INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lung transplantation has been developed to a viable therapeutic option in a selected number of patients with end-stage lung disease, including pulmonary emphysema, cystic fibrosis, idiopathic pulmonary fibrosis, and others (1). However, the increasing demand for suitable donor lungs far exceeds the number of donor lungs available (2). Limited ischemic tolerance of the lung has remained the limiting factor against the expansion of lung transplantation. At present, flush-perfusion with modified Euro-Collins solution (EC) and hypothermic storage continue to be used as standard preservation of donor lungs. With the currently applied methods, the safe period for reliable preservation is limited to 6 h (3). Furthermore, although early results have improved, postoperative morbidity due to early graft dysfunction remains to be a significant and unpredictable problem (4). Ischemia-reperfusion injury is one of the major problems associated with lung transplantation in the early postoperative course (5). The development of optimal lung preservation method is needed to prolong the storage period and to prevent ischemia-reperfusion injury which leads to an acute allograft dysfunction.

Although the pathogenesis of ischemia-reperfusion injury after lung transplantation remains unclear, reactive oxygen metabolites and neutrophils have been implicated in the pathogenesis (6). Reactive oxygen metabolites mediate the lipid peroxidation detected in postischemic tissues and promote the formation of inflammatory agents that recruit and activate neutrophils. Furthermore, activation of a large number of neutrophils results in the activation of NADPH oxidase and the production of significant amounts of superoxide anion and hydrogen peroxide (7). Many investigations about the effect of antioxidant intervention, such as superoxide dismutase, catalase, and glutathione, which prevent ischemia-reperfusion injury, have supported the role of oxidant-induced damage in lung preservation injury (8).

Since ischemia-reperfusion injury may be characterized by oxidant injury in which reactive oxygen metabolites can be implicated in tissue injury as well as in lipid peroxidation, lung levels of lipid peroxidation have been used as a quantitative measure of lung oxidant injury during reperfusion following ischemia (6, 7, 11, 12). Some investigators have referred to the association of lipid peroxidation with ischemia-reperfusion injury which suggests a causative role for oxidant injury in lung preservation and reperfusion in several animal models of lung transplantation (13). The correlation between lipid peroxide levels in lung tissue and measures of lung dysfunction, such as pulmonary hypertension, increased airway pressure, or impaired oxygenation, suggested a causative role of oxidant injury in lung preservation injury (14, 15). However, indicators of lipid peroxidation were measured in the samples from isolated-perfused lungs or homogenated lung tissue samples in most studies (8, 13). Moreover, the effect of preservation solutions on lung lipid peroxidation has not been investigated (16).

Some investigators showed that lung flush-perfusion and preservation with low-potassium dextran solution (LPD) provided better early pulmonary function than did EC, although the mechanisms responsible for improved storage with LPD and the final conclusion about the use of LPD are uncertain (2, 18).

In this study, in an animal model of orthotopic left-sided single lung transplantation, we measured lipid peroxide levels in bronchoalveolar lavage (BAL) fluid and pulmonary arterial effluent solutions and investigated the effect of flush-perfusion using EC or LPD on lung lipid peroxidation and lung functional parameters. The purpose of this study was to investigate whether changes of lipid peroxidation reflect the lung injury that occurs during lung preservation and transplantation. Furthermore, we investigated whether flush-perfusion and preservation with LPD can reduce lipid peroxidation and may become a strategy to prevent ischemia-reperfusion injury.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Twelve weight-matched pairs of domestic pigs weighing 26.4 ± 1.0 kg underwent left-sided unilateral orthotopic lung transplantation. In the EC group (n = 6), donor lungs were perfused with modified EC that was prepared in the pharmacy of our institute (Apotheke, Klinikum Innenstadt, Munich, Germany). In the LPD group (n = 6), donor lungs were perfused with LPD (Perfadex; Kabi Pharmacia AB, Uppsala, Sweden). The composition of these solutions is shown in Table 1. Six animals weighing 28.1 ± 1.3 kg underwent sham thoracotomy (Sham group). All animal care and procedures were in compliance with the guidelines published by the National Institutes of Health on the care and use of animals (22) and were approved by the local authorities (Regierung von Oberbayern). The experimental protocol is illustrated schematically in Figure 1.

