- Email: [email protected]
Equivalent eighteen-hour lung preservation with low-potassium dextran or Euro-Collins solution after prostaglandin E1 infusion
Equivalent eighteen-hour lung preservation with low-potassium dextran or Euro-Collins solution after prostaglandin E 1 infusion Improved techniques of pulmonary preservation would help alleviate the critical shortage of donor organs in lung transplantation and would improve early graft function. A previous study demonstrated that cold pulmonary artery flush with low-potassium dextran solution was superior to Euro-Collins solution in preservation of canine lung allografts stored for 12 hours when no pulmonary vasodilator was used before donor lung flush. The present study was designed to determine whether donor pretreatment with prostaglandin El would affect the superiority of low-potassium dextran as a preservation solution. Prostaglandin El was infused (50 ~g/min) in 12 donor dogs until potent vasodilatation was demonstrated. Low-pressure pulmonary artery flush (50 ml/kg) with either EuroCollins or low-potassium dextran solution (n = 6 for each group) was performed at 4° C in a randomized, blinded fashion. Heart-lung blocks were extracted and stored at 4° C for 18 hours before left lung allografting. Inflatable cuffs were placed around each pulmonary artery, allowing independent study of the native and transplanted lungs. All 12 recipient dogs survived the 3-day assessment period. Lungs flushed and stored in Euro-Collins or low-potassium dextran solution provided equivalent gas exchange function on day 0 (arterial oxygen tension: Euro-Collins 289 ± 105 mm Hg versus lowpotassium dextran 265 ± III mm Hg; mean ± standard error of the mean) and on day 3 (EuroCollins 516 ± 45 mm Hg versus low-potassium dextran 354 ± 77 mm Hg; p = 0.10). Mean pulmonary artery pressures in the transplanted lung were not significantly different in the Euro-Collins and low-potassium dextran groups on day 0 (21.4 ± 2 mm Hg versus 33.7 ± 5 mm Hg, respectively; p = 0.09) or on day 3 (20.2 ± 2.7 mm Hg versus 24.2 ± 5.1 mm Hg, respectively; p = 0.50). We conclude that there was no advantage of low-potassium dextran over Euro-Collins as a flush solution in this IS-hour canine single lung allograft model in which prostaglandin El was administered before pulmonary artery flush. (J THORAC CARDIOVASC SURG 1992;104:83-9)
John D. Puskas, MD,* Paulo F. G. Cardoso, MD, Eckhard Mayer, MD, Shiquig Shi, MD, Arthur S. Slutsky, MD, and G. Alexander Patterson, MD, FRCSC, Toronto, Ontario, Canada
Lng transplantation has become an effective option in the management of end-stage lung disease.!:" but widespread application of this procedure is limited by availFrom the Division of Thoracic Surgery, Department of Surgery, University of Toronto, Toronto General Hospital, Toronto, Ontario, Canada. Supported by Medical Research Council grant 10142. Received for publication Sept. II, 1990. Accepted for publication April 30, 1991. Address for reprints: G. A. Patterson, MD, FRCSC, Cardiothoracic Surgery, Washington University, 3108 Queeny Tower, I Barnes Hospital Plaza, St. Louis, MO 63110. 'Recipient of the E. D. Churchill Surgical Fellowship from the Massachusetts General Hospital, Harvard Medical School, Boston, Mass.
12/1/31885
ability of suitable donor lungs. Despite more than a quarter century of research in lung preservation.l there is no consistent clinical method of lung preservation beyond 6 to 9 hours. Improved techniques for donor lung preservation would extend allowable ischemic times, permit more distant procurement of donor lungs, allow human lymphocyte antigen crossmatching, facilitate appropriate pairing of donor and recipient cytomegalovirus serologies, and improve early function of transplanted lungs. Hypothermia is a fundamental tenet of organ preservation." Simple topical cooling of the human lung by immersion has provided satisfactory preservation for up to 5lf2 hours." Pulmonary artery (PA) flush is now more commonly used, and it allows 6 to 9 hours of cold ischemia, with superior preservation of pulmonary func-
83
The Journal of Thoracic and Cardiovascular Surgery
84 Puskas et al.
Table I. Composition offlush solutions
EC Na+ (mmol/L) K+ (rnmol/L)
cr (rnmol/L)
HC03- (mmoljL)
Mg++ (mmol/L) P04- (rnmol/L) Glucose (gm/L)
Dextran40 (gm/L) pH Osmolarity (mOsm/L)
10 115 15 10 58 35 7.3 355
LPD
168 4 103 2 37 20 7.45 280
EC. Euro-Collins; LPD, low-potassium dextran.
tion in dogs" and human beings." The consensus in the literature is that hypothermic PA flush produces its observed benefit principally by rapid, uniform cooling of the lung before the ischemic period.' Any salutory effect of flushing potentially harmful blood constituents from the pulmonary vasculature before lung preservation has generally been considered a secondary benefit.V' II Administration of a pulmonary vasodilator agent before cold P A flush has become routine, both clinically and in many experimental models. In particular, prostaglandin EI (PGEt) and prostacyclin (PGh) have been used to produce pulmonary vasodilatation before PA flushing and have improved preservation in lung l2• 13 and heartlung l 4-16 transplantation. The optimal composition and administration of PA flush solution remain controversial. A previous study!' from our laboratory demonstrated that PA flush with an extracellular solution (low-potassium dextran; LPD) provided better lung function than an intracellular solution (Euro-Collins; EC) after 12 hours of hypothermic preservation in a canine single lung allograft model in which no prostaglandin or other pulmonary vasodilator was given before PA flush. It is possible that these results were not due to any specific effect of the LPD solution during ischemic storage but simply that it caused less pulmonary vasoconstriction and hence more uniform flushing of the lung during the donor harvest. Thus the purpose of the present study was to determine whether administration of PGE I before PA flush would alter these findings. Because preliminary experiments with PGE 1 yielded uniformly excellent lung function after 12 hours of preservation with either EC or LPD flush, the ischemic period was extended to 18 hours to try to demonstrate an intergroup difference. Methods Donor procedure. Mongrel adult dogs were paired by sex and weight. All dogs received a standard intravenous anesthetic
sequence of meperidine (50 mg) and acepromazine maleate (I mg), followed by atropine (0.5 mg), cefazolin (l gm), and thiopental sodium (Pentothal) (LO mg/kg). Donors were then intubated (8.5F) and their lungs ventilated (Ventimeter Ventilator, Air-Shields, Hatboro, Pa.) with a tidal volume of 25 ml/kg at a rate of 12 breaths/min without positive end-expiratory pressure. An inspired oxygen concentration of 1.0 with halothane 0.5 to 1.5%was used to maintain anesthesia. Systemic blood pressure was continuously monitored (Hewlett-Packard 8A multichannel recorder; Hewlett-Packard Co., Palo Alto, Calif.) via a right femoral arterial catheter (No. 20 Abocath, Deseret Medical, Inc., Sandy, Utah). Arterial blood gases and hematocrit values were assayed (Corning model No. 178, Corning, Inc., Corning, N.Y.). After median sternotomy, the azygos vein was divided and the superior and inferior venae cavae were isolated. The aorta. main PA, and trachea were each encircled with umbilical tapes. A pursestring of 3-0 Prolene suture (Ethicon, Inc., Somerville, N.J.) was placed in the main PA proximally. After systemic heparinization with heparin (500 U/kg), the PA was cannulated with an 8 mm aortic arch cannula (Sarns Inc., Ann Arbor, Mich.). A rapid intravenous infusion ofpGE 1 (Prostin VR. The Upjohn Co., Kalamazoo, Mich.) (50 J.Lg/ min) was administered until systemic systolic blood pressure declined by at least 30%. After inflowocclusion the