Administration of prostaglandin E1 after lung transplantation improves early graft function

Administration of Prostaglandin E1 After Lung Transplantation Improves Early Graft Function Motoi Aoe, MO, Gregory D. Trachiotis, MO, Kan Okabayashi, MO, Jill K. Manchester, BS, Oliver H. Lowry, MO, PhD, Joel D. Cooper, MO, and G. Alexander Patterson, MO Division of Cardiothoracic Surgery, Department of Surgery, and Department of Pharmacology, Washington University School of Medicine, Barnes Hospital, St. Louis, Missouri

Early graft dysfunction continues to be a major clinical problem after lung transplantation. The objective of this experiment was to determine whether continuous administration of prostaglandin EI (PGE I) after lung transplantation has a beneficial effect on early graft function. Left lung allotransplantation was performed in 10 sizematched mongrel dogs (weight, 24.4 to 31.4 kg). Lung preservation consisted of a bolus injection of PGEI (250 p.g) into the pulmonary artery, followed by a pulmonary artery flush with 50 mL/kg of 4°C modified Euro-Collins solution. The lungs were then stored at rc for 12 hours. Left lung transplantation was performed using standard technique. The right pulmonary artery and right bronchus were ligated prior to chest closure. Animals were placed in the supine position and ventilated for 6 hours with 100% oxygen at a rate of 20 breaths/min, a tidal volume of 550 mL, and a positive end-expiratory pressure of 5 em H 2 0 . Animals were randomly allocated to one of two groups. Group 1 animals (n = 6) received continuous PGEI infusion from the onset of implantation. The dose was gradually increased and fixed when mean systemic pressure showed a 10%

decrease (mean PGEI dose, 31.7 ± 6.9 ng' kg-I. min-I). Group II animals (n = 4) received no PGEl" After the 6-hour assessment period, arterial oxygen tension and alveolar-arterial oxygen pressure difference were preserved in group I compared with group II (group I versus group II: arterial oxygen tension, 255.8 ± 37.6 mm Hg versus 64.7 ± 7.9 mm Hg [p < 0.05]; alveolar-arterial oxygen pressure difference, 411.1 ± 70.5 mm Hg versus 597.5 ± 1.3 mm Hg [p < 0.05]). There were no significant differences in pulmonary circulatory hemodynamics between the two groups. Wet to dry lung weight ratio and total volume of airway edema fluid were also significantly less in group I than in group II (group I versus group II: wet to dry ratio, 8.2 ± 0.9 versus 12.1 ± 0.7 [p < 0.01]; edema fluid, 106.7 ± 38.6 mL versus 375.0 ± 56.2 mL [p < 0.01]). There was no difference in lung myeloperoxidase activity between the two groups. We conclude that PGEI significantly improved early lung function after transplantation and that this improvement is not due to pulmonary vasodilatation, improved pulmonary circulation, or inhibition of leukocyte activation. (Ann Thorae Surg 1994;58:655-61)

A

ameliorated ischemic-reperfusion injury in an in-situ warm ischemic rabbit lung model. The beneficial effect did not appear to be due to the vasodilating properties of PCEI or to the known effects of PCEI on platelets, neutrophils, and cytokines. The purpose of this study was to determine whether administration of PCE] improves canine lung allograft and pulmonary circulatory hemodynamics during the early reperfusion period after prolonged hypothermic preservation.

lthou g h the performance of successful lung transplantation has become more common [1], early graft dysfunction remains a major clinical problem. Superior preservation strategies have lessened the incidence of allograft dysfunction. Early lung allograft dysfunction can occur unpredictably after satisfactory preservation, a short period of ischemia, and a technically satisfactory operation. This clinical entity is characterized by persistent moderate elevation in pulmonary artery (PA) pressure, pulmonary edema, and hypoxemia. Currently, prostaglandin E] (PCE]) or prostacyclin is used clinically at the time of donor lung flush and harvest [1, 2]. The vasodilatory effects of these agents are thought to improve pulmonary flush. lt is possible that the agents have some beneficial effect in reducing reperfusion injury and improving allograft function after transplantation. Matsuzaki and associates [3] recently reported that PCE] Presented at the Poster Session of the Twenty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 25-27, 1993. Address reprint requests to Dr Patterson, Division of Cardiothoracic Surgery, Washington University School of Medicine, Suite 3108, Queeny Tower, One Barnes Hospital Plaza, St. Louis, MO 63110.

