- Email: [email protected]
Establishment of an Ex Vivo Lung Perfusion Model Using Non-Heart-Beating Large Pigs
Establishment of an Ex Vivo Lung Perfusion Model Using Non-HeartBeating Large Pigs T. Okamoto, F. Chen, J. Zhang, T. Yamada, E. Nakayama, H. Morikawa, T. Bando, and H. Date ABSTRACT Background. Functional evaluation of potentially damaged lungs donated after cardiac death is crucial for widespread clinical transplantation. To date, the mean weight of animals used in studies of ex vivo lung perfusion (EVLP) has been 60 kg; however, in the clinical setting, donor weight may be greater. Objective. To investigate EVLP using lungs from large pigs (mean weight, 115 kg) to simulate human adult lungs donated after cardiac death. Materials and Methods. Five heart-lung blocks were obtained at 20 minutes after death at the slaughterhouse. The lungs were flushed and preserved on ice for 6 hours before being connected to an ex vivo lung circuit, and were perfused for at least 2 hours. Results. In all cases, perfusion was sustained for at least 2 hours. Mean (SEM) final flow rate was 4.9 (0.1) L/min, pulmonary artery pressure was 14.8 (1.7) mm Hg, and oxygen tension/fraction of inspired oxygen was 518.0 (18.0) mm Hg. The shunt fraction was 20.5% (4.0%). Histologic analysis demonstrated no significant pulmonary edema at the end of perfusion. Conclusion. We successfully completed EVLP using lungs from large pigs. UNG TRANSPLANTATION is now standard therapy for end-stage disease; however, because of a shortage of brain-dead donors, many patients do not receive this life-saving treatment. To compensate in part for this shortage, researchers have focused on refining preservation protocols.1,2 In addition, “marginal” donor lungs have begun to be used, but their number is also limited.3 Another potential way to compensate for the shortage of brain-dead donors could be use of lungs from donation after cardiac death, which has been performed in some countries.4,5 However, such use is hampered not only by ethical issues but also by lack of an evaluation system like that for assessment of brain-dead donors.3 In 2001, Steen et al4 introduced an ex vivo evaluation system for donation after cardiac death. Several groups have investigated similar systems using porcine models and observed that ex vivo perfusion systems have the potential to both evaluate and recondition donor lungs.5–7 In those studies, the weight of the pigs ranged from 30 to 72 kg, and total perfusion flow rate was less than 4.0 L/min; however, in the clinical setting, donor body weight and perfusion flow rate may be much greater. An experimental study using larger animals and higher flow rates is mandatory for clinical evaluation of ex vivo systems. From
L
the viewpoint of animal protection, use of larger laboratory animals has been limited.8 By obtaining specimens from a slaughterhouse, no animals are sacrificed for the experiments, resulting in efficient use of resources. Some researchers have already used cells and organs from animals obtained from slaughterhouses.9,10 The primary objective of the present study was to perform an ex vivo evaluation of larger pig lungs, simulating donor lungs from adult human beings, using specimens obtained from a slaughterhouse. MATERIAL AND METHODS Heart-Lung Blocks Five heart-lung blocks from domestic pigs with a mean weight of 115 kg were obtained from a local slaughterhouse where pigs are From the Department of Thoracic Surgery, Graduate School of Medicine, Kyoto University, Kyoto, Japan. This study was supported by grant 20390367 from the Japan Society for the Promotion of Science. Address reprint requests to Toshihiro Okamoto, MD, Department of Thoracic Surgery, Graduate School of Medicine, Kyoto Univeristy, 54 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 6068507, Japan. E-mail: [email protected]
0041-1345/10/$–see front matter doi:10.1016/j.transproceed.2010.03.140
© 2010 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
1598
Transplantation Proceedings, 42, 1598 –1601 (2010)
ESTABLISHMENT OF EX VIVO MODEL
1599
