INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

None of the clinically applied lung preservation techniques permit reliable preservation of human lung allografts for longer than 6 h (1, 2). This is likely due to the susceptibility of lung tissue to ischemia-reperfusion (I/R) injury (3, 4). Reduction in I/R injury would greatly improve the early lung function after transplantation and allow more lungs to become available for transplantation.

The most common approach to improve lung preservation is hypothermia. Hypothermia slows down cellular metabolism, enzyme activity, and energy consumption, all of which could improve preservation of a donor organ. Hypothermia by flushing cooling preservation solution in to the donor organ can provide some advantages, but it may also be harmful. For example, flushing of the graft organ with solutions at low temperature may facilitate the rate of cooling, but it may also cause vasoconstriction, with resulting incomplete perfusion of some areas of the lung as well as cellular edema (5). Despite these problems, hypothermia has a protective effect on I/R injury, and this approach has been well explored. PGE₁ can be used as an additive to promote the protection of I/R injury by University of Wisconsin solution (UW) (6, 7). Because PGE₁ has an effect of vasodilatation (1, 7), which might inhibit the vasospasm of cooling, we postulated that PGE₁ and hypothermia might have a synergistic effect and produce a further protection of I/R injury. In order to test this hypothesis, we designed an I/R injury study using perfused rat lungs with UW at different temperatures (30° C, 22° C, or 10° C), respectively, to evaluate whether hypothermia attenuates the I/R lung injury. Furthermore, PGE₁ was added in the cooling UW solution to test whether the combination of hypothermia and PGE₁ produced a further reduction of I/R lung injury.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Preparation of Isolated and Perfused Rat Lungs

Our isolated-perfused lung in situ I/R model has been previously described (7). Briefly, male Sprague-Dawley rats (weighing 250 to 350 g) were anesthetized intraperitoneally with sodium pentobarbital (20 to 25 mg). A tracheotomy was performed to permit ventilation with a Harvard rodent ventilator (Model 683; Harvard Apparatus, South Natick, MA) at 55 breaths/min, at a tidal volume at 2.5 ml, and a positive end-expiratory pressure of 2 cm H₂O. The inspired gas mixture contained 5% CO₂ and 95% air. After the median sternotomy was performed, heparin (1 U/g) was injected into the right ventricle. Blood was drawn from the right ventricle and discarded. A cannula was placed into the pulmonary artery through a puncture into the right ventricle, and a tight ligature was placed around the main trunk of the pulmonary artery. A large catheter was inserted into the left atrium through the left ventricle and mitral valve, fixed by ligature at the apex of the heart, and used to divert the pulmonary venous outflow into a reservoir. A third ligature was placed above the atrioventricular junction to prevent perfusate flowing back into the ventricles. The lungs were perfused with the chosen perfusate using a peristaltic pump (Minipulse 2; Gilson Medical Electronics, Middleton, WI) at a constant flow of 0.03 ml/min/g body weight. An initial 75 ml of lactate ringer solution perfusate, which contained residual blood cells and plasma, were discarded and not recirculated. An additional 25 ml of the chosen perfusate at different temperature was recirculated in the lung. Pulmonary artery (Ppa) and pulmonary venous (Ppv) pressures were continuously monitored with pressure transducers (P23 1D; Statham, Oxmard, CA) from a sidearm of the inflow and outflow cannulas and continuously recorded on a polygraph recorder (Gould Instruments, Cleveland, OH). The Ppv was set at 2.5 mm Hg by adjusting the height of the venous outflow reservoir and zone III flow conditions (arterial > venous > alveolar pressures) were maintained in all experiments.

The isolated perfused lung remained in situ, and the weight of the whole rat was monitored on an electronic balance and recorded on an oscillograph after digital-to-analog conversion. Any change in the preparation weight (body weight) was considered a result of changes in lung weight (9). Three criteria were used to continue the isolated lung preparation experiment: (1) no leakage was observed at the sites of cannula insertion, (2) no evidence of edema was present, and (3) the lung attained an isogravimetric state, i.e., the lung was neither gaining nor losing weight.