© 1994 by The Society of Thoracic Surgeons

Material and Methods All animals received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Cuide 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 85-23, revised 1985).

Donor Procedure Ten adult mongrel dogs underwent lung harvest. The donor procedure was performed in the same manner as 0003-4975/94/$7.00

656

AOE ET AL PGE, IN LUNG TRANSPLANTATION

previously described [4]. Briefly, the dogs were anesthetized with sodium thiopental (10 mg/kg intravenously) followed by atropine sulfate (0.04 mg/kg intramuscularly) and were intubated with a 9F endotracheal tube. The lungs were ventilated with a tidal volume of 750 ml., a respiratory rate of 12 breaths/min, an inspired oxygen fraction of 1.0, and no positive end-expiratory pressure (Bennett MAl respirator; Puritan Bennett, Inc, Overland Park, KS). Systemic arterial pressure was monitored through a femoral artery cannula. After a median sternotomy, the thymus was removed and the azygos vein, divided. At this point, a tiny section of the donor right lung was excised as a control sample for tissue myeloperoxidase (MPO) assays (described later). Both venae cavae, the aorta, the PA, and the trachea were isolated. Heparin sodium (400 U /kg) was administered systemically. A 6.5-mm curved, metal-tipped perfusion cannula (Sames, Inc, Ann Arbor, MI) was inserted into the main PA just distal to the valve and connected by tubing to a bag of cold (1°C) modified (4 mmol/L of MaS04 and 32.7 gil of glucose) Euro-Collins solution, which was hung at a height of 40 ern above the heart. After four 1,600-mL tidal volume breaths were administered to avoid atelectasis, a I-minute wait allowed the lungs to reach the previous ventilatory state. Just prior to inflow occlusion, a bolus injection of 250 JLg of PGE 1 (Prostin VR Pediatric; The Upjohn Company, Kalamazoo, MI) was administered directly into the trunk of the P A. After a IS-second wait, cardiac inflow occlusion was accomplished by ligation of both venae cavae. The left atrial appendage was amputated, and the aorta was opened to decompress the left atrium and bronchial circulation during PA flush. The lungs were then flushed with 50 mL/kg of modified Euro-Collins solution. Topical cooling was achieved at the same time by immersing the lungs in cold (1° to 4°C) saline solution. At the completion of the flush, the trachea was clamped at end-inspiration. The heart-lung block was excised, placed in a plastic bag containing cold (10c) modified Euro-Collins solution, and preserved at 1°C for 12 hours before implantation.

Recipient Procedure Left single-lung implantation was performed in the usual fashion. All 10 adult mongrel dogs were anesthetized in the same manner as the donors and ventilated with an adjustable-rate Harvard pump respirator (model 613; Harvard Apparatus, South Natick, MA) with 98.5% oxygen and 1.5% halothane. Systemic, main PA, and central venous pressures were obtained through a femoral artery cannula and a Swan-Ganz catheter by means of a transducer (model 1290A; Hewlett-Packard, Andover, MA). A thoracotomy was performed in the left fifth intercostal space, and a left pneumonectomy was done. At this time, each dog was randomly assigned to one of two study groups. In group I (n = 6), systemic administration of PGE I was started by continuous syringe pump infusion (Orion Sage pumps model 355; Orion Research Inc, Boston, MA) after completion of the left pneumonectomy. The infusion was maintained throughout implantation and the subsequent assessment period. The initial

Ann Thorac Surg 1994;58:655-61

dose (15 ng· kg- 1 • min -1) was gradually increased and fixed when a 10% decrease in systemic pressure was observed. This dose adjustment was completed prior to initial graft reperfusion. In group II (n = 4), no PGE 1 was used. The left lung of the donor was trimmed from the heart-lung block in a basin containing cold saline solution. The left lung was covered with cold gauze during implantation. The left atrial artery, P A, and bronchus were anastomosed consecutively. A continuous everting mattress suture was used for the left atrial anastomosis and continuous over-and-over suture, for the other two anastomoses. After completion of the anastomoses, left lung reperfusion was initiated. The right upper and lower bronchi were ligated separately so as to avoid distortion of the left main bronchus. Occlusion was confirmed with a fiberoptic bronchoscope. The right PA was also ligated. Two chest tubes were inserted and connected to negativesuction (-15 em H 20) closed drainage. The chest was then closed.

