R of tolerable lung ischemia may be prolonged if the

Transcription

1 ORIGINAL ARTICLES Effect of Inflation on Adenosine Triphosphate Catabolism and Lactate Production During Normothermic Lung Ischemia Paul R. J. De Leyn, MD, Tony E. Lerut, MD, Hans H. J. Schreinemakers, MD, Dirk E. M. Van Raemdonck, MD, Kanigula Mubagwa, MD, and Willem Flameng, MD Division of Thoracic Surgery and Center of Experimental Surgery and Anesthesiology, Katholieke Universiteit Leuven, Leuven, Belgium Although few biochemical data comparing adenosine triphosphate (ATP) catabolism or lactate production in isolated deflated versus inflated lung tissue are available, most transplant centers preserve their donor lungs inflated. We measured ATP level (using high-performance liquid chromatography), energy charge, and lactate level during 2 hours of normothermic ischemia in deflated lung tissue (n = 6), in lung tissue inflated with room air (n = 6), and in lung tissue inflated with 1% oxygen (n = 6). To determine the onset of anaerobic metabolism in lung tissue inflated with 1%,, ATP and lactate levels were measured in another group (n = 6) during 8 hours of normothermic ischemia. Rabbit lungs were flushed in situ with a modified Krebs-Henseleit solution (6 ml/kg). They were isolated and immersed in.9% NaCl at 37 C. In deflated lung tissue, ATP level (control value, 9.4 f.58 pmoyg dry wt) decreased and lactate level (control value, 5.6 f 1.16 pmol/g dry wt) increased after 15 minutes of ischemia (ATP, pmol/g dry wt; lactate, 13.3 f 1.58 pmoyg dry wt). When the lung was stored inflated with room air, ATP breakdown and increase of lactate concentration only occurred after 9 minutes of normothermic ischemia (at 6 minutes: ATP, 8. f.58 pmol/g dry wt; lactate, 6.3 f 1.1 pmoyg dry wt). In lungs stored inflated with 1%,, ATP breakdown and lactate accumulation only occurred after 5 hours of normothermic ischemia (at 4 hours: ATP, pmoyg dry wt; lactate, 5.9 f 1.28 pmol/g dry wt). These data suggest that the breakdown of ATP and the onset of anaerobic metabolism in lung tissue is related to the alveolar oxygen concentration during storage, and that in the isolated inflated lung, aerobic metabolism continues for 4 hours even when perfusion has stopped. (Ann Thoruc Surg 1993;55:173-9) esearch in the early 197s indicated that the duration R of tolerable lung ischemia may be prolonged if the lung remained inflated [l] or ventilated [2] during the ischemic interval. Based on these empirical findings, most lung transplant centers preserve their donor lungs inflated [3-61. However, contradictory results concerning the recovery of pulmonary function have been reported with isolated lungs preserved either ventilated or inflated with 1% oxygen [7, 81. It is assumed that the beneficial effect of inflation during storage is a result of ongoing aerobic metabolism. However, few biochemical data comparing adenosine triphosphate (ATP) catabolism or lactate production in isolated deflated versus inflated lung tissue are available in the literature. It was the aim of this study to compare ATP catabolism, energy charge, and lactate production in lungs stored deflated, in lungs stored inflated with room air, and in lungs stored inflated with 1% oxygen. Presented at the Twenty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 25-27, Address reprint requests to Dr De Leyn, Center of Experimental Surgery and Anesthesiology, Minderbroederstraat 17, Provisorium 1, 8-3 Leuven, Belgium. Material and Methods Experimental Groups Adenosine triphosphate catabolism and lactate production were studied over a period of 2 hours of normothermic (37 C) ischemia in three groups of lungs. In group 1 (n = 6), the lungs were stored deflated, in group 2 (n = 6) the lungs were stored inflated with room air, and in group 3 they were stored inflated with 1% 2. After it was observed that ATP catabolism did not occur within 2 hours in lungs preserved inflated with 1% oxygen, a fourth group (n = 6) was added in which the isolated lungs were stored inflated with 1% oxygen, and ATP catabolism and lactate were measured during 8 hours of normothermic ischemia. Lung Retrieval and Biopsies All animals received humane care in compliance with the Principles of Laboratory Animal Care formulated by the US National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the US National Academy of Sciences and published by the US National Institutes of Health (NIH publication no , revised 1985). New Zealand white rabbits weighing 2.5 to 3 kg were premedicated with.3 ml/kg of Hypnorm (1 mg fluan- Q 1993 by The Society of Thoracic Surgeons /93/$6.

