Effect of Prostaglandin I, and Superoxide Dismutase on Reperfusion Injury of Warm Ischemic Lung Chojiro Yamashita, MD, Hidefumi Oobo, MD, Fukumasa Tsuji, MD, Satosi Tobe, MD, Hidehiro Yamamoto, MD, Hiroomi Nakamura, MD, Masayosi Okada, MD, and Kazuo Nakamura, MD Second Division, Department of Surgery, Kobe University School of Medicine, Kobe, Japan
A prostaglandin I, (PGI,) analogue and superoxide dismutase (SOD) were administered to dogs with pulmonary denervation, and their effects on warm ischemic damage to the lung were studied. Twenty-seven adult mongrel dogs were divided into a control group (6 dogs), a PGI, group (7 dogs), an SOD group (6 dogs), and a heparin group (8 dogs). The left pulmonary hilum was dissected, with PGI, (1pg/kg) being administered to the PGI, group and heparin (100U/kg) to the heparin group. Then the left lung was placed in a warm ischemic state for 1 hour. The SOD group also received 20 mg/kg of SOD intravenously 1 minute before reperfusion. Before warm ischemia, immediately after reperfusion, and 1 hour and 2 hours afterward, the blood gases, left pulmonary vascular resistance, and other data were measured under right pulmonary artery clamping. Arterial oxygen tension showed significantly better values in the SOD
and PGI, groups than in the control and heparin groups. The left pulmonary vascular resistance increased with time in the control group but did not increase in the PGI, group. Pulmonary microangiography showed that dilatation of the pulmonary arterioles was prominent in the PGI, group. The quantity of pulmonary extravascular fluid was significantly less in the PGI, and SOD groups than in the control and heparin groups. Histological examination showed marked collapse of capillaries, intraalveolar hemorrhage, and edema in the control and heparin groups, whereas these changes were only slight in the PGI, and SOD groups. Thus, both PGI, and SOD appeared to have a protective effect on the pulmonary microcirculation and tissue, and may be effective in preventing warm ischemic damage in lung transplantation. (Ann Thorac Surg 1992;54:9214)
I
effective against warm ischemic reperfusion disturbance, and we studied its role in a pulmonary denervation model.
n lung transplantation, the appearance of infiltrates in the transplanted lung is seen on postoperative chest roentgenograms along with a decrease in respiratory function and worsening of the blood gases, and this transient pulmonary impairment is called the reimplantation response. It would be desirable to distinguish these events from rejection, elucidate the underlying mechanism, and develop methods to prevent the occurrence of these early postoperative phenomena. Warm ischemic damage, denervation, and the interruption of lymphatics have all been considered as possible causes of the reimplantation response. Prostaglandin I, (PGI,) inhibits platelet aggregation, causes vasodilatation, is cytoprotective, and stabilizes lysosome membranes, so we considered that it might be possible to reduce the reimplantation response by using a PGI, analogue. In unilateral pulmonary transplantation, unlike cardiac transplantation, blood at a temperature of about 36°C enters the transplanted lung at the time of reperfusion, and damage from oxygen radicals is likely to occur. Therefore, the radical scavenger superoxide dismutase (SOD) might also be Accepted for publication March 6, 1992. Address reprint requests to Dr Yamashita, Second Division, Department of Surgery, Kobe University School of Medicine, 7-5-2 Kusunoki-cho Chuoku, Kobe, Japan 650.
0 1992 by The Society of Thoracic Surgeons
Material and Methods Twenty-seven adult mongrel dogs weighing 10 to 15 kg were divided into four groups: 6 dogs formed the control group, 7 dogs were in the PGI, group, 6 dogs formed the SOD group, and 8 dogs were in the heparin group. All animals received humane care in compliance with the ”Guide for the Care and Use of Laboratory Animals” published by the National Institute of Health (NIH publication No. 85-23, revised 1985). General anesthesia was induced with ketamine hydrochloride (Ketalar; ParkeDavis, Morris Plains, NJ; 10 mg/kg), thiamylal sodium (Yoshitomi Inc, Osaka, Japan; 20 mgkg), and pancuronium bromide (Mioblock; Organon Inc, West Orange, NJ; 1 mg/kg), and a volume-controlled ventilator was used to provide ventilation (inspired oxygen fraction, 0.60; internal airway pressure, 15 cm H,O; ventilation rate, 20/min). Next, an arterial line was inserted into the right carotid artery and a Swan-Ganz catheter was advanced to the pulmonary artery trunk through the jugular vein. Arterial and mixed venous blood gases were measured, and then thoracotomy was performed through the fourth left inter0003-4975/92/$5.00
922
Ann Thorac Surg 1992;54:9214
YAMASHITA ET AL REPERFUSION INJURY OF WARM ISCHEMIC LUNG
reo
1'
1 hr
2hr
control G (n = 6)
dard error. Statistical significance was determined by two-factorial analysis of variance and the t test. A p value less than 0.05 was considered to be significant.
