INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In 1967, Ashbaugh and colleagues (1) described 12 patients who died of acute respiratory failure with hypoxemia and decreased lung compliance and they subsequently termed this disease the adult respiratory distress syndrome. Now referred to as the acute respiratory distress syndrome (ARDS) (2), ARDS represents a complex lung injury that may result from a variety of pulmonary insults, both direct and indirect (2). Despite the diverse nature of initiating events that can lead to ARDS, there exists a final common pathway resulting in diffuse alveolar damage and significant lung dysfunction. Although research efforts over the past several years have provided a greater understanding of the potential mechanisms leading to ARDS, adequate treatments have not yet been developed. Consequently, the mortality rate of these patients remains unacceptably high at 40-60% (2).

Pulmonary surfactant is the material that coats the inner surface of the alveoli and reduces the surface tension in the lungs to prevent alveolar collapse. Since the physiologic and morphologic findings in patients with ARDS resembled those found in the surfactant deficient state of premature infants with the respiratory distress syndrome (RDS), it was postulated that surfactant abnormalities may contribute to the pathophysiology of this disease (3). Data available from several animal models of ARDS have confirmed that alterations of the endogenous surfactant system contributed to the lung dysfunction associated with this condition (3).

Of the many etiologies leading to ARDS, sepsis is the most common predisposing factor and is associated with the greatest mortality rate compared with other causes of this disease (7). Pulmonary involvement in sepsis, specifically the development of alveolar edema and respiratory failure represents just one manifestation of a systemic response to a peripheral focus of infection. It has been suggested that the lungs themselves may represent an inflammatory organ, contributing to the systemic septic response by releasing lung-specific mediators into the circulation (8, 9). The lungs may therefore play an important role in the initiation and progression of multiple system organ failure (MSOF) associated with systemic sepsis.

One factor that has limited our ability to extensively study ARDS is the lack of a suitable animal model that adequately represents the clinical course of patients with this disease. For example, most animal models of acute lung injury have utilized direct insults to the lungs to cause lung dysfunction. Although some of these models may reflect patients that actually have direct insults to their lungs, such as aspiration, near-drowning or toxic fume inhalation (10, 11), these models do not represent the majority of patients that develop ARDS from indirect insults.

The objective of the present study was to characterize the lung injury associated with systemic sepsis in adult rats. In addition to demonstrating physiologic and morphologic changes in these animals, we focused on characterizing changes in the endogenous surfactant system. A greater understanding of these alterations in systemic sepsis-induced lung injury may lead to investigating therapeutic strategies aimed at improving the outcome of patients with ARDS.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal Model

Male Sprague Dawley rats weighing 350 to 425 g were acclimatized to the laboratory environment for 1 wk prior to surgery. Free access to food and water was available for this time period. For preparation of surgery, rats were anaesthetized with 4% halothane in oxygen in an anaesthetizing box. Once induced, the rats were transferred to a surgical table with a nose cone set up to maintain anaesthesia. Insertion of required monitoring catheters in the right external jugular vein and right carotid artery were performed. Both were cannulated with a PE-50 tube. The lines were then routed subcutaneously to the back of the neck and attached to a 3 fluid channel (22 G) swivel system. The incision under the neck was closed with 3-0 silk suture.

To induce sepsis, the cecal ligation and perforation (CLP) technique was performed as previously described (12). Based on previous experiments, two different surgical techniques to induce sepsis was utilized. In one group (septic), the CLP procedure consisted of ligating the cecum at the ileocecal valve, while still maintaining bowel continuity. The cecum was then punctured twice with a 16-gauge needle, one just above the ligature and the other at the tip of the cecum. The cecum was then compressed to extrude bowel contents into the peritoneum. A separate group of animals were characterized that had identical cannulation procedures, but only the proximal one-third of the cecum was ligated. The ligated portion of the cecum was punctured twice in these animals, similar to the aforementioned group and bowel contents extruded into the peritoneum. This procedure not only lead to the development of sepsis similar to the septic group but also resulted in a relatively mild degree of lung injury. This group was designated as septic + LI. Of note, animals in the sepsis groups were randomly assigned to one of the two surgical procedures that induced sepsis plus or minus lung injury. A sham operated group was used as a control group for the animals undergoing the CLP procedure. In sham animals, induction of anaesthesia and catheter placement were identical to the septic animals, although no laparotomy or CLP procedure was performed.