Assessment of Lung Function After chest closure, the dogs were placed in the supine position. The respirator was changed to a Bennett MAl, and the transplanted left lung was ventilated with a tidal volume of 550 ml., a respiratory rate of 20 breaths/min, an inspired oxygen fraction of 1.0, and 5 ern H 20 of positive end-expiratory pressure. The level of anesthesia was maintained with periodic intravenous administration of thiopental during the assessment. Systemic, PA, and central venous pressures were continuously recorded. Arterial and central venous blood gases were examined every 15 minutes from the start of the assessment. Cardiac output was determined hourly in triplicate by the thermodilution method, and pulmonary wedge pressure was obtained by inflating the balloon of the Swan-Ganz catheter, which was temporarily advanced peripheral to the left PA anastomosis. The assessment period covered 6 hours. During the 6-hour period, blood pH was normalized with administration of bicarbonate, and venous infusion volume of lactated Ringer's solution was regulated based on central venous pressure. Edema fluid from the left lung was suctioned using a fiberoptic bronchoscope 10 minutes before each assessment as necessary, and the total volume of edema fluid during the entire assessment period was measured. After the 6-hour period, the animals were sacrificed, and both lungs were excised for pathologic study, tissue MPO assay, and wet to dry lung weight ratio measurement.

Measurement of Tissue Myeloperoxidase Activity A lung tissue sample ranging from 0.2 to 0.4 g for MPO assay was obtained from the donor right upper lobe during harvest and from the recipient right and left middle lobes after the 6-hour assessment. Each sample was immediately quick frozen by immersing it in dichlorodifluoromethane (CCI2F2 ) that had been precooled to the freezing point. Samples were subsequently stored at -70°C until assay [5]. The MPO activity was measured in lung tissue by a modified method previously described by Henson and

AOE ET AL PGE I IN LUNG TRANSPLANTATION

Ann Thorac Surg 1994;58:655-61

657

Table 1. Characteristics of Experimental Groups"

(s)

Flushing Pressure (mm Hg)

Harvest Time b (min)

Preservation Time" (h)

Implantation Timed (min)

Operation Time" (h)

Intravenous Fluid Replacement During Operation (mL)

113 ± 15 90 ± 10

16.3 ± 1.7 17.0 ± 1.1

9 ± 1.1 11 ± 1.5

12:19 ± 0:06 12:05 ± 0:09

63.2 ± 3.7 63.2 ± 2.2

2:31 ± 0:04 2:42 ± 0:06

1,000 ± 160 875 ± 125

Flushing Time Group I II

Data are shown as the rnean z; the standard error of the mean. b Harvest time was from inflow occlusion to storage. c Preservation time was from inflow occlusion to reperfusion. d Implantation time was from start of left atrial anastomosis to reperfusion. e Operation time was from initial skin incision to first assessment.

a

co-workers [6]. Briefly, 100 mg of frozen lung tissue was homogenized in 1 mL of 0.5% HTAB, 5 mmol/L EDTA (ethylenediaminetetraactic acid), and 50 mmol/L potassium phosphate buffer (pH 6.2) with a Broeck tissue grinder (Kontes Glass Co, Vineland, NJ). The homogenate was centrifuged at 10,000 g for 15 minutes at 4°C. The supernatant was subsequently assayed for total soluble protein by the method of Pierce Laboratories [7]. The remaining supernatant was diluted lO-fold in an extraction buffer and assayed for MPO activity. Five microliters of the diluted extract was combined with 0.6 mL of Hanks' bovine serum albumin (0.25% bovine serum albumin added to Hanks' solution), 0.5 mL of 100 mmol/L potassium phosphate buffer (pH 6.2), 0.1 mL of 0.05% HzO z, and 0.1 mL of 1.25 mg/mL o-dianisidine. Color development was stopped after 5 minutes and 20 minutes at room temperature by adding 0.1 mL of 1% NaN 3 , and the optical density was measured at 460 nm with a spectrophotometer (model PMQ II; Carl Zeiss, Germany). The color development from 5 minutes to 20 minutes was linear and directly proportional to the amount of extract added. One unit of enzyme activity was defined as the amount of MPO that produced an absorbance change of 1.0 optical density unit per minute per milligram of tissue protein at room temperature.

