Effects of Lung Preservation With Euro-Collins and University of Wisconsin Solutions on Endothelium- Dependent Relaxations

Transcription

1 Effects of Lung Preservation With Euro-Collins and University of Wisconsin Solutions on Endothelium- Dependent Relaxations Martin Stri.iber, MD, Klaus A. Ehlers, MD, Folke N. Nilsson, MD, PhD, Virginia M. Miller, PhD, Christopher G. A. McGregor, FRCS, and Axel Haverich, MD Department of Thoracic and Cardiovascular Surgery, Hannover Medical School, Hannover, Germany, and Department of Cardiovascular Surgery, and Physiology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota Background. This study compares the effect of lung preservation using flush perfusion of Euro-Collins or University of Wisconsin solution on the pulmonary vascular function of endothelium-dependent and endothelium-independent relaxations. Methods. Rings of canine intrapulmonary arteries were studied after 6 hours of cold ischemia in Euro-Collins or University of Wisconsin preservation solution. Endothelium-dependent and endothelium-independent relaxations were induced in organ chamber experiments. To also study pulmonary resistance vessels, endotheliumdependent relaxations were induced in in vitro perfused intact rabbit lungs. Results. In the organ chamber experiments, a moderate but significant (p < 0.05) reduction in endotheliumdependent relaxations were found in the perfused and stored vessels. In perfused rabbit lungs, a decrease in the endothelial response occurred immediately after perfu- sion with Euro-Collins solution. However, a recovery and overshooting response was found after preservation with either solution and 6 hours of cold ischemia. A significant increase in the sensitivity of smooth muscle cells to nitric oxide was shown in both preparations. Conclusions. Both crystalloid perfusion fluids cause a decrease in endothelial function during the perfusion procedure. In contrast, endothelial function is well preserved during the ischemic time. University of Wisconsin solution induced a higher sensitivity of the vascular smooth muscle to the endothelium-derived relaxing factor nitric oxide. A reduction in pulmonary vascular resistance after University of Wisconsin preservation may be of importance in subsequent clinical lung transplantation. (Ann Thorac Surg 1997;63: ) 1997 by The Society of Thoracic Surgeons S uccessful organ transplantation depends on reliable preservation of cellular function of the graft. Donor lungs are particularly vulnerable with tolerable ischemic times reported to be shorter than those of other transplantable solid organs [1]. In addition, reperfusion injury remains a significant factor for morbidity after lung transplantation [2] and probably also for the long-term outcome. Many studies [3, 4] have demonstrated a loss of endothelium-dependent vasodilation and nitric oxide (NO) release after myocardial preservation for transplantation using University of Wisconsin (UW) solution or other crystalloid solutions. We [5] reported earlier a decrease of endothelium-dependent vasodilation after flush perfusion of the lung with Euro-Collins (EC) solution. Endothelial cells not only have an important role in the adhesion of blood cells, but also modulate the muscular tone of the vessel wall [6]. Injury to the endothelium from preservation and ischemia could increase the vascular resistance in the postoperative period, thus promoting reperfusion injury. The aim of this study was to Accepted for publication Nov 16, Address reprint requests to Dr Striiber, Department of Thoracic and Cardiovascular SurgeD,, Hannover Medical School, Konstanty Gutschow Str 8, Hannover, Germany. compare the impact of cold flush with either Euro-Collins or UW solution and 6 hours of ischemia on endotheliumdependent vasodilation. Euro-Collins and UW solutions were used because they are currently the most common choice for preservation. Material and Methods Experiments were performed using rings of canine intrapulmonary artery and in vitro perfused intact rabbit lungs to distinguish between effects on large pulmonary arteries and arteriolar resistance vessels. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH publication 80-23, revised 1978). Intrapulmonary Artery Rings DESmN. Male beagle dogs weighing 8 to 12 kg were divided into three groups of 7 animals each. In group 1 (control), left pneumonectomy was performed, and that nonperfused lung was studied immediately after re by The Society. of Thoracic Surgeons /97/$17.00 Published by Elsevier Science Inc PII S (97)