Ex Vivo Lung Perfusion Circuit The ex vivo lung perfusion circuit consisted of a hard-shelled reservoir, a centrifugal pump (CAPIOX; Terumo Corp, Tokyo, Japan), an artificial lung (␣Cube 4000; Nipro Medical Industries Ltd, Gunma, Japan) connected to a heat exchanger and gas supply, and a leukocyte/arterial filter (LG 6; Nihon Pall Ltd, Tokyo, Japan) (Fig 2). Sensors were used to monitor arterial and venous flow rates and temperature. The priming solution was 2 L of STEEN solution (Vitrolife AB), concentrated autologous red blood cells retrieved with a cell saver (Cell Saver 5; Haemonetics Corp, Braintree, Massachusetts), and a leukocyte filter (Sepacell Integra CA; Asahi Kasei Pharm, Tokyo, Japan). The perfusate included 500 mg of imipenem (Tienam; Banyu Pharmaceutical Co Ltd, Tokyo, Japan), 10,000 IU of heparin (Mochida Pharmaceutical Co, Ltd, Tokyo, Japan), THAM SET (Otsuka Pharmaceuticl Factory, Inc), and 20 U of insulin (Novolin R 100; Novo Nordisk Pharma Ltd, Tokyo, Japan). The priming circulation was performed at 24°C for 20 minutes until the following data were confirmed: perfusate hematocrit, 10% to 15%; pH, 7.35 to 7.45; and base excess, 0. Three gases (oxygen, nitrogen, and carbon dioxide) were adjusted until blood gas analysis demonstrated arterial oxygen tension of 140 to 150 mm Hg and arterial carbon dioxide tension of 34 to 45 mm Hg.
Fig 1. Experiment timetable. In a local slaughterhouse, heartlung blocks were recovered after 20 minutes of warm ischemia. Immediately thereafter, antegrade flushing of the pulmonary artery (PA) was performed, and the lungs were preserved in ice slush. After 6 hours of cold ischemia, ex vivo lung perfusion (EVLP) was maintained for at least 2 hours.
treated according to laws governing such facilities. The experimental timetable is shown in Fig 1. Pigs were rendered unconscious via electric shock. The carotid artery and jugular vein were severed with a knife, and death was by exsanguination. The blood was collected and mixed with an anticoagulant (S-400; JMS Co, Ltd, Tokyo, Japan), and preserved at 4°C. Heart-lung blocks wre excised by a slaughterhouse worker at 15 minutes after pig death. The weight of each heart-lung block was recorded, and those that weighed more than 2.0 kg were excluded from the study. The blocks were visually examined for signs of lung injury or lesions such as from pneumonia or blood aspiration. The trachea was opened, and a tracheal tube was inserted and secured. The pulmonary artery was immediately cannulated, and the left atrium was opened. After the lungs were ventilated several times, 2 L of 4°C electrolyte solution (Perfadex; Vitrolife AB, Kungsbacka, Sweden) supplemented with 10,000 IU of heparin and 0.6 mL of THAM SET (Otsuka Pharmaceutical Factory, Inc, Tokushima, Japan) was flushed antegrade through the pulmonary artery. Blocks with massive pleural and tracheal air leakage were excluded. Mean total warm ischemia time was approximately 20 minutes. The lungs were kept semi-inflated with room air by clamping the tracheal tube, soaked in Perfadex surrounded by ice slush, transported to our laboratory, and preserved for 6 hours at 4°C before ex vivo evaluation.
Fig 2. Ex vivo lung perfusion circuit. The circuit consists of a hard-shelled reservoir, a centrifugal pump, an artificial lung, a leukocyte/arterial filter, and a heart-lung block on the box. Sensors for flow rate and temperature are set on the pulmonary artery (PA) and the left atrium (LA) tubes. The tracheal tube is connected to the ventilator. CO2, carbon dioxide; N2, nitrogen; O2, oxygen.
1600
OKAMOTO, CHEN, ZHANG ET AL Table 1. Results of Ex Vivo Lung Perfusion Pig No. Variable
Weight, kg Hct Duration of perfusion, hr Maximum flow rate, L/min PAP, mm Hg Maximum At 2 hr PaO2, mm Hg Shunt fraction
1
2
3
4
5
Mean (SEM)
1.5 12.2 2.0 5.0
1.5 11.6 4.0 5.0
1.4 12.5 2.0 5.0
1.5 14.9 3.0 5.0
1.9 14.4 2.3 4.5
1.6 (0.1) 13.4 (0.6) 2.7 (0.4) 4.9 (0.1)
20 17 518 12.0
17 10 460 32.8
13 13 522 23.0
16 14 573 13.8
20 20 517 12.6
17.2 (1.3) 14.8 (1.7) 517.8 (17.9) 20.6 (4.0)
Hct, hematocrit; PaO2, arterial oxygen tension; PAP, pulmonary arterial pressure.