Results
PGI, G (n = 7) SOD I -rep
..
SOD G (n = 6) heparin
rep
heparin G (n = 8 )
IiJOITJO
r t PA clamp for 5min to measure
blood gas. PApressure. PAflow
1 EVLW. PAgraphy. histology
Fig 1. Protocol of the experiment. Arrows indicate the time of administration of prostaglandin l2 (PGI,), superoxide dismutase (SOD), and heparin. Shaded squares show the time of measurement. (EVLW = extravascular lung water; 1-HS = left hilar stripping; PA = pulmonary artery; PAgraphy = pulmonary arteriography; rep = reperfusion; WIT = warm ischemic time.)
costal space. The left pulmonary artery, pulmonary vein, and main bronchus were dissected out, and after the nerves, bronchial arteries, and lymphatics were completely transected, the right and left pulmonary arteries were taped. During this procedure, a PGI, analogue (1pg/kg) was administered through the Swan-Ganz catheter in the pulmonary artery over a 30-minute period in the PGI, group. In the heparin group, heparin (100 Ulkg) was injected intravenously during dissection of the hilum. Next, an electromagnetic flow meter was attached to the pulmonary artery trunk, and the right pulmonary artery was clamped to evaluate hemodynamics in the left lung. Ventilation was performed at an inspired oxygen fraction of 1.0 for 5 minutes and then the blood gases, pulmonary artery pressure, and pulmonary arterial blood flow were measured to provide standard values. Thereafter, the right pulmonary artery was released and the left lung was deflated. Then the left pulmonary artery and vein and the main bronchus were clamped with a vascular clamp for 1 hour to produce warm ischemia of the left lung. In the SOD group, 20 mg/kg of SOD was injected intravenously 1 minute before reperfusion. Immediately after reperfusion as well as 1 hour and 2 hours afterward, measurements were performed to investigate lung function, and then the heart and lungs were excised. Extravascular fluid in the left lung was determined by the dry weight method in 5 animals from each group. In the remaining dogs of each group (1in the control group, 2 in the PGI, group, 1 in the SOD group, and 3 in the heparin group), 50% barium solution was injected from the pulmonary artery trunk into both lungs at a pressure of 30 cm H,O to perform soft roentgenography. The plates were enlarged 10 times for comparative study (microangiography). Finally, a histological study was performed (Fig 1). The results were expressed as a mean value and stan-
The following data were measured under clamping of the right pulmonary artery to evaluate left lung function after warm ischemia. Arterial oxygen tension decreased with time in both the control and heparin groups, but remained stable at 280 to 360 mm Hg in both the PGI, and SOD groups. Significant differences were noted between the PGI, and SOD groups and the control and heparin groups at 1 hour and 2 hours after reperfusion (Fig 2). Arterial carbon dioxide tension showed a slow increase with time in all four groups, but no significant difference was seen among them. The left pulmonary vascular resistance was high in all groups, and it was noted that the cardiac output was only about 1 L/min in small dogs weighing 10 to 15 kg. Before warm ischemia, left pulmonary vascular resistance was 1,800 to 2,200 dyne s * cm? in all four groups, but it increased sharply in the control and SOD groups after 2 hours of reperfusion to become 5,780 f 681 and 5,005 ? 1,459 dyne s cmp5, respectively. However, in the PGI, group it increased only slightly to 2,137 f 205 dyne s cmP5 immediately after reperfusion and to 2,748 & 347 dyne s * cmP5 after 2 hours. In the heparin group, left pulmonary vascular resistance increased moderately to 2,775 ? 261 dyne * s * cmP5 immediately after reperfusion and to 3,695 278 dyne * s * cmP5 after 2 hours. But significant differences were noted between PGI, group and control group only at 2 hours after reperfusion (Fig 3). The left extravascular lung water content was 6.99 mL/kg in the control group and 6.02 mLlkg in the heparin group. It was significantly lower in the PGI, and SOD groups, being 4.93 mL/kg and 4.36 mL/kg, respectively (Fig 4).
