Comparison of the University of Wisconsin preservation solution and other crystalloid perfusates in a 30-hour rabbit lung preservation model The University of Wisconsin solution, which contains a high potassium concentration (120 mmoljL), was evaluated for rabbit lung preservation by comparing it with a modified University of Wisconsin solution with low potassium (4 mmol/L), a low-potassium dextran solution (4 mmoljL), and simple surface cooling. In the first three groups rabbit lungs were flushed in situ with the solution (n = 5 in each group); then the lung-heart block was harvested and stored at 10 0 C for 30 hours. In the surface cooling group the lungs were harvested without flushing and then simply immersed in saline and stored. For assessment, the stored lung was ventilated with room air and perfused with fresh venous blood at a rate of 40 m1jmin for 10 minutes. Assessment of lung function included gas analysis of effluent blood, mean pulmonary artery perfusion pressure, and peak airway pressure. Among these parameters, oxygen tension was most sensitive. Oxygen tension at 10 minutes' perfusion in the modified University of Wisconsin (95 ± 6 mm Hg) and low-potassium dextran (99 ± 4 mm Hg) groups was significantly higher than that in the surface cooling (61 ± 7 mm Hg) and University of Wisconsin (51 ± 7 mm Hg) groups. There was no difference between the modified University of Wisconsin and low-potassium dextran groups or between the surface cooling and University of Wisconsin groups. We conclude that the low-potassium University of Wisconsin solution is superior to the high-potassium University of Wisconsin ~olution and that the lactobionate and raffinose included in the University of Wisconsin solution as impermeants do not improve lung preservation in this model. (J THORAC CARDIOVASC SURG 1992;103:27-32)
Shinichiro Miyoshi, MD, * Shinji Shimokawa, MD, Hans Schreinemakers, MD, Hiroshi Date, MD, Walter Weder, MD, Baron Harper, MD, and Joel D. Cooper, MD, St. Louis, Mo.
h e University of Wisconsin (UW) group recently developeda flushing solution for solid organ preservation. This UW solution includes lactobionate and raffinose as impermeants that prevent hypothermia-induced cell swelling.1 Because this solution has been shown to preserve effectively the pancreas.? liver;' and kidney" of experimental animals, we wished to evaluate it for lung preservation. From the Division of Cardiothoracic Surgery, Washington University School of Medicine, St. Louis, Mo. Received for publication Feb. 14, 1990. Accepted for publication Sept. 21, 1990. Address for reprints: Joel D. Cooper, MD, Suite 3108 Queeny Tower, #1 Barnes Hospital Plaza, St. Louis, MO 63110. 'Present address: First Department of Surgery, Osaka University Medical School, Fukushima-ku, Osaka 553, Japan.
Recently Fujimura and coworkers' reported that a crystalloid solution with electrolytes of extracellular composition, along with dextran, glucose, and phosphate buffer, could be used successfully to preserve canine lungs for up to 48 hours. We have used a modification of this solution, which we refer to as low-potassium dextran solution (LPD), and have found it superior to Euro-Collins solution in a canine model of lung transplantation." Similar results were obtained in an isolated rabbit lung preservation model'? On the basis of these studies we have concluded that, for lung preservation, a solution high in potassium content is less suitable than one low in potassium. Because the UW solution contains potassium 120 mmoljL and sodium 30 mmoljL, we decided to compare the UW solution with a modified solution in which the sodium and potassium concentrations were reversed. 27
The Journal of Thoracic and Cardiovascular
Miyoshi et al.
Table I. Composition ofpulmonary flush solutions
s: en :::>
10 0 0
Sodium (mmol/L) Potassium (rnmol/L) Chloride (rnrnol/L) Phosphate radical (rnmol/L) Magnesium sulfate (rnmol/L) Lactobionate (mmol/L) Raffinose (mmol/L) Hydroxyethyl starch (gm/L) Dextran 40 (gm/L) Allopurinol (gm/L) Adenosine (gm/L) Glutathione (gm/L) pH Osmolality (mOsm/L)
27 115 0 25.0 5 100 30 50
150 4 0 25.0 5 100 30 50
168 4 103 36.7 2
0.1 1.3 0.9 7.4 318
0.1 1.3 0.9 7.4 338
Fig. 1. Mean pulmonary artery pressure (Pl'A) during flushing for group 2 (UW), group 3 (M-UW), and group 4 (LPD).