2 Ann Thorac Surg STRUBER ET AL ;63: LUNG PRESERVATION AND ENDOTHELIUM moval. In group 2, lungs were perfused with EC solution and stored at 4 C, and in group 3, UW solution was used for perfusion and storage. Intrapulmonary arteries of groups 2 and 3 were prepared and studied after 6 hours of cold storage (Fig 1). PRESERVATION PROCEDURE. The dogs were anesthetized with ketamine hydrochloride and sodium pentobarbital, intubated, and ventilated with 40% oxygen. In the control group, a left thoracotomy was performed, and the pulmonary artery and veins were isolated and ligated. The left bronchus was divided and tied, and the lung was removed and placed in cold (4 C) modified Krebs- Ringer's solution (millimolar): NaCl, 118.3; KC1, 4.0; CaCl, 2.5; MgSO 4, 0.7; KH2PO4, 1.2; NaHCO3, 25.0; and glucose, In the EC and UW groups, the pericardium was opened through a median sternotomy. The azygos vein was ligated. The aorta and the pulmonary artery were dissected, and heparin sodium (3 mg/kg) was administered intravenously. Catheters were inserted into the main pulmonary artery to monitor pulmonary artery pressure and to infuse the preservation solution. Both venae cavae were ligated, and after four to five beats to empty the heart of blood, the aorta was clamped. The inferior vena cava and the left atrial appendage were cut. Then the preservation solution was infused into the pulmonary artery. The flow was adjusted to maintain pulmonary artery pressure at 16 to 20 mm Hg. The perfusion was continued to a volume of 60 ml/kg. After removal of the heart-lung block, the right lung was dissected free in an inflated state and stored for 6 hours at 4 C in the preservation solution. ORGAN CHAMBER EXPERIMENTS. Intrapulmonary arteries were dissected and cleaned of connective tissue. The blood vessel was cut into rings 5 mm in length. In some rings, the endothelium was deliberately removed by gently rubbing the luminal surface with a stainless steel wire. The rings were suspended between a clip and a force transducer (Gould UC2; Viggio-Spectramed Inc, Critical Care Division, Oxnard, CA) for the measurement of isometric force. The rings were placed in organ chambers filled with 25 ml of the modified Krebs-Ringer's solution at 37 C and bubbled with 95% oxygen and 5% carbon dioxide. They were equilibrated at a passive tension of less than 5 g for 30 minutes. The rings were stretched and stimulated with KC1 (20 mmol/l) at each level of stretch until the increase in tension was maximal (optimal length). Rings with and without endothelium were studied in parallel. Maximal contractions to 60 mmol/l KC1 were measured. This agonist contracts the smooth muscle by depolarization. Then cumulative dose-response curves were obtained to the receptor operated norepinephrine (10 9 to 10 5 mol/l). To study endothelium-dependent relaxations, the rings were contracted with a submaximal concentration ( to 10 6 mol/l) of phenylephrine hydrochloride. Responses to acetylcholine (10 9 to 10 5 mol/l), adenosine diphosphate (10 s to 10-4 mol/l), bradykinin (10 ao to 10 6 mol/l), and the calcium ionophore A (10 -l to 10 6 mol/l) were determined. The former agonists stimulate receptors on the endothelial cells to synthesize NO, which causes relaxation of the smooth muscle through activation of cyclic guanosine monophosphate synthesis. The calcium ionophore A releases NO and prostaglandin 12 by a nonreceptor-dependent mechanism. Responses to A were measured in the presence and absence of the cyclooxygenase inhibitor indomethacin (10 s mol/l). Adenosine diphosphate was also studied with indomethacin to prevent endothelial prostaglandin release, but because of the limited number of organ chambers, this agonist was not studied without indomethacin. To determine possible differences in the response of the smooth muscle to NO, a concentration dose-response curve was obtained for each group in rings without endothelium. CHEMICAL AND DRUGS. The following drugs were used: acetylcholine chloride, adenosine diphosphate, bradykinin, calcium ionophore A23187, indomethacin, L-norepinephrine bitartrate, and phenylepinephrine (all, Sigma Chemicals, St. Louis, MO). Indomethacin