-
-
- -
-
*
control-G ( n = 6 ) (n = 7 ) (n=6) 0 heparin-G ( n = 8 ) 0
o PGln-G A SOD-G
O/ntubation before WI
after WI
1 hour
2 hour
Fig 2 . Arterial oxygen tension (Pa02) during occlusion of the right pulmonary artery before warm ischemia of the left lung, just after reperfusion, and 1 and 2 hours afterward. Significant difierences were noted between the prostaglandin l2 (PGI,) and superoxide dismutase (SOD) groups and the control group at 1 and 2 hours after reperfusion. (HS = hilar stripping; WI = warm ischemia.)
YAMASHITA ET AL REPERFUSION INJURY OF WARM ISCHEMIC LUNG
Ann Thorac Surg 1992;54:9214
923
0 control-G
(n = 6 ) (n = 7) (n=6) heparin-(; (n = 8)
o PG12-G ASODG
I-F'VR
(dyneis,ec/cm-?
* : P < 0.001
50004000-
30002000 -
intubation before WI
after WI
1 hour
2 hour
Fig 3. Left pulmonary vascular resistance 0-PVR) during occlusion of the right pulmonary artery before warm ischemia of the left lung, just after reperfusion, and 1 and 2 hours afterward. A significant difference was seen between the prostaglandin I , (PGI,) group and control group at 2 hours after reperfusion. (HS = hilar stripping; SOD = superoxide dismutase; WI = warm ischemia.)
In postmortem pulmonary microangiography, the contrast medium did not reach the periphery of the left pulmonary arterial vasculature in both the control and heparin groups, and occlusion of vessels of 100 to 200 pm in diameter was seen. The SOD group showed similar findings. In the PGI, group, however, the vascular tree was visualized clearly to a level of less than 100 pm in diameter, and little difference was noted when compared with the right lung, which had not been manipulated (Fig 5). In the control and heparin groups, alveolar hemorrhage, exudate, and interstitial edema were prominent on histologic examination. The capillaries were narrow and collapsed. Sludging in the capillaries was seen in the control group. In the SOD group, alveolar edema and hemorrhage were very slight and the alveolar architecture was almost normal. In the PGI, group, alveolar hemor-
Yd : P < 0.05
M E A N f SD
EVLW
Fig 5 . By postmortem pulmona y microangiography, the pulmonary vascular tree was visualized clearly to the level of less than 100 p m in the prostaglandin I, (PGI,) group, but contrast medium did not reach these small arterioles in the superoxide dismutase, control, and heparin groups.
4-
rhage and exudates were noted occasionally. Dilatation of the pulmonary arterioles was prominent, and a marked increase of blood flow was noted as far distally as the capillaries.
2-
Comment
6-
Cause of the Reimplantation Response n "
control-G PG1,-G
SOD-G heparin-G
Fig 4. The left extravascular lung water (EVLW) content: EVLW was significantly lower in the prostaglandin I, (PGI,) and superoxide dismutase (SOD) groups than in the control group.
The reimplantation response is a transient state of pulmonary edema occurring after transplantation, and warm ischemic reperfusion injury, chemical mediators, denervation, or operative manipulation are all regarded as possible causes [l-31. Warm ischemic reperfusion injury
924
YAMASHITA ET AL REPERFUSION INJURY OF WARM ISCHEMIC LUNG
includes cellular disturbances due to ischemia itself and other disturbances occurring at the time of reperfusion [4]. Because the lung contains air, unlike other organs, cellular function is maintained even after warm ischemia lasting for 2 to 3 hours [5-71, but interstitial edema, hemorrhage, and the destruction and swelling of mitochondria have been shown to develop after 60 minutes of warm ischemia [8]. Furthermore, because pulmonary edema becomes more marked with the increase in warm ischemic time, there is no doubt that warm ischemia is an important cause of the reimplantation response. When an organ undergoes warm ischemia, lipid peroxidation and the no-reflow phenomenon can occur. Free radicals are produced in the endothelial cell membranes and in neutrophils at the time of reperfusion and cause an increase in cell membrane permeability [9, 101. Superoxide dismutase is reported to scavenge free radicals produced by the hypoxanthine-xanthine oxidase system and by neutrophils, and to inhibit reperfusion injury by suppressing membrane lipid peroxidation [ll-131. The no-reflow phenomenon occurs when, after warm ischemia and reperfusion, the blood does not flow sufficiently from the arterioles to the capillaries due to sludging of red blood cells, obstruction by white blood cells, vasospasm, and narrowing of the vessel lumen due to edema of the vascular endothelium [14]. These reperfusion disturbances are also regarded as important causes of the reimplantation response. In this study, we investigated the effect of PGI, and SOD on reperfusion injury of the warm ischemic lung. We used a hilar-stripped lung model with additional warm ischemia instead of an autotransplantation model to make warm ischemic time equal. However, we believe that the purpose of this study concerning reperfusion injury was achieved by comparison with the control group.