View larger version (22K): [in this window] [in a new window]   Figure 1.   Experimental protocol. Donor lungs were perfused with modified Euro-Collins solution (EC; n = 6) or low-potassium dextran solution (LPD; n = 6). Blood gas = arterial blood gas; MAP = mean arterial pressure; mPAP = mean pulmonary arterial pressure; PCWP = pulmonary capillary wedge pressure; PVR = pulmonary vascular resistance. BAL = bronchoalveolar lavage; Effluent = pulmonary arterial effluent solutions. Lung-1, BAL-1, and Effluent-1: native lung (baseline). Lung-2, BAL-2, and Effluent-2: postischemic lung (after ischemia). Lung-3 and BAL-3: postischemia-reperfusion lung (after ischemia-reperfusion).

Donor Procedures

Premedication was achieved by intramuscular injection of azaperone (3 mg/kg) and ketamine hydrochloride (25 mg/kg). After anesthesia was induced with 6 mg/kg of thiopental sodium, the animals were intubated and mechanically ventilated with a Servo 900 C Ventilator (Siemens-Elema, Solna, Sweden). The ventilator was set to an FI_O2 of 1.0, a respiratory rate of 12 breaths/min, a tidal volume of approximately 20 ml/kg, and a positive end-expiratory pressure of 5 mm Hg. Anesthesia was maintained with piritramide (0.5 mg/kg/h) and pancuronium bromide (0.3 mg/kg/h) by continuous intravenous infusion. In addition, pentobarbital sodium (4 mg/kg) was intravenously injected when needed. A triple-lumen central venous catheter for infusion was placed in the external jugular vein. An arterial catheter for measuring mean arterial pressure was placed in the internal carotid artery. After sternotomy and heparinization (500 U/kg), the main pulmonary artery (PA) was cannulated with a 24-French catheter connected to the perfusion system. Both venae cavae were ligated and divided, and 250 µg of prostacyclin was injected directly into the main PA. After clamping of the aorta and incision of the left atrial appendage for venting, lung perfusion with either EC or LPD was started and cardiac arrest was achieved by the injection of 20 mEq of KCl into the ascending aorta. With the lungs being continuously ventilated, perfusion was accomplished through the PA catheter at a pressure of 30 to 40 cm H₂O. The temperature of the perfusate was 4° C. The amount of perfusion with either LPD and EC was 100 ml/kg. After the perfusion had been completed, ventilation was stopped when the lungs were inflated with a mean airway pressure of 20 cm H₂O, and the trachea was clamped. The double lung block was excised and stored in Ringer's solution at 4° C for 18 h.

Recipient Procedures

After induction of anesthesia as described for the donor, the recipient animals were intubated with a double-lumen tube and the ventilator was set to an FI_O2 of 1.0, a respiratory rate of 12 breaths/min, a positive end-expiratory pressure of 5 mm Hg, and a tidal volume of approximately 20 ml/kg. A central venous catheter in the external jugular vein and an arterial catheter in the internal carotid artery were placed essentially as described in the donor procedure. A 7F thermodilution Swan-Ganz catheter (Baxter Deutschland GmbH, Unterschleissheim, Germany) was placed in the internal jugular vein and was passed to the left main branch of the PA. Fluid was replaced through an intravenous line with Ringer's solution (25 ml/kg during the operation and 10 ml/kg/h during the observation period) to maintain a constant central venous pressure.