Ex Vivo Lung Perfusion After 6 hours of cold preservation, the heart-lung block in the evaluation box (Vitrolife AB) was cannulated (28F) via the pulmonary artery through the right ventricular outflow. A 36F/48F 2-stage cannula was inserted through the left ventricle, placed in the left atrium, and secured after clots were removed. Pressure catheters were placed in the pulmonary artery and left atrium. Perfusion was started at a low flow rate, 0.1 L/min. The first 100 mL of blood from the left atrium was discarded. The flow rate was gradually decreased to maintain pulmonary arterial pressure at less than 20 mm Hg.11 When the temperature of blood from the left atrium was 32°C, ventilation was started using respirator settings as follows: tidal volume, twice the flow rate (in liters per minute); respiratory rate, 15/min; positive end-expiratory pressure, 5 cm water; and fraction of inspired oxygen, 1.0. The gas to the artificial lung was changed to nitrogen (93%) and carbon dioxide (7%) to deoxygenate the perfusate. When the temperature of the left atrium reached 37°C, the flow rate was increased stepwise, and finally set at 5 L/min. Within the first 50 minutes, the flow rate reached the final amount. Left arterial pressure was maintained between 0 and 3 mm Hg by changing the height of the reservoir and partially clamping the venous tube. Perfusion was continued for 2 hours.
remaining 7 blocks, 5 were randomly selected for study. Their mean (SEM; range) weight was 1.6 (0.1; 1.4 –1.9) kg (Table 1). Ex Vivo Lung Perfusion of Large-Pig Lungs
In all 5 heart-lung blocks, ex vivo perfusion was performed successfully (Table 1). The hematocrit concentration in the perfusate was 13.3% (0.6%). Mean (SEM) final flow rate was 4.9 (0.1) mL/min; in 1 block weighing 1.9 kg, the final flow rate was 4.5 L/min. Maximum pulmonary arterial pressure ranged from 13 to 20 mm Hg, and pulmonary arterial pressure after 2 hours of perfusion was 14.8 (1.7) mm Hg. At 10 minutes after reaching the final flow rate, the arterial oxygen tension/fraction of inspired oxygen was 518 (18) mm Hg. The shunt fraction (Qs/Qt) was 20.5% (4.0%). The duration of perfusion in all blocks was at least 2 hours (range, 2– 4 hours), although the end of perfusion was not decided by deterioration of the perfused lungs. In all studied specimens, no significant findings were observed in the pulmonary architecture including the pulmonary vessels and the alveola and bronchi.
Analysis of Donor Lungs
DISCUSSION
The hematocrit concentration and blood gases in the perfusate to and from the lung were analyzed at 2 hours after the start of perfusion. The data were used to calculate the intrapulmonary shunt fraction (shunted blood to total blood flow ratio [Qs/Qt]), calculated as follows: Qs/Qt (%) ⫽ (Cc ⫺ Ca)/(Cc ⫺ Cv) ⫻ 100, where Cc, Ca, and Cv represent the oxygen content in the pulmonary capillary blood, the deoxygenerator effluent, and the pulmonary vein effluent, respectively. Pulmonary arterial pressure, left arterial pressure, and perfusion flow rate were monitored continuously. Histologic analysis of the lungs after perfusion was also performed. At the end of perfusion, the lungs were immersed in 10% formalin, embedded in paraffin, and stained with hematoxylineosin. Data are given as mean (SEM).