2 174 DE LEYN ET AL 1%3;55 17S9 isone +.2 mg fentanyl per milliliter; Janssen Pharmaceuticals, Bearse, Belgium) intramuscularly and anesthetized with 25 mg/kg sodium pentobarbital (Nembutal; Abbott Laboratories, North Chicago, IL) intravenously. The animals were intubated via a cervical tracheostomy, and the lungs were ventilated (small animal ventilator, model 683; Harvard Apparatus, Inc, South Natick, MA) with room air (respiratory rate = 28 breathdmin; tidal volume = 1 mlkg body weight; positive end-expiratory pressure = 2 cm H2). Rabbits of group 3 and group 4 were ventilated with 1% oxygen. Heparin (7 IU/kg) was administered via an ear vein. The chest was opened through a median stemotomy. Thymic tissue was excised while keeping both pleurae intact. Both superior venae cavae, the inferior vena cava, the ascending aorta, and the pulmonary artery were dissected free and loosely encircled by individual ligatures. The main pulmonary artery was cannulated through the right ventricular outflow tract using a 1-gauge cannula (Insyte Viallon; Becton Dickinson) inserted via the ventricular wall. The superior and inferior venae cavae were ligated and divided individually. The pulmonary artery was ligated around the cannula, the latter being kept in place for subsequent pulmonary flushing. The tip of the left atrial appendage was transected to allow free drainage of the flush solution. The mean time from the onset of inflow occlusion to the initiation of lung flush was 9 seconds and did not differ between the experimental groups. The lungs were flushed to remove the blood elements. Both lungs were flushed via the pulmonary artery with 6 mug cold (4 C) modified Krebs-Henseleit solution (composition in millimoles per liter: NaCI, 13; KCl, 5.6; CaCI,, 2.9; MgCl,,.6; NaH,PO,, 1.2; NaHCO,, 25; sucrose, 26; and D-glucose, 11). Flushing was by gravity at 6 cm H,O. During the flush the lungs were continuously ventilated and topically cooled with ice-cold water. The mean flushing time was 2 minutes in all groups. Immediately after completion of the lung flush, a biopsy sample of peripheral lung tissue was taken for the determination of ATP and lactate levels. This biopsy sample was considered to be normoxic because the lungs had been continuously ventilated until the end of flush. For lungs stored deflated, the tracheal cannula was disconnected after this first biopsy was performed. For lungs stored inflated, the trachea was clamped at endtidal volume and, before biopsy was performed, the lung tissue at the site of prelevation was ligated with a broad umbilical cord to prevent deflation. The heart-lung block was rapidly extracted, care being taken not to damage the lung tissue. The whole procedure was camed out under clean but not sterile conditions. Storage Lungs were stored immersed in.9% NaCl at 37 C. They were covered with wet gauzes to ensure that inflated lungs were completely immersed. In preliminary experiments, a temperature probe was inserted into the lung tissue. After excision, temperature of lung tissue averaged 1 C. After immersion of the lungs in saline solution, a tissue temperature of 37 C was obtained within 5 minutes. In group 1, the trachea was left open to the air, allowing total collapse of the lungs. In group 2 the lungs were stored inflated at end-tidal volume with room air. In groups 3 and 4, they were stored inflated at end-tidal volume with 1% oxygen. In all groups, biopsy specimens of normoxic lung tissue were taken. In groups 1, 2, and 3 samples of peripheral lung tissue were taken at 15, 3, 6, 9, and 12 minutes of ischemia. To determine the onset of ATP catabolism and anaerobic metabolism in lung tissue stored with 1% oxygen, 6 other rabbit lungs (group 4) were stored inflated with 1% 2. Biopsy specimens were taken immediately after lung flush (normoxic) and after 4, 5, 6, 7, and 8 hours of normothermic ischemia. Measurement of Adenosine Triphosphate and Lactate Levels The adenine nucleotides (ATP, adenosine diphosphate [ADP], and adenosine monophosphate [AMP]) and their catabolites (inosine monophosphate, adenosine, inosine, and hypoxanthine) were determined by high-performance liquid chromatography (HPLC) as described by Wynants and Van Belle [9] using a Varian HPLC model 55 equipped with an ultraviolet detector (Varian model UV nm; Varian Associates, Sunnyvale, CA). All chromatograms were monitored on a microcomputer data system (DS 61; Varian). The separation