left atrial appendage was amputated, and the lungs were inflated once to a pressure of 30 em of water pressure to minimize atelectasis. The PA was flushed by gravity drainage from a height of 30 em with 50 ml/kg of cold (4 0 C) flush solution, either EC or LPD (Table I), in a randomized, blinded fashion. The PA pressure was continuously recorded during each flush via a catheter introduced into the PA cannula through a side port. P A flush pressure was carefully maintained below 12 mm Hg by adjusting the flow rate with a clamp on the flush tubing. Ventilation of the lungs was continued during the PA flush. A temperature probe in the parenchyma of the right middle lobe recorded the cooling of the lungs. The cold left atrial effluent was allowed to pool in the chest to provide additional topical cooling. The trachea was clamped at end-inspiration, so that the lungs were maintained in an inflated state. The heart-lung block was then expeditiously excised, with minimal handling of the lungs, and placed in a sterile plastic bag containing 1 Lofcold (4 0 C) solution identical to thePA flush solution. A second sterile bag of cold (4 0 C) Ringer's lactate solution provided further cooling and mechanical cushioning. Finally, the double-bagged specimen was stored for 18 hours in a monitored cold room at 4 0 C. Recipient procedure. Weight-matched recipients received oral cyclosporine (15 mg/kg) and azathioprine (I mg/kg), exactly as for the donor procedure, before undergoing general anesthesia. All recipients received methylprednisolone (1000 mg intravenously after induction. A No.9 Fogarty venous occlusioncatheter (Baxter Healthcare Corp., Edwards Division, Irvine, Calif.) was advanced into the left main-stem bronchus before intubation. Right femoral venous Swan-Ganz (Baxter) and arterial catheters were introduced. After 10 minutes of supine ventilation, arterial blood gases, hematocrit values, and hemodynamic assessments were recorded. A thoracotomy was performed in the left fifth intercostal space and the left and right PAs were isolated. An inflatable silicone rubber cuff was placed around the proximal right PA. The left pneumonectomy was then completed. The bronchus was divided proximal to the first bifurcation, and the bronchial
Volume 104 Number 1 JUly 1992
Lung preservation with LPD or EC solution after PGE j
blockerballoon was inflated to occlude the amputated bronchus. A long vascular clamp was placed across the lateral wall of the left atrium, incorporating all three left pulmonary veins. The donor left lung, while floating in cold (4 0 C) flush solution in a sterile ice bath, was then trimmed from the preserved heart-lung block. The orifice of the right mediastinal vein(s) was closed with an everting, running 6-0 Prolene suture and was incorporated into the left atrial cuff. The donor bronchus was trimmed back to a single cartilaginous ring proximal to the first bifurcation. The PA stump was left as long as possible. The atrial cuff in the recipient was then fashioned with an incision joining the three pulmonary veins. The atrial anastomosis was performed first, with use of an everting, running, horizontal, 5-0 Prolene mattress suture. Special care was taken to exclude all atrial muscle from the atrial lumen, to avoid atrial thrombosis. An end-to-end PA anastomosis was performed with running 6-0 Prolene sutures, and it was left untied to allow back-bleeding. Finally, the bronchial anastomosis was performed with three running 4-0 Vicryl sutures (Ethicon, Inc., Somerville, N.J.) in triangular fashion. After the transplanted lung was reinflated, the atrial clamp was released. When back-bleeding appeared at the PA anastomosis, the PA clamp was released, and the PA suture was tied after the restoration of flow to prevent anastomotic narrowing. A second inflatable silicone rubber cuff was placed around the left PA. The injection ports connected to each of the PA cuffs were then implanted subcutaneously. A 20F chest tube was inserted, and -15 ern suction was applied. An intercostal block of 0.5% bupivacaine hydrochloride Marcaine, 10m!) was administered, and the chest was closed in layers in standard fashion. Ringer's lactate solution was used for volume replacement during the procedure; no pressor substances were given at any time. Each dog received cyclosporine (15 mg/kg orally), azathioprine (I mg/kg), prednisone (I mg/kg orally), and cefazolin (I gm intramuscularly) daily. Buprenorphine (0.3 mg intramuscularly) was given daily, as necessary, for analgesia. Assessment. Routine assessments were made before and after the operation on day 0 and were repeated on day 3. The dogs' lungs were ventilated in the supine position without positiveend-expiratory pressure, and arterial blood gases (inspired oxygen fraction = 1.0) and systemic and pulmonary hemodynamics were recorded after routine bronchoscopic examination and suctioning of any secretions or edema fluid. Similar determinations were repeated after a 10-minute occlusion of the left PA and again after a I O-minuteocclusion of the right PA. After each occlusion, flow was restored to both lungs, and two-lung assessmentswere performed. At the conclusion of assessment on day 0, each recipient was administered 40 mg furosemide (Lasix) intravenously, and the chest tube was removed. On the third postoperativeday, the dogs were anesthetized as described, and they underwent blood gas and hemodynamic assessment of combined and unilateral lung function as on day O. The dogs werethen killed, and all the anastomoses were carefully examined. All dogs received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Societyfor Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH Publication No. 80-23, revised 1978). Statistical analysis of the data was performed by paired and unpaired, double-tailed t tests, with use of the commercially
85
LUNG TEMP (GC) 40,.----------------------,
.+. LPD Group
~ EC Group
30
20
10 p-0.24
O'--------------------_----.J
Start Flush
30
60
90
120
FLUSH TIME (sec)
Fig. 1. End-flush lung temperature. EC versus LPD groups; 50 ml/kg, 4 0 C flush. All values are mean ± standard error of the mean; n = 6 for each group. Ee, Euro-Collins solution; LPD, low-potassium dextran.
available SAS software package (SAS, Inc., Cary, N.C.). All data are presented as mean ± standard error of the mean.
Results Dogs randomly assigned to the EC and LPD groups had similar weights (l9.3 ± 1.2kg versus 18.9 ± 0.6 kg, respectively). The EC and LPD groups did not differ in total PG E I dose administered to donor dogs (141.7 ± 10 f.Lg versus 177.5 ± 38 f.Lg; p = 0.38). The decline in systemic arterial blood pressure, used as an index of vasodilatation produced by this prostaglandin infusion, was also similar in the two groups (41% ± 4% for the EC group; 43% ± 3% for the LPD group). The mean time for delivery of the pulmonary flush solution was 101 ± 18 seconds in the EC group and 76.5 ± 11 seconds in the LPD group (p = 0.28). The flush pressures did not differ between the groups (7.2 ± 0.7 mm Hg in the EC group versus 6.7 ± 0.5 mm Hg in the LPD group). The final temperatures measured at the end of P A flush by a digital probe placed in the right middle lobe of each donor showed no significant intergroup difference: 11.7 0 ± 2 0 C for the EC group and 8.7 0 ± 1 0 C for the LPD
86
The Journal of Thoracic and Cardiovascular Surgery
Puskas et al.
PaO. (mmHg) 600 -,-----=--c.-_--=--c.-
-,
500
400 300+-------;; 200 100+---
o-L-_DAY 0
DAY 3
ASSESSMENT (TIME) _EC
_LPD
Fig. 2. Arterial oxygen tension of transplanted lung. EC versus LPD groups. All values are mean ± standard error of the mean; n = 6 for each group. EC, Euro-Collins solution; LPD, low-potassium dextran; Pa02, arterial oxygen tension during isolatedperfusion of the transplanted lung alone; inspired oxygen fraction = 1.0.