Statistical Analysis All data are presented as the mean ± the standard error of the mean. One-way analysis of variance with repeated measures was used to determine whether an overall difference existed in graft function between the two groups during the assessment. When a difference was obtained, contrast was performed to determine where significant differences arose. The other data were analyzed with one-way analysis of variance. Significance was accepted at the 95% confidence limit (p < 0.05).

plantation times, operation times, and total intravenous fluid replacement volumes during implantation and assessment between groups (Table 1). The mean time from reperfusion to initial assessment was 15.6 ± 1.2 minutes in group I and 14.7 ± 1.7 minutes in group II. All group I animals tolerated the 6-hour assessment period. In group II, 1 animal sustained hypoxic cardiac arrest 5 hours after reperfusion. Thus, for group II, gas exchange and hemodynamic data are available for 5 hours for 4 animals and for 6 hours for only 3 animals. Wet to dry lung weight ratio and tissue MPO activity data from both lungs were obtained from all experimental animals and analyzed.

Gas Exchange During Assessment During the first 45 minutes of reperfusion, there were no significant differences between the two groups with respect to arterial oxygen tension (PaOz), alveolar-arterial oxygen pressure difference (P[A-a]Oz), arterial carbon dioxide pressure, and intrapulmonary shunt. However, gas exchange deteriorated rapidly in group II, and significant differences were apparent between the two groups after 90 minutes of assessment (group I versus group II: PaOz, 446.3 ± 37.9 mm Hg versus 221.2 ± 41.4 mm Hg; P(A-a)Oz, 227.8 ± 53.3 versus 444.1 ± 42.9 mm Hg; intrapulmonary perfusion shunt: 13.3% ± 2.3% versus 29.3% ± 3.6% [p < 0.05]) (Fig 1). During the remainder of the 6-hour reperfusion, gas exchange in group II continued to deteriorate. The PaOz and P(A-a)Oz in group I remained significantly better than in group II (group I versus group II at 3-hour assessment: PaO z, 372.2 ± 37.8 mm Hg versus 122.8 ± 32.6 mm Hg; P(A-a)Oz, 300.1 ± 65.9 mm Hg versus 538.8 ± 35.9 mm Hg [p < 0.0ll; at 6-hour assessment: PaOz, 255.8 ± 37.6 mm Hg versus 64.7 ± 7.9 mm Hg; P(A-a)Oz, 411.1 ± 70.5 mm Hg versus 597.5 ± 1.3 mm Hg [p < 0.05]) (see Fig 1).

Results Twenty adult mongrel dogs were paired by weight, and the 10 recipients were randomly assigned to one of the two groups. The mean weight of donor dogs was 26.8 ± 3.5 kg in group I (n = 6) and 25.5 ± 2.6 kg in group II (n = 4). The mean weight of recipient dogs was 25.4 ± 0.7 kg in group I and 25.9 ± 1.2 kg in group II. There were no significant differences in flushing times, flushing pressures, harvest times, preservation times, im-

Hemodynamic Data Aortic pressure in group I was lower than that in group II at several assessment points. Mean PA pressures in both groups fell gradually from the initial assessment and became stable after the 90-minute assessment. There were no significant differences between the two groups in mean PA pressure, central venous pressure, cardiac index, and pulmonary vascular resistance (Fig 2).

658

I

0'"

=-= '@

1

cS Q

=

-< I

'@

1

cS

U

=-=

...

AOE ET AL PGE I IN LUNG TRANSPLANTATION

:1 200

0

•

t

~

t t

t ,

p
••. •

"'o-~~-6.o

• t t =!L.&:-...... • t · .......... t

tttt

ttt •••••

t t t •••••• •••••••••••

~~ • t t t t t t t t

20 0

NS

p
NS

-o-Groupl --Group II

o

01 -.. '" 01

....·:v;-~:a;;::gr~::'ZiZia1GC¥QNol:t

~W· ~~h:2.-91"9Y:1.~ &::£.:z:;t..::1.."!;t:. t t t • • • •

T:

4°j~~f

40

"t7"~

~J. t

~.:*
W

;;?