was achieved within 35 minutes on glass columns with C-18 reversedphase particles of 5 pm and ammonium dihydrogen phosphate (.15 mom; ph 6.) and a slow linear gradient of methanollacetonitrile (to 15%) as eluting solvent. Lactate in lung tissue was measured in the same biopsy specimens using an L-lactate analyzer (Model 23; YSI, Inc, Yellow Springs,. OH). The biopsy specimens were immediately cooled in liquid nitrogen, lyophilized overnight, and further stored at -8 C until analysis. The dry tissue was then weighed and homogenized with an Ultra-Turrax for 2 x 1 seconds in ice-cold.6 N HC1, (1 mu1.5 mg dry wt). After standing in ice for a few minutes, the material was centrifuged (Eppendorf centrifuge; 1 minute) and to 3 pl of the supernatant, kept in ice water, 2 pl of ice-cold 1 N KHCO, was added dropwise for neutralization. After centrifugation, 2 pl of the supernatant was injected into the HPLC column for ATP determination and 25 pl was injected into the L-lactate analyzer. Tissue levels of ATP and lactate are expressed in micromoles per gram of dry weight. Statistics All values are expressed as mean f standard deviation. In the different groups, values at ischemic times were compared with the normoxic value using one-way analysis of variance with repeated measurements followed by Scheffe's multiple comparison test. For comparison between different experimental groups, factorial one-way analysis of variance was used. Statistical significance was evaluated with Scheff6's test.

3 ~... p 1993;55: DE LEYN ET AL 175 Table 1. Preischemic Levels of Adenine Nucleotides, Energy Charge, and Lactate in Lung Tissues of the Different Experimental Groups Group Energy Charge Lactate (Folk dry wt) Stored deflated (group 1) 9.42 t t.19.43? ? 1.16 Stored inflated with room air 8.83? t ? ? 1.23 (group 2) Stored inflated with 1% O C ? ? ? 1.3 (group 3) Stored inflated with 1% O t ?.1.5?.1.89? (group 4) * Values are means 2 standard deviation from six experiments. No significant difference between groups was found for any parameter. ADP = adenosine diphosphate; + ADP + AMP). Results AMP = adenosine monophosphate; ATP = adenosine triphosphate; energy charge = (ATP +.5. ADP)/(ATP The mean preischemic level of ATP was 9.42 f.58 pmoyg dry wt; the preischemic level of lactate was 5.63 * 1.16 pmoyg dry wt. The mean energy charge [lo] was.9 t.1. As shown in Table 1, no difference in preischemic levels of ATP, ADP, AMP, lactate, and energy charge was found between the different experimental groups. Figure 1 shows that there was no significant difference in the distribution of bases and nucleosides in normoxic biopsy p 6- )--- Inflated Room Air %. T i... * specimens of the different experimental groups Figure 2 shows the time course of changes in tissue ATP content, lactate content, and energy charge with isch- emia. As we have previously shown (De Leyn P, Lerut T, - I... /*.*---* * Schreinemakers J, et al; unpublished results), ATP level in deflated lung tissue significantly decreased to 5.17 f.86 pmol/g dry wt (p <.1) after 15 minutes of normothermic ischemia and further to 3.42 f.24 pmol/g dry wt at 3 minutes of ischemia. At this time, a significant increase in AMP and IMP was found (De Leyn P, Lerut T, Schreinemakers J, et al; unpublished results). After 2 hours of normothermic ischemia, only 1.82 f.97 pmol/g dry wt ATP was left in the deflated lung tissue. The energy charge of deflated lung tissue decreased from.9 *.1 to.6 f.6 at 15 minutes (v <.1) and further to ( p <.1) at 2 hours i 7 1) *,/* I Deflated Inflated Room Alr Inflated 1% 2 (Group 1) (Group 2) (Group 3 + Group 4) Fig 1. Distribution of nucleosides and bases in normoxic lung tissue stored deflated (group I), stored inflated with room air (group 2). or stored inflated with 1%, (group 3 + 4). No statistical diference betuwen groups was found for any parameter. Bars are standard dmiation Fig 2. Adenosine triphosphate (ATP), lactate, and energy charge levels during 2 hours of norrnothermic ischemia (37 C) in lung tissue stored deflated (group I, open circles), stored inflated with room air (group 2, filled squares), and stored inflated 7cdh 1% Oz (group 3, filled circles). Values (? standard dmiation) at different ischemic times of group 2 and group 3 zuere compared with znlues of group 1 by anal!/sis of rwiance and factorial analysis: * p <.5, ** p <.1, *** p <.1 (group 2 uersus group + I); < ++.5, p < +++.2, p <.1 (group.? verstis group I).