was dissolved in NaECO3 (10 -s mol/l), and A23187 was dissolved in dimethyl sulfoxide ( mol/l). All drugs were prepared daily in distilled water. Nitric oxide was prepared by the method of Palmer and associates [7]. Nitric oxide from a cylinder (Union Carbide, Chicago IL) was injected into glass bulbs filled with 100 ml of distilled water (bubbled with helium for 3 hours) to give stock solutions of NO. Concentrations of the drugs are reported as the final molar concentration in the organ chamber. Euro-Collins solution was obtained from Fresenius AG, Bad Homburg, Germany, and UW solution from Du Pont Pharmaceuticals, Wilmington, DE, and Bad Homburg, Germany. Continuously Perfused Lungs DESIGN. Rabbits weighing 3.5 to 4.0 kg were divided into four groups of 7 animals each. In the control group, the heart-lung block was excised, and the main pulmonary artery was cannulated and perfused with modified Krebs-Ringer's solution. The heart-lung block was suspended in a warm and moist chamber, and the lungs were ventilated. During continuous perfusion, contraction of the pulmonary vasculature was induced by prostaglandin F2~. Then dose-response curves to the relaxation to the endothelium-dependent vasodilator substance P were obtained. Substance P was chosen because it was found to induce endothelium-dependent relaxations more reliably than other agonists in this preparation. All experiments were performed in the presence of indomethacin to prevent prostaglandin release from aggregated thrombocytes. Maximum relaxation to nitroglycerin (glyceryl trinitrate) by bolus injection was measured thereafter. In the second group (EC0), the lungs were perfused with EC solution (80 ml/kg) in situ. After perfusion in the third group (EC6), the lungs were stored

3 1430 STRLTBER ET AL Ann Thorac Surg LUNG PRESERVATION AND ENDOTHELIUM 1997;63: Design Control n=7 no preservation no storage Group 1 n=7 Euro-Collins preservation 6h storage Group 2 n=7 UW preservation 6h storage maximal contraction 60 mm KCl relaxation submaximal contraction phenylephrine cumulative dose response curve 1 ~ relaxation to 2 ~. relaxation to acetylcholine bradykinin +-indomethacin -indomethacin 3 ~ relaxation to adenosine diphosph. + inclomethacin 4 relaxation to A indomethacin 5 ~ relaxation to Nitric oxide - indomethacin 6 ~) contraction to norepinephrine - indomethacin i ~ with endothelium without endothelium Fig 1. Design of organ chamber experiments. The three groups, the curves qf maximal contraction and cumulative dose-response, the drugs, the presence or absence qf imhnnethacin, and the presence or absence of endothelium or the rings are shown. (UW = University of Wisconsin.) for 6 hours in the preservation solution at 4 C. Perfusion of the last group (UW6) was performed using UW solution, and then the lungs were stored as in the EC6 group. The heart-lung blocks were excised after perfusion, and tests were performed as in the control group right after excision (EC0) or storage (EC6, UW6) (Fig 2). PRESERVATION PROCEDURE. The rabbits were anesthetized using ketamine hydrochloride and sodium pentobarbital and ventilated through a tracheotomy with 40% oxygen. The pericardium was opened through a median sternotomy. Heparin (3 mg/kg) was administered intravenously. In the control group, the aorta and the pulmonary artery

4 Ann Thorac Surg STRUBER ET AL ;63: LUNG PRESERVATION AND ENDOTHELIUM J 10C'-~ 90' 8O 7O 6O contraction PGF 2alpha relaxation Substance P 10 nanomol 100 nanomol 1000 nanomol relaxation Glyceroltrinitrate 1 mg 50 4O 3O Design Control, ECO n=6 n=6 no Euro-Collins preservation preservation no storage no storage r EC6 n=6 Euro-Collins preservation 6h storage UW6 n=6 UW preservation 6h Storage Fig 2. Design of continuously peucused lung experiments. The doseresponse carve shows relaxations caused by Substance P and nitroglycerin (glyceryl trinitrate). (PGF = prostaglandin F; UW = University of Wisconsin.) were dissected. The venae cavae were ligated, and the aorta was clamped. Then a perfusion cannula was inserted through a right ventricular incision into the main pulmonary artery. After incision of the left atrium, the heart-lung block was excised. In all other groups, preservation solution was applied prior to excision. The perfusion cannula was connected to a pressure-monitoring line. During infusion of