Efects of Prostaglandin l2 and Superoxide Dismutase In the control group, deterioration of blood gases, an increase in pulmonary vascular resistance, an increase in the pulmonary extravascular water content, microcirculatory damage, and red cell sludging were noted immediately after operation. Arterial oxygen tension in the PGI, and SOD groups was well maintained even after reperfusion, and there was a significant difference between the control group and the PGI, and SOD groups at 1 and 2 hours after reperfusion. Pulmonary vascular resistance in the PGI, group did not increase after reperfusion, and a significant difference was noted between the PGI, group and the control group at 2 hours after reperfusion. From these phenomena, the microcirculation was probably improved by such actions of PGI, as the inhibition of platelet aggregation, relaxation of vascular smooth muscle, inhibition of white cell adhesion, and its cytoprotective effect [15-181. Furthermore, it was considered that tissue damage was probably reduced along with the improvement of blood gases through the prevention of an increase in capillary hydrostatic pressure by obviating a rise in pulmonary artery pressure. Also, the inhibition of chemical mediators such as thromboxanes and the cytoprotective
Ann Thorac Surg 1992;54:9214
action of PGI, would have had some effect. Although red cell sludging in the capillaries was prevented by the administration of heparin, an increase in pulmonary vascular resistance and the deterioration of blood gases were still noted. Accordingly, the heparin could not maintain pulmonary function after warm ischemia and hilar stripping of the lung. With the administration of SOD, as with PGI, pulmonary function was well maintained after warm ischemia, and the histological study showed that pulmonary edema was inhibited and the alveolar architecture remained nearly normal. These findings indicated that reperfusion injury was reduced by the administration of PGI, or SOD. However, because an increase in pulmonary vascular resistance was noted in the SOD group and also because microangiography did not visualize the smaller pulmonary arterioles, the action of SOD on the pulmonary circulation was considered to be less effective than that of PGI,.
References 1. Veith FJ, Koerner SK. The present status of lung transplantation. Arch Surg 1974;109:734-40. 2. Veith FJ, Kamholz SL, Mollenkoph FP, Montehsco CM. Lung transplantation 1983. Transplantation 1983;35:271-7. 3. Prop J, Ehrie MG, Crapo JD, Nieuwenhuis P, Wildevuur CRH. Reimplantation response in isografted rat lung. Analysis of causal factors. J Thorac Cardiovasc Surg 1984;87: 702-11. 4. Handa M, Fujimura S, Kondo 0, Nakata Y. Lung preservation. Jpn Ann Thorac Surg 1987;40:174-9. 5. Hitomi S, Wada H, Aoki M, et al. Lung transplantation. Jpn J Thorac Surg 1987;40:180-5. 6. Itinose T, Handa M, Hujimura S, et al. Warm ischemic time in lung transplantation. J Jpn Assoc Chest Surg 1987;l:lOl. 7. Toledo-Pereyra LH, Hau T, Simmons RL, Najarian JS. Lung preservation techniques. Ann Thorac Surg 1977;23:487-94. 8. Nishida T, Shibata H, Koseki M. Peroxidative injury of the mitochondria1 respiratory chain during reperfusion of hypothermic rat liver. Biochem Biophys Acta 1987;890:82-8. 9. Mead JF. Free radical mechanisms of lipid damage and consequences for cellular membranes. Free Radic Biol 1976; 1:51-68. 10. McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 1985;312:159-63. 11. Parks DA, Granger DN. Xanthine oxidase: biochemistry, distribution and physiology. Acta Physiol Scand 1986;548: 87-99. 12. Makino R, Tanaka T, Iizuka T, Ishimura Y, Kanegasaki S . Stoichiometric conversion of oxygen to superoxide anion during the respiratory burst in neutrophils. J Biol Chem 1986;261:11444-7. 13. Bannister JV, Bannister WH, Rotilio G. Aspects of structure, function and applications of superoxide dismutase. CRC Crit Rev Biochem 1987;22:111-80. 14. Braunwald E, Kloner RA. Myocardial reperfusion: a doubleedged sword. J Clin Invest 1985;76:1713-9. 15. Hirose T. Pulmonary edema. Gendaiiryo 1986;18:2642-51. 16. Hirose T. Respiratorv organ and prostaglandin. Rinsyoukagaku 1981;17:i047-54. 17. Czer GT. Marsh I. Konouka R, Moser KM. Low-dose PGI, prevents’ monociotaline-induced thromboxane production and lung injury. J Appl Physiol 1986;60:464-71. 18. Kambayashi J, Hatayama K, Kajiwara M, Sakon M, Oshiro T, Mori T. Mechanism of the cytoprotective effect of prostaglandin I, and its analogue in human platelets. Thromb Res 1986;44:4394.