Originally, Steen et al11 established an ex vivo pig lung evaluation system at a flow rate of 4.0 L/min using pigs weighing, on average, 60 kg, and while the vivo pulmonary flow rate was 4.9 L/min. It can be speculated that this flow rate simulated the physiologic condition of pig lungs. However, the optimal flow rate for ex vivo evaluation of donor lungs has not been established.12 According to one study, pulmonary flow rate in a 20-kg pig should be 2.0 L/min, and in 40-kg pigs should be 4.0 L/min.13 It is generally believed that a massively high flow rate in the pulmonary artery may damage lung function.12 Pulmonary edema develops gradually if ex vivo perfusion is performed with nonpulsatile flow using a centrifugal pump with pulmonary artery pressure greater than 20 mm Hg.14 In the present study, we chose to use a nonpulsatile flow rate of 5.0 L/min because it matched the cardiac output in 115-kg pigs. To determine the optimal flow rate, more experiments are necessary including lung transplantation after assessment using an ex vivo lung perfusion system. However, decreased pulmonary
RESULTS Pig Heart-Lung Blocks
Ten heart-lung blocks were obtained from a slaughterhouse. Three were excluded because they weighed more than 2.0 kg or there was lung injury or pneumonia. Of the
ESTABLISHMENT OF EX VIVO MODEL
flow can cause an abnormally high shunt fraction and hypoxemia in patients with normal lung function.15 Thus, the minimum perfusion flow rate that provides reliable functional assessment of donor lungs must also be investigated. A number of studies have used pig organs obtained from a slaughterhouse, including cells and organs such as the kidney, for several purposes.9,10 We recommend use of such heart-lung blocks from the viewpoint of laboratory animal protection.8 It is also preferable because in the present study, the cost of a heart-lung block from a slaughterhouse was only $30.00. In addition, in Japan, use of pig lungs is the only way to study lung transplantation because no rejected brain-dead donor lungs are available for research because there is no national agreement on the issue. The present study has several limitations. First, we obtained heart-lung blocks from a slaughterhouse, where death was induced by loss of blood. This model simulated the conditions under which non-heart-beating donor lungs are harvested after a gunshot; these lungs can be classified as Maastricht category I. The cause of death is important in non-heart-beating donors. This model with 20 minutes of warm ischemia time may have only slight injury compared with other lungs such as those obtained after hypoxic cardiac death.16 Second, because the slaughter of pigs is performed in a slaughterhouse by its workers, occasionally there may be unexpected damage to the lungs. It is important to inspect the heart-lung blocks and to measure their weight. Any lungs with signs of pneumonia or blood aspiration should not be used. In conclusion, we successfully performed an ex vivo lung evaluation with large-pig lungs, which provided meaningful evidence in favor of clinical use of non-heart-beating lung donors. Use of pig lungs obtained from a slaughterhouse not only decreases the cost of performing such studies but also provides laboratory animal protection. More studies are needed to determine the ideal minimum and maximum perfusion flow rates. ACKNOWLEDGEMENTS We thank Kazuo Umihira for providing pig organs from the slaughterhouse, Kenji Kita for preparation of the ex vivo lung perfusion circuit, and Amy Moore for proofreading the manuscript.
1601
REFERENCES 1. Chen F, Nakamura T, Wada H: Development of new organ preservation solutions in Kyoto University. Yonsei Med J 45:1107, 2004 2. de Perrot M, Keshavjee S: Lung preservation. Semin Thorac Cardiovasc Surg 16:300, 2004 3. Pierre A, Sekine Y, Hutcheon M, et al: Marginal donor lungs: a reassessment. J Thorac Cardiovasc Surg 123:421, 2002 4. Steen S, Sjöberg T, Pierre L, et al: Transplantation of lungs from a non-heart-beating donor. Lancet 357:825, 2001 5. Fernández E, Calatayud J, Jarabo J, et al: Profitability of our lung retrieval program from non-heart-beating donors. Eur J Cardiothorac Surg 35:287, 2009 6. Erasmus M, Fernhout M, Elstrodt J, et al: Normothermic ex vivo lung perfusion of non-heart-beating donor lungs in pigs: from