~

t

0

=~

1994;58:655-61

"O".:o~~~ ~ :bit .!.')tl -0;.':6;

200

400

Ann Thorae Surg

m.

p
.j-.;i:1J".i~ 4~ . ~...T.:~~

~~~~~~~~~~

=> U

"~·~'''-'''''-'fl-'rt'l::!::!:!:!. NS

U

-
~i~~~~~'1.o;Q.~~~~ * *

o

60

1lI

t

120

*

180

TIME (minutes)

240

300

360

.: p
Fig 1. Gas exchange during 6-hour assessment after ligation of right bronchus and pulmonary artery. There were significant differences in arterial oxygen tension (P(A-a)02)' alveolar-arterial oxygen pressure difference (A-aD02), and intrapulmonary perfusion shunt (Qs/Qt) between the two groups. (NS = not significant.)

Wet to Dry Lung Weight Ratio After the 6-hour assessment, the wet to dry lung weight ratios of the transplanted left lungs in group II were significantly higher than those in group I (group I versus group II: 8.2 ±: 0.9 versus 12.1 ±: 0.7; P < 0.01). The right lungs in both groups showed a minimal gain in this ratio (group I and group II: 5.2 ±: 0.9 and 5.7 ±: 0.6). There were no significant differences between transplanted left lungs and native right lungs in group I (Fig 3).

Total Volume of Airway Edema Fluid All group II animals had massive airway edema fluid that required periodic suction. Four animals in group I had no edema fluid from the transplanted left lung throughout the 6-hour assessment period. The total volume of airway edema fluid suctioned during the 6-hour assessment period in group II was significantly greater than that in group I (group I versus group II: 106.7 ±: 38.6 mL versus 375.0 ±: 56.2 mL; p < 0.01) (Fig 4).

Gross Appearance of Lungs at Postmortem Examination The transplanted left lung in both groups appeared slightly hyperinflated and edematous. In particular, the lower posterior portion of the lungs in both groups had severe pulmonary edema formation. The area of edema formation in group II looked much larger than that in group I, but no quantitative comparisons were performed in this study.

I

o

60

120

I

I

I

I

180

240

300

360

TIME (minutes) *: p
Each native right lung in both groups was completely atelectatic.

Histologic Findings After the 6-hour assessment, transplanted left lungs in group II typically had a greater degree of alveolar edema I

[]llJlJ Lt. Transplanted Lungj

, CJ

15

Rt. Native Lung

.S

~ :=

.~ 10 ~

-

p
~

~ ~

CJ)

5

= =

'""

o Group I

Group II

Fig 3. Wet to dry lung weight ratio after 6-hour assessment. After the assessment, the wet to dry weight ratio of the transplanted left (Lt.) lungs in group II was significantly greater than that in group I. (Rt, = righi.)

AOE ET AL PGE, IN LUNG TRANSPLANTATION

Ann Thorae Surg 1994;58:655-61

mt

l

·-.~ -~ft;~~~~lant~-lung --,

unit/mg 0.5

400

...

659

~ Right native lung -_~

.. - - - - , -

0.4

Ei

300 -a=

. =200

c 0;:

"Cl

oS

;:

.<

os

o ~

Ei ...

...

"Cl

...e

'i 100

0.3 0.2

0.1

Eo-<

0

0.0

Group I

Group II

Fig 4. Total volume of airway edema fluid suctioned during assessment. The volume of suctioned pulmonary edema fluid in group II was greater than that in group I during the 6-hour period of reperfusion.

and thickened alveolar septa (Fig 5). However, we were unable to quantitate any significant difference in histologic appearance between the two groups.

Control

Group I

Group II

Fig 6. Tissue myeloperoxidase (MPO) activity. After the 6-hour assessment, tissue MPO activity in the right and left lungs increased in both groups. There was no significant difference (NS) between groups.

Tissue Myeloperoxidase Activity The tissue MPO activity of the transplanted left and native right lungs in both groups after the 6-hour assessment period were significantly higher than in normal control lung biopsy specimens obtained from donor right lungs immediately after thoracotomy. There were no significant differences in the transplanted left lungs between the groups
Comment

A

B

Fig 5. Histologic findings. (A) Transplanted left lung in group I after 6-hour assessment. Minimal edema with preservation of alveolar architecture is evident. (B) Transplanted left lung in group II after 6-hour assessment. lntraalveolar edema and thickened alveolar septa are apparent. (Both, X 100 before 51 % reduciion.)