Perfusates

UW and modified UW solutions were used as perfusates. UW (DuPont-Merk Pharmaceuticals, Wilmington, DE) is composed of 50 g/L pentastarch, 35.83 g/L lactobionic acid, 3.4 g/L potassium phosphate monobasic, 1.23 g/L magnesium sulfate heptahydrate, 17.83 g/L raffinose pentahydrate, 1.34 g/L adenosine, 0.136 g/L allopurinol, and 0.922 g/L glutathione. Perfusate osmolarity was 320 mosmol, sodium concentration was 29 mEq/L, potassium concentration was 125 mEq/L at pH 7.4 (Table 1). The modified UW solution (UW+PGE₁) was prepared by the addition of the protective agent (20 µg/L PGE₁).

Determination of Pulmonary Capillary Pressure

The pulmonary capillary pressure (Ppc) was estimated using the double-occlusion method (10). Arterial inflow and venous outflow lines were simultaneously occluded and the equilibrium Ppa and Pp were measured. This equilibration pressure is the same as the isogravimetric measure of Ppc and also reflects the prevailing capillary pressure when the lungs are damaged.

Calculation of Pulmonary Vascular Resistance

The pulmonary arterial (Ra) and venous (R) resistances were calculated using the following equations: Ra = (Ppa  Ppc)/, and R = (Ppc  Pp)/ respectively, where is perfusate flow and Ppa and Pp are the arterial and venous pressures.

Measurement of Microvascular Permeability

The pulmonary capillary filtration coefficient (K_fc) was used as an index of the microvascular permeability to solvent. The K_fc was measured using a method described previously and used as an index of permeability in many published studies (11). Briefly, after an isogravimetric state was attained in the lung, Pp was rapidly elevated to 6 to 8 cm H₂O for 15 min. The increase in lung weight was recorded. The recording shows a characteristic rapid weight gain (vascular filling), which is followed by a slower rate of weight gain. The rate of weight change (W/t) occurring in the 6- to 14-min interval was analyzed using linear regression of the log-10-transformed rates of weight changes calculated at each minute. The initial rate of weight gain was then determined by extrapolation of (W/t) to zero time. K_fc was then calculated by dividing (W/t) at time 0 by the change in Ppc that was imposed after venous outflow pressure was increased. The K_fc value was normalized using the baseline wet lung weight and expressed as ml/min/cm H₂O/100 g lung tissue.

Experiment Protocols

In order to study the temperature effect on I/R injury, perfusates at different temperature (10° C, 22° C, or 30° C) were used. The animals were divided into six groups: (1) 30° C UW (n = 5), (2) 22° C UW (n = 5), (3) 10° C UW (n = 4), (4) 30° C UW+PGE₁ (n = 4), (5) 22° C UW+PGE₁ (n = 4), (6) 10° C UW+PGE₁ (n = 4). The isolated lungs were perfused with one of the above-designated perfusates. The closed system of circulation was maintained at constant flow, volume, and temperature. The experiment began after a hemodynamic stability period of 15 min. The protocol used to produce I/R lung injury was as follows: the isolated lung was neither ventilated nor perfused for 1 h. This ischemia was then followed by the reinstitution of ventilation and perfusion (reperfusion) for 60 min at different temperatures (10° C, 22° C, and 30° C), respectively.

Lung Histopathology

After termination of each experiment, the whole lungs were dissected and immediately fixed in 10% neutral buffered Formalin. After fixation, the right middle lobes were dehydrated through a graded series of alcohol, cleared in xylene, and embedded in paraffin. All sections were cut at 5 µm and stained with hematoxylin-eosin.