Lung preservation with these two solutions was compared with our previous best solution, the LPD solution, and with simple cold immersion. Materials and methods Lung preservation was evaluated in an isolated rabbit lung model, which we have described previously," Four different preservation methods were used, with five animals in each group: Group I: Simple immersion of the lung in saline at 10° C for 30 hours Group II: Pulmonary flushing with 200 ml of UW solution at 10° C followed by 30 hours of cold storage at 10°C Group III: Same as Group II, but with the modified UW solution (M-UW) Group IV: Same as Group II, but with the LPD solution The composition of the various flush solutions is shown in Table 1. Operative procedure. The operative procedures for harvest were identical to those previouslydescribed'' except for a flushing technique. After the pericardium was opened, a 14-gauge catheter was introduced into the main pulmonary artery. A drainage tube for the eflluent fluid was placed in the apex of the left ventricle. After the superior and inferior venae cavae were ligated, both lungs were flushed through the pulmonary artery catheter with 200 ml of the preservation solution at 10° C. The flush solution was placed in a plastic bag, hung 60 cm above the level of the pulmonary artery. The flushing was by gravity drainage with a 4.5 mm inside diameter tube. During flushing of the lungs the pulmonary artery pressure was monitored by a pressure transducer that was attached to the pulmonary artery catheter. After the lungs were flushed the left ventricular drainage catheter was removed and the lung-heart block was excised and stored at 10° C for 30 hours. Assessment. After 30 hours of storage, the left lung was reperfused with freshly drawn homologous venous blood at a
rate of 40ml/min for 10 minutes with a roller pump. A sample of this blood perfusate was analyzed for blood gases at the onset of reperfusion. During reperfusion the pulmonary artery and airway pressure were continuously recorded. The pulmonary venous eflluent was drained through the left ventricular drainage tube, and blood samples for blood gas analysis were taken from this after 1, 5, and 10 minutes of reperfusion. The details of the assessment procedure were also described previously.f Statistics. Statistical analysis was performed by means of analysis of variance and Tukey's multiple comparisons method; a p value less than 0.05 was considered statistically significant. All values are presented as mean ± standard error of the mean. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH Publication No. 80-23 revised 1978). Results The four groups of animals did not differ from each other in terms of weight of the animals, total ischemic time, or blood gas and hemoglobin determinations of the homologous blood used for perfusion of the lungs (Table II). For the groups in which the lungs were flushed before storage (groups II, III, and IV), the mean pulmonary artery pressure during flushing and the duration of flushing differed significantly (Table II and Fig. 1). The lowest perfusion pressure and shortest perfusion times were obtained with the LPD solution (12 mm Hg; 108 seconds). In contrast, the UW solution had values more than twice those of the LPD solution (27 mm Hg, 338 seconds). The M-UW solution had values between those of the other two (16 mm Hg; 150 seconds). The differences among the three groups were all statistically significant except for the flushing time between the M -UW and the LPD groups, as shown in Table II. A typical mean pulmonary artery pressure tracing during flushing with each of the three solutions is shown in Fig. 1.
Volume 103 Number 1 January 1992
UW solution for rabbit lung preservation
Table II. Characteristics of the four groups Rabbit weight (kg) Group I (SC) Group 2 (UW)
Group 3 (M-UW)
Group 4 (LPD)
t 3.0 ± 0.1
[[338 ± 45 [[27 ± I 150±7 108 ±
All values are mean ± standard error of the mean. P'PA,
12 ± I
Ischemic time (min)
Mean pulmonary artery
Peal (mm Hg)
1798 ± 3 1787 ± 10
7.21 ± 0.Q2 7.22 ± 0.02
20 ± I 17 ± 2
62 ± 2 62 ± 2
10.7 ± 0.4 10.3 ± 0.2
1801 ± II
7.18 ± 0.04
15 ± 2
64 ± 4
10.4 ± 0.2
1780 ± II
7.21 ± 0.02
15 ± I
61 ± 2
10.3 ± 0.2
during flushing; P02, oxygen tension; Pcoj, carbon dioxide tension; Hgb, hemoglobin.