In recent years, lung transplantation has become an established treatment for end-stage lung disease [8]. However, in the early postoperative period, graft dysfunction manifested by a persistent increase in pulmonary vascular resistance and a decrease in arterial oxygenation because of pulmonary edema frequently develops [9]. Numerous studies [4, 10, 11] have examined strategies to improve lung preservation so as to obtain better graft function after transplantation and better overall morbidity and mortality with lung transplantation. A number of specific strategies have been developed in an attempt to reduce ischemiareperfusion injury after lung transplantation, eg, leukocyte depletion [12, 13], oxygen radical scavengers [14-16], and some inhibition of lipid peroxidation [17]. Currently, many clinical lung transplantation programs use PGE 1 as a vasodilator prior to donor lung flush and harvest [2, 18]. Prostaglandins are products of arachidonic acid metabolism and have a wide variety of biologic effects in addition to vasodilation [19] that potentially could reduce clinical lung ischemia-reperfusion injury. Long and Rubin [20] reported that PGE 1 inhibits aggregation of platelets and leukocytes in pulmonary hypertension. Prostaglandin E1 also suppresses tumor necrosis factor production from macrophages [21], inhibits T-lymphocyte cytotoxicity and mitogenesis [22], diminishes production of interleukin-I [23], and prolongs the survival of allografts in

660

AOE ET AL PGE I IN LUNG TRANSPLANTATION

rat renal and cardiac transplantation [24, 25]. In addition, a "direct cytoprotective" effect of PGEl has been reported [26], but this effect is not understood completely. Several studies have reported amelioration of pulmonary ischemic injury by prostaglandin administration. Hooper and colleagues [27] demonstrated that prostacyclin reduces ischemic lung injury after 60 minutes of in situ warm ischemia in a canine model. Recently, Matsuzaki and co-workers [3] reported that there is a marked beneficial effect without any change in hemodynamics when PGEI is administered during reperfusion in an in situ rabbit lung model subjected to 2 hours' warm ischemia. These authors also measured blood platelet count, tissue MPO activity, and plasma tumor necrosis factor activity. They speculated that the improvement in graft function by PGEI administration after transplantation was not explained by its vasodilating properties and was not related to the known effects of PGEI on platelets, neutrophils, and cytokines but rather was due to its "cytoprotective" effect. In the present study, we demonstrated improved Pa0 2 and P(A-a)02 in animals receiving PGEI during a 6-hour posttransplantation reperfusion period. We suspect that the sporadic differences in shunt were caused by the intrapulmonary shunt created by edema fluid in the alveolar space, which was difficult to suction entirely. Despite the continuous infusion of PGEI and the significantly better gas exchange in group I, there were no differences between groups in pulmonary circulatory hemodynamics, particularly pulmonary vascular resistance. Matsuzaki and associates [3] also found no difference in pulmonary vascular resistance between PGEI and control groups. This may be explained by a variable pulmonary vasodilator response to PGEI and the consequently large standard errors. In the current study, aortic pressure dropped in both groups after ligation of the right P A. Then it gradually recovered to its previous level in group II but remained low in animals receiving PGEI. However, variability in aortic pressure in both groups accounts for the statistical difference at only several time points. The central venous pressure in group II tended to be lower than that in group I, perhaps because of volume loss from the airway as edema fluid, but there were no significant differences between the two groups. Thus, these results support the conclusion that administration of PGEI after transplantation maintains early graft function even at doses insufficient to effect any changes in pulmonary circulatory hemodynamics. It is widely known that neutrophils play an important role in ischemia-reperfusion injury [28, 29] and that PGEI affects neutrophils. In the present study, we measured tissue MPO activity to quantitate pulmonary neutrophil sequestration that may be a consequence of the effects of PGEI on cell-mediated factors in ischemic-reperfusion injury and to determine whether PGEI prevents lung graft reperfusion injury by affecting leukocytes. Significant increases in MPO activity were found in the transplanted reperfused left lungs in both groups and even in the native nonreperfused right lungs in both groups. This increase in MPO activity in the right lungs was not seen in the study by Matsuzaki and colleagues [3]. In the present study, it