-
A prostaglandin I, (PGI,) analogue and superoxide dismutase (SOD) were administered to dogs with pulmonary denervation, and their effects on warm ischemic damage to the lung were studied. Twenty-seven adult mongrel dogs were divided into a control group (6 dogs), a PGI, group (7 dogs), an SOD group (6 dogs), and a heparin group (8 dogs). The left pulmonary hilum was dissected, with PGI, (1pg/kg) being administered to the PGI, group and heparin (100U/kg) to the heparin group. Then the left lung was placed in a warm ischemic state for 1 hour. The SOD group also received 20 mg/kg of SOD intravenously 1 minute before reperfusion. Before warm ischemia, immediately after reperfusion, and 1 hour and 2 hours afterward, the blood gases, left pulmonary vascular resistance, and other data were measured under right pulmonary artery clamping. Arterial oxygen tension showed significantly better values in the SOD
and PGI, groups than in the control and heparin groups. The left pulmonary vascular resistance increased with time in the control group but did not increase in the PGI, group. Pulmonary microangiography showed that dilatation of the pulmonary arterioles was prominent in the PGI, group. The quantity of pulmonary extravascular fluid was significantly less in the PGI, and SOD groups than in the control and heparin groups. Histological examination showed marked collapse of capillaries, intraalveolar hemorrhage, and edema in the control and heparin groups, whereas these changes were only slight in the PGI, and SOD groups. Thus, both PGI, and SOD appeared to have a protective effect on the pulmonary microcirculation and tissue, and may be effective in preventing warm ischemic damage in lung transplantation. (Ann Thorac Surg 1992;54:9214)
I
effective against warm ischemic reperfusion disturbance, and we studied its role in a pulmonary denervation model.
n lung transplantation, the appearance of infiltrates in the transplanted lung is seen on postoperative chest roentgenograms along with a decrease in respiratory function and worsening of the blood gases, and this transient pulmonary impairment is called the reimplantation response. It would be desirable to distinguish these events from rejection, elucidate the underlying mechanism, and develop methods to prevent the occurrence of these early postoperative phenomena. Warm ischemic damage, denervation, and the interruption of lymphatics have all been considered as possible causes of the reimplantation response. Prostaglandin I, (PGI,) inhibits platelet aggregation, causes vasodilatation, is cytoprotective, and stabilizes lysosome membranes, so we considered that it might be possible to reduce the reimplantation response by using a PGI, analogue. In unilateral pulmonary transplantation, unlike cardiac transplantation, blood at a temperature of about 36°C enters the transplanted lung at the time of reperfusion, and damage from oxygen radicals is likely to occur. Therefore, the radical scavenger superoxide dismutase (SOD) might also be Accepted for publication March 6, 1992. Address reprint requests to Dr Yamashita, Second Division, Department of Surgery, Kobe University School of Medicine, 7-5-2 Kusunoki-cho Chuoku, Kobe, Japan 650.
0 1992 by The Society of Thoracic Surgeons
Material and Methods Twenty-seven adult mongrel dogs weighing 10 to 15 kg were divided into four groups: 6 dogs formed the control group, 7 dogs were in the PGI, group, 6 dogs formed the SOD group, and 8 dogs were in the heparin group. All animals received humane care in compliance with the ”Guide for the Care and Use of Laboratory Animals” published by the National Institute of Health (NIH publication No. 85-23, revised 1985). General anesthesia was induced with ketamine hydrochloride (Ketalar; ParkeDavis, Morris Plains, NJ; 10 mg/kg), thiamylal sodium (Yoshitomi Inc, Osaka, Japan; 20 mgkg), and pancuronium bromide (Mioblock; Organon Inc, West Orange, NJ; 1 mg/kg), and a volume-controlled ventilator was used to provide ventilation (inspired oxygen fraction, 0.60; internal airway pressure, 15 cm H,O; ventilation rate, 20/min). Next, an arterial line was inserted into the right carotid artery and a Swan-Ganz catheter was advanced to the pulmonary artery trunk through the jugular vein. Arterial and mixed venous blood gases were measured, and then thoracotomy was performed through the fourth left inter0003-4975/92/$5.00
922
Ann Thorac Surg 1992;54:9214
YAMASHITA ET AL REPERFUSION INJURY OF WARM ISCHEMIC LUNG
reo
1'
1 hr
2hr
control G (n = 6)
dard error. Statistical significance was determined by two-factorial analysis of variance and the t test. A p value less than 0.05 was considered to be significant.