the preservation solution for 3 to 6 minutes to a volume of 80 ml/kg, care was taken not to exceed a perfusion pressure of 24 mm Hg. The heart-lung blocks of the EC6 and UW6 groups were stored in their preservation solution in an inflated state. PERFUSION EXPERIMENTS. After excision or storage, the whole heart-lung blocks were mounted in a warm (37.5 C) and moist chamber. The lungs were ventilated with 40% oxygen at a rate of 40 per minute and a maximum inspiratory pressure of 25 mm Hg. The pulmonary cannula was connected to a pump that delivered a nonpulsatile flow of 40 ml/min of the modified Krebs- Ringer's solution [plus dextran 3% (wt/vol) to achieve isooncotic pressure] at 37.5 C. The cyclooxygenase inhibitor indomethacin was added to the perfusion fluid (10 5 mol/l). The perfusion cannula was connected to a transducer for measurement of pressure. After 30 minutes of equilibration, a bolus injection of prostaglandin F2, (10 5 mol/l) was administered to define maximum pulmonary artery pressure. When tension had returned to resting levels (3 to 7 mm Hg) a continuous infusion of prostaglandin F2~ (10 7 to 10 5 tool/l) was started to increase pulmonary artery pressure to 60% to 80% of the contraction to prostaglandin F2~ (10-5 mol/l). Concentrationresponse curves to substance P (10 to 1,000 pmol/l) were obtained. When the substance P-induced relaxation was finished, endothelium-independent relaxation was induced by a bolus injection of nitroglycerin (10 rag). 2O I0 0-8 Control --0- UW ~ EC "~ p = 0,031 "s :, without Endothelium Fig 3. Cumulative dose-response curves of relaxations induced by adenosine diphosphate (10-8 to 10 _4 tool~l) plus indomethacin UO -5 tool/l). Rings were contracted with phenylephrine, and the data are presented as a percent decrease in tension from that contraction. Values are shown as the mean + the standard error of the mean. Contractions were not different between groups and were not altered by indomethacin. The responses of rings without endothelium did not differ between groups and were combined for clarity. Relaxations in rings with endothelium in the two study groups were significantly reduced compared with the control group (p = ). (EC = Euro-Collins solution; UW = University of Wisconsin solution.) CHEMICALS AND DRUGS. Substance P, indornethacin, prostaglandin F2~, dextran (60 to 90 Ka) (all, Sigma Chemicals) and nitroglycerin (Perlinganit; Schwarz Pharma, Germany) were used. Indomethacin was dissolved in.e 7O 60 5o O 10 ~),,, l -I without -II-- Control -'Oh- UW ~ EC - - Endothelium Fig 4. Cumulative dose-response curves of relaxations induced by bradykinin (10 lo to 10-6 tool/l). Contractions were not different between groups. The responses of rings without endothelium did not differ between groups and were combined. Abbreviations are the same as in Figure 3, as are the protocol details and the presentation of the data.

5 1432 STRUBER ET AL Ann Thorac Surg LUNG PRESERVATION AND ENDOTHELIUM 1997;63: , E 60-._~ ~ o without -I~ Control -~P UW ~ EC -- Endothelium Fig 5. Cumulative dose-response curves of relaxations induced by acetylcholine (10 9 to 10 5 tool/l). Contractions were not different between groups. The responses qf rings without endothelium did not differ between groups and were combined. Abbreviations are the same as in Figure 3, as are the protocol details and the pres~ttation of the data. Na2CO3 (10 5 tool/l). All drugs were prepared daily in distilled water. Data Analysis and Statistics The results are expressed as the mean _+ the standard error of the mean. Responses were compared by maximal t Drug Concentration in log mol "D" Control -4P UW "llt- EC -- without Endothelium " " p = 0,049 Fig 6. Cumulative dose-response curves of relaxations induced by acetylcholine (10 9 to 10 5 tool/l) plus indomethacin (10 ~ tool/l). Contractions were not different between groups and were not altered by indomethacin. The responses of rings without endothelium did not differ between groups and were combined. Relaxations in rings with endothelium in the two study groups were significantly reduced compared with the control group (p ~ 0.049). Abbreviations are the same as in Figure 3, as are the protocol details and the presentation of the data o i =, v = I, i,, = = I Control -,qp UW "~- EC -- without Endotll"elium "~ p = 0,009 Fig 7. Cumulative dose-response