After a left thoracotomy and encircling of the right PA to clamp the blood flow of the contralateral right lung reversibly, a left pneumonectomy was performed. Excised left lungs were used as baseline lungs that are described below as "Lung-1." At that point, the donor lungs were divided from the organ block. The right lung was used as the postischemic lung, described below as "Lung-2." The trimmed left donor lung was implanted in the standard fashion with running suture for the PA and atrium and interrupted suture for the main bronchus. Control measurements of hemodynamic parameters and respiratory functions were performed immediately before reperfusion. During the 6-h reperfusion period, the thoracotomy was approximated. After completion of the 6-h observation period, the animals were killed with 20 mEq of KCl. The transplanted lung was excised and harvested. This lung was used as the post-ischemia-reperfusion lung, described below as "Lung-3."

Sham Group

Left-sided thoracotomy was done in six animals under the same conditions and using the same procedure as in the recipient animals, in terms of premedication and anesthesia, tracheostomy and ventilation, arterial and PA catheterization, and encircling of the right PA to clamp the blood flow of the contralateral right lung reversibly. After these procedures, the thoracotomy was closed, followed by a 6-h observation period. After completion of the 6-h observation period, the animals were killed. The left lung was harvested and used as "Lung-3."

Measurements of Parameters

Physiologic parameters during reperfusion. Mean arterial pressure (MAP), mean pulmonary arterial pressure (mPAP), pulmonary capillary wedge pressure (PCWP), cardiac output (CO), and partial arterial oxygen pressure (Pa_O2) were measured at times 0, 1, 2, 3, 4, 5, and 6 h after reperfusion. At each point, the right PA was clamped and the right lung was excluded from ventilation such that the respiratory rate was adjusted to 20 breaths/min at a tidal volume of approximately 30 ml to allow perfusion and ventilation of the transplanted lung only. In the Sham group, similarly, the right PA was clamped and the setting of the ventilator was adjusted at the measuring points. All hemodynamic measurements were performed with the chest closed. Pressure recordings were made on a Hellige hemodynamic monitor (Hellige GmbH, Freiburg, Germany). CO was determined in triplicate by the single-indicator, thermal-dilution technique using a Baxter REF-1 Ejection Fraction/Cardiac Output Computer (American Edwards Laboratories, Santa Ana, CA). The pulmonary vascular resistance (PVR) was calculated by using the formula: PVR = 79.9 × (mPAP  PCWP)/ CO (dyne · s · cm⁵).

Arterial blood gases were measured using a Model 238 pH/Blood Gas Analyzer (Ciba-Corning, Medfield, MA). Leukocyte counts and hematocrits in blood were determined at times 0, 1, 2, 3, 4, 5, and 6 h after reperfusion on EDTA-anticoagulated blood samples using a Coulter T 540 Counter (Coulter, Krefeld, Germany). At times 0 and 6 h, additional blood samples were drawn for measurement of lipid peroxide.

Lung water measurement. Each harvested lung (Lung-1, Lung-2, and Lung-3) was weighed and a total of four pieces, cut to size of 1 × 1 × 1 cm, were sampled from each lobe and placed in glass tubes. Each lung sample was weighed for the determination of wet weight and then dried in an oven at 70° C for 48 h. The dry tissue weight was determined and the lung wet-to-dry weight (W/D) ratio was calculated to assess pulmonary edema.

BAL and pulmonary arterial effluent solution ("Effluent"). The lung tissue sampling sites were sutured to prevent the loss of BAL fluids (BALF). The BAL procedures were performed on each harvested lung (Lung-1, Lung-2, and Lung-3). The BALF were named BAL-1, BAL-2, and BAL-3, respectively. A syringe with 90 ml of physiologic saline was inserted via the main bronchus and wedged; the recovery amount of BAL was routinely > 70%. The BALF was centrifuged at 400 × g for 10 min. The cell pellet was resuspended in 1 ml of saline containing 10% of albumin, and total cells in the recovered BALF were counted using a Neubauer Chamber. For differential counting of leukocytes in BALF, the smear was prepared in duplicate by centrifuging 100 µl of the BALF at 400 × g for 5 min using a Cytospin centrifuge (Cytospin 2; Shandon Southern Instruments, Pittsburgh, PA). Slides were stained with modified Wright's Giemsa Stain (Diff Quik; American Scientific Products, McGaw Park, IL). The supernatants were stored at 70° C until use.