*p < 0.01. tp < 0.001.
For the purposes of comparison of pulmonary gas exchange among the groups, the blood gases of the fluent bloodwere compared at 10 minutes after onset of reperfusion. Results are shown in Table III. The mean oxygen tensionsat 10 minutes for the surface cooling (SC), UW, M-UW,andLPDgroupswere61 ± 7,51 ± 7,95 ± 6, and 99 ± 4 mm Hg, respectively. The mean values for the M-UW and LPD groups were significantly higher than either the SC or UW groups, where there was no significant difference between the M-UW and LPD groups or between the SC and UW groups. The oxygen tension and carbon dioxide tension of the effluent blood at I, 5, and 10 minutes are graphically depicted in Figs. 2 and 3, respectively. Immediately after the onset of blood reperfusion through the isolated lung, the mean pulmonary artery pressure rose sharply within the first 30 seconds and then gradually declined. This is depicted in Fig. 4. The mean values for the pulmonary pressure at 10 minutes for the SC,UW, M-UW, and LPD groups were 15 ± 1,19 ± 2, 14 ± I, and 13 ± I mm Hg, respectively, as shown in Table III. There was a significant difference between the UW and LPD groups. The peak airway pressure of the M-UW and LPD groups declined initially and then stabilized. In the SC and UW groups, however, the peak airway pressure gradually rose, associated with increased tracheal fluid and pulmonary edema in most of the animals in the UW group. The peak airway pressures for the four groups of animals are depicted in Fig. 5. Discussion The isolated perfused rabbit lung model used for this study was previously developed as a screening technique for the many factors affecting lung preservation. Using this model we8 have previously demonstrated its sensitiv-
Table III. Summary of results at 10 minutes Group Group I
Peal (mm Hg)
Paw (mm Hg)
61 ± 7
37 ± I
15 ± I
16 ± 2
19 ± 2
21 ± 5
35 ± 2
14 ± I
II ± I
Group 2 (UW) Group 3 (M-UW) Group 4
* [40 ±
All values are mean ± standard error of the mean. POl, oxygen tension; Pcoj, carbon dioxide tension; PPA, mean pulmonary artery pressure during perfusion; Paw, peak airway pressure during perfusion. *p < 0.01. tp < 0.001.
ity to duration of ischemic time and to the temperature at which the lungs are preserved. In the current experiments we used preservation at 10° C, because this had previously been shown to be the optimal temperature for preservation with this model. We chose 30 hours of preservation, because previous experiments had demonstrated that lung function after preservation with simple SC began to deteriorate when preservation time was extended to 30 hours. The reproducibility of this rabbit model was demonstrated by the similar results obtained with 30 hours of SC in the current experiments compared with our previous series. The current experiments demonstrated that rabbit lungs were better preserved with the M-UW and LPD solutions than with simple SC, which confirms the value of a flush technique for extended preservation. On the
The Journal of Thoracic and Cardiovascular Surgery
Miyoshi et al.
o UW • M-UW '" LPD
Fig. 2. Oxygen tension (P02) at 1,5, and 10 minutes after start
60 40 20 0
Perfusion Time (min)
Perfusion Time (min)
Fig. 4. Mean pulmonary artery pressure (PPA) showing sharp
rise immediately after onset of blood reperfusion.