pretransplant function analysis towards a 6-h machine preservation. Transpl Int 19:589, 2006 7. Inci I, Zhai W, Arni S, et al: Fibrinolytic treatment improves the quality of lungs retrieved from non-heart-beating donors. J Heart Lung Transplant 26:1054, 2007 8. Louhimies S: Directive 86/609/EEC on the protection of animals used for experimental and other scientific purposes. Altern Lab Anim 30:217, 2002 9. Grosse-Siestrup C, Unger V, Fehrenberg C, et al: A model of isolated autologously hemoperfused porcine slaughterhouse kidneys. Nephron 92:414, 2002 10. Noguchi H, Ueda M, Nakai Y, et al: Modified two-layer preservation method (M-Kyoto/PFC) improves islet yields in islet isolation. Am J Transplant 6:496, 2006 11. Steen S, Liao Q, Wierup P, et al: Transplantation of lungs from non-heart-beating donors after functional assessment ex vivo. Ann Thorac Surg 76:244, 2003 12. Cypel M, Yeung J, Hirayama S, et al: Technique for prolonged normothermic ex vivo lung perfusion. J Heart Lung Transplant 27:1319, 2008 13. Kuwahara M, Hirose H, Sugano S: Alteration in cardiovascular function of piglets during the growth process. Nippon Juigaku Zasshi 50:990, 1988 14. Steen S, Ingemansson R, Eriksson L, et al: First human transplantation of a nonacceptable donor lung after reconditioning ex vivo. Ann Thorac Surg 83:2191, 2007 15. Casthely P, Lear S, Cottrell J, et al: Intrapulmonary shunting during induced hypotension. Anesth Analg 61:231, 1982 16. Van de Wauwer C, Neyrinck AP, Geudens N, et al: The mode of death in the non-heart-beating donor has an impact on lung graft quality. Eur J Cardiothorac Surg 35:919, 2009
L
the viewpoint of animal protection, use of larger laboratory animals has been limited.8 By obtaining specimens from a slaughterhouse, no animals are sacrificed for the experiments, resulting in efficient use of resources. Some researchers have already used cells and organs from animals obtained from slaughterhouses.9,10 The primary objective of the present study was to perform an ex vivo evaluation of larger pig lungs, simulating donor lungs from adult human beings, using specimens obtained from a slaughterhouse. MATERIAL AND METHODS Heart-Lung Blocks Five heart-lung blocks from domestic pigs with a mean weight of 115 kg were obtained from a local slaughterhouse where pigs are From the Department of Thoracic Surgery, Graduate School of Medicine, Kyoto University, Kyoto, Japan. This study was supported by grant 20390367 from the Japan Society for the Promotion of Science. Address reprint requests to Toshihiro Okamoto, MD, Department of Thoracic Surgery, Graduate School of Medicine, Kyoto Univeristy, 54 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 6068507, Japan. E-mail: [email protected]
0041-1345/10/$–see front matter doi:10.1016/j.transproceed.2010.03.140
© 2010 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
1598
Transplantation Proceedings, 42, 1598 –1601 (2010)
ESTABLISHMENT OF EX VIVO MODEL
1599
Ex Vivo Lung Perfusion Circuit The ex vivo lung perfusion circuit consisted of a hard-shelled reservoir, a centrifugal pump (CAPIOX; Terumo Corp, Tokyo, Japan), an artificial lung (␣Cube 4000; Nipro Medical Industries Ltd, Gunma, Japan) connected to a heat exchanger and gas supply, and a leukocyte/arterial filter (LG 6; Nihon Pall Ltd, Tokyo, Japan) (Fig 2). Sensors were used to monitor arterial and venous flow rates and temperature. The priming solution was 2 L of STEEN solution (Vitrolife AB), concentrated autologous red blood cells retrieved with a cell saver (Cell Saver 5; Haemonetics Corp, Braintree, Massachusetts), and a leukocyte filter (Sepacell Integra CA; Asahi Kasei Pharm, Tokyo, Japan). The perfusate included 500 mg of imipenem (Tienam; Banyu Pharmaceutical Co Ltd, Tokyo, Japan), 10,000 IU of heparin (Mochida Pharmaceutical Co, Ltd, Tokyo, Japan), THAM SET (Otsuka Pharmaceuticl Factory, Inc), and 20 U of insulin (Novolin R 100; Novo Nordisk Pharma Ltd, Tokyo, Japan). The priming circulation was performed at 24°C for 20 minutes until the following data were confirmed: perfusate hematocrit, 10% to 15%; pH, 7.35 to 7.45; and base excess, 0. Three gases (oxygen, nitrogen, and carbon dioxide) were adjusted until blood gas analysis demonstrated arterial oxygen tension of 140 to 150 mm Hg and arterial carbon dioxide tension of 34 to 45 mm Hg.