The pulmonary arterial effluent solutions were obtained from donor lungs before and after the ischemic preservation, referred to as Effluent-1 and Effluent-2, respectively. At the end point of lung-flush perfusion in the donor operation, 10 to 40 ml of Effluent-1 was recovered from the left atrium with sterile syringes. After 18 h of the bilateral lung storage, the right donor lung was flushed via the pulmonary artery by injecting 50 ml of the pulmoplegic solution that was used for the initial flush, and then 15 to 40 ml of Effluent-2 was recovered from the pulmonary vein. The effluent solution was centrifuged at 1,800 rpm for 10 min, and aliquots of the supernatants were stored at 70° C until use.

Determination of lipid peroxides. Lipid peroxides in blood plasma, effluent, and BALF were measured as thiobarbituric acid-reactive materials (TBARM) (11, 12). Each sample was measured in triplicate. TBARM levels in the blood plasma and the effluent were determined using the method of Yagi (23) with minor modifications, allowing exclusion of impurities such as bilirubin, glucose, sialic acid, and so on. Briefly, 0.25 ml of the samples and 2.0 ml of 1:12 N H₂SO₄ were added to a 15-ml screw-capped centrifuge tube. Then, 0.25 ml of 10% phosphotungstic acid was added and the mixture was centrifuged at 3,000 rpm for 10 min. The sediment was suspended in 2.0 ml of distilled water, and 0.5 ml of 0.5% thiobarbituric acid (TBA) reagent was added. The reaction mixture was heated for 60 min at 95° C in a water bath. After cooling, 2.5 ml of n-butanol was added and the mixture was shaken vigorously. After centrifugation at 3,000 rpm for 15 min, and n-butanol layer was fluorometrically measured at 553 nm with 515 nm excitation (FluoroSkan II; Labsystems Bornheim, Hersel, Germany). A standard cure was constructed using solutions of zero to 10 nmol/ml of malondialdehyde (MDA; Sigma Chemical Co., St. Louis, MO), first diluted 1:5 in 0.5 M H₂SO₄ and then diluted to a final concentration in deionized water. The data are expressed in terms of MDA (nmol MDA/ml).

TBARM levels in the BALF were determined using the method of Takeda and colleagues (24). Based on our preliminary experiment, small amounts of TBARM (< 1 nmol/ml) can be detected by this method. In summary, 0.1 of 7% sodium dodecyl sulfate was added to 0.25 ml of BALF. Then, 1 ml of 0.1 N HCl and 0.25 ml of 10% phosphotungstic acid were added. Finally, 0.5 ml of TBA reagent was added and the mixture was heated for 90 min at 95° C. After cooling, 2.5 ml of n-butanol was added and the mixture was shaken vigorously. After centrifugation at 3,000 rpm for 15 min, TBARM concentration in the n-butanol layer was measured fluorometrically in the above- described manner.

Statistical Analysis

All data are presented as mean ± SEM. One-way analysis of variance and the Fisher least-significant difference test were used to analyze differences in lung W/D ratios, neutrophil counts in BALF, and TBARM levels among the three groups. The hemodynamic data (MAP, CO, mPAP, and pulmonary vascular resistance [PVR], Pa_O2/ FI_O2, and leukocyte counts in blood were analyzed by repeated measures of ANOVA (Stat View 4.02; Abacus Concepts, Berkeley, CA). Simple regression test was used to search the correlation between lung W/D ratios or neutrophil counts and TBARM levels in BALF. A cut off value of p < 0.05 was used for detection of statistically significant differences.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Physiologic Parameters during Reperfusion