of reperfusion. mmHg
o SC o UW
5 Perfusion Time (min)
Fig. 3. Carbon dioxide tension (Peo2) at 1, 5, and 10 minutes
after start of reperfusion. other hand, the absence of difference between SC and flushing with the UW solution suggests that the composition of the flushing solution is as important as use of a flushing technique. For solid organ preservation, the use of an intracellular fluid type of solution was an issue thought to be a key for success. Collins, Bravo-Shugarman, and Terasaki" believed that eliminating the concentration gradients for sodium and potassium ions between the inside and outside of the cell would prevent ionic shifts and subsequent cell damage. However, ionic shifts between the inside and
outside of the cell were subsequently demonstrated not to be the cause of deteriorating renal function during preservation.l'" II Jamieson and colleagues I2also showed that there was no difference in rabbit liver preservation between the UW solution with high potassium (120 mmoljL) and low sodium (30 mmoljL) content and the modified UW solution in which the potassium content was lowered to 40 mmoljL and the sodium content raised to 100 mmoljL. Very recently Moen and colleaguesl'' and Sumimoto and colleagues!" reported that'the lowpotassium UW solution more effectively preserved the liver than the high-potassium UW solution in dog and rat models, respectively. Fujimura and coworkers'< demonstrated that the extracellular fluid type solutions were preferable to the intracellular fluid type for 24-hour canine lung preservation. When we initially developed the isolated rabbit perfusion model, our attempts to flush the rabbit lungs with intracellular fluid solution, such as Euro-Collins, proved impossible because of very high perfusion pressures. On the other hand, the low-potassium LPD solution could readily be used to flush the lungs at low perfusion pressure." We conjectured that the high potassium content of intracellular fluid type solutions caused severe pulmonary vasoconstriction in the rabbit model. This was not the case with flushing of canine lungs, but, nonetheless, we recently have demonstrated the superiority of an extracellular fluid type solution to an intracellular fluid type for canine lung preservation." Because of this experience we anticipated that the UW solution might not prove to be a suitable perfusate because of its high potassium content. It was for this reason that we chose to eval-
Volume 103 Number 1
UW solution for rabbit lung preservation
uate both standard UW solutionand a modified solution withlowpotassiumand high sodiumcontent. The current experiments did in fact confirm that the M-UW solution was superior to the standard UW solution. The potentiallyadverseeffectof a high-potassiumflush solution was reported by Kohno and coworkers.l" who demonstrated that perfusion pressure was markedly elevated during flushing of the rat coronary artery with Euro-Collins solution, which contains a 117 mmoljL concentrationof potassium.A postulated potential mechanism for calcium influx was a high-potassium-induced depolarization of the cell membrane, which activates the slow calcium channel.l? The augmented calcium influx constricts the coronary vascular smooth muscle and the myocardium. Because of this hypothesis, the optimal concentration of potassium in cardioplegic solutions is now widely accepted to be in the range of 10 to 40 mrnol/L." The same mechanism as found in the coronary artery and myocardium might occur in the pulmonary smooth muscle. The high potassiumconcentration wasthought to be important to preserve the parenchymal cells of solid organs(but it was not). In lung preservation, the optimum protection.of pulmonary vessels may require a low-potassium flush solution. Thus the superiorresultsobtained with the M -UW and LPD solutions compared with the standard UW solutions might be explained by the low potassium content of the former two solutions compared with the latter solution. Because the M-UW and LPD solutions both had low potassium content, whereas the M-UW solution containedthe impermeants that have provedso important for extending solidorgan preservation,such as pancreas, liver, and kidney, one might speculate as to why the M-UW solution did not givesuperior results in these experiments. Because lungfunctionwasas wellpreservedwith the LPD solution as with the M-UW solution,we have speculated that hypothermia-induced cell swelling may, in fact, not be occurring despite the absence of impermeants in lung preservation. The lungs were stored at 10° C, at which temperature activity of adenosinetriphosphatase may have been preserved to maintain the sodium pump. In these experimentsthe lungswere stored in the inflated condition,and oxygen in the alveoli may have been suppliedby diffusion notonly to the lung parenchyma but also to the endothelial cells to produce the adenosine triphosphatase. Our recentfindings that lungs inflated with room air are better preserved than lungsinflatedwith 100% nitrogenseem to support these speculations. Thus maintenance of optimum conditions for continued metabolism in the cold, preserved lungs may be important.
5 Perfusion Time
Fig. 5. Peak airway pressures (Paw) for groups 1 to 4.