Ann Thorac Surg 1994;58:655-61

may have been caused by longer assessment times, total atelectasis of the right lungs, and persistent bronchial artery-pulmonary venous circulation in our model. However, despite the improved pulmonary graft function in group I, there was no difference in MPO activity in the transplanted reperfused left lungs. This finding suggests that PGEI did not lessen lung injury by inhibition of cell-mediated factors, ie, leukocyte aggregation. On the basis of these results, we speculate that PGEI improves posttransplantation lung function mainly because of its "cytoprotective" effects, as reported by Rovert [26]. In the present study, we tried to create an experimental harvest and preservation technique similar to that employed in clinical lung transplantation. A bolus PGEl injection prior to donor lung flushing with modified EuroCollins solution and cooling of the graft to about 1°C with ice slush during transportation are now widely accepted as standard clinical technique [2]. We have evaluated the use of PGEI in an acute animal model of lung transplantation. The ligation of the contralateral PA and bronchus created an overperfused state in the transplanted lung, which is similar to the clinical situation of single-lung transplantation for pulmonary hypertension or like the situation in the first lung while the second lung is being extracted and implanted during sequential bilateral single-lung transplantation. The model permits lengthy assessment of allograft function alone. The 6-hour duration of the assessment was considered appropriate based on our previous work on lung reperfusion injury in a rabbit paracorporeal circulation model [30] in which significant alterations in pulmonary function had occurred by that time. It is possible that our measurement of pulmonary wedge pressure with a Swan-Ganz catheter after pneumonectomy affects cardiac output and may be misleading [31]. Animals were closely monitored to avoid a change in hemodynamics during inflation of the Swan-Ganz catheter balloon. At the time of measurement of pulmonary wedge pressure, the Swan-Ganz catheter was slowly advanced to the periphery until a blunt pressure wave was obtained, and then its balloon was inflated. If systemic pressure changed at that point, the balloon was deflated and the catheter advanced further to a point that was peripheral enough so that balloon inflation did not affect systemic circulation. In regard to the ventilatory setting during the assessment, the most important concern was stabilization of blood pH. Excessive acidosis can cause PA spasms and can affect graft function. Correcting the pH solely with bicarbonate sometimes requires massive infusions and makes graft function worse. We believe that a tidal volume of 550 mL and 20 breaths/min provide a certain degree of hyperventilation for a single left lung, but we found these settings essential to avoid hypercarbia and minimize bicarbonate infusion. We conclude that administration of PGEI after transplantation improved early postoperative lung function and that this improvement cannot be attributed to pulmonary vasodilatation, improved pulmonary circulation, or inhibition of leukocyte activation.

Ann Thorae Surg 1994;58:655-61

Supported by National Institutes of Health grants 1 ROI HL41281 and ROI HL41943. We acknowledge the expert technical assistance of Donna Marquart, Timothy Morris, Dennis Gordon, Steven Labarbera, and Duane Probst and the secretarial support of Mrs. Dawn Schuessler. Preservation solutions were provided by Gary Queensen, RPh. Statistical advice was obtained from Richard B. Schuessler, PhD.