Fig 1. Experiment timetable. In a local slaughterhouse, heartlung blocks were recovered after 20 minutes of warm ischemia. Immediately thereafter, antegrade flushing of the pulmonary artery (PA) was performed, and the lungs were preserved in ice slush. After 6 hours of cold ischemia, ex vivo lung perfusion (EVLP) was maintained for at least 2 hours.
treated according to laws governing such facilities. The experimental timetable is shown in Fig 1. Pigs were rendered unconscious via electric shock. The carotid artery and jugular vein were severed with a knife, and death was by exsanguination. The blood was collected and mixed with an anticoagulant (S-400; JMS Co, Ltd, Tokyo, Japan), and preserved at 4°C. Heart-lung blocks wre excised by a slaughterhouse worker at 15 minutes after pig death. The weight of each heart-lung block was recorded, and those that weighed more than 2.0 kg were excluded from the study. The blocks were visually examined for signs of lung injury or lesions such as from pneumonia or blood aspiration. The trachea was opened, and a tracheal tube was inserted and secured. The pulmonary artery was immediately cannulated, and the left atrium was opened. After the lungs were ventilated several times, 2 L of 4°C electrolyte solution (Perfadex; Vitrolife AB, Kungsbacka, Sweden) supplemented with 10,000 IU of heparin and 0.6 mL of THAM SET (Otsuka Pharmaceutical Factory, Inc, Tokushima, Japan) was flushed antegrade through the pulmonary artery. Blocks with massive pleural and tracheal air leakage were excluded. Mean total warm ischemia time was approximately 20 minutes. The lungs were kept semi-inflated with room air by clamping the tracheal tube, soaked in Perfadex surrounded by ice slush, transported to our laboratory, and preserved for 6 hours at 4°C before ex vivo evaluation.
Fig 2. Ex vivo lung perfusion circuit. The circuit consists of a hard-shelled reservoir, a centrifugal pump, an artificial lung, a leukocyte/arterial filter, and a heart-lung block on the box. Sensors for flow rate and temperature are set on the pulmonary artery (PA) and the left atrium (LA) tubes. The tracheal tube is connected to the ventilator. CO2, carbon dioxide; N2, nitrogen; O2, oxygen.
1600
OKAMOTO, CHEN, ZHANG ET AL Table 1. Results of Ex Vivo Lung Perfusion Pig No. Variable
Weight, kg Hct Duration of perfusion, hr Maximum flow rate, L/min PAP, mm Hg Maximum At 2 hr PaO2, mm Hg Shunt fraction
1
2
3
4
5
Mean (SEM)
1.5 12.2 2.0 5.0
1.5 11.6 4.0 5.0
1.4 12.5 2.0 5.0
1.5 14.9 3.0 5.0
1.9 14.4 2.3 4.5
1.6 (0.1) 13.4 (0.6) 2.7 (0.4) 4.9 (0.1)
20 17 518 12.0
17 10 460 32.8
13 13 522 23.0
16 14 573 13.8
20 20 517 12.6
17.2 (1.3) 14.8 (1.7) 517.8 (17.9) 20.6 (4.0)
Hct, hematocrit; PaO2, arterial oxygen tension; PAP, pulmonary arterial pressure.
Ex Vivo Lung Perfusion After 6 hours of cold preservation, the heart-lung block in the evaluation box (Vitrolife AB) was cannulated (28F) via the pulmonary artery through the right ventricular outflow. A 36F/48F 2-stage cannula was inserted through the left ventricle, placed in the left atrium, and secured after clots were removed. Pressure catheters were placed in the pulmonary artery and left atrium. Perfusion was started at a low flow rate, 0.1 L/min. The first 100 mL of blood from the left atrium was discarded. The flow rate was gradually decreased to maintain pulmonary arterial pressure at less than 20 mm Hg.11 When the temperature of blood from the left atrium was 32°C, ventilation was started using respirator settings as follows: tidal volume, twice the flow rate (in liters per minute); respiratory rate, 15/min; positive end-expiratory pressure, 5 cm water; and fraction of inspired oxygen, 1.0. The gas to the artificial lung was changed to nitrogen (93%) and carbon dioxide (7%) to deoxygenate the perfusate. When the temperature of the left atrium reached 37°C, the flow rate was increased stepwise, and finally set at 5 L/min. Within the first 50 minutes, the flow rate reached the final amount. Left arterial pressure was maintained between 0 and 3 mm Hg by changing the height of the reservoir and partially clamping the venous tube. Perfusion was continued for 2 hours.