50
mPAP (mmHg)
-,----'-----~----~---------------,
*
immediate assessment, with perfusion to the transplanted lung alone, the arterial oxygen tensions in the EC and LPD groups were similar (289 ± 105 mm Hg versus 265 ± 112 mm Hg, respectively). Oxygenation in both groups improved from day 0 to day 3 but did not reach statistical significance. In the EC group, arterial oxygen tension was 289 ± 104mmHgondayOversus516 ± 45 mm Hg on day 3 (p = 0.10); in the LPD group, arterial oxygen tension was 265 ± III mm Hg on day 0 versus 354 ± 77 mm Hg on day 3 (p = 0.63). On day 3, arterial oxygen tension generated by the transplanted lung alone was higher in the EC than in the LPD group, but not significantly so (515 ± 45 mm Hg versus 354 ± 77 mm Hg; p = 0.10; Fig. 2). On day 0, with flow to the transplanted lung alone, the mean PA pressure was lower in the EC group than in the LPD group (21.4 ± 2 versus 33.7 ± 5 mm Hg; Fig. 3). This difference was not significant (p = 0.09) and disappeared by day 3 (EC group, 20.2± 3 mm Hg versus LPD group, 24.2 ± 5 mm Hg; p = 0.50). There were no significant intergroup differences in cardiac output through the transplanted lung on either day 0 or day 3.
P-O.086
Discussion
40+---~
30+----~
20+--10 0-'---DAY 0
DAY 3
ASSESSMENT (TIME) _EC
_LPD
Fig. 3. Mean pulmonaryartery pressureof transplantedlung. EC versus LPD groups.All valuesare mean ± standard error of the mean; n = 6 for each group. EC, Euro-Collins solution; LPD, low-potassium dextran; mPAP, mean pulmonaryartery pressure during isolated perfusion of the transplanted lung alone.
group (p = 0.24) (Fig. I). The mean total ischemic time in the EC group was 18.6 ± 0.1 hours and in the LPD group, 18.5 ± 0.1 hours. In this 18-hour canine lung preservation model, a large variation in posttransplantation lung function was observed (Table 11). However, no significant intergroup differences were shown. In four dogs (two EC and two LPD), occlusion of the native PA was not well tolerated on day 0, and hemodynamic instability required abbreviation of the assessment period for the transplanted lung at periods ranging from 2.5 to 5 minutes. At the time of
PGEI and PGh are believed to improve lung preservation largely, but perhaps not exclusively, by theirvasodilatory effects.!? There is both direct and indirect evidence that, in both normal and disease states, prostaglandins are important modulators of pulmonary vascular tone." The pulmonary and systemic vasodilator effects of these agents have been well documented, both experimentally and clinically. 18-20 In addition, PGE 1 has been shown to inhibit pulmonary intravascular leukocyte aggregation-' and to suppress vascular endothelial permeability induced by vasoactive inflammatory mediators.F whereas PGh reduces the adhesion of leukocytes to injured vascular endothelium.P PGE 1 and PGh are potent inhibitors of aggregation of human platelets in vitro,24 and they have a variety of "cytoprotective" effects on gastric and intestinal mucosa exposed to noxious agents.P PGE 1 inhibits T-lymphocyte cytotoxicity and mitogenesis and diminishes production of interleukin-Zr'" These actions may explain its significant immunosuppressive effects; it prolongs the survival of both renal 26 and cardiac'? allografts in rats. Our results confirm the utility of this model of canine single lung allotransplantation.P which allows independent assessment of native and transplanted lung function and permits study of both acute preservation-related lung injury and the delayed manifestations of ischemia-reperfusion injury after a 3-day period of recovery. The present randomized, blinded experimental groups showed no sig-
Volume 104 Number 1
Lung preservation with LPD or EC solution after PGEj
July 1992
87
Table II. Assessment of transplanted lung function Day 3
Day 0
Pa02 (mm Hg) Paco2 (rnm Hg) mPAP (mm Hg)
co (Lzmin)
EC
LPD
EC
LPD
289 ± 105
265 ± 112 48 ± 6 34 ± 5
515 ± 45 46 ± 5 20 ± 3
354 ± 77
49 ± 4
21 ± 2 1.5 ± 0.3
1.4 ± 0.2
1.9 ± 0.2
61 ± 6 24 ± 5 2.1 ± 0.3
All values are mean ± standard error of the mean; n = 6 for each group. EC, Euro-Collins; LPD, low-potassium dextran; mPAP, mean pulmonary artery pressure; CO, cardiac output.
nificant differences in donor or preoperative recipient lung function, dose of and response to PGEI administration, flush pressures, flows and times, or parenchymal cooling of donor lungs during harvesting, as measured by a single thermistor probe in the right middle lobe. Mean values for transplanted lung arterial oxygen tension and mean pulmonary artery pressure were actually superior in the EC than in the LPD group on both days 0 and 3, but these differences did not reach statistical significance. With this small sample size (n = 6 in each group), it is not clear whether these differences represent a real trend toward statistical significance. The large individual variations observed in this 18-hour canine lung preservation model may be due to multiple factors, including the use of mongrel dogs, differences in donor-recipient matching, and the long preservation period. These results are in contrast to other data recently reported from our laboratory, II which showed superior gas exchange on day 0 (but not on day 3) in canine lungs transplanted after 12 hours of hypothermic preservation after PA flush with LPD in comparison with Ee. No prostaglandin or other pulmonary vasodilator was used in this earlier study, which may explain these different results. The vasoconstrictive response of pulmonary vessels to high-potassium flush solutions is widely known.I I, 29 Thus, in the present study, PGE I was administered in a rapid intravenous infusion to achieve a decline in systemic systolic blood pressure of approximately 40% in each group. (In all animals, the duration of this systemic hypotension was less than 20 seconds.) This potent dose of vasodilator appears to have largely overcome the vasoconstrictive effects of EC solution, as evidenced by the similar flush pressures, times, and cooling in the EC and LPD groups. The quality of lung preservation, as measured by early and late gas exchange and hemodynamics, was not statistically different in the two groups. Vasoconstriction during PA flushing may be undesirable for several reasons. First, it may produce acutely high PA flush pressures, which may disrupt endothelial integrity and may contribute to pulmonary edema both during
PA flushing and during postischemic reperfusion. Second, it may limit the uniformity of distribution of the cold flush solution by causing significant shunting in the lung, thereby producing an irregular and incomplete cooling of the lung parenchyma. Finally, this uneven flush distribution may remove blood constituents from the pulmonary vasculature less completely. Leukocytes, platelets, and complement may be retained in the ischemic lung and potentially become activated, contributing to the ischemia/reperfusion injury associated with poor pulmonary preservation. Recent histologic and autoradiographic studies by Dr. K. DeCampos in our laboratory have demonstrated that some leukocytes are retained in the pulmonary vascular bed of rat lungs flushed with EC (50 ml/kg, no PGE 1) and that these cells are metabolically active, still capable of incorporating thymidine into deoxyribonucleic acid after lung preservation (unpublished data). Thus a hyperkalemic electrolyte composition may constitute a poor pulmonary flush solution in the absence of a pulmonary vasodilator agent. The administration of a potent dose ofPGE t beforePAflush in the present study seems to have Overcome this difficulty, rendering EC as good as or better than LPD as a pulmonary preservation solution. This is the case despite recent evidence from a rat type II pneumocyte cell culture model, both in our own laboratory'? and elsewhere, 3 I suggesting that incubation in high-potassium intracellular solutions may be more injurious to this single pulmonary cell type than a variety oflow-potassium preservation solutions under cell culture conditions. It may be that the quality of PA flush-Iowpressure, relatively high volume (50 nil/kg) flush through a well-dilated pulmonary vascular bed, removing all blood constituents-is more important, within currently applied ischemic periods, than the differences in electrolyte composition of these two flush solutions. The relative importance of rapid organ cooling versus clearance of blood constituents from the pulmonary vasculature by pulmonary artery flush remains to be elucidated. The long-held belief that rapid cooling is the sole or predominant benefit obtained'' is challenged by the
8 8 Puskas et al.