Results
PGI, G (n = 7) SOD I -rep
..
SOD G (n = 6) heparin
rep
heparin G (n = 8 )
IiJOITJO
r t PA clamp for 5min to measure
blood gas. PApressure. PAflow
1 EVLW. PAgraphy. histology
Fig 1. Protocol of the experiment. Arrows indicate the time of administration of prostaglandin l2 (PGI,), superoxide dismutase (SOD), and heparin. Shaded squares show the time of measurement. (EVLW = extravascular lung water; 1-HS = left hilar stripping; PA = pulmonary artery; PAgraphy = pulmonary arteriography; rep = reperfusion; WIT = warm ischemic time.)
costal space. The left pulmonary artery, pulmonary vein, and main bronchus were dissected out, and after the nerves, bronchial arteries, and lymphatics were completely transected, the right and left pulmonary arteries were taped. During this procedure, a PGI, analogue (1pg/kg) was administered through the Swan-Ganz catheter in the pulmonary artery over a 30-minute period in the PGI, group. In the heparin group, heparin (100 Ulkg) was injected intravenously during dissection of the hilum. Next, an electromagnetic flow meter was attached to the pulmonary artery trunk, and the right pulmonary artery was clamped to evaluate hemodynamics in the left lung. Ventilation was performed at an inspired oxygen fraction of 1.0 for 5 minutes and then the blood gases, pulmonary artery pressure, and pulmonary arterial blood flow were measured to provide standard values. Thereafter, the right pulmonary artery was released and the left lung was deflated. Then the left pulmonary artery and vein and the main bronchus were clamped with a vascular clamp for 1 hour to produce warm ischemia of the left lung. In the SOD group, 20 mg/kg of SOD was injected intravenously 1 minute before reperfusion. Immediately after reperfusion as well as 1 hour and 2 hours afterward, measurements were performed to investigate lung function, and then the heart and lungs were excised. Extravascular fluid in the left lung was determined by the dry weight method in 5 animals from each group. In the remaining dogs of each group (1in the control group, 2 in the PGI, group, 1 in the SOD group, and 3 in the heparin group), 50% barium solution was injected from the pulmonary artery trunk into both lungs at a pressure of 30 cm H,O to perform soft roentgenography. The plates were enlarged 10 times for comparative study (microangiography). Finally, a histological study was performed (Fig 1). The results were expressed as a mean value and stan-
The following data were measured under clamping of the right pulmonary artery to evaluate left lung function after warm ischemia. Arterial oxygen tension decreased with time in both the control and heparin groups, but remained stable at 280 to 360 mm Hg in both the PGI, and SOD groups. Significant differences were noted between the PGI, and SOD groups and the control and heparin groups at 1 hour and 2 hours after reperfusion (Fig 2). Arterial carbon dioxide tension showed a slow increase with time in all four groups, but no significant difference was seen among them. The left pulmonary vascular resistance was high in all groups, and it was noted that the cardiac output was only about 1 L/min in small dogs weighing 10 to 15 kg. Before warm ischemia, left pulmonary vascular resistance was 1,800 to 2,200 dyne s * cm? in all four groups, but it increased sharply in the control and SOD groups after 2 hours of reperfusion to become 5,780 f 681 and 5,005 ? 1,459 dyne s cmp5, respectively. However, in the PGI, group it increased only slightly to 2,137 f 205 dyne s cmP5 immediately after reperfusion and to 2,748 & 347 dyne s * cmP5 after 2 hours. In the heparin group, left pulmonary vascular resistance increased moderately to 2,775 ? 261 dyne * s * cmP5 immediately after reperfusion and to 3,695 278 dyne * s * cmP5 after 2 hours. But significant differences were noted between PGI, group and control group only at 2 hours after reperfusion (Fig 3). The left extravascular lung water content was 6.99 mL/kg in the control group and 6.02 mLlkg in the heparin group. It was significantly lower in the PGI, and SOD groups, being 4.93 mL/kg and 4.36 mL/kg, respectively (Fig 4).
-
-
- -
-
*
control-G ( n = 6 ) (n = 7 ) (n=6) 0 heparin-G ( n = 8 ) 0
o PGln-G A SOD-G
O/ntubation before WI
after WI
1 hour
2 hour
Fig 2 . Arterial oxygen tension (Pa02) during occlusion of the right pulmonary artery before warm ischemia of the left lung, just after reperfusion, and 1 and 2 hours afterward. Significant difierences were noted between the prostaglandin l2 (PGI,) and superoxide dismutase (SOD) groups and the control group at 1 and 2 hours after reperfusion. (HS = hilar stripping; WI = warm ischemia.)