Statistical Analysis

Values are expressed as mean ± SD. Comparisons among all groups for a given variable were done using a one-way analysis of variance and Dunnett's method. Comparison between baseline and post-reperfusion, within each group for given variables were made by using Student's paired t test, and a p < 0.05 was considered a statistically significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lung Weight Gains

All groups showed a reduction in postreperfusion lung weight gain at lower temperatures (22° C or 10° C) as compared with the UW perfusate group at high temperature (30° C) (Table 2 and Figure 1). However, the LWG was significantly greater in the animals receiving UW at 10° C (Group 3) than in those receiving UW at 22° C (Group 2). In contrast, in animals perfused with modified UW solutions, the LWG at 10° C UW+ PGE₁ (Group 6) was less than that seen in the animals perfused with 22° C UW+PGE₁ (Group 5). At 30° C or 10 ° C, modified UW (UW+PGE₁) produced less LWG than pure UW alone. Among these six groups, 10° C UW+PGE₁ produced the greatest reduction of LWG.

View larger version (24K): [in this window] [in a new window]   Figure 1.   Changes in lung weight gain (LWG) during reperfusion. LWG was measured during a 60-min period of ischemia and a 60 min period of reperfusion. In the 30° C UW group, LWG was markedly increased during reperfusion. In animals perfused with pure UW at hypothermia (22° or 10° C) there was less LWG than at 30° C; however, UW at 22° C was superior to UW at 10° C in preventing edema formation. Perfusate with modified UW (UW+ PGE1) produced less LWG than did pure UW alone. LWG in animals perfused with 10° C UW+PGE1 was less than that of animals perfused at 22° C UW+PGE1. Modified UW at 10° C produced the most reduction of LWG.

Capillary Filtration Coefficient

In comparison with the animals receiving 30° C UW solution (Group 1), animals perfused with UW solution at low temperatures (22° C or 10° C; Groups 2 and 3 ) had a significant reduction in microvascular permeability at 60 min after reperfusion, as shown by a smaller K_fc (Table 3). As was seen with lung weights, the K_fc of animals receiving UW at 22° C was less than the value seen in the animal receiving UW solution at 10° C, indicating that too low a temperature was not as effective in preventing microvascular leakage. Again, as was seen with lung weights for the modified UW groups (UW+PGE₁), the K_fc was significantly lower in animals perfused with modified UW at 30° C or 10° C than in those receiving UW alone, indicating PGE₁ showed protection from I/R injury.

Hemodynamics

Before I/R, a significant increase of Ppa, Pc, Ra, and R was observed in Group 3 (10° C UW) as compared with Group 1 (30° C UW). PGE₁ added to the UW perfusate in Group 5 (22° C UW+PGE₁) and Group 6 (10° C UW+PGE₁) caused no reduction of Ppa, Pc, Ra, and R as compared with those in Group 2 (22° C UW) and Group 3 (10° C UW), respectively.

Histologic Findings

In Group 1 (30° C UW), marked perivascular edema, focal intra-alveolar hemorrhage, interstitial infiltrate, proteinous exudate, and intra-alveolar debris were identified (Figure 2A). In Group 5 (10° C UW+PGE₁), lung tissue appeared normal except for slight perivascular edema (Figure 2B).

View larger version (63K): [in this window] [in a new window]   Figure 2.   Histologic findings. Left lungs of rats in the 30° C UW group showed (A) marked perivascular edema, focal intra-alveolar protein exudates, and focal interstitial and intra-alveolar leukocytic infiltration (original magnification: ×200). Left lungs of rats in the 10° C UW+PGE1 group (B) had a normal appearance except for mild perivascular edema (original magnification: ×200).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Currently, organ preservation is primarily based on reversible inhibition of the isolated organ's metabolism by hypothermia. There are two ways to achieve hypothermia: surface cooling and/or flush-perfusion cooling. Surface cooling is an inefficient method for reducing the core temperature of large organs, especially for inflated lungs with the insulating effect of air. This inefficient cooling results in a longer period of warm ischemia (16). Flush preservation of the lungs may provide several advantages, including: (1) rapid cooling and shortening of the period of normothermic ischemia; (2) washout of pulmonary vascular blood, thus possibly preventing the production of toxic substances from the ischemic blood components; (3) minimizing the adverse effect of cooling on cell physiology by the use of an appropriate flush solution (5).