significant incremental improvement in lung preservation provided by the addition of prostanoid vasodilators to the PA flush protocolI2-14, 17 in the absence of any significant increase in the rate or degree of organ cooling.V A significant cytoprotective effect of PGE), obviating a possible hyperkalemic injury by EC solution, is possible. In the earlier study by Keshavjee and coworkers, II lungs were flushed with either EC or LPD, and all were stored in Ringer's lactate solution. In the present study, lungs were both flushed and stored in either EC or LPD; the possible effects of this difference are unclear. In summary, all 12 recipient dogs survived the 3-day assessment period; we were able to demonstrate reproducible preservation of canine lungs for more than 18 hours by PA flush after PGE I infusion. While large individual variations were observed in posttransplantation lung function, no significant intergroup differences could be shown. Advantages previously reported for LPD over EC as a P A flush solution are likely related to the lesser degree of pulmonary vasoconstriction observed with the low-potassium solution.. When a potent dose of the pulmonary vasodilator PGE I was administered before PA flushing, posttransplantation lung function was similar in the EC and LPD groups. Presently, the clinical donor lung extraction technique used in Toronto includes a bolus of 1000 J.Lg of PGEI (Prostin VR) injected directly into the main P A before high-flow (50 ml/kg), low-pressure PA flush with cold (4 0 C) modified EC solution that is supplemented with an additional 1000 J.Lg of PGEI. We acknowledge the superb technical assistance of J. Mates and S. Diamant. Dr. K. DeCampos provided expert assistance in statistical analysis. We thank Ethicon Canada Ltd., Peterborough, Canada, for providing the suture material and Upjohn Canada Ltd., Don Mills, Canada, for providing the PGE 1 for this study. REFERENCES 1. Cooper JD, Patterson GA, Grossman RF, Maurer J, Toronto Lung Transplant Group. Double-lung transplant for advanced chronic obstructive lung disease. Am Rev Respir Dis 1989;139:303-7. 2. Grossman RF, Frost A, Zamel N, et al. Results of singlelung transplantation for bilateral pulmonary fibrosis. N Engl J Med 1990;322:727-33. 3. Pasque MK, Cooper JD, Kaiser LR, Haydock DA, Triantaf A, Trulock EP. Improved technique for bilaterallung transplantation: rationale and initial clinical experience. Ann Thorac Surg 1990;49:785-91. 4. Calhoon JH, Grover FL, Gibbons WJ, et al. Single lung transplantation: alternative indications and technique. J THORAC CARDIOVASC SURG 1991;101:816-25. 5. Haverich A, Scott WC, Jamieson SW. Twenty years of
The Journal of Thoracic and Cardiovascular Surgery
lung preservation: a review. J Heart Transplant 1985; 4:234-40. 6. Belzer FO, Southard JH. Principles of solid-organ preservation by cold storage. Transplantation 1988;45:673-6. 7. Toronto Lung Transplant Group. Experience with singlelung transplantation for pulmonary fibrosis. JAMA 1990; 259:2258-62. 8. Locke TJ, Hooper TL, Flecknell PA, McGregor CGA. Preservation of the lung. J THORAC CARDIOVASC SURG 1988;96:789-95. 9. Starkey TD, Sakakibara N, Hagberg RC, Tazelaar HD, Baldwin JC, Jamieson SW. Successful six-hour cardiopulmonary preservation with simple hypothermic crystalloid flush. J Heart Transplant 1986;5:291-7. 10. Harjula A, Baldwin JC, Stinson EB, Oyer PE, Shumway NE. Clinical heart-lung preservation with prostaglandin E j • Transplant Proc 1987;19:4101-2. 11. Keshavjee S, Yamazaki F, Cardoso PF, McRitchie DI, Patterson GA, Cooper JD. A method for safe twelve-hour pulmonary preservation. J THORAC CARDIOVASC SURG 1989;89:529-34. 12. Klepetko W, Muller MR, Khunl-Brady G, et al. Beneficial effect of Iloprost on early pulmonary function after lung preservation with modified Euro-Collins solution. Thorac Cardiovasc Surg 1989;37:174-9. 13. Hooper TL, Thomson DS, Jones MT, et al. Amelioration of lung ischemic injury with prostacyclin. Transplantation 1990;49:1031-5. 14. Jurmann MJ, Dammenhayn L, Schafers H-J, Wahlers T, Fieguth H-G, Haverich A. Prostacyclin as an additive to singlecrystalloid flush:improved pulmonary preservationin heart-lung transplantation. Transplant Proc 1987;19: 4103-4. 15. Harjula A, Baldwin JC, Shumway NE. Donor deep hypothermia or donor pretreatment with prostaglandin E 1 and single pulmonary artery flush for heart-lung graft preservation: an experimentalprimate study. Ann Thorac Surg 1988;46:553-5. 16. Baldwin JC, Frist WH, Starkey TD, et al. Distant graft procurement for combined heart and lung transplantation using pulmonary artery flush and simple topical hypothermia for graft preservation. Ann Thorac Surg 1987;43: 670-3. 17. Mulvin D, Jones K, Howard R, Grosso M, Repine J. The effect of prostacyclin as a constituent of a preservation solution in protecting lungs from ischemic injury becauseof its vasodilatory properties. Transplantation 1990;49:82830. 18. Rabinovitch M. Prostaglandins and structural changes in pulmonary arteries. Am Rev Respir Dis 1987;136:777-9. 19. Kadowitz PJ, Chapnick BM, Feigen LP, Hyman AL, Nelson PK, Spannhake EW. Pulmonary and systemic vasodilator effects of the newly discoveredprostaglandin, PGI2. J Appl PhysioI1978;45:408-13. 20. Long WA, Rubin LJ. Prostacyclin and PGE j treatment of pulmonary hypertension. Am Rev Respir Dis 1987; 136:773-6.
Volume 104 Number 1 July 1992
Lung preservation with LPD or Ee solution after PGE/
21. Bolanowski PJP, Bauer J, Machiedo G, Neville WE. Prostaglandin influence on pulmonary intravascular leukocytic aggregation during cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1977;73:221-4. 22. Fantone JC, Kunkel SL, Ward PA, Zurier RB. Suppression by prostaglandin E, of vascular permeability induced by vasoactive inflammatory mediators. J Immunol 1980; 125:2591-6. 23. Jones G, Hurley JV. The effect of prostacyclin on the adhesion of leukocytes to injured vascular endothelium. J PathoI1984;142:51-9. 24. Moncada S, Flower RJ, Vane JR. Prostaglandins, prostacyclin, and thromboxane A2. In: Goodman AG, Goodman LS, Rail TW, Murad F, eds. The pharmacological basis of therapeutics. 7th ed. New York: Macmillan, 1985:660-73. 25. Robert A. Cytoprotection by prostaglandins. Gastroenterology 1979;77:761-7. 26. Strom TB, Carpenter CB. Prostaglandin as an effective antirejection therapy in rat renal allograft recipients. Transplantation 1983;35:279-81. 27. Imura M, Higashi K, Yada I, Namikawa S, Yuasa H, Kusagawa M. Effect of prostaglandin E, on the prolonga-
28.
29.
30.
31.