After attachment of a harness to secure the swivel system, each rat was placed in a plastic cage to recover. The swivel device allowed rats unlimited movement within the cage and free access to rat chow and water. Post-operatively all animals received a continuous infusion of saline at 7.5 ml/kg/h containing 2 µg/ml of fentanyl for analgesia via the venous line. The arterial line was flushed with heparinized saline (1 U/ ml) at 1 ml/h to maintain patency. Measurements of Pa_O2 and PCO₂ were performed on a ABL 500 blood gas analyzer (Radiometer, Copenhagen, Denmark). Arterial lactate levels were measured on a YSI 2300 STAT Plus glucose/lactate analyzer. Mean arterial blood pressure (MAP) and heart rate (HR) were recorded by attaching a pressure transducer to the arterial line and reading the resulting MAP and HR on a blood pressure monitor. The respiratory rate (RR) was recorded by counting the number of breaths in 15 s and multiplying by 4. All these measurements were recorded 4-5 h after surgery and at 24-28 h after surgery, which was immediately prior to death. All animals were killed by deeply anaesthetizing the rat with halothane, incising the abdominal cavity, and transecting the descending aorta.

Lung Lavage Analysis

Immediately after death, the chest cavity was opened to visualize the lungs while a whole lung lavage procedure was performed as previously described (4). The lungs were infused with 0.15 NaCl at room temperature until fully distended, withdrawn, and reinfused two more times. This procedure was repeated a total of five times and the combined volume of the crude alveolar wash (CAW) was recorded. There were no differences in the volume of saline infused or recovered after the whole lung lavage between the three experimental groups.

A 10 ml aliquot of the CAW was used for analysis of total surfactant pool size, total surfactant protein-A (SP-A) levels and total protein levels. The remainder of the CAW was centrifuged at 150 × g for 10 min to yield a pellet containing cellular debris. The 150 × g supernatant was then spun at 40,000 × g for 15 min, yielding a supernatant that was called the small aggregate surfactant fraction (SA). The 40,000 g pellet was suspended in 2 ml of saline and called the large aggregate surfactant fraction (LA). This technique for separating alveolar surfactant subtypes has been previously described (4).

To measure surfactant pool sizes, aliquots of the CAW, 150 g pellet, LA, and SA fractions were extracted using the method of Bligh and Dyer (13). Phospholipid levels in each lipid extract were determined using a modification of the Duck-Chong phosphorous assay (14). Briefly, 100 µl of 10% magnesium nitrate in methanol was added to the extracted lipids. After drying, the samples were ashed in a fume hood on an electric rack on high for approximately 1 min. After 1 ml of 1 M HCl was added, the samples were reheated on a heating block while covered for 15 min at 95 to 100° C. After cooling, 1 ml of 4.2% (w/v) ammonium molybdate in 4.5 N HCl and 500 µl of 10% (w/v) ascorbic acid in dH₂O were added and samples were heated again in a water bath at > 60° C for 15 min. After samples had cooled the absorbency was read at 820 nm. The total protein content of the CAW was determined by the method of Lowry and colleagues (15) using bovine serum albumin as a standard.

Surfactant-Associated Protein-A Analysis

An enzyme-linked immunoabsorbent assay (ELISA) was used to quantitate SP-A in the CAW obtained from the three experimental groups. Wells were coated with a 1/200 dilution of concentrated mouse lung lavage in 0.1 NaHCO₃, overnight. Samples and standards were diluted with water and mixed with a 1/8,000 dilution of rabbit anti-mouse SP-A in 82 b (0.15 M NaCl, 0.01 Tris (pH = 7.4), 5 µg/ml BSA) with 5% goat serum and incubated overnight at 37° C. The next day, the ELISA plates were washed with wash buffer (0.01 Tris, 0.05% Tween 20) and conditioned briefly with 82 b containing 5% goat serum. The samples and standards were added to the wells and incubated with goat anti-rabbit IgG conjugated to peroxidase (Calbiochem, La Jolla, CA) in Dilution Buffer (0.05 M NaPO₄, pH = 6.3, 0.15 M NaCl, 0.05% Tween 20) for 1 h. Color was developed using a standard orthophenylene diamine (OPD) substrate system. The SP-A concentration was estimated by comparing the binding of the antibody in the presence and absence of sample or standard by graphing the concentration of SP-A versus the percent inhibition. A linear standard curve was used and when the samples were too concentrated, they were diluted until they fit on the most linear portion of the standard curve.