4 176 DE LEYN ET AL 1993;55:17>9 L T 3 Deflated Inflated Room Air Inflated 1% 2 (Group 1) (Group 2) (Group 3) Fig 3. Distribution of niicleosides and bases after 2 hours of normothermic ischemia (37 C) in lung tissue stored deflated (group I), stored inflated with room air (group 2), or stored inflated with 1% Oz (group 3). Bars are standard deviation. Significantly less hypoxanthine (p <.1) was foiind in lung tissue stored inflated with room air (group 2) and in lung tissue stored inflated with 1%, (group 3) as compared with deflated lung tissue (group I). Together with the decrease in ATP content in deflated lung tissue, a significant increase in lactate concentration was found after 15 minutes of normothermic ischemia, suggesting the onset of anaerobic metabolism (see Fig 2). Tissue lactate level increased from the preischemic value of 5.63 * 1.16 pmol/g dry wt to * 1.58 PmoVg dry wt at 15 minutes of ischemia (p <.1) and further to 33.8 * 2.36 PmoVg dry wt at 1 hour of ischemia ( p <.1). Adenosine triphosphate catabolism and lactate production were significantly delayed in lung tissue stored inflated compared with lung tissue stored deflated. When lung tissue was stored inflated with room air, levels of ATP, lactate, and energy charge remained stable during 6 minutes of normothermic ischemia (see Fig 2). Only after 9 minutes of ischemia did ATP level significantly decrease to 4.99? 1.18 pmol/g dry wt (p <.1) and lactate level increase to pmovg dry wt (p <.1). After 2 hours of normothermic ischemia, ATP concentration in lung tissue stored inflated with room air further decreased to 3.3? 1.25 pmovg dry wt and lactate concentration increased to t 1.95 pmovg dry wt. The energy charge of lung tissue stored inflated with room air decreased to.68 *.4 (p <.1) after 9 minutes of ischemia and to.51 *.6 (p <.1) after 12 minutes. When the lung was stored inflated with 1% oxygen, levels of ATP, energy charge, and lactate remained completely stable during 2 hours of normothermic ischemia (see Fig 2). Figure 3 shows the distribution of nucleosides and bases after 2 hours of normothermic ischemia in lung tissue stored deflated compared with lung tissue stored inflated with room air or inflated with 1% oxygen. after 2 hours of ischemia, hypoxanthine was by far the most important catabolite in deflated lung tissue, with levels of 3.3 *.46 pmovg dry wt. In lung tissue stored inflated with room air, hypoxanthine only increased to 1.16?.51 pmovg dry wt. As there was no breakdown of ATP in lung tissue stored inflated with 1%, at 2 hours of ischemia, there was also no significant increase in hypo- xanthine concentration as compared with nonischemic lung tissue. The onset of ATP catabolism and lactate production in lung tissue stored inflated with 1% oxygen is shown in Figure 4. The level of ATP decreased significantly to 7.42?.57 pmovg dry wt (p <.5) only after 5 hours of normothermic ischemia. Also, an increase in lactate concentration from 4.39?.97 to 9.91? 1.28 pmol/g dry wt (p <.1) was obtained after this period. The energy charge decreased after 6 hours of ischemia to.76?.3 (p <.1). At 8 hours of normothermic ischemia, the level of ATP further decreased to 4.28 f.6 pmol/g dry wt (p <.1), lactate concentration increased to 2.83? 1.73 pmovg dry wt (p:..1), and the energy charge decreased to.56 *.8 ( p <.1). The results shown above provide evidence that the half-life time for ATP decrease in deflated lung tissue at 37 C was 19.8? 3.5 minutes. When the lung was stored inflated with room air, the half-life time for ATP decrease - B = 7i \ a [- ATP Inflated 1%2 I u 1 4 I I Fig 4. Adenosine triphosphate (ATE ), lactate, and energy charge levels during 8 hours of normothermic ischemia (37 C) in lung tissue stored inflated with 1% O2 (group 4). Values f? standard deviation) at different ischemic times were compared with normoxic value by analysis of variance for repeated measures: * p <.5, ** p <.1, *** p <.1.