curves of relaxations induced by A23187 (10 9 to 10 ~ tool~l). Contractions were not different between groups. The responses qf rings without endothelium did not differ between groups and were combined. Relaxations in rings with endothelium in the t'wo study groups were significantly reduced compared with the control group (p = 0.009). Abbreviations are the same as in Figure 3, as are the protocol details and the presentation of the data. response, areas under the curve, or, when appropriate, the means of the concentration producing 50% of the maximal response. Statistical analysis was performed using multivariate analysis. Scheff6's test for multiple comparisons was used to identify differences between mean values. Differences were considered significant when the value of p was less than Results Organ-Chamber Experiments ENDOTHELIUM-DEPENDENT RELAXATIONS. Adenosine diphosphate caused concentration-dependent relaxations only in rings with endothelium (Fig 3). These relaxations were significantly less in both the EC and UW groups after 6 hours of cold storage (p = 0.031). The curves of EC solution- and UW solution-perfused and stored rings were comparable. Relaxations to another receptordependent agonist, bradykinin, were similar in the EC and UW groups (Fig 4). Acetylcholine induced relaxations only in rings with endothelium and were analyzed in the presence (Fig 5) and absence of indomethacin (Fig 6). The curves of the UW and EC groups shifted to the right. This was significant in the presence of indomethacin (p = 0.049) but not in its absence (p = 0.077). The receptor-independent calcium ionophore A23187 caused concentration-dependent relaxations in rings with endothelium but not in those without it. Rings in the EC and UW groups were significantly less sensitive than the control rings (p ) (Fig 7). In the presence of indomethacin, all curves shifted significantly to the left

6 Ann Thorac Surg STRUBER ET AL ;63: LUNG PRESERVATION AND ENDOTHELIUM (p < 0.05) (Fig 8). In the presence of that cyclooxygenase inhibitor, relaxations of the EC and UW groups were significantly reduced compared with the control group (p = 0.02). Differences between the EC and the UW group were not significant. ENDOTHELIUM-INDEPENDENT RELAXATIONS. Nitric oxide decreased isometric force in rings without endothelium in a dose-dependent fashion in all groups (Fig 9). However, the maximal relaxations in the UW group were significantly greater than in the control and EC groups (p < 0.05). CONTRACTILE RESPONSES. Potassium chloride caused concentration-dependent contractions in all rings. The contractions were comparable among the groups. Norepinephrine increased isometric force in rings with and without endothelium in a concentration-dependent way. Without endothelium, all curves shifted to the right, thereby indicating a higher sensitivity, so that there were no significant differences. In the UW group, rings with endothelium contracted less than in the other groups, but this finding also was not significant. Continuously Perfused Lungs ENDOTHELIUM-DEPENDENT RELAXATIONS. Substance P caused a concentration-dependent reduction in pulmonary artery pressure in all lungs (Fig 10). The response in EC solution-perfused but not stored lungs (EC0) was less than that of the control group. The lungs of the perfused and stored groups (EC6, UW6) were significantly more ~ 0 0 : \\\ 3 0 ~ i i i i i i i i i i i i i i i tl- Control -(P UW -~- EC -- without Endothelium "~ p = 0,02 Fig 8. Cumulative dose-response curves qf relaxah'ons induced by A23187 (10 9 to 10 6 tool/l) plus indomethacin (10 -~ tool~l). Contractions were not different between groups and were not altered by indomethacin. The responses of rings without endothelium did not differ between groups and were combined. Relaxations in rings wi& endothelium in the study groups were significantly reduced compared with the control group (p = 0.002). Abbreviations are the same as in Figure 3, as are the protocol details and the presentation qf the data i i i i i i Control -'~ UW ~ EC "~ p = 0,038 Fig 9. Dose-response curves of relaxations induced by nitric oxide (10 ~ to 10 ~ tool~l). Only rings without endothelium were used. Contractions were not different between groups. Maximum relaxations in the University of Wisconsin solution (UW) group were significantly increased compared with