Biophysical Analysis

Surface tension measurements of unextracted samples of LA were performed. Aliquots of the appropriate samples were resuspended to a final concentration of 1 mg phospholipid/ml in 0.15 NaCl/1.5 mM CaCl₂. All samples were incubated for 90 min at 37° C and analyzed using a pulsating bubble surfactometer as described by Enhorning (16).

Morphological Assessment

Separate animals within each group were studied for histological analysis. Animals underwent either the sham or CLP procedures and just before death were anaesthetized with halothane, and a tracheotomy was performed. The endotracheal tube was connected to a pressure gauge attached to a self-inflating anaesthesia bag. An abdominal incision was made to expose the caudal vena cava, and a 20 gauge catheter was inserted into the vena cava at the level of the portal vein. The abdominal aorta was then severed, and the blood was cleared from the pulmonary vessels via antegrade perfusion of the vena cava with physiological buffered saline (0.9% NaCl, 0.3% NaNO₂, and 100 U/ml of heparin) at a hydrostatic pressure of 10-15 cm H₂O. The animals were ventilated with an opened chest cavity using the self-inflating anaesthesia bag (10-20 cycles/min). Fifteen minutes after the initiation of clearance, the lungs were maintained at a constant intratracheal pressure of 12 cm H₂O and perfused with fixative solution (2.5% paraformaldehyde and 1.5% glutaraldehyde in 1 M phosphate buffer, 440 mosmol, pH 7.4). Infusion of fixative was performed at a hydrostatic pressure of 10-12 cm H₂O for 15 min to ensure adequate fixation. The heart and lungs were then removed en bloc and stored in fixative solution.

Subsequently, the lobes of the lungs were separated and sectioned sagittally. After gross examination, one section was taken from each lobe for light-microscopic examination. The sections were embedded in paraffin, cut at 5 µm thickness, and stained with haematoxylin and eosin. Sections were examined for features of lung injury including, congestion, atelectasis, accumulation of inflammatory cells, alveolar edema, and hyaline membrane formation. The degree of injury and the distribution of the injury, being uniform or nonuniform within the various lobes of the lung was assessed. All morphologic assessments were done by a pathologist blinded with respect to the different experimental groups studied.

Calculations and Statistics

The alveolar-arterial O₂ gradient (PA_O2 - Pa_O2) was calculated using the alveolar gas equation to determine the alveolar PO₂ (PA_O2). PA_O2 = (PB-PH₂O) × FI_O2  (PA_CO2/R), where PB is barometric pressure, PH₂O is the partial pressure of water vapor (47 cm H₂O), FI_O2 is the inspired O₂ fraction, PA_CO2 is the mean alveolar PCO₂ (assumed to be equal to the arterial PCO₂ or PA_CO2), and R is the respiratory quotient (assumed to be equal to 0.8). Data is expressed as means ± standard error (SE). Values between groups were compared using one-way analysis of variance (ANOVA). The multiple comparison test by Bonferroni was used to test for differences among groups. An unpaired Student t test (two-tailed) was used to compare within groups. A probability level of p < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Physiologic Analysis

Table 1 shows the body weights and baseline physiological measurements taken 4 to 5 h after recovery from surgery for the sham, septic, and septic + LI experimental groups. There were no significant differences in A-a gradients, heart rates, respiratory rates, and arterial lactate levels between the three experimental groups at baseline. However, the PO₂ values in the septic + LI group were significantly higher compared to the sham control group, and the mean arterial pressure was significantly lower in the septic group compared to the sham group.

Over the 24 to 28 h period following surgery, there were obvious differences in the gross appearance of the animals among the different experimental groups. Animals in the septic and septic + LI groups developed lethargy, crusty haemorrhagic exudates around the eyes and nose, abdominal distension, and piloerection. These changes were not present in the sham group. Table 2 shows the physiologic parameters of the three experimental groups measured just prior to sacrifice. The septic and septic + LI groups had a decrease in mean arterial blood pressure over the experimental period compared with their respective baseline values. These values were also significantly lower (p < 0.05) then the sham group prior to death. Although, heart rates and respiratory rates were not significantly different among the three experimental groups at baseline, just prior to death, within both septic groups, mean heart rate (p < 0.05) and mean respiratory rate (p < 0.05) was significantly higher compared to the sham group. Arterial lactate levels also increased in both the septic and septic + LI groups 24-28 h after surgery compared with baseline values and were significantly greater than the sham group (p < 0.05). In addition to the changes in these parameters, the two septic groups undergoing the CLP procedure, also had positive blood cultures growing gram negative enteric bacteria, in contrast to animals in the sham group that had negative blood cultures (data not shown).