5 1993;55:17>9 DE LEYN ET AL 177 averaged minutes. From six experiments during which the lung tissue was stored inflated with 1% oxygen, this time was extended to 457 * 21 minutes in four experiments. In two experiments of inflation with 1% oxygen, the half-life time for ATP decrease was beyond the 8-hour sampling time. Comment In our experiments the mean ATP content of normoxic lung rabbit tissue was pmoyg dry wt, with an ATP/ADP ratio of 8.26 and a calculated energy charge ([ATP ADP]/[ATP + ADP + AMP], as defined by Atkinson and Walton [lo]) of.9. Thus the normal lung exists in a highly energized state. Our results are in agreement with those obtained by Bassett and associates [ll] and Fisher [12]. In the present study, ATP content in deflated, normothermic rabbit lung tissue was significantly decreased by 45% after 15 minutes of ischemia and further by 64% after 3 minutes of ischemia. After 12 minutes of ischemia, only 19.3% of the ATP was left in the examined lung tissue. On the other hand, lactate concentration started to increase significantly from 15 minutes of ischemia, suggesting an early onset of anaerobic metabolism. After 1 hour of ischemia the lactate level reached a steady state, because after 2 hours the levels were not significantly different from those measured at 1 hour. Our data on ATP catabolism and lactate production in deflated rabbit lung tissue are in agreement with the literature. Von Wichert [ 131 enzymatically measured ATP, ADP, AMP, and lactate levels in normothermic deflated rabbit lung tissue. He described a 7% decrease in ATP concentration after 3 minutes of ischemia and a fivefold increase in lactate concentration at that time. Measuring ATP degradation in degassed ischemic rabbit lung tissue by phosphorus 31 nuclear magnetic resonance, Hall and colleagues [14] demonstrated a 53% decrease in ATP level after 15 minutes and a 77% decrease after 2 hours of normothermic ischemia. Date and associates [15] measured ATP level by means of magnetic resonance analysis in isolated canine lungs preserved for 24 hours at 1" and 4 C. The lungs were flushed with low-potassium dextran solution and stored inflated with 1% oxygen. No breakdown of ATP during cold storage was observed in lung tissue inflated with 1% oxygen. Date and associates suspected that ATP was generated by oxidative phosphorylation during the preservation period. We studied ATP catabolism during normothermic (37 C) ischemia and compared deflated lung tissue, lung tissue inflated with room air, and lung tissue inflated with 1%,. When the lung was stored inflated with room air, an ATP decrease and a lactate increase were only seen after 9 minutes of normothermic ischemia. When the lung was stored inflated with 1% oxygen, ATP and lactate levels remained stable during 24 minutes of warm ischemia, and only at 5 hours of ischemia were a significant decrease in ATP level and an increase in lactate level found. During the period of stable ATP content, no increase in lactate concentration was found. These biochemical data suggest that the isolated inflated lung continues its aerobic metabolism, even without perfusion. The onset of anaerobic metabolism is related to the oxygen content of the alveolar space during storage. A possible explanation is the very thin alveolocapillary membrane; all parts of the lung are within a very short distance (.3 mm) of the gas in the air spaces [16]. The lung, which receives blood that is relatively low in oxygen content under normal circumstances, probably extracts the oxygen for the metabolic requirements of its component cells as much from inspired air as from blood. Several authors have shown that lung tissue requires oxidative metabolism to sustain essential functions including the uptake of serotonin from pulmonary capillary bed [17], the incorporation of carbon 14 into phospholipids [17] necessary for the maintenance of the alveolar surface tension [18], and the preservation of microfilament structure [19], which play an important role for the integrity of the basement membrane [2]. Exposure of the basement membrane during reperfusion could initiate platelet deposition and associated inflammatory response Our biochemical data support the empirical findings that inflation during storage prolongs the duration of tolerable ischemia. Veith and associates [l] compared the function of dog lungs preserved deflated versus the function of dog lungs stored inflated with 4% oxygen. The collapsed, nonventilated lung could tolerate only V2 hour of warm ischemia. In contrast, lungs stored inflated with 