the control and Euro-Collins solution (EC) groups. (p = 0.038). The protocol details and the presentation of the data are the same as in Figure 3. sensitive to the induced relaxation (p = 0.001). Relaxations of the stored lungs were not significantly different whether they were perfused with EC or UW solution. ENDOTHELIUM-INDEPENDENT RELAXATIONS. Bolus injection of 1 mg of nitroglycerin caused a profound relaxation in all lungs. The response was significantly increased (p = 0.017) in the UW6 group (Fig 11). All other mean values were comparable. Comment This study demonstrates a reduction in endotheliumdependent NO release after 6 hours of ischemia in pulmonary arteries of canine lungs preserved for transplantation with flush perfusion using EC or UW solution. The moderate but significant decrease in the endotheliurn-dependent response was comparable to the response seen right after perfusion without storage, which was reported previously [5]. Therefore, we conclude that both EC and UW solutions preserve endothelial function well for up to 6 hours. In the perfused rabbit lungs, a decrease in endothelial response was present after perfusion with EC solution. Similar results were seen in organ chamber experiments as well [5]. A recovery and overshooting response was found after additional ischemia in both the EC and UW groups. Part of this may be due to an increased sensitivity of the smooth muscle cell to NO after preservation with UW solution because an increased response to NO was found in the canine pulmonary artery as well as to the NO donor nitroglycerin in

7 1434 STROBER ET AL Ann Thorac Surg LUNG PRESERVATION AND ENDOTHELIUM 1997;63: O 0~ 80.E i ' o 10C 6O 5O % equals Prostaglandin F 2 alpha contraction i Substance P bolus concentration in nanomol -- Control... EC 0 -'~ EC 6 "~ UW 6 p = Fig 10. Cumulative dose-response curves of relaxations induced by substance P (10, 100, and 1,000 nmol/l) plus indomethacin (10 ~5 tool/l). Lungs were continuously perfused with Krebs solution. Pulmonary artery pressure was increased by prostaglandin F2~, infusion, and the data are presented as a percent decrease in tension from that pressure. Values are shown as the mean + the standard error of the mean. The contractions were not different between groups. Relaxations in the University of Wisconsin solution (UW) 6 and Euro- Collins solution (EC) 6 groups were significantly increased compared with the control group (p ). the perfused rabbit lungs. The arteriolar resistance vessels in the perfused rabbit lungs may respond differently than do the large pulmonary artery rings used in the organ-chamber preparation. 4._ I[l I~ Control I~ EC 0 ~ EC 6 m UW 6 p=0017 Fig 11. Relaxation to nitroglycerin bolus (1 rag) plus indomethacin (10 "~ tool/l). The contractions were not different between groups. Relaxations in the University of Wisconsin solution (UW) 6 group were significantly increased compared with all other groups (p = 0.017). Abbreviations are the same as in Figure 10, as are the protocol details and the presentation of the data. t Endothelial dysfunction after preservation using crystalloid solutions has been reported for the coronary circulation. Mankad and associates [8] found a temperature-dependent endothelial dysfunction in the rat heart as a result of preservation in UW solution. Nilsson and co-workers [4] reported a decrease in endotheliumdependent relaxations in porcine coronary arteries after perfusion with crystalloid solution. Additional cold storage for 5 hours had no further effect, and no endothelial dysfunction was found with the use of blood cardioplegia. Sellke and coauthors [9] looked at microvascular relaxation in the porcine coronary system. They also demonstrated a reduction in endothelium-dependent relaxation after preservation with crystalloid cardioplegia. However, an additional depression of the endothelial response was observed after reperfusion. Speciesdependent differences in endothelium-dependent relaxation must be considered [10] as well as differences in the different blood vessels of the same animal [11]. However, the results of this study and all investigations of the coronary vasculature suggest a moderate loss of endothelium-dependent relaxation after perfusion with crystalloid solutions. As the differences were found in the presence of indomethacin and the response