger version (32K): [in this window] [in a new window]   Figure 4.   Blood-free wet weight-to-dry weight (wet/dry) ratios of right lung. Both the control and MEG groups received a 40% TBSA third- degree burn and were insufflated with smoke. The MEG group was treated with MEG and the control group was treated with 0.9% NaCl. The sham group consisted of animals that received neither burn nor smoke injury (n = 6). *p < 0.05 versus control group,  p < 0.05 versus sham group.

The prefemoral lymph flow, which represents bQ_L, increased significantly in both study groups after the combined injury, and no statistical difference was observed between the two groups (Table 3). There was also no difference in the bPI between the study groups (Table 3).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although mortality from burn injury has decreased, it is still high if the injury is combined with smoke inhalation (26). In these patients the lung is a critical organ, because the progressive respiratory failure associated with lung edema is an important determinant of mortality (10, 26). The lung edema has been attributed to pulmonary microvascular hyperpermeability and subsequent iatrogenic fluid overload (10). In cutaneous burns that exceed to 25% TBSA, vascular hyperpermeability occurs not only at burned skin, but also in regions distant from the injury (6, 7). The lung is a primary target organ, since the blood that drains burned tissue enters the pulmonary circulation before passing through any other organ (29). Demling and colleagues, investigating lymph flow in the subject tissue, reported that pulmonary microvascular permeability (30) and the permeability of unburned soft tissue (19) increased after a major cutaneous burn. The resulting vascular leakage leads to a huge amount of fluid flux from circulating blood into interstitial space, and the consequent edema formation is more severe when thermal injury is associated with smoke inhalation injury (12). However, the mechanism responsible for these alterations in vascular permeability of the burned site itself and of distant regions in this kind of injury has not been defined.

Recent investigations of the vascular hyperpermeability seen with sepsis or burns have focused on NO as an important factor (4, 5, 31). However, there is still controversy about whether NOS inhibition prevents or augments changes in vascular permeability. Kubes and colleagues reported that NOS inhibition with L-nitroarginine methyl ester (L-NAME) a nonselective inhibitor of NOS, increased systemic microvascular permeability to protein in a feline model of sepsis (31). On the other hand, Hinder and colleagues reported that L-NAME did not affect the pulmonary microvascular hyperpermeability seen in an ovine sepsis model (4). Sozumi demonstrated that inhibition of NOS with either L-NAME or aminoguanidine suppressed the vascular permeability of second-degree burned skin in an ear thermal injury model of mice (5). The latter compound is a specific inhibitor of iNOS. The reason for these conflicting results is still unclear. Rodents are notorious for making NO in cells and in quantities far in excess of what is seen in larger mammals. Aminoguanidine was ineffective in reducing the increased albumin flux seen with burn injury until the dosage was very high. We are relatively certain that the dosage of MEG that we used in our study was effective in blocking iNOS, because the nitrate and nitrite concentration in the lymph draining the burn wound was not increased as it was in the untreated animals. Transvascular fluid flux is the result of an increased porosity of the microvasculature and an increased perfused surface area. Although we did not measure blood flow to the burned tissue, we might have seen, had we done so, that it increased with MEG administration, since CO and arterial pressure were elevated. In our study we may therefore have reduced the increased permeability seen with burn and smoke inhalation, but may also have increased blood flow as a result of increased microvascular pressure and an increase in the surface area perfused. Such changes in blood flow were unlikely to have occurred in the study by Sozumi, since the burn injury he used was only of second degree. As a result, the vessels would have been dilated. Moreover, the injury was too small (involving only the ears of the mouse) to affect the total cardiovascular picture. There is some evidence, however, that iNOS inhibition affected systemic transvascular fluid flux. In our study, hemoconcentration and fluid retention were seen in the injured animals after burn and smoke inhalation. These generic markers of hypovolemia were not seen when the animals were given MEG. Certainly these systemic aspects of MEG treatment will require further study.