4% oxygen at 38 C remained functional after 2 or 3 hours of ischemia (6 of 1 survivors). At the present moment, most transplantation centers preserve their donor lungs inflated. The ideal oxygen concentration before lung graft storage remains an open question. Weder and co-workers [8] reported an improved 24-hour hypothermic preservation of isolated rabbit lungs inflated with 1% oxygen, compared with room air or nitrogen. Weder and co-workers reasoned that intraalveolar oxygen may be used by the lung during cold storage and, therefore, that the availability of oxygen during ischemia may be important for optimal lung preservation. Contradictory to this recent report is the finding by Koyama and associates [7], who subjected canine left lower pulmonary lobes to 6 hours of room air ischemia while ventilating them with either 1%,, room air, or 1% N,. Lobes ventilated with 1% oxygen or room air incurred massive weight gain; lobes ventilated with 1% nitrogen were significantly less edema tous. Although we were able to show higher ATP level and energy charge in ischemic lungs stored inflated with 1% oxygen compared with lungs stored inflated with room air or lungs stored deflated, the results of Koyama and associates [7] suggest that oxygen toxicity might be important when the isolated lung is stored inflated. In vivo, oxygen toxicity occurs only after lungs are ventilated with 1% oxygen for more than 24 hours. Cell damage, caused by oxygen toxicity, is due primarily to the intra-

6 178 DE LEYN ET AL 1993;55 17S9 cellular production of oxygen free radicals or other chemically active oxygen metabolites [22,23]. The occurrence of O2 toxicity depends on the balance between the amount of available endogenous antioxidants and the rate of production of oxygen free radicals. Maintenance of the normal blood flow, therefore, appears to protect the lung against the development of, toxicity [22]. The isolated lung is devoid of its circulation, and its natural defense mechanisms such as catalase, superoxide dismutase, and glutathione peroxidase can become depleted. This could explain the damage in isolated lungs ventilated for 6 hours with 1% oxygen, as in the experiments of Koyama and associates [7]. Oxygen free radicals responsible for oxygen toxicity can be produced by different mechanisms. Superoxide anion, the most important agent in oxygen toxicity, is produced within respiring cells, both by autooxidation of reduced electron-transferring components and by enzymatic processes [22]. A very important enzymatic source of free radical generation is xanthine oxidase [24]. Endogenously formed hypoxanthine is oxidized to xanthine and uric acid; superoxide anion (2J and hydrogen peroxide (H22) are formed as by-products. Superoxide anion and hydrogen peroxide are known to destroy the tissue [25]. Our results indicate that when the lung is stored inflated with 1% oxygen, free radicals are probably generated by other mechanisms than xanthine oxidase because no hypoxanthine is formed. We conclude that lung inflation during storage leads to prolonged aerobic metabolism and maintenance of high ATP concentration and energy charge. These biochemical data could explain the better functional results of lungs stored inflated versus lungs stored deflated because essential functions of the lung depend on oxidative metabolism and ATP. In an isolated lung, however, 1% oxygen may be toxic due to formation of oxygen free radicals. Efforts must be done to further analyze and correlate biochemical data and functional results of isolated lungs stored with different oxygen concentrations. We thank 3M Belgium for funding this project. We are grateful to Andre Berghen, John Das, and Peter Lemmens for expert technical assistance. References Veith FJ, Sinha SBP, Graves IS, Boley SJ, Dougherty JC. Ischemic tolerance of the lung: the effect of ventilation and inflation. J Thorac Cardiovasc Surg 1971;61:841. Fonkalsrud EW, Sanchez M, Lassaletta L, Smeesters C, Higashijima I. Extended preservation of the ischemic canine lung by ventilation with PEEP. J Surg Res 1975;18: Baldwin JC, Frist WH, Starkey TD, et al. Distant graft procurement for combined heart and lung transplantation using pulmonary artery flush and simple topical hypothermia for graft preservation. 1987;43: Fragomeni LS, Bonser RS, Jamieson SW. Cardiopulmonary transplantation: current practice. Transplant Int 1988;l: Patterson GA, Cooper JD, Goldman 8, et al. Technique of successful clinical double lung transplantation. Ann Thorac Surg 1988;45: Keenan RJ, Griffith BP, Kormos RL, Armitage JM, Hardesty RL. Increased perioperative lung preservation injury with lung procurement by Euro-Collins solution flush. J Heart Lung Transplant 1991;1:65&5. 