32.
89
tion of rat cardiac allograft survival. Transplant Proc 1987;19:1312-5. Jones MT, Hsieh C, Yoshikawa K, Patterson GA, Cooper JD. A new model for assessment of lung preservation. J THORAC CARDIOVASC SURG 1988;96:608-14. Yamazaki F, Yokomise H, Keshavjee S, et al. The superiority of an extracellular fluid solution over Euro-Collins solution for pulmonary preservation. Transplantation 1990; 49:690-4. Puskas J, Maccherini M, Keshavjee S, Slutsky AS, Patterson GA, Edelson JD. Low potassium dextran is superior to Euro-Collins as a preservative solution in a type II pneumocyte cell culture model. Am Rev Respir Dis 1990; 141:A406. Hachida M, Hoon DSB, Morton DL. A comparison of solutions for lung preservation using pulmonary alveolar type II cell viability. Ann Thorae Surg 1988;45:643-6. Hooper TL, Fetherston GJ, Aecknell PA, Dark JH, McGregor CGA. The use of a prostacyclin analog, Iloprost, as an adjunct to pulmonary preservation with Euro-Collins solution. Transplantation 1990;49:495-9.
John D. Puskas, MD,* Paulo F. G. Cardoso, MD, Eckhard Mayer, MD, Shiquig Shi, MD, Arthur S. Slutsky, MD, and G. Alexander Patterson, MD, FRCSC, Toronto, Ontario, Canada
Lng transplantation has become an effective option in the management of end-stage lung disease.!:" but widespread application of this procedure is limited by availFrom the Division of Thoracic Surgery, Department of Surgery, University of Toronto, Toronto General Hospital, Toronto, Ontario, Canada. Supported by Medical Research Council grant 10142. Received for publication Sept. II, 1990. Accepted for publication April 30, 1991. Address for reprints: G. A. Patterson, MD, FRCSC, Cardiothoracic Surgery, Washington University, 3108 Queeny Tower, I Barnes Hospital Plaza, St. Louis, MO 63110. 'Recipient of the E. D. Churchill Surgical Fellowship from the Massachusetts General Hospital, Harvard Medical School, Boston, Mass.
12/1/31885
ability of suitable donor lungs. Despite more than a quarter century of research in lung preservation.l there is no consistent clinical method of lung preservation beyond 6 to 9 hours. Improved techniques for donor lung preservation would extend allowable ischemic times, permit more distant procurement of donor lungs, allow human lymphocyte antigen crossmatching, facilitate appropriate pairing of donor and recipient cytomegalovirus serologies, and improve early function of transplanted lungs. Hypothermia is a fundamental tenet of organ preservation." Simple topical cooling of the human lung by immersion has provided satisfactory preservation for up to 5lf2 hours." Pulmonary artery (PA) flush is now more commonly used, and it allows 6 to 9 hours of cold ischemia, with superior preservation of pulmonary func-
83
The Journal of Thoracic and Cardiovascular Surgery
84 Puskas et al.
Table I. Composition offlush solutions
EC Na+ (mmol/L) K+ (rnmol/L)
cr (rnmol/L)
HC03- (mmoljL)
Mg++ (mmol/L) P04- (rnmol/L) Glucose (gm/L)
Dextran40 (gm/L) pH Osmolarity (mOsm/L)
10 115 15 10 58 35 7.3 355
LPD
168 4 103 2 37 20 7.45 280
EC. Euro-Collins; LPD, low-potassium dextran.
tion in dogs" and human beings." The consensus in the literature is that hypothermic PA flush produces its observed benefit principally by rapid, uniform cooling of the lung before the ischemic period.' Any salutory effect of flushing potentially harmful blood constituents from the pulmonary vasculature before lung preservation has generally been considered a secondary benefit.V' II Administration of a pulmonary vasodilator agent before cold P A flush has become routine, both clinically and in many experimental models. In particular, prostaglandin EI (PGEt) and prostacyclin (PGh) have been used to produce pulmonary vasodilatation before PA flushing and have improved preservation in lung l2• 13 and heartlung l 4-16 transplantation. The optimal composition and administration of PA flush solution remain controversial. A previous study!' from our laboratory demonstrated that PA flush with an extracellular solution (low-potassium dextran; LPD) provided better lung function than an intracellular solution (Euro-Collins; EC) after 12 hours of hypothermic preservation in a canine single lung allograft model in which no prostaglandin or other pulmonary vasodilator was given before PA flush. It is possible that these results were not due to any specific effect of the LPD solution during ischemic storage but simply that it caused less pulmonary vasoconstriction and hence more uniform flushing of the lung during the donor harvest. Thus the purpose of the present study was to determine whether administration of PGE I before PA flush would alter these findings. Because preliminary experiments with PGE 1 yielded uniformly excellent lung function after 12 hours of preservation with either EC or LPD flush, the ischemic period was extended to 18 hours to try to demonstrate an intergroup difference. Methods Donor procedure. Mongrel adult dogs were paired by sex and weight. All dogs received a standard intravenous anesthetic
sequence of meperidine (50 mg) and acepromazine maleate (I mg), followed by atropine (0.5 mg), cefazolin (l gm), and thiopental sodium (Pentothal) (LO mg/kg). Donors were then intubated (8.5F) and their lungs ventilated (Ventimeter Ventilator, Air-Shields, Hatboro, Pa.) with a tidal volume of 25 ml/kg at a rate of 12 breaths/min without positive end-expiratory pressure. An inspired oxygen concentration of 1.0 with halothane 0.5 to 1.5%was used to maintain anesthesia. Systemic blood pressure was continuously monitored (Hewlett-Packard 8A multichannel recorder; Hewlett-Packard Co., Palo Alto, Calif.) via a right femoral arterial catheter (No. 20 Abocath, Deseret Medical, Inc., Sandy, Utah). Arterial blood gases and hematocrit values were assayed (Corning model No. 178, Corning, Inc., Corning, N.Y.). After median sternotomy, the azygos vein was divided and the superior and inferior venae cavae were isolated. The aorta. main PA, and trachea were each encircled with umbilical tapes. A pursestring of 3-0 Prolene suture (Ethicon, Inc., Somerville, N.J.) was placed in the main PA proximally. After systemic heparinization with heparin (500 U/kg), the PA was cannulated with an 8 mm aortic arch cannula (Sarns Inc., Ann Arbor, Mich.). A rapid intravenous infusion ofpGE 1 (Prostin VR. The Upjohn Co., Kalamazoo, Mich.) (50 J.Lg/ min) was administered until systemic systolic blood pressure declined by at least 30%. After inflowocclusion the left atrial appendage was amputated, and the lungs were inflated once to a pressure of 30 em of water pressure to minimize atelectasis. The PA was flushed by gravity drainage from a height of 30 em with 50 ml/kg of cold (4 0 C) flush solution, either EC or LPD (Table I), in a randomized, blinded fashion. The PA pressure was continuously recorded during each flush via a catheter introduced into the PA cannula through a side port. P A flush pressure was carefully maintained below 12 mm Hg by adjusting the flow rate with a clamp on the flush tubing. Ventilation of the lungs was continued during the PA flush. A temperature probe in the parenchyma of the right middle lobe recorded the cooling of the lungs. The cold left atrial effluent was allowed to pool in the chest to provide additional topical cooling. The trachea was clamped at end-inspiration, so that the lungs were maintained in an inflated state. The heart-lung block was then expeditiously excised, with minimal handling of the lungs, and placed in a sterile plastic bag containing 1 Lofcold (4 0 C) solution identical to thePA flush solution. A second sterile bag of cold (4 0 C) Ringer's lactate solution provided further cooling and mechanical cushioning. Finally, the double-bagged specimen was stored for 18 hours in a monitored cold room at 4 0 C. Recipient procedure. Weight-matched recipients received oral cyclosporine (15 mg/kg) and azathioprine (I mg/kg), exactly as for the donor procedure, before undergoing general anesthesia. All recipients received methylprednisolone (1000 mg intravenously after induction. A No.9 Fogarty venous occlusioncatheter (Baxter Healthcare Corp., Edwards Division, Irvine, Calif.) was advanced into the left main-stem bronchus before intubation. Right femoral venous Swan-Ganz (Baxter) and arterial catheters were introduced. After 10 minutes of supine ventilation, arterial blood gases, hematocrit values, and hemodynamic assessments were recorded. A thoracotomy was performed in the left fifth intercostal space and the left and right PAs were isolated. An inflatable silicone rubber cuff was placed around the proximal right PA. The left pneumonectomy was then completed. The bronchus was divided proximal to the first bifurcation, and the bronchial