YAMASHITA ET AL REPERFUSION INJURY OF WARM ISCHEMIC LUNG
Ann Thorac Surg 1992;54:9214
923
0 control-G
(n = 6 ) (n = 7) (n=6) heparin-(; (n = 8)
o PG12-G ASODG
I-F'VR
(dyneis,ec/cm-?
* : P < 0.001
50004000-
30002000 -
intubation before WI
after WI
1 hour
2 hour
Fig 3. Left pulmonary vascular resistance 0-PVR) during occlusion of the right pulmonary artery before warm ischemia of the left lung, just after reperfusion, and 1 and 2 hours afterward. A significant difference was seen between the prostaglandin I , (PGI,) group and control group at 2 hours after reperfusion. (HS = hilar stripping; SOD = superoxide dismutase; WI = warm ischemia.)
In postmortem pulmonary microangiography, the contrast medium did not reach the periphery of the left pulmonary arterial vasculature in both the control and heparin groups, and occlusion of vessels of 100 to 200 pm in diameter was seen. The SOD group showed similar findings. In the PGI, group, however, the vascular tree was visualized clearly to a level of less than 100 pm in diameter, and little difference was noted when compared with the right lung, which had not been manipulated (Fig 5). In the control and heparin groups, alveolar hemorrhage, exudate, and interstitial edema were prominent on histologic examination. The capillaries were narrow and collapsed. Sludging in the capillaries was seen in the control group. In the SOD group, alveolar edema and hemorrhage were very slight and the alveolar architecture was almost normal. In the PGI, group, alveolar hemor-
Yd : P < 0.05
M E A N f SD
EVLW
Fig 5 . By postmortem pulmona y microangiography, the pulmonary vascular tree was visualized clearly to the level of less than 100 p m in the prostaglandin I, (PGI,) group, but contrast medium did not reach these small arterioles in the superoxide dismutase, control, and heparin groups.
4-
rhage and exudates were noted occasionally. Dilatation of the pulmonary arterioles was prominent, and a marked increase of blood flow was noted as far distally as the capillaries.
2-
Comment
6-
Cause of the Reimplantation Response n "
control-G PG1,-G
SOD-G heparin-G
Fig 4. The left extravascular lung water (EVLW) content: EVLW was significantly lower in the prostaglandin I, (PGI,) and superoxide dismutase (SOD) groups than in the control group.
The reimplantation response is a transient state of pulmonary edema occurring after transplantation, and warm ischemic reperfusion injury, chemical mediators, denervation, or operative manipulation are all regarded as possible causes [l-31. Warm ischemic reperfusion injury
924
YAMASHITA ET AL REPERFUSION INJURY OF WARM ISCHEMIC LUNG
includes cellular disturbances due to ischemia itself and other disturbances occurring at the time of reperfusion [4]. Because the lung contains air, unlike other organs, cellular function is maintained even after warm ischemia lasting for 2 to 3 hours [5-71, but interstitial edema, hemorrhage, and the destruction and swelling of mitochondria have been shown to develop after 60 minutes of warm ischemia [8]. Furthermore, because pulmonary edema becomes more marked with the increase in warm ischemic time, there is no doubt that warm ischemia is an important cause of the reimplantation response. When an organ undergoes warm ischemia, lipid peroxidation and the no-reflow phenomenon can occur. Free radicals are produced in the endothelial cell membranes and in neutrophils at the time of reperfusion and cause an increase in cell membrane permeability [9, 101. Superoxide dismutase is reported to scavenge free radicals produced by the hypoxanthine-xanthine oxidase system and by neutrophils, and to inhibit reperfusion injury by suppressing membrane lipid peroxidation [ll-131. The no-reflow phenomenon occurs when, after warm ischemia and reperfusion, the blood does not flow sufficiently from the arterioles to the capillaries due to sludging of red blood cells, obstruction by white blood cells, vasospasm, and narrowing of the vessel lumen due to edema of the vascular endothelium [14]. These reperfusion disturbances are also regarded as important causes of the reimplantation response. In this study, we investigated the effect of PGI, and SOD on reperfusion injury of the warm ischemic lung. We used a hilar-stripped lung model with additional warm ischemia instead of an autotransplantation model to make warm ischemic time equal. However, we believe that the purpose of this study concerning reperfusion injury was achieved by comparison with the control group.