We found that perfusion of lungs with UW solution at 22° C or 10° C produced more attenuation of the I/R lung injury than perfusion with UW solution at 30° C. These data confirm that hypothermia provides some protection against I/R injury. Previous studies have shown that the optimal surface cooling temperature for lung preservation is in the vicinity of 10° C (12). However, the severity of I/R injury when UW was perfused at 22° C was less than that of UW perfused at 10° C, indicating that lowering the temperature of the perfusate to too low a level did not further protect the lung endothelium from the I/R injury. These results are similar to those of Wang and colleagues (5) who showed that flushing of lungs with preservation solution of 23° C resulted in superior postischemic lung function compared with flushing at 10° C. Our data suggest that the optimal surface-cooling temperature may not be the same as the optimal flush-cooling temperature. Both our results and those of Wang and colleagues (5) indicate that the optimal flush-cooling temperature is around 22° C.

The mechanism responsible for the temperature effect on I/R injury is unclear. Most likely, the attenuation on I/R injury by hypothermia occurred by slowing down cellular metabolism, enzyme activity, and energy consumption. In contrast, a lower temperature of 10° C in our studies produced a significant increase in vascular pressure (Ppa, Pc) and resistance (Ra, R) when compared with 30° C. The filtration coefficient (K_fc) was also increased at 10° C compared with that at 22° C UW. These data suggest that flushing of the lungs at very low temperatures causes vasoconstriction and increases in permeability of the pulmonary endothelial barrier. The reasons that a greater I/R injury occurred at very low temperatures (10° C) are presently unclear; however, previous studies have suggested that hypothermia produces detrimental effects on a number of cellular mechanisms such as membrane permeability, glucose metabolism, calcium sequestration, and cellular osmotic homeostasis (13). Several in vitro studies with cultured endothelial cells (EC) have confirmed loss of normal cell architecture and the creation of intercellular gaps after hypothermic exposures (14). In vivo studies showed that most enzyme activity and cellular metabolism are depressed to very low levels at very low temperatures. The lack of energy resources are postulated to result in EC losing their ability to regulate cellular volume. Small losses of cellular fluid, added to cytoskeletal reorganization, could lead to cellular shrinkage/retraction (15).

PGE₁ as an additive in UW solution caused a reduction of I/R injury as compared with pure UW solution alone. These data were similar to those from our previous study (6), which showed attenuation of I/R injury after addition of PGE₁ at higher temperatures. Our new results show that perfusion at lower temperatures induce higher pulmonary artery pressure and vascular resistance suggestive of the occurrence of vasospasm. Although PGE₁ has vasodilator effect, in our studies, PGE₁ as an additive to UW solution did not counteract the vasospasm induced by hypothermia. Our results showed that the lung perfused with pure UW at 22° C was superior to that at 10° C for preventing I/R injury. However, the lungs perfused with modified UW (UW+PGE₁) at 10° C produces more protection on I/R than does unmodified UW at 22° C. This finding indicates that PGE₁ is protective against I/R damage and minimizes the thermal injury. In addition to producing vasodilation and bronchodilation (17), previous studies suggested that the protective effects of PGE₁ was reduced platelet and leukocyte aggregation, a suppression of tumor necrosis factor (6, 18), immunosuppressive effects (18), and other "cytoprotective" effects (21). Using an orthotopic rat lung transplant model, Naka and colleagues showed that treatments inducing only vasodilatation are insufficient to enhance lung preservation (28). This suggests that PGE₁ promotes lung preservation by stimulating the cAMP-dependent protein kinase and promoting non-vasodilatory mechanisms of pulmonary protection (26, 27). Which effect of PGE₁ is responsible for protecting from I/R or cooling injury is still unknown and will require further investigation.

In conclusion, flushing lungs with tissue preservation solution (UW) at an optimal cooling temperature attenuates I/R injury, but too low a temperature results in adverse effects. The combination of hypothermia and PGE₁ produced a synergistic attenuation of I/R lung endothelial injury. We propose that pulmonary artery flushing of lungs with low-temperature UW+PGE₁ will provide better lung preservation for lung transplantation.