Oxygenation values of the three experimental groups just prior to sacrifice are shown in Figure 1. Pa_O2 values (Figure 1A) were significantly lower in the septic + LI group compared with both the sham (p < 0.05) and septic (p < 0.05) groups prior to death. This value in the septic + LI group was also significantly lower compared to its respective baseline value (p < 0.001). There were no significant differences in Pa_O2 values between the sham and septic groups (Figure 1A) and in both groups, these values were not significantly different compared to their respective baseline values. Although Pa_CO2 values were slightly higher in the septic + LI group compared to the other two groups (data not shown), the alveolar to arterial (A-a) oxygen gradient values (Figure 1B) for the septic + LI group was significantly greater than both the sham (p < 0.05) and septic (p < 0.05) groups at death. This value was also significantly higher compared with the respective baseline value (p < 0.05). There were no significant differences between the sham and septic groups at death, or within these groups compared with their respective baseline values.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Septic with lung injury (septic + LI) animals had significantly decreased (A) PaO2 (p < 0.05) and increased (B) alveolar- arterial (A-a) oxygen gradient (p < 0.05) values at time of death compared with septic and sham animals.

Morphological Analysis

The morphological appearance of lungs from the sham, septic, and septic + LI animals are shown in Figure 2. Septic animals were very similar to sham animals with no morphologic evidence of lung injury. Septic + LI animals on the other hand, showed features of mild lung injury with patchy atelectasis, congestion, and inflammatory infiltrates composed of neutrophils and macrophages. However, vascular thrombosis, edema and hyaline membrane were not present in these animals.

View larger version (96K): [in this window] [in a new window]   View larger version (106K): [in this window] [in a new window]   View larger version (119K): [in this window] [in a new window]   View larger version (143K): [in this window] [in a new window]   Figure 2.   Hematoxylin and eosin stained light micrographs of sham (A) and septic (B) rat lungs showing normal alveolar spaces and interstitium with no evidence of injury (original magnification ×112). Septic + LI lungs (C and D) showed early mild injury with patches of atelectasis (C, arrow, original magnification ×112), vascular congestion, and an inflammatory infiltrate composed of neutrophils and macrophages (D, original magnification ×450).

Lung Lavage and Surfactant Analysis

The total protein recovered from the CAW in the three experimental groups was 8.8 ± 1.4, 16.7 ± 5.2, and 17.6 ± 3.8 mg/kg of body weight for sham, septic, and septic + LI, respectively. Although protein values were higher in both septic and septic + LI groups compared with the sham group, due to the variance among groups, only the septic + LI was statistically significant compared with the sham group (p < 0.05).

Total surfactant phospholipid pool sizes of the CAW from the three experimental groups is shown in Figure 3A. The septic + LI group had significantly lower quantities of surfactant phospholipid recovered from the CAW than the sham group (p < 0.05). Although, the surfactant pool size in the septic group was lower than the sham group, there were no significant differences between this group and the other two groups.

View larger version (17K): [in this window] [in a new window]   Figure 3.   Septic + LI animals had significantly lower (A) total surfactant pool sizes (p < 0.05) and a significantly lower (B) small aggregate pool size (SA) (p < 0.05) recovered from the lung lavage compared with sham animals.

The measured pool sizes of the small aggregates (SA) and large aggregates (LA), are shown in Figure 3B. There was a decrease in the SA pool size in both septic groups and this was significant in the septic + LI group compared with the sham group (p < 0.05). There were no significant differences in LA pool sizes among the three experimental groups. These relative changes in surfactant aggregate pool sizes have previously been expressed as SA/LA ratios (5, 18). In this study, the SA/LA ratio for the sham, septic, and septic + LI groups were 0.93 ± 0.07, 0.62 ± 0.09, and 0.42 ± 0.08, respectively. Values for the septic and septic + LI groups were both significantly lower compared with the sham group (p < 0.05).