Volume 104 Number 1 JUly 1992
Lung preservation with LPD or EC solution after PGE j
blockerballoon was inflated to occlude the amputated bronchus. A long vascular clamp was placed across the lateral wall of the left atrium, incorporating all three left pulmonary veins. The donor left lung, while floating in cold (4 0 C) flush solution in a sterile ice bath, was then trimmed from the preserved heart-lung block. The orifice of the right mediastinal vein(s) was closed with an everting, running 6-0 Prolene suture and was incorporated into the left atrial cuff. The donor bronchus was trimmed back to a single cartilaginous ring proximal to the first bifurcation. The PA stump was left as long as possible. The atrial cuff in the recipient was then fashioned with an incision joining the three pulmonary veins. The atrial anastomosis was performed first, with use of an everting, running, horizontal, 5-0 Prolene mattress suture. Special care was taken to exclude all atrial muscle from the atrial lumen, to avoid atrial thrombosis. An end-to-end PA anastomosis was performed with running 6-0 Prolene sutures, and it was left untied to allow back-bleeding. Finally, the bronchial anastomosis was performed with three running 4-0 Vicryl sutures (Ethicon, Inc., Somerville, N.J.) in triangular fashion. After the transplanted lung was reinflated, the atrial clamp was released. When back-bleeding appeared at the PA anastomosis, the PA clamp was released, and the PA suture was tied after the restoration of flow to prevent anastomotic narrowing. A second inflatable silicone rubber cuff was placed around the left PA. The injection ports connected to each of the PA cuffs were then implanted subcutaneously. A 20F chest tube was inserted, and -15 ern suction was applied. An intercostal block of 0.5% bupivacaine hydrochloride Marcaine, 10m!) was administered, and the chest was closed in layers in standard fashion. Ringer's lactate solution was used for volume replacement during the procedure; no pressor substances were given at any time. Each dog received cyclosporine (15 mg/kg orally), azathioprine (I mg/kg), prednisone (I mg/kg orally), and cefazolin (I gm intramuscularly) daily. Buprenorphine (0.3 mg intramuscularly) was given daily, as necessary, for analgesia. Assessment. Routine assessments were made before and after the operation on day 0 and were repeated on day 3. The dogs' lungs were ventilated in the supine position without positiveend-expiratory pressure, and arterial blood gases (inspired oxygen fraction = 1.0) and systemic and pulmonary hemodynamics were recorded after routine bronchoscopic examination and suctioning of any secretions or edema fluid. Similar determinations were repeated after a 10-minute occlusion of the left PA and again after a I O-minuteocclusion of the right PA. After each occlusion, flow was restored to both lungs, and two-lung assessmentswere performed. At the conclusion of assessment on day 0, each recipient was administered 40 mg furosemide (Lasix) intravenously, and the chest tube was removed. On the third postoperativeday, the dogs were anesthetized as described, and they underwent blood gas and hemodynamic assessment of combined and unilateral lung function as on day O. The dogs werethen killed, and all the anastomoses were carefully examined. All dogs received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Societyfor Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH Publication No. 80-23, revised 1978). Statistical analysis of the data was performed by paired and unpaired, double-tailed t tests, with use of the commercially
85
LUNG TEMP (GC) 40,.----------------------,
.+. LPD Group
~ EC Group
30
20
10 p-0.24
O'--------------------_----.J
Start Flush
30
60
90
120
FLUSH TIME (sec)
Fig. 1. End-flush lung temperature. EC versus LPD groups; 50 ml/kg, 4 0 C flush. All values are mean ± standard error of the mean; n = 6 for each group. Ee, Euro-Collins solution; LPD, low-potassium dextran.
available SAS software package (SAS, Inc., Cary, N.C.). All data are presented as mean ± standard error of the mean.
Results Dogs randomly assigned to the EC and LPD groups had similar weights (l9.3 ± 1.2kg versus 18.9 ± 0.6 kg, respectively). The EC and LPD groups did not differ in total PG E I dose administered to donor dogs (141.7 ± 10 f.Lg versus 177.5 ± 38 f.Lg; p = 0.38). The decline in systemic arterial blood pressure, used as an index of vasodilatation produced by this prostaglandin infusion, was also similar in the two groups (41% ± 4% for the EC group; 43% ± 3% for the LPD group). The mean time for delivery of the pulmonary flush solution was 101 ± 18 seconds in the EC group and 76.5 ± 11 seconds in the LPD group (p = 0.28). The flush pressures did not differ between the groups (7.2 ± 0.7 mm Hg in the EC group versus 6.7 ± 0.5 mm Hg in the LPD group). The final temperatures measured at the end of P A flush by a digital probe placed in the right middle lobe of each donor showed no significant intergroup difference: 11.7 0 ± 2 0 C for the EC group and 8.7 0 ± 1 0 C for the LPD
86
The Journal of Thoracic and Cardiovascular Surgery
Puskas et al.
PaO. (mmHg) 600 -,-----=--c.-_--=--c.-
-,
500
400 300+-------;; 200 100+---
o-L-_DAY 0
DAY 3
ASSESSMENT (TIME) _EC
_LPD
Fig. 2. Arterial oxygen tension of transplanted lung. EC versus LPD groups. All values are mean ± standard error of the mean; n = 6 for each group. EC, Euro-Collins solution; LPD, low-potassium dextran; Pa02, arterial oxygen tension during isolatedperfusion of the transplanted lung alone; inspired oxygen fraction = 1.0.
50
mPAP (mmHg)
-,----'-----~----~---------------,
*
immediate assessment, with perfusion to the transplanted lung alone, the arterial oxygen tensions in the EC and LPD groups were similar (289 ± 105 mm Hg versus 265 ± 112 mm Hg, respectively). Oxygenation in both groups improved from day 0 to day 3 but did not reach statistical significance. In the EC group, arterial oxygen tension was 289 ± 104mmHgondayOversus516 ± 45 mm Hg on day 3 (p = 0.10); in the LPD group, arterial oxygen tension was 265 ± III mm Hg on day 0 versus 354 ± 77 mm Hg on day 3 (p = 0.63). On day 3, arterial oxygen tension generated by the transplanted lung alone was higher in the EC than in the LPD group, but not significantly so (515 ± 45 mm Hg versus 354 ± 77 mm Hg; p = 0.10; Fig. 2). On day 0, with flow to the transplanted lung alone, the mean PA pressure was lower in the EC group than in the LPD group (21.4 ± 2 versus 33.7 ± 5 mm Hg; Fig. 3). This difference was not significant (p = 0.09) and disappeared by day 3 (EC group, 20.2± 3 mm Hg versus LPD group, 24.2 ± 5 mm Hg; p = 0.50). There were no significant intergroup differences in cardiac output through the transplanted lung on either day 0 or day 3.