7. Koyama I, Toung TJK, Rogers MC, Gurtner GH, Traystman RJ., radicals mediate reperfusion lung injury in ischemic, ventilated canine pulmonary lobe. J Appl Physiol1987;63: Weder W, Harper B, Shimokawa S, et al. The influence of intraalveolar oxygen concentration on lung preservation in a rabbit model. J Thorac Cardiovasc Surg 1991;11: Wynants J, Van Belle H. Single-run high-performance liquid chromatography of nucleotides, nucleosides, and major purine bases and its application to different tissue extracts. Anal Biochem 1985;144: Atkinson DE, Walton GM. Adenosine triphosphate conservation in metabolic regulation. Rat liver cleavage enzyme. J Biol Chem 1967;242: Bassett DJP, Fisher AB, Rabinowitz JL. Effect of hypoxia on incorporation of glucose carbons into lipids by isolated rat lung. Am J Physiol 1974;227: Fisher AB. Intermediary metabolism of the lung. Environ Health Perspect 1984;55: Von Wichert P. Studies on the metabolism of ischemic rabbit lungs. J Thorac Cardiovasc Surg 1972;63: Hall TS, Buescher PC, Borkon AM, Reitz BA, Michael JR, Baumgartner WA. "P Nuclear magnetic resonance determination of changes in energy state in lung preservation. Circulation 1988;78(Suppl 3): Date H, Lima, Matsumura A, Tsuji H, davignon A, Cooper JD. In a canine model, lung preservation at 1 C is superior to that at 4 C. J Thorac Cardiovasc Surg 1992;13: Weber KC, Visscher MB. Metabolism of the isolated canine lung. Am J Physiol 1969; Fisher AB, Steinberg H, Bassett D. Energy utilization by the lung. Am J Med 1974; Buckingham S, Heinemann HO, Sommers SC, McNary WF. Phospholipid synthesis in the large pulmonary alveolar cell. Its relation to lung surfactants. Am J Pathol 1966;48: Hinshaw DB, Armstrong BC, Beak TF, Hyslop PA. A cellular model of endothelial cell ischemia. J Surg Res 1988;44: Hynes ON, Destree AT. Relationships between fibronectin (Lets protein) and actin. Cell 1978;15: Jaffe AE. Physiologic functions of normal endothelial cells. Ann N Y Acad Sci 1985;454: McCord JM, Fridovich I. The biology and pathology of oxygen radicals. Ann Intern Med 1978;89: Fridovich I. The biology of oxygen radicals. The superoxide radical is an agent of oxygen toxicity; superoxide dismutase provides an important defense. Science 1978;21: McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 1985;312: Granger DN. Role of xanthine oxidase and granulocytes in ischemia-reperfusion injury. Am J Physiol 1988;255: H

7 19!33;5517%9 DE LEYN ET AL 179 DISCUSSION DR J. KENT TRINKLE (San Antonio, TX): 1 think Dr De Leyn and associates are absolutely right. I would forget about oxygen toxicity to the lung. For many years we have harvested all of our human lungs in the inflated position with 5 cm H,O of positive end-expiratory pressure and 1% oxygen. On three occasions, when I was personally doing the harvest, I thought 1 had the lungs inflated, but when 1 lifted them out the left lower lobe was atelectatic. So I recreated your experimental model, and when I transplanted that lung, the lower lobe had terrible postoperative reperfusion edema whereas the upper lobe worked quite well. This is a nice controlled model, although 1 did not do it intentionally. But I think you are right. DR DE LEYN: Thank you for your remark, Dr Trinkle. Your personal clinical experience clearly shows that inflation with 1% oxygen leads to improved lung preservation, and this correlates with our biochemical findings. However, we think that oxygen toxicity exists and should be considered when isolated lungs are stored inflated with 1% oxygen. We performed reperfusion studies of isolated rabbit lungs that were stored inflated with 1% oxygen and were submitted to 9 minutes of normothermic ischemia. In those experiments, we found severe reperfusion injury with massive edema and bad oxygenation capacity. The function of these lungs was worse than the function of lungs stored deflated in the same experimental set-up. These experiments suggest that pulmonary oxygen toxicity does exist when lungs are stored inflated with 1% oxygen. The reason why we do not see that much oxygen toxicity in human grafts preserved inflated with 1% oxygen is probably because these lungs are stored at 4 C. At hypothermia, the metabolism and also the production of oxygen free radicals are delayed. However, as also suggested by Koyama and co-workers, oxygen toxicity does occur when lungs are stored inflated with 1% oxygen during normothermic ischemia.