Figure 4 shows the total quantity of SP-A recovered from the CAW of the sham, septic, and septic + LI groups. There were significantly greater quantities of SP-A recovered from both the septic (p < 0.05) and septic + LI (p < 0.05) groups compared to the sham group. There were no significant differences between the septic and septic + LI groups.

View larger version (13K): [in this window] [in a new window]   Figure 4.   Septic animals and septic + LI animals had significantly increased surfactant protein A (SP-A) levels (p < 0.05) in lung lavage compared with sham animals.

The surface tension measurements of the LA fraction isolated from the three experimental groups revealed that these fractions were all capable of reducing surface tension when tested on a pulsating bubble surfactometer. The surface tension measurements at adsorption for the sham, septic, and septic + LI were 44 ± 3.2, 48.5 ± 5.3, and 38.3 ± 3.7, respectively. The minimum surface tension measurements for the sham, septic, and septic + LI were 21.9 ± 1, 19.8 ± 2.4, and 14.8 ± 2.1, respectively. There were no significant differences in the function of LAs isolated from the sham, septic, and septic ± LI groups at any time point measured during the 5 min of pulsation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Currently, there are no effective treatment strategies available for patients with ARDS, and the mortality rate for this disease remains unacceptably high at 40-60% (2). One limitation to evaluating potential therapies is that there is no ideal animal model that precisely mimics the pathophysiology of ARDS. As a result, some therapeutic approaches have shown promise when tested in animal models but have not proven effective when evaluated in humans (17). Most animal models of lung injury have utilized direct insults to the lungs such as acid aspiration (10), repetitive saline lung lavage (18, 19), or smoke inhalation (11) to cause lung dysfunction. In the clinical setting however, ARDS is more commonly caused by indirect insults to the lungs such as trauma, burns, or, in most cases, sepsis.

Sepsis is not only the most common etiology of ARDS but also carries the highest mortality rate of any of the known causes of this disease (7). The septic response of animals in the present study resulted from a peripheral focus of infection induced by the CLP procedure, as initially described by Wichterman and colleagues (12). In this model, animals developed a systemic response to the CLP procedure which included increased respiratory rate, increased heart rate, increased lactate levels, decreased blood pressure and positive blood cultures. These changes are typical of the changes noted in patients with the systemic inflammatory response syndrome (SIRS), a condition that often develops prior to the onset of ARDS (7).

Despite similar systemic septic changes in the two groups undergoing different CLP procedures, there were significant differences noted in their pulmonary manifestations. The mechanisms responsible for these differences are unknown, but presumably are related to differences in the composition and/or quantities of the various inflammatory cytokines and mediators involved in the respective septic responses. It is possible, for example, that the less aggressive CLP procedure resulted in a systemic inflammatory cascade that specifically targeted the lung compared to the full cecal ligation procedure. Current studies are investigating potential mechanisms involved in these systemic responses, particularly the response resulting in lung injury.

In the septic + LI group, there was physiologic evidence of lung dysfunction reflected by a decreased Pa_O2 and increased A-a gradient values. These values were significantly different compared to both the sham and septic groups prior to sacrifice. Lung injury was also confirmed in the septic + LI by morphological assessment demonstrating some of the early histologic changes noted in patients with ARDS (20). These changes included, patchy atelectasis, congestion and an acute inflammatory response with increased number of neutrophils and macrophages in the interstitium and alveolar space. Finally, there were significantly greater amounts of protein recovered from the CAW of the septic + LI group compared to the sham group, a difference no doubt resulting from the leakage of serum proteins into the alveolar space due to increased pulmonary permeability in this group. These serum proteins have the potential to inhibit the surface reducing properties of surfactant and are felt to contribute to the lung dysfunction associated with lung injury (21, 22).

Surfactant analysis in these animals revealed differences in total surfactant pool sizes, surfactant aggregate pool sizes, and alveolar SP-A levels between the sham and septic groups. Alterations in alveolar surfactant have been described in several models of lung injury, but were interesting in this model for two reasons. First, these changes were noted at a very early stage of lung injury when lung dysfunction was relatively mild. To date, there is some evidence suggesting that the surfactant system is involved and/or contributes to early phases of lung injury. In septic adult sheep (23) and N-nitroso-N-methylurethane (NNMU) induced lung injury in adult rabbits (4), alterations of the surfactant system were demonstrated early in the coarse of lung injury, when lung dysfunction was relatively mild. In both these studies, changes in alveolar surfactant aggregate ratios were not only felt to contribute to the pathophysiology of the lung dysfunction, but also potentially represent a sensitive marker of lung injury. In the current study, alterations of the endogenous surfactant system were observed not only in the septic animals that had a relatively mild degree of lung injury, but also in the group of animals that had manifestations of systemic sepsis, but no evidence of lung injury. In general, studies designed to identify early markers of lung injury are important in order to identify patients potentially at risk of developing ARDS, which would then allow for more specific and earlier therapeutic interventions and prevention of the progression of lung injury.