Efects of Prostaglandin l2 and Superoxide Dismutase In the control group, deterioration of blood gases, an increase in pulmonary vascular resistance, an increase in the pulmonary extravascular water content, microcirculatory damage, and red cell sludging were noted immediately after operation. Arterial oxygen tension in the PGI, and SOD groups was well maintained even after reperfusion, and there was a significant difference between the control group and the PGI, and SOD groups at 1 and 2 hours after reperfusion. Pulmonary vascular resistance in the PGI, group did not increase after reperfusion, and a significant difference was noted between the PGI, group and the control group at 2 hours after reperfusion. From these phenomena, the microcirculation was probably improved by such actions of PGI, as the inhibition of platelet aggregation, relaxation of vascular smooth muscle, inhibition of white cell adhesion, and its cytoprotective effect [15-181. Furthermore, it was considered that tissue damage was probably reduced along with the improvement of blood gases through the prevention of an increase in capillary hydrostatic pressure by obviating a rise in pulmonary artery pressure. Also, the inhibition of chemical mediators such as thromboxanes and the cytoprotective
Ann Thorac Surg 1992;54:9214
action of PGI, would have had some effect. Although red cell sludging in the capillaries was prevented by the administration of heparin, an increase in pulmonary vascular resistance and the deterioration of blood gases were still noted. Accordingly, the heparin could not maintain pulmonary function after warm ischemia and hilar stripping of the lung. With the administration of SOD, as with PGI, pulmonary function was well maintained after warm ischemia, and the histological study showed that pulmonary edema was inhibited and the alveolar architecture remained nearly normal. These findings indicated that reperfusion injury was reduced by the administration of PGI, or SOD. However, because an increase in pulmonary vascular resistance was noted in the SOD group and also because microangiography did not visualize the smaller pulmonary arterioles, the action of SOD on the pulmonary circulation was considered to be less effective than that of PGI,.
References 1. Veith FJ, Koerner SK. The present status of lung transplantation. Arch Surg 1974;109:734-40. 2. Veith FJ, Kamholz SL, Mollenkoph FP, Montehsco CM. Lung transplantation 1983. Transplantation 1983;35:271-7. 3. Prop J, Ehrie MG, Crapo JD, Nieuwenhuis P, Wildevuur CRH. Reimplantation response in isografted rat lung. Analysis of causal factors. J Thorac Cardiovasc Surg 1984;87: 702-11. 4. Handa M, Fujimura S, Kondo 0, Nakata Y. Lung preservation. Jpn Ann Thorac Surg 1987;40:174-9. 5. Hitomi S, Wada H, Aoki M, et al. Lung transplantation. Jpn J Thorac Surg 1987;40:180-5. 6. Itinose T, Handa M, Hujimura S, et al. Warm ischemic time in lung transplantation. J Jpn Assoc Chest Surg 1987;l:lOl. 7. Toledo-Pereyra LH, Hau T, Simmons RL, Najarian JS. Lung preservation techniques. Ann Thorac Surg 1977;23:487-94. 8. Nishida T, Shibata H, Koseki M. Peroxidative injury of the mitochondria1 respiratory chain during reperfusion of hypothermic rat liver. Biochem Biophys Acta 1987;890:82-8. 9. Mead JF. Free radical mechanisms of lipid damage and consequences for cellular membranes. Free Radic Biol 1976; 1:51-68. 10. McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 1985;312:159-63. 11. Parks DA, Granger DN. Xanthine oxidase: biochemistry, distribution and physiology. Acta Physiol Scand 1986;548: 87-99. 12. Makino R, Tanaka T, Iizuka T, Ishimura Y, Kanegasaki S . Stoichiometric conversion of oxygen to superoxide anion during the respiratory burst in neutrophils. J Biol Chem 1986;261:11444-7. 13. Bannister JV, Bannister WH, Rotilio G. Aspects of structure, function and applications of superoxide dismutase. CRC Crit Rev Biochem 1987;22:111-80. 14. Braunwald E, Kloner RA. Myocardial reperfusion: a doubleedged sword. J Clin Invest 1985;76:1713-9. 15. Hirose T. Pulmonary edema. Gendaiiryo 1986;18:2642-51. 16. Hirose T. Respiratorv organ and prostaglandin. Rinsyoukagaku 1981;17:i047-54. 17. Czer GT. Marsh I. Konouka R, Moser KM. Low-dose PGI, prevents’ monociotaline-induced thromboxane production and lung injury. J Appl Physiol 1986;60:464-71. 18. Kambayashi J, Hatayama K, Kajiwara M, Sakon M, Oshiro T, Mori T. Mechanism of the cytoprotective effect of prostaglandin I, and its analogue in human platelets. Thromb Res 1986;44:4394.
-