The MAP, CO, mPAP, and PVR data after reimplantation of the left lung and during reperfusion are shown in Table 2. At 1, 3, and 4 h, MAP data were lower in the EC group than in the Sham group (p < 0.05). Both the EC and the LPD groups showed lower CO as compared with the Sham group at 1 and 3 h and at 1 and 2 h, respectively. In the LPD group, mPAP at 1 and 2 h after reperfusion was reduced as compared with that of the EC group (p < 0.05). At 6 h, the data of mPAP in the EC group were significantly higher than those in the Sham group (p < 0.05); however, those in the LPD group were not significantly different. In the LPD group, PVR at 1 h after reperfusion was lower than that in the EC group (p < 0.05). As shown in Figure 2, at the end of the experiment (6 h), Pa_O2/FI_O2 in the LPD group (436 ± 47) was improved as compared with that of the EC group (255 ± 69; p < 0.05 versus Sham). The EC group at 5 and 6 h and the LPD at 3, 4, 5, and 6 h had lower leukocyte counts than the Sham group (Table 3).

View larger version (17K): [in this window] [in a new window]   Figure 2.   Changes in PaO2/FIO2 after reperfusion. Data are shown as mean ± SEM. Open circles represent the EC group. Closed triangles represent the LPD group. Open triangles represent the Sham group. In the transplanted groups (EC and LPD), significant deterioration of arterial blood gas was observed as compared with the Sham group at 2, 4, and 5 h after reperfusion. In the LPD group, PaO2 was improved at 3 and 6 h after reperfusion (*p < 0.05 versus the Sham group).

Lung W/D Ratio

The lung W/D ratio of the three experimental groups is shown in Figure 3. The W/D ratios at baseline (Lung-1) did not differ significantly between the EC (5.1 ± 0.3) and the LPD (4.9 ± 0.2) groups. Additionally, the W/D ratios after ischemia (Lung-2) did not differ significantly between the two groups (EC: 6.2 ± 0.6, LPD; 5.8 ± 0.2). The W/D ratios of the EC group after ischemia-reperfusion (Lung-3; 7.5 ± 0.9) were increased as compared with that after ischemia (Lung-2; 6.2 ± 0.6), whereas the W/D ratios of the LPD group did not differ significantly between Lung-2 (5.8 ± 0.2) and Lung-3 (6.5 ± 0.2). In Lung-3, the W/D ratios of the EC group were significantly higher than those of the Sham group (6.2 ± 0.3) (p < 0.05), although those of the LPD group did not differ significantly from those of the Sham group.

View larger version (18K): [in this window] [in a new window]   Figure 3.   Lung wet-to-dry weight (W/D) ratio. Data are shown as mean ± SEM. The W/D ratios in the EC group after ischemia-reperfusion (Lung-3) were increased as compared with that after ischemia (Lung-2) (*p < 0.05). There were no significant differences (ns) between the W/D ratios at baseline (Lung-1) and those after ischemia (Lung-2). The W/D ratio of the LPD group did not differ significantly (n.s) between Lung-1 and Lung-2 or between Lung-2 and Lung-3. The W/D ratios after ischemia-reperfusion (Lung-3) were higher in the EC group than in the Sham group (#p < 0.05).

Neutrophil Counts in BALF

As shown in Figure 4, neutrophil counts in BAL-3 were significantly increased in both the EC ([5.8 ± 1.8] × 10⁶/ml) and the LPD ([4.5 ± 1.8] × 10⁶/ml) group as compared with those in the corresponding BAL-1 and BAL-2 samples. There were no statistically significant differences between the EC and the LPD groups in the neutrophil counts in BAL-1, BAL-2, and BAL-3.

View larger version (10K): [in this window] [in a new window]   Figure 4.   Neutrophil counts in BALF. Data are shown as mean ± SEM. BAL-3 neutrophil counts were significantly increased in both the EC and the LPD group as compared with BAL-1 and BAL-2 (p < 0.05).

Lipid Peroxidation

Plasma TBARM concentrations at times 0 and 6 h after reperfusion revealed no significant differences among the EC, LPD, and Sham groups, as shown in Table 4.