References 1. Patterson GA, Cooper JD, eds, Lung transplantation. Chest Surg Clin North Am 1993;3. 2. Sundaresan S, Trachiotis GD, Aoe M, Patterson GA, Cooper JD. Donor lung procurement: assessment and operative technique. Ann Thorac Surg 1994;56:1409-13. 3. Matsuzaki Y, Waddell TK, Puskas JD, et al. Amelioration of post-ischemic lung reperfusion injury by prostaglandin E1. Am Rev Respir Dis 1993;148:882-9. 4. Okabayashi K, Aoe M, DeMeester SR, Cooper JD, Patterson GA. Pentoxifylline reduces lung allograft reperfusion injury. Ann Thorac Surg 1994;58:50-6. 5. Mazzone RW, Durand CM, West JB. Electron microscopic appearances of rapidly frozen lung. J Microsc 1979;117: 269-84. 6. Henson PM, Zanolari B, Schwartzman NA, Hong SR. Intracellular control of human neutrophil secretion. I. C5a-induced stimulus-specific desensitization and the effects of cytochalasin. Br J Immunol 1978;121:851-5. 7. Smith PK, Krohn RI, Hemanson GT, et al. Measurement of protein using bicinchoninic acid. Anal Biochem 1985;150: 76-85. 8. International Lung Transplantation Registry: Washington University, St. Louis, MO, April 1993. 9. Pasque MK, Kaiser LR, Dresler CM, Trulock E, Triantafillou AN, Cooper JD. Single lung transplantation for pulmonary hypertension. Technical aspects and immediate hemodynamic results. J Thorac Cardiovasc Surg 1992;103:475-81. 10. Mayer E, Puskas JD, Cardoso PF, Shi S, Slutsky AS, Patterson GA. Reliable eighteen-hour lung preservation at 4°C and 10°C by pulmonary artery flush after high-dose prostaglandin El administration. J Thorac Cardiovasc Surg 1992;103:1136-42. 11. Puskas JD, Cardoso PF, Mayer E, Shi S, Slutsky AS, Patterson GA. Equivalent eighteen-hour lung preservation by pulmonary artery flush with low potassium dextran or Euro-Collins solution after prostaglandin El infusion. J Thorac Cardiovasc Surg 1992;104:83-9. 12. Breda MA, Hall TS, Stuart S, et al. Twenty-four hour lung preservation by hypothermia and leukocyte depletion. J Heart Transplant 1985;4:325-9. 13. Pillai R, Bando K, Schueler S, Zebley M, Reitz BA, Baumgartner WA. Leukocyte depletion results in excellent heart-lung function after 12 hours of storage. Ann Thorac Surg 1990;50: 211-4.

AOE ET AL PGE, IN LUNG TRANSPLANTATION

661

14. Detterbeck FC, Keagy BA, Paull DE, Wilcox BR. Oxygen free radical scavengers decrease reperfusion injury in lung transplantation. Ann Thorac Surg 1990;50:204-10. 15. Cremer J, [urmann M, Dammenhayn L, et al. Oxygen free radical scavengers to prevent pulmonary reperfusion injury after heart-lung transplantation. J Heart Transplant 1989;8: 330-6. 16. Lambert CJ, Egan TM. Optimal timing of administration of a free radical scavenger in lung preservation. Transplantation 1992;54:205-9. 17. Bryan CL, Cohen DJ, Dew JA, Trinkle JK, Jenkinson SG. Glutathion decreases the pulmonary reimplantation response in canine lung autotransplants. Chest 1991;100:1694-702. 18. Harjula A, Baldwin [C, Stinson EB, Oyer PE, Shumway NE. Clinical heart-lung preservation with prostaglandin E1. Transplant Proc 1987;19:4101-2. 19. Moncada S, Flower RJ, Vane JR. Prostaglandins, prostacyclin and thromboxane A2. In: Gilman AG, Goodman LS, Rail TW, Murad F, eds. The pharmacological basis of therapeutics. New York: Macmillan, 1985:660-76. 20. Long WA, Rubin IJ. Prostacyclin and PGEI treatment of pulmonary hypertension. Am Rev Respir Dis 1987;136:773-6. 21. Haynes DR, Whitehouse MW, Vernon RB. The prostaglandin El analogue, misoprostol, regulates inflammatory cytokines and immune functions in vitro like the natural prostaglandins El, E2 and E3. Immunology 1992;76:251-7. 22. Kelly CJ, Zurier RB, Krakauer KA, Blanchard N, Neilson EG. Prostaglandin El inhibits effector T cell induction and tissue damage in experimental murine interstitial nephritis. J Clin Invest 1987;79:782-9. 23. Kassis S, Lee JC, Hanna N. Effects of prostaglandins and cAMP levels on monocyte IL-l production. Agents Actions 1989;27:274-6. 24. Strom TB, Carpenter CB. Prostaglandin as an effective antirejecton therapy in rat allograft recipients. Transplantation 1983; 35:279-81. 25. Imura M, Higashi K Yada I, Namikawa S, Yuasa H, Kusagawa M. Effect of prostaglandin El on the prolongation of rat cardiac allograft survival. Transplant Proc 1987;19:1312-5. 26. Rovert A. Cytoprotection by prostaglandins. Gastroenterology 1979;77:761-7. 27. Hooper TL, Thomson DS, Jones MT, et al. Amelioration of lung ischemic injury with prostacyclin. Transplantation 1990; 49:1031-5. 28. McCord JM. Oxygen-derived free radicals in post-ischemic tissue injury. N Engl J Med 1985;312:159-63. 29. McCord JM. Oxygen-derived radicals: a link between reperfusion injury and inflammation. Fed Proc 1987;46:2402-6. 30. Miyoshi S, Shimokawa S, Schreinemakers H, et al. Comparison of the University of Wisconsin preservation solution and other crystalloid perfusates in a 30-hour rabbit lung preservation model. J Thorac Cardiovasc Surg 1992;103:27-32. 31. Wittnich C, Trudel J, Zidulka A, Chiu RC-J. Misleading "pulmonary wedge pressure" after pneumonectomy: its importance in postoperative fluid therapy. Ann Thorac Surg 1986;42:192-6.