P-O.086
Discussion
40+---~
30+----~
20+--10 0-'---DAY 0
DAY 3
ASSESSMENT (TIME) _EC
_LPD
Fig. 3. Mean pulmonaryartery pressureof transplantedlung. EC versus LPD groups.All valuesare mean ± standard error of the mean; n = 6 for each group. EC, Euro-Collins solution; LPD, low-potassium dextran; mPAP, mean pulmonaryartery pressure during isolated perfusion of the transplanted lung alone.
group (p = 0.24) (Fig. I). The mean total ischemic time in the EC group was 18.6 ± 0.1 hours and in the LPD group, 18.5 ± 0.1 hours. In this 18-hour canine lung preservation model, a large variation in posttransplantation lung function was observed (Table 11). However, no significant intergroup differences were shown. In four dogs (two EC and two LPD), occlusion of the native PA was not well tolerated on day 0, and hemodynamic instability required abbreviation of the assessment period for the transplanted lung at periods ranging from 2.5 to 5 minutes. At the time of
PGEI and PGh are believed to improve lung preservation largely, but perhaps not exclusively, by theirvasodilatory effects.!? There is both direct and indirect evidence that, in both normal and disease states, prostaglandins are important modulators of pulmonary vascular tone." The pulmonary and systemic vasodilator effects of these agents have been well documented, both experimentally and clinically. 18-20 In addition, PGE 1 has been shown to inhibit pulmonary intravascular leukocyte aggregation-' and to suppress vascular endothelial permeability induced by vasoactive inflammatory mediators.F whereas PGh reduces the adhesion of leukocytes to injured vascular endothelium.P PGE 1 and PGh are potent inhibitors of aggregation of human platelets in vitro,24 and they have a variety of "cytoprotective" effects on gastric and intestinal mucosa exposed to noxious agents.P PGE 1 inhibits T-lymphocyte cytotoxicity and mitogenesis and diminishes production of interleukin-Zr'" These actions may explain its significant immunosuppressive effects; it prolongs the survival of both renal 26 and cardiac'? allografts in rats. Our results confirm the utility of this model of canine single lung allotransplantation.P which allows independent assessment of native and transplanted lung function and permits study of both acute preservation-related lung injury and the delayed manifestations of ischemia-reperfusion injury after a 3-day period of recovery. The present randomized, blinded experimental groups showed no sig-
Volume 104 Number 1
Lung preservation with LPD or EC solution after PGEj
July 1992
87
Table II. Assessment of transplanted lung function Day 3
Day 0
Pa02 (mm Hg) Paco2 (rnm Hg) mPAP (mm Hg)
co (Lzmin)
EC
LPD
EC
LPD
289 ± 105
265 ± 112 48 ± 6 34 ± 5
515 ± 45 46 ± 5 20 ± 3
354 ± 77
49 ± 4
21 ± 2 1.5 ± 0.3
1.4 ± 0.2
1.9 ± 0.2
61 ± 6 24 ± 5 2.1 ± 0.3
All values are mean ± standard error of the mean; n = 6 for each group. EC, Euro-Collins; LPD, low-potassium dextran; mPAP, mean pulmonary artery pressure; CO, cardiac output.
nificant differences in donor or preoperative recipient lung function, dose of and response to PGEI administration, flush pressures, flows and times, or parenchymal cooling of donor lungs during harvesting, as measured by a single thermistor probe in the right middle lobe. Mean values for transplanted lung arterial oxygen tension and mean pulmonary artery pressure were actually superior in the EC than in the LPD group on both days 0 and 3, but these differences did not reach statistical significance. With this small sample size (n = 6 in each group), it is not clear whether these differences represent a real trend toward statistical significance. The large individual variations observed in this 18-hour canine lung preservation model may be due to multiple factors, including the use of mongrel dogs, differences in donor-recipient matching, and the long preservation period. These results are in contrast to other data recently reported from our laboratory, II which showed superior gas exchange on day 0 (but not on day 3) in canine lungs transplanted after 12 hours of hypothermic preservation after PA flush with LPD in comparison with Ee. No prostaglandin or other pulmonary vasodilator was used in this earlier study, which may explain these different results. The vasoconstrictive response of pulmonary vessels to high-potassium flush solutions is widely known.I I, 29 Thus, in the present study, PGE I was administered in a rapid intravenous infusion to achieve a decline in systemic systolic blood pressure of approximately 40% in each group. (In all animals, the duration of this systemic hypotension was less than 20 seconds.) This potent dose of vasodilator appears to have largely overcome the vasoconstrictive effects of EC solution, as evidenced by the similar flush pressures, times, and cooling in the EC and LPD groups. The quality of lung preservation, as measured by early and late gas exchange and hemodynamics, was not statistically different in the two groups. Vasoconstriction during PA flushing may be undesirable for several reasons. First, it may produce acutely high PA flush pressures, which may disrupt endothelial integrity and may contribute to pulmonary edema both during
PA flushing and during postischemic reperfusion. Second, it may limit the uniformity of distribution of the cold flush solution by causing significant shunting in the lung, thereby producing an irregular and incomplete cooling of the lung parenchyma. Finally, this uneven flush distribution may remove blood constituents from the pulmonary vasculature less completely. Leukocytes, platelets, and complement may be retained in the ischemic lung and potentially become activated, contributing to the ischemia/reperfusion injury associated with poor pulmonary preservation. Recent histologic and autoradiographic studies by Dr. K. DeCampos in our laboratory have demonstrated that some leukocytes are retained in the pulmonary vascular bed of rat lungs flushed with EC (50 ml/kg, no PGE 1) and that these cells are metabolically active, still capable of incorporating thymidine into deoxyribonucleic acid after lung preservation (unpublished data). Thus a hyperkalemic electrolyte composition may constitute a poor pulmonary flush solution in the absence of a pulmonary vasodilator agent. The administration of a potent dose ofPGE t beforePAflush in the present study seems to have Overcome this difficulty, rendering EC as good as or better than LPD as a pulmonary preservation solution. This is the case despite recent evidence from a rat type II pneumocyte cell culture model, both in our own laboratory'? and elsewhere, 3 I suggesting that incubation in high-potassium intracellular solutions may be more injurious to this single pulmonary cell type than a variety oflow-potassium preservation solutions under cell culture conditions. It may be that the quality of PA flush-Iowpressure, relatively high volume (50 nil/kg) flush through a well-dilated pulmonary vascular bed, removing all blood constituents-is more important, within currently applied ischemic periods, than the differences in electrolyte composition of these two flush solutions. The relative importance of rapid organ cooling versus clearance of blood constituents from the pulmonary vasculature by pulmonary artery flush remains to be elucidated. The long-held belief that rapid cooling is the sole or predominant benefit obtained'' is challenged by the
8 8 Puskas et al.