In the present study, the quantity of iNOS mRNA in lung tissue after combined burn and smoke inhalation injury was determined and the result revealed a significant increase in iNOS mRNA expression. Similarly, the immunostaining for iNOS was more intense. Additionally, we found no evidence for the presence of peroxynitrite, an oxidative metabolite of NO, in ovine pulmonary tissue after combined burn and smoke inhalation injury, as did Kooy and coworkers in human pulmonary tissue after smoke inhalation injury (32). These data are evidence for the production of iNOS-generated NO in lung tissue after this kind of injury. In our study, the group treated with MEG showed slightly greater expression of iNOS mRNA than did the untreated control group. This result may suggest the existence of a feedback system in the iNOS NO pathway, since MEG inhibits iNOS activity.

The control group in our study exhibited the typical hemodynamic response to combined burn and smoke inhalation injury. Despite receiving large amounts of fluid, they showed a significantly low CO associated with hemoconcentration. In contrast, neither the decrease in CO nor hemoconcentration occurred in the group in which iNOS was inhibited. In addition, fluid requirement was significantly less in the treated group than in the control group. These findings suggest that inhibition of iNOS prevents systemic vascular changes after combined cutaneous burn and smoke inhalation injury. The reduced CO may also have a myocardial contractility component, since both it and a reduced stroke volume were seen at a time when Pla was increased. Pla did not rise to the same extent in the group treated with the iNOS inhibitor, suggesting that NO may have been responsible for this apparent decrease in myocardial contractility. NO has been reported to depress myocardial contractility (33, 34).

The increase in lQL and LPI were also significantly lower in the MEG group than in the control group. Additionally, the lung wet/dry weight ratio was maintained in the normal range in the animals in which iNOS was inhibited. These data suggest that pulmonary microvascular hyperpermeability to both fluid and protein was also suppressed by iNOS inhibition.

In the present study, we also determined the total amount of NOx, in plasma, lung lymph, and burned-tissue lymph. The difference in the level of NOx between the control group and the MEG group (iNOS-inhibited group), should have been proportional to the amount of NO associated with iNOS. In the control group, an increase in iNOS-generated NO was detected as early as 3 h after injury, and achieved significance at 24 h. In contrast, the treatment group did not show changes in plasma NOx. In lung lymph, NO from iNOS was not identified in the initial 12 h, but increased significantly after 24 h. It peaked at 36 h after injury, which is compatible with the time of the peak lPI in our study. This marked increase in NO production the in later phase after injury may explain the delayed onset of pulmonary failure and pulmonary microvascular hyperpermeability after burn and/or smoke inhalation injury. In the control group, the LLNOx level in the later phase tended to be higher than the PNOx level, although the difference was not statistically significant. This finding may suggest that some part of iNOS-generated NO is produced in lung tissue.

Although our study was not designed to evaluate the changes in vascular permeability that occurred within 3 h after injury, there must have been some alternation in systemic vascular permeability, because significant increases in hct were seen at 3 h after injury. Our data cannot define why administration of MEG affected these early vascular changes, since iNOS is not expressed before cell activation by inflammatory cytokines, and this requires several hours (37). Our speculation is that the peroxynitrite that can be scavenged by MEG also plays some role on the microvascular changes seen after injury. It has been shown that peroxynitrite is generated from the NO synthesized by constitutive NOS after hemorrhagic shock, sepsis, or ischemia-reperfusion injury (35, 36). Another possibility is that arachidonic acid metabolism influences the early vascular alterations that follow burn/inhalation injury, since it has been suggested that MEG is also a cyclooxygenase (COX) inhibitor (37). Lipoxygenase inbhibitors have been shown to prevent the decrease in CO reported after smoke inhalation alone, without burn (38), and there is evidence that leukotriene D₄ may work via the release of COX products in sheep. Further investigations of these hypotheses will be needed.

From these data, we conclude that: (1) iNOS-generated NO has an important role in the changes in both systemic and pulmonary microvascular permeability that follow combined cutaneous burn and smoke inhalation injury. (2) Our third- degree burn model showed no evidence for a role of iNOS-generated NO in burned tissue permeability. (3) Inhibition of iNOS may have a beneficial therapeutic effect on the systemic and pulmonary microvascular hyperpermeability and edema formation that follow combined burn/inhalation injury.