of the smooth muscle to NO was not decreased, a reduction in NO release by the endothelium must be considered the most likely explanation. This hypothesis is to be verified in further experiments with measurement of NO release. Additional cold storage for up to 6 hours does not aggravate this dysfunction, but reperfusion may cause additional depression of the endothelial response. In some studies [12], histologic examination revealed injury to endothelial cells after cardioplegia, but this finding was not supported by Sellke and associates [9]. In addition, Unruh [13] reported a reduction in angiotensinconverting enzyme activity in canine lungs after perfusion with EC solution to be a metabolic endothelial dysfunction. The EC and UW preservation solutions have quite different compositions, but both contain a high concentration of potassium, as do most cardioplegic solutions: University of Wisconsin Pentafraction 50 g/l Lactobionate 100 mmol/l Raffinose 30 mmol/l Adenosine 5 mmol/l MgSO 4 5 mmol/l Glutathione 3 mmol/l Allopurinol 1 mmol/l KH2PO 4 25 mmol/l NaOH 29 mmol/l KOH 104 mmol/l ph 7.6 Na 29 meq/l K 125 meq/l Osm 300 mosm/l Euro-Collins KH2PO 4 15 mmol/l K2HPO mmol/l

8 Ann Thorac Surg STRUBER ET AL ;63: LUNG PRESERVATION AND ENDOTHELIUM KC1 15 mmol/l NaHCO 3 10 mmol/l Glucose 35 g/l ph 7.25 Na 10 meqll K 115 meq/l Osm 375 mosm/l In addition to the change in membrane potential, biochemical alterations with metabolic dysfunction will be considered. The reduction in endothelium-dependent relaxations will be measured after washout of the perfusion fluid, so that membrane potentials should be restored. The reduction in angiotensin-converting enzyme activity after flush perfusion supports this hypothesis. In conclusion, preservation of the lungs using EC or UW solution causes depression in endothelium-dependent relaxations. This may contribute to a higher vascular resistance during preservation and reperfusion. During cold ischemia for up to 6 hours, the endothelial response is well preserved, both with EC solution and UW solution. In addition, a higher sensitivity to NO is caused by preservation in UW solution in canine pulmonary arteries and in perfused rabbit lung. This finding may be of importance for a decrease in vascular resistance in UW solution-preserved lungs. Further studies are to be performed to elucidate the effect of the preservation procedure on endothelial function after reperfusion. References 1. Theodore J, Lewiston N. Lung transplantation comes of age. N Engl J Med 1990;332: Haverich A, Scott WC, Jamieson SW. Twenty years of lung preservation--a review. J Heart Transplant 1985;4: Pearl JM, Laks H, Drinkwater DC, et al. Loss of endotheliumdependent vasodilation and nitric oxide release after myocardial protection with University of Wisconsin solution. J Thorac Cardiovasc Surg 1994;107: Nilsson FN, Miller VM, Vanhoutte PM, McGregor CGA. Methods of cardiac preservation alter the function of the endothelium in porcine coronary arteries. J Thorac Cardiovasc Surg 1991;102: Str~iber M, McGregor CGA, Locke TJ, Miller VM. Effect of flush perfusion with Euro-Collins solution on pulmonary arterial function. Transplant Proc 1990;22: Furehgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980;288: Palmer RMJ, Ferrige AG, Moncada S. Nitric oxide accounts for the biological activity of the endothelium-derived relaxing factor. Nature 1987;327: Mankad P, Slavik Z, Yacoub M. Endothelial dysfunction caused by University of Wisconsin preservation solution in the rat heart. J Thorac Cardiovasc Surg 1992;104: Sellke FW, Shafique T, Schoen F, Weintraub RM. Impaired endothelium-dependent coronary microvaseular relaxation after cold potassium cardioplegia and reperfusion. J Thorac Cardiovasc Surg 1993;105: Fi~rstermann U, Trogisch G, Busse R. Species-dependent differences in the nature of endothelium-derived vascular relaxing factor. Eur J Pharmacol 1984;106: Houston DA, Burnstock G, Vanhoutte PM. Different P2- purinergic receptor subtypes of endothelium and smooth muscle in canine blood vessels. J Pharmacol Exp Ther 1987; 241: Harjula A, Mattila S, Mattila l, et al. Coronary endothelial damage after crystalloid cardioplegia. J Cardiovasc Surg (Torino) 1984;25: Unruh H. Pulmonary endothelial cell function after modified Eurocollins solution infusion. J Heart Lung Transplant 1993; 12:700-5.