In the search for the optimum conditions for lung preservation,further experiments will be required to systematically definethe nature of the injury occurring during ischemia and reperfusion and the optimum flushing and storage conditions required for extended preservation. One of the limitations of the current model is the relativelyshort period of reperfusion caused by the need for a large amount of autologous venous blood even for short perfusionperiods.Recirculation of the blood perfusate would eliminate the opportunity of studying gas exchangewith this model.To address this concern we are currently developing a similar model, using an anesthetized rabbit attached to the isolated lung by means of cross-circulation of venous blood from the rabbit to the lung and back to the rabbit. Such a model will permit evaluation of both the ischemic injury, as in these experiments, and any reperfusion injury caused by restoration of circulation to the preserved lungs. Wewould like to acknowledge the technical expertise of Ed Casabar, PharmD, and Jim Wilson, Supervisor, Pharmacy Manufacturing, forproviding theflush solutions. Wealso thank Dennis Gordon for his technical assistance. Statistical advice was obtained from Richard B. Schuessler, PhD. REFERENCES 1. Belzer OF, Southard JH. Principles of solid-organ preservation by cold storage. Transplantation 1988;45:673-6. 2. Wahlberg JA,Love R, Landegaad L,Southard JH, Belzer OF. 72-hour preservation of the canine pancreas. Transplantation 1987;43:5-8. 3. Jamieson NV,Sundberg R, Lindell S,et al.Preservation of the canine liver for 24-48 hours using simple cold storage with UW solution. Transplantation 1988;46:517-22. 4. Ploeg RJ, Goossens D, McAnulty JF, et al. Successful
The Journal of Thoracic and Cardiovascular Surgery
3 2 Miyoshi et al.
72-hour cold storage of the dog kidney with UW solution. Transplantation 1988;46:191-6. Fujimura S, Handa M, Kondo T, Ichinose T, Shiraishi Y, Nakada T. Successful 48-hour simple hypothermic preservation of canine lung transplants. Transplant Proc 1987; 19:1334-6. Keshavjee SH, Yamazaki F, Cardosa PF, McRitchie DI, Patterson GA, Cooper JD. A method for safe 12 hour pulmonary preservation. J THORAC CARDIOVASC SURG 1989; 98:529-34. Yamazaki F, Yokomise H, Keshavjee SH, et al. The superiorityon an extracellular fluid solution over Euro-Collins' solution for pulmonary preservation. Transplantation 1990;49:690-4. Wang LS, Yoshikawa K, Miyoshi S, et al. The effect of ischemic time and temperature on lung preservation in a simple ex vivo rabbit model used for functional assessment. J THORAC CARDIOVASC SURG 1989;98:333-42. Collins GH, Bravo-Shugarman MB, Terasaki PI. Kidney preservation for transplantation. Initial perfusion and 30 hour ice storage. Lancet 1969;2:1219-22. Downes GL, Hoffmann RM, Huang J, Belzer OF. Mechanism of action of washout solutions for kidney preservation. Transplantation 1973;16:45-53. Green CJ, Pegg DE. Mechanism of action of "intra-
cellular" renal preservation solutions. World J Surg 1979;3:115-20. Jamieson NY, Lindell S, Sundberg R, Southard JH, Belzer OF. An analysis of the components in UW solution using the isolated perfused rabbit liver. Transplantation 1988;46:512-6. Moen J, Claesson K, Pienaar H, et al. Preservation of dog liver, kidney, and pancreas using the Belzer-U'W solution with a high-sodium and low-potassium content. Transplantation 1989;48:1-5. Sumimoto R, Jamieson NY, Wake K, Kamada N. 24-Hour rat liver preservation using UW solution and some simplified variants. Transplantation 1989;48:1-5. Fujimura S, Kondo T, Handa M, et al. Successful 24-hour preservation of canine lung transplants using modified extracellular fluid. Transplant Proc 1985;17:1466-7. Kohno H, Shiki K, Ueno Y, Tokunaga K. Cold storage of the rat heart for transplantation: two types of solution required for optimal preservation. J THORAC CARDIOVASC SURG 1987;93:86-94. Reuter H. Properties of two inward membrane currents in the heart. Annu Rev PhysioI1979;41:413-24. Hearse DJ, Braimbridge MY, Jynge P. Protection of the ischemic myocardium: cardioplegia. New York: Raven Press, 1981;209-99.