significant incremental improvement in lung preservation provided by the addition of prostanoid vasodilators to the PA flush protocolI2-14, 17 in the absence of any significant increase in the rate or degree of organ cooling.V A significant cytoprotective effect of PGE), obviating a possible hyperkalemic injury by EC solution, is possible. In the earlier study by Keshavjee and coworkers, II lungs were flushed with either EC or LPD, and all were stored in Ringer's lactate solution. In the present study, lungs were both flushed and stored in either EC or LPD; the possible effects of this difference are unclear. In summary, all 12 recipient dogs survived the 3-day assessment period; we were able to demonstrate reproducible preservation of canine lungs for more than 18 hours by PA flush after PGE I infusion. While large individual variations were observed in posttransplantation lung function, no significant intergroup differences could be shown. Advantages previously reported for LPD over EC as a P A flush solution are likely related to the lesser degree of pulmonary vasoconstriction observed with the low-potassium solution.. When a potent dose of the pulmonary vasodilator PGE I was administered before PA flushing, posttransplantation lung function was similar in the EC and LPD groups. Presently, the clinical donor lung extraction technique used in Toronto includes a bolus of 1000 J.Lg of PGEI (Prostin VR) injected directly into the main P A before high-flow (50 ml/kg), low-pressure PA flush with cold (4 0 C) modified EC solution that is supplemented with an additional 1000 J.Lg of PGEI. We acknowledge the superb technical assistance of J. Mates and S. Diamant. Dr. K. DeCampos provided expert assistance in statistical analysis. We thank Ethicon Canada Ltd., Peterborough, Canada, for providing the suture material and Upjohn Canada Ltd., Don Mills, Canada, for providing the PGE 1 for this study. REFERENCES 1. Cooper JD, Patterson GA, Grossman RF, Maurer J, Toronto Lung Transplant Group. Double-lung transplant for advanced chronic obstructive lung disease. Am Rev Respir Dis 1989;139:303-7. 2. Grossman RF, Frost A, Zamel N, et al. Results of singlelung transplantation for bilateral pulmonary fibrosis. N Engl J Med 1990;322:727-33. 3. Pasque MK, Cooper JD, Kaiser LR, Haydock DA, Triantaf A, Trulock EP. Improved technique for bilaterallung transplantation: rationale and initial clinical experience. Ann Thorac Surg 1990;49:785-91. 4. Calhoon JH, Grover FL, Gibbons WJ, et al. Single lung transplantation: alternative indications and technique. J THORAC CARDIOVASC SURG 1991;101:816-25. 5. Haverich A, Scott WC, Jamieson SW. Twenty years of
The Journal of Thoracic and Cardiovascular Surgery
lung preservation: a review. J Heart Transplant 1985; 4:234-40. 6. Belzer FO, Southard JH. Principles of solid-organ preservation by cold storage. Transplantation 1988;45:673-6. 7. Toronto Lung Transplant Group. Experience with singlelung transplantation for pulmonary fibrosis. JAMA 1990; 259:2258-62. 8. Locke TJ, Hooper TL, Flecknell PA, McGregor CGA. Preservation of the lung. J THORAC CARDIOVASC SURG 1988;96:789-95. 9. Starkey TD, Sakakibara N, Hagberg RC, Tazelaar HD, Baldwin JC, Jamieson SW. Successful six-hour cardiopulmonary preservation with simple hypothermic crystalloid flush. J Heart Transplant 1986;5:291-7. 10. Harjula A, Baldwin JC, Stinson EB, Oyer PE, Shumway NE. Clinical heart-lung preservation with prostaglandin E j • Transplant Proc 1987;19:4101-2. 11. Keshavjee S, Yamazaki F, Cardoso PF, McRitchie DI, Patterson GA, Cooper JD. A method for safe twelve-hour pulmonary preservation. J THORAC CARDIOVASC SURG 1989;89:529-34. 12. Klepetko W, Muller MR, Khunl-Brady G, et al. Beneficial effect of Iloprost on early pulmonary function after lung preservation with modified Euro-Collins solution. Thorac Cardiovasc Surg 1989;37:174-9. 13. Hooper TL, Thomson DS, Jones MT, et al. Amelioration of lung ischemic injury with prostacyclin. Transplantation 1990;49:1031-5. 14. Jurmann MJ, Dammenhayn L, Schafers H-J, Wahlers T, Fieguth H-G, Haverich A. Prostacyclin as an additive to singlecrystalloid flush:improved pulmonary preservationin heart-lung transplantation. Transplant Proc 1987;19: 4103-4. 15. Harjula A, Baldwin JC, Shumway NE. Donor deep hypothermia or donor pretreatment with prostaglandin E 1 and single pulmonary artery flush for heart-lung graft preservation: an experimentalprimate study. Ann Thorac Surg 1988;46:553-5. 16. Baldwin JC, Frist WH, Starkey TD, et al. Distant graft procurement for combined heart and lung transplantation using pulmonary artery flush and simple topical hypothermia for graft preservation. Ann Thorac Surg 1987;43: 670-3. 17. Mulvin D, Jones K, Howard R, Grosso M, Repine J. The effect of prostacyclin as a constituent of a preservation solution in protecting lungs from ischemic injury becauseof its vasodilatory properties. Transplantation 1990;49:82830. 18. Rabinovitch M. Prostaglandins and structural changes in pulmonary arteries. Am Rev Respir Dis 1987;136:777-9. 19. Kadowitz PJ, Chapnick BM, Feigen LP, Hyman AL, Nelson PK, Spannhake EW. Pulmonary and systemic vasodilator effects of the newly discoveredprostaglandin, PGI2. J Appl PhysioI1978;45:408-13. 20. Long WA, Rubin LJ. Prostacyclin and PGE j treatment of pulmonary hypertension. Am Rev Respir Dis 1987; 136:773-6.
Volume 104 Number 1 July 1992
Lung preservation with LPD or Ee solution after PGE/
21. Bolanowski PJP, Bauer J, Machiedo G, Neville WE. Prostaglandin influence on pulmonary intravascular leukocytic aggregation during cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1977;73:221-4. 22. Fantone JC, Kunkel SL, Ward PA, Zurier RB. Suppression by prostaglandin E, of vascular permeability induced by vasoactive inflammatory mediators. J Immunol 1980; 125:2591-6. 23. Jones G, Hurley JV. The effect of prostacyclin on the adhesion of leukocytes to injured vascular endothelium. J PathoI1984;142:51-9. 24. Moncada S, Flower RJ, Vane JR. Prostaglandins, prostacyclin, and thromboxane A2. In: Goodman AG, Goodman LS, Rail TW, Murad F, eds. The pharmacological basis of therapeutics. 7th ed. New York: Macmillan, 1985:660-73. 25. Robert A. Cytoprotection by prostaglandins. Gastroenterology 1979;77:761-7. 26. Strom TB, Carpenter CB. Prostaglandin as an effective antirejection therapy in rat renal allograft recipients. Transplantation 1983;35:279-81. 27. Imura M, Higashi K, Yada I, Namikawa S, Yuasa H, Kusagawa M. Effect of prostaglandin E, on the prolonga-
28.
29.
30.
31.
32.
89
tion of rat cardiac allograft survival. Transplant Proc 1987;19:1312-5. Jones MT, Hsieh C, Yoshikawa K, Patterson GA, Cooper JD. A new model for assessment of lung preservation. J THORAC CARDIOVASC SURG 1988;96:608-14. Yamazaki F, Yokomise H, Keshavjee S, et al. The superiority of an extracellular fluid solution over Euro-Collins solution for pulmonary preservation. Transplantation 1990; 49:690-4. Puskas J, Maccherini M, Keshavjee S, Slutsky AS, Patterson GA, Edelson JD. Low potassium dextran is superior to Euro-Collins as a preservative solution in a type II pneumocyte cell culture model. Am Rev Respir Dis 1990; 141:A406. Hachida M, Hoon DSB, Morton DL. A comparison of solutions for lung preservation using pulmonary alveolar type II cell viability. Ann Thorae Surg 1988;45:643-6. Hooper TL, Fetherston GJ, Aecknell PA, Dark JH, McGregor CGA. The use of a prostacyclin analog, Iloprost, as an adjunct to pulmonary preservation with Euro-Collins solution. Transplantation 1990;49:495-9.