The second interesting observation from this study was that the SA/LA ratios were lower and SP-A levels were higher in the septic animals versus the sham group. This finding was in contrast to most other models of established lung injury that showed increased SA/LA ratios and decreased SP-A levels recovered in the lavage material obtained from injured lungs, compared to normal lungs.

Large surfactant aggregates, or heavy forms based on buoyant density, are the metabolic precursors to the smaller, vesicular subtypes or small surfactant aggregates. LA are functionally superior to the SA and can reduce surface tension to very low values when tested both in vivo (24) and in vitro (25). Given these functional differences, it has been postulated that an increased SA/LA ratio in animals or humans with lung injury contributed to lung dysfunction. It was somewhat surprising therefore that in the present study, the surfactant aggregate ratio was opposite to that previously reported. Although the reason for this finding is unknown, there are some potential explanations. First, the decreased SA/LA ratio could be a specific response to sepsis and/or lung injury in rats. However, in similar rats undergoing lung transplantation (26), the resultant ischemia-reperfusion injury associated with this injury model resulted in an increased SA/LA ratio. Another possibility was that the altered aggregate ratio in our study was specific to the sepsis model, however a similar septic model induced by CLP in adult sheep resulted in an increased SA/LA ratio (23). Perhaps a more likely explanation for the decreased ratio is related to the severity of illness observed in these animals. For example, in the septic rats with no lung impairment (septic), there was a lower SA/LA ratio than in the sham group. It is possible that a relative increase in the proportion of LA to SA in this situation represented a compensatory mechanism in which the host attempts to conserve alveolar surfactant in the more functional LA form. Consistent with this theory, is that previous studies that reported an increase in the SA/LA ratio utilized models with relatively severe lung injury, suggesting that the surfactant aggregates were isolated at a later stage of the disease process.

Another potential factor to consider when evaluating surfactant aggregates in lung injury is the effects of mechanical ventilation. Mechanical ventilation has been shown to damage the lungs itself (27), and has also been reported to result in the conversion of LA to SA within the airspace (28). In addition, the majority of animal models used to investigate lung injury, including most models of severe lung injury, have involved animals either requiring or already undergoing mechanical ventilation. Interestingly, studies using ozone exposed rats (29) and radiation exposure in mice (30) also reported a decrease in the relative proportion of SA to LA. Animals in these two studies were spontaneously breathing, similar to the septic rats in the present study. Current studies are examining the effects of mechanical ventilation on the severity of the lung injury and on the surfactant alterations observed in this model of sepsis.

As mentioned previously, the increased levels of SP-A observed in the CAW of the septic animals was also in contrast to previous reports of lung injury models (23, 31). SP-A has been reported to have a variety of functions in the lung including, maintaining LA integrity (32), reversing protein inhibition of surfactant (33), and participating in host defense mechanisms (34). The elevated levels of SP-A in the septic rats could be a response to the systemic effects of sepsis and/or a compensatory response in the lung, similar to that described for the decreased SA/LA ratio. Increased levels of SP-A may also have resulted in the stable LA pool size, and contributed to the similar surface tension reducing properties observed in the three experimental groups even though there was evidence of leakage of serum proteins in the septic groups. Further studies evaluating the significance of SP-A in lung function and surfactant metabolism in vivo are required.

In summary we have characterized changes in the endogenous surfactant system in both an animal model of systemic sepsis and sepsis with lung injury induced by a CLP procedure. In the septic + LI group, the lung injury was induced by a systemic septic response from a peripheral focus of infection, a situation common to patients developing SIRS and subsequently ARDS. Although the lung injury demonstrated in these animals was relatively mild, significant changes in the endogenous surfactant system were observed. Therefore, this model represents an ideal situation in which to investigate various issues involving surfactant aggregate conversion, the role of SP-A in lung injury, the effect of mechanical ventilation on a pre-existing lung injury, and different therapeutic strategies for lung injury.