remaining 7 blocks, 5 were randomly selected for study. Their mean (SEM; range) weight was 1.6 (0.1; 1.4 –1.9) kg (Table 1). Ex Vivo Lung Perfusion of Large-Pig Lungs
In all 5 heart-lung blocks, ex vivo perfusion was performed successfully (Table 1). The hematocrit concentration in the perfusate was 13.3% (0.6%). Mean (SEM) final flow rate was 4.9 (0.1) mL/min; in 1 block weighing 1.9 kg, the final flow rate was 4.5 L/min. Maximum pulmonary arterial pressure ranged from 13 to 20 mm Hg, and pulmonary arterial pressure after 2 hours of perfusion was 14.8 (1.7) mm Hg. At 10 minutes after reaching the final flow rate, the arterial oxygen tension/fraction of inspired oxygen was 518 (18) mm Hg. The shunt fraction (Qs/Qt) was 20.5% (4.0%). The duration of perfusion in all blocks was at least 2 hours (range, 2– 4 hours), although the end of perfusion was not decided by deterioration of the perfused lungs. In all studied specimens, no significant findings were observed in the pulmonary architecture including the pulmonary vessels and the alveola and bronchi.
Analysis of Donor Lungs
DISCUSSION
The hematocrit concentration and blood gases in the perfusate to and from the lung were analyzed at 2 hours after the start of perfusion. The data were used to calculate the intrapulmonary shunt fraction (shunted blood to total blood flow ratio [Qs/Qt]), calculated as follows: Qs/Qt (%) ⫽ (Cc ⫺ Ca)/(Cc ⫺ Cv) ⫻ 100, where Cc, Ca, and Cv represent the oxygen content in the pulmonary capillary blood, the deoxygenerator effluent, and the pulmonary vein effluent, respectively. Pulmonary arterial pressure, left arterial pressure, and perfusion flow rate were monitored continuously. Histologic analysis of the lungs after perfusion was also performed. At the end of perfusion, the lungs were immersed in 10% formalin, embedded in paraffin, and stained with hematoxylineosin. Data are given as mean (SEM).
Originally, Steen et al11 established an ex vivo pig lung evaluation system at a flow rate of 4.0 L/min using pigs weighing, on average, 60 kg, and while the vivo pulmonary flow rate was 4.9 L/min. It can be speculated that this flow rate simulated the physiologic condition of pig lungs. However, the optimal flow rate for ex vivo evaluation of donor lungs has not been established.12 According to one study, pulmonary flow rate in a 20-kg pig should be 2.0 L/min, and in 40-kg pigs should be 4.0 L/min.13 It is generally believed that a massively high flow rate in the pulmonary artery may damage lung function.12 Pulmonary edema develops gradually if ex vivo perfusion is performed with nonpulsatile flow using a centrifugal pump with pulmonary artery pressure greater than 20 mm Hg.14 In the present study, we chose to use a nonpulsatile flow rate of 5.0 L/min because it matched the cardiac output in 115-kg pigs. To determine the optimal flow rate, more experiments are necessary including lung transplantation after assessment using an ex vivo lung perfusion system. However, decreased pulmonary
RESULTS Pig Heart-Lung Blocks
Ten heart-lung blocks were obtained from a slaughterhouse. Three were excluded because they weighed more than 2.0 kg or there was lung injury or pneumonia. Of the
ESTABLISHMENT OF EX VIVO MODEL
flow can cause an abnormally high shunt fraction and hypoxemia in patients with normal lung function.15 Thus, the minimum perfusion flow rate that provides reliable functional assessment of donor lungs must also be investigated. A number of studies have used pig organs obtained from a slaughterhouse, including cells and organs such as the kidney, for several purposes.9,10 We recommend use of such heart-lung blocks from the viewpoint of laboratory animal protection.8 It is also preferable because in the present study, the cost of a heart-lung block from a slaughterhouse was only $30.00. In addition, in Japan, use of pig lungs is the only way to study lung transplantation because no rejected