The TBARM concentrations in Effluents are shown in Figure 5. In the EC group, the TBARM levels in Effluent-2 (1.7 ± 0.1 nmol/ml) were significantly higher than those in Effluent-1 (1.4 ± 0.2; p < 0.05), whereas, in the LPD group, the difference between Effluent-2 (1.3 ± 0.1) and Effluent-1 (1.1 ± 0.1) TBARM levels was not significant. The TBARM concentrations in Effluent-2 (after 18 h of ischemia) in the LPD group were lower than that in the EC group (p < 0.05).

View larger version (20K): [in this window] [in a new window]   Figure 5.   Thiobarbituric acid-reactive material (TBARM) concentrations in pulmonary arterial effluent solutions (Effluent). Data are shown as mean ± SEM. In the EC group, the TBARM levels in Effluent-2 (after ischemia) were significantly higher than those in Effluent-1 (baseline) (*p < 0.05). In the LPD group, there was no significant difference between Effluent-2 and Effluent-1 TBARM concentrations. TBARM levels in Effluent-2 were lower in the LPD group than in the EC group (#p < 0.05).

The TBARM concentrations in BALF are shown in Figure 6. In the EC group, there were significant differences in TBARM between BAL-1 (0.40 ± 0.03 nmol/ml) and BAL-2 (0.57 ± 0.06) and between BAL-2 and BAL-3 (0.85 ± 0.05) (p < 0.05). In the LPD group, however, there was no significant difference in TBARM between BAL-1 (0.41 ± 0.02) and BAL-2 (0.42 ± 0.02), although the TBARM in BAL-3 (0.61 ± 0.03) was higher than that in BAL-2 (p < 0.05). The BAL-2 TBARM level in the LPD group was lower than that in the EC group (p < 0.05). The BAL-3 TBARM level in the LPD group was also lower than that in the EC group (p < 0.05).

View larger version (13K): [in this window] [in a new window]   Figure 6.   TBARM concentrations in BALF. Data are shown as mean ± SEM. In the EC group, TBARM concentrations after ischemia (BAL-2) were significantly increased as compared with baseline (BAL-1) (*p < 0.05). TBARM levels in BAL after ischemia-reperfusion (BAL-3) were higher than those in Lung-2 (*p < 0.05). In the LPD group, TBARM levels did not differ significantly (n.s) between BAL-1 and BAL-2. BAL-3 TBARM levels were significantly increased as compared with BAL-2 TBARM levels (*p < 0.05). BAL-3 TBARM in both the LPD and the Sham groups were lower than in the EC group (#p < 0.05 versus BAL-3 of the EC group).

As shown in Figure 7, there was a significant correlation between the lung W/D ratio and BALF TBARM concentrations in all three groups (p < 0.05, r = 0.50). TBARM concentrations also correlated with BALF neutrophil counts in all three groups (p < 0.05, r = 0.60).

View larger version (12K): [in this window] [in a new window]   Figure 7.   Correlations of the lung W/D ratios with TBARM concentrations in BALF samples of the EC, LPD, and Sham groups. The correlation coefficient of the linear regression analysis (r = 0.50) was statistically significant at p = 0.0012.

To assess the effect of surgical stress on lung injury parameters, the W/D ratio and BALF neutrophil counts and TBARM concentrations in the Sham group were compared with those of Lung-1 in the EC and the LPD groups (each experimental group: n = 12). The W/D ratio and BALF neutrophil counts in the Sham group were significantly higher than those of Lung-1 and BAL-1 from the experimental groups (p < 0.05). The BALF TBARM concentrations in the Sham group were not significantly different from those of Lung-1 from the experimental groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, lung flush-preservation with EC solution resulted in increased lipid peroxidation of lungs after 18 h of cold ischemia as well as after ischemia followed by reperfusion, which leads to ischemia-reperfusion lung injury, as evidenced by increased mPAP, impaired arterial oxygenation, and increased lung W/D ratios. These effects were attenuated after preservation with LPD solution as compared with preservation with EC solution.