brain-dead donor lungs are available for research because there is no national agreement on the issue. The present study has several limitations. First, we obtained heart-lung blocks from a slaughterhouse, where death was induced by loss of blood. This model simulated the conditions under which non-heart-beating donor lungs are harvested after a gunshot; these lungs can be classified as Maastricht category I. The cause of death is important in non-heart-beating donors. This model with 20 minutes of warm ischemia time may have only slight injury compared with other lungs such as those obtained after hypoxic cardiac death.16 Second, because the slaughter of pigs is performed in a slaughterhouse by its workers, occasionally there may be unexpected damage to the lungs. It is important to inspect the heart-lung blocks and to measure their weight. Any lungs with signs of pneumonia or blood aspiration should not be used. In conclusion, we successfully performed an ex vivo lung evaluation with large-pig lungs, which provided meaningful evidence in favor of clinical use of non-heart-beating lung donors. Use of pig lungs obtained from a slaughterhouse not only decreases the cost of performing such studies but also provides laboratory animal protection. More studies are needed to determine the ideal minimum and maximum perfusion flow rates. ACKNOWLEDGEMENTS We thank Kazuo Umihira for providing pig organs from the slaughterhouse, Kenji Kita for preparation of the ex vivo lung perfusion circuit, and Amy Moore for proofreading the manuscript.
1601
REFERENCES 1. Chen F, Nakamura T, Wada H: Development of new organ preservation solutions in Kyoto University. Yonsei Med J 45:1107, 2004 2. de Perrot M, Keshavjee S: Lung preservation. Semin Thorac Cardiovasc Surg 16:300, 2004 3. Pierre A, Sekine Y, Hutcheon M, et al: Marginal donor lungs: a reassessment. J Thorac Cardiovasc Surg 123:421, 2002 4. Steen S, Sjöberg T, Pierre L, et al: Transplantation of lungs from a non-heart-beating donor. Lancet 357:825, 2001 5. Fernández E, Calatayud J, Jarabo J, et al: Profitability of our lung retrieval program from non-heart-beating donors. Eur J Cardiothorac Surg 35:287, 2009 6. Erasmus M, Fernhout M, Elstrodt J, et al: Normothermic ex vivo lung perfusion of non-heart-beating donor lungs in pigs: from pretransplant function analysis towards a 6-h machine preservation. Transpl Int 19:589, 2006 7. Inci I, Zhai W, Arni S, et al: Fibrinolytic treatment improves the quality of lungs retrieved from non-heart-beating donors. J Heart Lung Transplant 26:1054, 2007 8. Louhimies S: Directive 86/609/EEC on the protection of animals used for experimental and other scientific purposes. Altern Lab Anim 30:217, 2002 9. Grosse-Siestrup C, Unger V, Fehrenberg C, et al: A model of isolated autologously hemoperfused porcine slaughterhouse kidneys. Nephron 92:414, 2002 10. Noguchi H, Ueda M, Nakai Y, et al: Modified two-layer preservation method (M-Kyoto/PFC) improves islet yields in islet isolation. Am J Transplant 6:496, 2006 11. Steen S, Liao Q, Wierup P, et al: Transplantation of lungs from non-heart-beating donors after functional assessment ex vivo. Ann Thorac Surg 76:244, 2003 12. Cypel M, Yeung J, Hirayama S, et al: Technique for prolonged normothermic ex vivo lung perfusion. J Heart Lung Transplant 27:1319, 2008 13. Kuwahara M, Hirose H, Sugano S: Alteration in cardiovascular function of piglets during the growth process. Nippon Juigaku Zasshi 50:990, 1988 14. Steen S, Ingemansson R, Eriksson L, et al: First human transplantation of a nonacceptable donor lung after reconditioning ex vivo. Ann Thorac Surg 83:2191, 2007 15. Casthely P, Lear S, Cottrell J, et al: Intrapulmonary shunting during induced hypotension. Anesth Analg 61:231, 1982 16. Van de Wauwer C, Neyrinck AP, Geudens N, et al: The mode of death in the non-heart-beating donor has an impact on lung graft quality. Eur J Cardiothorac Surg 35:919, 2009