Lung levels of malondialdehyde, one of the main degeneration products of oxidatively peroxidized lipids, can be measured as TBARM (25). It is unclear whether the TBARM concentrations is a specific marker of lipid peroxidation. Other substances, such as bilirubin, glucose, and sialic acid in plasma, can react with TBA (23). However, these substances were essentially eliminated by the method used in this study (23, 24). Indeed, TBARM has been used as a quantitative measure of oxidant injury with ischemia-reperfusion or sepsis (17, 26).

There is some evidence that hypothermic or normothermic lung storage may cause an increase in TBARM (13). In our study, TBARM levels after ischemic storage, presented as BAL-2 and Effluent-2, were increased in the EC group as compared with baseline (Figures 5 and 6). It is believed that a reintroduction of oxygen to the cells play a major role in the production of oxidative injury during ischemia-reperfusion (27). During an ischemic period, ATP in tissue cells is catabolized to yield hypoxanthine. The hypoxic stress also triggers the conversion of NAD-reducing xanthine dehydrogenase to the oxygen radical-producing xanthine oxidase. During reperfusion, molecular oxygen is reintroduced into the tissue where it reacts with hypoxanthine and xanthine oxidase to produce a burst of superoxide anion and hydrogen peroxide (27).

The lung is the only organ to which oxygen can be supplied after its blood supply is interrupted. Oxygen that exists in alveoli may contribute to the reintroduction of oxygen to cells and the acceleration of oxidant formation (14, 28). The increase in TBARM after ischemia in the present study might be explained by such mechanisms. The levels of TBARM in BALF may reflect ischemic damage to alveolar wall cells or macrophages, resulting in oxidant injury, and the TBARM in Effluent, similarly, may reflect oxidant injury of the pulmonary vascular wall. However, Christie and associates (17) showed that hypothermic storage alone caused no significant effects on lipid perioxidation in lung tissue in a rat model of lung storage and reperfusion. These conflicting results may be due to differences in animal species, storage duration, or sites where we sampled BALF or Effluent.

BALF TBARM levels after ischemia-reperfusion (BAL-3) were elevated in the EC group, which showed more severe ischemia-reperfusion lung injury than the LPD group (Figures 3 and 6). In the EC group, the increase in lung water may have partially been due to transient systemic hypotension, which can accelerate pulmonary vascular permeability through the mechanism by which shock-induced adult respiratory distress syndrome is produced (29). However, the increase in TBARM in BAL suggests TBARM generation in lung tissue because plasma TBARM levels were not elevated during reperfusion (26).

Neutrophil counts in blood were lower in the EC and the LPD groups than in the Sham group (Table 3). Ischemic storage and reperfusion may activate neutrophils and pulmonary vascular endothelial cells and accelerate the expression of adhesion molecules on these cell types (30, 31). Since circulating neutrophils might have attached to pulmonary vessels to some extent or have marginated into the lung, neutrophil counts in blood may have been reduced during the reperfusion period.

Neutrophil counts in BALF were markedly increased after ischemia-reperfusion, in both the EC and the LPD groups (Figure 4). After initiating the reperfusion, a large circulating and marginated pool of neutrophils can be stimulated by various kinds of chemotatic or physical stimuli and these neutrophils may immigrate into the alveolar air space (30, 31). Based on the data obtained from postischemia samples (BAL-2 and Effluent-2), we may conclude that reactive oxygen metabolites, which are produced by ischemic preservation, would also contribute to the chemotatic stimulation. Activated neutrophils may release toxic products, including reactive oxygen metabolites, and thereby promote tissue injury (32). Rising neutrophil levels in BAL may be paralleled by increased TBARM levels and lung water. Indeed, we also found that lung W/D ratios correlated significantly with BALF neutrophil counts and BALF TBARM levels (Figure 7).

However, neutrophil counts in BALF did not differ significantly between the EC and LPD groups, although there were significant differences in pulmonary physiologic parameters of lung injury (mPAP, Pa_O2/FI_O2, and lung water) and TBARM levels (Table 2, Figures 3-6). These results may suggest that TBARM levels in BALF or Effluent are more sensitive than BALF neutrophil counts as indicators of ischemia-reperfusion injury after lung transplantation.

Since in most other studies lipid peroxides were measured in samples from isolated-perfused lungs or homogenated lung tissue samples (13), it may be difficult to compare these data directly with in vivo situations after lung transplantation. In our model, lipid peroxidation in BALF and Effluent could be detected during the course of lung transplantation and thus increases in lipid peroxidation in BALF or the Effluent of graft lungs may become a useful predictor of ischemia-reperfusion injury in the clinical setting of lung transplantation.

Moreover, in this study, we have shown that flush-perfusion and preservation of lungs with LPD can reduce TBARM levels and ischemia-reperfusion injury. Some investigators have indicated that an extracellular type solution (LPD provides significantly better immediate function of the preserved lung than an intracellular type EC solution, although many studies in other organs (kidney, liver, and so forth) have shown the theoretical advantage of an intracellular type solution over an extracellular solution (3, 18, 33). The mechanisms responsible for improved storage with LPD are uncertain, and there are a number of possible explanations. For example, low concentrations of potassium in LPD may prevent the induction of pulmonary vasoconstriction during lung flush-perfusion, leading to an increase in pulmonary vascular resistance and impaired distribution of the perfusion solutions (34, 35). Dextran, a component of LPD, can act as an oncotic agent and may reduce pulmonary edema and improve reperfusion lung function (18).

Our results suggest that LPD may prevent lipid peroxidation due to reactive oxygen metabolites released from activated neutrophils as compared with EC. LPD may have some favorable effect, in terms of preventing the neutrophil responses during ischemia-reperfusion, because of its fluid composition of electrocytes, osmotic pressure, or oncotic pressure, which may stabilize cell membranes or the cytoplasm. EC may have an effect as an endothelial cell depolarizing agent resulting in an increase in oxidant generation because of its high potassium content, as high extracellular potassium concentrations induce cell membrane depolarization and lead to the formation of hydrogen peroxide and other oxidants with depolarization (36).

Spaggiari and coworkers (37) reported that extracellular type solutions provided better cell preservation of human lung fibroblasts in vitro. Maccherini and colleagues (38) showed that LPD is less cytotoxic for isolated type II pneumocytes than EC and that LPD allows higher levels of metabolic activity in recovering epithelial cells. It has been suggested that dextran has a scavenging effect on toxic oxygen metabolites (39), but Keshavjee and associates (40) showed that the improved lung preservation with dextran 40 is probably not mediated by a superoxide radical-scavenging process involving dextran. Some studies showed that dextran can improve microcirculatory flow under various conditions and that this action is mediated by its surface coating of red blood cells, platelets, and endothelial cells (41). It is possible that dextran exerts a protective effect against the generation of toxic oxygen metabolites via a similar mechanism. Further investigation is needed to elucidate the cellular mechanism beneath the effect of LPD.

In the Sham group, the hemodynamic data, Pa_O2/FI_O2, and leukocyte counts in blood were essentially unchanged during the reperfusion period. These observations indicate that intermittent right pulmonary artery occlusion had little effect on changes in these parameters. The W/D ratio and BALF neutrophil counts in the SHAM group were significantly higher than those of Lung-1 or BAL-1 from the EC and LPD groups, although TBARM concentrations in BALF in the Sham group were not significantly different from those in the EC and LPD groups. These results suggest that the effect of surgical stress on lung injury parameters cannot be ignored in this model. However, these three parameters in the LPD group were significantly different from those in the EC group. This indicates that we can compare the effects of LPD with those of EC in the animal model used in this study.

In conclusion, our data suggest that lipid peroxidation measured in BALF and Effluent can reflect the ischemia and reperfusion-induced lung injury in a pig model of orthotopic left-sided single lung transplantation. Furthermore, lung flush-perfusion and preservation with LPD can reduce lipid peroxidation and the lung injury that appears after flush-perfusion and ischemic preservation with EC.